Content uploaded by Janaína Dernowsek
Author content
All content in this area was uploaded by Janaína Dernowsek on Sep 03, 2018
Content may be subject to copyright.
Content uploaded by Janaína Dernowsek
Author content
All content in this area was uploaded by Janaína Dernowsek on Dec 05, 2017
Content may be subject to copyright.
1
This article is protected by copyright. All rights reserved
Article
Posttranscriptional Interaction between miR-450a-5p and miR-28-5p and STAT1
mRNA Triggers Osteoblastic Differentiation of Human Mesenchymal Stem Cells†
Janaína A. Dernowsek1, Milena C. Pereira1, Thaís A. Fornari1, Claudia Macedo1,
Amanda F. Assis1, Paula B. Donate1, Karina F. Bombonato-Prado2, Maria Rita Passos-
Bueno3, Geraldo A. Passos1,2*
Short title: Posttranscriptional Control during Osteoblastic Differentiation of Stem Cells
1) Molecular Immunogenetics Group, Department of Genetics, Ribeirão Preto
Medical School, University of São Paulo (USP), Ribeirão Preto, São Paulo, Brazil.
2) Department of Morphology, Physiology and Basic Pathology, School of Dentistry
of Ribeirão Preto, USP, Ribeirão Preto, São Paulo, Brazil.
3) Department of Genetics and Evolutionary Biology, Institute of Biosciences, USP,
São Paulo, Brazil.
*correspondence to: Dr. Geraldo A. Passos, Department of Genetics, Ribeirão Preto
Medical School, University of São Paulo (USP), Via Bandeirantes, 3900 ZIP-CODE
14049-900, Ribeirão Preto, São Paulo, Brazil. E-mail: (passos@usp.br) Tel.: +55 16
3315-3030.
Competing interests: Authors declare no competing interests.
†This article has been accepted for publication and undergone full peer review but has not been
through the copyediting, typesetting, pagination and proofreading process, which may lead to
differences between this version and the Version of Record. Please cite this article as doi:
[10.1002/jcb.26060]
Additional Supporting Information may be found in the online version of this article.
Received 5.February 2017; Revised 11 April 2017; Accepted 12 April 2017
Journal of Cellular Biochemistry
This article is protected by copyright. All rights reserved
DOI 10.1002/jcb.26060
2
This article is protected by copyright. All rights reserved
Abstract
We demonstrate that the interaction between miR-450a-5p and miR-28-5p and signal
transducer and activator of transcription 1 (STAT1) mRNA correlates with the
osteoblastic differentiation of mesenchymal stem cells from human exfoliated
deciduous teeth (shed cells). STAT1 negatively regulates runx-related transcription
factor 2 (RUNX2), which is an essential transcription factor in this process. However, the
elements that trigger osteoblastic differentiation and therefore pause the inhibitory
effect of STAT1 need investigation. Usually, STAT1 can be posttranscriptionally regulated
by miRNAs. To test this, we used an in vitro model system in which shed cells were
chemically induced toward osteoblastic differentiation and temporally analyzed,
comparing undifferentiated cells with their counterparts in the early (2 days) or late (7
or 21 days) periods of induction. The definition of the entire functional genome
expression signature demonstrated that the transcriptional activity of a large set of
mRNAs and miRNAs changes during this process. Interestingly, STAT1 and RUNX2
mRNAs feature contrasting expression levels during the course of differentiation. While
undifferentiated or early differentiating cells express high levels of STAT1 mRNA, which
was gradually down-regulated, RUNX2 mRNA was up-regulated toward differentiation.
The reconstruction of miRNA-mRNA interaction networks allowed the identification of
six miRNAs (miR-17-3p, miR-28-5p, miR-29b, miR-29c-5p, miR-145-3p and miR-450a-
5p), and we predicted their respective targets, from which we focused on miR-450a-5p
and miR-28-5p STAT1 mRNA interactions, whose intracellular occurrence was validated
through the luciferase assay. Transfections of undifferentiated shed cells with miR-450a-
5p or miR-28-5p mimics or with miR-450a-5p or miR-28-5p antagonists demonstrated
that these miRNAs might play a role as posttranscriptional controllers of STAT1 mRNA
during osteoblastic differentiation. This article is protected by copyright. All rights reserved
Keywords: Osteoblastic differentiation, STAT1, RUNX2, miR-450a-5p,
posttranscriptional control, microRNA mimic, microRNA antagonist.
3
This article is protected by copyright. All rights reserved
Introduction
Bone formation is a complex phenomenon that involves sequential steps of cell
differentiation. Mesenchymal stem cells (MSCs) and bone marrow stromal cells produce
differentiated osteoblasts, which can mature into osteocytes a terminally differentiated
cell type involved in the inhibition of bone resorption, consequently preventing
osteoporosis and fracture [Bonewald et al., 2013; Sobacchi et al., 2013; Prideaux et al
2016].
The occurrence of these steps involves the activation of key signaling pathways,
including the transforming growth factor beta (TGFB1), bone morphogenetic protein
(BMP) [Santibanez and Koic, 2012], Wnt [Wang et al., 2014], and Notch [Regan and Long,
2013] signaling pathways, along with the transcription factors (TFs) runt-related
transcription factor 2, RUNX2 (also referred as CBFA1); SP7 (also referred as osterix);
and beta-catenin (CTNNBIP1) [Komori, 2011; Weivoda et al., 2016]
RUNX2 represents a pivotal element controlling MSC differentiation into
osteoblasts by regulating key genes associated with osteoblast fate commitment, such
as collagen, type 1, alpha 1 (COL1A1); alkaline phosphatase (ALPL); integrin-binding
sialoprotein (IBSP); osteopontin (SPP1); and osteocalcin (OC/BGLAP) [Komori et al.,
1997; Ducy et al., 1999; Lian et al., 2004; Zhang et al., 2006; Karsenty, 2008; Dalle
Carbonare et al., 2012; Bruderer et al., 2014; Vilmalraj et al., 2015; Wysokinski et al.,
2015; Gomes-Picos and Eames, 2015; Xu et al, 2015].
Osteoblast differentiation is a finely controlled process, in which RUNX2 is
attenuated by signal transducer and activator of transcription 1 (STAT1), a member of
the signal transducers and activators of transcription family [Kim et al., 2003]. This
process is complex as STAT1 is considered an essential TF for both type I and type II
interferon (IFN) responses, whose pathway establishes a connection between bone
formation and immune response cytokines, termed osteoimmunology [Greenblatt and
Shim, 2013; Walsh et al., 2014; Tompkins, 2016; Takayanagi et al., 2005; 2009; 2016].
The demonstration of increased bone formation after the inhibition of STAT1 by
fludarabine [Tajima et al., 2010] confirms that this transcription activator functions as
an upstream negative regulator in this process.
4
This article is protected by copyright. All rights reserved
However, a central question remains: what are the factors that physiologically
pause the inhibitory effect of STAT1 on RUNX2 during osteoblast differentiation? Using
chromatin immunoprecipitation, Fujie and colleagues [Fujie et al., 2015] have shown
that STAT1 is a direct target of the B-cell lymphoma 6 (BCL6) transcriptional repressor
and that it may function through negative regulation.
Nevertheless, we recognize that the factors involved in pausing the effects of
STAT1 in the context of osteoblastic differentiation still need to be clarified.
Posttranscriptional regulation involving microRNAs (miRNAs) is another plausible
pathway that may act on STAT1. The miRNAs make available a novel dimension to
posttranscriptional control and this species of RNA might play a role on STAT1 mRNA in
the context of osteoblastic differentiation. These molecules are single-stranded non-
coding RNAs that have emerged as regulators of diverse physiological and pathological
processes, such as cell proliferation, differentiation, apoptosis and cancer [Bartel, 2004;
Pasquinelli, 2012].
Osteogenic differentiation, a crucial biological process in bone development,
involves the activation of multiple signaling pathways, including TGFb, BMP, Wnt as well
transcription factors, which are rigidly regulated by miRNAs [Zhang et al., 2011; Vimalraj
et al., 2015]. In fact, we know a number of specific miRNAs, as for example, miR-20a,
miR-22, miR-27, miR-29b, miR-29c, miR-133, miR-141, miR-196a, miR-200a, miR-204,
miR-210, miR-211 and miR-378 that that play their role as posttranscriptional regulators
during osteogenic differentiation [Zhang et al., 2011, Xu et al., 2015; Lian et al., 2017].
Regarding STAT1, in a study using murine mesenchymal stem cells, it was demonstrated
that miR-194 interacts with the STAT1 3´ untranslated region (3´UTR) during the
induction of osteoblast differentiation [Li et al., 2015].
This observation led us to assess this possibility in an unbiased manner by
scanning the entire human functional genome, as well a large set of miRNAs, to identify
those miRNAs that interact with STAT1 mRNA during osteoblastic differentiation of
human stem cells because the information about this mechanism is lacking.
Our starting hypothesis featured two parts: first, we postulated that the
modulation (up- or down-regulation) of mRNAs or miRNAs would be sufficient to
5
This article is protected by copyright. All rights reserved
hierarchize the different periods of in vitro osteoblastic differentiation, and second, we
surveyed the interactions between the modulated mRNAs and miRNAs of these cells
during differentiation.
Materials and Methods (For a flow diagram of the methods employed, see
Supplemental material Figure 1).
Mesenchymal stem cells from human exfoliated deciduous teeth (shed cells)
Shed cells used in this study were isolated by Fanganiello et al (2015), according
to the protocol previously described [Miura et al., 2003]. In this study, we used the shed
cells of a normal male child and cultured them in DMEM medium with Ham F12
supplemented with 15% fetal bovine serum plus 100 U/ml each penicillin and
streptomycin (here referred to as control medium or CMD medium) at 37°C in an
incubator with a 5% CO2 atmosphere. These cells were assayed by flow cytometry using
a BD-FACSCalibur flow cytometer (Becton Dickinson and Company, Franklin Lakes, NJ,
USA) to confirm the expression of the surface markers CD29, CD44, CD13, CD90, and
CD73. To control the presence of hematopoietic stem cell contamination, we tested the
following surface markers: CD34 and CD45/CD14 (data not shown). Cells were labeled
with the monoclonal antibody anti-CD73-phycoerythrin (PE), CD45-fluorescein
isothiocyanate (FITC), CD14-PE, CD34-PE, CD44-FITC, CD29-PE, CD90-PE, CD31-FITC and
CD13-PE. The respective labeled monoclonal antibodies recognizing these markers were
purchased from Becton Dickinson and Company. These cells are referred to as
mesenchymal stem cells from human exfoliated deciduous teeth (shed cells). Shed cells
from fifth-passage cultures were used in all experiments. All of the procedures involving
human cell manipulations were approved by the Committee for Ethics in Research of
the Biosciences Institute, USP, São Paulo, Brazil (approval # 435054).
Monitoring the in vitro osteoblastic differentiation of shed cells
Induction of the osteoblastic differentiation of shed cells was chemically
activated by adding 10-7 M dexamethasone (Sigma-Aldrich), 5 µg/ml ascorbic acid
(Gibco) and 2.16 g/ml β-glycerophosphate (Sigma-Aldrich) to the shed cell cultures (here
referred to as induction medium or IMD medium). The IMD medium was changed every
three days, and the cells were harvested after 2, 7 or 21 days in culture. Undifferentiated
6
This article is protected by copyright. All rights reserved
control cells were cultured in CMD medium. Cell viability was determined by
conventional MTT (3-[4,5-dimethylthiazol-2-yl]-2,5- diphenyltetrazolium bromide,
Sigma-Aldrich, USA) assay [Mosmann, 1983].
The activity of alkaline phosphatase (ALPL) per µg of total protein was
determined by a conventional thymolphthalein (5′,5′′-diisopropyl-2′,2′′-
dimethylphenolphthalein) hydrolysis assay according to the manufacturer´s instructions
(Labtest Diagnostics AS, Belo Horizonte, MG, Brazil). ALPL activity was used as a
biochemical marker for osteoblastic differentiation.
Moreover, ALPL and bone osteocalcin/gamma-carboxyglutamate (gla) protein
(OC/BGLAP) were antibody stained and observed through conventional indirect
immunolocalization to monitor osteoblastic differentiation. The respective anti-bone
ALPL monoclonal antibody (B4-78 hybridoma, diluted 1:100 from Developmental
Studies Hybridoma Bank, University of Iowa, USA) and a goat anti-human OC/BGLAP
polyclonal antibody diluted 1:200 (Santa Cruz Biotechnology) were used. After
incubation with these primary antibodies, cells were stained with a mixture of Alexa
Fluor 594 (red fluorescence)-conjugated goat anti-rabbit secondary antibody (diluted
1:200) (Molecular Probes, Eugene, OR, USA) and Alexa Fluor 488 (green fluorescence).
Replacement of the primary monoclonal antibody with phosphate buffer (PB) was used
as a control.
ALPL was also assayed by conventional Western blotting (WB). A goat IgM
antibody recognizing human ALPL (Sigma-Aldrich, USA) was used as the primary
antibody. The respective anti-goat (IgM) or anti-rabbit horseradish peroxidase-
conjugated antibodies (Abcam) were used as secondary antibodies. WB membranes
were incubated with Pierce ® ECL Western blot substrate (Thermo Fisher Scientific) for
the chemiluminescent detection of protein bands using an ImageQuant LAS 500
apparatus (GE Healthcare). The expression levels of GAPDH or β-actin proteins were
used as references to normalize the WB data of ALPL. In addition, osteoblastic
differentiation was monitored to observe the formation of a mineralized matrix through
the quantification of alizarin red S staining according to a protocol previously described
[Gregory et al, 2004].
7
This article is protected by copyright. All rights reserved
miRNA mimic and miRNA antagonist transfection
The sequences of miRNA mimics and miRNA antagonists were defined and
synthesized by IDT Integrated DNA Technologies, Inc. (IDT, Coralville, IA, USA). The
miRNA antagonist was designed with 2´O-methylation modification (here indicated as
mBASE*) to assure better specificity, resistance to degradation and stability. These
sequences were as follows: miR-450a-5p mimic (MIMAT0001545,
UUUUGCGAUGUGUGUUCCUAAUAU) and miR-450a-5p antagonist
(mU*mA*mU*mU*mA*G*G*A*A*C*A*) and miR-28-5p mimic (MIMAT0000085,
AAGGAGCUCACAGUCUAUUGAG) and mir-28-5p antagonist
(mC*mA*mA*mU*mA*G*A*C*T*G*T*G*A*G*C*mU*mC*mC*mU*mU*).
Transfections of shed cells with these oligonucleotides were performed using
Hyperfect Transfection Reagent® (Qiagen, Hilden, Germany) according to the
manufacturer´s instructions. Briefly, the preparation of oligonucleotide-Hyperfect
reagent complex was made by mixing 40 nM oligonucleotide in 100 µl of serum-free
CMD medium plus 6 µl of Hyperfect reagent. The mixture was vortexed for 10 seconds
and was allowed to stand for 10 minutes at room temperature to form complexes. The
complexes were added to shed cells in a proportion of 40 nM oligonucleotide/1 x 105
cells. These were cultured in six-well Corning® plates for 48 hours for further total RNA
extraction for 72 hours for further biochemical assays. To monitor the oligo uptake we
used an irrelevant Cy3(*)-labeled modified oligo RNA
(T*CUUUCCUCUCUUUCUCUCCCUUGUG*AT*CACAAGGGAGAGAAAGAGAGGAAGG*A)
to transfect shed cells, which were then observed by confocal fluorescence microscopy
(Supplemental material Figure 2).
Functional assays for miRNAs
For the functional assay experiments, we established the following: a control
group (undifferentiated shed cells) and nine experimental groups (exp): exp 1
(undifferentiated shed cells transfected with miR-450a-5p mimic), exp 2
(undifferentiated shed cells transfected with miR-450a-5p antagonist), exp 3
(undifferentiated shed cells transfected with miR-28-5p mimic), exp 4 (undifferentiated
shed cells transfected with miR-28-5p antagonist), exp 5 (shed cells cultured for 14 days
8
This article is protected by copyright. All rights reserved
in IMD medium), exp 6 (shed cells cultured for 14 days in IMD medium transfected with
miR-450a-5p mimic), exp 7 (shed cells cultured for 14 days in IMD medium transfected
with miR-450a-5p antagonist), exp 8 (shed cells cultured for 14 days in IMD medium
transfected with miR-28-5p mimic) and exp 9 (shed cells cultured for 14 days in IMD
medium transfected with miR-28-5p).
Total RNA preparation and quality control
The total RNA of the shed cells was extracted from approximately 1 x 107 control
undifferentiated, differentiated or oligonucleotide-transfected cells using the mirVana
total RNA isolation kit (Ambion, NY, USA) according to the manufacturer’s instructions.
Quality control of the total RNA preparations was checked using microfluidic
electrophoresis in RNA Nano Chips and the Agilent 2100 Bioanalyzer apparatus (Agilent
Technologies, Santa Clara, CA, USA). RNA preparations were confirmed to be free of
proteins and phenol by UV spectrophotometry. Only RNA samples that were free of
proteins and phenol and featured an RNA Integrity Number (RIN) ≥ 9.0 were used.
Microarray hybridizations and data analysis
We performed this procedure as previously described by our group [Donate et
al., 2013; Fornari et al., 2015]. Briefly, for each RNA species, the hybridizations, data
analysis and validations of the miRNA interactions were as follows:
For miRNA expression profiling
We used total RNA for direct labeling with Cy3 using the Agilent miRNA Complete
Labeling and Hybridization Kit (Agilent Technologies, Mississauga, ON, Canada). The Cy3-
labeled RNA samples were hybridized to Agilent human 8 x 15 K format oligonucleotide
miRNA microarrays for 20 h at 55°C, and the slides were washed according to the
manufacturer’s instructions (Agilent Technologies) and scanned using an Agilent DNA
microarray scanner.
9
This article is protected by copyright. All rights reserved
For mRNA expression profiling
We used total RNA to synthesize dscDNA and Cy3-CTP-labeled complementary
amplified RNA (cRNA) using the Agilent Linear Amplification Kit (Agilent Technologies,
Santa Clara, CA, USA) according to the manufacturer´s instructions. The Cy3-cRNA
samples were hybridized to Agilent human 4 x 44 K format oligonucleotide microarrays
(Agilent Technologies) for 18 h at 60°C, washed according the manufacturer’s
instructions (Agilent Technologies) and scanned using an Agilent DNA microarray
scanner.
Microarray data analysis
Hierarchical clustering
The hybridization signals from the scanned miRNA or mRNA oligonucleotide
microarrays were extracted using Agilent Feature Extraction software, version 10.5. The
expression profiles from independent preparations of undifferentiated control,
differentiated or transfected shed cells were analyzed by comparing the microarray
hybridizations of the respective samples. The numerical quantitative microarray data
were normalized to the 75th percentile and analyzed using the GeneSpring GX
bioinformatics platform (http://www.agilent.com/chem/genespring) according to
default instructions, allowing hierarchical clustering of the samples of cells or types of
RNAs through Student's T-test analysis (P < 0.05), with a fold change ≥ 2.0, and
uncentered Pearson correlation metrics [Eisen et al., 1998].
Dendrograms were used to represent the similarities and dissimilarities in the
expression profiles between the samples, mRNAs or miRNAs, and a colored heat map
was used to represent the variability in RNA expression. The GeneSpring platform or
Panther gene classification system (http://www.pantherdb.org) was used to assess the
biological functions of mRNAs; for miRNAs, we used the miRBase databank
(www.mirbase.org).
The microarray data of this study are available at EMBL EBI ArrayExpress Data
Bank (https://www.ebi.ac.uk/arrayexpress) under Array Express accessions E-MTAB-
3010 (mRNAs) and E-MTAB-3077 (miRNAs).
10
This article is protected by copyright. All rights reserved
Reconstruction of the miRNA-mRNA interaction networks
Briefly, the miRNA-mRNA interaction networks were reconstructed based on the
respective miRNA and mRNA expression profiles to identify candidate miRNA-mRNA
target pairs that were best supported by the expression data. The GenMir++ algorithm
[Huang et al., 2007a,b], which is available at (http://www.psi.toronto.edu/genmir), was
used to establish such interactions. This algorithm calculates the scores to reproduce an
mRNA profile using a weighted combination of the genome-wide average normalized
expression profile and the negatively weighted profiles of a subset of miRNA regulators.
The networks were graphically represented using the Cytoscape version 3.3.0 program
(www.cytoscape.org). We used the TargetScanS database (http://www.targetscan.org)
to verify the prediction of the selected target mRNAs.
Validation of the miRNA-mRNA interactions
Determination of the hybridization minimum free energy (mfe)
Using the miRNA-mRNA interaction networks generated by the GenMir++
algorithm, we selected pairs based on the biological function of the mRNA target.
Annealing was validated using the RNA-Hybrid algorithm [Rehmsmeier et al., 2004;
Krüger and Rehmsmeier, 2006], which is available at (https://bibiserv2.cebitec.uni-
bielefeld.de/rnahybrid). This algorithm estimates the most favorable pairing between a
given miRNA and its mRNA target by calculating the minimum free energy based on a
thermodynamic state that postulates that an RNA duplex is more stable and
thermodynamically stronger when free energy is low [Lewis et al., 2005].
Luciferase reporter gene assay (LRGA)
The complementary ds-oligonucleotide pairs containing a portion of the
predicted miRNA binding sites of the BMP6, TM4SF1 or STAT1 3´ UTRs, as validated by
the RNA-Hybrid algorithm, were synthesized using GBlocks® technology by Integrated
DNA Technologies (IDT, Coralville, IA, USA). The oligonucleotides were cloned into the
pmirGLO vector (Promega Corporation, USA) between the XhoI/XbaI restriction sites of
the polycloning site (PCS), resulting in the miRNA target region assuming the correct 5´
to 3´ orientation immediately downstream of the luciferase gene. For selected targets,
we introduced point mutations into the 7-nucleotide seed-binding sequence. These
11
This article is protected by copyright. All rights reserved
constructs, named “pMIR-BMP6”, pMIR-TM4SF1” or “pMIR-STAT1” for the wild-type
sequences and “pMIR- BMP6 (m)”, “pMIR- TM4SF1 (m)” or “pMIR-STAT1(m)” for the
mutant sequences, were selected by colony polymerase chain reaction (PCR) using a pair
of primers flanking the vector poly cloning site. We used Escherichia coli DH5α for
cloning.
For the LRGA, 0.2 µg of each pmirGLO construct was transfected into human HEK-
293T cells (6 x 104 cells/well) with 1.6 pmol of miR-450a-5p, miR-28-5p or scrambled
control miRNA (Thermo Scientific Dharmacon, Waltham, MA, USA) in a 96-well plate.
Transfections were performed using Attractene Transfection Reagent (Qiagen, Hilden,
Germany) according to the manufacturer´s instructions. Transfected cells were
incubated at 37°C in a 5% CO2 incubator; 24 h after transfection, the cells were lysed in
Passive Lysis Buffer. Firefly and renilla luciferase activity were measured in a Synergy 2
luminometer (BioTek Instruments, Inc, Winooski, VT, USA) using the Dual-Luciferase
reporter system (Promega Corporation, USA) according to the manufacturer´s
instructions.
The LRGA results are presented as the standard error of the mean (SEM). The
differences were evaluated by one-way ANOVA followed by Student´s t test (two
groups). P < 0.05 was considered to be statistically significant.
Our laboratory has national biosafety permission (National Technical Committee
for Biosafety, Brasilia, Brazil, CTNBio Permit No. 0040/98).
Reverse transcription quantitative real-time PCR (RT-qPCR)
miRNAs
The miRNAs that were differentially expressed, as observed by microarray
hybridizations, and participating in miRNA-mRNA interaction networks had their
expression levels confirmed by RT-qPCR. Complementary DNA (cDNA) was synthesized
from 10 ng of total RNA using the High-Capacity RNA-to-cDNA kit (Applied Biosystems-
Life Technologies, USA). The PCR products were amplified from cDNA samples using the
TaqMan MicroRNA Assay kit (Applied Biosystems-Life Technologies, USA) according the
manufacturer´s instructions.
12
This article is protected by copyright. All rights reserved
We evaluated the expression levels of the following miRNAs (their respective
accession numbers and mature sequences are also informed): miR-29b
(MIMAT0000100, AACACTGATTTCAAATGGTGC), miR-29c-5p (MIMAT0004673,
UGACCGAUUUCUCCUGGUGUUC), miR-17-3p (MIMAT0000071,
ACUGCAGUGAAGGCACUUGUAG), miR-145-3p (MIMAT0004601,
GGAUUCCUGGAAAUACUGUUCU), miR-450a-5p (MIMAT0001545,
UUUUGCGAUGUGUUCCUAAUAU), miR-28-5p (MIMAT0000085,
AAGGAGCUCACAGUCUAUUGAG). These miRNAs were selected based on their targets
within miRNA-mRNA interaction networks. miR-26b-3p (MIMAT0004500,
CCUGUUCUCCAUUACUUGGCUC) was used as an endogenous control.
Thermal cycling was completed using a StepOne Real-Time PCR System (Applied
Biosystems, USA) as follows: 50°C for 2 min, 95°C for 15 min, and 60°C for 1 min (40
cycles). The 2 - ΔΔCT was used as a relative normalization method. We used the GraphPad
Prism 5.00 tool (http://www.graphpad.com/prism/Prism.html) to run one-way or two-
way ANOVA with Bonferroni´s correction.
mRNAs
RT-qPCR was also used to evaluate the transcriptional expression levels of those
differentially expressed mRNAs, as observed by microarray hybridizations, participating
in miRNA-mRNA interaction networks. The mRNAs (cDNAs) are indicated as follows. The
respective GenBank accession numbers and sequences of their respective 5´ to 3´
forward and reverse oligonucleotide primers are in parentheses: bone morphogenetic
protein 6 (BMP6, NM_001718, GACATGGTCATGAGCTTTGTGA), prostaglandin-
endoperoxide synthase 2 (PTGS2, NM_000963, ATAAGCGAGGGCCAGCTTTC), paired-like
homeodomain 1 (PITX1, NM_002653, GCGAGTCGTCTGACACGGAG,
TCTTCTTTGGCTGGGTCGTCTG), transmembrane 4 L six family member 1 (TM4SF1,
AL832780, CCGCTTCGTGTGGTTCTTTT, CAAACGATGTGCGATGCTTTC), signal transducer
and activator of transcription 1 (STAT1, NM_007315, TAATCAGGCTCAGTCGGGGA,
TCCAGGCTCTTGATTTCATGCT), alkaline phosphatase (ALPL, NM_000478,
CCACGTCTTCACATTTGGTG, AGACTGCGCCTGGTAGTTGT), runt-related transcription
factor 2 (RUNX2, NM_001024630, CCTTGGGAAAAATTCAAGCA,
AACACATGACCCAGTGCAAA) and osteocalcin (OC/BGLAP, NM_199173,
13
This article is protected by copyright. All rights reserved
GGCAGCGAGGTAGTGAAGAG, CTGGAGAGGAGCAGAACTGG). Glyceraldehyde-3-
phosphate dehydrogenase mRNA (cDNA) (GAPDH, NM_002046,
ACGACCAAATCCGTTGACTC, CTCTGCTCCTCCTGTTCGAC) was used as an endogenous
control. The primer sequences for each mRNA covering intron/exon junction from their
full length cDNAs with a fixed melting temperature of 60°C were defined using the
Primer3 program (http://frodo.wi.mit.edu/primer3).
Complementary DNA (cDNA) was synthesized from 2 µg of total RNA using the
High-Capacity RNA-to-cDNA kit (Applied Biosystems-Life Technologies, USA), according
to the manufacturer´s instructions. We used SYBR Green Real-Time PCR Master Mix, and
thermal cycling was performed using the StepOne Real-Time PCR System (Applied
Biosystems, USA). The 2 - ΔΔCT method was used for relative normalization. We used the
GraphPad Prism 5.00 tool to perform one-way or two-way ANOVA with Bonferroni´s
correction.
Results
Transcriptional profiles of mRNAs and miRNAs during osteoblastic differentiation as
evaluated by microarrays
We compared the microarray transcriptional profiles of the mRNAs and miRNAs
of shed cells cultured in CDM versus those cells cultured in IMD medium during time
courses of 2, 7 and 21 days. From the set of mRNAs and miRNAs present in each type of
microarray, we identified those that were differentially expressed in the course of
osteoblastic differentiation [P < 0.05, false discovery rate (FDR) = 0.05 and fold change
≥ 2.0]. The dendrograms and heat maps depicted in Figures 1-2 show that shed cells
feature unique mRNA or miRNA expression profiles according to each time point
analyzed, i.e., 2 days (Figure 1a), 7 days (Figure 1b) and 21 days (Figure 1c) for mRNAs
and 2 days (Figure 2a), 7 days (Figure 2b) and 21 days (Figure 2c) for miRNAs. The up- or
down-regulated mRNAs and miRNAs from each time point were then reanalyzed using
the GenMir++ algorithm.
14
This article is protected by copyright. All rights reserved
Reconstruction of miRNA-mRNA interaction networks
The respective sets of differentially expressed mRNAs and miRNAs were
separately imputed into the GenMir++ algorithm to reconstruct three miRNA-mRNA
interaction networks, as depicted according to each time point of the osteoblastic
differentiation of shed cells, i.e., 2 days (Figure 3a), 7 days (Figure 3b) and 21 days (Figure
3c) analyzed.
We note that in the early stage of differentiation (2 days), miR-450a-5p played a
role as a controlling node, establishing interactions with a set of 62 downstream mRNAs,
including those engaged in osteoblastic differentiation—for example, PTGS2, TM4SF1,
PITX1 and STAT1.
In the intermediary phase of differentiation (7 days), a more diverse number of
miRNAs (miR-28-5p, miR-7, miR-17-3p, miR-29c-5p, miR-369-5p and miR-145-3p)
established interactions with 181 downstream mRNAs. The interaction between miR-
28-5p-BMP6 mRNA is highlighted due to the functional role exerted by BMP6 protein on
the differentiation of osteoblasts from mesenchymal precursors [Rickard et al., 1998].
In the late phase of differentiation (21 days), which is characterized by the
formation of mineralized matrix, we note that miR-29b played a role as a controlling
node, establishing interactions with a set of 273 downstream mRNAs, including those
engaged in several signaling pathways that govern osteoblastic differentiation: CTNNA2,
CTNNAL1, FAT4, ITPR1, TL4 and PCDH18 (Wnt signaling pathway); COL5A2, LIMS2,
ASAP1, COL3A1, COL4A5, and COL4A4 (Integrin signaling pathway); and CTNNA2, FAT4
and PCDH18 ( Cadherin signaling pathway). Irrespective of their type of modulation—
i.e., up- or down-regulation—the miRNAs established interactions with mRNA targets
involved in major gene-ontology (GO) processes (Table 1).
During osteoblastic differentiation, shed cells cultured in IMD medium presented
several mRNA targets, and we highlight the following targets, with their
molecular/biological processes in parenthesis: early phase, 2 days: STAT1
(Developmental process GO:0032502), PTGS2, and PITX1 (Biological regulation
GO:0065007); intermediary phase, 7 days: BMP6 (Developmental process
GO:0032502), and PTGS (Developmental process GO:0032502); and late phase, 21 days:
15
This article is protected by copyright. All rights reserved
BMP2K (Metabolic process GO:0008152), CUL4B (Cellular process GO:0009987),
COL3A1, COL4A4, COL4A5 and COL5A2 (Biological adhesion GO:0022610 and Cellular
component organization or biogenesis GO:0071841).
Transcriptional expression and functionality of the in vitro osteoblastic differentiation
model system
To evaluate the relative expression levels of miR-450a-5p and miR-28-5p (Figure
4 a-d), as well as the expression levels of mRNAs encoding markers of osteoblast
differentiation, such as RUNX2, STAT1, ALPL and OC/BGLAP of shed cells transfected (or
not) with the respective miRNA mimics or antagonists, we performed RT-qPCR assays in
parallel with formation of mineralization nodules (Figure 5 a-d and Figure 6 a-d).
Expression of ALPL protein, as evaluated by western blotting, is increased in
undifferentiated shed cells transfected with miR-450a-5p or with miR-28-5p as well as
in shed cells cultured during 14 days in IDM (Figure 5 e and Figure 6 e).
Figure 5 f and Figure 6 f shows that expression of ALPL or OC/BGLAP protein can
be modulated by transfection of miR-450a-5p or miR-28-5p mimic or antagonist, as
determined by fluorescence immunolocalization. While in undifferentiated control shed
cells these proteins are not localized, their expression was clear when those cells were
transfected with miR-450a-5p or miR-28-5p mimic but not with miR-450a-5p or miR-28-
5p antagonist. However, shed cells during osteoblastic differentiation, i.e., cultured
during 14 days in IMD, clearly express these proteins, which were increased when
transfected with miR-450a-5p or miR-28-5p mimic and completely abolished by
transfection with miR-450a-5p or miR-28-5p antagonist.
Validation of miRNA-mRNA interactions
Minimal free energy (RNA-Hybrid algorithm)
The pairing of the Homo sapiens STAT1 (acc NM_007315) mRNA target region
and miR-450a-5p was predicted to occur at nucleotide positions 515-525 of its 2,254-bp
3´UTR; TM4SF1 mRNA (acc NM_014220.2) and miR-450a-5p, at nucleotide positions
253-262 of its 609-bp 3´UTR; and miR-28-5p, at nucleotide positions 400-406 of its 609-
bp 3´UTR.
16
This article is protected by copyright. All rights reserved
Because these interactions have not been previously reported, we used the
RNAHybrid (http://bibiserv.techfak.uni-bielefeld.de/rnahybrid/) algorithm, which
calculates the most favorable hybridization between the miRNA and the predicted
3´UTR segment of its mRNA target by calculating the thermodynamic minimum free
energy (MFE). Artificial point mutations changed the hybridization MFEs and,
consequently, the likelihood of miRNA-mRNA hybridizations (Figure 7 a-d).
Luciferase reporter gene assay (LRGA)
The validation of the occurrence of the miRNA-mRNA interactions within the
cellular milieu was performed using the LRGA. This assay allowed the confirmation that
miR-450a-5p interacts with STAT1 or TM4SF1 mRNA 3´UTRs and that miR-28-5p
interacts with TM4SF1 or BMP6 mRNA 3´UTRs. Artificially introduced point mutations
(base substitutions or deletions) in the original wild-type 3´UTR sequences significantly
abolished the miRNA interactions. This confirms the necessity of the specificity and
exactness of base pairing for hybridizations (Figure 7 e-h).
Discussion
In this study, we wondered whether miRNAs play a role in the osteoblastic
differentiation of mesenchymal stem cells from human exfoliated deciduous teeth (shed
cells) as upstream inhibitors of key mRNAs related to the stemness state and/or of those
considered inhibitors of differentiation. This served as a suitable model system to test
our starting hypothesis, as mentioned above.
We began these studies by establishing a model system, which consisted of
inducing shed cells into osteoblastic differentiation in vitro under the effects of ascorbic
acid, beta glycerophosphate and dexamethasone, with no significant changes in cell
viability. We were interested in using this cell type based on results previously published
[Faganiello et al., 2015], whose authors showed that these cells feature increased
osteopotential. In addition, we verified that these cells expressed mesenchymal cell
17
This article is protected by copyright. All rights reserved
surface markers, namely CD90, CD73, CD29, CD105, CD13, CD44 and not the
hematopoietic markers CD34 and CD14 (Supplemental material Figure 3).
In a previous study [Kim et al., 2003], a new regulatory mechanism in bone
remodeling was demonstrated, in which the transcription factor STAT1 negatively
regulates, in the cytoplasm, another transcription factor, RUNX2, which is essential for
osteoblast differentiation [Lian et al., 2004; Brudereret al., 2014; Vilmalraj et al., 2015].
In turn, Vishal et al (2017) demonstrated that miRNA 590-5p indirectly protects and
stabilizes RUNX2 by targeting SMAD7 expression. This represented evidence, but the
molecular details that trigger this differentiation and pause the inhibitory effect of
STAT1 remained an open question. Our suspicion was that miRNAs could act at that
stage, as had been suggested previously [Gao et al., 2011; Eguchi et al., 2013;
Papaioannou et al., 2014; Jing et al., 2015].
To test our hypothesis, we successfully used microarray hybridizations to explore
a large set of known mRNAs and miRNAs and to identify those that were differentially
expressed during the early (2 days) or middle/late (7 and 21 days) periods of osteoblastic
differentiation of shed cells. The dendrograms and heat maps depicted in Figures 1 and
2 show that the cell samples from these periods of osteoblast differentiation display
unique expression signatures and show sets of differentially expressed mRNAs or
miRNAs.
This approach was sufficiently robust to demonstrate that undifferentiated shed
cells and cells that were submitted to the osteoblastic differentiation exhibit different
patterns of transcriptional expression. Thus, the first part of the hypothesis was
confirmed.
Next, we used a method of data analysis that allows interactions between these
RNA species to identify miRNA-mRNA targets. Several target predictors are currently
available [Fan and Kurgan, 2015] that are still largely used based on the sequence and
conservation of 3' UTR regions among different species. In our view, the drawback of
this type of approach is that it also yields a large number of false positives.
Thus, we used the GenMir++ algorithm [Rehmsmeier et al., 2004; Huang et al,
2007a,b ], which uses Bayesian inference to process gene expression data, given that
18
This article is protected by copyright. All rights reserved
most target mRNAs are posttranscriptionally regulated by interaction with miRNAs,
leading to decreases in their expression levels.
Thus, we could reconstruct miRNA-mRNA interaction networks and predict
miRNA-mRNA targets based on Bayesian statistics of expression data and on annealing.
Shed cells feature a large set of mRNA targets, which interact with differentially
expressed (up- or down-regulated) miRNAs (Figure 3, Table 1) during osteoblastic
differentiation. The second part of the hypothesis was confirmed by these results.
By analyzing these interaction networks, we observed an inherent feature of this
type of interaction, i.e., each miRNA interacted simultaneously with multiple mRNA
targets. For example, miR-505, which was down-regulated in the early period of
differentiation (2 days), interacted with the following mRNA targets: C6orf176, ESM1,
KRTAP1-3, PTGS2, SLC14A1 and TM4SF1. However, a given mRNA target was regulated
by several miRNAs during the same period of differentiation. For example, TM4SF1
mRNA simultaneously interacted with the following miRNAs: miR-505 (down-regulated),
miR-543, miR-381 and miR-450a-5p (up-regulated). Of note, regardless of their
expression levels (up- or down-regulation), the miRNAs interacted with mRNA targets.
As shed cells differentiate into osteoblasts, they exhibit miRNA interactions
involving mRNA targets associated with responses to stimuli; cellular, immune,
metabolic and developmental processes; biological adhesion and organization of
cellular components. In the late phase of differentiation, in addition to these
interactions, they also exhibit mRNAs related to apoptotic, multicellular organismal,
locomotion and reproductive processes (Table 1).
The increasing expression of miR-450a-5p or miR-28-5p as shed cells
differentiate into osteoblasts, as well as the modulation of their intracellular levels by
transfection with the respective miRNA mimics or antagonist oligonucleotides, as well
as the functionality of the osteoblastic differentiation model system was confirmed by
different experimental approaches as RT-qPCR, western blotting, enzymatic activity,
formation of mineralization nodules and protein immunolocalization (Figures 4-6).
The most significant miRNA-mRNA interactions for this study were validated
using two approaches. First, we calculated the minimum free energy (mfe) of annealing
19
This article is protected by copyright. All rights reserved
between miR-28-5p or miR-450a-5p and the 3´UTR seed region of BMP6, TM4SF1 or
STAT1 mRNA targets. Next, we used the luciferase reporter gene assay (LRGA) to confirm
that these miRNA-mRNA interactions can occur within the cell milieu (Figure 7).
Of note, shed cells that were analyzed during different periods of osteoblast
differentiation carried out intricate posttranscriptional control. These cells were shown
to represent an adequate model system for studying the molecular genetic control of
osteoblast differentiation. Although efforts are being made to better understand the
molecular biology of these cells [Menicanin et al., 2010; Tamaoki et al, 2014], to date,
little is known about their gene expression and/or posttranscriptional control as they
differentiate into osteoblasts.
On closer examination of miRNA-mRNA interactions, we observed that miR-
450a-5p regulates, among other targets, STAT1 mRNA in the early period (2 days) of
differentiation (Figure 3, Table 1). This result was interesting because it has been
observed that STAT1 plays a role as a cytoplasmic attenuator of RUNX2 transcription
factors during osteoblast differentiation [Kim et al., 2003].
This led us to another hypothesis - i.e., miR-450a-5p or miR-28-5p triggers
osteoblast differentiation through posttranscriptional control. To the best of our
knowledge, this has not yet been described.
To test this, we transfected control undifferentiated or osteoblast differentiating
shed cells with the respective miR-450-5p or miR-28-5p mimic or antagonist
oligonucleotides.
Immunolocalization showed that transfection with miR-450a-5p mimic triggered
osteoblastic differentiation by stimulating the synthesis of two important protein
markers, alkaline phosphatase (ALPL) and osteocalcin (OC/BGLAP). This result supports
that miR-450a-5p plays a role in triggering osteoblast differentiation, thus confirming
the above hypothesis. Transfection with miR-450a-5p antagonist, however, did not
influence the expression of these proteins in either control or osteoblast differentiating
shed cells. Immunolocalization of ALPL or BGLAP in response to transfection with miR-
28-5p mimic or antagonist oligonucleotides showed that this miRNA also plays a role in
osteoblast differentiation but is, however, biologically less effective at inducing the
20
This article is protected by copyright. All rights reserved
synthesis of these proteins, although its interaction has been predicted in networks. The
Supplemental material Figure 4 illustrates through a graphical abstract the main finding
of this work.
These results open new perspectives for further studies involving the use of
miRNA mimics, for example, miR-450a-5p, miR-28-5p and other interacting miRNAs,
which can be further assayed as inducers of osteoblast differentiation either in basic
research or in clinical protocols studying bone remodeling.
Acknowledgements
We thank Drs Vanessa Fontana, Adriane F. Evangelista, Maidy R. Ferreira, Paulo
T. Oliveira and Adalberto L. Rosa for help and discussions. Mr Roger R. Fernandes and
Ms Tania M. Silva e Sousa for their technical assistance and Ms Sandra N. Bresciani for
drawing supplemental material Figure 4. This work was funded by Conselho Nacional de
Desenvolvimento Científico e Tecnológico, CNPq, Brasilia, Brazil, No. 552227/2005-6,
No. 573489/2008-4 and No. 306315/2013-0) and Fundação de Amparo à Pesquisa do
Estado de São Paulo, FAPESP, São Paulo, Brazil, No. 07/57658-7).
21
This article is protected by copyright. All rights reserved
References
Bartel DP (2004) MicroRNAs: genomics, biogenesis, mechanism, and function. Cell 116:
281-297.
Bonewald LF, Johnson ML (2008) Osteocytes, mechanosensing and Wnt signaling. Bone
42: 606-615.
Bruderer M, Richards RG, Alini M, Stoddart MJ (2014) Role and regulation of RUNX2 in
osteogenesis. Eur Cell Mater 28: 269-286.
Dalle Carbonare L, Innamorati G, Valenti MT (2012) Transcription factor Runx2 and its
application to bone tissue engineering. Stem Cell Rev 8: 891-897.
Donate PB, Fornari TA, Macedo C, Cunha TM, Nascimento DC, Sakamoto-Hojo ET,
Donadi EA, Cunha FQ, Passos GA (2013) T cell post-transcriptional miRNA-mRNA
interaction networks identify targets associated with susceptibility/resistance to
collagen-induced arthritis. PLoS One 8: e54803.
Ducy P, Starbuck M, Priemel M, Shen J, Pinero G, Geoffroy V, Amling M, Karsenty G
(1999) A Cbfa1-dependent genetic pathway controls bone formation beyond embryonic
development. Genes Dev 13: 1025-1036.
Eguchi T, Watanabe K, Hara ES, Ono M, Kuboki T, Calderwood SK (2013) OstemiR: a novel
panel of microRNA biomarkers in osteoblastic and osteocytic differentiation from
mesencymal stem cells. PLoS One 8: e58796.
Eisen MB, Spellman PT, Brown PO, Botstein D (1998) Cluster analysis and display of
genome-wide expression patterns. Proc Natl Acad Sci USA 95: 14863-14868.
Fan X, Kurgan L (2015) Comprehensive overview and assessment of computational
prediction of microRNA targets in animals. Brief Bioinform 16: 780-794.
Fujie A, Funayama A, Miyauchi Y, Sato Y, Kobayashi T, Kanagawa H, Katsuyama E, Hao
W, Tando T, Watanabe R, Morita M, Miyamoto K, Kanaji A, Morioka H, Matsumoto M,
Toyama Y, Miyamoto T (2015) BCL6 promotes osteoblastogenesis through STAT1
inhibition. Biochem Biophys Res Commun 457: 451-456.
Fanganiello RD, Ishiy FA, Kobayashi GS, Alvizi L, Sunaga DY, Passos-Bueno MR (2015)
Increased In Vitro Osteopotential in SHED Associated with Higher IGF2 Expression When
Compared with hASCs. Stem Cell Rev 11: 635-644.
Fornari TA, Donate PB, Assis AF, Macedo C, Sakamoto-Hojo ET, Donadi EA, Passos GA
(2015) Comprehensive Survey of miRNA-mRNA Interactions Reveals That Ccr7 and
Cd247 (CD3 zeta) are Posttranscriptionally Controlled in Pancreas Infiltrating T
Lymphocytes of Non-Obese Diabetic (NOD) Mice. PLoS One 10: e0142688.
Gao J, Yang T, Han J, Yan K, Qiu X, Zhou Y, Fan Q, Ma B (2011) MicroRNA expression
during osteogenic differentiation of human multipotent mesenchymal stromal cells
from bone marrow. J Cell Biochem 112: 1844-1856.
Greenblatt MB, Shim JH (2013) Osteoimmunology: a brief introduction. Immune Netw
13: 111-115. doi: 10.4110/in.2013.13.4.111 PMID: 24009537
Gómez-Picos P, Eames BF (2015) On the evolutionary relationship between
chondrocytes and osteoblasts. Front Genet 6: 297.
22
This article is protected by copyright. All rights reserved
Gregory CA, Gunn WG, Peister A, Prockop DJ (2004) An Alizarin red-based assay of
mineralization by adherent cells in culture: comparison with cetylpyridinium chloride
extraction. Anal Biochem 329: 77-84.
Huang JC, Babak T, Corson TW, Chua G, Khan S, Gallie BL, Hughes TR, Blencowe BJ, Frey
BJ, Morris QD. (2007a) Using expression profiling data to identify human microRNA
targets. Nat Methods 12: 1045-1049.
Huang JC, Morris QD, Frey BJ (2007b) Bayesian inference of MicroRNA targets from
sequence and expression data. J Comput Biol 5: 550-563.
Jing D, Hao J, Shen Y, Tang G, Li ML, Huang SH, Zhao ZH (2015) The role of microRNAs in
bone remodeling. Int J Oral Sci 7: 131-143.
Karsenty G (2008) Transcriptional control of skeletogenesis. Annu Rev Genomics Hum
Genet 9: 1831-1896. doi: 10.1146/annurev.genom.9.081307.164437 PMID: 18767962
Komori T, Yagi H, Nomura S, Yamaguchi A, Sasaki K, Deguchi K, Shimizu Y, Bronson RT,
Gao YH, Inada M, Sato M, Okamoto R, Kitamura Y, Yoshiki S, Kishimoto T (1997) Targeted
disruption of Cbfa1 results in a complete lack of bone formation owing to maturational
arrest of osteoblasts. Cell 89: 755-764.
Kim S, Koga T, Isobe M, Kern BE, Yokochi T, Chin YE, Karsenty G, Taniguchi T, Takayanagi
H (2003) Stat1 functions as a cytoplasmic attenuator of Runx2 in the transcriptional
program of osteoblast differentiation. Genes Dev 17: 1979-1991.
Komori T (2011) Signaling networks in RUNX2-dependent bone development. J Cell
Biochem 112: 750-755.
Krüger J, Rehmsmeier M (2006) RNAhybrid: microRNA target prediction easy, fast and
flexible. Nucleic Acids Res 34(Web Server issue): W451-4.
Lewis BP, Burge CB, Bartel DP (2005) Conserved seed pairing, often flanked by
adenosines, indicates that thousands of human genes are microRNA targets. Cell 120:
15-20.
Li J, He X, Wei W, Zhow X (2015) MicroRNA-194 promotes osteoblast differentiation via
downregulating STAT1. Biochem Biophys Res Commun 460: 482-488.
Lian JB, Javed A, Zaidi SK, Lengner C, Montecino M, van Wijnen AJ, Stein JL, Stein GS
(2004) Regulatory controls for osteoblast growth and differentiation: role of
Runx/Cbfa/AML factors. Crit Rev Eukaryot Gene Expr 14: 1-41.
Lian JB, Stein GS, van Wijnen AJ, Stein JL, Hassan MQ, Gaur T, Zhang Y (2012) MicroRNA
control of bone formation and homeostasis. Nat Rev Endocrinol 8: 212-227,
Menicanin D, Bartold PM, Zannettino AC, Gronthos S (2010) Identification of a common
gene expression signature associated with immature clonal mesenchymal cell
populations derived from bone marrow and dental tissues. Stem Cells Dev 19: 1501-
1510.
Miura M, Gronthos S, Zhao M, Lu B, Fisher LW, Robey PG, Shi S (2003) SHED: stem cells
from human exfoliated deciduous teeth. Proc Natl Acad Sci USA 100: 5807-5812.
Mosmann T (1983) Rapid colorimetric assay for cellular growth and survival: application
to proliferation and cytotoxicity assays. J Immunol Methods 65: 55-63. PMID: 6606682
Papaioannou G, Mirzamohammadi F, Kobayashi T (2014) MicroRNAs involved in bone
formation. Cell Mol Life Sci 24: 4747-4761.
23
This article is protected by copyright. All rights reserved
Pasquinelli AE (2012) MicroRNAs and their targets: recognition, regulation and an
emerging reciprocal relationship. Nat Rev Genet 13: 271-282.
Prideaux M, Findlay DM, Atkins GJ (2016) Osteocytes: The master cells in bone
remodelling. Curr Opin Pharmacol 28: 24-30.
Regan J, Long F (2013) Notch signaling and bone remodeling. Curr Osteoporos Rep 11:
126-129. doi: 10.1007/s11914-013-0145-4 PMID: 23519781
Santibanez JF, Koic J (2012) Transforming growth factor-β superfamily, implications in
development and differentiation of stem cells. Biomol Concepts 3: 429-445.
Rehmsmeier M, Steffen P, Hochsmann M, Giegerich R. (2004) Fast and effective
prediction of microRNA/target duplexes. RNA. 10: 1507-17.
Rickard DJ, Hofbauer LC, Bonde SK, Gori F, Spelsberg TC, Riggs BL. (1998) Bone
morphogenetic protein-6 production in human osteoblastic cell lines. Selective
regulation by estrogen. J Clin Invest 101: 413-422.
Sobacchi C, Schulz A, Coxon FP, Villa A, Helfrich MH (2013) Osteopetrosis: genetics,
treatment and new insights into osteoclast function. Nat Rev Endocrinol 9: 522-536.
Tamaoki N, Takahashi K, Aoki H, Iida K, Kawaguchi T, Hatakeyama D, Inden M, Chosa N,
Ishisaki A, Kunisada T, Shibata T, Goshima N, Yamanaka S, Tezuka K (2014) The
homeobox gene DLX4 promotes generation of human induced pluripotent stem cells.
Sci Rep 4: 7283 .
Tompkins KA (2016) The osteoimmunology of alveolar bone loss. Connect Tissue Res 57:
69-90.
Takayanagi H (2005) Osteoimmunological insight into bone damage in rheumatoid
arthritis. Mod Rheumatol 15: 225-231.
Takayanagi H, Sato K, Takaoka A, Taniguchi T (2005) Interplay between interferon and
other cytokine systems in bone metabolism. Immunol Rev 208: 181-193.
Takayanagi H (2009) Osteoimmunology and the effects of the immune system on bone.
Nat Rev Rheumatol 5: 667-676.
Tajima K, Takahashi H, Takito J, Tohmonda T, Yoda M, Ota N, Kosaki N, Matsumoto M,
Ikegami H, Nakamura T, Kimura T, Okada Y, Horiuchi K, Chiba K, Toyama Y (2010)
Inhibition of STAT1 accelerates bone fracture healing. J Orthop Res 28: 937-941.
Vilmalraj S, Arumugam B, Miranda PJ, Selvamurugan N (2015) Runx2: Structure,
function, and phosphorylation in osteoblast differentiation. Int J Biol Macromol 78: 202-
208.
Vishal M, Vilmalraj S, Ajeetha R, Gokulnath M, Keerthana R, He Z, Partridge NC,
Selvamurugan N (2017) MicroRNA 590-5p stabilizes Runx2 by targeting Smad7 during
osteoblast differentiation. J Cell Physiol 232: 371-380.
Xu J, Li Z, Hou Y, Fang W (2015) Potential mechanisms underlying the Runx2 induced
osteogenesis of bone marrow mesenchymal stem cells. Am J Transl Res 7: 2527-2535.
Walsh MC, Choi Y (2014) Biology of the RANKL-RANK-OPG system in immunity, bone,
and beyond. Front Immunol 5: 511.
24
This article is protected by copyright. All rights reserved
Wang Y, Li YP, Paulson C, Shao JZ, Zhang X, Wu M, Chen W (2014) Wnt and the Wnt
signaling pathway in bone development and disease. Front Biosci (Landmark Ed) 19:
379-407.
Weivoda MM, Ruan M, Hachfeld CM, Pederson L, Howe A, Davey RA, Zajac JD, Kobayashi
Y, Williams BO, Westendorf JJ, Khosla S, Oursler MJ (2016) Wnt Signaling Inhibits
Osteoclast Differentiation by Activating Canonical and Noncanonical cAMP/PKA
Pathways. J Bone Miner Res 31: 65-75.
Wysokinski D, Pawlowska E, Blasiak J (2015) RUNX2: A Master Bone Growth Regulator
That May Be Involved in the DNA Damage Response. DNA Cell Biol 34: 305-315.
Xu J, Li Z, HouY, Fang W (2015) Potential mechanisms underlying the Runx2 induced
osteogenesis of bone marrow mesenchymal stem cells. Am J Transl Res 7: 2527-2535.
Zhang X, Yang M, Lin L, Chen P, Ma KT, Zhou CY, Ao YF (2006) Runx2 overexpression
enhances osteoblastic differentiation and mineralization in adipose-derived stem cells
in vitro and in vivo. Calcif Tissue Int 79: 169-178.
Zhang J, Tu Q, Bonewald LF, He X, Stein G, Lian J, Chen J (2011) Effects of miR-335-5p in
modulating osteogenic differentiation by specifically downregulating Wnt antagonist
DKK1. J Bone Min Res 26: 1953-1963.
25
This article is protected by copyright. All rights reserved
Captions
Figure 1. Transcriptome (mRNAs) expression profiles of SHED cells subjected (or not)
to osteogenic induction. SHED cells were cultured in control medium or in induction
medium during 2 to 21 days. The total RNA samples from these cells were then analyzed
through mRNA microarray hybridizations, which allowed identify 8,376 (a = 2 days),
7,246 (b = 7 days) and 13,994 (c = 21 days) differentially expressed mRNAs. The
dendrograms and heat-maps were obtained using the cluster and tree view algorithm
considering 2.0 fold-change and 0.05 false discovery rate. Heat-map legend: red = up-
regulation, green = down-regulation, black = unmodulation (Pearson correlation
metrics).
Figure 2. Mirnome (miRNAs) expression profiles of SHED cells subjected (or not) to
osteogenic induction. SHED cells were cultured in control medium or in induction
medium during 2 to 21 days. The total RNA samples from these cells were then analyzed
through miRNA microarray hybridizations, which allowed identify 32 (a = 2 days), 17 (b
= 7 days) and 21 (c = 21 days) differentially expressed miRNAs. The dendrograms and
heat-maps were obtained using the cluster and tree view algorithm considering 2.0 fold-
change and 0.05 false discovery rate. Heat-map legend: red = up-regulation, green =
down-regulation, black = unmodulation (Pearson correlation metrics).
Figure 3. Posttranscriptional miRNA-mRNA interactions observed during osteoblastic
differentiation of SHED cells. a) 2 days, b) 7 days and c) 21 days of differentiation. Data
of the differentially expressed (up- or down-regulated) mRNAs and miRNAs were
reanalyzed using the GenMir++ and Cytoscape algorithms to establish interactions and
design networks, respectively.
Figure 4. Relative expression of miR-450a-5p and miR-28-5p during osteoblastic
differentiation of SHED cells and its alteration by miRNA mimic or miRNA antagonist
transfection. Expression of miR-450a-5p (a) or miR-28-5p (b) was measured by RT-qPCR
at day 0 (control), 2, 7 and 21 during osteogenic induction. SHED cells were subjected
(or not) to osteogenic differentiation during 7 days or transfected (or not) with miR-
450a-5p mimic (causing an increase of about 82% in the levels of this miRNA) or with
miR-450a-5p antagonist (causing a reduction of about 89% in the levels of this miRNA)
(c). SHED cells were subjected (or not) to osteogenic differentiation during 7 days or
transfected (or not) with miR-28-5p mimic (causing an increase of about 90% in the
levels of this miRNA) or with miR-28-5p antagonist (causing a reduction of about 93% in
the levels of this miRNA). Data are presented as the mean and standard error of mean
(SEM) from three independent determinations (n = 3). Difference between groups was
analyzed by ANOVA-Tukey´s range statistical test, comparing data from control vs
differentiating cells, transfected vs untransfected cells (* p < 0.05, ** p < 0.01, *** p <
0.001).
26
This article is protected by copyright. All rights reserved
Figure 5. Functionality of the in vitro osteoblastic differentiation model system,
including the effect of miR-450a-5p. a) Viability of SHED cells subjected (or not) to
osteoblastic induction. SHED cells were cultured in control medium (CMD) or in
induction medium (IMD) during 7 days in the presence (or not) of miRNA mimic or
miRNA antagonist. Cell viability was determined MTT assay. b) Relative expression of
RUNX2, STAT1, ALPL and BGLAP mRNAs as evaluated by RT-qPCR. SHED cells were
subjected (or not) to osteogenic induction (7 days) or transfected (or not) with miR-
450a-5p mimic or with miR-450a-5p antagonist. c) Alkaline phosphatase (ALPL) activity
in SHED cells subjected (or not) to osteoblastic induction. SHED cells were cultured in
control medium (CDM) or in induction medium (IDM) during 14 days and transfected (or
not) with miRNA mimic or miRNA antagonist oligonucleotides. d) Calcified matrix
production by SHED cells subjected (or not) to osteogenic induction. SHED cells were
cultured in control medium (CDM) or in induction medium (IMD) during 21 days and
transfected either with miR-450a-5p mimic or with miR-450a-5p antagonist.
Mineralization was determined by alizarin red-based assay and nodules images were
taken by light microscopy. e) Expression of alkaline phosphatase (ALPL) protein in SHED
cells subjected (or not) to osteoblastic induction as analyzed by western blotting. SHED
cells were cultured in control medium (CMD) or in induction medium (IMD) during 14
days and transfected (or not) with miR-450a-5p mimic. Western-blot for ALPL or for
GAPDH (used to normalize data). Bar graphs resulted from quantification of western-
blot bands show the effect of miR-450a-5p mimic transfection on ALPL expression. The
data are presented as the means and standard error of mean (SEM) from three
independent determinations. Difference between groups was analyzed by one-way
ANOVA-Tukey´s range statistical test, comparing control vs induced cells, transfected
cells vs untransfected cells (* p< 0.05, ** p<0.01, *** p < 0.001). f) Immunolocalization
of alkaline phosphatase (ALPL) or osteocalcin/BGLAP (OC) protein in SHED cells
subjected (or not) to osteogenic differentiation and undifferentiated SHED cells
transfected (or not) with miR-450a-5p mimic or miR-450a-5p antagonist. ALPL (green),
OC/BGLAP (red), nucleus (blue). Fluorescence microscopy, 40 x magnification.
Figure 6. Functionality of the in vitro osteoblastic differentiation model system,
including the effect of miR-28-5p. a) Viability of SHED cells subjected (or not) to
osteoblastic induction. SHED cells were cultured in control medium or in induction
medium during 7 days and transfected (or not) with miR-28-5p mimic or miR-28-5p
antagonist. Cell viability was determined MTT assay. b) Relative expression of RUNX2,
STAT1, ALPL or BGLAP mRNAs as evaluated by RT-qPCR. SHED cells were subjected (or
not) to osteogenic induction (7 days) or transfected (or not) with miR-28-5p mimic or
miR-28-5p antagonist. c) Alkaline phosphatase (ALPL) activity in SHED cells subjected (or
not) to osteoblastic induction. SHED cells were cultured in control medium or in
induction medium during 14 days and transfected (or not) with miR-28-5p mimic or miR-
28-5p antagonist. d) Calcified matrix production by SHED cells subjected (or not) to
osteogenic induction. SHED cells were cultured in control medium or in induction
medium during 21 days and transfected either with miR-28-5p mimic or miR-28-5p
antagonist. Mineralization was determined by alizarin red-based assay and nodules
images were taken by light microscopy. e) Expression of alkaline phosphatase (ALPL)
protein in SHED cells subjected (or not) to osteoblastic induction as analyzed by western
blotting. SHED cells were cultured in control medium or in induction medium during 14
27
This article is protected by copyright. All rights reserved
days and transfected (or not) with miR-28-5p mimic. Western-blot for ALPL or for GAPDH
(used to normalize data). Bar graphs resulted from quantification of western-blot bands
show the effect of miR-28-5p mimic transfection on ALPL expression. The data are
presented as the mean and standard error of mean (SEM) from three independent
determinations. Difference between groups was analyzed by one-way ANOVA-Tukey´s
range statistical test, comparing control vs induced cells, transfected cells vs
untransfected cells (* p< 0.05, ** p<0.01, *** p < 0.001). f) Immunolocalization of
alkaline phosphatase (ALPL) or osteocalcin/BGLAP (OC) protein in SHED cells subjected
(or not) to osteogenic differentiation and undifferentiated SHED cells transfected (or
not) with miR-28-5p mimic or miR-28-5p antagonist. ALPL (green), OC/BGLAP (red),
nucleus (blue). Fluorescence microscopy, 40 x magnification.
Figure 7. Hybridization likelihoods between the wild-type and mutant 3´UTRs and
luciferase reporter gene assay (LRGA). a) The miRNA-mRNA hybrid structures and their
respective minimum free energy (mfe) were calculated using the RNA-Hybrid algorithm
for interaction between STAT1 (a) or TM4SF1 (b) mRNA with miR-450a-5p or TM4SF1 (c)
or BMP6 (d) mRNA with miR-28-5p. Posttranscriptional interactions between miR-450a-
5p with STAT1 3´UTR were assayed by LRGA (e). pMIR-STAT1 (wild type 3´UTR) or pMIR-
STAT1(m) (mutant 3´UTR) luciferase vector constructs were transfected into human
HEK-293T cells to demonstrate 3´UTR sequence specificity. Posttranscriptional
interactions between miR-450a-5p with TM4SF1 3´UTR were assayed by LRGA (f). pMIR-
TM4SF1 (wild type 3´UTR) or pMIR-TM4SF1(m) (mutant 3´UTR) luciferase plasmid
constructs were transfected into human HEK-293T cells to show specificity of the 3´UTR
for the interactions. Posttranscriptional interactions between miR-28-5p with TM4SF1
3´UTR were assayed by LRGA (g). pMIR-TM4SF1 (wild type 3´UTR) or pMIR-TM4SF1(m)
(mutant 3´UTR) luciferase vector constructs were transfected into human HEK-293T
cells. Posttranscriptional interactions between miR-28-5p with BMP6 3´UTR were
assayed by LRGA (h). pMIR-BMP6 (wild-type 3´UTR) or pMIR-BMP6(m) (mutant 3´UTR)
luciferase vector constructs were transfected into human HEK-293T cells. pMIR (wild
type 3´UTR) or pMIR (mutant 3´UTR, m) were transfected into HEK-293T cells in order
to demonstrate the possibility of occurrence of the miRNA-mRNA interactions within
the cell milieu and the specificity of 3´UTR sequence. The data are presented as the
means and standard error of mean (SEM) from three independent experiments. The
difference between groups was analyzed by one-way ANOVA-Tukey´s range statistical
test (* p< 0.05, ** p<0.01, *** p < 0.001).
Supplemental material
Supplemental material Figure 1. Flow diagram of the study showing the methods used.
In this flow diagram are indicated the methods employed for the study of the gene
expression profiling during the osteoblastic differentiation of shed cells and its
posttranscriptional control involving miRNAs.
28
This article is protected by copyright. All rights reserved
Supplemental material Figure 2. Monitoring liposome (Hyperfect)-mediated
transfection of oligo RNA into shed cells. Shed cells were transfected with 40 nM
irrelevant Cy3-labeled oligo RNA and observed by confocal fluorescence microscopy.
Control untransfected SHED cells (a), Cy3-labeled oligo RNA transfected SHED cells.
DAPI-stained nuclei (blue), oligo RNA molecules (red). 40x magnification.
Supplemental material Figure 3. Expression of mesenchymal cell surface markers.
Undifferentiated shed cells were phenotyped for mesenchymal surface markers CD90,
CD73, CD29, CD105, CD13, CD44, CD34 and CD14 by fluorescence-activated cell sorting
(FACS).
Supplemental material Figure 4. Posttranscriptional control of miRNAs in pausing
STAT1 negative effect over RUNX2 during osteoblastic differentiation of shed cells.
Transcriptome analysis showed that shed cells feature differentially expressed mRNAs
and miRNAs at large scale during its in vitro differentiation into osteoblasts. Of note, two
mRNAs showed opposite modulation during differentiation; STAT1 mRNA was
expressed in high levels in undifferentiated or early differentiating cells and was
gradually down-regulated. Contrariwise, RUNX2 mRNA was up-regulated toward
differentiation. This observation has become of interest as STAT1 negatively regulates
RUNX2, an essential transcription factor in the differentiation process. The
reconstruction of miRNA-mRNA interaction networks allows us to identify that STAT1
mRNA is controlled by miR-450a-5p and miR-28-5p. This finding contributes for a better
understanding of the factors that pause the negative effect of STAT1 in triggering
osteoblastic differentiation.
29
This article is protected by copyright. All rights reserved
Figure 1
30
This article is protected by copyright. All rights reserved
Figure 2
31
This article is protected by copyright. All rights reserved
Figure 3
32
This article is protected by copyright. All rights reserved
Figure 4
33
This article is protected by copyright. All rights reserved
Figure 5
34
This article is protected by copyright. All rights reserved
Figure 6
35
This article is protected by copyright. All rights reserved
Figure 7
36
This article is protected by copyright. All rights reserved
Time point
miRNAs
mRNA targets
GO biological processes
2 days (98
interactions)
miR-140-5p (↓)
PTGS2
Metabolic process (GO:0008152)
miR-505 (↓)
C6orf176, ESM1, KRTAP1-3, PTGS2,
SLC14A1, TM4SF1
Metabolic process (GO:0008152)
miR-543 (↑)
C6orf176, ESM1, IL13RA2, KRTAP1-3,
PTGS2, SLC14A1, TM4SF1
Cellular process (GO:0009987),
Developmental process (GO:0032502),
Immune system process (GO:0002376),
Metabolic process (GO:0008152),
Response to stimulus (GO:0050896)
miR-381 (↑)
C6orf176, ESM1, IL13RA2, KRTAP1-3,
PTGS2, SLC14A1, TM4SF1
Cellular process (GO:0009987),
Developmental process (GO:0032502),
Immune system process (GO:0002376),
Metabolic process (GO:0008152),
Response to stimulus (GO:0050896)
miR-450a-5p (↑)
ACVR1C, ADAMTS6, ANKRD20A2, ARL14,
BCL2A1, BDKRB1, C20orf173, C6orf176,
CA9, CASC5, CSF2, CXCL12, DEFB103A,
DPF3, ECSCR, ESM1, F3, FAM46C,
FLJ30307, FLJ43390, FNDC5, FOXI1, G0S2,
GRPR, HNF1A, IL11, IL12A, IL13RA2, IL1B,
KCNS1, KRT34, KRTAP1-1, KRTAP1-3,
KRTAP1-5, LDB2, LIF, LOC283392,
LOC349160, LRRC15, MMP12, MMP3,
MSTN, MYOCD, STAT1, NAALADL2, NEFM,
NPTX1, NPTX2, NTNG1, PITX1, PTGS2,
PTPRR, RELN, RGMB, RSPO3, RTP3,
SLC14A1, SLC28A3, SUCNR1, TFPI2,
TM4SF1, TNFRSF6B
Apoptotic process (GO:0006915),
Biological adhesion (GO:0022610),
Biological regulation (GO:0065007),
Cellular component organization or
biogenesis (GO:0071840), Cellular
process (GO:0009987), Developmental
process (GO:0032502), Immune system
process (GO:0002376), Localization
(GO:0051179), Metabolic process
(GO:0008152), Multicellular organismal
process (GO:0032501), Response to
stimulus (GO:0050896)
Table 1. Selected miRNAs that were modulated during osteogenic differentiation of SHED cells, their mRNA targets, as predicted by the
GenMir++ algorithm and their respective gene ontology (GO) processes. (↑) up-regulated, (↓) down-regulated.
37
This article is protected by copyright. All rights reserved
7 days (181
interactions)
miR-29c-3p (↑)
ADRA1D, BDKRB1, C6orf176, C6orf176,
CDCP1, DPF3, ECSCR, FAM46C,
FLJ30307, FLJ43390, FST, GPRC5A,
KRTAP1-3, LOC729251, LPPR4, MATN2,
MMP1, MSTN, NAALADL2, PTGS2,
SLC14A1, TFPI2, TM4SF1
Apoptotic process (GO:0006915),
Biological adhesion (GO:0022610),
Biological regulation (GO:0065007),
Cellular process (GO:0009987),
Developmental process (GO:0032502),
Immune system process (GO:0002376),
Localization (GO:0051179), Metabolic
process (GO:0008152), Multicellular
organismal process (GO:0032501),
Response to stimulus (GO:0050896)
miR-654-3p (↑)
BMP6, CXCL6, DACH2, FKBP5, NRCAM,
OMD, PDGFD, PPP2R2C, SAA1, SAA2,
SEMA4D, SHISA9, TIMP4
Apoptotic process (GO:0006915),
Biological adhesion (GO:0022610),
Biological regulation (GO:0065007)
Cellular process (GO:0009987),
Developmental process (GO:0032502),
Immune system process (GO:0002376),
Localization (GO:0051179), Locomotion
(GO:0040011), Metabolic process
(GO:0008152), Multicellular organismal
process (GO:0032501), Response to
stimulus (GO:0050896)
miR-369-5p (↓)
ADRA1D, BDKRB1, C6orf176, DPF3,
ECSCR, FAM46C, FLJ30307, FLJ43390,
FST, GPRC5A, KRTAP1-3, LPPR4, MMP1,
MSTN, NAALADL2, PTGS2, SLC14A1,
TFPI2, TM4SF1
Apoptotic process (GO:0006915)
Biological regulation (GO:0065007)
Cellular process (GO:0009987),
Developmental process (GO:0032502),
Immune system process (GO:0002376),
Localization (GO:0051179), Metabolic
process (GO:0008152), Multicellular
38
This article is protected by copyright. All rights reserved
Organismal process (GO:0032501),
Response to stimulus (GO:0050896)
miR-17-3p (↑)
ADRA1D, BDKRB1, C6orf176, CDCP1,
DPF3, ECSCR, FAM46C, FLJ30307,
FLJ43390, FST, GPRC5A, KRTAP1-3,
LOC729251, LPPR4, MMP1, MSTN,
NAALADL2, PTGS2, SLC14A1, TFPI2
TM4SF1
Apoptotic process (GO:0006915),
Biological regulation (GO:0065007),
Cellular process (GO:0009987),
Developmental process (GO:0032502),
Immune system process (GO:0002376),
Localization (GO:0051179), Metabolic
process (GO:0008152), Multicellular
organismal process (GO:0032501),
Response to stimulus (GO:0050896)
miR-145-3p (↑)
BMP6, CADM3, CXCL6, DACH2, FKBP5,
NRCAM, OMD, PDGFD, PPP2R2C, SAA1,
SAA2, SHISA9, TIMP4
Apoptotic process (GO:0006915),
Biological adhesion (GO:0022610),
Biological regulation (GO:0065007),
Cellular process (GO:0009987),
Developmental process (GO:0032502),
Immune system process (GO:0002376),
Localization (GO:0051179), Locomotion
(GO:0040011), Metabolic process
(GO:0008152), Multicellular organismal
process (GO:0032501), Response to
stimulus (GO:0050896)
miR-32 (↓)
PTGS2
Metabolic process (GO:0008152)
miR-376a-3p (↓)
BMP6, FKBP5, OMD, PPP2R2C, TIMP4
Apoptotic process (GO:0006915),
Biological adhesion (GO:0022610),
Biological regulation (GO:0065007),
Cellular process (GO:0009987),
Developmental process (GO:0032502),
Immune system process (GO:0002376),
Metabolic process (GO:0008152),
Multicellular organismal process,
39
This article is protected by copyright. All rights reserved
(GO:0032501), Response to stimulus
(GO:0050896)
miR-28-5p (↑)
BATF, BMP6, CADM3, CHST2, CXCL6,
DACH2, FAM153B, FKBP5, MGP, NRCAM,
NRK, OLAH, OLFML2A, OMD, PDGFD,
PPP2R2C, SAA1, SAA2, SEMA4D, SHISA9,
TIMP4
Apoptotic process (GO:0006915),
Biological adhesion (GO:0022610),
Biological regulation (GO:0065007),
Cellular component organization or
biogenesis (GO:0071840), Cellular
process (GO:0009987), Developmental
process (GO:0032502), Immune system
process (GO:0002376), Localization
(GO:0051179), Locomotion
(GO:0040011), Metabolic process
(GO:0008152), Multicellular organismal
process (GO:0032501), Response to
stimulus (GO:0050896)
miR-7 (↓)
CHST2, CXCL6, DACH2, FAM153B, FKBP5,
NRCAM, OLAH, OLFML2A, OMD, PDGFD,
PPP2R2C, SAA1, SAA2, SAA2, SHISA9,
TIMP4, BATF, BMP6, CADM3
Apoptotic process (GO:0006915),
Biological adhesion (GO:0022610),
Biological regulation (GO:0065007),
Cellular component organization or
biogenesis (GO:0071840), Cellular
process (GO:0009987), Developmental
process (GO:0032502), Immune system
process (GO:0002376), Localization
(GO:0051179), Locomotion
(GO:0040011), Metabolic process
(GO:0008152), Multicellular organismal
process (GO:0032501), Response to
stimulus (GO:0050896)
miR-143- 3p (↓)
ADRA1D, C6orf176, C6orf176, DPF3,
ECSCR, KRTAP1-3, LPPR4, NAALADL2,
PTGS2, SLC14A1, TM4SF1
Apoptotic process (GO:0006915),
Biological regulation (GO:0065007),
Cellular process (GO:0009987),
Developmental process (GO:0032502),
40
This article is protected by copyright. All rights reserved
Immune system process (GO:0002376),
Localization (GO:0051179), Metabolic
process (GO:0008152), Multicellular
organismal process (GO:0032501),
Response to stimulus (GO:0050896)
miR-193a-5p (↓)
BMP6, CXCL6, DACH2, FKBP5, NRCAM,
OMD, PDGFD, PPP2R2C, SAA1, SAA2,
SHISA9, TIMP4
Apoptotic process (GO:0006915),
Biological adhesion (GO:0022610),
Biological regulation (GO:0065007),
Cellular process (GO:0009987),
Developmental process (GO:0032502),
Immune system process (GO:0002376),
Localization (GO:0051179), Locomotion
(GO:0040011), Metabolic process
(GO:0008152), Multicellular organismal
process (GO:0032501), Response to
stimulus (GO:0050896)
miR-193b (↓)
OMD, PDGFD, PPP2R2C, TIMP4
Biological adhesion (GO:0022610),
Biological regulation (GO:0065007),
Cellular process (GO:0009987),
Developmental process (GO:0032502),
Immune system process (GO:0002376),
Localization (GO:0051179), Locomotion
(GO:0040011), Metabolic process
(GO:0008152), Multicellular organismal
process (GO:0032501), Response to
stimulus (GO:0050896)
miR-548c-3p (↓)
BMP6, FKBP5, NRCAM, OMD, PPP2R2C,
SAA1, SHISA9, TIMP4
Apoptotic process (GO:0006915),
Biological adhesion (GO:0022610),
Biological regulation (GO:0065007),
Cellular process (GO:0009987),
Developmental process (GO:0032502),
Immune system process (GO:0002376),
41
This article is protected by copyright. All rights reserved
Localization (GO:0051179), Locomotion
(GO:0040011), Metabolic process
(GO:0008152), Multicellular organismal
process (GO:0032501), Response to
stimulus (GO:0050896)
miR-181d (↓)
PPP2R2C
Immune system process (GO:0002376),
Metabolic process (GO:0008152),
Response to stimulus (GO:0050896)
miR-377-3p (↓)
ADRA1D, C6orf176, DPF3, ECSCR,
KRTAP1-3, LPPR4, NAALADL2, PTGS2,
SLC14A1, TM4SF1
Apoptotic process (GO:0006915),
Biological regulation (GO:0065007),
Cellular process (GO:0009987),
Developmental process (GO:0032502),
Immune system process (GO:0002376),
Localization (GO:0051179), Metabolic
process (GO:0008152), Multicellular
organismal process (GO:0032501),
Response to stimulus (GO:0050896)
21 days (276
interactions)
miR-145-3p (↑)
CUL4B
Apoptotic process (GO:0006915),
Cellular process (GO:0009987),
Developmental process (GO:0032502),
Metabolic process (GO:0008152)
miR-28-5p (↑)
CUL4B, TM4SF1, LOC100130935
Apoptotic process (GO:0006915),
Cellular process (GO:0009987),
Developmental process (GO:0032502),
Metabolic process (GO:0008152)
miR-29b-3p (↓)
CUL4B, LOC100130935, HBB, MKI67,
ASB1, ATP8A2, BEND7, CDCA5, CEP110,
F3, H1F0, HBBP1, HCP5, LCORL,
LOC729595, RDH5, SMS, SNORA62, TK1,
VCAM1, WNK4, AADACL4, AKR1B10,
Apoptotic process (GO:0006915),
Biological adhesion (GO:0022610),
Biological regulation (GO:0065007),
Cellular component organization or
biogenesis (GO:0071840), Cellular
42
This article is protected by copyright. All rights reserved
AKR1B15, ASAP1, ATP2A3, C2orf72,
C4orf32, C5AR1, C5orf62, C6orf154,
CCND3, CD163, IGLON5, IL20RB,
KIAA0485, KIF6, KLF9, LOC171220,
LOC729378, MORN3, MYCT1, NBEA,
OR6Y1, PPP1R14A, PRG1, PTPRO,
RGS11, RNF32, SCN1A, SEMA4D,
SLC38A5, SLC44A1, SLC5A3, SNAI1,
SOD3, SPATA17, SPRY1, ST18, STON1,
TIMP4, TLR2, TPK1, TRAF3IP2, TRIM50,
UNC5CL, UNQ5815, VMO1, WDFY4,
ZFP36, CDCA7, NCF2, CDCP1, CDKN3,
CENPF, CENPM, CEP55, DIAPH3, ESCO2,
KRT34, RGMB, RTP3, C6orf173, DIAPH3,
OAS1, PCOLCE2, ABL2, ANKRD13A,
APOLD1, ARHGAP11A, ASPHD1, ATP8B1,
AURKAPS1, BYSL, C7orf58, CA13, CA8,
CAMK1, CASC5, CCDC80, CCDC85A,
DCAF8L1, DDIT4L, DRAP1, DSG2, ERCC2,
FAM72A, FAM72D, FANCD2, FRRS1,
GLT25D2, HERC4, HMGN2, HYI, IFIT2,
KCNN4, KCTD20, KIAA1632, KLHL7,
LOC100128960, LOC100131960,
LOC284232, LOC441795, LOC541472,
LOC644992, LOC646576, LOC729251,
LOC730456, LOC730961, MX1, MYOC,
MYPN, NEK10, NKX3-1, NPW, PSTPIP2,
PVR, RBM24, RP11-631M21.2, SLC4A1,
SOCS3, SUV39H1, SYT1, TM4SF4,
TM4SF1, TMEM217, TNC, TNFRSF6B, TTK,
TUBB, TUBB2C, TUBB3, TUBG1, WDR4,
ZNF316, ZNF774, ABCA6, ABCC4, ACOXL,
AKAP12, ARHGEF19, ASPH, C10orf41,
C11orf53, C1R, C7orf41, C8orf79, CANT1,
process (GO:0009987), Developmental
process (GO:0032502), Immune system
process (GO:0002376), Localization
(GO:0051179), Locomotion
(GO:0040011), Metabolic process
(GO:0008152), Multicellular organismal
process (GO:0032501), Reproduction
(GO:0000003), Response to stimulus
(GO:0050896)
43
This article is protected by copyright. All rights reserved
CC2D2A, CCDC152, CDAN1, CDON,
CEP110, CFH, COL3A1, COL4A4, COL4A5,
COL5A2, CPNE7, CTNNA2, CTNNAL1,
CTSD, CXCL1, CXCL14, CYP4V2, DEPDC6,
DFNB59, DIO2, DLC1, DLGAP4, DLL1,
DMGDH, DNAJC1, DOCK4, DOK3,
DUSP22, FAM107A, FAM177B, FAM179A,
FAM59A, FAM92A3, FAT4, FBLN1,
FRMPD4, GALNT7, GLIPR1L2, GLP2R,
GLRB, GTF2A1L, H19, HSP90AA1, HTRA1,
IL1R1, IL24, IL28RA, ITPR1, KCNE4,
KIF16B, KISS1R, LASS6, LEPR, LIMS2,
LIN7A, LOC100288755, LOC100289178,
LOC100289290, LOC390595, LOC391322,
LOC541471, LOC644189, LOC728431,
LRP4, LRRC66, MCC, MEGF10, MMP3,
MPP4, NFXL1, NR4A3, OGDHL, OLAH,
OR5C1, PAG1, PALM, PCDH18, PODN,
PODNL1, POM121L9P, PRKAR2B, PROC,
PTPLB, PTPRG, RASGRP2, REM2,
RNF152, SCARNA1, SERPINB7, SF3B3,
SGCG, SIK1, SIK3, SLC1A5, SSBP3,
THBS2, THRA, TLE4, TMEM159,
TNKS1BP1, TXLNB, UCN2,
ZBTB34, ZCCHC18, ZNF774
44
This article is protected by copyright. All rights reserved
