Regulation of osteogenic differentiation of human adipose-derived stem cells by controlling electromagnetic field conditions.

2 MB·English

by  Kang, Kyung Shin

Experimental&MolecularMedicine(2013)45,e6;doi:10.1038/emm.2013.11
&2013KSBMB. Allrightsreserved2092-6413/13
www.nature.com/emm
ORIGINAL ARTICLE
Regulation of osteogenic differentiation of human
adipose-derived stem cells by controlling
electromagnetic field conditions
Kyung Shin Kang1, Jung Min Hong1, Jo A Kang2, Jong-Won Rhie2, Young Hun Jeong3 and Dong-Woo Cho1,4
Many studies have reported that an electromagnetic field can promote osteogenic differentiation of mesenchymal stem cells.
However, experimental results have differed depending on the experimental and environmental conditions. Optimization of
electromagnetic field conditions in a single, identified system can compensate for these differences. Here we demonstrated that
specific electromagnetic field conditions (that is, frequency and magnetic flux density) significantly regulate osteogenic
differentiation of adipose-derived stem cells (ASCs) in vitro. Before inducing osteogenic differentiation, we determined ASC
stemness and confirmed that the electromagnetic field was uniform at the solenoid coil center. Then, we selected positive
(30/45Hz, 1mT) and negative (7.5Hz, 1mT) osteogenic differentiation conditions by quantifying alkaline phosphate (ALP)
mRNA expression. Osteogenic marker (for example, runt-related transcription factor 2) expression was higher in the 30/45Hz
condition and lower in the 7.5Hz condition as compared with the nonstimulated group. Both positive and negative regulation
of ALP activity and mineralized nodule formation supported these responses. Our data indicate that the effects of the
electromagnetic fields on osteogenic differentiation differ depending on the electromagnetic field conditions. This study
provides a framework for future work on controlling stem cell differentiation.
Experimental& Molecular Medicine (2013) 45, e6;doi:10.1038/emm.2013.11; published online25 January2013
Keywords: adipose-derived stem cells; electromagnetic field; frequency; magnetic flux density; optimization; osteogenic
differentiation
INTRODUCTION used for osteogenesis varied from 7.5 to
Low-frequency,low-energy,electromagneticfieldsarecommonly 75Hz.5–8,11,16,17 Unlike bio-mimetic stimuli (for example,
usedtopromotebonefracturehealing.1,2Theseelectromagnetic cyclic strain and shear stress), this time-varying
fields enhance osteogenesis, decrease osteoporosis3–6 and electromagnetic field does not have logical cues for each
regulate diverse osteoblastic responses for osteogenesis in vitro. stimulation parameter. Thus, studies aimed at identifying
DNAsynthesiscanalsobeenhancedbyelectromagneticfields.7,8 optimal parameters, at least within the ranges of previous
In addition, electromagnetic fields increase osteogenic marker investigations, are required.
gene expression during differentiation9 and affect calcified Previous studies on osteogenesis using electro-
matrix production during mineralization.10 magnetic fields have been performed under various
Two parameters primarilycharacterizeelectromagneticfield conditions,1,2,4–10,12,13,15–17 and the results have differed
properties:frequencyandmagneticfluxdensity.Themagnetic depending on the experimental and environmental
fluxdensityusedinmostpreviousstudiestoinduceosteogen- conditions. These differences may be attributable to different
esisvariedfrom0.1to3mT.6,8,9,11–13Suchstudieshaveshown experimentalsystems.Moreover,authorshaverarelydiscussed
that a low-frequency electromagnetic field is more effective why they selected specific conditions for stimulating cells in
thanastaticfieldincausingbiologicaleffects.14,15Frequencies their studies. Therefore, these conditions should be optimized
1Department of Mechanical Engineering, Pohang University of Science and Technology (POSTECH), Pohang, Korea; 2Department of Plastic Surgery,
CollegeofMedicine,TheCatholicUniversityofKorea,Seoul,Korea;3DepartmentofMechanicalEngineering,KoreaPolytechnicUniversity,Siheung,Korea
and4DivisionofIntegrativeBiosciencesandBiotechnology,PohangUniversityofScienceandTechnology(POSTECH),Pohang,Korea
Correspondence: Professor YH Jeong, Department of Mechanical Engineering, Korea Polytechnic University, 2121 Jeongwang-dong,
Siheung,Gyeonggi-do429-793,Korea.
E-mail:[email protected]
orProfessorD-WCho,DepartmentofMechanicalEngineeringandCenterforRapidPrototypingBased3DTissue/OrganPrinting,PohangUniversityof
ScienceandTechnology(POSTECH),San31,Hyoja-dong,Nam-gu,Pohang790-751,Korea.
E-mail:[email protected]
Received29August2012;revised29October2012;accepted5November2012
Electromagneticfieldforosteogenesis
KSKangetal
2
in a single, identified system where the experimental and Isolation, culture and characterization of ASCs
environmental conditions are fixed. ASCs were isolated and cultured as previously described.25 The
We examined the effects of various electromagnetic field isolated ASCs were cultured in Dulbecco’s modified Eagle’s medium
conditions on the osteogenic differentiation of mesenchymal (Gibco BRL, Grand Island, NY, USA) containing 10% (v/v) fetal
bovine serum (Gibco BRL), 100Uml(cid:2)1 penicillin and 100mgml(cid:2)1
stemcells.Weusedadipose-derivedstemcells(ASCs) because
streptomycin (Gibco BRL) at 371C in a humidified atmosphere
ASCs easily attach to surfaces and proliferate rapidly.18 In
containing 5% CO. All experiments were performed at passage
addition, ASCs can be derived from fat tissue, which afforded 2
three. To induce osteogenic differentiation, the culture medium
us an abundant source of cells. ASCs also have osteogenic
was changed to osteogenic induction medium (OM) containing
potential, among various differentiation lineages,18–22 and are
10nM dexamethasone (Sigma-Aldrich, St Louis, MO, USA), 10mM
known to play a role in bone regeneration, including the b-glycerol phosphate (Sigma-Aldrich) and 50mgml(cid:2)1 ascorbic
acceleration of bone fracture healing23,24 and promotion of acid-2-phosphate(Sigma-Aldrich).
bone regeneration in cases of calvarial defects without For flow cytometry, cells were harvested with trypsin/EDTA and
chemical induction.19 washed in 1% bovine serum albumin (Pierce, Rockford, IL, USA).
Here, we observed the effects of an electromagnetic field Cells were then incubated with fluorescently labeled monoclonal
under various conditions on osteogenic differentiation of antibodies against CD13, CD29, CD34, CD45, CD90 and CD105
(Serotec, Oxford, UK) for 30min at 41C and analyzed using a flow
ASCs. A fixed electromagnetic field distribution was used to
cytometer(Becton Dickinson,MountainView, CA, USA).
provide uniform field effects on cells. Among various combi-
nations of frequency and magnetic flux density, we selected
and investigated specific conditions that had positive (30– Exposure of ASCs to the electromagnetic field
45Hz, 1mT) or negative (7.5Hz, 1mT) effects on ASC A solenoid coil was used because it could generate a uniform
differentiation. electromagnetic field. A custom-made solenoid coil had the wire
wound 7300 times and the DC resistance was 450O for 10mT. The
innerdiameterandlengthwere180and400mm,respectively.Weused
MATERIALS AND METHODS 3-axismagneticfieldsensors(MFS-3A,Ametes,SanCarlos,CA,USA)
Ethics statement tomeasurethemagneticfluxdensityinsidethecoil.The24-wellplates
Thisstudy wasapproved by the Institutional ReviewBoard ofSeoul were used with a cell density of 2(cid:3)104 cells per well. Cells were
StMary’s Hospitalandconformedtothe principlesexpressedinthe exposed to the electromagnetic field for 8h per day in a humidified
Declaration of Helsinki. All patients provided written informed incubatorat371Cundera5%CO atmosphere.Allexperimentswere
2
consent for the collectionofcellsand subsequentanalysis. conductedin theconditions without heatgeneration.
Figure 1 Identification of the electromagnetic field generated by a solenoid coil. (a) The solenoid coil, which generated the
electromagnetic field, was located inside a CO incubator. The inner diameter and length were 180 and 400mm, respectively. (b) The
2
magnetic flux density inside the solenoid coil was calculated using commercial software. Axial and radial directions were visualized in a
solenoid coil cross-section view. (c) The sinusoidal magnetic flux density was measured using a three-axis sensor. (d) The three-
dimensional graph describes the measuredmagneticflux density along the cross-section insidethe solenoid coilwhenitwas set as1mT.
Bothcolorandheight(zaxis)indicatethemagneticfluxdensityateachpoint.
Experimental&MolecularMedicine
Electromagneticfieldforosteogenesis
KSKangetal
3
Real-time PCR wasaddedandincubatedfor1hatroomtemperature.Sampleswere
TotalRNA(1mg)extractedfromculturedcellsatday7usingTRIzol mounted with antifade reagent (Invitrogen) and observed using a
reagent (Invitrogen, Groningen, The Netherlands) was used as a FluoView 1000 confocal microscope (Olympus, Melville, NY, USA).
templateforcomplementaryDNAsynthesis(Invitrogen,Eugene,OR, All images were capturedwithout changingthe camerasettings.
USA). Real-time PCR was performed using the SYBR Green PCR
MasterMixassay(AppliedBiosystems,Warrington,UK)andtheABI
Statistical analysis
StepOnePlussystem(AppliedBiosystems,FosterCity,CA,USA).The
All experiments were performed in triplicate. Data are presented as
amplification reaction consisted of 40 cycles of 951C for 10min, the mean±s.d. Statistical significance was determined by a one-way
951C for 15s and 601C for 1min. The following primers
analysis of variance (n¼3). Differences were considered to be
were used: GAPDH sense, 50-CCAGGTGGTCTCCTCTGACTTC-30;
statisticallysignificant at *(or#) Po0.05.
GAPDH antisense, 50-GTGGTCGTTGAGGGCAATG-30; alkaline
phosphatase (ALP) sense, 50-ATGTCATCATGTTCCTGGGAGAT-30;
RESULTS
ALPantisense, 50-TGGAGCTGACCCTTGAGGAT-30; collagentype-I,
a1 (COL-I) sense, 50- CAAGACAGTGATTGAATACAAAACCA-30; Electromagnetic field identification
COL-I antisense, 50- ACGTCGAAGCCGAATTCCT-30; osteocalcin Before evaluating the effects of the electromagnetic field on
(OC) sense, 50- AAGAGACCCAGGCGCTACCT-30; OC antisense, ASCs, we characterized the electromagnetic field generated by
50- AACTCGTCACAGTCCGGATTG-30; osterix (OSX) sense, 50- GG the solenoid coil (Figure 1a). The magnetic flux density was
CAGCGTGCAGCAAATT-30; OSX antisense, 50- CCTGCTTTGCCC predicted using three-dimensional finite element analysis
AGAGTTGT-30; runt-related transcription factor 2 (RUNX2) sense, before development of the coil. The distribution of the
50-AACCCACGAATGCACTATCCA-30;RUNX2antisense,50-CGGA magnetic flux density in the cross-section of the coil was
CATACCGAGGGACATG-30.Geneexpressionlevelsofallexperimen-
tal groups were expressed as fold changes relative to levels of the
controlgroup.
ALP activity and staining
After culture for 10 days, cells were washed three times with
phosphate-bufferedsalineandthenlysedinlysisbuffer(RIPAbuffer,
Upstate, Temecula, CA, USA). Cell lysates were incubated with
p-nitrophenyl phosphate (Sigma-Aldrich) at 371C for 30min. The
reaction was stopped by adding 2N NaOH. Absorbances were
measured at 405nm using a microplate reader. ALP activity was
normalized to the total protein content of each sample. Protein
contents were measured using a BCA protein assay kit (Pierce). For
ALPstaining,cellswerefixedwith4%paraformaldehyde(WakoPure
Chemical, Osaka, Japan). ALP was then stained using a solution
containing 0.01% Naphthol AS-MX phosphate and 0.05% Fast Red
Violet LB Salt (Sigma-Aldrich). Fixed cells were incubated with
substratesolution atroom temperature for 30min.
Measurement of mineralization
AlizarinRedSwasusedtodetectmineralization.Cellswerefixedby
incubationwith4%paraformaldehydefor30minatday20.Samples
werethenimmersedinAlizarinRedSsolutionatroomtemperature
for 30min. After washing, stained cells were observed under an
opticalmicroscope.
For quantitative calcium measurement, cells were decalcified in
0.6NHCLatday20followedbyshakingfor24hatroomtemperature.
Then, their calcium content was measured using a QuantiChrom
Calcium Assay kit (BioAssay Systems, Hayward, CA, USA), as
previously reported.26 Briefly, 5ml of lysate was mixed with 200ml of
workingsolutionfromthekitandincubatedfor20min.Absorbances
were measured at 612nmusing amicroplate reader.
Immunostaining
After culture with OM for 4 days, cells were fixed with 4% Figure 2 Comparison of the predicted magnetic flux density with
paraformaldehyde for 30min, permeabilized with 0.1% Triton the measured data. The predicted magnetic flux density using
X-100 and blocked with 0.2% bovine serum albumin for 15min. finite element analysis was similar tothe measuredvalues at1mT.
Theywerethenincubatedwithanti-RUNX2(SantaCruz,Heidelberg, The magnetic flux density was measured and predicted along
Germany) (primary antibody). After washing with phosphate-buf- (a) the axial direction (at point 0 along the radial direction) and
feredsaline,secondaryantibody(Alexa488anti-rabbit,diluted1:200) (b)theradialdirection(atpoint0alongtheaxialdirection).
Experimental&MolecularMedicine
Electromagneticfieldforosteogenesis
KSKangetal
4
described in Figure 1b when we set the magnetic flux dimensions of a commercialized well-plate (128(cid:3)86mm).
density to 1mT. The magnetic flux density varied throughout The average magnetic flux density in this area had a 5%
the cross-section and was highest at the center along the margin of error.
y axis (axial direction). In contrast, it decreased from both
sides (walls) to the center point along the x axis (radial
direction). ASC characterization
The magnetic flux density inside the solenoid coil was Before inducing osteogenic differentiation, ASCs were char-
measured using a three-axis sensor. As described in Figure 1c, acterized using flow cytometry (Figure 3a). ASCs showed
the sinusoidal magnetic flux density generated by the coil was negative expression of hematopoietic markers CD34 and
controllable. All measured data along the cross-section of the CD45. In contrast, ASCs were positive for the cell adhesion
solenoidcoilwhenitwas setto1mTaredescribedina three- molecules CD29 (integrin b ) and CD90 (Thy-1), the metal-
1
dimensional graph (Figure 1d). The magnetic flux density at loproteinase CD13 (aminopeptidase N) and the receptor
thecentralareaalongtheaxialdirectionwashigherthaninthe molecule CD105 (endoglin).
other sections. Next, we evaluated the osteogenic potential of ASCs under
This measured magnetic flux density corresponded to both the electromagnetic field and OM (Figure 3b). Both the
predicted values from Figure 1b, as shown in Figure 2. These electromagnetic field (electromagnetic field-normal growth
data were predicted and measured under identical conditions medium (NM)) and OM (Control-OM) increased ALP
when cells were exposed to the electromagnetic field with an expression and matrix mineralization as compared with the
averagemagneticfluxdensityof1mT.Weestablishedanearly control-NM group. ALP was most strongly activated and
uniform area (based on these results) as the experimental mineralized nodules were most heavily formed if ASCs were
sectionwherecellswerepositioned.Thisexperimentalareawas treated with both the electromagnetic field and OM
atthecenteralongboththeaxialandradialdirectionswiththe simultaneously.
Figure 3 Characterization of adipose-derived stem cells (ASCs). (a) To characterize ASCs, CD (cluster of differentiation) marker proteins
were analyzed using fluorescent-conjugated antibodies. ASCs were not hemopoietic (that is, negative for CD34 and CD45), but positive
for adhesion (CD29 and CD90), metalloproteinase (CD13) and receptor molecules (CD105). (b) Osteogenic differentiation of ASCs was
evaluated by alkaline phosphate (ALP) staining on day 10 and Alizarin Red S staining on day 20. The frequency and magnetic flux
density were 30Hz and 1mT, respectively. Either osteogenic induction medium (OM) or electromagnetic field induced osteogenic
differentiationofASCs.
Experimental&MolecularMedicine
Electromagneticfieldforosteogenesis
KSKangetal
5
Condition screening expression (30Hz (1mT), 45Hz (1mT)) and one with
RelativeALPexpressionwasevaluatedusingreal-timePCRfor decreased ALP expression (7.5Hz (1mT)).
conditionoptimization.Themagneticfluxdensityvariedfrom
0.1 to 3mT, and the frequency ranged from 7.5 to 75Hz. All
groups were composed of subgroups (that is, NM; Figure 4a Confirmation of the selected conditions
and OM; Figure 4b) to evaluate the synergistic effects of the We measured RUNX2 expression using immunostaining to
induction medium and electromagnetic field. ALP expression compare positive and negative groups with the control group
increasedinthefrequencyrangeof30to50Hzandamagnetic (Figure5).RUNX2expressionwasincreasedbyOM,asshown
flux density of 0.8 to 2.4mT. The area in red represents the in the control group. At 45Hz, the electromagnetic field
most highly expressed conditions. Frequencies and magnetic strongly inducedRUNX2expressioncomparedwith the other
flux densities outside of these ranges made no change of ALP groups. At 30Hz, the electromagnetic field also increased
expression (blueandvioletcolors).Asshown inFigure4b, 30 RUNX2 expression compared with the control group. How-
and 45Hz (1mT) increased ALPexpression. In contrast, ALP ever, RUNX2 expression was reduced compared with the
expression was not increased (compared with the control control groups when cells were exposed to a 7.5-Hz electro-
treated with OM; 2.38) at 7.5Hz (1 and 2mT). Overall, we magnetic field.
selected three groups to further examine osteogenic differen- WecheckedRUNX2,COL-I,OSXandOCexpressionlevels
tiation of ASCs; namely, two groups with increased ALP using real-time PCR (Figure 6). Expression levels of all genes
wereincreasedundertheelectromagneticfieldat30and45Hz
compared with the control groups. RUNX2 and COL-I levels
at 30Hz were higher than those at 45Hz, whereas OSX and
OClevelsat30Hzwerelowerthanthoseat45Hz.Geneswere
upregulated by the electromagneticfieldatboth30 and 45Hz
even without OM. Gene expression was not significantly
increased at 7.5Hz compared with the control with NM.
Figure 4 Alkaline phosphate (ALP) expression measured during
specific frequency and magnetic flux density conditions. Relative
ALP expression was visualized using real-time PCR on day 7
depending on the frequency and magnetic flux density
combinations. All levels were normalized to those of the control Figure 5 Immunostaining of electromagnetic field induced-RUNX2
group exposed tonormalgrowthmedium (NM). (a) Adipose-derived (runt-related transcription factor 2) expression. Fluorescence
stem cells (ASCs) were treated with NM. ALP expression differed images of RUNX2 expression in the osteogenic induction medium
according to combinations of the two parameters. Colors indicate (OM) groups were captured using a confocal microscope on day 4.
the relative ALP expression levels. (b) Relative expression levels of The 30 and 45Hz electromagnetic fields (positive) increased
ASCs treated with osteogenic induction medium (OM) are RUNX2 expression, whereas the 7.5Hz electromagnetic field
expressed in the bar graph. The control (OM) ALP expression level (negative) inhibited RUNX2 expression compared with the control
was 2.38. *Statistically significant differences relative to the group. The magnetic flux density was fixed at 1mT. Original
controlatPo0.05. magnification (cid:3)2000.
Experimental&MolecularMedicine
Electromagneticfieldforosteogenesis
KSKangetal
6
Figure 6 Osteogenic marker expression. Osteogenic marker expression was measured at day 7 using real-time PCR under both normal
growth medium (NM) and osteogenic induction medium (OM). (a) Runt-related transcription factor 2 (RUNX2); (b) collagen type-I, a1
(COL-I);(c)osterix(OSX);(d)osteocalcin(OC).*Po0.05(comparedwithcontrol-NM);#Po0.05(comparedwithcontrol-OM).
To evaluate the positive and negative effects of the electro- electromagnetic field on osteogenic differentiation of ASCs
magnetic field on the osteogenic differentiation we measured based on screening results using real-time PCR.
ALP activity (Figure 7a). We found that OM increased ALP Beforescreeningconditions,itwasimportanttoestablishthe
activity compared with NM groups, regardless of the electro- experimentalrangesofeachparameterbasedonanosteogenesis
magnetic field. The ALP activity was higher in the 30/45Hz literaturereview.1,2,4–10,12,13,15–17Panagopoulosetal.14proposed
groups when ASCs were treated with both NM and OM as the use of a weak electromagnetic field with low
compared with the control group cultured with identical frequency based on a biophysical model for the action of
medium. The activity levels of the 30/45Hz groups with NM oscillating magnetic fields on cells. Additionally, it is
were lower than the controlwith OM. In contrast, lower ALP known that an electromagnetic field with a frequency below
activity was detected at 7.5Hz as compared with the control 100Hz induces physiological effects as a result of ionic
group.AlthoughASCsweretreatedwithOM,theALPactivity interactions.27,28 Thus, based on this information and other
at 7.5Hz was lower than the control group treated with NM. previousstudies,weestablishedtheexperimentalrangesofthe
Matrix production of differentiated osteoblasts was evaluated two parameters.
using a calcium content assay (Figure 7b). Similar to ALP To provide a uniform stimulus to the cells, we first
activity, increased amounts of mineralized nodules were characterized the generated stimulus before evaluating its
detectedwhenASCswereexposedto30/45Hzelectromagnetic effects on cells. We estimated the uniform magnetic flux
fields compared with the control group. The calcium levels of density area of the solenoid coil using finite element analysis
the 7.5Hz electromagnetic field were lower than the control todeterminedimensions ofthe solenoid coilwith an effective
group. Theseresultssuggest thatthe30/45Hzelectromagnetic stimulation area. The central area of the coil was reasonably
fields stimulated ASC osteogenic differentiation, whereas uniform, which is the location of the commercial well plate.
the 7.5Hz electromagnetic field inhibited ASC osteogenic Based on these results, an electromagnetic field generator was
differentiation. developed, as described in Figure 1a. When the electromag-
neticfieldwasgeneratedinsidethe solenoidcoil,the distribu-
tions of the measured data were in good agreement with the
DISCUSSION estimated values (Figure 2). We established the uniform area
Here,wedemonstratedthatosteogenicdifferentiationofASCs witha5%marginoferrorwherecellscouldbeexposedtothe
could be regulated by controlling the conditions of an uniform electromagnetic field.
electromagnetic field in vitro. After characterizing the electro- Before inducing the osteogenic differentiation of ASCs, we
magnetic field generated by a solenoid coil, we verified the confirmed that the ASCs used in this study had typical
effects of both positive and negative conditions of the properties of mesenchymal stem cells, that is, the ASCs
Experimental&MolecularMedicine
Electromagneticfieldforosteogenesis
KSKangetal
7
electromagnetic field did not show positive effects on ALP
expression at 7.5Hz, regardless of the medium type. Taken
together, we selected 30/45Hz (1mT) and 7.5Hz (1mT) as
positive and negative conditions for osteogenic differentiation
of ASCs, respectively.
The effect of the selected three conditions on osteogenic
differentiation was evaluated using the gene expression of
other osteogenic markers. Similar to ALP, RUNX2 is an
important early osteogenic marker.33 Both the 30 and 45Hz
groups had increased RUNX2 expression compared with the
nonstimulated group, whereas the electromagnetic field at
7.5Hz did not increase RUNX2 expression even though cells
were treated with OM (Figure 5). Other osteogenic markers
wereupregulatedwiththeelectromagneticfieldatboth30and
45Hz compared with the nonstimulated (control) group.
However, it was difficult to conclude which frequency had
stronger potential to osteogenic differentiation (Figure 6).
ALP activity and calcium content results corroborated the
RUNX2 expression data. Although the positive conditions of
theelectromagneticfieldgeneratedsynergisticeffectswithOM
on ALP activity and mineralization (Figure 7), the individual
effects of OM on osteogenic differentiation appear stronger
than the electromagnetic field. This synergistic effect with the
electromagnetic field can be highly induced by other growth
factorssuchasbonemorphogeneticprotein2.16WithoutOM,
ALPactivitiesandcalciumcontentinthepositivegroupswere
not higher than the control with OM.
The 7.5Hz electromagnetic field seemed to inhibit osteo-
genicdifferentiationofASCs.ALPactivityat7.5Hzwaslower
than the control group regardless of the medium type.
Calciumcontentwasnotincreasedatthisfrequencycompared
Figure 7 Alkaline phosphate (ALP) activity and matrix with the control groups under the identical medium. More-
mineralization. (a) ALP activity was measured on day 10. (b) over, lower expression levels of other osteogenic markers
Calcium content was measured on day 20 to evaluate matrix
showed that specific conditions had negative potential to
mineralization induced by differentiated adipose-derived stem cells
osteogenic differentiation, even though all markers were not
(ASCs). The magnetic flux density was fixed at 1mT. *Po0.05
inhibited significantly. This inhibitory effect (7.5Hz) did not
(comparedwithcontrol-NM);#Po0.05(comparedwithcontrol-OM).
NM,normalgrowthmedium;OM,osteogenicinductionmedium. seemtobeasstrongasthepositiveeffect(30/45Hz).Basedon
these data, we conclude that osteogenic differentiation of
ASCs can be regulated by controlling the electromagnetic
expressed negative hematopoietic markers and positive adhe- conditions.
sion/receptor molecules.29–31 These ASCs were osteogenically Previous papers reported that osteogenic activities can be
differentiatedinthepresenceofpermissiveinductionmedium affected by different frequencies such as 15 or 60Hz.12,16,17
and electromagnetic field treatment, as illustrated in Similarly, we observed that the electromagnetic field at these
Figure 3b,32 which indicates that the ASCs used in this study frequencies increased ALP expression (Figure 4b). However,
had osteogenic potential that can be accelerated using our results suggested that frequencies from 30 to 45Hz can
induction medium and electromagnetic field treatment. provide stronger osteogenic effects than either 15 or 60Hz.
Relative ALP mRNA expression was evaluated using real- This is meaningful because these results were obtained in a
time PCR on day 7 to optimize conditions for osteogenic single identified system among various conditions to avoid
differentiationofASCs.Twoparameters,thatis,magneticflux differences caused by experimental and environmental
density and frequency, were controlled independently during conditions.
the electromagnetic field treatment. When ASCs were stimu- Although many questions remain unclear, our study pro-
lated with an electromagnetic field (NM), ALP expression vides a framework for future work to control differentiation
levels were dependent on a combination of magnetic flux lineages of stem cells using biophysical stimulation. Several
density and frequency (Figure 4a). In Figure 4b, ALP was general mechanisms regarding how electromagnetic fields
significantly upregulated at a magnetic flux density between 1 influence cellshave been introduced; analyses suffer from lack
and 2mTand a frequency from 30 to 45Hz. In contrast, the of proof on how specific conditions affect osteogenic
Experimental&MolecularMedicine
Electromagneticfieldforosteogenesis
KSKangetal
8
differentiation.14,27,28,34,35 We guess that our finding might be 11Trock DH, Bollet AJ, Dyer Jr RH, Fielding LP, Miner WK, Markoll R.
related to moving ions with specific frequencies, for example, Adouble-blindtrialoftheclinicaleffectsofpulsedelectromagneticfields
calcium ions. The ratio of the efflux of Ca2þ through the inosteoarthritis.JRheumatol1993;20:456–460.
12Soda A, Ikehara T, Kinouchi Y, Yoshizaki K. Effect of exposure to an
membrane can be increased at the frequency from 30 to extremelylowfrequency-electromagneticfieldonthecellularcollagenwith
50Hz.36 Higher efflux of Ca2þ might influence osteoblastic respecttosignalingpathwaysinosteoblast-likecells.JMedInvest2008;
55:267–278.
viability and function because the released ions play a
13MiduraRJ,IbiwoyeMO,PowellKA,SakaiY,DoehringT,GrabinerMDetal.
structural role of extracellular matrix formation.36–38 Pulsed electromagnetic field treatments enhance the healing of fibular
Treatment using the electromagnetic field could be an osteotomies.JOrthopRes2005;23:1035–1046.
auxiliary approach to enhancing cellular activities for tissue 14PanagopoulosDJ,KarabarbounisA,MargaritisLH.Mechanismforaction
ofelectromagneticfieldsoncells.BiochemBiophysResCo2002;298:
regeneration. Stimulation of cells with both electromagnetic
95–102.
field and proper chemical signals will promote the cellular 15Carson JJ, Prato FS, Drost DJ, Diesbourg LD, Dixon SJ. Time-varying
reaction synergistically. Additionally, this inherently noninva- magneticfieldsincreasecytosolicfreeCa2þ inHL-60cells.AmJPhysiol
1990;259:C687–C692.
siveandnoncontacttreatmentmethodiseasytobeappliedto
16Schwartz Z, Simon BJ, Duran MA, Barabino G, Chaudhri R, Boyan BD.
tissue regeneration using three-dimensional scaffolds.39–43 Pulsed electromagnetic fields enhance BMP-2 dependent osteoblastic
Exposure of the electromagnetic field to cells on scaffolds differentiationofhumanmesenchymalstemcells.JOrthopRes2008;26:
1250–1255.
with specific electromagnetic conditions will accelerate tissue
17SunLY,HsiehDK,YuTC,ChiuHT,LuSF,LuoGHetal.Effectofpulsed
formation in vitro. electromagnetic field onthe proliferationand differentiation potentialof
humanbonemarrowmesenchymalstemcells.Bioelectromagnetics2009;
30:251–260.
CONFLICT OF INTEREST
18Gimble JM, Katz AJ, Bunnell BA. Adipose-derived stem cells for regen-
The authors declare no conflictof interest. erativemedicine.CircRes2007;100:1249–1260.
19Cowan CM, Shi YY, Aalami OO, Chou YF, Mari C, Thomas R et al.
Adipose-derived adult stromal cells heal critical-size mouse calvarial
ACKNOWLEDGEMENTS
defects.NatBiotechnol2004;22:560–567.
Thiswork was supported by the National ResearchFoundation of 20EstesBT,WuAW,GuilakF.Potentinductionofchondrocyticdifferentiation
Korea(NRF) grant funded by the Koreagovernment (MEST) (No. ofhumanadipose-derivedadultstemcellsbybonemorphogeneticprotein
2012-0001235)andWorldClassUniversity(WCU)programthrough 6.ArthritisRheum2006;54:1222–1232.
theNationalResearchFoundationofKoreafundedbytheMinistryof 21AngleSR,SenaK,SumnerDR,VirdiAS.Osteogenicdifferentiationofrat
bone marrow stromal cells by various intensities of low-intensity pulsed
Education, Science and Technology (Project No. R31-2008-000-
ultrasound.Ultrasonics2011;51:281–288.
10105-0). 22HalvorsenYD,FranklinD,BondAL,HittDC,AuchterC,BoskeyALetal.
Extracellular matrix mineralization and osteoblast gene expression by
human adipose tissue-derived stromal cells. Tissue Eng 2001; 7:
729–741.
23Lee SW, Padmanabhan P, Ray P, Gambhir SS, Doyle T, Contag C et al.
1 TrainaGC,FontanesiG,CostaP,MammiGI,PisanoF,GiancecchiFetal.
Stem cell-mediated accelerated bone healing observed with in vivo
Effectofelectromagneticstimulationonpatientssufferingfromnon-union.
molecularandsmallanimalimagingtechnologiesinamodelofskeletal
Aretrospectivestudywithacontrolgroup.ElectromagnBiolMed1991;
injury.JOrthopRes2009;27:295–302.
10:101–117.
24Shoji T, Ii M, Mifune Y, Matsumoto T, Kawamoto A, Kwon SM
2 FredericksDC,NepolaJV,BakerJT,AbbottJ,SimonB.Effectsofpulsed
et al. Local transplantation of human multipotent adipose-derived stem
electromagneticfieldsonbonehealinginarabbittibialosteotomymodel.
cells accelerates fracture healing via enhanced osteogenesis and angio-
JOrthopTrauma2000;14:93–100.
genesis.LabInvest2010;90:637–649.
3 Rubin CT, McLeod KJ,Lanyon LE. Preventionof osteoporosis by pulsed
25Hong JM, Kim BJ, Shim JH, Kang KS, Kim KJ, Rhie JW
electromagneticfields.JBoneJointSurgAm1989;71:411–417.
etal.Enhancementofboneregenerationthroughfacilesurfacefunctio-
4 Tabrah F, HoffmeierM,GilbertJrF, Batkin S,Bassett CA. Bonedensity
nalization of solid freeform fabrication-based three-dimensional
changesinosteoporosis-pronewomenexposedtopulsedelectromagnetic
fields(PEMFs).JBoneMinerRes1990;5:437–442. scaffolds using mussel adhesive proteins. Acta Biomater 2012; 8:
5 Lohmann CH, Schwartz Z, Liu Y, Guerkov H, Dean DD, Simon B et al. 2578–2586.
Pulsed electromagnetic field stimulation of MG63 osteoblast-like cells 26Kang KS, Lee SJ, Lee H, Moon W, Cho DW. Effects of combined
affectsdifferentiationandlocalfactorproduction.JOrthopRes2000;18: mechanical stimulation on the proliferation and differentiation of pre-
637–646. osteoblasts.ExpMolMed2011;43:367–373.
6 TsaiMT,LiWJ,TuanRS,ChangWH.Modulationofosteogenesisinhuman 27Gartzke J, Lange K. Cellular target of weak magnetic fields: ionic
mesenchymalstemcellsbyspecificpulsedelectromagneticfieldstimula- conductionalongactinfilaments ofmicrovilli.AmJ Physiol-CellPhysiol
tion.JOrthopRes2009;27:1169–1174. 2002;283:C1333–C1346.
7 De Mattei M, Caruso A, Traina GC, Pezzetti F, Baroni T, Sollazzo V. 28Funk RHW, Monsees TK. Effects of electromagnetic fields on cells:
Correlationbetweenpulsedelectromagneticfieldsexposuretimeandcell physiological and therapeutical approaches and molecular mechanisms
proliferationincreaseinhumanosteosarcomacelllinesandhumannormal ofinteraction.CellsTissuesOrgans2006;182:59–78.
osteoblastcellsinvitro.Bioelectromagnetics1999;20:177–182. 29GronthosS,FranklinDM,LeddyHA,RobeyPG,StormsRW,GimbleJM.
8 TsaiMT,ChangWHS,ChangK,HouRJ,WuTW.Pulsedelectromagnetic Surfaceproteincharacterizationofhumanadiposetissue-derivedstromal
fields affect osteoblast proliferation and differentiation in bone tissue cells.JCellPhysiol2001;189:54–63.
engineering.Bioelectromagnetics2007;28:519–528. 30Kolf CM, Cho E, Tuan RS. Mesenchymal stromal cells. Biology of adult
9 SelvarnuruganN,KwokS,VasilovA,JefcoatSC,PartridgeNC.Effectsof mesenchymalstemcells:regulationofniche,self-renewalanddifferentia-
BMP-2 and pulsed electromagnetic field (PEMF) on rat primary osteo- tion.ArthritisResTher2007;9:204.
blastic cell proliferation and gene expression. J Orthop Res 2007; 25: 31ZukPA,ZhuM,AshjianP,DeUgarteDA,HuangJI,MizunoHetal.Human
1213–1220. adiposetissueisasourceofmultipotentstemcells.MolBiolCell2002;
10FassinaL,VisaiL,BenazzoF,BenedettiL,CalligaroA,DeAngelisMGC 13:4279–4295.
etal.Effectsofelectromagneticstimulationoncalcifiedmatrixproduction 32Bunnell BA, Flaat M, Gagliardi C, Patel B, Ripoll C. Adipose-derived
bySAOS-2 cells overa polyurethane porousscaffold. TissueEng2006; stem cells: isolation, expansion and differentiation. Methods 2008; 45:
12:1985–1999. 115–120.
Experimental&MolecularMedicine
Electromagneticfieldforosteogenesis
KSKangetal
9
33LianJB,JavedA,ZaidiSK,LengnerC,MontecinoM,vanWijnenAJetal. scaffoldsfabricatedbysolidfreeformfabricationandpolymerreplication
Regulatorycontrolsforosteoblastgrowthanddifferentiation:roleofRunx/ method.IntJPrecisEngMan2011;12:695–701.
Cbfa/AMLfactors.CritRevEukarGene2004;14:1–41. 40KonradK,SevcicK,FoldesK,PiroskaE,MolnarE.Therapywithpulsed
34Halgamuge MN, Perssont BRR, Salford LG, Mendis P, Eberhardt J. electromagnetic fields in aseptic loosening of total hip protheses: a
Comparison between two models for interactions between electric and prospectivestudy.ClinRheumatol1996;15:325–328.
magneticfieldsandproteinsincellmembranes.EnvironEngSci2009; 41Liu FH, Shen YK, Lee JL. Selective laser sintering of a hydroxyapatite-
26:1473–1480. silicascaffoldonculturedMG63osteoblastsinvitro.IntJPrecisEngMan
35BelyaevIY,AlipovED.Frequency-dependenteffectsofELFmagneticfield 2012;13:439–444.
onchromatinconformationinEscherichiacolicellsandhumanlympho- 42TrockDH.Electromagneticfieldsandmagnets:Investigationaltreatment
cytes.BiochimBiophysActa2001;1526:269–276. for musculoskeletal disorders. Rheum Dis Clin North Am 2000; 26:
36Koch CLMB M, Persson BRR, Salford LG, Eberhardt JL. Interaction 51–62.
between weak low frequency magnetic fields and cell membranes. 43Yun JH, Yoo JH, Choi SH, Lee MH, Lee SJ, Song SU et al. Synergistic
Bioelectromagnetics2003;24:395–402. effectofbonemarrow-derivedmesenchymalstemcellsandplatelet-rich
37SunXH,McLamoreE,KishoreV,FitesK,SlipchenkoM,PorterfieldDM plasma on bone regeneration of calvarial defects in rabbits. Tissue Eng
et al. Mechanical stretch induced calcium efflux from bone matrix RegenMed2012;9:17–23.
stimulatesosteoblasts.Bone2012;50:581–591.
38Boonrungsiman S, Gentleman E, Carzaniga R, Evans ND, McComb DW,
PorterAEetal.Theroleofintracellularcalciumphosphateinosteoblast- ThisworkislicensedunderaCreativeCommons
mediated bone apatite formation. Proc Natl Acad Sci USA 2012; 109: Attribution-NonCommercial-NoDerivs 3.0 Un-
14170–14175.
ported License. To view a copy of this license, visit http://
39KimJ,LimD,KimYH,KohYH,LeeMH,HanIetal.Acomparativestudy
of the physical and mechanical properties of porous hydroxyapatite creativecommons.org/licenses/by-nc-nd/3.0/
Experimental&MolecularMedicine