Content uploaded by Timothée Houssin
Author content
All content in this area was uploaded by Timothée Houssin on Apr 26, 2015
Content may be subject to copyright.
ARTICLES
PUBLISHED ONLINE: 20 OCTOBER 2013 | DOI: 10.1038/NMAT3777
Biophysical regulation of epigenetic state and cell
reprogramming
Timothy L. Downing1,2, Jennifer Soto1,2, Constant Morez2,3†, Timothee Houssin2,4†, Ashley Fritz5,
Falei Yuan2,6, Julia Chu2, Shyam Patel2, David V. Schaffer1,2,5 and Song Li1,2*
Biochemical factors can help reprogram somatic cells into pluripotent stem cells, yet the role of biophysical factors
during reprogramming is unknown. Here, we show that biophysical cues, in the form of parallel microgrooves on the
surface of cell-adhesive substrates, can replace the effects of small-molecule epigenetic modifiers and significantly improve
reprogramming efficiency. The mechanism relies on the mechanomodulation of the cells’ epigenetic state. Specifically,
decreased histone deacetylase activity and upregulation of the expression of WD repeat domain 5 (WDR5)—a subunit of H3
methyltranferase—by microgrooved surfaces lead to increased histone H3 acetylation and methylation. We also show that
microtopography promotes a mesenchymal-to-epithelial transition in adult fibroblasts. Nanofibrous scaffolds with aligned
fibre orientation produce effects similar to those produced by microgrooves, suggesting that changes in cell morphology may
be responsible for modulation of the epigenetic state. These findings have important implications in cell biology and in the
optimization of biomaterials for cell-engineering applications.
Cell reprogramming represents a major advancement in cell
biology and has wide applications in regenerative medicine,
disease modelling and drug screening. Induced pluripotent
stem cells (iPSCs) can be produced from somatic cells by the forced
expression of four or fewer transcription factors (for example,
Oct4 (O), Sox2 (S), Klf4 (K) and c-Myc (M), or Oct4, Sox2,
Nanog and Lin28; refs 1–4). Recently, the overexpression of
epithelial-cadherin (E-cad) was shown to replace the need for
Oct4—the most critical factor of the OSKM cocktail—during
cell reprogramming5. In addition, the elimination of E-cad in
somatic cells prevents nuclear reprogramming to pluripotency all
together6. These results corroborate the importance of the reported
mesenchymal-to-epithelial transition (MET) that occurs during the
reprogramming process7.
Since the first report on iPSC generation, researchers have em-
ployed numerous methods to improve reprogramming efficiency.
Chemical compounds, in particular, have shown great promise
in enhancing the efficiency of iPSC formation and/or eliminating
the need for some oncogenic transcription factors such as c-Myc
and Klf4 (refs 8–12). For example, valproic acid (VPA), a histone
deacetylase (HDAC) inhibitor, was reported to increase the per-
centage of Oct4+cells generated during reprogramming11. Tranyl-
cypromine hydrochloride (TCP), an inhibitor of lysine-specific
demethylase, was also shown to significantly improve reprogram-
ming efficiency9. Researchers have more recently demonstrated
the ability of VPA to promote transcription-factor-free repro-
gramming through the overexpression of microRNAs (miRNAs),
specifically the miR302/367 cluster, in mouse fibroblasts13. Inter-
estingly, this miRNA-mediated reprogramming was shown to be
dependent on VPA-induced degradation of HDAC2, suggesting
that HDAC proteins act as inhibitors of cell reprogramming.
1UC Berkeley & UCSF Joint Graduate Program in Bioengineering, Berkeley/San Francisco, California 94720/94143, USA, 2Department of Bioengineering,
University of California, Berkeley, B108A Stanley Hall, Berkeley, California 94720-1762, USA, 3Ecole Polytechnique, 91128 Palaiseau, France, 4University
Lille Nord de France, F-59000 Lille, France, 5Department of Chemical and Biomolecular Engineering, University of California, Berkeley, California
94720-1762, USA, 6Med-X Center, Shanghai Jiao Tong University, Shanghai 200030, China. †These authors contributed equally to this work.
*e-mail: song_li@berkeley.edu
These studies, along with many others14–16, highlight the critical
role of histone modifications and other epigenetic factors in cell
reprogramming.
Although extensive studies have been performed on the roles
of transcription factors, miRNAs and chemical compounds in cell
reprogramming, the role of biophysical cues in this process has
yet to be revealed. Biophysical factors (that is, the topographical
and mechanical properties of cell-adhesive materials) have been
shown to regulate a variety of cellular functions such as migration,
proliferation and differentiation17–29. Importantly, these functions
have closely related influences in a broad range of complex
biological processes such as wound healing, tissue remodelling and
tumour growth. There is also growing evidence that biophysical
signals can be transmitted through the cytoskeleton, and into the
nucleus and chromatin30,31.
In addition, some studies have demonstrated that substrate
properties can help regulate adult stem cell self-renewal32,33, and
an increasing number of laboratories have investigated the role
of mechanotransduction in the maintenance of embryonic stem
cells6,20,34–36 (ESCs). These studies highlight the importance of
biomechanics in the regulation of pluripotency. For example, an
intact actin–myosin network is critical for the stable propagation
of human ESCs. Non-muscle myosin IIA, specifically, was found
to play a critical role in ESC maintenance because its inhibition,
by short hairpin RNA (shRNA) knockdown or treatment with
blebbistatin, downregulates the expression of E-cad, a transmem-
brane mechanosensor, leading to a deregulation of the pluripotency
circuitry6,34. Still, it is not known whether or how the biophysical
properties of cell-adhesive materials influence adult cell reprogram-
ming, a reverse process of cell differentiation. In addition, whether
mechanotransduction can induce critical epigenetic modifications
1154 NATURE MATERIALS |VOL 12 |DECEMBER 2013 |www.nature.com/naturematerials
© 2013 Macmillan Publishers Limited. All rights reserved.
NATURE MATERIALS DOI: 10.1038/NMAT3777 ARTICLES
Flat 40 μm
30 μm
20 μm 10 μm
Day 0 Day 2 Day 12¬14
Nanog+ colonies
per 105 cells
Nanog+ colonies
per 105 cells
OSKM OSK
∗
∗∗
DAPI+
Infection
(1 day)
Seed to
substrate
Colony
formation
MEF
medium
mESC
medium
10 μm
Flat
0
25
50
75
Flat 40 μm 20 μm 10 μm
0
20
40
60
Flat 10 μm
Oct4 Nanog SSEA-1Sox2 Rosette Cartilage Epithelium
gh
def
Nanog
DAPI
ab c
Figure 1 |Microgroove substrates altered fibroblast morphology and improved iPSC generation. a, Scanning electron micrograph of PDMS membranes
with a 10 µm groove width. All grooves were fabricated with a groove height of 3 µm. b, The top row shows phase contrast images of flat and grooved
PDMS membranes with various widths and spacings. The bottom row shows fibroblast morphology on various PDMS membranes. Images are
fluorescence micrographs of the nucleus (DAPI, blue) and actin network (phalloidin, green; scale bars, 100µm). c, Reprogramming protocol. Colonies were
subcultured and expanded or immunostained and quantified by day 12–14. d, Fluorescence micrograph showing the morphology of iPSC colonies generated
on flat and grooved membranes (scale bar, 1 mm). Groove dimensions were 10µm in width and spacing, denoted as 10 µm in this and the rest of the figures.
Double-headed arrow indicates microgroove orientation of alignment. e, Reprogramming efficiency of fibroblasts transduced with OSKM and cultured on
PDMS membranes with flat or grooved microtopography. The number of biological replicates, n, used for this experiment was equal to 6. Groove width and
spacing were varied between 40, 20 and 10µm. Differences of statistical significance were determined by a one-way ANOVA, followed by Tukey’s post-hoc
test. * indicates significant difference (p<0.05) compared with the control flat surface. f, Reprogramming efficiency in fibroblasts transduced with OSK
(n=4). ∗p<0.05 (two-tailed, unpaired t-test) compared with the control flat surface. Error bars represent one standard deviation. g, Immunostaining of a
stable iPSC line expanded from colonies generated on 10µm grooves. These cells express mESC-specific markers Oct4, Sox2, Nanog and SSEA-1 (scale
bar, 100 µm). h, The expanded iPSCs in gwere transplanted into SCID mice to demonstrate the formation of teratomas in vivo (scale bar, 50µm).
in somatic cells, which might facilitate the reprogramming process,
also remains unclear.
To elucidate the role of biophysical factors in cell reprogramming,
we have used reprogramming technology in conjunction with var-
ious bioengineered substrates. Our data show that the biophysical
microenvironment, in the form of micro- and nano-scale topogra-
phy on cell-adhesive substrates, can induce pronounced changes in
histone acetylation and methylation patterns, which are dependent
on cell morphological changes and actin–myosin tension. These
epigenetic changes can replace the effects of potent small-molecule
epigenetic modifiers and significantly improve iPSC generation
efficiency. Furthermore, we identify specific mediators of this epi-
genetic mechanomodulation. Our findings represent a significant
advancement in the understanding of mechanotransduction by
revealing a new relationship between the biophysical microenviron-
ment, epigenetic mechanomodulation and cell reprogramming.
Microtopography enhances cell reprogramming
To assess whether topography might influence the process of cell
reprogramming, primary fibroblasts, isolated from adult mouse
ears, were first transduced using polycistronic lentiviral vectors,
STEMCCA–loxP or STEMCCA–loxP–RedLight37, encoding repro-
gramming factors OSKM or OSK, respectively. Following transduc-
tion, cells were seeded onto flat poly(dimethyl siloxane) (PDMS)
membranes or those fabricated with parallel microgrooves of vari-
ous width and spacing (40, 20 and 10 µm). The three-dimensional
structure of PDMS membranes with a 10 µm groove width is
shown in Fig. 1a. Groove heights were maintained at 3 µm for
all cases. As expected, microgroove substrates had a pronounced
effect on fibroblast alignment and nuclear elongation (Fig. 1b).
In general, cells aligned with microgrooves, and cell spreading
decreased on microgrooves. In addition, nuclear shape index
decreased (more elongated) as microgroove width and spacing
decreased (Supplementary Fig. 1a). Cell proliferation, as quantified
by EdU (5-ethynyl-2’-deoxyuridine) incorporation, also decreased
slightly (Supplementary Fig. 1b).
After seeding, transduced cells were cultured in mouse ESC
(mESC) culture conditions (see Fig. 1c for the reprogramming
procedure). After 12–14 days, iPSC colonies were subcultured
and expanded or immunostained for Nanog protein expression
NATURE MATERIALS |VOL 12 |DECEMBER 2013 |www.nature.com/naturematerials 1155
© 2013 Macmillan Publishers Limited. All rights reserved.
ARTICLES NATURE MATERIALS DOI:10.1038/NMAT3777
and quantified. Colony formation was observed on all PDMS
membranes. Interestingly, colonies formed on microgrooves
exhibited a more elongated morphology relative to those formed
on flat surfaces, and favoured outgrowth along the axis of groove
alignment (Fig. 1d). In addition, we observed that microgroove
topography significantly enhanced cell reprogramming efficiency
in mouse fibroblasts transduced with OSKM, where microgrooves
of 10 µm width (and spacing) exhibited the most pronounced effect.
Here, the number of Nanog+colonies observed was more than
fourfold higher than that seen on flat PDMS membranes (Fig. 1e). A
similar trend was observed in mouse fibroblasts infected with only
3 factors (OSK; Fig. 1f). This enhancement of reprogramming effi-
ciency by microgrooves was also confirmed by using mouse fibrob-
lasts isolated from Oct4–GFP reporter mice (Supplementary Fig. 2).
Given that 10 µm-width grooves showed the greatest enhancement
in cell reprogramming, we used these membranes to investigate the
mechanism of action during further experimentation.
To further characterize iPSCs, colonies generated on 10 µm-
width grooved substrates were isolated and expanded for at least
six passages. Colonies were then immunostained to confirm their
expression of mESC-specific markers Oct4, Sox2, Nanog and SSEA-
1 (Fig. 1g and Supplementary Fig. 3). Before immunostaining, we
observed that iPSCs generated using STEMCCA–loxP–RedLight
did not exhibit mCherry fluorescence, suggesting that transgene
expression was silenced after six passages (data not shown). In
addition, we confirmed that the generated iPSCs were capable
of forming teratomas in vivo (Fig. 1h) and embryoid bodies
in vitro, with differentiation potential for all three germ layers
(Supplementary Fig. 4). Consistently, we also observed a similar
trend in reprogramming efficiency of normal neonatal human
dermal fibroblasts (NHDFs) cultured on flat and microgrooved
surfaces (Supplementary Fig. 5).
Biophysical regulation of histone modifications
Modulation of chromatin-modifying enzymes has a direct effect
on cell reprogramming16. In addition, somatic cells must undergo
pronounced changes in chromatin structure to overcome the
epigenetic barriers of cell reprogramming14,15. A number of
these barriers as they relate to histone modifications have been
well characterized. On the basis of our previous observations
in mesenchymal stem cells38, we postulated that microgroove
topography might increase the presence of histone H3 acetylation
(AcH3) marks in adult fibroblasts, which tend to promote cell
reprogramming. For this reason, we performed a western blotting
analysis to monitor global changes in AcH3 and other histone marks
shown to influence cell reprogramming.
Through our analysis, we found that microgrooves markedly
increased global AcH3 marks in non-transduced mouse fibroblasts
(Fig. 2a) in the absence of reprogramming factors. Interestingly,
we also observed an increase in methylation (both di- and
tri-methylation) of histone H3 at lysine 4 (H3K4me2 and
H3K4me3, respectively) on microgrooved substrates relative to
flat surfaces (Fig. 2a). A similar trend in histone modifications
was observed in fibroblasts infected with OSKM (Supplementary
Fig. 6). Confocal microscopy confirmed that this increase was
localized in the nucleus (Fig. 2b). As AcH3, H3K4me2 and
H3K4me3 marks are associated with transcript activation and/or
critical in the early phase of reprogramming39, we performed a
chromatin immunoprecipitation (ChIP)–quantitative polymerase
chain reaction (qPCR) assay at the promoter regions of Oct4, Sox2
and Nanog—the key regulators of the pluripotency network. ChIP
analysis revealed significant increases in AcH3 at all three promoter
regions on microgrooves relative to flat surfaces. The promoter
region of Nanog showed a significant increase in H3K4me2 marks,
whereas the promoter region of Sox2 showed a significant increase
in H3K4me3 marks (Fig. 2c). These data suggest that microgroove
topography can induce global and local changes in histone H3
acetylation and methylation that are highly favourable in the
activation of reprogramming genes.
Replacement of chemical compounds
To further test the hypothesis that the observed increases in
histone H3 acetylation and methylation on microgrooves play a
role in topography-enhanced cell reprogramming, we compared the
microgroove effect with that of potent small-molecule epigenetic
modifiers. In particular, VPA and TCP—HDAC and histone
demethylase enzyme inhibitors, respectively—have been shown to
greatly improve cell reprogramming efficiency9,11. Both VPA and
TCP induced global increases in AcH3, H3K4me2 and H3K4me3
in our system (Fig. 2a), suggesting that VPA and TCP had
some overlapping effects on histone modifications. To determine
whether microgrooves and VPA or TCP have similar effects on
cell reprogramming, we reprogrammed fibroblasts cultured on
flat and microgrooved membranes in the absence or presence of
chemical compounds—VPA and TCP, alone or combined. From
this experiment, we make several noteworthy observations. As
expected, VPA and TCP alone produced a significant increase
in the number of Nanog+colonies observed when the cells
were reprogrammed on flat membranes (Fig. 2d). Interestingly,
the number of colonies generated on flat membranes in the
presence of chemical compounds increased to levels comparable
with that of microgrooves in the absence of VPA and TCP,
indicating that microgroove substrates had a similar effect to the
chemical compounds in enhancing reprogramming efficiency. In
addition, VPA and TCP showed no effect on the number of
Nanog+colonies formed on microgrooves, suggesting that their
mechanism of action might share a similar pathway in improving
reprogramming efficiency.
As VPA was shown to be necessary for the generation of miRNA-
iPSCs in mouse fibroblasts13, although complete reprogramming
may not be consistently achieved40, we examined the ability of
microgrooved membranes to generate colonies under the forced
expression of the miRNA cluster 302/367 in the absence of
VPA. Remarkably, we were able to generate colonies only on
microgrooves but not on flat membranes (Supplementary Fig. 7a).
We were also unable to generate any colonies with VPA alone.
Although our microgrooved surfaces showed the ability to form
Nanog+and GFP+clones (using fibroblasts isolated from wild-type
and Oct4–GFP reporter mice, respectively) (Supplementary Fig.
7b,c), we were unable to expand these iPSC-like colonies, suggesting
that microgrooved surfaces could enhance the early stage of cell
reprogramming, and that additional chemical or reprogramming
factors may be necessary to achieve complete reprogramming
into iPSCs. These results demonstrated the ability of microgroove
topography to replace or mimic the effects of chemical compounds,
VPA and TCP, in cell reprogramming.
Mediators of mechanotransduction
To elucidate the upstream mechanotransduction pathways in-
volved in the regulation of the histone H3 modifications observed,
we first monitored nuclear HDAC activity in fibroblasts cultured on
microgrooved or flat membranes. Indeed, nuclear HDAC activity
was significantly reduced on microgrooved substrates relative to
the flat surfaces (Fig. 3a), which may account for the topography-
induced AcH3. Western blotting analysis was performed to examine
the expression of several HDAC proteins in mouse fibroblasts.
Our results indicate that microgrooves induce pronounced de-
creases in the expression of HDAC2 but not HDAC1 and HDAC3
(Fig. 3b). By using confocal microscopy, we observed a presence
of HDAC2 specifically within the nucleus of mouse fibroblasts,
which is markedly decreased on microgrooves (Fig. 3c). As miRNA-
mediated reprogramming was shown to be specifically dependent
1156 NATURE MATERIALS |VOL 12 |DECEMBER 2013 |www.nature.com/naturematerials
© 2013 Macmillan Publishers Limited. All rights reserved.
NATURE MATERIALS DOI: 10.1038/NMAT3777 ARTICLES
OSKM
Fold change
∗
∗
∗∗
Flat 10 μm
∗
AcH3
H3K4me2
H3K4me3
AcH3
H3K4me2
H3K4me3
AcH3 H3K4me2
Flat Flat
∗
10 μmFlat
Nanog+ colonies per 105 cells
Ctrl VPA TCP VPA + TCP
Flat +VPA
10 μm+TCP
0
1
2
0
2
4
6
8
10
0
5
10
15
20
AcH3
H3K4me2
H3K4me3
NanogOct4 Sox2
0
30
60
90
a
c
d
b
Flat
10 μm
H3K4me3
GAPDH
H3K4me2
AcH3
H3
Actin
H3K4me3
Figure 2 |Microtopography induced histone modifications and replaced VPA and TCP in reprogramming. a, Mouse fibroblasts were cultured on
microgrooved or flat surfaces in the absence or presence of VPA or TCP for 3 days, followed by western blotting analysis of histone modifications AcH3,
H3K4me2 and H3K4me3. Total histone H3, GAPDH and actin are shown as loading controls. b, Confocal microscopy shows the fluorescence intensity of
histone modifications localized in the nucleus (scale bar, 10 µm). Double-headed arrow indicates microgroove orientation of alignment. c, ChIP–qPCR
analysis shows fold enrichment of histone modifications at the promoter regions of Oct4, Sox2 and Nanog (n=3). d, Reprogramming efficiency of
fibroblasts transduced with OSKM in the presence of chemical compounds, VPA and TCP (n=5). ∗p<0.05 (two-tailed, unpaired t-test) compared with
the control flat surface. Error bars represent one standard deviation.
on VPA-induced degradation of the HDAC2 protein, microgrooved
topography, as with VPA, might promote the increase in global
AcH3 through the downregulation of HDAC2, which may lead to
the observed enhancement in cell reprogramming by OSKM and in
the partial reprogramming by miRNA302/367 transduction (Fig. 1e
and Supplementary Fig. 7a).
To investigate potential upstream mediators of the observed
increase in H3K4 methylation on microgrooves, we screened
for regulators of histone methylation. Given the recent evi-
dence that WD repeat domain 5 (WDR5)—a subunit of H3
methyltranferase—mediates both cell reprogramming and ESC
self-renewal41, we probed for its expression through western blot-
ting analysis and immunostaining. Interestingly, we observed that
microgrooves markedly upregulated the expression of WDR5 in
mouse fibroblasts (Fig. 3d,e). This result led us to perform RNA
interference on WDR5 in the presence of microgrooves to deter-
mine its effects on histone modifications. Sufficient knockdown of
WDR5 was confirmed at protein and gene expression levels (Fig. 3f
and Supplementary Fig. 8). Surprisingly, WDR5 knockdown on
grooves resulted in substantially lower levels of both H3K4me2 and
H3K4me3 as well as AcH3 in mouse fibroblasts (Fig. 3f), suggesting
that WDR5 plays a critical role in the modulation of both histone
NATURE MATERIALS |VOL 12 |DECEMBER 2013 |www.nature.com/naturematerials 1157
© 2013 Macmillan Publishers Limited. All rights reserved.
ARTICLES NATURE MATERIALS DOI:10.1038/NMAT3777
WDR5
WDR5
GAPDH
Actin
Flat 10 μm10 μm
Relative nuclear HDAC activity
HDAC2
0.0
0.5
1.0
Flat 10 μm
Flat 10 μm
HDAC2
HDAC1
HDAC3
GAPDH
Actin
Flat 10 μm
Flat 10 μm
GAPDH
WDR5
H3K4me3
H3
Actin
AcH3
H3K4me2
a
d
e
f
bc
∗
WDR5ControlsiRNA
Figure 3 |Microtopography reduced HDAC activity, downregulated HDAC2 and upregulated WDR5. For these experiments, mouse fibroblasts were
analysed after being cultured on flat or microgrooved surfaces for 3 days. a, Nuclear HDAC activity assay (n=3). ∗P<0.05 (two-tailed, unpaired t-test)
compared with the control flat surface. Error bars represent one standard deviation. b, Western blotting analysis of HDACs. c, Confocal microscopy of
HDAC2 staining. d, Western blotting analysis of WDR5. e, Confocal microscopy of WDR5. f, Western blotting analysis performed on fibroblasts cultured
on microgrooves after siRNA knockdown of WDR5. (Scale bars in c,e, 10 µm.) Double-headed arrows indicate microgroove orientation of alignment.
methylation and acetylation marks. This result also presents a new
role for WDR5 as a key mediator of mechanotransduction.
The mesenchymal-to-epithelial transition
MET is required for nuclear reprogramming in mouse fibroblasts7.
In addition, a number of studies have implicated the roles of
both HDAC2 and WDR5 in the expression of several epithe-
lial and mesenchymal-associated genes42,43. For these reasons,
we monitored the effect of microgroove topography on the
expression of several genes associated with a MET. Real-time
reverse-transcription PCR (qRT-PCR) analysis of non-transduced
fibroblasts cultured on microgrooves for 3 days showed increased
messenger RNA (relative to those cultured on flat membranes)
for several epithelial-related genes (for example, E-cad, epithe-
lial cell adhesion molecule (Ep-CAM), keratin 8 (Krt8), occludin
(Ocln) and claudin 3 (Cldn3)) and reduced mRNA for multiple
mesenchymal markers (for example, transforming growth factor β
receptor 1 (Tgfbr1), snail homologue 1 (Snai1), snail homologue
2 (Snai2), vimentin (Vim) and integrin β1 (Itgb1)). This result
suggests that microgroove topography promotes the initiation of
a MET in adult mouse fibroblasts (Fig. 4a). In an effort to link
the histone modifications observed on microgrooved surfaces to
the observed increases in epithelial-related gene expression, we
performed ChIP–qPCR analysis on histone H3 modifications at
the promoter of E-cad. Indeed, we observed increases in the
presence of H3K4me2 at the E-cad promoter (Fig. 4b). Other marks
showed no detectable change (Supplementary Fig. 9), suggesting
that H3K4me2 mediates the enhanced E-cad expression and MET
transition on microgrooves. Activation of transforming growth
factor-β(TGF-β) pathways has been shown to block MET in cell
reprogramming7. To study whether the observed MET plays a
role in the topographical enhancement of cell reprogramming,
we tested the effect of microgrooves on cell reprogramming in
the presence of TGF-β1. As shown in Fig. 4c, TGF-β1 completely
blocked the enhanced reprogramming by microgroove topogra-
phy. Furthermore, inhibition of the TGF-βpathway using Alk5i
increased reprogramming efficiency on flat surfaces as expected,
but showed no synergistic behaviour in reprogramming efficiency
on microgrooves (Fig. 4d). Taken together, these results suggest
that the MET induced by microgrooves plays an important role in
topography-enhanced cell reprogramming.
Actin–myosin tension and epigenetic mechanomodulation
Microgroove topography facilitates pronounced cytoskeleton reor-
ganization in fibroblasts. Still, it is not clear how or whether these
cytoskeletal changes are important in the mechanomodulation
of histone modifications and their upstream regulators. For this
reason, we disrupted actin–myosin contractility in fibroblasts by
treatment with blebbistatin, a non-muscle myosin-II inhibitor.
Remarkably, blebbistatin treatment eliminates the effect of mi-
crogrooves on the epigenetic modulation of AcH3, H3K4me2 and
H3K4me3 as well as the mediators HDAC2 and WDR5 (Fig. 4e).
These data strongly suggest that the cytoskeleton reorganization
observed in fibroblasts on microgrooves and the actin–myosin
contractility are critical in this mechanomodulation of epigenetic
state and thus cell reprogramming. Figure 4f summarizes the effects
1158 NATURE MATERIALS |VOL 12 |DECEMBER 2013 |www.nature.com/naturematerials
© 2013 Macmillan Publishers Limited. All rights reserved.
NATURE MATERIALS DOI: 10.1038/NMAT3777 ARTICLES
Fold change (10 μm/flat) of gene expression
Control + Alk5i
Flat 10 μm
Relative number of Nanog+ colonies
∗
Cell morphology change
on microgrooves
HDAC activity
(HDAC 2)
WDR5
Histone methylation
(H3K4me)
Histone acetylation
(AcH3)
MET
TCP
VPA Cytoskeleton reorganization
Adult cell iPSCs
HDM
E-cad
Fold change
H3K4me2
∗
∗
∗∗
+¬¬
H3K4me3
HDAC2
WDR5
AcH3
Actin
GAPDH
H3
Blebbistatin
H3K4me2
1
10
100
1,000
10,000
Tgfbr1 Snai1 Snai2 Vim Itgb1
0.0
0.1
1.0
0
1
2
3
4
5
Relative number of Nanog+ colonies
0
4
8
12
Flat 10 μm
0
1
2
3
4
5
c
efd
ab
Flat 10 μm
Control + TGF-β1
E-cad Ep-CAM Krt8 Ocln Cldn3
Figure 4 |Initiation of MET and contractility-dependent histone modifications. a, Mouse fibroblasts were cultured for three days on flat or microgrooved
surfaces, followed by qRT–PCR analysis of epithelial-related genes and mesenchymal genes (n=3). b, ChIP–qPCR analysis for the fold enrichment of
H3K4me2 modifications at the promoter region of E-cad (n=3). ∗p<0.05 (two-tailed, unpaired t-test) compared with the control flat surface. c,d, Mouse
fibroblasts were reprogrammed with OSKM in the absence or presence of MET inhibitor, TGF-β1 (c), or TGF-βinhibitor, A-83-01 (Alk5i) (d), and the
number of Nanog+colonies generated was quantified (n=5 and n=4, respectively). ∗p<0.05 (one-way ANOVA followed by Tukey’s post-hoc test).
Error bars represent one standard deviation. e, Western blotting analysis of mouse fibroblasts cultured on flat or microgrooved surfaces for three days in
the absence or presence of blebbistatin. f, A summary of microtopographical regulation of histone modifications and cell reprogramming in adult mouse
fibroblasts. HDM, histone demethylase.
of microgrooves on histone modifications and cell reprogramming
in adult mouse fibroblasts.
Nanoscale epigenetic regulation
Nanofibrous scaffolds can regulate cell morphology, cell organiza-
tion and cell migration, as can microgrooves23,25. To generalize the
role of topography in epigenetic mechanomodulation, we employed
the use of nanofibrous membranes in our studies. Figure 5a shows
nanofibre structure in aligned and random orientations and its
effect on fibroblast morphology. As with microgrooves, aligned
nanofibres encouraged fibroblast alignment and elongation over the
random nanofibres. Western blotting analysis revealed that fibrob-
lasts cultured on aligned nanofibres also exhibit global increases in
AcH3, H3K4me2 and H3K4me3 marks over fibroblasts cultured on
random fibres (Fig. 5b). In addition, these changes correlated well
with the decrease in HDAC2 and the increase in WDR5 protein
expression. Next, we seeded fibroblasts transduced with OSKM
onto nanofibre surfaces to determine the effects of random and
aligned fibre orientation on reprogramming efficiency. Indeed,
aligned nanofibres generated significantly more Nanog+colonies
than random fibres (Fig. 5c,d and Supplementary Fig. 10). These re-
sults confirm that nanofibre alignment produces an effect similar to
microgrooves in epigenetic modification and cell reprogramming.
Cell shape effect on histone modifications
A common effect of microgrooves and aligned nanofibres is cell
elongation. To directly determine whether this cell-morphology
change is sufficient to induce histone modifications or whether
topography is required, we used micropatterning techniques to gain
tight control over cell shape while culturing cells on topographically
flat surfaces. Fibroblasts were cultured on micropatterned islands
with well-defined cell shapes (round or elongated). Interestingly,
fluorescence microscopy revealed that cells in an elongated shape
(cell shape index or CSI =0.1) exhibited significantly higher levels
of nuclear AcH3, H3K4me2 and H3K4me3 as compared with cells
in a circular shape (CSI =1; Fig. 5e,f) with an equal spreading
area. These data suggest that the morphological change experienced
by fibroblasts, when interacting with the nano- and micro-scale
features of cell-adhesive materials is sufficient to induce epigenetic
modifications. Cell elongation also significantly correlated with an
elongation of the cell nucleus (Supplementary Fig. 11).
Our results demonstrate, for the first time, that biophysical
cues, in the form of parallel microgrooves or aligned nanofibres
on the surface of cell-adhesive substrates, can significantly improve
reprogramming efficiency and replace the effects of small-molecule
epigenetic modifiers. The mechanism behind this biophysical
enhancement of cell reprogramming relies on mechanomodulation
NATURE MATERIALS |VOL 12 |DECEMBER 2013 |www.nature.com/naturematerials 1159
© 2013 Macmillan Publishers Limited. All rights reserved.
ARTICLES NATURE MATERIALS DOI:10.1038/NMAT3777
Nanog+ colonies per 105 cells
RandomAligned
RandomAligned
H3K4me3
CSI = 1
CSI = 1
CSI = 1
AcH3H3K4me2
AcH3
Phalloidin
CSI = 0.1
H3K4me2
Phalloidin
CSI = 0.1
Phalloidin
CSI = 0.1
Random Aligned
0
5
10
15
20
Random Aligned
0
2
4
6
1.0 0.1
1.0 0.1
0
1
2
3
H3K4me3
Relative fluorescent intensity
1.0 0.1
CSI
CSI
CSI
0
1
2
3
∗
∗
∗
∗
Nanog DAPI
abc
de f
WDR5
AcH3
Actin
GAPDH
HDAC2
H3
H3K4me2
H3K4me3
Figure 5 |Nanoscale and morphological regulation of histone modifications and cell reprogramming. a, Scanning electron micrograph of nanofibres
showing fibre morphology in aligned and random orientations (scale bar, 20 µm). Confocal fluorescence micrograph of fibroblasts cultured on nanofibres
(DAPI (blue) and phalloidin (green) staining; scale bar, 100 µm). b, Western blotting analysis for fibroblasts cultured on random and aligned nanofibres for
three days. c, Fibroblasts were transduced with OSKM and seeded onto nanofibre surfaces, followed by immunostaining for Nanog expression (red) at
day 12. Nuclei were stained with DAPI in blue; scale bar, 500 µm. d, Quantification of colony numbers in c(n=5). ∗p<0.05 (two-tailed, unpaired t-test)
compared with the control surface with random nanofibres. e, Fibroblasts were micropatterned into single-cell islands of 2,000 µm2area with a CSI value
of 1 (round) or 0.1 (elongated). After 24 h, cells were immunostained for AcH3, H3K4me2 or H3K4me3 (in green). Phalloidin staining (red) identifies the
cell cytoskeleton for cell shape accuracy. The white arrowhead indicates the location of the nucleus (scale bars, 20 µm). f, Quantification of fluorescence
intensity in e(n=34, 20 and 34 for AcH3, H3K4me2 and H3K4me3, respectively). ∗p<0.05 (two-tailed, unpaired t-test) compared with the circular
micropatterned cells (CSI =1). Error bars represent one standard deviation.
of H3 acetylation and methylation marks, which are regulated
by HDAC2 and WDR5. These epigenetic changes may also play
a role in the increased expression of epithelial-related genes
on microgrooved membranes. We believe this new biophysical
regulation of epigenetics has important implications in the broad
scope of cell biology, and provides a rational basis for the
optimization of biomaterials and the cellular microenvironment for
a number of biological applications.
In any given physiological microenvironment, cells may experi-
ence a number of different biophysical inputs, which, as we show,
might induce critical changes in epigenetic signature. Given the
broad influence of epigenetics in cell behaviour and phenotype
determination, our results have interesting implications beyond
this field. It will be useful to explore, for example, whether these
microtopography-induced changes in histone marks poise adult
fibroblasts into a more plastic or unstable epigenetic state, allow-
ing them to more readily transition into other phenotypes. Such
evidence could greatly advance the field of cell biology by helping
understand how mechanomodulated epigenetic changes alter cell
behaviour in cell–material interactions, and direct cell reprogram-
ming into cardiomyocytes and neurons44–49, tumour growth and
metastasis, and adult stem/progenitor cell differentiation. This
information is critical to the medical field as it presents new
ways to maximize the potential of stem cell technologies, mitigate
pathophysiological response in injured and diseased states, and
greatly improve the design of next-generation biomaterials.
1160 NATURE MATERIALS |VOL 12 |DECEMBER 2013 |www.nature.com/naturematerials
© 2013 Macmillan Publishers Limited. All rights reserved.
NATURE MATERIALS DOI: 10.1038/NMAT3777 ARTICLES
Methods
Fibroblast isolation, culture and reprogramming. Ear tissues from adult C57BL/6
mice were minced and then partially digested in a solution of 3 mg ml−1collagenase
type IV (Invitrogen) and 0.25% trypsin–EDTA (GIBCO) for 30 min under constant
agitation at 37 ◦C. Partially digested tissues were plated and fibroblasts were
allowed to migrate out (passage 0). Isolated fibroblasts were expanded in MEF
medium (DMEM (Hyclone) +10%FBS (Hyclone) and 1% penicillin/streptomycin
(GIBCO)) and used exclusively at passage 2 for all experiments. Fibroblasts from
Oct4–GFP reporter mice (008214; The Jackson Laboratory) were also isolated as
described above. NHDFs were purchased from Lonza (CC-2509) and expanded in
FGM-2 BulletKit (CC-3132) for 1–2 passages. After transduction, mouse fibroblasts
were seeded onto PDMS membranes with flat and microgroove topographies or
nanofibres of aligned and random orientation. After seeding, transduced fibroblasts
were maintained in MEF medium for 48 h. For the next 10–12 days, cells were
fed with mouse mESC medium (DMEM +15% mouse ESC maintenance FBS
(STEMCELL Technologies), 100 units ml−1ESGRO mouse leukaemia inhibitory
factor (mLIF; Millipore), 0.1 mM β-mercaptoethanol (Sigma) and 1×MEM
non-essential amino acid (GIBCO)). Reprogrammed NHDFs were cultured in
mTeSR 1 complete medium from STEMCELL Technologies. Human iPSCs were
reprogrammed for 1 month after the infection with OSKM before colonies were
expanded and/or analysed. Culture medium was replenished every 1–2 days during
reprogramming. VPA (0.5 mM; Sigma) and TCP (5 µM; Sigma) were used for
biochemical modulation of epigenetics. Cells were exposed to TGF-β1 (2 ng ml−1)
for 5 days for activation of the TGF-βpathway. A-83-01 (0.5µM; Millipore)
was used for all 14 days during the reprogramming. Blebbistatin (10 µM; or
dimethylsulphoxide) was administered for 3 days.
Lentiviral production and cell transduction. Lentiviral vectors were used to
transduce mouse fibroblasts for ectopic expression of three (OSK) or four
(OSKM) of the key reprogramming genes or miRNAs (miR302/367cluster).
STEMCCA–loxP and STEMCCA–loxP–RedLight plasmids were a generous gift
from G. Mostoslavsky at Boston University. Lentivirus was produced using
common, well-established calcium phosphate transfection methods. Viral particles
were collected and concentrated using Lenti-X Concentrator (Clontech) according
to the manufacturer’s protocol. Stable virus was aliquoted and stored at −80 ◦C.
For viral transduction, fibroblasts were seeded and allowed to attach overnight.
Cells were then incubated with virus for 24 h. After incubation, transduced cells
were reseeded onto either PDMS or nanofibre surfaces.
ChIP–qPCR and qRT–PCR. ChIP–qPCR was performed by using a
high-throughput ChIP kit (EZ-Magna ChIP HT96; Millipore), according to
the manufacturer’s instructions. Sonicated chromatin prepared from 100,000
mouse fibroblasts was subjected to ChIP by using 3 µg of normal rabbit IgG
(CS200581; Millipore) or specific antibodies for AcH3 (06-599; Millipore),
H3K4me2 (05-1338; Millipore) or H3K4me3 (07-473, Millipore). Substantial fold
enrichment was observed for each experimental condition (Supplementary Fig. 12).
ChIP–qPCR data were analysed by normalizing the DNA concentration to the
percentage input using a relative standard curve method. Primers for ChIP–qPCR
are included as (Supplementary Table 1). qRT–PCR was performed using a
customized StellARay Gene Expression System (Lonza) to probe for multiple
genes at once. RNA was isolated using a NucleoSpin RNA II kit from Clontech
and converted to complementary DNA using the RT2First Strand Kit (Qiagen).
qRT–PCR data were analysed using the 11Ct method.
RNA interference. RNA interference was performed using control (Cat. No.
4390853) and WDR5 (Cat. No. 4390771, ID: s100547, Lot: ASO0R7LI) Silencer
Select short interfering RNAs (siRNA) from Ambion. Transfections were carried
out using Lipofectamine RNAiMAX Reagent (Invitrogen) according to the
manufacturer’s instructions.
Western blotting analysis. Fibroblasts were lysed and collected in a buffer
containing 20 mM Tris-HCl, 150 mM NaCl, 1% Triton X-100, 0.1% SDS and
10 mM NaF along with protease inhibitors (phenylmethyl sulphonyl fluoride,
Na3VO4and leupeptin). Protein lysates were centrifuged to pellet cell debris, and
the supernatant was removed and quantified by DC Protein Assay (Bio-Rad).
Protein samples (15 mg per well) were run using SDS–PAGE and transferred to
polyvinylidene fluoride membranes. Membranes were blocked in 3% non-fat
milk and incubated with primary antibodies. Primary antibodies include
AcH3, H3K4me2, H3K4me3, HDAC 1, 2, 3, WDR5, H3, actin and GAPDH.
Refer to Supplementary Table 2 for all antibody information. All antibodies
were used at a 1:1,000 dilution and incubated overnight. Next, membranes
were incubated with HRP-conjugated IgG secondary antibodies (Santa Cruz
Biotechnologies) for one hour. Protein bands were visualized using Western
Lightning Plus-Enhanced Chemiluminescence Substrate (Perkin Elmer Life
& Analytical Sciences).
Nuclear HDAC activity. Nuclear HDAC activity was performed as
previously described38.
Microgroove, nanofibre and micropatterned substrate fabrication.
Bioengineered substrates were fabricated as previously described23,38,50,51.
Briefly, PDMS membranes were fabricated using well-established soft lithography
procedures. Poly-l-lactic acid or poly(l-lactide-co-caprolactone) nanofibres were
fabricated using electrospinning technology where random fibres were ejected onto
a grounded collector. Aligned fibres were produced by collector modifications or
mechanical stretch. Micropatterned islands on PDMS substrates were fabricated
using oxygen plasma treatment through the windows of a PDMS membrane mask
that produced hydrophilic areas of different shapes and sizes. All substrates were
coated with 2% gelatin for 1 h to promote cell attachment.
Immunostaining, imaging and quantifications, teratoma formation and
in vitro differentiation. For immunostaining, cells were fixed with 4%
paraformaldehyde, permeabilized with 0.5% Triton X-100 (Sigma) and blocked
with 1% BSA (Sigma). For actin-cytoskeleton staining, samples were incubated with
fluorescein-isothiocyanate-conjugated phalloidin (Invitrogen) for 2 h. Primary
antibodies were incubated overnight at 4 ◦C, followed by 1h incubation with Alexa
488- and/or Alexa 546-labelled secondary antibodies (Molecular Probes). Nuclei
were stained with 4’,6-diamino-2-phenylindole (DAPI; Invitrogen).
For counting iPSC colonies, entire wells were imaged using an ImageXpress
Micro System (Molecular Devices). Colonies were determined positive for
Nanog protein expression on the basis of positive (mESCs) and negative
(uninfected/non-reprogrammed mouse fibroblasts) thresholds. Confocal and
epifluorescence images were collected using a Zeiss LSM710 microscope and Zeiss
Axio Observer.A1, respectively. DAPI images were used to determine nuclear shape
index. For quantification of AcH3, H3K4me2 and H3K4me3 fluorescence intensity,
×40 epifluorescence images were used. DAPI staining was used as a mask and
fluorescence intensity was averaged over the entire nucleus. All image analyses were
performed using a Matlab script or ImageJ software.
Teratoma formation was performed by injecting 1 ×106cells into the
abdominal cavity of SCID/NOD mice. After 1 month mice were euthanized, and
the formed teratomas were explanted and fixed using paraffin-embedding and
sectioned using a microtome. Haematoxylin and eosin staining was performed
for histological analysis.
Embryoid body formation for in vitro differentiation was performed using the
hanging-drop method. Mouse iPSCs were cultured in mESC maintenance media
without LIF. Embryoid bodies were plated onto gelatin-coated surfaces and allowed
to spontaneously differentiate. After 2 weeks of differentiation, samples were fixed
and immunostained. Human iPSCs were overgrown in a feeder-free system using
mTeSR 1 complete medium to initiate spontaneous differentiation.
Statistical analysis. All data are presented as mean, plus one standard deviation,
where n≥3. Comparisons among values for groups greater than two were
performed using a one-way analysis of variance (ANOVA). Differences between
groups were then determined using Tukey’s post-hoc test. For two-group analysis,
a two-tailed, unpaired t-test was used. For all cases, p-values less than 0.05 were
considered statistically significant. GraphPad Prism 6.0 software was used for
all statistical evaluations.
Received 5 February 2013; accepted 11 September 2013;
published online 20 October 2013
References
1. Takahashi, K. & Yamanaka, S. Induction of pluripotent stem cells from
mouse embryonic and adult fibroblast cultures by defined factors. Cell 126,
663–676 (2006).
2. Yu, J. et al. Induced pluripotent stem cell lines derived from human somatic
cells. Science 318, 1917–1920 (2007).
3. Park, I-H. et al. Reprogramming of human somatic cells to pluripotency with
defined factors. Nature 451, 141–146 (2008).
4. Wernig, M. et al.In vitro reprogramming of fibroblasts into a pluripotent
ES-cell-like state. Nature 448, 318–324 (2007).
5. Redmer, T. et al. E-cadherin is crucial for embryonic stem cell pluripotency
and can replace OCT4 during somatic cell reprogramming. EMBO Rep. 12,
720–726 (2011).
6. Li, D. et al. Integrated biochemical and mechanical signals regulate multifaceted
human embryonic stem cell functions. J. Cell Biol. 191, 631–644 (2010).
7. Li, R. et al. A mesenchymal-to-epithelial transition initiates and is required
for the nuclear reprogramming of mouse fibroblasts. Cell Stem Cell 7,
51–63 (2010).
8. Feng, B., Ng, J-H., Heng, J-C. D. & Ng, H-H. Molecules that promote or
enhance reprogramming of somatic cells to induced pluripotent stem cells.
Cell Stem Cell 4, 301–312 (2009).
9. Li, Y. et al. Generation of iPSCs from mouse fibroblasts with a single gene,
Oct4, and small molecules. Cell Res. 21, 196–204 (2011).
10. Huangfu, D. et al. Induction of pluripotent stem cells from primary human
fibroblasts with only Oct4 and Sox2. Nature Biotechnol. 26, 1269–1275 (2008).
NATURE MATERIALS |VOL 12 |DECEMBER 2013 |www.nature.com/naturematerials 1161
© 2013 Macmillan Publishers Limited. All rights reserved.
ARTICLES NATURE MATERIALS DOI:10.1038/NMAT3777
11. Huangfu, D. et al. Induction of pluripotent stem cells by defined factors
is greatly improved by small-molecule compounds. Nature Biotechnol. 26,
795–797 (2008).
12. Zhu, S. et al. Reprogramming of human primary somatic cells by OCT4 and
chemical compounds. Cell Stem Cell 7, 651–655 (2010).
13. Anokye-Danso, F. et al. Highly efficient miRNA-mediated reprogramming
of mouse and human somatic cells to pluripotency. Cell Stem Cell 8,
376–388 (2011).
14. Pasque, V., Jullien, J., Miyamoto, K., Halley-Stott, R. P. & Gurdon, J. B.
Epigenetic factors influencing resistance to nuclear reprogramming. Trends
Genet. 27, 516–525 (2011).
15. Plath, K. & Lowry, W. E. Progress in understanding reprogramming to the
induced pluripotent state. Nature Rev. Genet. 12, 253–265 (2011).
16. Onder, T. T. et al. Chromatin-modifying enzymes as modulators of
reprogramming. Nature 483, 598–602 (2012).
17. Engler, A. J., Sen, S., Sweeney, H. L. & Discher, D. E. Matrix elasticity directs
stem cell lineage specification. Cell 126, 677–689 (2006).
18. Paszek, M. J. et al. Tensional homeostasis and the malignant phenotype.
Cancer Cell 8, 241–254 (2005).
19. McBeath, R., Pirone, D. M., Nelson, C. M., Bhadriraju, K. & Chen, C. S. Cell
shape, cytoskeletal tension, and RhoA regulate stem cell lineage commitment.
Dev. Cell 6, 483–495 (2004).
20. Chowdhury, F. et al. Material properties of the cell dictate stress-induced
spreading and differentiation in embryonic stem cells. Nature Mater. 9,
82–88 (2010).
21. Engler, A. J., Humbert, P. O., Wehrle-haller, B. & Weaver, V. M. Multiscale
modeling of form and function. Science 324, 208–212 (2009).
22. DuFort, C. C., Paszek, M. J. & Weaver, V. M. Balancing forces: Architectural
control of mechanotransduction. Nature Rev. 12, 308–319 (2011).
23. Patel, S. et al. Bioactive nanofibers: Synergistic effects of nanotopography and
chemical signaling on cell guidance. Nano Lett. 7, 2122–2128 (2007).
24. Thakar, R. G., Ho, F., Huang, N. F., Liepmann, D. & Li, S. Regulation of
vascular smooth muscle cells by micropatterning. Biochem. Biophys. Res.
Commun. 307, 883–890 (2003).
25. Huang, N. F. et al. Myotube assembly on nanofibrous and micropatterned
polymers. Nano Lett. 6, 537–542 (2006).
26. Park, J. S. et al. The effect of matrix stiffness on the differentiation
of mesenchymal stem cells in response to TGF-β.Biomaterials 32,
3921–3930 (2011).
27. Guilak, F. et al. Control of stem cell fate by physical interactions with the
extracellular matrix. Cell Stem Cell 5, 17–26 (2009).
28. Discher, D. E., Mooney, D. J. & Zandstra, P. W. Growth factors, matrices, and
forces combine and control stem cells. Science 324, 1673–1677 (2009).
29. Wozniak, M. A. & Chen, C. S. Mechanotransduction in development: A
growing role for contractility. Nature Rev. 10, 34–43 (2009).
30. Wang, N., Tytell, J. D. & Ingber, D. E. Mechanotransduction at a distance:
Mechanically coupling the extracellular matrix with the nucleus. Nature Rev.
10, 75–82 (2009).
31. Poh, Y-C. et al. Dynamic force-induced direct dissociation of protein complexes
in a nuclear body in living cells. Nature Commun. 3, 866 (2012).
32. McMurray, R. J. et al. Nanoscale surfaces for the long-term maintenance
of mesenchymal stem cell phenotype and multipotency. Nature Mater. 10,
637–644 (2011).
33. Gilbert, P. M. et al. Substrate elasticity regulates skeletal muscle stem cell
self-renewal in culture. Science 329, 1078–1081 (2010).
34. Li, D. et al. Role of mechanical factors in fate decisions of stem cells. Regen Med.
6, 229–240 (2011).
35. Chowdhury, F. et al. Soft substrates promote homogeneous self-renewal of
embryonic stem cells via downregulating cell-matrix tractions. PLoS ONE 5,
e15655 (2010).
36. Ji, L., LaPointe, V. L. S., Evans, N. D. & Stevens, M. M. Changes in embryonic
stem cell colony morphology and early differentiation markers driven by
colloidal crystal topographical cues. Europ. Cells Mater. 23, 135–146 (2012).
37. Sommer, C. A. et al. Excision of reprogramming transgenes improves the
differentiation potential of iPS cells generated with a single excisable vector.
Stem Cells 28, 64–74 (2010).
38. Li, Y. et al. Biophysical regulation of histone acetylation in mesenchymal stem
cells. Biophys. J. 100, 1902–1909 (2011).
39. Buganim, Y., Faddah, D. a & Jaenisch, R. Mechanisms and models of somatic
cell reprogramming. Nature Rev. Genet. 14, 427–439 (2013).
40. Tavernier, G., Mlody, B., Demeester, J., Adjaye, J. & De Smedt, S. C.
Current methods for inducing pluripotency in somatic cells. Adv. Mater. 25,
2765–2771 (2013).
41. Ang, Y-S. et al. WDR5 mediates self-renewal and reprogramming
via the embryonic stem cell core transcriptional network. Cell 145,
183–197 (2011).
42. Peinado, H., Ballestar, E., Esteller, M. & Cano, A. Snail mediates
E-cadherin repression by the recruitment of the Sin3A/histone deacetylase 1
(HDAC1)/HDAC2 complex. Mol. Cell. Biol. 24, 306–319 (2004).
43. Wu, M-Z. et al. Interplay between HDAC3 and WDR5 is essential
for hypoxia-induced epithelial-mesenchymal transition. Mol. Cell 43,
811–822 (2011).
44. Zhou, Q., Brown, J., Kanarek, A., Rajagopal, J. & Melton, D. A. In vivo
reprogramming of adult pancreatic exocrine cells to beta-cells. Nature 455,
627–632 (2008).
45. Ieda, M. et al. Direct reprogramming of fibroblasts into functional
cardiomyocytes by defined factors. Cell 142, 375–386 (2010).
46. Pang, Z. P. et al. Induction of human neuronal cells by defined transcription
factors. Nature 476, 220–223 (2011).
47. Karow, M. et al. Reprogramming of pericyte-derived cells of the adult human
brain into induced neuronal cells. Cell Stem Cell 11, 471–476 (2012).
48. Thier, M. et al. Direct conversion of fibroblasts into stably expandable neural
stem cells. Cell Stem Cell 10, 473–479 (2012).
49. Qian, L. et al.In vivo reprogramming of murine cardiac fibroblasts into induced
cardiomyocytes. Nature 485, 593–598 (2012).
50. Downing, T. L. et al. Drug-eluting microfibrous patches for the local
delivery of rolipram in spinal cord repair. J. Control Release 161,
910–917 (2012).
51. Cheng, Q., Komvopoulos, K. & Li, S. Surface chemical patterning for long-term
single-cell culture. J. Biomed. Mater. Res. A 96, 507–512 (2011).
Acknowledgements
This work is supported in part by grants from the California Institute of Regenerative
Medicine (RB3-05232) and the National Institute of Health (EB012240; to S.L.),
fellowships from the Ford Foundation and Siebel Scholars Foundation (to T.L.D.), a
fellowship from the National Science Foundation GRFP (to J.S.), and the Fulbright-France
Commission (to T.H.). The authors would like to thank M. West at the CIRM/QB3
Shared Stem Cell Facility of UC Berkeley, I. Conboy, D. Wang, C. Elabd, T. Yamaguchi,
C. W. Huang, I. Grubisic, W. C. Huang, E. Su, A. Chang and J. Zhang for their
assistance and fruitful discussions. The pLOVE 302/367 plasmid was provided by the
laboratory of R. Blelloch.
Author contributions
S.L., T.L.D., J.S., C.M., T.H. and D.V.S. designed the experiments. T.L.D., J.S., C.M.,
T.H., F.Y., J.C., A.F. and S.P. carried out experiments and analysed the data. S.L., T.L.D.
and J.S. wrote the manuscript.
Additional information
Supplementary information is available in the online version of the paper.Reprints and
permissions information is available online at www.nature.com/reprints. Correspondence
and requests for materials should be addressed to S.L.
Competing financial interests
The authors declare no competing financial interests.
1162 NATURE MATERIALS |VOL 12 |DECEMBER 2013 |www.nature.com/naturematerials
© 2013 Macmillan Publishers Limited. All rights reserved.
