Induction of Stem Cell Gene Expression in Adult Human
Fibroblasts without Transgenes
Raymond L. Page,1,2,3,4Sakthikumar Ambady,1,2William F. Holmes,2Lucy Vilner,1
Denis Kole,2Olga Kashpur,2Victoria Huntress,2Ina Vojtic,1Holly Whitton,2and Tanja Dominko1,2,4
Reprogramming of differentiated somatic cells into induced pluripotent stem (iPS) cells has potential for deri-
vation of patient-specific cells for therapy as well as for development of models with which to study disease
progression. Derivation of iPS cells from human somatic cells has been achieved by viral transduction of human
fibroblasts with early developmental genes. Because forced expression of these genes by viral transduction results
in transgene integration withunknownandunpredictablepotential mutageniceffects,identificationofcellculture
conditions that can induce endogenous expression of these genes is desirable. Here we show that primary adult
human fibroblasts have basal expression of mRNA for OCT4, SOX2, and NANOG. However, translation of these
messages into detectable proteins and their subcellular localization depends on cell culture conditions. Manip-
genes at the transcriptional, translational, and cellular localization level. Changing cell culture condition pa-
rameters led toexpression ofREX1, potentiationof expressionof LIN28, translation of OCT4, SOX2, andNANOG,
and translocation of these transcription factors to the cell nucleus. We also show that culture conditions affect the
in vitro lifespan of dermal fibroblasts, nearly doubling the number of population doublings before the cells reach
replicative senescence. Our results suggest that it is possible to induce and manipulate endogenous expression of
stem cell genes in somatic cells without genetic manipulation, but this short-term induction may not be sufficient
for acquisition of true pluripotency. Further investigation of the factors involved in inducing this response could
forced expression by transgenesis, thus eliminating the risk of mutagenic effects due to genetic manipulation.
and embryonic carcinoma cells (Taranger et al., 2005), expo-
sure to extracts from pluripotent cell types (Byrne et al., 2003;
Miyamoto et al., 2007; Tada et al., 2001), and most recently
viral transduction of stem cell genes coding for transcription
factors (Maherali et al., 2007; Meissner et al., 2007; Takahashi
et al., 2007b; Yu et al., 2007) have demonstrated that the de-
It has been shown that transfection of genes coding for tran-
scription factors into fibroblasts may be sufficient to induce
pluripotency in both mouse (Takahashi et al., 2007a) and
human (Takahashi et al., 2007b; Yu et al., 2007) fibroblasts.
This achievement has groundbreaking implications for cell
uclear transplantation into oocytes (Campbell
et al., 2007), hybrid formation with embryonic stem cells
therapy research because cells with nearly equivalent differ-
entiation potential as embryonic stem cells (ESC) could be
of OCT4, SOX2, NANOG, and LIN28 appears to be sufficient
to set in motion a cascade of molecular events leading to ac-
quisition of pluripotency with characteristics of ESCs (Yu
et al., 2007). Transfection of OCT4 and SOX2 was absolutely
required to achieve this transformation, whereas the addition
of NANOG was beneficial, but not essential and alone not
capable of achieving pluripotency (Yu et al., 2007). Interest-
ingly, Takahashi et al. (2007b) showed that pluripotency
could be induced in mouse fibroblasts by transduction with a
different set of four factors, but included OCT4 and SOX2.
Recently, pluripotency was achieved by addition of the his-
and SOX2 expressing trasngenes (Huangfu et al., 2008).
1CellThera, Inc., Worcester, Massachusetts.
2Department of Biology and Biotechnology,3Department of Biomedical Engineering,4Bioengineering Institute, Worcester Polytechnic
Institute, Worcester, Massachusetts.
CLONING AND STEM CELLS
Volume 11, Number 3, 2009
ª Mary Ann Liebert, Inc.
Maintenance of inherent (ESC) or induced [induced plu-
ripotent stem (iPS)] pluripotency in human cells depends on
continuous presence of FGF2 (Levenstein et al., 2006), as its
withdrawal leads to spontaneous cell differentiation (Diecke
et al., 2008). FGF2 induces differential expression of members
of the transforming growth factor (TGF)-b family of proteins.
Upregulation of TGF-b1, activin A (TGF-b receptor and ALK
receptor ligand, respectively) and gremlin1 (BMP4 inhibitor),
and downregulation of BMP4 leads to SMAD2=3-driven ex-
pression of OCT4, SOX2, and NANOG (Suzuki et al., 2006).
These three transcription factors co-occupy promoters of
several genes, including FGF2. Increased expression of FGF2
completes this autoregulatory loop that is perpetuated in the
presence of exogenous FGF2. Interestingly, a very similar ef-
fect of FGF2 on expression of the TGF-b family members was
observed in fibroblasts used to support growth and mainte-
nance of hESCs (Greber et al., 2007a). In addition to supple-
mentation with FGF2, reducing oxygen concentration during
culture is becoming increasingly more appreciated in hESC
laboratories (Ezashi et al., 2005). Adult and embryonic stem
cells cultured in a reduced oxygen atmosphere have been
shown to maintain their undifferentiated state more effi-
ciently andincrease the efficiency oftheir differentiation upon
induction (Chakravarthy et al., 2001; Covello et al., 2006; Fink
et al., 2004; Forsyth et al., 2006; Grayson et al., 2006, 2007;
Pistollato et al., 2007). However, the effects of a combination
of these easily controllable factors on the phenotype of cul-
tured human adult fibroblasts have not been investigated.
Therefore, we sought to evaluate the potential effects of these
factors on the in vitro life span of primary human fibroblasts,
and on the expression and localization of genes often associ-
ated with pluripotency.
Materials and Methods
Primary adult human dermal fibroblasts from connective
tissue isolated from tissue biopsy from a below-knee ampu-
tation of a 24-year-old male (CRL-2352) were obtained from
American Tissue Culture Collection (ATCC; Manassas, VA)
at passage 2. Cells were cultured in medium consisting of
DMEM:Ham’s F12 (60:40, MediaTech) with 10% Fetalclone III
(Hyclone, Logan, UT). The DMEM (without L-Gln or phenol
red) was supplemented with 4mM fresh L-Gln (MediaTech,
Manassas, VA) prior to use. Cultures were carried out in
a 378C incubator in a humidified environment of 5% CO2,
5% O2, and 90% N2. The number of population doublings
was calculated as log2(#final=#initial). Cells were seeded at
100,000cells per T25 flask at each passage (Falcon, Oxnard,
CA). When used, human recombinant FGF2 (Chemicon, Te-
mecula, CA, or Protide, Lake Zurich, IL) and BMP-2 (R&D
Systems, Minneapolis, MN) was supplemented into medium
at 4 and 1ng=mL, respectively. Human muscle fibroblasts
were derived from surplus muscle tissue from the calf flexor
muscle used for a surgical knee repair in a 59-year-old adult
male (CT0706). Tissue was rinsed in Leibowitz L-15 medium
(MediaTech) containing 10mg=mL gentamicin (Invitrogen,
Carlsbad, CA) and 2.5mg=mL fungizone (Hyclone) and
minced using a sterile scalpel and digested with 1800units=
mL collagenase Type IV for 1h at 378C. Cell lines were es-
tablished by culture at 5% O2from the beginning and work-
ing stocks cryopreserved at passage 2. Teratocarcinoma cells
(CRL-2073) were grown as recommended by the supplier
(ATCC). Human embryonic stem cells (H9, WiCell, Madison,
WI) were cultured on mitomycin C-treated mouse embryonic
fibroblasts seeded onto 0.1% gelatin coated six-well plates
using 80% Knockout? DMEM (Invitrogen), 20% Knockout?
serum replacement supplemented with 2.0mM L-Gln,
0.055mM 2-mercaptoethanol, and 4.0ng=mL FGF2, as re-
commended by the supplier.
Fibroblasts were seeded into 24-well plates (BD Falcon)
onto 12-mm round glass coverslips (VWR, West Chester, PA)
at 5000cells per well in medium consisting of DMEM with
4mM fresh L-Gln: Ham’s F12 medium (60:40) supplemented
with 1? TCH serum replacement (Protide Pharmaceuticals),
1X ITS-X (Invitrogen), 2% FetalClone III (Hyclone) and cul-
tured at 378C, 5% CO2, 5% O2and 90% N2. For FGF2 treat-
ments, FGF2 (4ng=mL) was added to the medium at the time
of seeding. After 7 days, the cells were washed with DPBS
w=o Ca=Mg (Mediatech) and fixed in methanol (?208C) for
10min, washed with DPBS, and stored in DPBS at 48C until
use. Cells were washed with phosphate-buffered saline
Hercules, CA) and blocked for 30min at room temperature
with PBS containing 0.05% Tween, 5% fetal bovine serum
(FBS), and 1% bovine serum albumin (BSA). Primary anti-
bodies (2.5mg=mL) were added in blocking solution for
30min at room temperature. Cells were washed four times in
PBS=Tween and Alexafluor-568 labeled appropriate second-
ary antibody(4mg=mL)inblockingsolution addedfor 30min.
Cells were washed four times in PBS=Tween and stored in
PBS at 48C until image analysis. Antibodies used were: OCT4
(Abcam, Cambridge, MA, ab19857), SOX2 (Abcam, ab15830),
NANOG (Abcam, ab21624), FGFR-1 (Abgent, San Diego, CA,
AP7636a), FGFR-2 (Abgent, AP7636a), FGFR-3 (Abcam
ab10651), FGFR-4 (Abcam ab 41948), FGF2 (Santa Cruz, Santa
Cruz, CA, sc-1390 or sc-79). Coverslips were removed and
mounted onto glass slides in 80% glycerol in PBS containing
0.1% Na-Azide and sealed with nail polish. Cells were visu-
alized using an Olympus IX81 inverted microscope with
epi-fluorescence using appropriate filters (Semrock, Inc.,
Rochester, NY) and phase contrast. Images were collected
using a 12 bit Hamamatzu CCD camera and processed using
Total protein was isolated from subconfluent fibroblasts
with RIPA cell lysis buffer (Santa Cruz Biotechnology), sup-
plemented with complete protease inhibitor cocktail (PIC,
Santa Cruz Biotechnology) and 1mM DTT. Lysates were in-
cubated on ice for 30min and vortexed every 10min. Lysates
were centrifuged at 13,000?g and supernatants stored at
?808C. Protein concentration was determined with Quant-iT
protein assay kit (Invitrogen). Equal amounts of protein su-
pernatant and denaturing 2?sample buffer (BioRad Labora-
separated on 4–20% gradient SDS-PAGE gels and trans-
ferred to nitrocellulose membranes (BioRad Laboratories)
using Towbins transfer buffer (25mM Tris, 192mM glycine,
20% methanol, and 0.037% SDS). The membranes were
blocked with Tween-Tris-buffered saline (TTBS: 25mM Tris,
418 PAGE ET AL.
137mM NaCl, 2.7mM KCl, 0.2% Tween), 5% dry milk (Santa
Cruz) and 5% FBS. The same buffer was used for primary and
secondary antibody incubations. Antibodies used were:
OCT4 (Abcam), SOX2 (Abcam), NANOG (Santa Cruz), and
fibroblast specific protein FSP (Sigma, St. Louis, MO). HRP-
conjugated secondary antibodies were used (Invitrogen). In
between antibody incubations, membranes were washed
(Santa Cruz Biotechnologies) and luminescence detected us-
ing a Kodak 4000MM Image Station. All images were ac-
quired after 30-sec exposure and processed using Kodak
Reverse Transcription PCR (RT-PCR)
Total RNA was isolated using Trizol (Invitrogen) follow-
ing the manufacturer’s protocol. Four micrograms of total
RNA was used to perform first strand cDNA synthesis using
Superscript (Invitrogen). For RT-PCR, 0.5mL of the first-
strand cDNA was used as a template. PCR was performed in
Mg2þfree PCR buffer (TaKara, Shiga, Japan) supplemented
with 1.5mM MgCl2, 200mM each of dNTPs, 25pmol each of
forward and reverse primers and 0.5U of TaKara ExTaq
polymerase per reaction. PCR cycling was done as follows:
initial denaturation at 958C for 2min, followed by 35 cycles
of denaturation at 958C for 15 sec; annealing at primer-spe-
cific annealing temperature for 1min; and extension at 728C
for 1:30min. Final extension was done at 728C for 10min and
the samples held at 48C until use. Amplification products
were resolved on 2% agarose gels containing 0.5mg=mL
ethidium bromide in 1?TAE buffer and photographed using
a Kodak 4000MM Image Station. Human OCT4 (POU5F1)
gene transcribes two mRNA variants (NM_002701 and
NM_203289), that translates to 360 amino acid and 265
amino acid proteins respectively. Of these, 225 amino acids
at the C-terminal are common to both isoforms. The RT-PCR
primers used in our study was specifically designed to am-
plify from the transcript for the 360 amino acid variant
(NM_002701). This variant has been shown to be expressed
by human ESCs, wheres the 265 amino acid variant was not,
suggesting that the former is important for maintaining
‘‘stemness’’ in human ESCs (Cauffman et al., 2006). RT-PCR
for the 265 aa variant in the dermal fibroblasts, including
human embryonic carcinoma cells (ATCC, CRL2073) did not
show any amplification. Primers used are listed in Table 1.
Quantitative reverse transcription-PCR (qRT-PCR)
RNA was extracted using TRIzol? reagent (Invitrogen)
according to the manufacturer’s protocol and quantified by
spectrophotometry. Two micrograms of RNA was subjected
to DNase I digestion, followed by a reverse transcription us-
ing a QuantiTech? Reverse transcription kit (Qiagen, Chats-
worth, CA) with a mixture of oligo-dT and random hexamers
primers. cDNA (50ng=well) was used as a template in qPCR
reactions with oligonucleotides specific for the genes of in-
terest (Table 2). A nontemplate control and an RNA sample
without reverse transcription for each sample were used to
control for potential contaminating DNA. All qPCR reactions
were performed in triplicate with the resultant values being
combined into an average cycle threshold (CT). The efficiency
of qPCR was calculated from the slope of a relative standard
Table 1. DNA Sequences Used for RT-PCR Primers
na, reverse primer spans the intron–exon junction so it will not amplify the genomic DNA.
STEM CELL GENE EXPRESSION IN FIBROBLASTS419
curve using GAPDH primers. Relative quantification was
determined using a 7500 Real Time PCR system (Applied
Biosystems, Bedford, MA) measuring SYBR green fluores-
cence (PerfeCTa? SYBR Green FastMix, Low ROX, Quanta
Biosciences, Gathersburg, MD). Expression profiles for the
mRNA transcripts are shown as inverted cycle threshold (CT)
relative to those of ACTIN. This was done by subtracting the
CTvalue from each gene from the total number of cycles run
(40), which enables the relative values to be plotted such that
genes detected at larger cycles are accurately represented as
being present in less abundance.
Severe combined immunodeficiency (SCID)
Animal studies were done with IACUC approved proto-
Worcester Polytechnic Institute, Worcester, MA. One million
of control and one million of FGF2-treated fibroblasts were
mixed with 8–12mm diameter carbon beads in sterile DPBS
and injected into the hind leg muscle of SCID mice (Charles
River Laboratories, Wilmington, MA). Animals were eutha-
nized 6 weeks after injection, the muscle excised, and pro-
cessed for histology. Tissues were fixed in 4% formaldehyde
in DPBS and embedded in paraffin. Sections were stained
with H&E and the injection site located by microscopic visu-
alization of the carbon beads.
Adult human fibroblasts grown in DMEM=F12, 10% serum
substitute (Fetal Clone III, Hyclone), with FGF2 (4ng=mL;
Chemicon) at 378C and 5% O2, 5% CO2, 90% N2cultured
continuously with a rigorously controlled passage schedule
underwent 70 population doublings (PDs) before reaching
senescence compared to 33 PDs for cells cultured in atmo-
spheric oxygen without FGF2 (Fig. 1). Cells cultured without
FGF2 in 5% O2underwent 50 PDs. The addition of FGF2
and culture with atmospheric oxygen resulted in 50 PDs.
The increase over the expected 33 PDs (according to the cell
supplier) with FGF2 supplementation was accompanied by
change in morphology to smaller cells with a more spindle
Table 2. DNA Sequences Used for qRT-PCR Primers
geneForward primerReverse primer
Days in Culture
5% O2 (No FGF2)
5% O2 (FGF2)ATM O2 (No FGF2)
ATM O2 (FGF2)
saged at regular intervals and seeded at the same density. ATM, atmospheric oxygen.
Adult human dermal fibroblasts (CRL-2352) were grown with (FGF2) or without 4ng=mL FGF2 (no FGF2), pas-
420 PAGE ET AL.
shape. The FGF2=low oxygen-treated fibroblasts maintained
normal karyotype at 65 PDs (44 XY) and entered replicative
senescence as indicated by cessation of further cell division
and b-galactosidase staining (not shown). These data indi-
cated both an individual and synergistic role for both reduced
oxygen and FGF2 supplementation for increasing cell life
span, which prompted us to investigate expression of genes
related to undifferentiated cells.
RT-PCR using primers designed to recognize embryonic
forms of transcripts for the stem cell genes (Table 1) was
performed on day 7 of the initial culture to obtain control
baseline values for the cells’ transcriptional activity. Contrary
to the expected absence of stem cell gene transcripts, RT-PCR
amplified detectable amounts of OCT4, NANOG, and KLF4
mRNA in fibroblasts grown under 5% O2. Transcription of
these genesdidnotchange whenculturesweresupplemented
with FGF2 and the amplified transcripts were of the same size
as those present in human teratocarcinoma cell controls (Fig.
2). Other stem cell genes SOX2, REX1, and LIN28 showed
transcriptional dependence on FGF2 (Fig. 2). Neither lowered
oxygen nor FGF2 supplementation had an obvious effect on
expression of OCT4, NANOG, KLF4, or hTERT, when com-
pared to GAPDH controls (Fig. 2). To quantify and validate
these observations, quantitative real-time PCR (qRT-PCR)
was employed and the analysis confirmed our RT-PCR data
(Fig. 3). The quantitation of mRNA expression from genes
associated with pluripotency revealed that normal human
fibroblasts have a basal expression of these genes (OCT4,
SOX2, NANOG, and hTERT), albeit much lower than that
expressed in hESCs (Fig. 3b).. However, REX1 and LIN28
transcript levels were induced by culture in reduced oxygen
with additional upregulation of LIN28 in the presence of
in hTERT mRNA, which was not further affected by FGF2
line (Fig. 3c). Primary muscle-derived human fibroblasts
(CT0706) contained detectable amounts of NANOG, OCT4,
and hTERT mRNA, which were not affected by reduced O2or
low O2with FGF2. In contrast to the dermal fibroblasts, REX1
To evaluate the potential significance of the presence
of OCT4, SOX2, and NANOG mRNA in fibroblasts and to
determine whether the messages were being translated, we
performed a Western blot analysis. OCT4, NANOG, and
SOX2 proteins were detected only in the presence of FGF2
(Fig. 4a). Their appearance was not a consequence of tran-
scriptional upregulation as the levels of their mRNA re-
mained unchanged in control untreated fibroblasts and did
not increase after treatment (Fig. 3). In order to perform their
function as transcription factors, OCT4, SOX2, and NANOG
need to translocate to the nucleus. Immunocytochemistry re-
vealed that when cultured on untreated glass coverslips in the
presence of FGF2, OCT4, SOX2, and NANOG antigens were
detected in the nucleus (Fig. 4b) and identical cultures on
tissue culture plastic did not show this pattern of expression
(Fig. 4b). Preliminary observations show that NANOG and
SOX2 are colocalizing in the nuclei and the perinuclear re-
gion of expressing cells (Fig 4c). Not all of the cells within the
FGF2-treated population responded by detectable nuclear
expression of OCT4, SOX2, and NANOG. Expressing cell
populations were clustered in discrete groups, indicating that
distinct subpopulations of adult human dermal fibroblasts
may be more susceptible to these changes. These observations
were reproduced in an independent primary fibroblast cell
population isolated from adult muscle tissue (CT0706).
The unexpected detection of apparent translation and nu-
clear translocation of OCT4, SOX2, and NANOG in fibro-
blasts due to FGF2 prompted us to evaluate the expression
of FGF2 itself and its receptors in cells cultured under these
conditions. Supplementation of FGF2 had no significant ef-
fecton transcription ofFGFreceptors orFGF2itselfregardless
of the culture conditions (Fig. 5a). However, immunocyto-
chemistry showed that in cells supplemented with FGF2,
FGF2 itself, and both FGFR1 and FGFR2 translocated to the
nucleus, whereas FGFR3 was detected in the nucleus in the
absence of FGF2 (Fig. 5b). FGF2 addition had no effect on
FGFR4 localization (Fig. 5b). Addition of BMP-2 (1ng=mL) to
the medium prevented FGF2-induced OCT4 nuclear detec-
tion (Fig. 6).
Recent efforts in derivation of iPS cells have demonstrated
that a relatively small number of genes when introduced into
differentiated somatic cells have the ability to reprogram
nuclear memory and cause acquisition of a pluripotent phe-
notype. These observations suggest that there may be only a
few prime upstream regulators of the stability of differentia-
tion phenotype, particularly in fibroblasts. We show that by
manipulating culture conditions alone, we can achieve chan-
ges in fibroblasts that would be beneficial in development
of LIN28, REX1, OCT4, NANOG, KLF4, SOX2, TERT, and
GAPDH in fibroblasts grown with (FGF2) or without FGF2
(no FGF2), teratocarcinoma cells (TC), and no template
control. No reverse transcriptase controls were negative (not
shown). Assays were performed after 7 days of culture. No
FGF2 and FGF2 lanes contain fibroblasts grown under 5% O2
Expression of stem cell genes in fibroblasts. RT-PCR
STEM CELL GENE EXPRESSION IN FIBROBLASTS 421
2352, c,d—CT0706). Primary fibroblasts grown in ambient O2conditions were used as control cells. Experimental fibroblasts
were grown at 5% O2with or without 4ng=mL FGF2 (5% O2and 5% O2þFGF2, respectively), and compared to human
embryonic stem cells H9 (hES). No template and no reverse transcriptase controls were performed under identical PCR
conditions. Assays were performed after 7 days of culture. Error bars represent standard error of measurement (SEM).
qRT-PCR results comparing relative gene expression levels in two primary human fibroblast cell lines (a,b—CRL-
from control or FGF2-treated fibroblasts. Western blots were probed with primary antibodies against OCT4 (Abcam), SOX2
(Abcam), and NANOG (Santa Cruz), and detected with HRP-conjugated secondary antibodies (Invitrogen). Fibroblast
specific protein FSP (Sigma) was used as a loading control. All cultures were grown under 5% O2. (b) For immunocyto-
chemistry, control (no FGF2), or FGF2-treated fibroblasts were labeled with primary rabbit antibodies against OCT4 (Abcam),
SOX2 (Abcam), and NANOG (Abcam). Alexafluor-568 conjugated goat antirabbit antibody (Invitrogen) was used for de-
tection (red). Staining with secondary antibody alone was used as a negative control (2nd Ab). (c) FGF2-treated fibroblast
grown on glass were double stained with anti-SOX2 (rabbit polyclonal, Abcam) and anti-NANOG antibody (goat polyclonal;
Santa Cruz). SOX2 was detected with Alexafluor 568 conjugated goat antirabbit secondary antibody and NANOG was
detected with Alexafluor488 conjugated donkey antigoat secondary antibody (both from Invitrogen). The images were
overlaid to produce a composite (merged) image. Assays were performed after 7 days of culture. Images were captured using
identical capture settings and processed using Slidebook?.
Expression and localization of stem cell transcription factors in human fibroblasts. (a) Total protein was isolated
422PAGE ET AL.
of patient-specific cell therapy approaches. First, we demon-
strate an increase in the number of population doublings of
adult dermal fibroblasts in vitro; second, fibroblasts acquire
expression and nuclear localization of several stem cell spe-
cific transcription factors, and third, the cells maintain low
TERT levels and are not tumorigenic.
Experimental evidence gathered over the past decades for
involvement of FGF2 in a number of cellular and develop-
mental processes is extensive. Among others, FGF2 has been
shown to regulate human ES cell self-renewal (Levenstein
et al., 2006), is a potent mitogen and morphogen for a variety
of cell types (reviewed in Ornitz and Itoh, 2001), has been
blasts, and teratocarcinoma cells (TC). No template control was performed under identical conditions. (b) Immuno-
fluorescence staining of control (no FGF2) and FGF2-treated human adult dermal fibroblasts with antibodies against FGFR1
(Abgent), FGFR2 (Abgent), FGFR3 (Abcam), FGFR4 (Abcam), and FGF2 (Santa Cruz). Secondary antibodies were conjugated
to Alexafluor 568. Secondary antibody incubations alone were used as negative controls (control). Assays were performed
after 7 days of culture. Images were captured using identical capture settings and processed using Slidebook?.
Expression of FGF2 and its receptors in fibroblasts. (a) RT-PCR results in control fibroblasts, FGF2-treated fibro-
fluorescence staining of adult human fibroblasts for OCT4 (Abcam) cultured with FGF2 (4ng=mL) alone (a) and with FGF2
(4ng=mL) and BMP-2 (1ng=mL) (b). Secondary antibodies were conjugated to Alexafluor 568. Assays were performed after
7 days of culture. Images were captured using identical capture settings and processed using Slidebook?.
BMP2 prevents FGF2-induced expression and nuclear localization of OCT4 in dermal fibroblasts. Immuno-
STEM CELL GENE EXPRESSION IN FIBROBLASTS 423
shown to initiate regeneration in the eye (Hayashi et al.,
2004), and reprograms primordial germ cells to pluripotency
(Durcova-Hills et al., 2006).
The increase in in vitro longevity of fibroblasts cultured in
reduced oxygen were consistent with those reported previ-
ously (Saito et al., 1995). An increased life span has been as-
sociated with FGF2-facilitated selection and enrichment of
multipotent cells (Bianchi et al., 2003; Quarto and Longaker
2006); however, FGF2 has not been shown to impact the
in vitro life span of terminally differentiated cell types. The
initial FGF2 effect is likely mediated via FGF receptors at
the plasma membrane. Nuclear translocation of FGFR1 and
FGFR2 upon FGF2 treatment indicates a possibility for in-
volvement of both FGF2 and these receptors at the chromatin
level. FGF receptors have been observed in 3T3cells (Maher,
1996) and mammary epithelial cells (Bryant and Stow, 2005),
may act through a number of FGFR isoforms. Therefore,
the increased number of population doublings in our FGF2-
treated fibroblasts could be a consequence of FGF2 mitogenic
stimulation. The changes we observed in FGFR and FGF2
localization did not appear to coincide with an increase in
their mRNA levels. In addition to their mitogenic effects, both
FGFR1 and 2 have been associated with FGF2-mediated
maintenance of pluripotency in hES cells (Babaie et al., 2007;
Greber et al., 2007a). Although FGF2 has not been implicated
previously in transcriptional activation of OCT4 or SOX2,
it has been determined that the maintenance of expression
of these genes and cell pluripotency is dependent on FGF2
(Levenstein et al., 2006). The proposed action of FGF2 in-
volves induction of members of TGF-b pathway; TGF-b li-
gands maintain expression of OCT4, SOX2, and NANOG,
which in turn, activate expression of endogenous FGF2 that
completes this regulatory loop (Greber et al., 2007b). Our data
is in agreement with previous hypothesis that this circuit can
be initiated and perpetuated by exogenous FGF2, leading to
autocrine signaling by endogenous FGF2 (Dvorak et al., 2005;
Dvorak and Hampl, 2005; Greber et al., 2007a; Levenstein
et al., 2006; Xu et al., 2005). BMP signaling has been shown
to antagonize FGF2 signaling in maintaining the pluripotent
state of human ES cells (Xu et al., 2008). Inhibition of OCT4
nuclear detection in our culture system in the presence of
BMP-2 suggests that a similar pathway may be involved.
Recently, FGF2 has been shown to be involved in re-
modeling of the chromatin in rat cortical neuronal progenitor
cells by methylation of histone H3K4 (K4me3) and repression
of methylation of H3K9 (Song and Ghosh, 2004). Both of
these posttranslational histone H3 modifications have been
associated with transcriptionally active chromatin (Kimura
et al., 2004).
It was surprising that detectable levels of stem cell tran-
scription factors were present in control fibroblasts and that
at least one of these genes (OCT4) represented the true em-
bryonic form. Despite the presence of transcription factor
mRNAs, however, no detectable level of these proteins could
be found either by Western blotting or ICC. Although we
have not yet determined the functional relationship between
FGF2 and stem cell gene expression in dermal fibroblasts, it
has become apparent that a combination of FGF2 supple-
mentation, low oxygen culture conditions, and cell culture
surface triggered translation of these proteins and their ap-
propriate translocation to cells’ nuclei. Our results suggest
for the first time that alteration of cell fate may depend not
only on induction of new transcription, but on posttran-
scriptional regulation as well.
The absence of tumor formation after injection into SCID
mice indicates that despite stem cell gene expression, subse-
quent translation and appropriate nuclear localization after
7 days of culture, these cells have not yet acquired a plurip-
otent phenotype. Events downstream of OCT4, SOX2, and
NANOG that may be critically important for acquisition and
maintenance of pluripotency may require extended culture of
cells under appropriated conditions (Takahashi et al., 2007b;
Yu et al., 2007). Because only up to 30% of cells demonstrated
stem-like nuclear localization of the transcription factors, ab-
sence of tumor formation may have been due to insufficient
numbers of OCT4=NANOG=SOX2 positive cells injected.
The ability oflowered oxygentogether with FGF2toinduce
expression of stem cell genes in adult human fibroblasts
without hTERT protein expression and with a significantly
increased replication potential suggests that sufficient num-
bers of cells could be produced for therapeutic applications.
Our data suggests that mechanisms regulating translation and
posttranslational modifications may be critically important in
induction of a stem cell phenotype. This suggests that there
may be a subpopulation of fibroblasts capable of responding
to FGF2 at the translational or signaling level. In addition, the
culture conditions warrants further exploration.
The mechanism of induction of key regulatory genes in-
volved in pluripotency by nontransgenic methods to create
truly pluripotent cells will require further investigation. The
published studies on transgene induced pluripotency in fi-
broblasts show that forced expression of exogenous plur-
ofphenotypicchangesin thecellsfollowedbyamplification of
colonies of cells with truly pluripotent properties (Takahashi
et al., 2007b; Yu et al., 2007). The long-term stability of this
phenotype will likely involve introduction of extra cellular
components and specialized media formulations similar to
those employed for derivation and in vitro maintenance of
hESCs and IPS cells, and possibly factors yet to be identified.
However, this work suggests that it may be possible to de-
differentiate adult human somatic cells by modifying the
in vitro culture conditions. The ability to dedifferentiate so-
matic cells to a less differentiated (not necessarily pluripotent)
state by simply modifying the culture conditions may have
value in the utilization of autologous or primary cells for cell
therapy and diagnostic applications.
We thank Prof. Marsha Rolle and Prof. Charles Murry for
reviewing the manuscript. We thank Sharon Shaw for as-
sistance with histological analysis of injected muscle. We
thank Amanda Blackwood and Cara Messier from Prof. G.
Pins’ lab at WPI for providing primary human dermal fi-
broblasts used as controls. We thank Professor Raymond
Dunn and Dr. Ronald Ignotz for assistance with muscle tis-
sue. This work was supported by the WPI start up funds to
T.D., NIH R01GM085456, and by the Defense Advanced
Research Projects Agency and US Army Research Office. The
424PAGE ET AL.
content of the information does not necessarily reflect the
position or the policy of the Government, and no official
endorsement should be inferred. Any intellectual property
arising from this communication has been assigned to
Worcester Polytechnic Institute.
Author Disclosure Statement
The authors declare that no conflicting financial interests
Babaie, Y., Herwig, R., Greber, B., et al. (2007). Analysis of Oct4-
dependent transcriptional networks regulating self-renewal
and pluripotency in human embryonic stem cells. Stem Cells
Bianchi, G., Banfi, A., Mastrogiacomo, M., et al. (2003). Ex vivo
enrichment of mesenchymal cell progenitors by fibroblast
growth factor 2. Exp. Cell Res. 287, 98–105.
Bryant, D.M., and Stow, J.L. (2005). Nuclear translocation of cell-
surface receptors: lessons from fibroblast growth factor. Traffic
Byrne, J.A., Simonsson, S., Western, P.S., et al. (2003). Nuclei of
adult mammalian somatic cells are directly reprogrammed to
oct-4 stem cell gene expression by amphibian oocytes. Curr.
Biol. 13, 1206–1213.
Campbell, K.H., Fisher, P., Chen, W.C., et al. (2007). Somatic cell
nuclear transfer: past, present and future perspectives. Ther-
iogenology 68(Suppl. 1), S214–S231.
Cauffman, G., Liebaers, I., Van Steirteghem, A., et al. (2006).
POU5F1 isoforms show different expression patterns in hu-
man embryonic stem cells and preimplantation embryos. Stem
Cells 24, 2685–2691.
Chakravarthy, M.V., Spangenburg, E.E., and Booth, F.W. (2001).
Culture in low levels of oxygen enhances in vitro proliferation
potential of satellite cells from old skeletal muscles. Cell Mol.
Life Sci. 58, 1150–1158.
Covello, K.L., Kehler, J., Yu, H., et al. (2006). HIF-2alpha regu-
lates Oct-4: effects of hypoxia on stem cell function, embryonic
development, and tumor growth. Genes Dev. 20, 557–570.
Diecke, S., Quiroga-Negreira, A., Redmer, T., et al. (2008). FGF2
signaling in mouse embryonic fibroblasts is crucial for self-
renewal of embryonic stem cells. Cells Tissues Organs 188,
Durcova-Hills, G., Adams, I.R., Barton, S.C., et al. (2006). The
role of exogenous fibroblast growth factor-2 on the repro-
gramming of primordial germ cells into pluripotent stem cells.
Stem Cells 24, 1441–1449.
Dvorak, P., and Hampl, A. (2005). Basic fibroblast growth factor
and its receptors in human embryonic stem cells. Folia His-
tochem. Cytobiol. 43, 203–208.
Dvorak, P., Dvorakova, D., Koskova, S., et al. (2005). Expression
and potential role of fibroblast growth factor 2 and its recep-
tors in human embryonic stem cells. Stem Cells 23, 1200–1211.
Ezashi, T., Das, P., and Roberts, R.M. (2005). Low O2 tensions
and the prevention of differentiation of hES cells. Proc. Natl.
Acad. Sci. USA 102, 4783–4788.
Fink, T., Abildtrup, L., Fogd, K., et al. (2004). Induction of adi-
pocyte-like phenotype in human mesenchymal stem cells by
hypoxia. Stem Cells 22, 1346–1355.
Forsyth, N.R., Musio, A., Vezzoni, P., et al. (2006). Physiologic
oxygen enhances human embryonic stem cell clonal recovery
and reduces chromosomal abnormalities. Cloning Stem Cells
Grayson, W.L., Zhao, F., Izadpanah, R., et al. (2006). Effects of
hypoxia on human mesenchymal stem cell expansion and
plasticity in 3D constructs. J. Cell Physiol. 207, 331–339.
Grayson, W.L., Zhao, F., Bunnell, B., et al. (2007). Hypoxia
enhances proliferation and tissue formation of human mes-
enchymal stem cells. Biochem. Biophys. Res. Commun. 358,
Greber, B., Lehrach, H., and Adjaye, J. (2007a). Fibroblast growth
factor 2 modulates transforming growth factor beta signaling
in mouse embryonic fibroblasts and human ESCs (hESCs) to
support hESC self-renewal. Stem Cells 25, 455–464.
Greber, B., Lehrach, H., and Adjaye, J. (2007b). Silencing of core
transcription factors in human EC cells highlights the impor-
tance of autocrine FGF signaling for self-renewal. BMC Dev.
Biol. 7, 46.
Hayashi, T., Mizuno, N., Ueda, Y., Okamoto, M., et al. (2004).
FGF2 triggers iris-derived lens regeneration in newt eye.
Mech. Dev. 121, 519–526.
Huangfu, D., Osafune, K., Maehr, R., et al. (2008). Induction of
pluripotent stem cells from primary human fibroblasts with
only Oct4 and Sox2. Nat. Biotechnol.
Kimura, H., Tada, M., Nakatsuji, N., et al. (2004). Histone code
modifications on pluripotential nuclei of reprogrammed so-
matic cells. Mol. Cell Biol. 24, 5710–5720.
Levenstein, M.E., Ludwig, T.E., Xu, R.H., et al. (2006). Basic fi-
broblast growth factor support of human embryonic stem cell
self-renewal. Stem Cells 24, 568–574.
Maher, P.A. (1996). Nuclear translocation of fibroblast growth
factor (FGF) receptors in response to FGF-2. J. Cell Biol. 134,
Maherali, N., Sridharan, R., Xie, W., et al. (2007). Directly re-
programmed fibroblasts show global epigenetic remodel-
ing and widespread tissue contribution. Cell Stem Cell 1,
Meissner, A., Wernig, M., and Jaenisch, R. (2007). Direct repro-
gramming of genetically unmodified fibroblasts into pluripo-
tent stem cells. Nat. Biotechnol. 25, 1177–1181.
Miyamoto, K., Furusawa, T., Ohnuki, M., et al. (2007). Repro-
gramming events of mammalian somatic cells induced by Xe-
nopus laevis egg extracts. Mol. Reprod. Dev. 74, 1268–1277.
Ornitz, D.M., and Itoh, N. (2001). Fibroblast growth factors.
Genome Biol. 2, REVIEWS3005.
Pistollato, F., Chen, H.L., Schwartz, P.H., et al. (2007). Oxygen
tension controls the expansion of human CNS precursors and
the generation of astrocytes and oligodendrocytes. Mol. Cell
Neurosci. 35, 424–435.
Quarto, N., and Longaker, M.T. (2006). FGF-2 inhibits osteogen-
esis in mouse adipose tissue-derived stromal cells and sustains
their proliferative and osteogenic potential state. Tissue Eng.
Saito, H., Hammond, A.T., and Moses, R.E. (1995). The effect of
low oxygen tension on the in vitro-replicative life span of
human diploid fibroblast cells and their transformed deriva-
tives. Exp. Cell Res. 217, 272–279.
Song, M.R., and Ghosh, A. (2004). FGF2-induced chromatin
remodeling regulates CNTF-mediated gene expression and
astrocyte differentiation. Nat. Neurosci. 7, 229–235.
Suzuki, A., Raya, A., Kawakami, Y., et al. (2006). Nanog binds
to Smad1 and blocks bone morphogenetic protein-induced
differentiation of embryonic stem cells. Proc. Natl. Acad. Sci.
USA 103, 10294–10299.
STEM CELL GENE EXPRESSION IN FIBROBLASTS425
Tada, M., Takahama, Y., Abe, K., et al. (2001). Nuclear repro-
gramming of somatic cells by in vitro hybridization with ES
cells. Curr. Biol. 11, 1553–1558.
Takahashi, K., Okita, K., Nakagawa, M., et al. (2007a). Induction
of pluripotent stem cells from fibroblast cultures. Nat. Protoc.
Takahashi, K., Tanabe, K., Ohnuki, M., et al. (2007b). Induction
of pluripotent stem cells from adult human fibroblasts by
defined factors. Cell 131, 861–872.
Taranger, C.K., Noer, A., Sorensen, A.L., et al. (2005). Induction
of dedifferentiation, genomewide transcriptional program-
ming, and epigenetic reprogramming by extracts of carcinoma
and embryonic stem cells. Mol. Biol. Cell 16, 5719–5735.
Xu, R.H., Peck, R.M., Li, D.S., et al. (2005). Basic FGF and sup-
pression of BMP signaling sustain undifferentiated prolifera-
tion of human ES cells. Nat. Methods 2, 185–190.
Xu, R.H., Sampsell-Barron, T.L., Gu, F., et al. (2008). NANOG is
a direct target of TGFbeta=activin-mediated SMAD signaling
in human ESCs. Cell Stem Cell 3, 196–206.
Yu, J., Vodyanik, M.A., Smuga-Otto, K., et al. (2007). Induced
pluripotent stem cell lines derived from human somatic cells.
Science 318, 1917–1920.
Address correspondence to:
Raymond L. Page, Ph.D.
Biology and Biotechnology Department
Worcester Polytechnic Institute
100 Institute Road
Worcester, MA 01609
426PAGE ET AL.