Experimental Neurology 207 (2007) 267 274

www.elsevier.com/locate/yexnr

Adipose-derived stem cells differentiate into a Schwann cell phenotype and

promote neurite outgrowth in vitro
Paul J. Kingham a,, Daniel F. Kalbermatten a,b , Daljeet Mahay a , Stephanie J. Armstrong a ,
Mikael Wiberg b , Giorgio Terenghi a
a

Blond McIndoe Research Laboratories, The University of Manchester, Room 3.106 Stopford Building, Oxford Road, Manchester M13 9PT, UK
b
Departments of Surgical and Perioperative Science (Handsurgery) and Integrative Medical Biology (Anatomy), Umea University, Sweden
Received 31 March 2007; revised 11 June 2007; accepted 29 June 2007
Available online 2 August 2007

Abstract
Experimentally, peripheral nerve repair can be enhanced by Schwann cell transplantation but the clinical application is limited by donor site
morbidity and the inability to generate a sufficient number of cells quickly. We have investigated whether adult stem cells, isolated from adipose
tissue, can be differentiated into functional Schwann cells. Rat visceral fat was enzymatically digested to yield rapidly proliferating fibroblast-like
cells, a proportion of which expressed the mesenchymal stem cell marker, stro-1, and nestin, a neural progenitor protein. Cells treated with a
mixture of glial growth factors (GGF-2, bFGF, PDGF and forskolin) adopted a spindle-like morphology similar to Schwann cells.
Immunocytochemical staining and western blotting indicated that the treated cells expressed the glial markers, GFAP, S100 and p75, indicative of
differentiation. When co-cultured with NG108-15 motor neuron-like cells, the differentiated stem cells enhanced the number of NG108-15 cells
expressing neurites, the number of neurites per cell and the mean length of the longest neurite extended. Schwann cells evoked a similar response
whilst undifferentiated stem cells had no effect. These results indicate adipose tissue contains a pool of regenerative stem cells which can be
differentiated to a Schwann cell phenotype and may be of benefit for treatment of peripheral nerve injuries.
2007 Elsevier Inc. All rights reserved.
Keywords: Adult stem cell; Axon; Glia; Peripheral nerve; Regeneration; Schwann cell

Introduction
Peripheral nerve injuries are a common occurrence and
represent a major economic burden for society (Wiberg and
Terenghi, 2003). Treatment usually involves direct endend
surgical repair of the damaged nerves for minor defects whereas
autologous nerve grafts are required for longer gaps (Lundborg,
2000). Despite good surgical advances, functional recovery is
often poor. Tissue engineering techniques which enhance the
beneficial endogenous responses to nerve injury could provide
an alternative repair strategy.
Schwann cells play a pivotal role in peripheral nerve
regeneration (Ide, 1996) and are thus an attractive therapeutic
target. Nerve injury disrupts the normal Schwann cellaxon
Corresponding author. Fax: +44 161 275 1814.
E-mail address: paul.j.kingham@manchester.ac.uk (P.J. Kingham).
0014-4886/$ - see front matter 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.expneurol.2007.06.029

interaction resulting in dedifferentiation of the Schwann cells

and activation of a growth promoting phenotype (Hall, 2005).
Proliferating Schwann cells release neurotrophic factors (Frostick et al., 1998) and form the bands of Bngner to direct
regenerating axons across the lesion. When seeded in artificial
nerve conduits, Schwann cells have been shown to enhance
nerve regeneration (Li et al., 2006; Mosahebi et al., 2002;
Rutkowski et al., 2004). However, cultured Schwann cells have
limited clinical application. The requirement for nerve donor
material evokes additional morbidity and the time required to
culture and expand the cells would delay treatment. Instead, the
ideal transplantable cell should be easily accessible, proliferate
rapidly in culture and successfully integrate into host tissue with
immunological tolerance (Tohill and Terenghi, 2004).
Mesenchymal stem cells (MSC) are an attractive cell source
for the regeneration of nerve tissue as they are able to self-renew
with a high growth rate and possess multi-potent differentiation

268

P.J. Kingham et al. / Experimental Neurology 207 (2007) 267274

properties (Pittenger et al., 1999). There is also some evidence

that MSC may be non-immunogenic or hypo-immunogenic
(Barry and Murphy, 2004). MSC fibroblast-like cells can be
isolated from the stromal cell population found in a number of
tissues (Barrilleaux et al., 2006). Bone marrow has been used as
the main source of MSC and under appropriate conditions we
(Caddick et al., 2006; Tohill et al., 2004) and other groups
(Dezawa et al., 2001; Keilhoff et al., 2006) have shown they can
be selectively differentiated into Schwann cells. However, the
harvest of bone marrow MSC is a highly invasive and painful
procedure and alternative sources from which to isolate MSC
should be investigated.
In the last few years, adipose tissue has been identified as
possessing a population of multi-potent stem cells (Gimble and
Guilak, 2003; Strem et al., 2005) to which we assign the generic
nomenclature, adipose-derived stem cells (ADSC). The phenotypic and gene expression profiles of ADSC are similar to MSC
obtained from bone marrow (De Ugarte et al., 2003a,b; Strem
et al., 2005) and these cells can be expanded in culture for
extended periods (Zuk et al., 2002). Humans have abundant
subcutaneous fat deposits and ADSC can easily be isolated by
conventional liposuction procedures, thus overcoming the
tissue morbidity associated with bone marrow aspiration.
Furthermore the frequency of MSC in bone marrow is between
1 in 25,000 and 1 in 100,000 cells (Banfi et al., 2001; D'Ippolito
et al., 1999; Muschler et al., 2001) whereas ADSC constitute
approximately 2% of lipoaspirate cells (Strem et al., 2005).
The apparent advantages of ADSC have led us to investigate
whether they can be differentiated to a Schwann cell phenotype
which could ultimately provide functional benefits for peripheral
nerve repair. Animal models are critical to the development of
applications utilising ADSC so in this study we have used stem
cells isolated from rats. Since rats possess little subcutaneous fat,
we used adipose tissue from the abdominal cavity. Previous
studies have shown that stem cells isolated from rat visceral fat
mimic the differentiation process of human ADSC and can adopt
adipocyte, osteoblast, chondrocyte and neural phenotypes
(Tholpady et al., 2003).
Materials and methods
Isolation and culture of adipose-derived stem cells (ADSC)
ADSC were isolated from adult Sprague-Dawley rats euthanized by a schedule 1 method according to the UK Animal
Scientific Procedures Act 1986. Visceral fat encasing the stomach
and intestines was carefully dissected and minced using a sterile
razor blade. Tissue was then enzymatically dissociated for 60 min at
37 C using 0.15% (w/v) collagenase type I (Invitrogen, UK). The
solution was passed through a 70-m filter to remove undissociated
tissue, neutralized by the addition of Modified Eagle Medium (MEM; Invitrogen, UK) containing 10% (v/v) foetal bovine serum
(FBS) and centrifuged at 800g for 5 min. The stromal cell pellet
was resuspended in MEM containing 10% (v/v) FBS and 1% (v/v)
penicillin/streptomycin solution. Cultures were maintained at subconfluent levels in a 37 C incubator with 5% CO2 and passaged
with trypsin/EDTA (Invitrogen, UK) when required.

Other cell culture

Bone marrow-derived stem cells were harvested from adult
rat femoral bones (Caddick et al., 2006) and maintained under
the same conditions as ADSC. Schwann cells were isolated from
the sciatic nerves of 1- to 2-day-old rat pups (Caddick et al.,
2006) and cultured in Dulbecco's modified Eagle's medium
(DMEM) containing 10% (v/v) FBS, 1% (v/v) penicillin/
streptomycin, 10 M forskolin (Sigma, UK) and 63 ng/ml
glial growth factor-2 (GGF-2; Acorda Therapeutics Inc., USA).
The NG108-15 cell line was purchased from ECACC (Porton
Down, UK) and was maintained in DMEM growth medium.
3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide
(MTT) cell proliferation assay
At either passage 1 or passage 5, cells were trypsinised and
plated at a density of 5000 cells/35 mm2 dish. Cells were allowed
to settle for 1 h after which time MTT was added (1 mg/ml final
concentration) for a period of 2 h. The resulting formazan
precipitate was then solubilised using 20% (v/v) Triton X-100
and the absorbance at 570 nm measured. This value was
recorded as the baseline and further measurements were taken
every 24 h.
Characterisation of stem cell properties
At passage 1, sub-confluent cultures were treated with agents
to induce the phenotype of mesoderm-derived cells. For
osteogenic induction, cultures were treated with 100 g/ml
ascorbate, 0.1 M dexamethasone and 10 mM -glycerophosphate for a period of 3 weeks. Cells were then fixed with 4%
(w/v) paraformaldehyde for 30 min at room temperature,
washed 3 times with phosphate-buffered saline (PBS) containing 1% (w/v) bovine serum albumin and then incubated with
1% (w/v) Alizarin Red solution to stain for calcium deposition.
For induction of a chondrocyte phenotype, cells were treated
with 0.1 M dexamethasone, 50 g/ml ascorbate, 10 ng/ml
TGF-1, 40 g/ml proline and 1% (v/v) ITS + premix (BD
Biosciences, UK) for 3 weeks. Cells were then fixed with 10%
(v/v) formalin for 60 min, washed in H2O and stained for
proteoglycan with 1% (w/v) toluidine blue.
For immunocytochemical assessment of stem cell markers,
ADSC were cultured on slide flasks (Nunc-Fisher Scientific, UK)
for 24 h and then fixed in 4% (w/v) paraformaldehyde for 20 min.
Fixative was removed and cells washed 1 10 min in PBS and
permeabilised using 0.2% (v/v) Triton X-100 for 20 min. The
cells were washed 2 10 min in PBS and 5% (v/v) normal goat
serum blocking solution (Sigma, UK) was added for 1 h at room
temperature. Monoclonal stro-1 (1:50; R&D Systems, UK) or
nestin (1:500; Chemicon, USA) antibodies were added and
incubated at 4 C overnight. Cells were washed 3 10 min in PBS
and goat anti-mouse Cy3-labelled secondary antibody (1:200;
Amersham Biosciences, UK) added for 1 h at room temperature.
The cells were washed 3 10 min in PBS and slides mounted
using an anti-fading Vectashield solution containing DAPI
(Vector labs, UK). Slides were examined using an Olympus

P.J. Kingham et al. / Experimental Neurology 207 (2007) 267274

IX51 inverted fluorescence microscope and the number of

immuno-positive cells counted from a minimum total of
100 cells per experiment.

269

posure to Kodak X-OMAT light-sensitive film. Antibody was

stripped from the membranes using 100 mM glycine pH 2.9
and the blots re-probed with -tubulin antibody (1:1000;
Abcam, UK) as a loading control.

Differentiation to a Schwann cell phenotype

Stem cell and NG108-15 neuron co-culture
Growth medium was removed from sub-confluent ADSC
cultures at passage 2 and replaced with medium supplemented
with 1 mM -mercaptoethanol (Sigma-Aldrich, UK) for 24 h.
Cells were then washed and fresh medium supplemented with
35 ng/ml all-trans-retinoic acid was added. A further 72 h later,
cells were washed and medium replaced with differentiation
medium; cell growth medium supplemented with 5 ng/ml
platelet-derived growth factor (PDGF; PeproTech Ltd., UK),
10 ng/ml basic fibroblast growth factor (bFGF; PeproTech Ltd.,
UK), 14 M forskolin and 252 ng/ml GGF-2. Cells were
incubated for 2 weeks under these conditions with fresh medium added approximately every 72 h.
Immunocytochemistry and western blotting
Undifferentiated (uADSC) and differentiated (dADSC)
cultures were trypsinised and replated on slide flasks for
immunostaining as above. Cells were incubated with mouse
anti-glial fibrillary acidic protein (GFAP; 1:200; Chemicon,
USA), rabbit anti-S100 (1:500; Dako, Denmark) and rabbit antip75 (1:500; Promega, USA) overnight at 4 C. Washed slides
were then treated with goat anti-rabbit FITC- (1:100; Vector
Labs, UK) and goat anti-mouse CY3 (1:200; Amersham, UK)conjugated secondary antibodies. Slides were examined using
an Olympus IX51 inverted fluorescence microscope and the
number of immuno-positive cells counted in a minimum total of
100 cells per experiment.
For western blotting, individual lysates were prepared from
one 75-cm2 flask of confluent cultures. Cells were washed in
PBS and then scraped into buffer containing 100 mM PIPES,
5 mM MgCl, 20% (v/v) glycerol, 0.5% (v/v) Triton X-100,
5 mM EGTA and protease inhibitors (Sigma, UK). Lysates
were incubated for 15 min on ice and then subjected to 2
freezethaw cycles prior to analysis of protein content using a
commercially available protein assay kit (Bio-Rad, UK). 15 g
protein was prepared per sample, combined with Laemmli
buffer and denatured at 95 C for 5 min. Proteins were resolved
at 120 V on 15% (for S100) or 10% (for GFAP) sodium
dodecyl sulphatepolyacrylamide gels. Following electrophoretic transfer to nitrocellulose, membranes were blocked for
1 h in 5% (w/v) non-fat dry milk in TBSTween (10 mM Tris
pH 7.5, 100 mM NaCl, 0.1% (v/v) Tween), and then incubated
overnight at 4 C with either monoclonal anti-GFAP (1:200;
LabVison, USA), monoclonal anti-S100 (1:750; Chemicon,
UK) or polyclonal anti-p75 (1:250; Santa Cruz Biotechnology,
USA) antibodies. Following 6 5 min washes in TBSTween,
membranes were incubated for 1 h with HRP-conjugated
secondary antibodies [goat anti-mouse and goat anti-rabbit
(1:1000; Cell Signalling Technology, USA)]. Membranes were
washed as previously and treated with ECL chemiluminescent
substrate (Amersham, UK) for 1 min and developed by ex-

uADSC and dADSC were plated at a density of 10,000 cells

per slide flask and allowed to settle for 24 h. NG108-15 cells
were then added to the ADSC monolayer at a density of 1000
cells and the co-cultures maintained for a further 24 h followed
by fixation with 4% (w/v) paraformaldehyde for 20 min at room
temperature. Fluorescent immunocytochemistry (as above)
using mouse anti-neurofilament protein antibody (1:500;
BioMol, UK) incubated at 4 C overnight followed by
secondary labelling with goat anti-mouse Cy3 conjugate
(1:200; Amersham Biosciences, UK) was used to visualise
NG108-15 neurite outgrowth on the ADSC. Co-culture with
NG108-15 cells did not alter the differentiation status of ADSC
(as measured by expression of glial protein markers). Cultures
were examined using an Olympus IX51 inverted fluorescence
microscope and images captured using Image ProPlus software
(MediaCybernetics, Marlow, UK). The trace function was used
to determine the percentage of NG108-15 cells expressing

Fig. 1. Proliferation of cells isolated from adipose tissue and bone marrow. An
MTT assay was used to determine the growth rate of cells obtained from adipose
tissue () and bone ( ) plated at passage 1 (A) and passage 5 (B). Data are
expressed as the mean % increase S.E.M. in absorbance (570 nm). P b 0.05
significantly higher in adipose tissue derived cultures compared with those from
bone.

270

P.J. Kingham et al. / Experimental Neurology 207 (2007) 267274

Results
Characterisation of stem cell cultures
Rat visceral adipose tissue was enzymatically digested and
then centrifuged to isolate the stromal cell fraction from mature
adipocytes. After approximately 1 week in culture, cells from
the stromal fraction formed confluent fibroblast-like monolayers on 75-cm2 flasks. Cells were then trypsinised and plated
for MTT proliferation assays (Fig. 1). There was an apparent lag
phase in growth of cells up to 48 h after which time the rate of
proliferation expanded more rapidly. Cells isolated from

Fig. 2. Adipose tissue-derived cells exhibit properties of stem cells. (A) Cultures
were treated with agents to induce differentiation to cells of mesoderm origin.
Alizarin red-stained and toluidine blue-stained cells indicate cells of osteogenic
and chondrogenic lineages, respectively. Scale bar = 40 m. (B) Adipose tissuederived cells at passage 1 were stained with anti-stro-1 and anti-nestin antibodies
and CY3-conjugated secondary antibody (red). DAPI staining (blue) indicates
the total number of cells in the field and it was used to quantify the percentage of
cells positive for each antigen. Scale bar = 40 m. (C) Data are expressed as the
mean % positive S.E.M. P b 0.01 significantly higher levels of nestinpositive cells in adipose tissue-derived cultures compared with those from bone.

neurites and the number of neurites expressed per cell. For each
experiment neurite data were ordered according to length (m).
The longest neurite in each experiment was thus identified and
the mean of these calculated from 5 independent experiments.
An average of one hundred NG108-15 cell bodies was analysed
for each condition in each experiment.
Statistical analysis
Data are presented as mean S.E.M. from 4 to 5 independent
cell cultures. KruskalWallis one way ANOVA with Dunn's
comparison test was used to determine the statistical significance between data, p b 0.05; p b 0.01.

Fig. 3. Adipose-derived stem cells differentiate to a Schwann cell phenotype.

(A) Cultures of undifferentiated ADSC (uADSC) showed a flattened fibroblastlike morphology which adopt a spindle elongated shape characteristic of
Schwann cells (dADSC) upon treatment with glial growth factors. Scale
bar = 40 m. (B) Immunofluorescence staining indicated differentiated ADSC
(dADSC) expressed GFAP, S100 and p75 proteins. Scale bar = 30 m. (C)
Quantitative analysis of morphology and expression of GFAP, S100 and p75
proteins (data are mean % cells S.E.M.).

P.J. Kingham et al. / Experimental Neurology 207 (2007) 267274

Fig. 4. Western blot analysis of Schwann cell proteins. Undifferentiated ADSC

(uADSC), differentiated ADSC (dADSC) and Schwann cell (SC) lysates were
blotted for GFAP, S100 and p75 proteins. -Tubulin was used as a loading
control.

271

expressed GFAP (Fig. 3C). Almost all of these GFAP-positive

cells also stained for S100 protein (42.9 3.3% positive). p75
expression was occasionally observed in uADSC but was readily
apparent in the cultures treated with glial growth factors (36.38
3.3% positive). A small fraction (16.8 1.4%) of treated ADSC
retained a fibroblast-like morphology, some of which expressed
GFAP (5.9 1.3%), S100 (3.4 1.0%) and p75 (1.23 0.38%)
proteins (Fig. 3C). A minority of cells (1.7 0.5%) displayed a
rounded cell body with multiple processes.
To confirm the results obtained by immunocytochemistry
were not due to an artefact of cellular shrinkage, western
blotting was performed (Fig. 4). Lysates of dADSC but not
uADSC showed a GFAP-immunoreactive band corresponding
to a molecular weight of 55 kDa. This was present in Schwann
cell lysates together with an additional lower band which is
likely to represent a proteolytic fragment or alternate transcript.

bone marrow exhibited a similar growth pattern; however, the

overall proliferation rate of cells taken from adipose tissue was
significantly faster in passage 1 cultures (Fig. 1A). Although the
growth rate of passage 5 adipose cells in the lag phase was not
different from bone-derived cells, they still proliferated significantly faster after 48 h (Fig. 1B).
In order to determine whether the cells isolated from adipose
tissue exhibited properties of mesenchymal stem cells, they
were treated with agents known to induce differentiation to cells
originating from the mesoderm. Osteogenic differentiation was
confirmed by the production of calcium deposits detected with
Alizarin Red (Fig. 2A) and chondrocyte differentiation by the
presence of toluidine blue-positive proteoglycans (Fig. 2A). We
also examined the passage 1 cultures for the presence of the
stem cell marker, stro-1. A small proportion (11.38 0.87%) of
cells were positive for stro-1 in adipose cell cultures, a similar
number to that found in bone marrow cultures (Figs. 2B and C).
Since we were interested in deriving cells of a neural lineage we
also determined whether the cells expressed nestin, a putative
marker of neural progenitors (Dahlstrand et al., 1995). There
were approximately three times more nestin-positive cells in
cultures of cells taken from adipose tissue compared with those
from bone (Fig. 2C; 14.6 0.8% vs. 5.1 1.6%, P b 0.01).
Differentiation to a Schwann cell phenotype
Having determined our cultures contained a population of
stem-like cells, we assign the term adipose-derived stem cell
(ADSC) to describe them. ADSC at passage 2 were treated with a
mixture of glial growth factors for a period of 2 weeks after which
time they were analysed morphologically and for the expression
of the Schwann cell proteins, GFAP, S100 and p75. Cells cultured
in the differentiation media changed from a fibroblast-like
morphology to an elongated spindle shape (Fig. 3A), similar to
that of Schwann cells. Neither GFAP nor S100 protein was
detected in undifferentiated cultures (uADSC) (data not shown)
but both were expressed in differentiated (dADSC) cells
(Fig. 3B). Quantitative analysis indicated that 81.5 1.5% of the
cells adopted a spindle-like morphology of which 44.5 3.7%

Fig. 5. Differentiated ADSC promote neurite outgrowth in NG108-15 cells. (A)

NG108-15 cells were grown alone (con NG108-15) or on a monolayer of
undifferentiated (NG108-15 + uADSC) or differentiated (NG108-15 + dADSC)
stem cells. Cultures were stained with neurofilament antibody (red) to visualise
neurites and DAPI (blue) to highlight individual cells. Scale bar = 60 m.
(B) The percentage of NG108-15 cells expressing neurites, number of neurites
per cell and length of longest neurite were determined in NG108-15 cells grown
alone (con NG108-15) and co-cultures with stem cells (+uADSC, +dADSC)
and Schwann cells (+SC). Data are mean S.E.M. P b 0.05; P b 0.01
significantly different compared with NG108-15 grown alone.

272

P.J. Kingham et al. / Experimental Neurology 207 (2007) 267274

S100 and p75 proteins were also detected in dADSC but were
absent in uADSC.
Functional properties of differentiated cells
The ability of ADSC to promote neurite outgrowth was
determined by examining their interaction with NG108-15 cells, a
motor neuron-like cell line (Jiang et al., 2003). uADSC and
dADSC were plated on slide flasks to form monolayers and then
NG108-15 cells were added (Fig. 5A). Computerised image
analysis of co-cultures after 24 h was used to quantify three
separate parameters: percentage of cells extending neurites,
number of neurites per cell and length of longest neurite (Fig. 5B).
Comparisons were made with control cultures of NG108-15 cells
grown alone and NG108-15 cells seeded with Schwann cells. A
small fraction (22.0 2.5%) of control NG108-15 cells extended
neurites which was significantly increased to 69.1 4.1%
(P b 0.05) and 57.6 5.3% (P b 0.05) in the presence of dADSC
and Schwann cells respectively. uADSC had no significant effect.
Likewise the number of neurites extended per cell was significantly (P b 0.05) increased in co-cultures of NG108-15 cells
with dADSC or Schwann cells, when compared with NG108-15
cells grown alone. The mean longest neurite extended by control
cultures of NG108-15 cells was 67.5 7.5 m and in co-culture
with uADSC it was 74.3 9.7 m. In contrast, dADSC evoked a
significant (P b 0.01) increase in neurite length to 205.2 2.7 m
and Schwann cells stimulated an increase to 309.8 31.5 m
(P b 0.01).
Discussion
We have recently shown that MSC derived from bone
marrow can undergo differentiation to a Schwann cell phenotype
(Caddick et al., 2006; Tohill et al., 2004). Given the clinical
advantages of adipose tissue as an alternative source of stem
cells (Gimble and Guilak, 2003; Strem et al., 2005), we have
now investigated whether it is also possible to derive Schwann
cells from adipose tissue. We found that rat ADSC treated with a
mixture of glial growth factors expressed GFAP, S100 and p75
proteins and enhanced neurite outgrowth in vitro, suggesting
transition to a Schwann cell phenotype.
As part of an initial characterisation of our cultures, we
compared the growth rate of cells isolated from adipose tissue and
bone marrow. We found that ADSC proliferated significantly faster
than bone MSC. This is consistent with a recent study comparing
various sources of rat MSC (Yoshimura et al., 2007) and if
translated to human studies could mean a reduction in the time
required to generate a therapeutically useful stock of cells. In order
to determine the stem-ness of our cells we investigated the
expression of stro-1. In contrast to the study by Ning et al. (2006)
which identified all ADSC as stro-1 positive, only a small fraction
of our cultures expressed this marker. This apparent discrepancy
has also been reported by different groups examining human
ADSC (De Ugarte et al., 2003a,b; Gronthos et al., 2001) and might
reflect a difference in the region from which the tissue was obtained.
Nestin is a protein commonly used to identify proliferating
adult neural progenitor cells in the central nervous system

(Dahlstrand et al., 1995). More recently nestin expression has

been observed in bone marrow MSC (Caddick et al., 2006;
Wislet-Gendebien et al., 2004) and its importance in controlling
commitment to differentiation along the glial lineage has been
demonstrated (Wislet-Gendebien et al., 2004). Our results
showed that when compared with bone marrow MSC, a significantly greater proportion of ADSC expressed nestin protein.
Murine ADSC have also been shown to express low levels of
nestin which can be up-regulated upon neurogenic differentiation (Safford et al., 2002). These results suggest that ADSC are
not restricted towards specific mesodermal cell lineages and
rather they retain some ability for differentiation along a neuroglial lineage.
To investigate this, we exposed the ADSC to a differentiation
media we have previously used to induce Schwann cells from
bone marrow MSC (Caddick et al., 2006). This produced a
morphological change in the majority of the cells, toward an
elongating spindle phenotype, characteristic of Schwann cells.
Many of these cells also began to express the glial markers, GFAP,
S100 and p75. The co-expression of these proteins taken with the
morphological changes indicates we can produce high yields of
Schwann-like cells from ADSC. In contrast, previous reports
have shown that neural induction media converts ADSC to cells
of a neuronal morphology with co-expression of GFAP, S100 and
neuronal proteins including III-tubulin and neurofilament (Ning
et al., 2006; Safford et al., 2004). Krampera et al. (2007) recently
showed these changes were rapid and reversible, suggesting
against a specific, full differentiation process. To induce a more
selective Schwann cell differentiation, the authors co-cultured
various MSC with Schwann cells and this produced a long lasting
expression of PMP-20 and S100 proteins in the absence of other
CNS glial and neuronal markers (Krampera et al., 2007).
Interestingly, of all the MSC tested, those derived from adipose
tissue produced the best response. These effects could not be
induced by Schwann cell factors alone, indicating contact
between the two cell types was necessary (Krampera et al.,
2007). Whilst this suggests that some form of trans-differentiation
might occur if ADSC were transplanted at a nerve injury site in
vivo, the methodology does not provide a suitable approach for
the generation of clinically useful cells. Instead, we have used a
defined mixture of GGF-2, bFGF and PDGF, molecules which are
known to play a role in the differentiation and proliferation of
Schwann cells (Jessen and Mirsky, 1999; Li et al., 1998). These
molecules together with forskolin are responsible for the
induction of the glial protein expression we observed by
immunocytochemistry. These results were confirmed by western
blotting; arguing against the notion suggested by Lu et al. (2004)
that increased staining is merely the result of an increase in
antigen levels per unit area, due to disruption of the cytoskeleton.
We also tested the function of our differentiated ADSC using
a co-culture with the NG108-15 motor neuron-like cell line. We
found that differentiated ADSC evoked a similar response to
Schwann cells, in that they promoted neurite outgrowth and
elongation. NG108-15 cells have been shown to express low
levels of trkA (Fu et al., 1997) and GFR-1 (Lee et al., 2006),
receptors for nerve growth factor (NGF) and glial-derived
neurotrophic factor (GDNF), respectively, two proteins which

P.J. Kingham et al. / Experimental Neurology 207 (2007) 267274

are known to be released by Schwann cells (Frostick et al.,

1998). To date, there have been no reports of the expression of
neurotrophic factors produced by ADSC. However, a recent
study has shown that rat bone marrow cells constitutively
express NGF, GDNF and brain-derived neurotrophic factor
(BDNF) (Chen et al., 2007). Human MSC also express NGF
and BDNF although this is restricted to certain sub-populations
(Crigler et al., 2006). These and other unidentified factors
promoted neurite outgrowth and survival in a co-culture system
with SH-SY5Y cells (Crigler et al., 2006). In our model,
undifferentiated ADSC had no effect on neurite outgrowth
suggesting that the process of differentiation led to the upregulation of nerve growth factors. In on-going experiments to
identify the factors involved we have determined that
conditioned medium from dADSC increases the number of
NG108-15 cells producing neurites and neurite number per cell
but does not enhance neurite length (data not shown). Thus
direct contact (the model used in this study) between the cells is
required for neurite elongation and this suggests that dADSC
might also up-regulate extracellular matrix molecules such as
laminin or fibronectin, which have also been shown to be
expressed by rat bone marrow MSC (Chen et al., 2007).
Our results suggest that ADSC differentiated to a Schwann cell
phenotype might have a beneficial role for treatment of peripheral
nerve injuries. We have previously shown that only differentiated
bone marrow MSC, rather than untreated MSC, can stimulate
nerve regeneration (Tohill et al., 2004), a result in agreement with
other groups (Dezawa et al., 2001; Keilhoff et al., 2006). In
contrast, other studies indicate that undifferentiated bone marrow
MSC can also enhance regeneration and lead to improved motor
function (Chen et al., 2007; Pereira Lopes et al., 2006). This
suggests that some form of trans-differentiation might occur in
vivo as the result of local signals from injured Schwann cells and
axons. However, the long-term effect of undifferentiated cells
remains to be determined and therefore we will continue to study
the mechanisms of differentiation of ADSC, in the hope of
generating clinically useful cells, for the treatment of peripheral
nerve injuries.
Acknowledgments
This work was supported by grants from the UK Northwest
Regional Development Agency, The Wellcome Trust, the UK
and Swedish Medical Research Councils and the County of
Vasterbotten and the Aners Foundation. We also gratefully
acknowledge the continuing supply of GGF-2 from Acorda
Therapeutics, USA.
References
Banfi, A., Bianchi, G., Galotto, M., Cancedda, R., Quarto, R., 2001. Bone
marrow stromal damage after chemo/radiotherapy: occurrence, consequences and possibilities of treatment. Leuk. Lymphoma 42, 863870.
Barrilleaux, B., Phinney, D.G., Prockop, D.J., O'Connor, K.C., 2006. Review:
ex vivo engineering of living tissues with adult stem cells. Tissue Eng.
30073019.
Barry, F.P., Murphy, J.M., 2004. Mesenchymal stem cells: clinical applications
and biological characterization. Int. J. Biochem. Cell Biol. 36, 568584.

273

Caddick, J., Kingham, P.J., Gardiner, N.J., Wiberg, M., Terenghi, G., 2006.
Phenotypic and functional characteristics of mesenchymal stem cells
differentiated along a Schwann cell lineage. Glia 54, 840849.
Chen, C.J., Ou, Y.C., Liao, S.L., Chen, W.Y., Chen, S.Y., Wu, C.W., Wang,
C.C., Wang, W.Y., Huang, Y.S., Hsu, S.H., 2007. Transplantation of bone
marrow stromal cells for peripheral nerve repair. Exp. Neurol. 204,
443453.
Crigler, L., Robey, R.C., Asawachaicharn, A., Gaupp, D., Phinney, D.G., 2006.
Human mesenchymal stem cell subpopulations express a variety of neuroregulatory molecules and promote neuronal cell survival and neuritogenesis.
Exp. Neurol. 198, 5464.
D'Ippolito, G., Schiller, P.C., Ricordi, C., Roos, B.A., Howard, G.A., 1999.
Age-related osteogenic potential of mesenchymal stromal stem cells from
human vertebral bone marrow. J. Bone Miner. Res. 14, 11151122.
Dahlstrand, J., Lardelli, M., Lendahl, U., 1995. Nestin mRNA expression
correlates with the central nervous system progenitor cell state in many, but
not all, regions of developing central nervous system. Brain Res. Dev. Brain
Res. 84, 109129.
De Ugarte, D.A., Alfonso, Z., Zuk, P.A., Elbarbary, A., Zhu, M., Ashjian, P.,
Benhaim, P., Hedrick, M.H., Fraser, J.K., 2003a. Differential expression of
stem cell mobilization-associated molecules on multi-lineage cells from
adipose tissue and bone marrow. Immunol. Lett. 89, 267270.
De Ugarte, D.A., Morizono, K., Elbarbary, A., Alfonso, Z., Zuk, P.A., Zhu, M.,
Dragoo, J.L., Ashjian, P., Thomas, B., Benhaim, P., Chen, I., Fraser, J.,
Hedrick, M.H., 2003b. Comparison of multi-lineage cells from human
adipose tissue and bone marrow. Cells Tissues Organs 174, 101109.
Dezawa, M., Takahashi, I., Esaki, M., Takano, M., Sawada, H., 2001. Sciatic
nerve regeneration in rats induced by transplantation of in vitro
differentiated bone-marrow stromal cells. Eur. J. Neurosci. 14,
17711776.
Frostick, S.P., Yin, Q., Kemp, G.J., 1998. Schwann cells, neurotrophic factors,
and peripheral nerve regeneration. Microsurgery 18, 397405.
Fu, A.K., Ip, F.C., Lai, K.O., Tsim, K.W., Ip, N.Y., 1997. Muscle-derived
neurotrophin-3 increases the aggregation of acetylcholine receptors in
neuronmuscle co-cultures. Neuroreport 8, 38953900.
Gimble, J., Guilak, F., 2003. Adipose-derived adult stem cells: isolation,
characterization, and differentiation potential. Cytotherapy 5, 362369.
Gronthos, S., Franklin, D.M., Leddy, H.A., Robey, P.G., Storms, R.W., Gimble,
J.M., 2001. Surface protein characterization of human adipose tissue-derived
stromal cells. J. Cell. Physiol. 189, 5463.
Hall, S., 2005. The response to injury in the peripheral nervous system. J. Bone
Jt. Surg., Br. 87, 13091319.
Ide, C., 1996. Peripheral nerve regeneration. Neurosci. Res. 25, 101121.
Jessen, K.R., Mirsky, R., 1999. Developmental regulation in the Schwann cell
lineage. Adv. Exp. Med. Biol. 468, 312.
Jiang, J.X., Choi, R.C., Siow, N.L., Lee, H.H., Wan, D.C., Tsim, K.W., 2003.
Muscle induces neuronal expression of acetylcholinesterase in neuron
muscle co-culture: transcriptional regulation mediated by cAMP-dependent
signaling. J. Biol. Chem. 278, 4543545444.
Keilhoff, G., Goihl, A., Langnase, K., Fansa, H., Wolf, G., 2006. Transdifferentiation of mesenchymal stem cells into Schwann cell-like myelinating
cells. Eur. J. Cell Biol. 85, 1124.
Krampera, M., Marconi, S., Pasini, A., Galie, M., Rigotti, G., Mosna, F., Tinelli,
M., Lovato, L., Anghileri, E., Andreini, A., Pizzolo, G., Sbarbati, A.,
Bonetti, B., 2007. Induction of neural-like differentiation in human
mesenchymal stem cells derived from bone marrow, fat, spleen and thymus.
Bone 40, 382390.
Lee, R.H., Wong, W.L., Chan, C.H., Chan, S.Y., 2006. Differential effects of
glial cell line-derived neurotrophic factor and neurturin in RET/GFRalpha1expressing cells. J. Neurosci. Res. 83, 8090.
Li, H., Wigley, C., Hall, S.M., 1998. Chronically denervated rat Schwann cells
respond to GGF in vitro. Glia 24, 290303.
Li, Q., Ping, P., Jiang, H., Liu, K., 2006. Nerve conduit filled with GDNF genemodified Schwann cells enhances regeneration of the peripheral nerve.
Microsurgery 26, 116121.
Lu, P., Blesch, A., Tuszynski, M.H., 2004. Induction of bone marrow stromal
cells to neurons: differentiation, transdifferentiation, or artifact? J. Neurosci.
Res. 77, 174191.

274

P.J. Kingham et al. / Experimental Neurology 207 (2007) 267274

Lundborg, G., 2000. A 25-year perspective of peripheral nerve surgery:

evolving neuroscientific concepts and clinical significance. J. Hand Surg.
[Am.] 25, 391414.
Mosahebi, A., Fuller, P., Wiberg, M., Terenghi, G., 2002. Effect of allogeneic
Schwann cell transplantation on peripheral nerve regeneration. Exp. Neurol.
173, 213223.
Muschler, G.F., Nitto, H., Boehm, C.A., Easley, K.A., 2001. Age- and genderrelated changes in the cellularity of human bone marrow and the prevalence
of osteoblastic progenitors. J. Orthop. Res. 19, 117125.
Ning, H., Lin, G., Lue, T.F., Lin, C.S., 2006. Neuron-like differentiation of adipose
tissue-derived stromal cells and vascular smooth muscle cells. Differentiation
74, 510518.
Pereira Lopes, F.R., Camargo de Moura, C.L., Dias Jr., C.J., Balduino, A., Lora,
S., Langone, F., Borojevic, R., Blanco Martinez, A.M., 2006. Bone marrow
stromal cells and resorbable collagen guidance tubes enhance sciatic nerve
regeneration in mice. Exp. Neurol. 198, 457468.
Pittenger, M.F., Mackay, A.M., Beck, S.C., Jaiswal, R.K., Douglas, R., Mosca,
J.D., Moorman, M.A., Simonetti, D.W., Craig, S., Marshak, D.R., 1999.
Multilineage potential of adult human mesenchymal stem cells. Science 284,
143147.
Rutkowski, G.E., Miller, C.A., Jeftinija, S., Mallapragada, S.K., 2004.
Synergistic effects of micropatterned biodegradable conduits and Schwann
cells on sciatic nerve regeneration. J. Neural Eng. 1, 151157.
Safford, K.M., Hicok, K.C., Safford, S.D., Halvorsen, Y.D., Wilkison, W.O.,
Gimble, J.M., Rice, H.E., 2002. Neurogenic differentiation of murine and
human adipose-derived stromal cells. Biochem. Biophys. Res. Commun.
294, 371379.

Safford, K.M., Safford, S.D., Gimble, J.M., Shetty, A.K., Rice, H.E., 2004.
Characterization of neuronal/glial differentiation of murine adipose-derived
adult stromal cells. Exp. Neurol. 187, 319328.
Strem, B.M., Hicok, K.C., Zhu, M., Wulur, I., Alfonso, Z., Schreiber, R.E.,
Fraser, J.K., Hedrick, M.H., 2005. Multipotential differentiation of adipose
tissue-derived stem cells. Keio J. Med. 54, 132141.
Tholpady, S.S., Katz, A.J., Ogle, R.C., 2003. Mesenchymal stem cells from rat
visceral fat exhibit multipotential differentiation in vitro. Anat. Rec.,
A Discov. Mol. Cell Evol. Biol. 272, 398402.
Tohill, M., Terenghi, G., 2004. Stem-cell plasticity and therapy for injuries of the
peripheral nervous system. Biotechnol. Appl. Biochem. 40, 1724.
Tohill, M., Mantovani, C., Wiberg, M., Terenghi, G., 2004. Rat bone marrow
mesenchymal stem cells express glial markers and stimulate nerve regeneration. Neurosci. Lett. 362, 200203.
Wiberg, M., Terenghi, G., 2003. Will it be possible to produce peripheral
nerves? Surg. Technol. Int. 11, 303310.
Wislet-Gendebien, S., Bruyere, F., Hans, G., Leprince, P., Moonen, G., Rogister,
B., 2004. Nestin-positive mesenchymal stem cells favour the astroglial
lineage in neural progenitors and stem cells by releasing active BMP4. BMC
Neurosci. 5, 33.
Yoshimura, H., Muneta, T., Nimura, A., Yokoyama, A., Koga, H., Sekiya, I.,
2007. Comparison of rat mesenchymal stem cells derived from bone
marrow, synovium, periosteum, adipose tissue, and muscle. Cell Tissue Res.
327, 449462.
Zuk, P.A., Zhu, M., Ashjian, P., De Ugarte, D.A., Huang, J.I., Mizuno, H., Alfonso,
Z.C., Fraser, J.K., Benhaim, P., Hedrick, M.H., 2002. Human adipose tissue is a
source of multipotent stem cells. Mol. Biol. Cell 13, 42794295.