Advanced Drug Delivery Reviews 61 (2009) 10841096

Contents lists available at ScienceDirect

Advanced Drug Delivery Reviews

j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / a d d r

Electrospun scaffolds for stem cell engineering

Shawn H. Lim a, Hai-Quan Mao b,
a

Department of Biomedical Engineering, Johns Hopkins University School of Medicine, 3400 North Charles Street, 102 Maryland Hall, Baltimore, MD 21218, USA
Department of Materials Science and Engineering and Whitaker Biomedical Engineering Institute, Johns Hopkins University, 3400 North Charles Street, 102 Maryland Hall,
Baltimore, MD 21218, USA

a r t i c l e

i n f o

Article history:
Received 23 January 2009
Accepted 16 July 2009
Available online 30 July 2009
Keywords:
Electrospinning
Nanober
Stem cell niche
Topographical cue
Regenerative medicine
Stem cell expansion
Fate specication
Directed differentiation

a b s t r a c t
Stem cells interact with and respond to a myriad of signals emanating from their extracellular microenvironment. The ability to harness the regenerative potential of stem cells via a synthetic matrix has
promising implications for regenerative medicine. Electrospun brous scaffolds can be prepared with high
degree of control over their structure creating highly porous meshes of ultrane bers that resemble the
extracellular matrix topography, and are amenable to various functional modications targeted towards
enhancing stem cell survival and proliferation, directing specic stem cell fates, or promoting tissue
organization. The feasibility of using such a scaffold platform to present integrated topographical and
biochemical signals that are essential to stem cell manipulation has been demonstrated. Future application of
this versatile scaffold platform to human embryonic and induced pluripotent stem cells for functional tissue
repair and regeneration will further expand its potential for regenerative therapies.
2009 Elsevier B.V. All rights reserved.

Contents
1.
2.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Electrospun nanobers recapitulate features of the stem cell niche . . . .
2.1.
Electrospun scaffolds as stem cell culture supports . . . . . . . . .
2.2.
Stem cell interactions with ECM-mimetic electrospun bers . . . .
2.3.
Nanober modication for presentation of biochemical cues to stem
3.
Electrospun scaffolds directly inuence stem cell/progenitor differentiation
3.1.
Effects of ber alignment . . . . . . . . . . . . . . . . . . . . .
3.2.
Effects of ber diameter . . . . . . . . . . . . . . . . . . . . .
3.3.
Release of bioactive molecules for control of stem cell fate . . . . .
4.
Rational design of stem cell constructs for tissue engineering . . . . . . .
4.1.
Engineering cardiovascular tissue . . . . . . . . . . . . . . . . .
4.2.
Myotube formation for skeletal muscle engineering . . . . . . . .
4.3.
Nanocomposites for osteogenesis . . . . . . . . . . . . . . . . .
5.
Conclusions and future perspectives . . . . . . . . . . . . . . . . . . .
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . .
. . .
. . .
. . .
cells .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

1084
1085
1085
1086
1088
1088
1088
1090
1090
1092
1092
1094
1094
1094
1095
1095

1. Introduction

This review is part of the Advanced Drug Delivery Reviews theme issue on Nanobers
in Regenerative Medicine & Drug Delivery.
Corresponding author. Tel.: +1 410 516 8792; fax: +1 410 516 5293.
E-mail addresses: shlim@jhu.edu (S.H. Lim), hmao@jhu.edu (H.-Q. Mao).
0169-409X/$ see front matter 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.addr.2009.07.011

The aim of regenerative medicine is to repair or replace damaged

or diseased tissues in the human body. Progress in the eld has been
achieved at an accelerated pace, largely due to the improved understanding of the cell and tissue development process. Much of the focus
has been on the development of cell transplantation strategies, where
implanted cells either replace the loss-of-function, such as insulin-

S.H. Lim, H.-Q. Mao / Advanced Drug Delivery Reviews 61 (2009) 10841096

secreting cells for treatment of diabetes [1], or provide appropriate

signals to recruit host cells to repair the damage site, or in the case of
stem cell transplantation to differentiate into the desired lineages and
contribute to formation of new tissue [24]. Cell therapy based upon
stem and progenitor cells have many distinct advantages and offer
tremendous potential for regenerative medicine. The multipotency
and proliferative nature of stem cells makes them a more reliable cell
source than terminally differentiated phenotypes. Autologous adult
tissue-derived stem cells have the additional advantage of being
immune-compatible, although they are lineage-restricted. On the
other hand, embryonic stem (ES) cells are pluripotent, but their
derivation is ethically controversial, and much work remains to be
done to rene the control over ES cell differentiation prior to and after
implantation before these therapies can be put to widespread use.
Nevertheless, the exibility and promise of stem cells has generated
great scientic interest and they are currently a major focus in
regenerative medicine.
Regenerative medicine has much to benet from rapid advances in
the biomaterials engineering toolbox. For example, stem cell therapy
can be complemented by combining the cells with a supportive
scaffold to ll the damage site and assist in tissue repair. Rational
design of cell-supportive scaffolds has led to the evolution of scaffolds
beyond that of merely a passive support, towards systems that can
interact with and direct stem cell fates, in addition to promoting
integration with the host tissue. Much experimental evidence exists
to support the notion that stem cell fate can be controlled via
interactions with a synthetic scaffold, either prior to or after in vivo
implantation [58]. A widely-cited example is the work reported by
Silva et al. demonstrating that neural stem cells encapsulated within a
peptide-based amphiphilic hydrogel presenting an articially high
density of the laminin-inspired signaling epitope isolucinelysine
valinealaninevaline (IKVAV) selectively differentiated into neurons
at the expense of astrocyte differentiation [9]. Indeed, an ideal
implantable scaffold would recapitulate many of the salient features
of the native extracellular matrix (ECM) in the target tissue, including
but not limited to presentation of adhesion molecules, topographical
and biochemical cues.
Aside from creating cellscaffold constructs for implantation,
another potential contribution of biomaterials science to regenerative
medicine is the development of ex vivo methods for efcient
expansion and differentiation of stem cells. Due to the relative
scarcity of adult stem cells, access to an abundant supply of cells could
easily become a therapeutic bottleneck. Current methods for stem cell
manipulation have low efciencies for either expansion or differentiation, yielding mixed populations of cells that require the additional
step of separation and purication [1013]. Uncontrolled differentiation of neural stem cells following transplantation has also been
shown to generate undifferentiated mitotic neuroepithelial cells,
which may be tumorigenic [14]. From both a fundamental and clinical
perspective, further understanding of the interactions of stem cells
with articial scaffold platforms in vitro will provide instructive
insights towards the development of new technologies for stem cell
manipulation.
Over the past decade, electrospinning has gained rising popularity
as a means of fabricating scaffolds with micro to nanoscale features
similar to the hierarchical structure of the ECM. The ability to mimic
the ECM structural organization is an important consideration in
rational design of a cell-responsive scaffold platform upon which
additional functionalities can be incorporated. Electrospinning offers
great exibility in terms of the choice of scaffold material, as well
as ner control over the scaffold geometry. Fibers with diameters
ranging from tens to hundreds of nanometers can be easily produced,
where the diameter is adjusted empirically via modulation of spinning
parameters such as ow rate and collecting distance, and polymer
solution properties such as solvent, concentration, conductivity, and
surface tension [1517]. Electrospun polymeric brous meshes also

1085

offer a higher surface area for cell attachment, are easy and inexpensive to fabricate and scale-up, and are relatively reproducible.
Synthetic polymer-based systems offer additional advantages with
their adjustable mechanical properties, as well as ease of surface
functionalization via protein coatings, or chemical conjugation of
specic signaling molecules.
In this review, we will provide an overview of recent efforts and
current progress in manipulation and control of stem cell fates via
cellular interactions with electrospun brous scaffolds for ex vivo
stem cell expansion and differentiation, as well as applications of stem
cell-ber scaffold constructs for tissue engineering and regenerative
medicine. We will rst present the earlier applications of unmodied
electrospun scaffolds for stem cell culture, followed by discussion of
the various modications to the basic electrospun ber platform, such
as introduction of ECM molecules and ECM-like analogs, surface
presentation of biochemical cues, and controlled release of bioactive
molecules for initiation of stem cell signaling. Finally, we will outline
several future perspectives on engineering functional tissue constructs from stem cells seeded on electrospun scaffolds.
2. Electrospun nanobers recapitulate features of the stem
cell niche
Stem cells exist in vivo within a unique tissue-specic unique
microenvironment commonly termed the stem cell niche (Fig. 1). The
ECM comprises the structural component of the niche; in addition to
providing the physical support matrix for cell attachment, migration
and division, it also presents biochemical signals to cells that are
modulated through molecular interactions with ECM proteins such as
heparin sulfate proteoglycans (HSPGs) or through adjacent cells
[18,19]. Electrospun nanobrous scaffolds are able to recapitulate
both the structural features of the ECM, and, via various modications
to the ber material or surface, the biochemical cues as well. This type
of articial scaffold with enhanced biofunctionality would comprise a
more biomimetic microenvironment for ex vivo stem cell culture.
2.1. Electrospun scaffolds as stem cell culture supports
One of the major motivating factors supporting the use of electrospun brous scaffolds as supports for stem cell culture is the
observation that these brous scaffolds recapitulate the scale and
three-dimensional arrangement of collagen brils in the ECM.
Electrospun meshes generally comprise of nonwoven bers with
diameters in the hundreds of nanometers, and highly interconnected
pores that are tens of micrometers in diameter [16,20]. The high
surface areavolume ratio of these brous meshes also ensures
abundant area for cell attachment, which allows for a higher density
of cells to be cultured as compared with a at, two-dimensional
surface. The morphological resemblance of electrospun nanobers to
native ECM suggests their natural application as a supportive matrix
for creating scaffold constructs from stem cells.
Much of the earlier work on electrospun brous scaffolds as
matrices for stem cell culture was aimed at establishing the biocompatibility of such a scaffold architecture and its suitability for the
creation of tissue-like constructs in vitro. In an effort to assess bone
formation from human MSCs (hMSCs), Vacanti et al. used a rotational
oxygen-permeable bioreactor to provide uniform oxygen tension and
mechanical stress during osteogenic induction of hMSCs seeded on
polycaprolactone (PCL) nanobrous scaffolds [21]. After 4 weeks of
culture, scaffolds were found to be qualitatively stiffer, and Von Kossa
staining revealed that signicant calcication was present throughout
the construct. Constructs fabricated from nanobrous scaffolds
maintained their original size and shape, demonstrating improved
structural integrity over the culture period as compared with
previously tested macroporous poly(lactic-co-glycolic acid) foams,
which had a tendency to collapse or exhibit severe shrinkage. When

1086

S.H. Lim, H.-Q. Mao / Advanced Drug Delivery Reviews 61 (2009) 10841096

Fig. 1. The concept of the stem cell niche, and role of the extracellular matrix in regulating stem cell survival and signaling. ECM: extracellular matrix; GF: growth factor; [GF]: growth
factor concentration. Matrix topography may facilitate these processes. Adapted with permission from Macmillan Publishers Ltd [18].

this study was extended a further four weeks following implantation

of constructs into the omenta of rats, constructs adopted a rigid and
bone-like appearance, with histological evidence of extensive alkaline
phosphatase and collagen I deposition both on the outer and inner
portions of the scaffolds [22].
Concurrent with the efforts to engineer bone tissue using
nanobrous scaffolds, Li et al. also showed that these substrates
were suitable for supporting chondrogenic induction of human bone
marrow-derived MSCs in vitro [23]. Culturing MSCs on the scaffolds
not only produced comparable quantities of GAGs, but also exhibited
similar levels of chondrogenic gene expression as MSCs cultured as in
a cell pellet, an established model of chondrogenic induction. Of note,
culture on scaffolds abolished the requirement for the extremely high
cell densities previously shown to be critical for MSC differentiation.
The same group later demonstrated the general exibility of this
scaffold system in supporting multiphasic MSC differentiation by
successfully achieving differentiation of MSCs into chondrogenic,
osteogenic and adipogenic phenotypes [24] (Fig. 2).
Recently, the feasibility of this technology was tested in vivo by
evaluating hMSC-nanobrous scaffold constructs in the repair of a
full-thickness articular cartilage defect created in a swine model [25].
Cartilage repair is a particularly challenging problem in regenerative
medicine due to the avascular nature of the tissue, with limited access
to nutrients as well as a cell source for repopulating the defect.
Currently explored therapies involve implantation of tissue plugs or
autologous cells [26], which necessitate donor site morbidity; articial
scaffolds thus hold great promise as an alternative therapy for
cartilage repair. Electrospun PCL nanobrous scaffolds were seeded
with either hMSCs or porcine chondrocytes and allowed to repopulate
the scaffold for 3 weeks in vitro, following which time the constructs
were sutured onto defects created on the load-bearing portion of the
femoral condyle of the animals' hind knees. The most complete repair
was observed in defects treated with the hMSC-seeded scaffolds,
which regenerated into a smooth surface with hyaline cartilage-like
tissue formation. In contrast, scaffolds seeded with chondrocytes
produced tissue resembling brocartilage and the repaired surface
was discontinuous. In both test conditions, signicant remodeling of
subchondral bone was observed, with the loss of the tidemark
separating bone and cartilage. hMSC-seeded constructs also showed
superior equilibrium compressive stress compared with chondrocyteseeded or acellular scaffolds, albeit still inferior to the mechanical
properties of native cartilage. The authors proposed that hMSCs
supported by an electrospun nanobrous scaffold is a promising
approach for treatment of cartilage defects, as the stem cells are able

to proliferate within the scaffold, resulting in higher cell density as

well as matrix biosynthesis and deposition.
2.2. Stem cell interactions with ECM-mimetic electrospun bers
Structural proteins of the ECM such as collagen, bronectin and
laminin present cells with a myriad of recognition sites for binding cell
surface integrins, and HSPGs sequester and present growth factors
and cytokines. For example, bronectin contains the peptide sequence
arginineglycineaspartic acid (RGD) and its synergy site proline
histidineserinearginineasparagine (PHSRN), which is widely
implicated in integrin-mediated cell adhesion across a variety of cell
types [27,28]. Interactions between stem cells and ECM molecules are
implicated in supporting normal cell fates, including cell migration,
proliferation, and differentiation. The incorporation of ECM molecules
or ECM-inspired peptide analogs into electrospun scaffolds is thus a
means of combining the structural features of the scaffolds with the
biofunctionality of ECM proteins. Such biologically active nanobers
can better support stem cell attachment and growth. For example,
bers with surface laminin coating are favorable for the attachment,
proliferation and differentiation of neural stem cells and progenitor
cells [29].
A number of research groups have successfully fabricated electrospun scaffolds comprised of type I collagen, and evaluated their
effectiveness at supporting stem cell differentiation. Shih et al. compared the growth and osteogenic induction of bone marrow-derived
hMSCs on type I collagen nanobers [30]. Nanobrous collagen
scaffolds supported greater cell proliferation over uncoated tissue
culture polystyrene (TCPS) surfaces; hMSCs on collagen bers also
had elevated expression of differentiated osteogenic markers. hMSCs
on collagen nanobers showed fewer focal adhesions than on TCPS,
hinting at the possible involvement of cytoskeletal-linked signaling
pathways.
A study by Sefcik et al. suggested that simply presenting collagen
in nanober form better supports osteogenic differentiation than
coating collagen onto 2D tissue culture surfaces [31]. Using a novel
serum-free osteogenic induction medium, adipose-derived stem cells
cultured on electrospun collagen nanobers enhanced their expression of osteogenic genes compared to 2D collagen coating. Although
the integrity of the collagen nanobers did not persist beyond 9 days
in culture, the initial cellscaffold interactions were ostensibly
sufcient to account for the observed differences in degree of osteogenic differentiation. However, the difculty of achieving a relatively
stable scaffold in culture underlines the fundamental difculties in

S.H. Lim, H.-Q. Mao / Advanced Drug Delivery Reviews 61 (2009) 10841096

1087

Fig. 2. Multiphasic differentiation of hMSCs on electrospun brous scaffold. Scanning electron micrographs (SEM) of PCL nanobrous scaffold at (a) low magnication, scale bar =
30 m; (b) high magnication, scale bar = 10 m. (cj) SEM of scaffolds after differentiation for 21 days. (c, e, g, i) Cross sections; (d, f, h, j) top view of scaffolds. (c, d) Constructs
cultured in basal medium without supplements. (e, f) constructs cultured in adipogenic medium had globular cells; (g, h) constructs cultured in chondrogenic medium exhibited
round chondrocytelike cells with thick ECM; (i, j) constructs cultured in osteogenic medium showed nodules that were mineralized. Reprinted from [24], with permission from
Elsevier.

handling of electrospun collagen bers. In theory, scaffolds for tissue

regeneration should be able to support cells post-implantation for a
sufciently long time to allow cells to proliferate and repopulate
the defect site. A report by Li et al. investigating the response of
chondrocytes and hMSCs to a variety of scaffolds with different
degradation rates revealed that cells had higher proliferation rate on

the more stable scaffolds [32]. Rapidly degrading scaffolds quickly

lost their porosity and structural integrity, to the detriment of cell
adhesion and ingrowth. Other reports likewise conclude that electrospun collagen scaffolds are insufciently stable to carry out this
role without some form of post-spinning modication, such as
chemical crosslinking with gluteraldehyde, which introduces

1088

S.H. Lim, H.-Q. Mao / Advanced Drug Delivery Reviews 61 (2009) 10841096

cytotoxicity issues [33,34]. The question is also raised whether simply

coating a thin layer of collagen onto electrospun polymeric bers
would elicit the same effects as observed in these previous studies. An
alternative approach would be to blend ECM proteins with a second
polymeric component prior to electrospinning. Blending gelatin with
PCL in a 30:70 ratio produced bers that were structurally stable in
culture, and still supported signicantly greater cell adhesion and
proliferation of C17.2 neural stem cells compared with pure PCL bers
alone [35].
Laminin are second only to collagen in terms of its ubiquity in the
basal lamina, can self-assemble into networks either independently,
or in close association with other ECM components such as collagen
IV and nidogen, as found in the commercial ECM substrate Matrigel.
Through integrin-mediated interactions, laminin plays a critical role
in cell fate specication. Indeed, the expression of laminin-1 has
been established to be essential during early embryogenesis [36].
In vitro culture on laminin-coated substrates has also been shown to
be highly efcacious in the adhesion and expansion of neural progenitor cells. With this in mind, Neal et al. found that electrospun
laminin supported a higher degree of attachment and neurite
extension of adipose tissue-derived stem cells than laminin lms,
although there was not a statistically signicant difference in the
percentage of cells that expressed 3-tubulin, an early neuronal
marker [37].
More widespread adoption of ECM molecule electrospinning has
generally been hindered by the cost of acquiring sufcient material.
The issue of maintaining scaffold stability is not a trivial one as well.
Furthermore, there is considerable resistance to using animal-derived
proteins in an implantable scaffold, due to issues of immunogenicity;
for example, laminin is generally puried from tumor tissue [38]. One
might also argue the necessity for electrospinning ECM proteins when
studies have showed that coating the ber or conjugating the ber
surface with ECM proteins is sufcient to induce the appropriate cell
response.
2.3. Nanober modication for presentation of biochemical cues to stem
cells
Synthetic electrospun polymer scaffolds possess an advantage
over naturally-derived biomaterials as the polymeric platform is
amenable to a variety of functional modications. Substrate presentation of functionally relevant biochemical cues or surface chemistry
has been extensively investigated and shown to inuence cell
response. For example, surface modication with amine, hydroxyl,
and carboxylic acid groups revealed that integrin binding specicity to
different functional groups correlated with differential patterns of
focal adhesion and matrix deposition in immature osteoblast-like cells
[39]. Especially within the context of stem cell engineering, regulating
stem cell fate decision via manipulation of biochemical interactions
is an area of great interest. The ability to combine topographical and
biochemical cues within a single scaffold presents a valuable opportunity to evaluate their synergistic impact.
Achieving efcient ex vivo expansion of hematopoietic stem/
progenitor cells (HSCs) is an attractive option for the derivation of a
consistent and reliable cell supply for the treatment of a variety of
hematological disorders. In an expansion culture, the combination of
nanober topography and surface functional groups have been shown
to synergistically improve the self-renewal and proliferation of
human cord blood-derived HSCs [40], as shown in Fig. 3. Amination
of electrospun poly(ether sulfone) nanobers resulted in a scaffold
that supported the highest expansion fold of cryopreserved HSCs
compared with carboxylation and hydroxylation (195-fold vs. 40-fold
and 60-fold, respectively). Expansion of HSCs on aminated nanobers
resulted in the highest expansion of CFU-GEMM cells compared with
similarly modied 2-D lm and other substrates. More interestingly,
HSCs showed better adhesion to this nanober matrix and prolifer-

ated as colonies with greater preservation of the proportion of stem

and progenitor cells. This result suggests a selective enrichment effect
by these functional nanobers during expansion. The clinical potential
of this technology was further demonstrated by the successful
engraftment of the bone marrow of NOD/SCID mice by HSCs expanded on aminated nanober substrates [41].
3. Electrospun scaffolds directly inuence stem cell/progenitor
differentiation
When attempting to replicate the structural hierarchy of the ECM
in a synthetic scaffold, we should appreciate that the natural ECM
contains components with different organizational length scales.
Organization on the macro- and micro-scale is largely responsible for
the structural integrity, porosity and other physical properties of the
matrix. Such a structural organization may play an important role in
dictating the cellular responses elicited by the local ECM components.
It is worth noting that most cellcell and cellscaffold communications occur via nanoscale molecular presentation and organization,
such as in the cases of receptor clustering and focal adhesion complex
formation. Due to the ability to modulate structural parameters of the
electrospun brous scaffold, the inuence of topographical cues on
the control of stem cells cultured on or in these scaffolds can be
explored in detail.
3.1. Effects of ber alignment
One unique feature of electrospun scaffolds is the ability to
manipulate the deposition of bers so as to achieve meshes of highly
aligned bers. Instead of using a grounded stationary collector, bers
can be deposited onto a continuously rotating mandrel, or onto the
edge of a spinning disc, which preferentially orients the bers in the
direction of the axis of rotation (Fig. 4). Cells cultured on such
scaffolds are shown to adhere and elongate along the long axis of ber
alignment [4244]. These brous meshes present a simple, scalable
and straightforward strategy to induce stem cell alignment and
introducing directionality of cell growth in the cell-seeded constructs.
This principle of contact guidance is particularly relevant in the
engineering of tissues that intrinsically possess highly anisotropic cell
organization, such as skeletal muscle tissue, ligaments, articular cartilage and blood vessel walls. The anisotropic nature of these tissues is
critical for their proper development of function in vivo, for example,
the alignment of skeletal muscle cells permits the fusion of myoblasts
into polynucleated myotubes that make up the structural building
blocks of the densely packed muscle bers that generate longitudinal
muscle contraction [45]. It is generally hypothesized that using the
substrate to induce an elongated, possibly more physiologically relevant cell morphology could alter, or even improve the responsiveness
of stem cells to extrinsically applied cues, resulting in enhanced
differentiation and improved tissue organization and function. An
example of this was shown by Baker and Mauck using aligned PCL
bers to replicate the network of radially aligned collagen brils that
comprise the brocartilaginous menisci of the knee [43]. They found
that MSCs seeded on aligned bers resulted in a tissue with higher
stiffness and modulus after 10 weeks of culture than the MSC-random
ber construct, even though both groups yielded comparable
productions of glycosaminoglycan (GAG) and total collagen. This
was likely due to the more organized deposition of ECM in aligned
ber construct. A recent report by Wise et al. demonstrated that
culturing hMSCs on aligned electrospun nanobers for the purpose of
engineering the supercial zone of articular cartilage enhanced the
chondrogenic differentiation of hMSCs compared with culturing them
on 2-dimensional lms, as evidenced by signicantly higher sulfatedGAG production and collagen II mRNA expression [44].
Besides complementing stem cell differentiation and tissue organization, recent evidence has emerged to suggest that interactions

S.H. Lim, H.-Q. Mao / Advanced Drug Delivery Reviews 61 (2009) 10841096

1089

Fig. 3. Aminated nanobers support the in vitro expansion of HSCs. (a). SEM images showing that human cord blood HSCs cultured on aminated nanobers formed circular colonies
after 10 days of expansion, whereas random adherent cells were found at the cracks on 2-D lm (b). Functionalized PES bers have an average diameter of 529 nm and the average
surface amino group concentration was 55 nmol/cm2. (c). Aminated nanobers promoted the expansion of CFUGEMM colony forming cells (CFUGEMM: colonyforming unit
granulocyte, erythrocyte, monocyte and megakaryocyte). (d). Phenotypic comparison between adherent and suspension populations of the cells expanded on aminated PES
nanober scaffolds, lm and TCPS. Adapted from [40], with permission from Elsevier.

with the underlying substrate topography can be exploited to actively

inuence stem cell fate decisions. Work done by Yim et al.
demonstrated that hMSCs cultured on a uniform 350 nm wide grating
exhibited elongated and orientated morphology and slower proliferation; more signicantly, hMSCs on nanogratings showed signicant
upregulation of expression of the neuronal marker MAP2 over those on
unpatterned surfaces [46]. The presence of the nanotopography alone
was sufcient for induction of neuronal differentiation of hMSCs even
without the addition of the neuronal inducer retinoic acid. This work is
of particular interest because hMSCs do not typically differentiate into
a neuronal lineage, suggesting that topographical cues might present a
unique paradigm of achieving neural transdifferentiation. Although
the exact signaling pathways responsible for this phenomenon have
yet to be elucidated, such a fate specication mechanism might be
related to the observed cytoskeletal rearrangement and nuclei
elongation. Work published by Chen et al. appears to support this
hypothesis. They demonstrated that changes to hMSC cell shape and

cytoskeletal tension signal through the Rho-ROCK pathway to dictate

the fate choice between adipogenesis and osteogenesis [47].
Despite these promising results, most of the research conducted
to date has been limited to using microfabricated substrates to
demonstrate the effects of micro- and nanotopography on stem cell
differentiation. Although a convenient and relatively inert platform
for in vitro evaluation, microfabricated substrates are time-consuming
to prepare and are not easily scaled-up for the preparation of
implantable cell scaffolds. Electrospinning of aligned nanobers is a
practical solution to this obstacle. Although such scaffolds possess
relatively lower degree of homogeneity in presenting the aligned
topography as compared with micropatterned substrates, it is
reasonable to assume that similar outcomes can be expected. A recent
report published by Dang and Leong highlighted a unique application
of electrospinning for the creation of highly aligned stem cell sheets
[48]. Thermosensitive hydroxybutyl chitosan (HBC) was electrospun
into aligned nanober meshes, with the aim of modulating hMSC

1090

S.H. Lim, H.-Q. Mao / Advanced Drug Delivery Reviews 61 (2009) 10841096

Fig. 4. Control over ber alignment and geometry. Fiber alignment can be achieved by switching from a stationary collector to a rotating disk collector. Fiber diameter can be
controlled via changing parameters such as polymer solution concentration or ow rate.

response through the substrate topography. Human MSCs cultured on

aligned HBC nanobers exhibited an upregulation of the myogenic
genes collagen IV, desmin, Pax-3, Pax-7 and myogenin compared with
hMSCs on HBC lms. The most interesting advantage of this system
was that after the cells reached conuent density, the entire construct
could be transferred to medium cooled to 4 C, which dissolved the
bers and released a cell sheet. Multiple such cell sheets can
conceivably be stacked to produce a thicker polymer-free tissue
suitable for implantation.
In neurogenesis and neural regeneration, a single axon often has to
extend over relatively long distances, seeking out and forming
connections with other neurons that are both meaningful and
functional. During normal embryonic development, axonal pathnding in the central nervous system is facilitated by a chemoattractants
such as netrins and semaphorins. Unfortunately the same complement of spatially dened biochemical signals may not be available
after traumatic injury to the nervous system. Neural regeneration in
the central nervous system is complicated by the formation of the glial
scar that constitutes a physical barrier to regenerating axons. On the
other hand, long nerve gaps are often characteristic of peripheral
nervous system injury. The use of aligned nanobers as a physical
guidance cue to simulate anisotropic directionality for regenerating
axons is postulated to be a strategy that likely improves the efciency
of nerve regeneration.

and nanober effects on stem cells. In a recent paper, Lin et al.

employed a facile modication to the electrospinning technique to
control ber diameter and dimension variability [17]. The authors
demonstrated that addition of a cationic amphiphile to the polymer
solution helped to reduce the surface tension of the solution, resulting
in a thinner yet more stable charged polymer jet that deposited
narrower, more uniform electrospun bers. This technique was used
by Christopherson et al. to investigate the impact of electrospun ber
diameter on the differentiation of adult rat hippocampal-derived
neural stem cells (NSCs) [29]. Rat NSCs were cultured on laminincoated electrospun poly(ether sulfone). (PES) ber scaffolds with
average diameters of 283 nm and 749 nm, and differentiated in the
presence of serum and retinoic acid (Fig. 5). NSCs cultured on the
smaller diameter bers differentiated preferentially along the oligodendrocyte lineage, whereas on the larger diameter bers, a higher
degree of neuronal differentiation was observed. This difference was
attributed to the effects of the ber diameter on limiting cell
spreading and migration. On the 283 nm nanobers, neurites were
guided by the underlying ber matrix and assumed a glial-like
morphology, while cells on the 749 nm bers were restricted to
spreading along single bers and thus did not favor differentiation
into glial lineages.

3.2. Effects of ber diameter

Tissue regeneration and repair can often be complemented by

targeted delivery of a specic signaling molecules that enhance tissue
growth rate, promote vascularization, or serve as chemoattractive
signals for stem or progenitor cells homing to the repair site. For
example, broblast growth factors are a mitogenic signal for a variety
of cell types, and members of the transforming growth factor-beta
superfamily such as BMP-2 are widely implicated in the regulation of
skeletal development, including proliferation and differentiation of
MSCs towards chondrogenic and osteogenic fates. Encapsulation of
signaling proteins within an electrospun scaffold and their subsequent release allows the scaffold to act as a vehicle for sustained
delivery of growth factors to the local cell microenvironment. This
approach has been explored with a particular focus on bone
regeneration. BMP-2 has been shown to have osteoinductive activity
for bone formation; numerous studies demonstrated that scaffolds
with encapsulated BMP-2 signicantly improved bone regeneration
in vivo, including extensive segmental defects in long bone [50]. Li
et al. electrospun silk broin scaffolds containing BMP-2 and
evaluated their effect on in vitro bone formation from bone marrowderived hMSCs [51]. BMP-2 incorporation elicited a four-fold higher
amount of calcium deposition than scaffolds without the protein.

Varying the size scale of topographical cues by changing electrospun ber diameter is another potential means of imposing a spatial
restriction on stem cells. It is well-established that electrospun ber
diameter can be varied by changing the concentration of the polymer
spinning solution, changing the solution ow rate, or regulating the
distance between the needle and collector. It is postulated that
through interactions with bers on different length scales, one might
signicantly inuence stem cell spreading, migration, proliferation
and differentiation. In one study, Yang et al. evaluated the impact of
ber diameter on the morphology and neurite extension of C17.2
mouse neonatal cerebellum stem cells [49]. Poly(l-lactic acid) (PLLA)
was electrospun into micro- and nanober scaffolds by changing PLLA
solution concentration. The authors found that C17.2 cells cultured on
nanober scaffolds showed increased neurolament 200 kD staining
than cells on microber scaffolds; furthermore, aligned nanober
scaffolds promoted neurite extension from C17.2 cells over random
nanober scaffolds.
Controlling the uniformity of electrospun ber diameter is
essential to facilitating a rigorous comparison between microber

3.3. Release of bioactive molecules for control of stem cell fate

S.H. Lim, H.-Q. Mao / Advanced Drug Delivery Reviews 61 (2009) 10841096

1091

Fig. 5. Impact of ber diameter on the differentiation of rNSCs. Rat adult NSCs were differentiated on laminincoated PES bers with different diameters, in comparison with tissue
culture polystyrene. (ab) SEM images of 749 nm (a); and 283 nm (b) bers. (c) NSCs on 749 nm bers immunostained for progenitor (Nestin+) and neuronal (Tuj1+) markers.
(d) NSCs on 283 nm bers immunostained for oligodendrocyte (RIP+) and astrocyte (GFAP+) markers. (e) Quantication of NSC differentiation on all substrates. Adapted from [29],
with permission from Elsevier.

Furthermore, mRNA transcript analysis showed that exposure of

hMSCs to BMP-2 resulted in signicantly higher expression of
osteogenic genes BMP-2, bone sialoprotein-II and collagen I than the
brous scaffold alone. Although the results appeared to demonstrate
that the encapsulated BMP-2 retained some of its biological activity,
a direct comparison with soluble BMP-2 supplemented into the
culture media would be more indicative of the relative potency of
the released growth factor. It is also uncertain at this point whether
the encapsulation is sufcient to protect the growth factor from
degradation over extended periods of time, a condition which is
necessary for an in vivo delivery system.
Gene therapy offers an attractive alternative to drug delivery,
especially in instances where the signaling protein of interest is highly
unstable or difcult to formulate in vitro. Additionally, gene therapy
provides a means of altering cellular activity at a genetic level, which
cannot be achieved by application of an external drug. For example,
delivering small interfering RNA can knockout or knockdown
expression of a particular gene product [52,53], thereby promoting
or inhibiting a particular differentiation pathway. On the other hand,

sustained release of plasmid DNA from a scaffold can provide

continuous transfection of the cells cultured on the scaffold [54].
Transfected stem cells, particularly MSCs, can then be used as
localized bioreactors for therapeutic delivery of bioactive gene
products [55]. Particularly, in the case of protein therapeutics with
short half life or requiring supraphysiological levels to elicit desired
bioactivity in vivo, secretion by genetically transduced cells can result
in higher concentrations locally over a prolonged time period. Work
by Nie and Wang have demonstrated that it is possible to electrospin
plasmid DNA in the form of chitosan nanoparticles with PLGA into
nanobers as a means of achieving scaffold-based gene delivery [56]
(Fig. 6). Chitosan-DNA nanoparticles were incorporated into nanobers with high encapsulation efciency (N65%). Gradual release of
nanoparticles occurred over a 60-day time period. Formulation of
plasmid DNA into nanoparticles protected the DNA from degradation
during the electrospinning process; nanoparticles released from the
bers successfully transfected bone marrow MSCs seeded on the
scaffold as measured by protein secretion. Further investigation is
merited to determine whether transfected hMSCs are indeed more

1092

S.H. Lim, H.-Q. Mao / Advanced Drug Delivery Reviews 61 (2009) 10841096

Fig. 6. Incorporation of chitosanDNA nanoparticles into electrospun PLGA for scaffoldbased gene delivery of BMP2. (a) SEM micrographs of brous scaffolds encapsulating
nanoparticles, 3 types of scaffolds were spun containing different proportions of hydroxyapatite; (b) cross section of a single ber, with encapsulated nanoparticles indicated by
arrows; (c) continuous release of nanoparticles was achieved over a 66day time period. Increasing the amount of hydroxyapatite in the scaffolds quickened the rate of release.
Reprinted from [56], with permission from Elsevier.

effective at promoting bone regeneration in vivo, and if so, whether

such an effect is exerted through osteogenic differentiation or by
chemotactic recruitment of host stem cells.
4. Rational design of stem cell constructs for tissue engineering
The ultimate goal of tissue engineering is to design and fabricate a
cellscaffold construct that, post-implantation, facilitates regeneration of neo-tissue that is both functional and well-integrated with the
host. Electrospun scaffolds have been demonstrated to possess
suitable biocompatibility and mechanical strength to act as temporary
scaffolding; furthermore they are easily processed into templated
shapes such as tubes and patches to suit the dimensions of the defect.
Continuing efforts that combine electrospinning technology with stem
cells for tissue regeneration are currently under way. Here we will
summarize several case studies that exemplify such efforts.
4.1. Engineering cardiovascular tissue
Currently, engineering of smaller diameter grafts remains a serious
challenge, primarily because the acellular grafts are prone to occlusion
and thrombosis following implantation. The presence of the endothelial intimal layer in vascular grafts is believed to be antithrombogenic. It was shown that grafts seeded with endothelial progenitor
cells prior to implantation in an ovine model maintained graft patency
over 130 days [57]. However, reproducing the endothelial layer is only
part of the solution, as graft strength and contractility are provided by
the medial layer of aligned smooth muscle cells and connective tissue.
Due to their ability to differentiate into vascular phenotypes, bone

marrow-derived MSCs were explored as a strategy for repopulation of

a decellularized matrix in a vascular graft [57]. Although there was
successful formation of both the endothelial and smooth muscle cell
layers, it was unclear whether MSCs differentiated into these cell
types following implantation, or had recruited host cells to repopulate
the graft. Nevertheless, this study established the feasibility of MSCs
as a cell source for regenerating bioarticial vascular grafts.
Decellularized tissue matrices pose several attractive features
as scaffolds because they already possess the essential complement
of relative tissue ECM. However, donor tissue is likely to elicit immunogenic responses, and contain dramatic batch-to-batch variation.
Several elements of electrospinning technology can overcome these
challenges, while still fullling the same desirable properties of a
vascular scaffold. The vascular tissue macrostructure can be recapitulated by electrospinning bers onto a thin mandrel to fabricate a
tubular structure with suitable mechanical strength and exibility.
Selection of an elastomeric polymer will provide the scaffold with
compliance comparable to native tissues. The porosity of the scaffold
can be adjusted to facilitate cell inltration and attachment. Using a
combined electrospinning/electrospraying technique, Wagner and
colleagues have taken a unique approach to rapid fabrication of a cellintegrated vascular graft with retention of high levels of cell viability
[58,59]. This technique has the advantage of homogenous distribution
of cells, but the rather harsh processing conditions during electrospraying might hinder its application to stem cells.
Zhang et al. investigated small diameter vascular tissue engineering by seeding bone marrow MSCs on an electrospun poly(propylene
carbonate) scaffold [60]. In light of the fact that nitric oxide (NO)
production is correlated with vessel patency following bypass surgery,

S.H. Lim, H.-Q. Mao / Advanced Drug Delivery Reviews 61 (2009) 10841096

1093

Fig. 7. Remodeling and cellularization of aligned electrospun vascular grafts without MSCs (a, c, e, g) and with preseeded MSCs (b, d, f, h) after 60 days in vivo. (a, b) Hematoxylin/
eosin staining of graft crosssection; (c, d) CD31+ staining indicates endothelial cell monolayer; (e, f) MHC staining for SMCs; (g,h) Verhoeffs staining indicates collagen in pink and
elastin in black. Scale bars: (ab), 100 m; (ch), 100 m. (i) Cellseeded aligned brous sheet is rolled around a mandrel into a tube, followed by (j) implantation as a common cartid
artery bypass in a rat. Reproduced from reference [62].

1094

S.H. Lim, H.-Q. Mao / Advanced Drug Delivery Reviews 61 (2009) 10841096

rat MSCs were virally transfected with a gene encoding for endothelial
nitric oxide synthase (eNOS), with the hypothesis that eNOS gene
delivery would be a suitable means of inhibiting neointimal
formation. The authors demonstrated that the electrospun scaffold
facilitated survival and proliferation of transfected MSCs, and MSCs
produced NO at a level comparable to endothelial cells from a freshly
harvested artery. It remains to be veried whether MSCs differentiate
into the smooth muscle cell layer and NO production induces
endothelialization following in vivo implantation.
In a more comprehensive study, Hashi et al. seeded bone marrow
MSCs on an aligned PLLA nanobrous mesh, allowed the cells to grow
to conuence, and then rolled the mesh into a tubular structure
reminiscent of a vascular graft [61]. The grafts were implanted as a
common carotid artery bypass in rats. After 60 days in vivo, the MSCseeded tubes stained positively for a CD31+ endothelial layer as well
as myosin heavy chain (MHC+) smooth muscle cell layer (Fig. 7). In
contrast to acellular tubes, SMCs in the grafts were arranged in a tight
band, with collagen and elastin deposition patterns mimicking the
structure of native arteries. More interestingly, there was very little
neointimal formation in the cell-seeded grafts. Further experiments
showed that MSCs seeded on nanobrous scaffolds did not support
platelet aggregation in the short term. The authors attributed the
antithrombogenic property of MSCs to the presence of heparin sulfate
proteoglycans on the cell surface, rather than endogenous production
of nitric oxide. Contrary to their initial hypothesis, most of the cells
within the graft at 60 days post-implantation were not originated
from the seeded MSCs, indicating their role of recruiting host cells to
the implantation site likely through MSC-released trophic factors. The
contribution of MSCs to maintaining patency of the graft was thus
postulated to be short-term prevention of thrombus formation.
4.2. Myotube formation for skeletal muscle engineering
Mature skeletal muscle consists of bundles of terminally differentiated and multinucleated muscle bers that have formed via myoblast
fusion. Damaged muscle has a limited capacity for regeneration;
however, there exists a small population of quiescent muscle progenitor
cells, known as satellite cells, that are activated in response to muscle
damage and can fuse with damaged bers or form new myotubes
themselves [62]. Satellite cells are committed to the myogenic lineage,
but still retain proliferative capabilities, and are thus considered
the prototypical muscle stem cells. One paradigm for muscle reconstruction and repair is ex vivo expansion of autologously derived satellite
cells, followed by scaffold-supported differentiation into muscle tissue.
Riboldi et al. electrospun DegraPol, a commercially available and
degradable block copolymer polyesterurethane, into aligned and
random brous meshes and compared the differentiation and
myotube formation from C2C12 and L6 myoblast cell lines [63].
These scaffolds were pre-coated with Matrigel, an animal-derived
puried ECM, to facilitate cell adhesion. Cultured myoblasts exhibited
a steady proliferation on both aligned and random meshes over
1 week; however, rate of proliferation on aligned meshes was
signicantly slower than on random meshes. The authors took this
to be indicative of increased differentiation induced by aligned
morphology. Both myoblast lines on aligned bers had upregulated
gene expression of myogenin and myosin heavy chain (MHC), and
differentiation was further conrmed by anti-MHC immunouorescence staining. Putative multinucleated myotubes were highly
oriented in parallel arrays on aligned bers, which was not the case
on random bers. In an earlier study, Huang et al. evaluated the
differentiation of C2C12 myoblasts on aligned PLLA nanobers and
showed similar results in myoblast gene expression [64]. The authors
also quantied the average length of differentiated myotubes and
found that aligned brous scaffolds induced myotube formation that
was at least twice the length of myotubes generated on random bers.
Striated sarcomeres assembled in the myotubes, although the relative

proportions of striated myotubes were not signicantly different on

the two types of ber meshes.
These studies demonstrated that aligned electrospun scaffolds
have the potential to direct myoblast organization and assembly via
simple contact guidance. As a follow-up work, satellite cells cultured
on such substrates may be subjected to mechanical or electrical
stimulation reminiscent of in vivo conditions that might potentially
enhance myogenic differentiation. In order to realize the potential of
this strategy, the next challenges to overcome would be to increase
the differentiation efciency of the myoblasts, as well as assembly
of myotubes into compact, elongated bundles capable of contracting
in unison.
4.3. Nanocomposites for osteogenesis
A tested strategy in developing articial scaffolds for bone regeneration is the incorporation of bioceramics in order to better mimic
the mineral composition of native bone as well as promote osteointegration and osteoinduction [65]. Mixing a more ductile and
exible polymer with ceramics such as calcium phosphate or
hydroxyapatite (HAp) helps to reduce the brittleness of ceramic
structures and also facilitates shaping the scaffold to match the
defect size. The versatility of the electrospinning technique allows
for incorporation of bioactive ceramics into the polymeric nanobers. Schneider et al. electrospun bers of PLGA containing up to
40% nanoparticles of aerosolized amorphous tricalcium phosphate
(ATCP), and found that the meshes formed had a texture similar to
cotton wool [66]. Incubation of this mesh in simulated body uid
promoted deposition of a dense layer of hydroxyapatite in an amount
correlated with the doping level of ATCP. This nding demonstrated
that incorporation of ATCP into the scaffold promotes mineralization.
The mineralized scaffolds did not have an adverse effect on
biocompatibility, as indicated by the fact that the nanocomposite
scaffolds supported hMSC proliferation and osteogenic differentiation to the same extent as the polymer scaffold without ATCP. A
variation of ATCP encapsulation is to load scaffolds with HAp
[51,56,67]. Similarly, the encapsulation of HAp particles had no
deleterious effects on hMSC viability and differentiation in culture.
When both HAp and BMP-2 were encapsulated into electrospun
bers, HAp particles acted as an adjuvant to modulate the release
rate of growth factor from the bers. Fibers with higher doping level
of HAp had a more rapid release rate of BMP-2 over a 60-day test
period (Fig. 6).
As an alternative to using bone mineral composite scaffolds, Ko
et al. created a novel type of nanobrous composite scaffold by electrospinning a mixture of demineralized bone powder (DBP) with poly
(l-lactide) (PLA) [68]. Although there was no signicant difference in
the expression of osteogenic markers in mandible bone-derived
hMSCs on PLA scaffolds versus PLA/DBP scaffolds, calcium deposition
onto PLA/DBP scaffolds was signicantly higher after 14 days and
21 days of culture. PLA/DBP scaffolds also showed enhanced healing in
a rat cranial defect model, with almost complete bone healing
achieved in 12 weeks.
5. Conclusions and future perspectives
Over the past decade, the body of work encompassing electrospun scaffolds for regenerative medicine has expanded exponentially. The versatility of this platform is in no small part responsible
for this surge in scientic interest. Electrospun nanobers recapitulate several key features of the stem cell niche, are relatively easy to
fabricate, and most importantly are amenable to various functional
modications targeted towards enhancing stem cell survival
and proliferation, directing fate decisions, or promoting tissue
organization. Although much work has been done to establish the
biocompatibility of scaffolds for stem cell culture, there remains a

S.H. Lim, H.-Q. Mao / Advanced Drug Delivery Reviews 61 (2009) 10841096

critical need to demonstrate in a rigorous manner that such an

electrospun ber platform can augment stem cell-based regeneration in vivo. More signicantly, the vast majority of existing work has
focused primarily on inuencing differentiation of MSCs. Research
into human embryonic stem cells (hESCs) for regenerative medicine
has progressed at a rapid pace over the last ten years, ever since the
establishment of the rst hESC lines [69,70]. hESCs offer a signicant
advantage over the lineage-restricted adult-derived stem cell lines
due to their pluripotency as well as superior expansion capability.
The goal of bringing human pluripotent stem cells into the clinic
is now brought closer to realization by the successful derivation
of human induced pluripotent stem cells (iPSCs) from terminally
differentiated adult somatic cells. True validation of the electrospinning platform will come when it is successfully used to support
development of hESCs and iPSCs into functional tissue in vivo, paving
the way for widespread adoption in a variety of biomedical
applications.
Acknowledgements
This work is partially supported by National Science Foundation
Faculty Early Career Award (DMR-0748340) and Maryland Stem Cell
Research Commission (2007-MSCRFE-018).
References
[1] M.E. Furth, A. Atala, Stem cell sources to treat diabetes, J. Cell Biochem. 106 (2009)
507511.
[2] I.L. Weissman, J.A. Shizuru, The origins of the identication and isolation of
hematopoietic stem cells, and their capability to induce donor-specic transplantation tolerance and treat autoimmune diseases, Blood 112 (2008) 35433553.
[3] H. Taylor, S.L. Minger, Regenerative medicine in Parkinson's disease: generation
of mesencephalic dopaminergic cells from embryonic stem cells, Curr. Opin.
Biotechnol. 16 (2005) 487492.
[4] K. Fukuda, Progress in myocardial regeneration and cell transplantation, Circ. J. 69
(2005) 14311446.
[5] S.M. Dellatore, A.S. Garcia, W.M. Miller, Mimicking stem cell niches to increase
stem cell expansion, Curr. Opin. Biotechnol. 19 (2008) 534540.
[6] W. Potter, R.E. Kalil, W.J. Kao, Biomimetic material systems for neural progenitor
cell-based therapy, Front. Biosci. 13 (2008) 806821.
[7] K. Ghosh, D.E. Ingber, Micromechanical control of cell and tissue development:
implications for tissue engineering, Adv. Drug Deliv. Rev. 59 (2007) 13061318.
[8] C. Chai, K.W. Leong, Biomaterials approach to expand and direct differentiation of
stem cells, Mol. Ther. 15 (2007) 467480.
[9] G.A. Silva, C. Czeisler, K.L. Niece, E. Beniash, D.A. Harrington, J.A. Kessler, S.I. Stupp,
Selective differentiation of neural progenitor cells by high-epitope density
nanobers, Science 303 (2004) 13521355.
[10] T.C. McDevitt, S.P. Palecek, Innovation in the culture and derivation of pluripotent
human stem cells, Curr. Opin. Biotechnol. 19 (2008) 527533.
[11] B.M. Abdallah, M. Kassem, Human mesenchymal stem cells: from basic biology to
clinical applications, Gene. Ther. 15 (2008) 109116.
[12] P. Herve, Donor-derived hematopoietic stem cells in organ transplantation: technical
aspects and hurdles yet to be cleared, Transplantation 75 (2003) 55S57S.
[13] A. Alhadlaq, J.J. Mao, Mesenchymal stem cells: isolation and therapeutics, Stem
Cells. Dev. 13 (2004) 436448.
[14] N.S. Roy, C. Cleren, S.K. Singh, L. Yang, M.F. Beal, S.A. Goldman, Functional engraftment
of human ES cell-derived dopaminergic neurons enriched by coculture with
telomerase-immortalized midbrain astrocytes, Nat. Med. 12 (2006) 12591268.
[15] S.G. Kumbar, R. James, S.P. Nukavarapu, C.T. Laurencin, Electrospun nanober
scaffolds: engineering soft tissues, Biomed. Mater. 3 (2008) 034002.
[16] T.J. Sill, H.A. von Recum, Electrospinning: applications in drug delivery and tissue
engineering, Biomaterials 29 (2008) 19892006.
[17] K. Lin, K.N. Chua, G.T. Christopherson, S. Lim, H.Q. Mao, Reducing electrospun
nanober diameter and variability using cationic amphiphiles, Polymer 48 (2007)
63846394.
[18] M.P. Lutolf, J.A. Hubbell, Synthetic biomaterials as instructive extracellular
microenvironments for morphogenesis in tissue engineering, Nat. Biotechnol.
23 (2005) 4755.
[19] A. Spradling, D. Drummond-Barbosa, T. Kai, Stem cells nd their niche, Nature 414
(2001) 98104.
[20] D. Liang, B.S. Hsiao, B. Chu, Functional electrospun nanobrous scaffolds for
biomedical applications, Adv. Drug Deliv. Rev. 59 (2007) 13921412.
[21] H. Yoshimoto, Y.M. Shin, H. Terai, J.P. Vacanti, A biodegradable nanober scaffold
by electrospinning and its potential for bone tissue engineering, Biomaterials 24
(2003) 20772082.
[22] M. Shin, H. Yoshimoto, J.P. Vacanti, In vivo bone tissue engineering using
mesenchymal stem cells on a novel electrospun nanobrous scaffold, Tissue Eng.
10 (2004) 3341.

1095

[23] W.J. Li, R. Tuli, C. Okafor, A. Derfoul, K.G. Danielson, D.J. Hall, R.S. Tuan, A threedimensional nanobrous scaffold for cartilage tissue engineering using human
mesenchymal stem cells, Biomaterials 26 (2005) 599609.
[24] W.J. Li, R. Tuli, X. Huang, P. Laquerriere, R.S. Tuan, Multilineage differentiation of
human mesenchymal stem cells in a three-dimensional nanobrous scaffold,
Biomaterials 26 (2005) 51585166.
[25] W.J. Li, H. Chiang, T.F. Kuo, H.S. Lee, C.C. Jiang, R.S. Tuan, Evaluation of articular
cartilage repair using biodegradable nanobrous scaffolds in a swine model: a
pilot study, J. Tissue Eng. Regen. Med. 3 (2009) 110.
[26] S.N. Redman, S.F. Oldeld, C.W. Archer, Current strategies for articular cartilage
repair, Eur. Cell Mater. 9 (2005) 2332 discussion 2332.
[27] S. Aota, M. Nomizu, K.M. Yamada, The short amino acid sequence ProHisSer
ArgAsn in human bronectin enhances cell-adhesive function, J. Biol. Chem. 269
(1994) 24,75624,761.
[28] S.K. Akiyama, Integrins in cell adhesion and signaling, Hum. Cell 9 (1996)
181186.
[29] G.T. Christopherson, H. Song, H.Q. Mao, The inuence of ber diameter of
electrospun substrates on neural stem cell differentiation and proliferation,
Biomaterials 30 (2009) 556564.
[30] Y.R. Shih, C.N. Chen, S.W. Tsai, Y.J. Wang, O.K. Lee, Growth of mesenchymal stem
cells on electrospun type I collagen nanobers, Stem Cells 24 (2006) 23912397.
[31] L.S. Sefcik, R.A. Neal, S.N. Kaszuba, A.M. Parker, A.J. Katz, R.C. Ogle, E.A. Botchwey,
Collagen nanobres are a biomimetic substrate for the serum-free osteogenic
differentiation of human adipose stem cells, J. Tissue Eng. Regen. Med. 2 (2008)
210220.
[32] W.J. Li, J.A. Cooper Jr., R.L. Mauck, R.S. Tuan, Fabrication and characterization of six
electrospun poly(alpha-hydroxy ester)-based brous scaffolds for tissue engineering applications, Acta Biomater. 2 (2006) 377385.
[33] K.J. Shields, M.J. Beckman, G.L. Bowlin, J.S. Wayne, Mechanical properties and
cellular proliferation of electrospun collagen type II, Tissue Eng. 10 (2004)
15101517.
[34] C.P. Barnes, C.W. Pemble, D.D. Brand, D.G. Simpson, G.L. Bowlin, Cross-linking
electrospun type II collagen tissue engineering scaffolds with carbodiimide in
ethanol, Tissue Eng. 13 (2007) 15931605.
[35] L. Ghasemi-Mobarakeh, M.P. Prabhakaran, M. Morshed, M.H. Nasr-Esfahani, S.
Ramakrishna, Electrospun poly(epsilon-caprolactone)/gelatin nanobrous scaffolds for nerve tissue engineering, Biomaterials 29 (2008) 45324539.
[36] H. Colognato, P.D. Yurchenco, Form and function: the laminin family of
heterotrimers, Dev. Dyn. 218 (2000) 213234.
[37] R.A. Neal, S.G. McClugage, M.C. Link, L.S. Sefcik, R.C. Ogle, E.A. Botchwey, Laminin
nanober meshes that mimic morphological properties and bioactivity of
basement membranes, Tissue Eng. Part C 15 (2008) 1121.
[38] H.K. Kleinman, Preparation of basement membrane components from EHS
tumors, Curr. Protoc. Cell Biol. (1998) 10.2.110.2.10.
[39] B.G. Keselowsky, D.M. Collard, A.J. Garcia, Surface chemistry modulates focal
adhesion composition and signaling through changes in integrin binding,
Biomaterials 25 (2004) 59475954.
[40] K.N. Chua, C. Chai, P.C. Lee, Y.N. Tang, S. Ramakrishna, K.W. Leong, H.Q. Mao,
Surface-aminated electrospun nanobers enhance adhesion and expansion of
human umbilical cord blood hematopoietic stem/progenitor cells, Biomaterials 27
(2006) 60436051.
[41] K.N. Chua, C. Chai, P.C. Lee, S. Ramakrishna, K.W. Leong, H.Q. Mao, Functional
nanober scaffolds with different spacers modulate adhesion and expansion of
cryopreserved umbilical cord blood hematopoietic stem/progenitor cells, Exp.
Hematol. 35 (2007) 771781.
[42] W.J. Li, R.L. Mauck, J.A. Cooper, X. Yuan, R.S. Tuan, Engineering controllable
anisotropy in electrospun biodegradable nanobrous scaffolds for musculoskeletal tissue engineering, J. Biomech. 40 (2007) 16861693.
[43] B.M. Baker, R.L. Mauck, The effect of nanober alignment on the maturation of
engineered meniscus constructs, Biomaterials 28 (2007) 19671977.
[44] J.K. Wise, A.L. Yarin, C.M. Megaridis, M. Cho, Chondrogenic differentiation of
human mesenchymal stem cells on oriented nanobrous scaffolds: engineering
the supercial zone of articular cartilage, Tissue Eng. Part A 15 (2009) 913921.
[45] J.C. Chen, D.J. Goldhamer, Skeletal muscle stem cells, Reprod. Biol. Endocrinol. 1
(2003) 101.
[46] E.K. Yim, S.W. Pang, K.W. Leong, Synthetic nanostructures inducing differentiation
of human mesenchymal stem cells into neuronal lineage, Exp. Cell. Res. 313
(2007) 18201829.
[47] R. McBeath, D.M. Pirone, C.M. Nelson, K. Bhadriraju, C.S. Chen, Cell shape,
cytoskeletal tension, and RhoA regulate stem cell lineage commitment, Dev. Cell 6
(2004) 483495.
[48] J.M. Dang, K.W. Leong, Myogenic induction of aligned mesenchymal stem cell
sheets by culture on thermally responsive electrospun nanobers, Adv. Mater. 19
(2007) 27752779.
[49] F. Yang, R. Murugan, S. Wang, S. Ramakrishna, Electrospinning of nano/micro scale
poly(l-lactic acid) aligned bers and their potential in neural tissue engineering,
Biomaterials 26 (2005) 26032610.
[50] C.M. Cowan, C. Soo, K. Ting, B. Wu, Evolving concepts in bone tissue engineering,
Curr. Top. Dev. Biol. 66 (2005) 239285.
[51] C. Li, C. Vepari, H.J. Jin, H.J. Kim, D.L. Kaplan, Electrospun silk-BMP-2 scaffolds for
bone tissue engineering, Biomaterials 27 (2006) 31153124.
[52] A. Aigner, Nonviral in vivo delivery of therapeutic small interfering RNAs, Curr.
Opin. Mol. Ther. 9 (2007) 345352.
[53] J.N. Moreira, A. Santos, V. Moura, M.C. Pedroso de Lima, S. Simoes, Non-viral lipidbased nanoparticles for targeted cancer systemic gene silencing, J. Nanosci.
Nanotechnol. 8 (2008) 21872204.

1096

S.H. Lim, H.-Q. Mao / Advanced Drug Delivery Reviews 61 (2009) 10841096

[54] L. De Laporte, L.D. Shea, Matrices and scaffolds for DNA delivery in tissue
engineering, Adv. Drug Deliv. Rev. 59 (2007) 292307.
[55] C. Conrad, R. Gupta, H. Mohan, H. Niess, C.J. Bruns, R. Kopp, I. von Luettichau, M.
Guba, C. Heeschen, K.W. Jauch, R. Huss, P.J. Nelson, Genetically engineered stem
cells for therapeutic gene delivery, Curr. Gene Ther. 7 (2007) 249260.
[56] H. Nie, C.H. Wang, Fabrication and characterization of PLGA/HAp composite
scaffolds for delivery of BMP-2 plasmid DNA, J. Control. Release 120 (2007)
111121.
[57] S.W. Cho, H.J. Park, J.H. Ryu, S.H. Kim, Y.H. Kim, C.Y. Choi, M.J. Lee, J.S. Kim, I.S. Jang, D.I.
Kim, B.S. Kim, Vascular patches tissue-engineered with autologous bone marrowderived cells and decellularized tissue matrices, Biomaterials 26 (2005) 19151924.
[58] J.J. Stankus, J. Guan, K. Fujimoto, W.R. Wagner, Microintegrating smooth muscle
cells into a biodegradable, elastomeric ber matrix, Biomaterials 27 (2006)
735744.
[59] J.J. Stankus, L. Soletti, K. Fujimoto, Y. Hong, D.A. Vorp, W.R. Wagner, Fabrication of
cell microintegrated blood vessel constructs through electrohydrodynamic
atomization, Biomaterials 28 (2007) 27382746.
[60] J. Zhang, H. Qi, H. Wang, P. Hu, L. Ou, S. Guo, J. Li, Y. Che, Y. Yu, D. Kong, Engineering
of vascular grafts with genetically modied bone marrow mesenchymal stem cells
on poly (propylene carbonate) graft, Artif. Organs 30 (2006) 898905.
[61] C.K. Hashi, Y. Zhu, G.Y. Yang, W.L. Young, B.S. Hsiao, K. Wang, B. Chu, S. Li,
Antithrombogenic property of bone marrow mesenchymal stem cells in
nanobrous vascular grafts, Proc. Natl. Acad. Sci. U S A 104 (2007) 11,91511,920.
[62] P.M. Wigmore, G.F. Dunglison, The generation of ber diversity during myogenesis, Int. J. Dev. Biol. 42 (1998) 117125.

[63] S.A. Riboldi, N. Sadr, L. Pigini, P. Neuenschwander, M. Simonet, P. Mognol, M.

Sampaolesi, G. Cossu, S. Mantero, Skeletal myogenesis on highly orientated microbrous polyesterurethane scaffolds, J. Biomed. Mater. Res. A 84 (2008) 10941101.
[64] N.F. Huang, S. Patel, R.G. Thakar, J. Wu, B.S. Hsiao, B. Chu, R.J. Lee, S. Li, Myotube
assembly on nanobrous and micropatterned polymers, Nano. Lett. 6 (2006)
537542.
[65] M. Wang, Developing bioactive composite materials for tissue replacement,
Biomaterials 24 (2003) 21332151.
[66] O.D. Schneider, S. Loher, T.J. Brunner, L. Uebersax, M. Simonet, R.N. Grass, H.P.
Merkle, W.J. Stark, Cotton wool-like nanocomposite biomaterials prepared by
electrospinning: in vitro bioactivity and osteogenic differentiation of human
mesenchymal stem cells, J. Biomed. Mater. Res. B Appl. Biomater. 84 (2008)
350362.
[67] H. Nie, B.W. Soh, Y.C. Fu, C.H. Wang, Three-dimensional brous PLGA/HAp
composite scaffold for BMP-2 delivery, Biotechnol. Bioeng. 99 (2008) 223234.
[68] E.K. Ko, S.I. Jeong, N.G. Rim, Y.M. Lee, H. Shin, B.K. Lee, In vitro osteogenic
differentiation of human mesenchymal stem cells and in vivo bone formation in
composite nanober meshes, Tissue Eng. Part A 14 (2008) 21052119.
[69] J.A. Thomson, J. Itskovitz-Eldor, S.S. Shapiro, M.A. Waknitz, J.J. Swiergiel, V.S.
Marshall, J.M. Jones, Embryonic stem cell lines derived from human blastocysts,
Science 282 (1998) 11451147.
[70] M.J. Shamblott, J. Axelman, S. Wang, E.M. Bugg, J.W. Littleeld, P.J. Donovan, P.D.
Blumenthal, G.R. Huggins, J.D. Gearhart, Derivation of pluripotent stem cells
from cultured human primordial germ cells, Proc. Natl. Acad. Sci. U S A 95 (1998)
13,72613,731.