Optimization strategies for electrospun silk fibroin tissue engineering scaffolds
Anne J. Meinela, Kristopher E. Kubowb, Enrico Klotzschb, Marcos Garcia-Fuentesa, Michael L. Smithb,
Viola Vogelb, Hans P. Merklea, Lorenz Meinela,*
aInstitute of Pharmaceutical Sciences, ETH Zurich, Department for Chemistry and Applied Biosciences, HCI J 390.1, Wolfgang-Pauli-Strasse 10, CH-8093 Zurich, Switzerland
bLaboratory for Biologically Oriented Materials, ETH Zurich, Wolfgang-Pauli-Strasse 10, CH-8093 Zurich, Switzerland
a r t i c l e i n f o
Received 13 January 2009
Accepted 29 January 2009
Available online 23 February 2009
a b s t r a c t
As a contribution to the functionality of scaffolds in tissue engineering, here we report on advanced
scaffold design through introduction and evaluation of topographical, mechanical and chemical cues. For
scaffolding, we used silk fibroin (SF), a well-established biomaterial. Biomimetic alignment of fibers was
achieved as a function of the rotational speed of the cylindrical target during electrospinning of a SF
solution blended with polyethylene oxide. Seeding fibrous SF scaffolds with human mesenchymal stem
cells (hMSCs) demonstrated that fiber alignment could guide hMSC morphology and orientation
demonstrating the impact of scaffold topography on the engineering of oriented tissues. Beyond
currently established methodologies to measure bulk properties, we assessed the mechanical properties
of the fibers by conducting extension at breakage experiments on the level of single fibers. Chemical
modification of the scaffolds was tested using donor/acceptor fluorophore labeled fibronectin. Fluore-
scence resonance energy transfer imaging allowed to assess the conformation of fibronectin when
adsorbed on the SF scaffolds, and demonstrated an intermediate extension level of its subunits. Biological
assays based on hMSCs showed enhanced cellular adhesion and spreading as a result of fibronectin
adsorbed on the scaffolds. Our studies demonstrate the versatility of SF as a biomaterial to engineer
modified fibrous scaffolds and underscore the use of biofunctionally relevant analytical assays to opti-
mize fibrous biomaterial scaffolds.
? 2009 Elsevier Ltd. All rights reserved.
purposes should be designed such that they control cellular adhe-
sion, proliferation and differentiation, thereby guiding new tissue
formation and function . To meet such specifications, it is
necessary to control the chemical, topographical and mechanical
properties of the scaffold [2,3]. This is particularly important for
cells with high plasticity such as stem cells, which may give rise to
the generation of various tissues depending on the environment
which they are exposed to [4–6].
In this context, silk fibroin (SF) has been established as an
attractive biomaterial for scaffolding . Versatile processing
options allow the engineering of tailored architecture, mechanical
properties and surface modifications. Furthermore, as a biomate-
rial, SF features excellent biocompatibility, adaptable biodegrad-
ability and good oxygen/water vapor permeability [8–10]. Several
studies have detailed its suitabilityas a template for stem cell based
tissue engineering and have suggested its potential towards the
controlled generation of bone-, cartilage- or ligament-like tissues
Recent studies demonstrated the impact of topographical cues
on cellular performance. For example, nanocolumns affected
fibroblast morphology and gene expression, and synthetic nano-
gratings improved differentiation of human mesenchymal stem
cells (hMSCs) into neuronal lineages as compared to unpatterned
controls [6,12]. Various techniques to fabricate topographical
patterns have been reported, including lithography, polymer
demixing, electrospinning, and self-assembly . Electrospinning
is a straightforward technique for the fabrication of nanofibrous
scaffolds for tissue engineering , and can be used to generate 2D
and 3D constructs, such as sheets, tubes, stacked sheets and
wrapped sheets. Electrospun scaffolds can be manufactured to
mimic topographic features of the extracellular matrix (ECM), for
example of in vivo collagen fibers . Typically, randomly
distributed fibers are collected during electrospinning, forming
non-woven fibrous mats. However, control of fiber alignment
during electrospinning offers the potential to mimic oriented tissue
architecture such as found in ligament or muscle tissue [16,17]. So
far the potential of biomimetic fiber alignment on the formation of
* Corresponding author. Tel.: þ41 44 633 73 11; fax: þ41 44 633 13 14.
E-mail address: firstname.lastname@example.org (L. Meinel).
Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/biomaterials
0142-9612/$ – see front matter ? 2009 Elsevier Ltd. All rights reserved.
Biomaterials 30 (2009) 3058–3067
oriented tissue has been studied with synthetic polymer nanofibers
or rapidly degrading biopolymers of lowmechanical resilience such
as type I collagen . Within this context, studies using more
stable biopolymer scaffolds, such as SF, are missing.
Scaffolds for tissue engineering may also be tailored regarding
mechanical cues. For example, it is well-known that the stiffness of
an underlying matrix impacts a number of fundamental biological
processes of cells, including differentiation, matrix remodeling and
cell migration [4,19,20]. The mechanical properties of electrospun
fibers are a result of material selection, or can be controlled by
modifying the processing parameters and postspinning treatments
. Current mechanical assessments of fibrous scaffolds are based
on bulk measurements and do not provide information on the
single-fiber level. Beyond that, an evaluation of single-fiber
mechanics could be a valuable supplement to current bulk
assessments, particularly when it comes to probe the microenvi-
ronment cells encounter upon adhesion.
Finally, introducing chemical cues into a scaffold requires
a substrate that allows physical adsorption or covalent decoration
with short recognition sequences or ECM molecules to the surface,
as previously shown for SF [22,23]. For instance, to mediate
outside-in and inside-out signaling, short Arg-Gly-Asp (RGD)
motifs as well as whole ECM molecules such as fibronectin may
serve as cellular binding sites for integrin receptors and impact
cellular adhesion, migration, growth and differentiation .
Despite a limited number of studies describing the use of SF fibers
coated with ECM proteins for the support of cellular interactions
[25,26], more detailed studies about the conformation and bio-
logical impact of adsorbed proteins on SF fibers are critical to
improve control and guidance of the cell response to such scaffolds.
For instance, fibronectin is a multidomain ECM proteinwith several
molecular recognition sites functioning as chemical cues. Confor-
mational changes to fibronectin upon adsorption to model surfaces
with different chemical compositionwereshown tocause exposure
of cryptic binding sites which led to different integrin binding
specificity affecting myoblast differentiation . Therefore,
precise control of the conformational state of such ECM proteins,
when decorating biomaterials, may be essential to control the
biological performance of such implants and their in vivo success
In this study, we introduce and evaluate multiple means to
engineer environmental cues into SF scaffolds. A rotating target is
used to control scaffold topography by SF fiber alignment during
electrospinning. Postspinning treatments are applied to SF fibers,
and their mechanical properties are tested as extension at breakage
of single fibers. To improve cellular spreading on SF fibers we
decorate them with fibronectin as a chemical cue, and study its
conformational state upon adsorption. The cellular response to
such scaffolds is analyzed in vitro using hMSC cultures, an impor-
tant human progenitor cell source for tissue engineering.
2. Materials and methods
Bombyx mori (silk worm) cocoons were from Trudel Inc., Zurich, Switzerland.
Surface antigen antibodies for flow cytometry were obtained from Becton Dickinson
(Allschwil, Switzerland) and papainwas fromWorthington Biochemical Corporation
(Allschwil, Switzerland). Fetal bovine serum (FBS), Dulbecco’s modified eagle
medium (DMEM), Roswell Park Memorial Institute medium (RPMI-1640), basic
fibroblast growth factor (bFGF), penicillin, D-streptomycin, fungizone, nonessential
amino acids (NEAA, consisting of 8.9 mg/l
13.3 mg/l L-aspartic acid, 14.7 mg/l L-glutamic acid, 7.5 mg/l glycine, 11.5 mg/l L-
proline, 10.5 mg/l L-serine) and trypsin were purchased from Gibco (Carlsbad, CA).
Transforming growth factor-b1 (TGF-b1) was obtained from R&D Systems (Abing-
don, UK) and BMP-2 was kindly supplied by Wyeth (Andover, MA). 3-[2-(2-Ami-
L-alanine, 13.21 mg/l
(Geel, Belgium). All other substances were of analytical or pharmaceutical grade and
obtained from Sigma (St. Louis, MO).
2.2. Preparation of regenerated B. mori silk fibroin solution
SF was prepared using a modification of our earlier procedure . Briefly,
cocoons from B. mori were boiled in an aqueous solution of 0.02 M Na2CO3, rinsed
with ultrapurified water (UPW) and dissolved in 9 M LiBr at 55?C to generate a 10%
w/v solution. This solution was dialyzed (Pierce, Rockfort, IL, MWCO 3.500 Da)
against UPW for 48 h. After desalination a second dialysis step against PEG 6000
(200 g/1.5 l UPW) was performed to generate a SF solution of higher concentration
which was determined by weighing the remaining solid after drying. SF solution of
12.5% w/w was obtained by diluting the concentrated SF solution with UPW.
A SF/PEO blend was used for electrospinning to enable stable, continuous
spinning . PEO 900,000 solution of 5% w/w was prepared by directly adding PEO
to UPW and stirring for 5 days at room temperature. The solution was filtered
through a 5 mm syringe filter to remove remaining insoluble materials. 2 ml of PEO
solution (5% w/w) were mixed with 5 ml of SF solution (12.5%) by moderate stirring
for further use in the electrospinning process.
Electrospinning was performed in a fume hood using a cylindrical target to
collect fibers except for mechanical measurements using a similar setup as
described before . Relative humidity was adjusted by flushing the hood with dry
air. A volume flow rate of 1.2 ml/h of the SF/PEO blend through a steel capillary tube
was maintained using a syringepump. Forelectrospinning avoltage of 12–15 kV was
applied to the capillary tube using a high voltage power supply (Fabrimex, Vol-
ketswil, Switzerland). The distance between the capillary tube and the grounded
target was 19 cm. Randomly oriented fibrous scaffolds were collected on a cylin-
drical target of d¼3.8 cm when rotating at 200–250 min?1, whereas aligned fibrous
scaffolds were collected on a cylindrical target of d¼14 cm rotating at 1000 up to
2.4. Fiber treatment and fiber characterization
Electrospun fibrous scaffolds from the SF/PEO blend were either immersed in
90/10 (v/v) methanol/water for 30 min (MeOH treatment) or stored for 12 h in
a desiccator containing a saturated aqueous solution of Na2SO4?10 H2O at room
temperature (93% relative humidity, water vapor treatment) to induce an amor-
phous to beta-sheet transition of SF . Fibrous scaffolds were washed with UPW
for 48 h at 37?C to remove PEO .
Scanning electron microscopy (SEM) was used to determine the diameter of
electrospun fibers and surface texture of fibers was examined after treatments with
methanol or water vapor and washing for 48 h. Samples were coated with platinum
prior to evaluation with a LEO 1530 GEMINI scanning electron microscope (Zeiss,
Fourier transformed infrared spectroscopy (FT-IR) data were gathered on
a Nicolet 5.700-spectrometer (Thermo Fisher Scientific, Waltham, MA). Samples
were compressed into KBr pellets and each spectrum was acquired in transmittance
mode by accumulation of 256 scans with a resolution of 2 cm?1and a spectral range
of 4000–400 cm?1.
2.5. Mechanical measurements on single fibers
Thin polydimethylsiloxane (PDMS) grids with 40 mm wide and 12 mm deep
trenches were prepared similar to a method previously described . For covalent
attachment of SF fibers onto the grids, plasma cleaned grids were functionalized
with 3-[2-(2-aminoethyl-amino)ethyl-amino]propyl-trimethoxysilane (3% in UPW;
15 min) and activated with glutaraldehyde (0.5% in UPW, 30 min). Finally, the grids
were rinsed with UPWand dried. Fibers were electrospun from a SF/PEO blend with
a flow rate of 0.8 ml/h onto the PDMS grids which were fixed between two parallel
grounded metal clamps to deposit a few isolated fibers perpendicular to the grid
structure. Postspinning treatments of fibers on PDMS grids were performed as
described above using either MeOH or water vapor and subsequent washing with
A 3-axis micromanipulator (Sutter Instrument, Novato, CA) with a thin tungsten
probe attached (Nanoprobes, NY) was used to stretch single SF fibers . A sche-
matic image of the setup is shown as insert in Fig. 3. Tip displacement along the
trenches was programmed to proceed at 2 mm/step. DIC movies were recorded with
an Olympus FV 1000 confocal microscope (Olympus, Volketswil, Switzerland) to
analyze maximum fiber extension at breakage, and samples were kept in UPW
throughout. 4 batches of fibers were prepared and only fibers which showed no
detachment from PDMS grids before breakage were included in the analysis
(n ¼10–17 per batch).
A.J. Meinel et al. / Biomaterials 30 (2009) 3058–30673059
2.6. Fibronectin adsorption onto electrospun SF scaffolds and evaluation by
fluorescence resonance energy transfer (FRET) imaging
Human plasma fibronectin was isolated from fresh human plasma  and
double labeled with acceptor fluorophores (Alexa 546, Molecular Probes, Eugene,
OR) on free cysteines, and with donor fluorophores on random amines (Alexa 488,
Molecular Probes, Eugene, OR), as previously described [35–37]. Labeling yielded an
average of 6.3 donor and 3.6 acceptor moieties per fibronectin molecule. Acquisition
of FRET images was by means of confocal laser scanning microscopy (CLSM).
Acceptor and donor intensities were separately detected using 10 nm donor
(515–525 nm) and acceptor (567–577 nm) emission peaks. Since individual fibro-
nectin molecules within the multi-labeled fibronectin population carried different
fluorophore labeling ratios, absolute distances between donor and acceptor fluoro-
phores could not be calculated. Instead the ratio of acceptor to donor fluorophore
intensities (IA/ID) was measured as an indication of average fluorophore separation
and hence fibronectin conformation.
The sensitivity of double-labeled fibronectin (fibronectin-D/A) to unfolding was
evaluated by progressively denaturing fibronectin-D/A in a series of GdnHCl
concentrations from 0 to 4 M . For IA/IDmeasurements of fibronectin-D/A on SF
scaffolds, scaffolds were immersed in fibronectin solution (20 mg/ml in phosphate
buffer solution, PBS) for 75 min and labeled fibronectin was diluted with unlabeled
fibronectin (10:90) to avoid intermolecular FRET . Scaffolds were washed with
PBS and placed in a MatTek glass bottom culture dish (MatTek, Ashland, MA) filled
with PBS for imaging.
All imageswereprocessed with
Switzerland). Dark current background values were subtracted from donor and
acceptor images and a threshold mask of 100 relative intensity units was applied to
all images of Fn-D/A on SF scaffolds . Then, acceptor images were divided pixel
by pixel by donor images and histograms were computed from all data pixels within
each field of view.
2.7. Primary human mesenchymal stem cell (hMSC) isolation and expansion
Total human bone marrow (25 cm3, Cambrex, Walkersville, MD) was diluted in
isolation medium (5% FBS in RPMI-1640 medium) and centrifuged at 300?g for
10 min. Cells were pelleted and resuspended in expansion medium (DMEM,10% FBS,
100 U/ml penicillin, 1000 U/ml streptomycin, 0.5 mg/ml fungizone antimycotic, 1%
NEAA, 1 ng/ml bFGF) and seeded in 175 cm2flasks at a density of 5?104cells/cm2.
Adherent cells were allowed to reach 80% confluence (15 days for passage 0). Cells
were trypsinized and replated every 7–8 days. Second or third passage (P2, P3) cells
were used for cell culture experiments. The expression of the surface antigens CD31,
CD34 and CD105 was characterized by flowcytometry (FacsCanto, Becton Dickinson,
Basel, Switzerland), similar to what was previously described . Briefly, trypsi-
nized cells were pelleted and resuspended in RPMI medium with 10% FBS at
a concentration of 1?107cells/ml. Aliquots of the cell suspension were incubated
with R-Pe conjugated anti-CD31 (PECAM-1/endothelial cells), APC conjugated anti-
CD34 (sialomucin/hematopoietic progenitor cells) and anti-CD105 (endoglin/
endothelial cells and hMSCs) with a secondary goat-antimouse IgG FITC-conjugated
antibody. Cells were washed and fixed with 2% formalin before analysis.
To assess the potential of hMSCs for osteogenic and chondrogenic differentia-
tion, the cells were cultured in 12-well plates as micro-mass cultures (2 drops of
15 ml of 2 ?107cells/ml per well) in either control medium (DMEM,10% FBS,100 U/
ml penicillin,1000 U/ml streptomycin, 0.5 mg/ml fungizone antimycotic), osteogenic
medium (control medium supplemented with 50 mg/ml ascorbic acid-2-phosphate,
10 nM dexamethasone, 10 mM ß-glycerolphosphate, and 1 mg/ml BMP-2) or chon-
drogenic medium (control medium supplemented with 50 mg/ml ascorbic acid-2-
phosphate, 10 nM dexamethasone, 1% NEAA, 5 mg/ml insulin, and 5 ng/ml TGF-b1).
Medium was exchanged 3 times per week. After 3 weeks of culture the pellets
were washed in PBS and the amounts of glycosaminoglycans (GAG) and calcium
were measured. For the determination of calcium content micro-mass cultures
were extracted with 0.5 ml 5% trichloroacetic acid and calcium was measured
spectrophotometrically at 570 nm (Thermomax microplate reader, Molecular
Devices, Sunnyvale, CA) following the reaction with o-cresophthalein complexone
according to the manufacturer’s protocol (Rolf Greiner Biochemica, Flacht,
Germany). To measure the amount of GAG, samples were digested for 16 h with
0.5 ml papain solution (2.4 U/ml) in buffer (0.1 M disodium hydrogen phosphate,
0.01 M EDTA disodium salt, 14.4 mM
determined spectrophotometrically (Cary 300, Varian, Palo Alto, CA) at 525 nm
following binding to the dimethylene blue dye using chondroitin sulphate as
L-cysteine) at 60?C. GAG content was
2.8. Cell culture on electrospun SF scaffolds
Disc shaped scaffolds were punched (15 mm in diameter), water vapor treated
and used for cell culture after steam autoclaving. For fibronectin coating, scaffolds
were immersed in 20 mg/ml fibronectin in PBS (0.5 ml) for 75 min at RT and rinsed
with PBS. hMSCs were seeded at a density of 1 ?105cells/scaffold and cultured in
control medium at 37?C, 5% CO2on fibronectin-coated or uncoated scaffolds.
2.9. Cell adhesion assay
Prior to biochemical analysis of adherent cells measured as DNA content on
scaffolds, loosely adherent or unbound cells were removed by washing twice with
PBS. Scaffolds were transferred into a 0.1% Triton X-100 solution and disintegrated
by using steel balls and a Minibead-beater (Biospec, Bartlesville, OK). DNA content
was measured using the PicoGreen assay (Molecular Probes, Eugene, OR) according
to the manufacturer’s protocol. Aliquots of the solutions prepared from the samples
(n ¼4 per group and timepoint) were measured fluorometrically at an excitation
wavelength of 480 nm and an emission wavelength of 520 nm (FluoroCount, Pack-
ard BioScience, Meriden, CT).
2.10. Cell spreading analysis
To determine cell morphology on electrospun SF scaffolds by SEM, samples
were rinsed in 0.1 M sodium cacodylate buffer and fixed in glutaraldehyde (1.5%
glutaraldehyde in 0.1 M sodium cacodylate solution) at RT for 2 h. Samples were
rinsed in 0.1 M sodium cacodylate, dehydrated through exposure to a gradient of
alcohol and HMDS and dried in a fume hood before sputter coating with platinum
Fluorescent images were acquired after labeling cell membranes with DiI (1,10-
dicotadecyl-3,3,30,30-tetramethyl-indocarbocyanine perchlorate, Molecular Probes,
Eugene, OR). For this purpose hMSCs (5?105cells/ml) were resuspended in 4 mg/ml
DiI in control medium, and the cell suspension was incubated for 30 min on a gen-
tly rotating shaker at 37?C. Cells were washed 5 times in culture medium and
seeded on fibronectin-coated or uncoated scaffolds. Cells were then cultured for 2 h
at 37?C, 5% CO2, rinsed with PBS and fixed with 2% formalin for 20 min before
2.10.1. Statistical analysis
Presentation of data is as means?standard deviation (Fig. 7). For statistical
significance, samples were evaluated using the log rank test based on the Kaplan
Meier plot (Fig. 3) and the Student t-test (Fig. 7). Differences between groups were
considered significant for p?0.05.
3.1. Characterization of fibrous scaffolds
Non-woven fibrous scaffolds were prepared by electrospinning
of aqueous SF/PEO blends. When using a slowly rotating cylindrical
target we obtained porous scaffolds with randomly oriented
Depending on the relative humidity, uneven and beaded fibers
were obtained at 60% RH, while electrospinning below 30% RH
resulted in uniform and bead-free fibers that were 530?100 nm in
diameter (Fig. 1A,B). Therefore, fibers prepared at <30% RH were
used for further experiments. The control of fiber network orien-
tation was achieved as a function of the rotational speed of the
target (1000 up to 4000 min?1). Optimum fiber alignment was
achieved at 4000 min?1as qualitatively assessed (Fig. 1C–E).
To investigate conformational changes of SF in fibrous scaffolds
after MeOH or water vapor treatment, FT-IR structural analysis
was performed (Fig. 2A–C). MeOH treatment of electrospun SF
scaffolds resulted in a N–H bending vibration bond (amide II)
intensity shift from 1539 to 1520 cm?1compared to untreated
scaffolds. For water vapor treated SF scaffolds a peak shift with
a maximum at 1531 cm?1for the amide II band and two shoul-
ders at 1537 and 1520 cm?1was observed. Similarly, the amide I
peak was shifted from 1651 to 1632 cm?1after treatments with
MeOH or water vapor, with the shift being more pronounced for
the MeOH treated SF scaffolds. Washing of MeOH or water vapor
treated scaffolds at 37?C for 2 d resulted in the essential absence
of typical PEO peaks at 1101 (C–O–C stretching), 953 cm?1(CH2
rocking) and 843 cm?1(CH2rocking), suggesting that the water
soluble PEO had been extracted. Surface texture of electrospun
scaffolds was assessed by SEM (Fig. 2D–F). A smooth fiber surface
was observed after treatment with water vapor whereas MeOH
treatment resulted in an increase of surface roughness compared
to non-treated fibers.
A.J. Meinel et al. / Biomaterials 30 (2009) 3058–30673060
3.2. Extension at breakage of single fibers
Extension at breakage in water was measured for single elec-
trospun fibers after MeOH or water vapor treatment and PEO
extraction. This protocol has been previously shown to be appli-
cable for manually deposited fibers of fibronectin . Median
extension at breakage was 77–229% for MeOH treated and 152–
364% for water vapor treated fibers depending on the variability
between batches of measured fibers and without a consistent
statistical difference between the two treatments (Fig. 3).
3.3. Fibronectin adsorption onto fibrous scaffolds
In order to visualize the spatial distribution of fluorescently
labeled fibronectin on SF scaffolds we performed CLSM. We
observed uniform coatings of the SF fibers with labeled fibro-
nectin-D/A, as reflected by the net-like fluorescence of the scaf-
folds (Fig. 4A). Scaffolds coated with unlabeled fibronectin did not
show any fluorescent signal under the same conditions (negative
control, data not shown). To measure the structural changes of
fibronectin upon adsorption to the SF surface, we analyzed FRET
RH < 30%
RH ~ 60%
4 µm 20 µm
Fig. 1. SEM micrographs of electrospun SF/PEO fibers at different humidities (A, B). SEM micrographs of SF/PEO fibers electrospun on a cylindrical target rotating with increasing
MeOH treated and
PEO extracted fibers A
Water vapor treated and
PEO extracted fibers B
Fibers as spun C
Amide I PEO
Fig. 2. FT-IR spectra and SEM micrographs of electrospun fibers. MeOH (A,D) and water vapor treated fibers (B,E) after PEO extraction were compared to fibers as spun (C,F).
A.J. Meinel et al. / Biomaterials 30 (2009) 3058–30673061
between multiple donor and acceptor fluorophores attached to
fibronectin. Fig. 4B, C shows a color-coded ratiometric image of
IA/IDvalues of fluorescently labeled fibronectin-D/A adsorbed on
SF, in combination with a histogram indicating a relatively broad
distribution for all data pixels. Intensity ratios (IA/ID) were
correlated to known conformations of the protein in increasing
concentrations of GdnHCl (Fig. 4C) [35,37]. Median IA/IDvalues
of fibronectin adsorbed on SF scaffolds corresponded to IA/ID
scale values between 0 and 1 M GdnHCl. This indicated a partial
separation of its dimeric arms, which are crossed over in solution
3.4. hMSC characterization
hMSCs were characterized with respect to their expression of
surface antigens and the ability toselectivelydifferentiatealong the
chondrogenic and osteogenic lineages in response to environ-
mental stimuli. GAG accumulation was significantly higher in
chondrogenic medium as compared to control medium or osteo-
genic medium (Fig. 5A). Calcium depositionwas observedonly with
hMSCs cultured in osteogenic medium, but not in control or
chondrogenic medium (Fig. 5B). Flow cytometry showed that more
than 95% of the cells expressed the surface antigen CD105,
a descriptive yet not causal marker for hMSCs (Fig. 5C). Negative
expression of surface antigens CD34 and CD31 suggested the
absence of hematopoietic progenitor cells and cells of endothelial
origin (Fig. 5D,E) [40,41].
3.5. hMSC response to modified fibrous scaffolds
Modulation of cell–matrix interaction upon fibronectin coating
and fiber alignment, was investigated using hMSCs on SF scaffolds.
Two hours after seeding, cells were well spread on fibronectin-
coated random scaffolds and had a flattened shape with cellular
extensions (Fig. 6). In contrast, cells on scaffolds without fibro-
nectin coating had a spherical morphology. After72 h differences in
cell shape were indistinguishable among groups with a well spread
morphology on all scaffolds.
Measurements of the total DNA content indicated that signifi-
cantly more cells adhered to fibronectin-coated scaffolds as
compared to non-coated scaffolds at the timepoints of 0.5 and 2 h
(p<0.01) and 48 h after seeding (p<0.05, Fig. 7). No significant
differences were observed after 1, 8, 24 and 72 h.
The response of hMSCs to fiber alignment was studied after 2 h
on fibronectin-coated scaffolds. Both SEM and CLSM demonstrated
an elongated and spindle-shaped morphology of hMSCs on aligned
SF fibers with their orientation being parallel to that of the fibers
Batch 1Batch 2Batch 3
Fraction of intact fibers
Fig. 3. Extensibility of single electrospun SF fibers after MeOH (dashed line) or water vapor (solid line) treatment and PEO extraction. Data are shown for four different batches of
fibers. A schematic image of the experimental setup is depicted in the insert.
0 M 1 M 2 M 4 M
0.4 0.60.8 11.2
Intensity ratio (IA/ID)
Fig. 4. CLSM image of fluorescently labeled fibronectin adsorbed on electrospun random SF scaffold (A). IA/IDratiometric image of acceptor to donor fluorophore intensity of
adsorbed fibronectin-D/A (B). Histogram of all pixels in the field of view with overlaid solution denaturation values for fibronectin-D/A in 0–4 M GdnHCl. Hypothetical solution
structures of fibronectin in denaturant are indicated by cartoons of compact, extended and partially unfolded fibronectin (C) and were presented before .
A.J. Meinel et al. / Biomaterials 30 (2009) 3058–30673062
tension), processing parameters (e.g. electric field strength, flow
rate and collector-set up), as well as ambient parameters (e.g.
temperature and humidity) impact the process of electrospinning
[42–48]. In the present study, air humidity was critical for elec-
trospinning SF fibers. Low humidity was a prerequisite to achieve
bead-free fibers (Fig. 1A, B). This may be explained by an increased
rate of solvent (water) evaporation at dry conditions. At high
humidity, solvent removal may have been insufficient between
when the jet fluid stream left the spinneretand when it reached the
target.This incomplete removal possibly resulted in bead formation
due to residual solvent and the resulting high surface tension as has
been suggested for the electrospinning of hyaluronic acid .
Fiber alignment in scaffolds is considered as a beneficial topo-
graphical cue for the engineering of structurally oriented tissue
architecture, and may serve as a biomimetic tool to induce
phenotypic differentiation of the cells and ensure overall tissue
function [16,50]. Depending on the rotational speed of the cylin-
drical target the alignment of electrospun SF fibers could be well
controlled (Fig.1C–E) as recentlyalso shown for synthetic polymers
[51–53]. Limitations in fiber alignment by high rotational speeds, as
previously reported for collagen, potentially related to the low fiber
strength and elasticity of collagen, were not observed for SF .
The results thus demonstrate the superior processability of SF as
compared to collagen, and add another biomaterial option to the
currently available set of suitable synthetic polymers.
Treatment of SF scaffolds with methanol is frequently used to
crystallize SF rendering it insoluble in aqueous media . It is well
established that MeOH treatment triggers the transition of SF from
a predominantly random coil or silk I structure into a water insol-
uble b-sheet enriched structure [32,44]. This also holds true for our
electrospun fibrous SF scaffolds as shown by the shifts of the amide
I and II bands in their FT-IR spectra (Fig. 2A) [30,32]. Water vapor
treatment of the scaffolds resulted in similar shifts of the amide
bands (Fig. 2B) and presents a milder alternative as compared to
treatments with organic solvents, e.g. when labile growth factors
are embedded into fibers.
Bulk mechanical properties of electrospun SF scaffolds showed
4.4% and 8.5% extension at breakage for MeOH and water vapor
treated scaffolds . Starting off from these observations we
scaled the mechanical assessment down to the single-fiber level. In
these experiments, median extensions at breakage were much
higher (77–229% for MeOH treated and 152–364% for water vapor
treated fibers). In contrast to the reported bulk assessments, which
were performed in dry state at 50% RH , single-fiber measure-
ments were conducted in water. Similarly high extensions as
reported here were previously observed for SF films in water and
hydrated electrospun SF tubes after water annealing and MeOH
treatment, respectively [55,56]. A likely explanation for high
extensions at breakage in an aqueous environment is the plasti-
cizingeffect of wateras indicated for fibers from regenerated spider
silk . When compared to bulk measurements in literature ,
our single-fiber measurements further suggest that extension at
breakage in an aqueous environment is mainly governed by single-
fiber behavior (or below) and less influenced by fiber orientation,
fiber overlay or branch points in the fibrous scaffold. Nevertheless,
as concluded from the observed large variabilities of the extension
at breakage data, our experimental setup may require further
optimization and validation, and different methods for single-fiber
and bulk measurements have to be taken into account when
comparing fibrous material properties. Further use of this method
at the interface of single-fiber mechanics and cellular performances
may extend our understanding how to modulate electrospun
biomaterials towards optimized cellular interaction. Within the
context of SF, future experiments building off from these findings
will correlate the impact of elastic components (such as elastin) or
cross-linking of fibers with cell cytoskeleton assembly and differ-
P2 P3P2 P3 P2 P3
P2 P3P2 P3 P2 P3
GAG (µ µg/cell pellet)
Calcium (µ µg/cell pellet)
Fig. 5. Characterization of hMSCs. Deposition of GAG (A) and calcium (B) in passages 2 and 3 hMSCs (P2, P3) in pellets cultured in either control, osteogenic or chondrogenic
medium for 3 weeks. Flow cytometry with positive expression of the surface antigen CD105 (C) and negative expression of CD34 (D) and CD31 (E) on P2 hMSCs.
A.J. Meinel et al. / Biomaterials 30 (2009) 3058–30673063
2 h after
2 h after
2 h after
72 h after
72 h after
Silk fibroin scaffolds
Silk fibroin scaffolds
150 µm150 µm
15 µm15 µm
150 µm150 µm
15 µm15 µm
15 µm 15 µm
Fig. 6. SEM and CLSM (membrane staining with DiI) micrographs of hMSCs cultured for 2 and 72 h on electrospun random scaffolds with and without fibronectin coating.
A.J. Meinel et al. / Biomaterials 30 (2009) 3058–30673064
Decoration of scaffolds with ECM ligands as chemical cues, such
as fibronectin, is a biomimetic approach to affect scaffold func-
tionality, e.g. towards an increase in cellular adhesion. The RGD
sequence is a well-known binding site on fibronectin and other
matrix proteins for cell-surface presented integrin receptors ,
but is not present in B. mori SF . Therefore, adsorption of
fibronectin may offer the potential to increase cellular adhesion to
SF scaffolds. However, fibronectin also displays a number of other
molecular recognition sites for cells, e.g. the synergy site (PHSRN),
which enhances a5b1integrin binding to the RGD site, and for other
ECM components. It is noteworthy that some of these sites are
cryptic and that each of them must be in a certain conformation in
order to function . Thus, adsorption of fibronectin to a substrate
may not only render the surface amenable to cell adhesion, but also
changes the conformation of the molecule, which could in turn
alters its epitope exposure, leading to altered cell behavior
[27,37,62]. To investigate the adsorption of fibronectin to the silk
scaffolds and the resulting conformational change we used FRET
imaging of double fluorophore fibronectin-D/A molecules and
studied fibronectin adsorption and related structural changes of
the protein. The median intensity ratios of fibronectin-D/A on SF
scaffolds (Fig. 4B, C) were intermediate between those previously
reported for adsorption onto (hydrophilic) glass and (hydrophobic)
fluorosilane surfaces . Previous studies showed that cellular
adhesion and proliferation increased on more hydrophilic fibro-
nectin-coated surfaces. This was explained by an extension of
fibronectin subunits that promotes exposure of previously cryptic
cell integrin binding sites . Other experiments have shown that
surface chemistry, charge and topography of a substrate may affect
the ability of fibronectin to mediate cell spreading, proliferation
and differentiation. In fact, C2C12 myoblasts were reported to
proliferate or differentiate on fibronectin-coated model surfaces as
a function of fibronectin conformation [27,63,64]. Further studies
are needed to correlate fibronectin conformation on different
scaffold materials with cell response to these substrates in vitro and
in vivo in order to use FRETas a routine probe to predict a scaffold’s
Cell culture experiments were performed with hMSCs isolated
fromadult humanbone marrow. Thiscell sourceis well-established
for tissue engineering applications, based on its marked prolifera-
tion and differentiation potential (Fig. 5) . hMSCs also express
numerous integrins on their cell surface and secret various ECM
components [66,67]. In this study on fibronectin-coated fibrous SF
undergoing rapid adhesion and advanced spreading within 2 h of
culture as compared tonon fibronectin-coated scaffolds (Figs. 5 and
6). This contrasts with a previous study, in which only minor
differences in terms of adhesion and spreading were observed for
normal human epidermal keratinocytes or fibroblasts after 1 h of
cell culture on fibronectin-coated or non-coated SF fibers . A
possible explanation for the discrepancy may be that keratinocytes
their surfaces as compared to hMSCs. Furthermore, in the previous
study the aforementioned surface-dependent fibronectin confor-
mation and especially the low coating density may not have sup-
ported cellular adhesion to the same extent as observed here .
After an extended culture period of 72 h, differences in hMSC
adhesion between fibronectin-coated and non-coated scaffolds lev-
coated or non-coated scaffolds(Figs.6 and7). Thisfinding could have
been the result of cellular secretion of ECM molecules from hMSCs
overtimeand/orserumproteinadsorptionfrom the culture medium,
both of which would quickly outweigh the initial advantage of
a fibronectin-coated surface. A similar effect has been observed on
RGD-functionalized surfaces, which facilitated enhanced fibroblast
adhesion relative to unmodified surfaces up to 2 h in serum con-
taining medium, but exhibited no significant difference thereafter
hMSCmorphologyon electrospun scaffolds largelydepended on
scaffold architecture, namely whether it consisted of randomly
Silk fibroin scaffolds without fibronectin
Silk fibroin scaffolds with fibronectin
Fig. 7. DNA content on electrospun random scaffolds after 0.5–72 h of hMSC culture.
Asterisks indicate significant differences between scaffolds with and without fibro-
nectin coating (*p <0.01, **p<0.05).
150 µm 15 µm 15 µm
Fig. 8. SEM and CLSM (membrane staining with DiI) micrographs of hMSCs cultured for 2 h on fibronectin-coated aligned scaffolds.
A.J. Meinel et al. / Biomaterials 30 (2009) 3058–3067 3065
oriented or aligned SF fibers. The latter typically resulted in elon-
gated cell morphology and orientation along the main fiber axis
(Fig. 8). This corroborates previous studies made with other cells,
such as coronary artery endothelial cells, cardiomyocytes and
ligament fibroblasts [16,69–71]. The preferred cell orientation on
aligned fibers was also observed after hMSC differentiation (in
preliminary experiments in mixed adipogenic/osteogenic differ-
entiation medium, data not shown). Given the expansion and
multilineage potential of hMSCs to generate tissues of different
type and structure, scaffolds with topographical guidance may
prove useful in the engineering of structured tissues, e.g. ligament
or muscle. Changes beyond cellular morphology were demon-
strated for other cell types, such as fibroblasts and MC3T3-E1 cells
aligned on nanofibers or microgrooves impacting cellular metab-
olism and resulting in an increased and oriented deposition of
a collagen matrix [16,72]. Similar effects have yet to be studied for
hMSCs on aligned fibers.
This study explored optimization strategies for scaffold design
by introduction and evaluation of topographical, mechanical and
chemical cues. We used advanced analytical tools shifting
mechanical evaluation from bulk properties down to the single-
fiber level. The topography of electrospun scaffolds was impacted
by electrospinning conditions, particularly the rotational speed of
the cylindrical target. SF fiber alignment functioned as topo-
graphical cue leading to elongated and oriented cellular morphol-
ogies and mayopen an interesting avenue to use SF scaffolds for the
de novo engineering of structurally aligned tissues. Fibronectin
adsorbed on SF scaffolds was demonstrated by FRET to exhibit
partial extension of its dimerarms andfunctioned as a chemical cue
to enhance hMSC adhesion and spreading. Relating biologically
relevant protein conformations on scaffolds to cell responses can
provide a useful tool to further optimize scaffold surfaces towards
enhanced biological features. In conclusion, our findings pave the
way for more generally applicable optimization strategies for
biomaterial scaffold design. We thus suggest to further explore the
potential of these scaffolds for tissue engineering applications in
vitro and in vivo.
We thank Marc Simonet for helpful advice with the electro-
spinning and Sheila Luna for providing the schematic images in
Figs. 3 and 4. We further thank Trudel Inc. (Zurich, Switzerland) for
the supply of silk cocoons and Wyeth pharmaceuticals (Andover,
MA) for providing BMP-2. Funding by the BEST Bioengineering
Cluster for AJM and by the Human Frontier Science Program
Organization for MLS is greatly acknowledged.
Figures with essential color discrimination. Certain figures in
this article, in particular parts of Figs. 4 and 8, may be difficult to
interpret in black and white. The full color images can be found in
the on-line version, at doi: 10.1016/j.biomaterials.2009.01.054.
 Muschler GF, Nakamoto C, Griffith LG. Engineering principles of clinical cell-
based tissue engineering. J Bone Joint Surg Am 2004 Jul;86-A(7):1541–58.
 Toh Y, Ng S, Khong YM, Zhang X, Zhu Y, Lin P, et al. Cellular response to
a nanofibrous environment. Nanotoday 2006;1(3):34–43.
 Wong YW, Leach JB, Brown XQ. Balance of chemistry, topography, and
mechanics at the cell–biomaterial interface: issues and challenges for
assessing the role of substrate mechanics on cell reponse. Surf Sci
 Engler AJ, Sen S, Sweeney HL, Discher DE. Matrix elasticity directs stem cell
lineage specification. Cell 2006 Aug 25;126(4):677–89.
 Salasznyk RM, Williams WA, Boskey A, Batorsky A, Plopper GE. Adhesion to
vitronectin and collagen I promotes osteogenic differentiation of human
mesenchymal stem cells. J Biomed Biotechnol 2004;2004(1):24–34.
 Yim EKF, Pang SW, Leong KW. Synthetic nanostructures inducing differenti-
ation of human mesenchymal stem cells into neuronal lineage. Exp Cell Res
 Altman GH, Diaz F, Jakuba C, Calabro T, Horan RL, Chen J, et al. Silk-based
biomaterials. Biomaterials 2003 Feb;24(3):401–16.
 Meinel L, Hofmann S, Karageorgiou V, Kirker-Head C, McCool J, Gronowicz G,
et al. The inflammatory responses to silk films in vitro and in vivo. Biomate-
rials 2005 Jan;26(2):147–55.
 Wang Y, Kim HJ, Vunjak-Novakovic G, Kaplan DL. Stem cell-based tissue
engineering with silk biomaterials. Biomaterials 2006 Aug 4.
 Minoura N, Tsukada M, Nagura M. Physico-chemical properties of silk fibroin
membrane as a biomaterial. Biomaterials 1990;11(6):430–4.
 Meinel L, Karageorgiou V, Hofmann S, Fajardo R, Snyder B, Li C, et al. Engi-
neering bone-like tissue in vitro using human bone marrow stem cells and silk
scaffolds. J Biomed Mater Res A 2004 Oct 1;71(1):25–34.
 Dalby MJ, Riehle MO, Sutherland DS, Agheli H, Curtis AS. Morphological and
microarray analysis of human fibroblasts cultured on nanocolumns produced
by colloidal lithography. Eur Cell Mater 2005;9:1–8. discussion 8.
 Stevens MM, George JH. Exploring and engineering the cell surface interface.
Science 2005 Nov 18;310(5751):1135–8.
 Ma Z, Kotaki M, Inai R, Ramakrishna S. Potential of nanofiber matrix as tissue-
engineering scaffolds. Tissue Eng 2005 Jan–Feb;11(1–2):101–9.
 Nair LS, Bhattacharyya S, Laurencin CT. In: Kumar CE, editor. Nanotechnology
and tissue engineering: the scaffold based approach. Tissue Cell Organ Eng
 Lee CH, Shin HJ, Cho IH, Kang YM, Kim IA, Park KD, et al. Nanofiber alignment
and direction of mechanical strain affect the ECM production of human ACL
fibroblast. Biomaterials 2005 Apr;26(11):1261–70.
 Baker SC, Atkin N, Gunning PA, Granville N, Wilson K, Wilson D, et al. Char-
acterisation of electrospun polystyrene scaffolds for three-dimensional in vitro
biological studies. Biomaterials 2006 Jun;27(16):3136–46.
 Zhong S, Teo WE, Zhu X, Beuerman RW, Ramakrishna S, Yung LY. An aligned
nanofibrous collagen scaffold by electrospinning and its effects on in vitro
fibroblast culture. J Biomed Mater Res A 2006 Dec 1;79(3):456–63.
 Halliday NL, Tomasek JJ. Mechanical properties of the extracellular matrix
 Pelham Jr RJ, Wang Y. Cell locomotion and focal adhesions are regulated by
substrate flexibility. Proc Natl Acad Sci U S A 1997 Dec 9;94(25):13661–5.
 Wang M, Jin HJ, Kaplan DL, Rutledge GC. Mechanical properties of electrospun
silk fibers. Macromolecules 2004;37:6856–64.
 Karageorgiou V, Meinel L, Hofmann S, Malhotra A, Volloch V, Kaplan D. Bone
morphogenetic protein-2 decorated silk fibroin films induce osteogenic
differentiation of human bone marrow stromal cells. J Biomed Mater Res A
2004 Dec 1;71(3):528–37.
 Karageorgiou V, Tomkins M, Fajardo R, Meinel L, Snyder B, Wade K, et al.
Porous silk fibroin 3-D scaffolds for delivery of bone morphogenetic protein-2
in vitro and in vivo. J Biomed Mater Res A 2006 Aug;78(2):324–34.
 Giancotti FG, Ruoslahti E. Integrin signaling. Science 1999 Aug 13;285(5430):
 Min BM, Lee G, Kim SH, Nam YS, Lee TS, Park WH. Electrospinning of silk
fibroin nanofibers and its effect on the adhesion and spreading of normal
human keratinocytes and fibroblasts in vitro. Biomaterials 2004 Mar–
 Bondar B, Fuchs S, Motta A, Migliaresi C, Kirkpatrick CJ. Functionality of
endothelial cells on silk fibroin nets: comparative study of micro- and nano-
metric fibre size. Biomaterials 2007 Oct 15.
 Garcia AJ, Vega MD, Boettiger D. Modulation of cell proliferation and differ-
entiation through substrate-dependent changes in fibronectin conformation.
Mol Biol Cell 1999 Mar;10(3):785–98.
 Vogel V, Baneyx G. The tissue engineering puzzle: a molecular perspective.
Annu Rev Biomed Eng 2003;5:441–63.
 Sofia S, McCarthy MB, Gronowicz G, Kaplan DL. Functionalized silk-based
biomaterials for bone formation. J Biomed Mater Res 2001 Jan;54(1):
 Li C, Vepari C, Jin HJ, Kim HJ, Kaplan DL. Electrospun silk-BMP-2 scaffolds for
bone tissue engineering. Biomaterials 2006 Jun;27(16):3115–24.
 Hino T, Tanimoto M, Shimabayashi S. Change in secondary structure of silk
fibroin during preparation of its microspheres by spray-drying and exposure
to humid atmosphere. J Colloid Interface Sci 2003 Oct 1;266(1):68–73.
 Jin HJ, Chen J, Karageorgiou V, Altman GH, Kaplan DL. Human bone marrow
stromal cell responses on electrospun silk fibroin mats. Biomaterials 2004
 Ochsner M, Dusseiller MR, Grandin HM, Luna-Morris S, Textor M, Vogel V,
et al. Micro-well arrays for 3D shape control and high resolution analysis of
single cells. Lab Chip 2007 Aug;7(8):1074–7.
 Sun Y, Nelson BJ. MEMS capacitive force sensors for cellular and flight
biomechanics. Biomed Mater 2007;(2):S16–22.
invitro. ExpCell Res 1995
A.J. Meinel et al. / Biomaterials 30 (2009) 3058–30673066
 Smith ML, Gourdon D, Little WC, Kubow KE, Eguiluz RA, Luna-Morris S, et al. Download full-text
Force-induced unfolding of fibronectin in the extracellular matrix of living
cells. PLoS Biol 2007 Oct 2;5(10):e268.
 Baneyx G, Baugh L, Vogel V. Coexisting conformations of fibronectin in cell
culture imaged using fluorescence resonance energy transfer. Proc Natl Acad
Sci U S A 2001 Dec 4;98(25):14464–8.
 Baugh L, Vogel V. Structural changes of fibronectin adsorbed to model surfaces
probed by fluorescence resonance energy transfer. J Biomed Mater Res A 2004
 Farndale RW, Buttle DJ, Barrett AJ. Improved quantitation and discrimination
of sulphated glycosaminoglycans by use of dimethylmethylene blue. Biochim
Biophys Acta 1986 Sep 4;883(2):173–7.
 Klotzsch E, Smith ML, Kubow KE, Gourdon D, Muntwyler S, Beyerle F, et al.
Fibronectin forms the most elastic biological fibers, unpublished results.
 Negrin RS, Atkinson K, Leemhuis T, Hanania E, Juttner C, Tierney K, et al.
Transplantation of highly purified CD34þ Thy-1þ hematopoietic stem cells in
patients with metastatic breast cancer. Bio Blood Marrow Transplant
 DeLisser HM, Newman PJ, Albelda SM. Platelet endothelial cell adhesion
molecule (CD31). Curr Top Microbiol Immunol 1993;184:37–45.
 Zong XH, Kim K, Fang DF, Ran SF, Hsiao BS, Chu B. Structure and process
relationship of electrospun bioabsorbable nanofiber membranes. Polymer
 Fong H, Chun I, Reneker DH. Beaded nanofibers formed during electro-
spinning. Polymer 1999;40(16):4585–92.
 Jin HJ, Fridrikh SV, Rutledge GC, Kaplan DL. Electrospinning Bombyx mori silk
with poly(ethylene oxide). Biomacromolecules 2002 Nov–Dec;3(6):1233–9.
 Sukigara S, Gandhi M, Ayutsede J, Micklus M, Ko F. Regeneration of Bombyx
mori silk by electrospinning-part 1: processing parameters and geometric
properties. Polymer 2003;44(19):5721–7.
engineering applications: a review. Tissue Eng 2006 May;12(5):1197–211.
 Mit-Uppatham C, Nithtanakal M, Supaphol P. Ultrafine electrospun poly-
amide-6 fibers: effect of solution conditions on morphology and average fibers
diameter. Macromol Chem Phys 2004;205(17):2327–38.
 Caspar CL, Stephens JS, Tassi NG, Chase DB, Rabolt JF. Controlling surface
morphology of electrospun polystyrene fibers: effect of humidity and
molecular weight in the electrospinning process. Macromolecules 2004;37(2):
 Um IC, Fang D, Hsiao BS, Okamoto A, Chu B. Electro-spinning and electro-
blowing of hyaluronic acid. Biomacromolecules 2004 Jul–Aug;5(4):1428–36.
 Li WJ, Mauck RL, Cooper JA, Yuan X, Tuan RS. Engineering controllable
anisotropy in electrospun biodegradable nanofibrous scaffolds for musculo-
skeletal tissue engineering. J Biomech 2006 Oct 20.
 Courtney T, Sacks MS, Stankus J, Guan J, Wagner WR. Design and analysis of
tissue engineering scaffolds that mimic soft tissue mechanical anisotropy.
Biomaterials 2006 Jul;27(19):3631–8.
 Xu CY, Inai R, Kotaki M, Ramakrishna S. Aligned biodegradable nanofibrous
structure: a potential scaffold for blood vessel engineering. Biomaterials 2004
 Baker BM, Mauck RL. The effect of nanofiber alignment on the maturation of
engineered meniscus constructs. Biomaterials 2007 Apr;28(11):1967–77.
 Min BM, Jeong L, Lee KY, Park WH. Regenerated silk fibroin nanofibers: water
vapor-induced structural changes and their effects on the behavior of normal
human cells. Macromol Biosci 2006 Apr 12;6(4):285–92.
 Jin HJ, Park J, Karageorgiou V, Kim UJ, Valluzzi R, Cebe P, et al. Water-stable silk
films with reduced b-sheet content. Adv Funct Mater 2005;15:1241–7.
 Soffer L, Wang X, Zhang X, Kluge J, Dorfmann L, Kaplan DL, et al. Silk-based
electrospun tubular scaffolds for tissue-engineered vascular grafts. J Biomater
Sci Polym Ed 2008;19(5):653–64.
 Shao Z, Vollrath F, Yang Y, Thogersen HC. Structure and behavior of regen-
erated spider silk. Macromolecules 2003;36:1157–61.
 Peyton SR, Raub CB, Keschrumrus VP, Putnam AJ. The use of poly(ethylene
glycol) hydrogels to investigate the impact of ECM chemistry and mechanics
on smooth muscle cells. Biomaterials 2006 Oct;27(28):4881–93.
 Pierschbacher MD, Ruoslahti E. Cell attachment activity of fibronectin can
be duplicated by small synthetic fragments of the molecule. Nature 1984 May
 Mita K, Ichimura S, James TC. Highly repetitive structure and its organization
of the silk fibroin gene. J Mol Evol 1994 Jun;38(6):583–92.
 Vogel V. Mechanotransduction involving multimodular proteins: converting
 Keselowsky BG, Collard DM, Garcia AJ. Integrin binding specificity regulates
biomaterial surface chemistry effects on cell differentiation. Proc Natl Acad Sci
U S A 2005 Apr 26;102(17):5953–7.
 McClary KB, Ugarova T, Grainger DW. Modulating fibroblast adhesion,
spreading, and proliferation using self-assembled monolayer films of
alkylthiolates on gold. J Biomed Mater Res 2000 Jun 5;50(3):428–39.
 Keselowsky BG, Collard DM, Garcia AJ. Surface chemistry modulates fibro-
nectin conformation and directs integrin binding and specificity to control cell
adhesion. J Biomed Mater Res A 2003 Aug 1;66(2):247–59.
 Bobis S, Jarocha D, Majka M. Mesenchymal stem cells: characteristics and
clinical applications. Folia Histochem Cytobiol 2006;44(4):215–30.
 Docheva D, Popov C, Mutschler W, Schieker M. Human mesenchymal stem
cells in contact with their environment: surface characteristics and the
integrin system. J Cell Mol Med 2007 Jan–Feb;11(1):21–38.
 Majumdar MK, Keane-Moore M, Buyaner D, Hardy WB, Moorman MA,
McIntosh KR, et al. Characterization and functionality of cell surface molecules
on human mesenchymal stem cells. J Biomed Sci 2003 Mar–Apr;10(2):228–41.
 Karakecili AG, Demirtas TT, Satriano C, Gumusderelioglu M, Marletta G.
Evaluation of L929 fibroblast attachment and proliferation on Arg-Gly-Asp-Ser
J Biosci Bioeng 2007 Jul;104(1):69–77.
 Zong X, Bien H, Chung CY, Yin L, Fang D, Hsiao BS, et al. Electrospun fine-
textured scaffolds for heart tissue constructs. Biomaterials 2005 Sep;26(26):
 Yang F, Murugan R, Wang S, Ramakrishna S. Electrospinning of nano/micro
scale poly(L-lactic acid) aligned fibers and their potential in neural tissue
engineering. Biomaterials 2005 May;26(15):2603–10.
 He W, Yong T, Ma ZW, Inai R, Teo WE, Ramakrishna S. Biodegradable polymer
nanofiber mesh to maintain functions of endothelial cells. Tissue Eng 2006
 Wang JH, Jia F, Gilbert TW, Woo SL. Cell orientation determines the alignment
of cell-produced collagenous matrix. J Biomech 2003 Jan;36(1):97–102.
A.J. Meinel et al. / Biomaterials 30 (2009) 3058–30673067