Acta Biomaterialia xxx (2013) xxxxxx

Contents lists available at ScienceDirect

Acta Biomaterialia
journal homepage: www.elsevier.com/locate/actabiomat

Jellysh collagen scaffolds for cartilage tissue engineering

Birgit Hoyer a,, Anne Bernhardt a, Anja Lode a, Sascha Heinemann b, Judith Sewing c,d, Matthias Klinger e,
Holger Notbohm c, Michael Gelinsky a
a

University Hospital and Medical Faculty, Technische Universitt Dresden, Centre for Translational Bone, Joint and Soft Tissue Research, Fetscher Str. 74, D-01307 Dresden, Germany
Max Bergmann Center of Biomaterials and Institute for Materials Science, Technische Universitt Dresden, Budapester Str. 27, 01069 Dresden, Germany
c
Institute of Virology and Cell Biology, University of Lbeck, Ratzeburger Allee 160, 23562 Lbeck, Germany
d
CRM Coastal Research & Management GmbH, Tiessenkai 12, 24159 Kiel, Germany
e
Institute of Anatomy, University of Lbeck, Ratzeburger Allee 160, 23562 Lbeck, Germany
b

a r t i c l e

i n f o

Article history:
Received 27 May 2013
Received in revised form 28 August 2013
Accepted 22 October 2013
Available online xxxx
Keywords:
Marine collagen
Fibril formation
Jellysh
Mesenchymal stem cells
Chondrogenic differentiation

a b s t r a c t
Porous scaffolds were engineered from rebrillized collagen of the jellysh Rhopilema esculentum for
potential application in cartilage regeneration. The inuence of collagen concentration, salinity and
temperature on bril formation was evaluated by turbidity measurements and quantication of
brillized collagen. The formation of collagen brils with a typical banding pattern was conrmed by
atomic force microscopy and transmission electron microscopy analysis. Porous scaffolds from jellysh
collagen, rebrillized under optimized conditions, were fabricated by freeze-drying and subsequent
chemical cross-linking. Scaffolds possessed an open porosity of 98.2%. The samples were stable under
cyclic compression and displayed an elastic behavior. Cytotoxicity tests with human mesenchymal stem
cells (hMSCs) did not reveal any cytotoxic effects of the material. Chondrogenic markers SOX9, collagen II
and aggrecan were upregulated in direct cultures of hMSCs upon chondrogenic stimulation. The formation of typical extracellular matrix components was further conrmed by quantication of sulfated
glycosaminoglycans.
2013 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

1. Introduction
Collagen is the prevailing component of extracellular matrices
in connective tissues. Due to its biocompatibility, biodegradability,
low immunogenicity and cell-adhesive properties, collagen is one
of the most frequently utilized materials in the eld of tissue
engineering [1]. Collagen can be extracted from a variety of
organisms. Preferential sources of collagen for tissue engineering
applications are bovine skin and tendon as well as porcine skin.
However, collagen of bovine origin involves the risk of infection
with diseases such as bovine spongiform encephalopathy. Furthermore, mammalian collagens, especially of porcine origin, are
increasingly rejected for religious reasons [2]. Marine organisms
are an alternative natural source of collagen and, presumably, are
safer compared to mammals. Recent publications focus mainly
on the isolation and characterization of collagen from different sh
species, such as salmon [3], shark [4] or deep sea redsh [5] and
marine sponges [6]. Another attractive marine source for the
extraction of collagen is jellysh. The global increase in jellysh
population causes major problems in the ecological environment,
and their potential use in tissue engineering, next to food industry
Corresponding author. Tel.: +49 351 458 6694; fax: +49 351 458 7210.
E-mail address: birgit.hoyer@tu-dresden.de (B. Hoyer).

and medicine, may help to reduce their further expansion [7]. With
a collagen content of more than 60% [8], jellysh has the potential
to become a signicant source of collagen in biomedical
applications [9]. Isolation and molecular characterization of jellysh collagen derived from Stomophulus nomurai has been reported
decades ago [10,11]. More recent investigations are concerned
with collagen from other jellysh species, e.g. Rhopilema asamushi,
Stomolophus meleagris, Catostylus tagi and Rhizostoma pulmo [7
9,12]. High collagen recovery rates have consistently been
reported. Amino acid analyses revealed a composition similar to
vertebrate collagen with, however, a lower content of hydroxyproline, which leads to relatively low denaturation temperatures
between 26 and 29.9 C. Differences in the subunit composition
of collagens from different species are detectable in their respective electrophoretic pattern. Hence, it may be stated that collagens
of different jellysh species show similarities to different vertebrate collagen types. Some jellysh collagens are comparable to
vertebrate collagen IV or V [10,11], others seem to resemble vertebrate collagen I [7,12,13] and some show a unique structure with a
fourth a-chain [9]. Hsieh [14] postulates that jellysh collagen of S.
meleagris is similar to vertebrate collagen type II according to the
molecular mobility, salting-out concentration, high content of
hydroxylysine, solubility properties, absence of disulde bonds
and highly hygroscopic nature. Similar ndings are described by

1742-7061/$ - see front matter 2013 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.actbio.2013.10.022

Please cite this article in press as: Hoyer B et al. Jellysh collagen scaffolds for cartilage tissue engineering. Acta Biomater (2013), http://dx.doi.org/10.1016/
j.actbio.2013.10.022

B. Hoyer et al. / Acta Biomaterialia xxx (2013) xxxxxx

Bermueller et al. for jellysh collagen of Rhopilema esculentum,

which consists of only one type of a-chain and shows a degree of
glycosylation similar to that of vertebrate collagen type II [15].
Since collagen II, as the main component of cartilage extracellular
matrix, has been shown to support the chondrogenic differentiation and maintenance of chondrogenic phenotype to a higher
extent compared to collagen I [16], jellysh collagen derived from
R. esculentum might be a predominant prospective material for the
preparation of scaffolds in cartilage tissue engineering.
So far only a few publications have reported on the preparation of jellysh collagen-based scaffolds for tissue engineering
applications. Song et al. [17] generated porous scaffolds by
freeze-drying and subsequent chemical cross-linking of acidsolubilized jellysh collagen. Biocompatibility investigations
comprised the attachment of human broblasts as well as the
immune response after implantation of the scaffolds in vivo,
being similar to other collagen sources. Another type of porous
scaffold was established by the same group, combining jellysh
collagen and hyaluronic acid [18]. Tubular porous scaffolds from
jellysh collagen reinforced with poly(lactic-co-glycolic) acid
bers were developed by freeze-drying and electrospinning techniques [19,20]. Seeded with osteosarcoma cells, these constructs
were cultivated to test the inuence of electrospinning parameters on cell adhesion and proliferation, or, when seeded with
endothelial cells and smooth muscle cells in a perfusion system,
to generate vascular grafts.
There are three different types of cartilage: hyaline, brocartilaginous and elastic. All cartilage types consist of chondrocytes
and an extracellular matrix with collagens, proteoglycans and
water, differing in protein types and proportions [21]. Joint surfaces are covered by hyaline cartilage, which is predominantly collagen type II (9095%), followed by collagen XI and IX [22,23]. On
the contrary, in brocartilage (in knee joint menisci and intervertebral discs), which has properties of both connective tissues and
hyaline cartilage, collagen type I and fewer proteoglycans are present [24]. Analogous to hyaline cartilage, elastic cartilage is mainly
composed of collagen type II. The main difference to the other two
types is elastin and a higher number of cells [21]. Cartilage repair is
very challenging since all three types of cartilage tissue are nonvascularized and have therefore a poor intrinsic regeneration
capacity.
One approach to the regeneration of cartilage defects is the
implantation of a tissue-engineered scaffold colonized with cells.
In clinical practice autologous chondrocytes are used for this
purpose. There are, however, limitations to this procedure due to
the induction of morbidity at the donor site and instability in
monolayer cultivation [25]. Mesenchymal stem cells from various
tissues, such as bone marrow or adipose tissue [25], represent an
alternative source of superior availability and are for that reason
the object of research in cartilage tissue engineering [26,27].
Materials for scaffolds in cartilage tissue engineering consist
mainly of natural or synthetic polymers [26], both in the shape
of either hydrogels [26,28] or porous matrices [27,29,30]. Natural
polymers include agarose, collagen, silk, alginate or chitosan
[28,3032]. In the clinical environment, mammalian collagen is already used in the form of membranes, e.g. CaReS (Arthro Kinetics
AG, Germany), Chondro-Gide (Geistlich Pharma AG, Switzerland),
Cartimaix (Matricel GmbH, Germany), Novocart 3D (TETEC Tissue Engineering Technologies AG, Germany) or hydrogels, e.g.
ChondroFiller (Amedrix, Germany).
The aim of the present study was to characterize bril formation parameters of jellysh collagen to nd optimal parameters
for the fabrication of stable porous scaffolds from rebrillized
jellysh collagen. Furthermore we wanted to evaluate the
suitability of the scaffolds for potential usage in cartilage tissue
engineering.

2. Materials and methods

2.1. Collagen
Collagen was extracted by pepsin digestion from cured jellysh
R. esculentum (LiroyBV, Rotterdam, The Netherlands) as described
before [33]. Briey, salted jellysh was cut into pieces and extensively rinsed with cold water until salinity was 60.01. After
equilibration in 0.5 M acetic acid for at least 30 min, pieces of jellysh were homogenized. Following 60 h of pepsin digestion at
4 C, the solution was centrifuged for cleaning. Collagen was then
precipitated from the supernatant by adding NaH2PO4 and KCl at
pH 7 for 12 h. After another centrifugation step the collected collagen was redissolved in 0.05% acetic acid and dialyzed against 0.05%
acetic acid. It was stored at 20 C and lyophilized only when
required to avoid unwanted cross-linking. Prior to use, a stock
solution was generated under constant stirring at 4 C with a concentration of 5 mg ml1 lyophilized collagen dissolved in 0.01 M
HCl.
2.2. Fibril formation
For reassembly analysis, 500 ll collagen stock solution in
graded concentrations of 5, 4, 3, 2 and 1 mg ml1 were thoroughly
mixed with 500 ll of 50 mM tris-(hydroxymethyl)-aminomethane
(Tris) buffer containing graded sodium chloride concentrations of
20, 40, 60, 80 and 100 mM; nal pH was adjusted to 7.4. Turbidity
was measured at 313 nm over 60 min at 4 C or 25 C using a UV
Vis spectrophotometer Cary 50 Bio (Varian, Germany). Fibril
formation was quantied indirectly by measuring the collagen
concentration in the supernantant using the modied Bradford assay, as described previously [34,35]. In brief, upon completion of
the bril formation process over 4 h at 4 or 25 C, the suspension
was centrifuged for 15 min at 10,000g, at 4 C. 5 ll of the supernantant were mixed with 250 ll Bradford reagent (SigmaAldrich,
USA) containing 0.035 mg ml1 SDS (SigmaAldrich). Absorbance
was measured after 15 min at 590 nm using a microplate reader
Innite M200Pro (Tecan, Switzerland). For calibration, a graded
series of jellysh collagen stock solution was used. Finally, the ratio
of brillized to total initial collagen was calculated, revealing the
degree of bril formation. The graphs show the mean standard
deviation (n = 6).
2.3. Scaffold preparation
Based on the results of the bril formation studies we developed the following protocol for scaffold preparation. A 5 mg ml1
stock solution of jellysh collagen was merged 1:1 with 50 mM
Tris buffer, pH 8. This preparation with a resulting pH 7.4 was stirred for 12 h at 4 C. Upon centrifugation at 5000g and 4 C, the
pellet was resuspended in a small amount of supernantant and
transferred to 96-well cell culture dishes. Three-dimensional (3D) sponge-like, porous scaffolds were obtained after freezing at a
speed of 1 K min1 and subsequent freeze-drying (Alpha 12,
Christ, Germany) for 24 h. The scaffolds were chemically crosslinked in a 1 wt.% solution of N-(3-dimethylaminopropyl)-N0 -ethylcarbodiimide hydrochloride (EDC, Fluka, Germany) in 80 vol.% ethanol for 2 h. After careful rinsing in deionized water, in 1 wt.%
glycine solution, and once more in deionized water, a nal
freeze-drying step concluded the procedure.
2.4. Atomic force microscopy (AFM)
Droplets of rebrillized jellysh collagen suspended in 50 mM
Tris buffer, pH 8, were transferred onto the surface of mica discs.

Please cite this article in press as: Hoyer B et al. Jellysh collagen scaffolds for cartilage tissue engineering. Acta Biomater (2013), http://dx.doi.org/10.1016/
j.actbio.2013.10.022

B. Hoyer et al. / Acta Biomaterialia xxx (2013) xxxxxx

The collagenous material was allowed to adsorb for 30 min followed by rinsing with deionized water and drying in air. AFM
imaging was performed in tapping mode in air using a Nanoscope
IIIa Bioscope (Digital Instruments/Veeco, USA) and aluminum reex coated silicon tips (force constant 40 N m1). Both deection
and height images were captured simultaneously at imaging
speeds of 1.2 Hz, scanning 512 lines. Measurements were taken
from height images using the Nanoscope software.
2.5. Transmission electron microscopy (TEM)
5 ll graded dilutions of rebrillized jellysh collagen were applied to formvar-coated copper grids (SF 162-3, Plano GmbH,
Germany), negatively stained with 1% phosphotungstic acid in distilled water, and allowed to dry. Grids were examined with a
Philips TEM 400 at 60 kV. Fibril diameter was measured semiquantitatively from two pictures with Axio Vision 3.1 software
(Zeiss, Germany), in which several sections of the widest brils
were measured.
2.6. Scanning electron microscopy (SEM)
Cell-seeded scaffolds were xed with 2.5% glutaraldehyde in
phosphate-buffered saline (PBS), followed by dehydration in
graded series of ethanol and critical point drying (BAL-TEC CPD
30, Liechtenstein). All samples were xed on carbon pads and sputter-coated with gold. A Philips XL 30/ESEM with eld emission gun
operated in SEM mode was used.
2.7. Confocal laser scanning microscopy (cLSM)
Cell-seeded scaffolds were xed with 3.7% formaldehyde in PBS.
After treatment with 0.2% Triton X-100 in PBS for 5 min, samples
were rinsed ve times with PBS, followed by blocking of autouorescence with 3% bovine serum albumin (SigmaAldrich) in PBS.
The cytoskeleton of the cells was stained using Alexa Fluor 488
phalloidin (Invitrogen, USA) and nuclei with DAPI (SigmaAldrich).
Samples were imaged using a Zeiss cLSM 510.
2.8. Porosity
The real volume of freeze-dried jellysh collagen sponges was
evaluated in a helium pycnometer (Ultrapyc1200e, Quantachrome
Instruments, USA). Height, diameter and weight were measured to
calculate porosity.
2.9. Differential scanning calorimetry (DSC)
After thoroughly rinsing in deionized water, samples were
sealed in 50 ll melting pans of aluminum (BO14-3003 and BO
14-3017, Perkin Elmer, USA) and nally measured in a Pyris 6
DSC, software version 9.1 (Perkin Elmer), with a heating rate of
2 K min1. An empty melting pan was used as reference sample.

2.11. Cell culture

Human mesenchymal stem cells (hMSCs) (kindly provided by
Professor Martin Bornhuser and co-workers, Medical Clinic I,
Dresden University Hospital Carl Gustav Carus) were isolated from
bone marrow aspirate of two healthy male donors (age 30 and 32)
after providing written informed consent. The application of
hMSCs for in vitro experiments was approved by the ethics committee of the Medical Faculty of Technische Universitt Dresden.
Cells were expanded in Dulbeccos modied Eagles medium
(DMEM) low glucose (Biochrom, Germany), containing 10% fetal
calf serum, 100 U ml1 penicillin and 100 lg ml1 streptomycin
(Biochrom) at 37 C in a humidied, 5% CO2/95% air incubator.
Prior to cell seeding, cylindrical, gamma-sterilized jellysh collagen scaffolds (diameter: 6 mm; height: 3 mm) were incubated in
cell culture expansion medium for 24 h. Upon removal of excess liquid, each scaffold was seeded with 1.2  106 cells in expansion
medium. After 24 h the culture medium was replaced with chondrogenic medium containing 100 U ml1 penicillin and
100 lg ml1 streptomycin, 10 lg ml1 insulin, 5.5 lg ml1 transferrin, 6.7 ng ml1 selenium, 0.2 lg ml1 ethanolamine (ITS-X
mix, Gibco, Germany), 107 M dexamethasone (SigmaAldrich),
0.2 mM ascorbic acid-2-phosphate (SigmaAldrich), 10 ng ml1
TGF-b3 (Milteny Biotec, Germany) and 35 lM proline (SigmaAldrich). The rst medium change was 1 day after seeding and later
twice a week over 21 days. Samples for biochemical analysis (n = 3)
were washed twice with PBS and frozen at 80 C in 2 ml Nalgene
tubes containing six ceramic beads (Peqlab, Germany). For SEM
investigations, samples were washed with PBS and xed in 2.5%
glutaraldehyde in PBS. Samples for gene expression analysis
(n = 3) were taken after 1 and 21 days of cultivation, washed with
PBS and immediately subjected to RNA isolation as described
below.

2.12. Cytotoxicity test

Cytotoxicity of jellysh collagen scaffolds was tested by indirect
cultivation of cells with scaffold extracts. Jellysh collagen scaffolds (diameter: 6 mm; height: 3 mm) were incubated in 1 ml cell
culture medium (DMEM containing 10% fetal calf serum,
100 U ml1 penicillin and 100 lg ml1 streptomycin) at 37 C.
8  103 hMSCs were seeded in 48-well tissue culture plates and
cultivated for 1, 7 or 14 days, respectively, either with scaffoldconditioned medium or normal medium. On completing the
respective cultivation period, samples were washed twice in PBS
and frozen at 80 C for later analysis. After thawing, samples were
lysed with 1% Triton X-100 in PBS for 50 min on ice. During lysis,
samples were incubated for 10 min in an ice-cooled ultrasonic
bath. The CytoTox 96 non-radioactive cytotoxicity assay (Promega, USA) was used for determination of the lactate dehydrogenase
(LDH) activity according to the manufacturers instructions. Absorbance was read at 492 nm in a microplate reader. LDH activity of
the samples was correlated with the number of cells using a calibration line of dened cell numbers. Graphs show mean standard
deviation (n = 3).

2.10. Mechanical measurements

2.13. MTT staining
Uniaxial compressive tests were conducted using an Instron
5566 testing machine (Instron GmbH, Germany). Ten cylindrical
samples (diameter 10.3 0.3 mm, height 10.7 0.2 mm) were
incubated in modied simulated body uid (SBF) [36] for 24 h
prior to measurements. Wet samples were compressed to 50% of
their initial height. For static tests, a velocity of 0.1 mm s1 and
for cyclic tests (50 cycles) a velocity of 0.3 mm s1 was used.

After 21 days of cultivation, a cell-seeded and a non-seeded

scaffold were incubated in culture medium with 1.2 mM 3-(4,5dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT;
SigmaAldrich) at 37 C for 4 h. Intracellular conversion of MTT
into a dark blue formazan derivative was macroscopically
evaluated.

Please cite this article in press as: Hoyer B et al. Jellysh collagen scaffolds for cartilage tissue engineering. Acta Biomater (2013), http://dx.doi.org/10.1016/
j.actbio.2013.10.022

B. Hoyer et al. / Acta Biomaterialia xxx (2013) xxxxxx

2.14. Determination of sulfated glycosaminoglycan (sGAG) and DNA

content
1000 ll papain digestion solution (125 lg ml1 papain, 100 mM
Na2HPO4, 5 mM ethylenediaminetetraacetic acid (EDTA), 5 mM cystein in deionized water) was added to the frozen cell-seeded scaffolds. The samples were homogenized (2  10 s at 5900 rpm) using
a Precellys 24 apparatus (Peqlab) and were then incubated at
60 C for 24 h. Subsequently, 50 ll of the samples were applied
for quantication using the sGAG assay (Kamiya Biomedical Company, USA) according to the manufacturers instructions. Absorbance was assessed at 610 nm in a microplate reader. Another
aliquot of the cell lysate was mixed with the Quant-iT PicoGreen dsDNA reagent (Molecular Probes, USA), diluted 1:800 in
TE buffer (10 mM TRIS, 1 mM EDTA) and incubated for 5 min in
the dark. The intensity of uorescence was measured at excitation/emission wavelengths of 485/535 nm, respectively. Relative
uorescence units were correlated to the amount of DNA using a
calf thymus DNA calibration line. Graphs of sGAG concentration
in relation to DNA content show mean standard deviation (n = 6).
2.15. Reverse transcriptase PCR (RT-PCR)
RNA was extracted from cell-seeded scaffolds using the
peqGOLD Micro Spin Total RNA Kit (Peqlab) according to the manufacturers instructions. During the RNA isolation procedure, cell
lysates of three samples cultured under identical conditions were
pooled. 100 ng of total RNA were reverse transcribed into cDNA
in a 20 ll reaction mixture containing 200 U of Superscript II
Reverse Transcriptase (Invitrogen, Germany), 0.5 mM dNTPs (Invitrogen), 12.5 ng ll1 random hexamers (Eurons MWG Operon,
Germany) and 40 U of RNase inhibitor RNase OUT (Invitrogen).
1 ll cDNA in 20 ll reaction mixtures containing specic primer
pairs were used for amplication in PCR analysis to detect transcripts of collagen I, collagen IIa, collagen X, aggrecan, SOX9,
RUNX2 and b-actin, respectively. Primer sequences (Eurons
MWG Operon), annealing temperatures and amplicon sizes for
each gene are summarized in Table 1. The PCR experiments were
carried out in a Thermocycler (Vapo-Protect Mastercycler Pro,
Eppendorf, Germany) and the resulting PCR products were visualized using the FlashGel Dock system (Cambrex Bio Science, USA).
2.16. Statistics
Mean and standard deviation are shown in the gures. The difference in the proportion of reassembled collagen within each
parameter was statistically evaluated with two-way ANOVA and
post hoc Tukey test via Origin 8.6.

3. Results
3.1. Inuence of collagen concentration, ionic strength and
temperature on bril formation
3.1.1. Inuence of nal collagen concentration
The inuence of the nal collagen concentration on bril formation was investigated in the range of 0.52.5 mg ml1 by monitoring the change of turbidity at 313 nm (Fig 1A and B) as well as by
indirect evaluation of the degree of bril formation via modied
Bradford assay (Fig. 1C). Turbidity started to rise shortly after
mixing the solutions and was too fast to be monitored in the
spectrometer. Nevertheless, a clear difference in the turbidity as
a function of the collagen concentration could be observed in the
plateau phases. The degree of bril formation was not signicantly
inuenced by the collagen concentration.
3.1.2. Inuence of NaCl concentration
With an increasing amount of sodium chloride from 0 to
100 mM (Fig. 2A and B) in the suspension, the turbidity plateau decreased. Sodium chloride had a signicant inuence (p < 0.001) on
the degree of bril formation (Fig. 2C). Increase of the salinity over
20 mM reduced the amount of jellysh collagen brils.
3.1.3. Inuence of temperature
Comparing the collagen concentrations of 2.5 mg ml1 at 4 and
at 25 C (Fig. 3A) revealed slightly lower turbidity plateaus at 25 C.
A similar tendency was observed with different TRIS buffer concentration (data not shown). At 25 C the amount of reassembled collagen was signicantly (p < 0.05) diminished (Fig. 3B).
Investigations of varying sodium chloride concentration did not
show such a difference between the two tested temperatures.
3.2. Morphology of jellysh collagen brils
The morphology of reassembled jellysh collagen brils was
visualized by AFM. As shown in Fig. 4A, branched brils, varying
in thickness and length, were observed. The small brils measured
32 5 nm in width, 1.5 0.3 nm in height and 469 103 nm in
length. On the other hand, large brils exhibited a width of
49 6 nm, a height of 3.1 0.5 nm and a length of
4871 683 nm. Higher magnication (Fig. 4B and C) exhibited
irregular banding patterns of the brils with 3040 nm gaps between overlap regions occurring most frequently.
TEM analysis of brils from jellysh collagen (Fig. 5) revealed
also a banding pattern; however, here a period of 67 nm was observed. Thick representatives of jellysh collagen brils had a
width of 53.4 8.6 nm.

Table 1
Primers for RT-PCR.
Marker
Collagen I
Collagen IIa
Collagen X
Aggrecan
SOX9
b-Actin
RUNX2

Primer sequences
0

Tannealing (C)
0

For:5 -CCAGAAGAACTGGTACATCA-3
rev: 50 -CCGCCATACTCGAACTGGAA-30
For: 50 -GAACATCACCTACCACTGCAAG-30
rev: 50 -GCAGAGTCCTAGAGTGACTGAG-30
For: 50 -GCCCACTACCCAACACCAAGAC-30
rev: 50 -CCTGGCAACCCTGGCTCTC-30
For: 50 -ACCACCGAGCCAGAAAACCAGAC-30
rev: 50 -CCTCTGAGGGGAACAGCTCCAC-30
For: 50 -GTGCTCAAAGGCTACGACTG-30
rev: 50 -CGTTCTTCACCGACTTCCTC-30
For: 50 -GGACTTCGAGCAAGAGATGG-30
rev: 50 -AGCACTGTGTTGGCGTACAG-30
For: 50 -GGTAACGATGAAAATTATTCTGCTG-30
rev: 50 -CCGAGGTCCATCTACTGTAAC-30

Amplicon size (bp)

60

96

60

488

50

196

64

268

62

316

55

234

55

201

Please cite this article in press as: Hoyer B et al. Jellysh collagen scaffolds for cartilage tissue engineering. Acta Biomater (2013), http://dx.doi.org/10.1016/
j.actbio.2013.10.022

B. Hoyer et al. / Acta Biomaterialia xxx (2013) xxxxxx

Fig. 1. Inuence of the nal collagen concentration on bril reassembly of jellysh collagen: turbidity measurement at 4 C (A) and 25 C (B), and degree of bril formation
measured by Bradford assay (C).

Fig. 2. Inuence of the NaCl concentration on bril reassembly of jellysh collagen: turbidity measurement at 4 C (A) and 25 C (B), and degree of bril formation measured
by Bradford assay (C).

Fig. 3. Inuence of the reaction temperature on the bril reassembly of jellysh collagen exemplary for a nal collagen concentration of 2.5 mg ml1: turbidity measurement
(A) and degree of bril formation measured by Bradford assay (B).

3.3. Characterization of porous jellysh collagen scaffolds

Porous 3-D jellysh collagen scaffolds were engineered in different sizes (Fig. 6A) by means of lyophilization. SEM analysis
(Fig. 6B) revealed an open, interconnected pore structure. A porosity of 98.2 0.4% was assessed by measuring the pore volume with
a helium pycnometer.
DSC (Fig. 7) was used to determine the effect of chemical crosslinking on the denaturation temperature of jellysh collagen. An
increase of 12 K of the peak temperature was achieved by chemical cross-linking of jellysh collagen with 1% EDC.
Mechanical properties were evaluated under wet conditions,
with scaffolds incubated in SBF. In static tests (Fig. 8A), a compressive modulus of 9.98 0.93 kPa for 3-D jellysh collagen scaffolds
was measured. The compressive stress at 20% compression was
1.79 0.20 kPa. Cyclic tests (Fig. 8B) revealed a highly elastic
behavior of the jellysh collagen scaffolds.

3.4. Viability and chondrogenic differentiation of hMSC in scaffolds

from jellysh collagen
Initially, the cytotoxicity of jellysh collagen was tested. Cells
on polystyrene were cultivated with either scaffold extract medium or fresh medium (Fig. 9). There was no signicant difference
in the number of cells in both examined media, showing that jellysh collagen did not release cytotoxic substances.
Human mesenchymal stem cells were seeded on porous
jellysh collagen scaffolds and cultivated under chondrogenic
stimulation for 21 days. Cell distribution was evaluated macroscopically by MTT staining (Fig. 10). Only viable cells metabolize
the yellowish MTT into a dark blue formazan derivative. The staining showed viable cells throughout the whole scaffold. Gene
expression of b-actin, collagen I, collagen IIa, collagen X, aggrecan,
RUNX2 and SOX9 was investigated (Fig. 11A). The housekeeping
gene b-actin and collagen I were expressed at similar levels on

Please cite this article in press as: Hoyer B et al. Jellysh collagen scaffolds for cartilage tissue engineering. Acta Biomater (2013), http://dx.doi.org/10.1016/
j.actbio.2013.10.022

B. Hoyer et al. / Acta Biomaterialia xxx (2013) xxxxxx

Fig. 4. AFM images (left: height image, right: amplitude image) of brillized jellysh collagen. Insets in A and B represent the height scale. Panel C is a magnication of the
marked section in B and was used for the section analysis.

According to SEM and cLSM images after 21 days of cultivation

(Fig. 12AC), a large number of cells were found on the seeding
side of the scaffolds, which spread over the pores, similar to broblast phenotype. Yet in between, cells of a more spherical shape
(indicated by arrows) were discovered which resemble the chondrogenic phenotype. Fig. 12D shows the opposing seeding side of
the jellysh collagen scaffold. Here only spherical cells are found,
arranged in small clusters.
4. Discussion

Fig. 5. TEM image of stained jellysh collagen brils. Bar shows 200 nm.

day 1 and day 21. Genes coding for collagen II, collagen X,
aggrecan, RUNX2 and SOX9 were upregulated. Furthermore, sGAG
concentration was biochemically quantied, showing a signicant
increase (p < 0.001) from day 1 to day 21 (Fig. 11B).

Sufcient regeneration of cartilage defects is still one of the

major challenges in orthopedics. One strategy to meet this
challenge is tissue engineering. The development of suitable matrices for hosting chondrogenically stimulated cells is a prerequisite
for this purpose. In this study, we describe the manufacture of
brillized jellysh collagen to be used as a biomimetic matrix for
cartilage tissue engineering.
Collagen is a well-investigated biomaterial that has been studied since the beginning of the last century. Although bril formation of collagen has been extensively examined, it is still not
common to use brillized collagen for scaffold fabrication. Porous

Please cite this article in press as: Hoyer B et al. Jellysh collagen scaffolds for cartilage tissue engineering. Acta Biomater (2013), http://dx.doi.org/10.1016/
j.actbio.2013.10.022

B. Hoyer et al. / Acta Biomaterialia xxx (2013) xxxxxx

Fig. 6. (A) Scaffolds of different sizes and (B) SEM image of a 3-D jellysh collagen scaffold.

Fig. 7. DSC measurement of jellysh collagen without and with 1% EDC crosslinking.

collagen-based scaffolds are more frequently prepared by freezedrying solubilized collagen monomers. There are a few reports
on brillized collagen matrices in the form of lyophilized sponges
[37], hydrogels [28,38], lms [39] or as coatings [40,41]. In the eld
of bone tissue engineering, some studies focus on mineralization of
brillized collagen to produce a biomaterial which mimics the
mineralized bone matrix [35,4244]. Scaffolds made of brillized
marine collagens derived from salmon skin have been reported
only by a Japanese research group [45] and recently by us [34].
In contrast to structures of freeze-dried monomeric collagen, structures of brillized collagen mimic the natural extracellular matrix
and can provide additional mechanical strength [46].
A challenge in the handling of marine collagens is their lower
denaturation temperature compared to collagens from mammalian
sources. For collagen derived from R. esculentum we determined a
denaturation temperature of 32 C by means of circular dichroism (data not shown). Consequently, we conducted the bril formation studies at 4 and 25 C, respectively, both temperatures
being below the denaturation temperature of jellysh collagen.
The reaction temperature had a considerable inuence on the

Fig. 9. Viability of human mesenchymal stem cells cultured with scaffold extract
(scaffolds) or fresh medium (control). Number of viable cells was calculated from
LDH activity after total lysis.

extent of bril formation (Fig. 3), which was signicantly increased

(p < 0.05) at 4 C compared to 25 C. Wood and Keech [47] reported
a similar temperature effect on the bril formation of collagen
from calf skin caused by different bril widths of 500 and
1000 nm, respectively.
Changes in the collagen concentration (Fig. 1) in the range of
0.52.5 mg ml1 did not signicantly inuence the percentage of
brillized collagen, while the absorbance values of the turbidity
plateaus increased with increasing collagen concentration. Increasing turbidity values of the plateau phase were also documented for
salmon collagen type I in the range of 0.22.5 mg ml1 [34] and
bovine collagen type II in the range of 0.20.5 mg ml1 [48]. Fibril
formation of jellysh collagen was further inuenced by ionic
strength (Fig. 2). With increasing sodium chloride concentration
(0100 mM) the absorbance values of the turbidity plateaus decreased, and consequently the amount of brillized collagen was
reduced. This was a surprising result since sea water contains
455 mM sodium chloride [49], four times higher than the highest
concentration investigated in this study. Based on those ndings a

Fig. 8. Compressive stressstrain curve of 3-D jellysh collagen scaffolds: static compression (A) and cyclic compression (B) of wet samples.

Please cite this article in press as: Hoyer B et al. Jellysh collagen scaffolds for cartilage tissue engineering. Acta Biomater (2013), http://dx.doi.org/10.1016/
j.actbio.2013.10.022

B. Hoyer et al. / Acta Biomaterialia xxx (2013) xxxxxx

Fig. 10. MTT staining of viable human mesenchymal stem cells cultured with chondrogenic supplements after 21 days of cultivation. Jellysh collagen scaffolds without (left)
and with (right) cells (A), and cross-section of a cell-seeded scaffold (B). Bars show 3 mm.

Fig. 11. Analysis of unstimulated hMSCs on jellysh collagen scaffolds after 1 day of
cultivation and chondrogenically stimulated hMSCs on jellysh collagen scaffolds
after 21 days of cultivation. RT-PCR expression analysis of different genes (A) and
biochemical quantication of sulfated glycosaminoglycan concentration (B).

protocol was established for bril formation of jellysh collagen at

a temperature of 4 C: 5 mg ml1 collagen dissolved in hydrochloric acid was mixed 1:1 with 50 mM Tris buffer at pH 8, yielding a
solution with the resultant pH of 7.4.
Fibril formation was veried with AFM (Fig. 4) and TEM (Fig. 5)
images showing the typical banding of collagen brils. It is well
known that the glycosylation grade is inversely related to the bril
diameter [50]. This was already reported for jellysh collagen with
a glycosylation grade of 2325 residues/1000 amino acids, leading
to brils with a diameter of 1030 nm [11]. Collagen from R. esculentum showed a lower glycosylation grade of 9 residues/1000
amino acids, which could explain the slightly thicker brils with
diameters between 30 and 60 nm, which are in the range of native
mammalian cartilage collagen type II brils (5120 nm) [51]. The
D-periodic banding pattern of 67 nm could not be clearly identied
in AFM for jellysh collagen. This is a well-known phenomenon of
collagen brils with diameters below 30 nm [51]. For example, Tsai
et al. [52] demonstrated a banding pattern of 48.6 nm in porcine
cartilage collagen. In TEM, however, stained jellysh collagen brils showed a periodic banding of 67 nm, which is typical for
collagens of type I and II [53,54]. Nevertheless, banding of jellysh
collagen brils was less distinct, indicating that bril association
was slightly disordered. To sum up, jellysh collagen is a brilforming collagen with a distinct banding pattern.

3-D porous scaffolds (Fig. 6) made of brillized jellysh collagen

were prepared by a freeze-drying technique, resulting in 98%
porosity comparable to bovine collagen scaffolds produced by
Dawson et al. [55]. The issue of the denaturation temperature being
below 37 C was solved by cross-linking with a carbodiimid derivative veried with DSC measurements (Fig. 7).
Compressive modulus (Fig. 8) of jellysh collagen scaffolds was
increased 13-fold compared to salmon collagen scaffolds [34].
However, collagen scaffolds generally have, compared to other
materials, limited mechanical properties, owing to the high porosity of 98% and the use of a natural protein.
Cytotoxicity of jellysh collagen (Fig. 9) was indirectly assessed
by cultivation of cells with scaffold extracts. No signicant difference in the proliferation rate of hMSCs cultivated with scaffold extracts or fresh medium was observed, indicating the absence of
cytotoxic effects of the scaffold extracts. Similar non-cytotoxic effects on human broblasts and smooth muscle cells incubated with
extracts of collagen of Stomolophus nomurai meleagris were obtained by Song et al. [17].
Direct cultivation of hMSCs under chondrogenic stimulation revealed viable cells (Fig. 10) and upregulation of chondrogenic
markers on mRNA level (Fig. 11A). Other studies have demonstrated chondrogenic differentiation of MSCs in the 3-D environment of scaffolds. Collagen-based matrices were applied in
particular by Bosnakovski et al. [28], who cultivated bovine MSCs
in collagen type I and II gels, and by Chen and co-workers [56],
who studied the chondrogenic differentiation of rabbit MSCs in
porous scaffolds from bovine collagen type II. Both research groups
reported in accordance a substantial increase of collagen type II
and aggrecan mRNA after 3 weeks of cultivation. Additionally, we
found collagen type X to be upregulated, which is also synthesized
in hypertrophic chondrocytes and is therefore associated with
hypertrophic induction [57]. However, some researchers express
doubts as to whether an increased collagen type X transcript level
in chondrogenically stimulated MSC cultures is really associated
with hypertrophy. Mwale et al. [58] detected an increase of collagen type X mRNA earlier than collagen II mRNA in chondrogenically stimulated pellet cultures of hMSCs, suggesting that collagen
type X is no reliable hypertrophy marker in stem cell differentiation. Similar results were reported by Jakobsen et al. [59], who observed a simultaneous upregulation of the chondrogenic markers
SOX9 and aggrecan on the one hand and collagen type X on the

Fig. 12. SEM images (A, C, D) and cLSM image (B) of chondrogenically stimulated hMSCs cultivated on jellysh collagen scaffolds for 21 days. (AC) Cell-seeding side and (D)
opposite side of scaffold. Bars show 50 lm.

Please cite this article in press as: Hoyer B et al. Jellysh collagen scaffolds for cartilage tissue engineering. Acta Biomater (2013), http://dx.doi.org/10.1016/
j.actbio.2013.10.022

B. Hoyer et al. / Acta Biomaterialia xxx (2013) xxxxxx

other hand when cultivating chondrogenically stimulated hMSCs

in hyaluronic acid scaffolds. In our experiments a tendency to
hypertrophic development is furthermore indicated by the upregulation of RUNX2. An upregulated expression of RUNX2 in later
chondrocytes regulates hypertrophy [60]. On the contrary, RUNX2
is expressed in bone-marrow-derived MSCs even without osteogenic or chondrogenic differentiation, as conrmed by immunouorescent staining and gene expression analysis [61]. This is in
accordance with our study, where RUNX2 was already expressed
after seeding of the scaffolds. Redifferentiation of chondrocytes is
accompanied by upregulation of collagen type II and downregulation of collagen type I expression. In our study chondrogenic differentiation of hMSCs led to an increased collagen type II expression;
however, the mRNA level of collagen type I did not decrease. Other
groups reported likewise no downregulation of collagen type I
expression during chondrogenic differentiation of MSCs [59,62].
The chondrogenic induction was probably not sufcient to completely suppress the broblast-like MSC phenotype, which could
be observed in SEM and cLSM (Fig. 12AC).
Chondrogenic differentiation of hMSCs was further indicated by
a 23-fold increase of deposited sGAG in jellysh collagen scaffolds
after 21 days of cultivation (Fig. 11B). An increase of sGAG deposition was also observed in chondrocytes on collagen matrices
[16,63]. Furthermore SEM images (Fig. 12) revealed a large number
of hMSCs with a spherical phenotype, similar to the ndings of Pieper et al. [63] for chondrocytes on collagen matrices. Obviously not
all cells have differentiated towards the spherical phenotype yet,
requiring further adjustments in chondrogenic stimulation and
scaffold modication. However, the potential of brillized jellysh
collagen to support chondrogenically stimulated hMSCs was
shown beyond dispute.

5. Conclusions
This is the rst report on brillized collagen scaffolds based on
collagen from jellysh R. esculentum. This jellysh species is a prospective collagen source due to the similarity of its collagen to
mammalian collagen type II. We analyzed different parameters
inuencing the bril formation of jellysh collagen. The use of collagen brils for producing a matrix mimics the cartilage extracellular matrix, which is quite often not regarded in other publications.
Porous 3-D jellysh collagen scaffolds with an interconnected pore
structure were manufactured by freeze-drying and subsequent
chemical cross-linking. We successfully engineered a cytocompatible matrix with the potential to support and maintain chondrogenic stimulation of human mesenchymal stem cells.
Acknowledgements
The authors wish to thank the German Federal Ministry for Education and Research (BMBF) for nancial support (Regeneration
with cell specic matrices, BMBF contract No. 01GN0962) and Prof.
Martin Bornhuser, Medical Clinic I, University Hospital Carl Gustav Carus, Dresden, for providing hMSCs. Special thanks to Dr. Silke
Erdmann for her enlightening and encouraging discussions and
support. We also thank Kristin Faulwasser and Rudolf Beiermeister
for the preparation and bril formation analysis. Furthermore, we
are grateful for the excellent technical assistance of Sophie Brggemeier and Ortrud Zieschang.

Appendix A. Figures with essential colour discrimination

Certain gures in this article, particularly Figs. 4, 10, and 12, are
difcult to interpret in black and white. The full colour images can

be found in the on-line version, at http://dx.doi.org/10.1016/

j.actbio.2013.10.022.

References
[1] Glowacki J, Mizuno S. Collagen scaffolds for tissue engineering. Biopolymers
2008;89:33844.
[2] Gmez-Guilln MC, Gimnez B, Lpez-Caballero ME, Montero MP. Functional
and bioactive properties of collagen and gelatin from alternative sources: a
review. Food Hydrocolloids 2011;25(8):181327. http://dx.doi.org/10.1016/
j.foodhyd.2011.02.007.
[3] Yunoki S, Suzuki T, Takai M. Stabilization of low denaturation temperature
collagen from sh by physical cross-linking methods. J Biosci Bioeng
2003;96(6):5757.
[4] Nomura Y, Yamano M, Hayakawa C, Ishii Y, Shirai K. Structural property and
in vitro self-assembly of shark type I collagen. Biosci Biotechnol Biochem
1997;61(11):191923.
[5] Wang L, An X, Xin Z, Zhao L, Hu Q. Isolation and characterization of collagen
from the skin of deep sea redsh (Sebastes mentella). J Food Sci
2007;72(8):E4505. http://dx.doi.org/10.1111/j.1750-3841.2007.00478.x.
[6] Heinemann S, Ehrlich H, Douglas T, Heinemann C, Worch H, Schatton W, et al.
Ultrastructural studies on the collagen of the marine sponge Chondrosia
reniformis Nardo. Biomacromolecules 2007;8(11):34527. http://dx.doi.org/
10.1021/bm700574y.
[7] Addad S, Exposito JY, Faye C, Ricard-Blum S, Lethias C. Isolation,
characterization and biological evaluation of jellysh collagen for use in
biomedical applications. Mar Drugs 2011;9(6):96783.
[8] Nagai T, Ogawa T, Nakamura T, Ito T, Nakagawa H, Fujiki K, et al. Collagen of
edible jellysh exumbrella. J Sci Food Agric 1999;79:8558.
[9] Nagai T, Worawattanamateekul W, Suzuki N, Nakamura T, Ito T, Fujiki K, et al.
Isolation and characterization of collagen from rhizostomous jellysh
(Rhophilema asamushi). Food Chem 2000;70:2058. http://dx.doi.org/
10.1016/S0308-8146(00)00081-9.
[10] Kimura S, Miura S, Park YH. Collagen as the major edible component of
jellysh (Stomolophus nomural). J Food Sci 1983;48:175860.
[11] Miura S, Kimura S. Jellysh mesogloea collagen. J Biol Chem
1985;260(28):153526.
[12] Calejo MT, Morais ZB, Fernandes AI. Isolation and biochemical characterisation
of a novel collagen from Catostylus tagi. J Biomater Sci 2009;20:207387.
http://dx.doi.org/10.1163/156856208X399125.
[13] Zhuang Y, Hou H, Zhao X, Zhang Z, Li B. Effects of collagen and collagen
hydrolysate from jellysh (Rhopilema esculentum) on mice skin photoaging
induced by UV irradiation. J Food Sci 2009;74(6):H1838. http://dx.doi.org/
10.1111/j.1750-3841.2009.01236.x.
[14] Hsieh YHP. Use of jellysh collagen (type II) in the treatment of rheumatoid
arthritis. U.S. Patent 6,894,029 B1; 17 May 2005.
[15] Bermueller C, Schwarz S, Elsaesser AF, Sewing J, Baur N, von Bomhard A, et al.
Marine collagen scaffolds for nasal cartilage repair: prevention of nasal septal
perforations in a new orthotopic rat model using tissue engineering
techniques. Tissue eng 2013;19(19/20):114. http://dx.doi.org/10.1089/
ten.tea.2012.0650.
[16] Nehrer S, Breinan HA, Ramappa A, Shortkroff S, Young G, Minas T, et al. Canine
chondrocytes seeded in type I and type II collagen implants investigated
in vitro. J Biomed Mater Res 1997;38(2):95104. http://dx.doi.org/10.1002/
(SICI)1097-4636(199722)38:2<95::AID-JBM3>3.0.CO;2-B.
[17] Song E, Yeon Kim S, Chun T, Byun HJ, Lee YM. Collagen scaffolds derived from a
marine source and their biocompatibility. Biomaterials 2006;27(15):295161.
http://dx.doi.org/10.1016/j.biomaterials.2006.01.015.
[18] Lee SJ, Kim SY, Lee YM. Preparation of porous collagen/hyaluronic acid hybrid
scaffolds for biomimetic functionalization through biochemical binding
afnity. J Biomed Mater Res B 2007;82(2):50618. http://dx.doi.org/10.1002/
jbm.b.30756.
[19] Jeong SI, Kim SY, Cho SK, Chong MS, Kim KS, Kim H, et al. Tissue-engineered
vascular grafts composed of marine collagen and PLGA bers using pulsatile
perfusion bioreactors. Biomaterials 2007;28(6):111522. http://dx.doi.org/
10.1016/j.biomaterials.2006.10.025.
[20] Park JS, Choi JB, Jo SY, Lim YM, Gwon HJ, Khil MS, et al. Characterization and
structure analysis of PLGA/collagen nanobrous membranes by
electrospinning. J Appl Polym Sci 2012;125:E595603. http://dx.doi.org/
10.1002/app.36833.
[21] Wachsmuth L, Sder S, Fan Z, Finger F, Aigner T. Immunolocalization of matrix
proteins in different human cartilage subtypes. Histol Histopathol
2006;21:47785.
[22] Eyre D. Collagen of articular cartilage. Arthritis Res 2002;4:305.
[23] Temenoff JS, Mikos AG. Review: tissue engineering for regeneration of
articular cartilage. Biomaterials 2000;21:43140.
[24] Benjamin M, Evans EJ. Fibrocartilage. J Anat 1990;171:115.
[25] Keeney M, Lai JH, Yang F. Recent progress in cartilage tissue engineering. Curr
Opin
Biotechnol
2011;22(5):73440.
http://dx.doi.org/10.1016/
j.copbio.2011.04.003.
[26] Williams CG, Kim TK, Taboas A, Malik A, Manson P, Elisseeff J. In vitro
chondrogenesis of bone marrow-derived mesenchymal stem cells in a
photopolymerizing hydrogel. Tissue Eng 2003;9(4):67988. http://dx.doi.org/
10.1089/107632703768247377.

Please cite this article in press as: Hoyer B et al. Jellysh collagen scaffolds for cartilage tissue engineering. Acta Biomater (2013), http://dx.doi.org/10.1016/
j.actbio.2013.10.022

10

B. Hoyer et al. / Acta Biomaterialia xxx (2013) xxxxxx

[27] Li WJ, Tuli R, Okafor C, Derfoul A, Danielson KG, Hall DJ, et al. A threedimensional nanobrous scaffold for cartilage tissue engineering using human
mesenchymal stem cells. Biomaterials 2005;26(6):599609. http://dx.doi.org/
10.1016/j.biomaterials.2004.03.005.
[28] Bosnakovski D, Mizuno M, Kim G, Takagi S, Okumura M, Fujinaga T.
Chondrogenic differentiation of bovine bone marrow mesenchymal stem
cells (MSCs) in different hydrogels: inuence of collagen type II extracellular
matrix on MSC chondrogenesis. Biotechnol Bioeng 2006;93(6):115263.
http://dx.doi.org/10.1002/bit.20828.
[29] Steck E, Bertram H, Walther A, Brohm K, Mrozik B, Rathmann M, et al.
Enhanced biochemical and biomechanical properties of scaffolds generated by
ock technology for cartilage tissue engineering. Tissue Eng A
2010;16(12):3697707. http://dx.doi.org/10.1089/ten.tea.2009.0817.
[30] Wang Y, Kim UJ, Blasioli DJ, Kim HJ, Kaplan DL. In vitro cartilage tissue
engineering with 3D porous aqueous-derived silk scaffolds and mesenchymal
stem cells. Biomaterials 2005;26(34):708294. http://dx.doi.org/10.1016/
j.biomaterials.2005.05.022.
[31] Awad HA, Wickham MQ, Leddy HA, Gimble JM, Guilak F. Chondrogenic
differentiation of adipose-derived adult stem cells in agarose, alginate, and
gelatin
scaffolds.
Biomaterials
2004;25(16):321122.
10.1016/
j.biomaterials.2003.10.045.
[32] Ragetly GR, Griffon DJ, Lee HB, Fredericks LP, Gordon-Evans W, Chung YS.
Effect of chitosan scaffold microstructure on mesenchymal stem cell
chondrogenesis. Acta Biomater 2010;6(4):14306. http://dx.doi.org/10.1016/
j.actbio.2009.10.040.
[33] Sttzel S, Schurink M, Wienk H, Siebler U, Burg-Roderfeld M, Eckert T, et al.
Molecular organization of various collagen fragments as revealed by atomic
force microscopy and diffusion-ordered NMR spectroscopy. Chemphyschem
2012;13(13):311725. http://dx.doi.org/10.1002/cphc.201200284.
[34] Hoyer B, Bernhardt A, Heinemann S, Stachel I, Meyer M, Gelinsky M.
Biomimetically mineralized salmon collagen scaffolds for bone tissue
engineering. Biomacromolecules 2012;13(4):105966. http://dx.doi.org/
10.1021/bm201776r.
[35] Bradt JH, Mertig M, Teresiak A, Pompe W. Biomimetic mineralization of
collagen by combined bril assembly and calcium phosphate formation. Chem
Mater 1999;11(10):2694701.
[36] Oyane A, Kim HM, Furuya T, Kokubo T, Miyazaki T, Nakamura T. Preparation
and assessment of revised simulated body uids. J Biomed Mater Res A
2003;65(2):18895.
[37] Allemann F, Mizuno S, Eid K, Yates KE, Zaleske D, Glowacki J. Effects of
hyaluronan on engineered articular cartilage extracellular matrix gene
expression in 3-dimensional collagen scaffolds. J Biomed Mater Res
2001;55(1):139.
[38] Ben-Yishay A, Grande DA, Schwartz RE, Menche D, Pitman MD. Repair of
articular cartilage defects with collagenchondrocyte allografts. Tissue Eng
1995;1(2):11933.
[39] Shanmugasundaram N, Ravikumar T, Babu M. Comparative physico-chemical
and in vitro properties of brillated collagen scaffolds from different sources. J
Biomater
Appl
2004;18(4):24764.
http://dx.doi.org/10.1177/
0885328204040945.
[40] Heinemann C, Heinemann S, Bernhardt A, Worch H, Hanke T. Novel textile
chitosan scaffolds promote spreading, proliferation, and differentiation of
osteoblasts. Biomacromolecules 2008;9(10):291320. http://dx.doi.org/
10.1021/bm800693d.
[41] Kim HW, Li LH, Lee EJ, Lee SH, Kim HE. Fibrillar assembly and stability of
collagen coating on titanium for improved osteoblast responses. J Biomed
Mater Res A 2005;75(3):62938.
[42] Gelinsky M, Welzel P, Simon P, Bernhardt A, Knig U. Porous three dimensional
scaffolds made of mineralised collagen: preparation and properties of a
biomimetic nanocomposite material for tissue engineering of bone. Chem Eng
J 2008;137:8496.
[43] Heinemann S, Heinemann C, Jger M, Neunzehn J, Wiesmann HP, Hanke T.
Effect of silica and hydroxyapatite mineralization on the mechanical
properties and the biocompatibility of nanocomposite collagen scaffolds.
Appl Mater Interfaces 2011;3(11):432331. http://dx.doi.org/10.1021/
am200993q.
[44] Roveri N, Falini G, Sidoti MC, Tampieri A, Landi E, Sandri M, et al. Biologically
inspired growth of hydroxyapatite nanocrystals inside self-assembled collagen

[45]

[46]

[47]
[48]

[49]
[50]

[51]

[52]

[53]

[54]

[55]

[56]

[57]
[58]

[59]

[60]

[61]

[62]

[63]

bers. Mater Sci Eng C 2003;23:4416. http://dx.doi.org/10.1016/S09284931(02)00318-1.

Yunoki S, Nagai N, Suzuki T, Munekata M. Novel biomaterial from reinforced
salmon collagen gel prepared by bril formation and cross-linking. J Biosci
Bioeng 2004;98(1):407. http://dx.doi.org/10.1016/S1389-1723(04)70240-6.
Yunoki S, Marukawa E, Ikoma T, Sotome S, Fan H, Zhang X, et al. Effect of
collagen bril formation on bioresorbability of hydroxyapatite/collagen
composites. J Mater Sci Mater Med 2007;18(11):217983. http://dx.doi.org/
10.1007/s10856-007-3011-z.
Wood GC, Keech MK. The formation of brils from collagen solutions. Biochem
J 1960;78:58898.
Kuo SM, Wang YJ, Niu GC, Lu HE, Chang SJ. Inuences of hyaluronan on type II
collagen brillogenesis in vitro. J Mater Sci Mater Med 2008;19(3):123541.
http://dx.doi.org/10.1007/s10856-007-3205-4.
Hodgkin AL, Katz B. The effect of sodium ions on the electrical activity of the
giant axon of the squid. J Physiol 1949;108:3777.
Notbohm H, Nokelainen M, Myllyharju J, Fietzek PP, Mller PK, Kivirikko KI.
Recombinant human type II collagens with low and high levels of
hydroxylysine and its glycosylated forms show marked differences in
brillogenesis in vitro. J Biol Chem 1999;274(13):898892.
Ortolani F, Marchini M. Cartilage type II collagen brils show distinctive
negative-staining band patterns differences between type II and type I unxed
or glutaraldehyde-xed collagen brils. J Electron Microsc 1995;44(5):36575.
Tsai CL, Hsu Sh SH, Cheng WL. Effect of different solvents and crosslinkers on
cytocompatibility of type II collagen scaffolds for chondrocyte seeding. Artif
Organs 2002;26(1):1826.
Engel J, Bchinger HP. Structure, stability and folding of the collagen triple
helix. In: Brinckmann J, Notbohm H, Mller PK, editors. Collagen: primer in
structure, processing and assembly. Berlin: Springer Verlag; 1993. p. 733.
http://dx.doi.org/10.1007/b103818.
Kadler KE, Hill A, Canty-Laird EG. Collagen brillogenesis: bronectin,
integrins, and minor collagens as organizers and nucleators. Curr Opin Cell
Biol 2008;20(5):495501. http://dx.doi.org/10.1016/j.ceb.2008.06.008.
Dawson JI, Wahl DA, Lanham SA, Kanczler JM, Czernuszka JT, Oreffo RO.
Development of specic collagen scaffolds to support the osteogenic and
chondrogenic differentiation of human bone marrow stromal cells.
Biomaterials
2008;29(21):310516.
http://dx.doi.org/10.1016/
j.biomaterials.2008.03.040.
Chen WC, Yao CL, Wei YH, Chu IM. Evaluating osteochondral defect repair
potential of autologous rabbit bone marrow cells on type II collagen scaffold.
Cytotechnology 2011;63(1):1323. http://dx.doi.org/10.1007/s10616-0109314-9.
Pelttari K, Steck E, Richter W. The use of mesenchymal stem cells for
chondrogenesis. Injury 2008;39S1:S5865.
Mwale F, Stachura D, Roughley P, Antoniou J. Limitations of using aggrecan and
type X collagen as markers of chondrogenesis in mesenchymal stem cell
differentiation. J Orthop Res 2006;24(8):17918. http://dx.doi.org/10.1002/
jor.20200.
Jakobsen RB, Shahdadfar A, Reinholt FP, Brinchmann JE. Chondrogenesis in a
hyaluronic acid scaffold: comparison between chondrocytes and MSC from
bone marrow and adipose tissue. Knee Surg Sports Traumatol Arthrosc
2010;18(10):140716. http://dx.doi.org/10.1007/s00167-009-1017-4.
Gross S, Krause Y, Wuelling M, Vortkamp A. Hoxa11 and Hoxd11 regulate
chondrocyte differentiation upstream of Runx2 and Shox2 in mice. PLoS ONE
2012;7(8):e43553. http://dx.doi.org/10.1371/journal.pone.0043553.
Ulrich C, Rolauffs B, Abele H, Bonin M, Nieselt K, Hart ML, et al. Low osteogenic
differentiation potential of placenta-derived mesenchymal stromal cells
correlates with low expression of the transcription factors Runx2 and
Twist2. Stem Cells Dev 2013;22(21):114. http://dx.doi.org/10.1089/
scd.2012.0693.
Perrier E, Ronzire MC, Bareille R, Pinzano A, Mallein-Gerin F, Freyria AM.
Analysis of collagen expression during chondrogenic induction of human bone
mesenchymal stem cells. Biotechnol Lett 2011;33(10):2091101. http://
dx.doi.org/10.1007/s10529-011-0653-1.
Pieper JS, van der Kraan PM, Hafmans T, Kamp J, Buma P, van Susante JL, et al.
Crosslinked type II collagen matrices: preparation, characterization, and
potential for cartilage engineering. Biomaterials 2002;23(15):318392.
http://dx.doi.org/10.1016/S0142-9612(02)00067-4.

Please cite this article in press as: Hoyer B et al. Jellysh collagen scaffolds for cartilage tissue engineering. Acta Biomater (2013), http://dx.doi.org/10.1016/
j.actbio.2013.10.022