Self-Assembled Peptide-Based Hydrogels as Scaffolds for Proliferation and Multi-Differentiation of Mesenchymal Stem Cells

Article abstract of DOI:10.1002/mabi.201600192

Fluorenyl-9-methoxycarbonyl (Fmoc)-diphenylalanine (Fmoc-FF) and Fmoc-arginine-glycine--aspartate (Fmoc-RGD) peptides self-assemble to form a 3D network of supramolecular hydrogel (Fmoc-FF/Fmoc-RGD), which provides a nanofibrous network that uniquely presents bioactive ligands at the fiber surface for cell attachment. In the present study, mesenchymal stem cells (MSCs) in Fmoc-FF/Fmoc-RGD hydrogel increase in proliferation and survival compared to those in Fmoc-FF/Fmoc-RGE hydrogel. Moreover, MSCs encapsulated in Fmoc-FF/Fmoc-RGD hydrogel and induced in each defined induction medium undergo in vitro osteogenic, adipogenic, and chondrogenic differentiation. For in vivo differentiation, MSCs encapsulated in hydrogel are induced in each defined medium for one week, followed by injection into gelatin sponges and transplantation into immunodeficient mice for four weeks. MSCs in Fmoc-FF/Fmoc-RGD hydrogel increase in differentiation into osteogenic, adipogenic, and chondrogenic differentiation, compared to those in Fmoc-FF/Fmoc-RGE hydrogel. This study concludes that nanofibers formed by the self-assembly of Fmoc-FF and Fmoc-RGD are suitable for the attachment, proliferation, and multi-differentiation of MSCs, and can be applied in musculoskeletal tissue engineering. (Figure presented.).

Full text of DOI:10.1002/mabi.201600192

Macromolecular
Bioscience
Full Paper
Self-Assembled Peptide-Based Hydrogels as
Scaffolds for Proliferation and Multi-
Differentiation of Mesenchymal Stem Cells
Yung-Li Wang, Shih-Pei Lin, Srinivasa Rao Nelli, Fu-Kai Zhan, Hsun Cheng,
Tsung-Sheng Lai, Mei-Yu Yeh, Hsin-Chieh Lin,* Shih-Chieh Hung*
Fluorenyl-9-methoxycarbonyl (Fmoc)-diphenylalanine (Fmoc-FF) and Fmoc-arginine-glycine-
aspartate (Fmoc-RGD) peptides self-assemble to form a 3D network of supramolecular
hydrogel (Fmoc-FF/Fmoc-RGD), which provides a nanofibrous network that uniquely presents
bioactive ligands at the fiber surface for cell attachment. In the present study, mesenchymal
stem cells (MSCs) in Fmoc-FF/Fmoc-RGD hydrogel increase in proliferation and survival com-
pared to those in Fmoc-FF/Fmoc-RGE hydrogel. Moreover, MSCs encapsulated in Fmoc-FF/
Fmoc-RGD hydrogel and induced in each defined induction medium undergo in vitro osteo-
genic, adipogenic, and chondrogenic differentiation. For in vivo differentiation, MSCs encapsu-
lated in hydrogel are induced in each defined medium for one week, followed by injection into
gelatin sponges and transplantation into immunodeficient
mice for four weeks. MSCs in Fmoc-FF/Fmoc-RGD hydrogel
increase in differentiation into osteogenic, adipogenic, and
chondrogenic differentiation, compared to those in Fmoc-
FF/Fmoc-RGE hydrogel. This study concludes that nanofibers
formed by the self-assembly of Fmoc-FF and Fmoc-RGD are
suitable for the attachment, proliferation, and multi-differen-
tiation of MSCs, and can be applied in musculoskeletal tissue
engineering.
Y.-L. Wang, Dr. S.-P. Lin, Dr. S.-C. Hung
Institute of Clinical Medicine
National Yang-Ming University
Taipei 112, Taiwan
Dr. M.-Y. Yeh, Dr. S.-C. Hung
Graduate Institute of Biomedical Sciences
China Medical University
Taichung 40402, Taiwan
S. R. Nelli, F.-K. Zhan, H. Cheng,
T.-S. Lai, Prof. H.-C. Lin
Dr. S.-C. Hung
Institute of New Drug Development
China Medical University
Department of Materials Science
and Engineering
National Chiao Tung University
Hsinchu 300, Taiwan
Taichung 40402, Taiwan
E-mail: hung3340@gmail.com
Dr. S.-C. Hung
E-mail: hclin45@nctu.edu.tw
Dr. M.-Y. Yeh, Dr. S.-C. Hung
Integrative Stem Cell Center
Department of Orthopedics
China Medical University Hospital
Taichung 40447, Taiwan
Department of Medical Research and Department of
Orthopaedics and Traumatology
Taipei Veterans General Hospital, Taipei 112, Taiwan
Dr. S.-C. Hung
Institute of Biomedical Sciences
Academia Sinica, Taipei 11529, Taiwan
Macromol. Biosci. 2016, DOI: 10.1002/mabi.201600192
© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1
DOI: 10.1002/mabi.201600192
wileyonlinelibrary.com
Early View Publication; these are NOT the final page numbers, use DOI for citation !!

Macromolecular
Bioscience
Y.-L. Wang et al.
www.mbs-journal.de
portion. However, RGD peptide is very hydrophilic so
Fmoc-RGD is not a very efficient hydrogelator. In contrast,
Fmoc-FF has the ability to self-assemble into a nanofi-
brous hydrogel,[25] which offers a stable structural compo-
nent to mix with bio-active Fmoc-RGD for the formation
of scaffolds for anchorage-dependent cells.[26] Moreover,
these cells deposit ECM components of both fibronectin
and collagen I within the Fmoc-FF/Fmoc-RGD hydrogel,
which form mesh-like structures and increase the stiff-
ness of the gel, leading to the strengthening of the cell-gel
constructs.[27] Most of the human MSCs used in previous
tissue engineering studies have been derived from the
primary cell cultures of patients,[28] which may have dis-
ease- or age-related differences in terms of proliferation
and differentiation potential.[29,30] To provide more con-
sistent results for the parametric studies designed here,
we therefore chose to use a well-studied human bone
marrow-derived MSC cell line, 3A6.[31] We then investi-
gated the effects of Fmoc-FF/Fmoc-RGD hydrogel on the
attachment ability, proliferation, viability, and multi-dif-
ferentiation, both in vitro and in vivo, of this immortal-
ized cell line.[26]
1. Introduction
One approach to mimicking natural extracellular matrix
(ECM), which provides a 3D network of interlaced pro-
tein-nanofibers to communicate with resident cells, is
the development of hydrogels through molecular self-
assembly of peptides and their derivatives.[1–3] Hydrogels
fabricated through molecular self-assembly comprise
nanofibrous networks, which imitate the structural archi-
tecture of ECM.[4–6] In order to incorporate biologically-
active components, the termini of certain oligopeptide
derivatives may be modified with bioactive ligands and
upon assembly these groups are usually presented as flex-
ible pendant groups.[7–12]
Human mesenchymal stem cells (MSCs), isolated from
human tissues, such as bone marrow, have become a
promising cell source in combination with biomaterials
for clinical application. MSCs are characterized by the
capacity for self-renewal and multi-differentiation into
mesenchymal and non-mesenchymal tissues.[13] MSCs are
attractive for the treatment of musculoskeletal disorders,
because they are easy to obtain and expand in culture,
and only induce a mild immune response after transplan-
tation in vivo.[14] For application in musculoskeletal tissue
engineering, MSCs have been delivered in natural hydro-
gels or synthetic polymers and induced for differentiation
into bone, fat, and cartilage.[15,16]
2. Experimental Section
2.1. Synthesis of FmocFF, Fmoc-RGD, and Fmoc-RGE
Enhancement of the biofunctionality of materials is an
important step for biomaterial development.[17] The argi-
nine-glycine-aspartate (RGD) sequences of many adhesive
extracellular matrices[18,19] are recognized by members of
the integrin superfamily on the surface of cells and allow
cells to adhere.[20] The incorporation of the RGD bioactive
motif is one of the most adapted strategies to enhance
the functionality of synthetic biomaterials. Specifically,
RGD has been incorporated into 3D hydrogels such as
alginate,[21,22] poly(ethylene glycol),[23] poly(organophosp
hazene),[24] and PhaP-coated poly(3-hydroxybutyrate-co-
3-hydroxyhexanoate)[22] to interact with MSCs for tissue
engineering in bone healing, cartilage regeneration, and
the repair of myocardial infarction.
Compared with conventional polymeric hydrogels,
supramolecular hydrogelators possess several inherent
advantages:[11] (i) they have a well-defined configura-
tion, structure, and conformation; (ii) they are more bio-
compatible and biodegradable than polymeric hydrogels
because of the non-covalent interactions between mol-
ecules; (iii) they mimic the ECM more closely because the
self-assembled nanostructures are usually in the range
of 5–300 nm. In this work, we have synthesized Fmoc-FF,
Fmoc-RGD, and Fmoc-RGE (an analogue of RGD that does
not promote cell adhesion). Fmoc-peptide is an amphi-
phile where the Fmoc group forms the relatively hydro-
phobic part and a short peptide serves as the hydrophilic
Fmoc-FF, Fmoc-RGD, and Fmoc-RGE were synthesized by typical
solid-phase peptide synthesis and structures were confirmed by
nuclear magnetic resonance (NMR) and mass spectra.
2.1.1. Synthesis of Fmoc-FF
2-Chlorotrityl chloride resin (1.20 g) was swelled in anhydrous
CH2Cl2 for 30 min and then Fmoc-L-phenylalanine (0.775 g,
2.000 × 10–3 mol) was loaded onto the resin in anhydrous N,N-
dimethylformamide (DMF) and N,N-diisopropylethylamine
(DIEA; 0.830 mL, 5.000 × 10–3 mol) over a period of 1 h. For depro-
tection of the Fmoc group, piperidine (20% in DMF) was added
and the sample was left for 20 min; this procedure was repeated
twice (each time for 2 min). Fmoc-L-phenylalanine (0.775 g,
2 × 10–3 mol) was coupled to the free amino group using O-(ben-
zotriazol-1-yl)-N,N,N′,N′-tetramethyluraniumhexafluorophosp
hate (HBTU; 0.76 g, 2 × 10–3 mol) and DIEA (0.83 mL, 5 × 10–3 mol)
as coupling agents over a period of 1 h. The peptide derivative
was cleaved using 90% trifluoroacetic acid in deionized water
for 3 h. The resulting solution was air-dried, and then diethyl
ether was added to precipitate the target product. The solid was
dried under a vacuum to remove residual solvent (white solid:
0.244 g).
1H NMR (300 MHz, [D6]DMSO): δ = 2.70–3.20 (m, 4H, CH2),
4.00–4.20 (br, 3H; CH, CH2), 4.30–4.40 (m, 1H; CH), 4.50–4.60 (m,
1H; CH), 7.20–7.40 (m, 12H; CH), 7.44 (t, J = 7.2 Hz, 2H; CH), 7.66
(t, J = 7.5 Hz, 3H; CH, NH), 7.91 (d, J = 7.5 Hz, 2H; CH), 8.30 (d, J =
6.0 Hz, 1H; NH). MS [ESI–]: m/z Calcd. 534.22, Found 533.21.
Macromol. Biosci. 2016, DOI: 10.1002/mabi.201600192
© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
2
www.MaterialsViews.com
Early View Publication; these are NOT the final page numbers, use DOI for citation !!

Macromolecular
Bioscience
Self-Assembled Peptide-Based Hydrogels as Scaffolds for Proliferation and Multi-Differentiation of Mesenchymal Stem Cells
www.mbs-journal.de
J = 7.35 Hz, 2H; CH2), 3.11 (d, J = 6.00 Hz, 2H; CH2), 3.77 (d, J = 4.80
Hz, 2H; CH2), 4.02 (t, J = 3.00 Hz, 1H; CH); 4.24 (t, J = 7.50 Hz, 1H;
CH), 4.24 (t, J = 7.50 Hz, 1H; CH), 4.32 (d, J = 6.60 Hz, 2H; CH2), 7.36 (t,
J = 7.50 Hz, 1H; CH), 7.45 (t, J = 7.50 Hz, 1H; CH), 7.65 (d, J =
7.20 Hz, 1H; NH), 7.75 (t, J = 7.50 Hz, 2H; CH), 7.93 (d, J = 12.60 Hz,
2H; NH), 8.06 (d, J = 7.80 Hz, 1H; NH), 8.25 (d, J = 5.40 Hz, 1H; NH).
MS [FAB+]: m/z Calcd. 582.24, Found 583.70.
2.1.2. Synthesis of Fmoc-RGD
2-Chlorotrityl chloride resin (1.2 g) was swelled in anhy-
drous CH2Cl2 for 30 min and then Fmoc-l-aspartic acid
(0.823 g, 2.000 × 10–3 mol) was loaded onto the resin in
anhydrous DMF and DIEA (0.830 mL, 5.000 × 10–3 mol) for
1 h. For deprotection of the Fmoc group, piperidine (20% in
DMF) was added and the sample was left for 20 min; this
procedure was repeated twice (each time for 2 min). Fmoc-
l-alanine (0.595 g, 2.000 × 10–3 mol) was coupled to the
free amino group using HBTU (0.758 g, 2.000 × 10–3 mol)
and DIEA (0.830 mL, 5.000 × 10–3 mol) as coupling agents
for 30 min. Again, the sample was treated with piperidine
(20% in DMF) for 20 min; this procedure was repeated
twice (each time for 2 min). Finally, Fmoc-l-arginine
(1.298 g, 2.000 × 10–3 mol) was coupled to the free amino
group using HBTU (0.758g, 2.000 × 10–3 mol) and DIEA
(0.83 mL, 5 × 10–3 mol) as coupling agents. After the reac-
tion mixture had been stirred overnight, the peptide deriv-
ative was cleaved through treatment with CF3CO2H (90% in
deionized water) for 3 h. The resulting solution was dried
in air and then Et2O was added to precipitate the target
product. The solid was dried under a vacuum to remove
residual solvent (white solid: 0.402g).
2.2. Hydrogel Preparation and Stiffness Test
To prepare a mixed solution of Fmoc-FF/Fmoc-RGD or Fmoc-FF/
Fmoc-RGE, 170 µL of double distilled water (ddH2O) and 16 µL
of 0.5 m sodium hydroxide (NaOH) solution were pipetted into
a 2 mL Eppendorf tube that contained 1 mg of each of Fmoc-FF
and Fmoc-RGD or Fmoc-FF and Fmoc-RGE, followed by ultra-
sonication (Bioruptor® Plus sonication device, Denville, USA)
until a clear solution was obtained. This solution was subse-
quently neutralized to pH 7.0 by addition of 2 µL of 0.1 m hydro-
chloric acid (HCl) solution and 4 µL of 0.5 m HCl solution. The
final concentration of each of Fmoc-FF and Fmoc-RGD or Fmoc-
FF and Fmoc-RGE was 5.2 mg mL–1. Formation of the Fmoc-FF/
Fmoc-RGD and Fmoc-FF/Fmoc-RGE hydrogels depends on the
pH sensitivity of the Fmoc-peptide self-assembly process.[26]
The mixture of Fmoc-FF/Fmoc-RGD (or Fmoc-FF/Fmoc-RGE) has
low solubility at neutral pH but dissolves in alkaline solution;
upon approaching pH 7.0, the mixture self-assembles into a
macroscopic hydrogel (Figure S1A, Supporting Information). The
Transmission electron microscopy (TEM) images revealed that
the nanofibers were ’flat ribbons’ comprising parallel-aligned
fine fibrils across their width. Stiffness tests were conducted
using a TA rheometer (DHR-1) and a 40 mm parallel plate.
200 µL of hydrogel sample (1 wt%) was placed on the parallel
plate. In the angular frequency sweep test, the test range was
0.1 to 100 rad s–1 frequency, with strain of 0.8% and 15 points
per decade. Sweep mode is “log” and measurements were com-
pleted at 25 °C and 37 °C.
1H NMR (300 MHz, [D6]DMSO): δ = 1.56–1.72 (br, 4H; CH2),
2.66–2.71 (br, 2H; CH2), 3.42 (d, J = 7.20 Hz, 2H; CH2), 3.79 (s, 2H;
CH2), 4.04–4.06 (br, 1H; CH), 4.24–4.26 (br, 1H; CH), 4.27–4.32 (br,
2H; CH2), 4.58–4.61 (br, 1H; CH), 7.37 (t, J = 7.20 Hz, 2H; CH2), 7.46
(t, J = 7.20 Hz, 2H; CH2), 7.64 (d, J = 7.50 Hz, 2H; NH), 7.77 (t, J =
7.20 Hz, 1H; NH), 7.94 (d, J = 7.20 Hz, 2H; CH2), 8.25 (d, J = 7.50 Hz,
2H; CH2). MS [FAB+]: m/z Calcd. 571.21, Found 570.1.
2.1.3. Synthesis of Fmoc-RGE
2-Chlorotrityl chloride resin (1.20 g) was swelled in anhydrous
CH2Cl2 for 30 min and then Fmoc-Glu(Otbu)-OH (0.850 g, 2.00 ×
10–3 mol) was loaded onto the resin in anhydrous DMF and DIEA
(0.830 mL, 5.00 × 10–3 mol) over a period of 1 h. For deprotection
of the Fmoc group, piperidine (20% in DMF) was added and the
sample was left for 20 min; this procedure was repeated twice
(each time for 2 min). Fmoc-Gly-OH (0.594 g, 2.000 × 10–3 mol)
was coupled to the free amino group, using HBTU (0.759 g,
2.00 × 10–3 mol) and DIEA (0.830 mL, 5.00 × 10–3 mol) as coupling
agents over a period of 1 h. Again, the sample was treated
with piperidine (20% in DMF) for 20 min; this procedure was
repeated twice (each time for 2 min). Finally, Fmoc-Arg(Pbf)-
OH (3.876 g, 3.000 × 10–3 mol) was coupled to the free amino
group using HBTU (1.14 g, 3.00 × 10–3 mol) and DIEA (1.25 mL,
7.50 × 10–3 mol). After the reaction mixture had been stirred
overnight, the peptide derivative was cleaved through treatment
with CF3CO2H for 3 h. The resulting solution was concentrated
through exposure to the air and precipitated the target product.
The solid was dried under a vacuum to remove any residual sol-
vent, providing a white solid (0.230 g).
2.3. Cell Culture
The 3A6 cell line was developed by immortalization of human
primary MSCs with HPV-16 E6E7 and human telomerase reverse
transcriptase (hTERT) genes. 3A6 has been well characterized to
have increased stem-like properties and decreased spontaneous
differentiation possibility.[31] Cells were seeded at a density of
4000 cells cm–2 and grown in DMEM-LG medium (Invitrogen,
Carlsbad, CA) supplemented with 10% fetal bovine serum (FBS;
Hyclone, Logan, UT) and streptomycin/penicillin, with medium
changes every 2 d. Cells were split at the ratio of 1:5 when they
reached sub-confluence, generally every 3–5 d. Different ali-
quots of 3A6 cells (2.5 × 105, 8 × 105, or 2.5 × 106) suspended in
100 µL of serum free DMEM-LG medium were gently mixed with
100 µL of immediately prepared Fmoc-FF/Fmoc-RGD or Fmoc-
FF/Fmoc-RGE solution, followed by an immediate yet gentle
shaking to ensure the homogeneity of these cell mixtures.
Then the cell-gel mixture was transfered to the upper wells of
12 well transwell plates (Falcon, BD Biosciences, San Jose, CA).
After incubation for 30 min at 37 °C, 200 µL and 2 mL of growth
medium (GM) or induction medium were loaded to the upper
1H NMR (300 MHz, [D6]DMSO): δ = 1.55 (q, J = 6.66 Hz, 2H;
CH2), 1.65–1.90 (m, 2H; CH2), 2.00 (q, J = 7.07 Hz, 2H; CH2), 2.29 (t,
Macromol. Biosci. 2016, DOI: 10.1002/mabi.201600192
© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
3
www.MaterialsViews.com
Early View Publication; these are NOT the final page numbers, use DOI for citation !!

Macromolecular
Bioscience
Y.-L. Wang et al.
www.mbs-journal.de
and lower well, respectively. The medium was changed 2 h post
culture and then every 2 d.
supplemented with ITS+ Premix (41400-045, Gibco), 10 ng mL–1
TGF-β1 (240-B, R&D, Minneapolis, MN), 10−7 m dexamethasone,
40 µg m–1 L-Proline (P5607, Sigma), 50 µg mL–1 ascorbate-2 phos-
phate] to differentiate into osteoblasts, adipocytes, and chondro-
cytes, respectively.
2.4. Lentivirus Mediated Overexpression
Lentivirus carried gene encoding red fluorescent protein
(pLAS2w.RFP-C.Ppuro) was purchased from the National RNAi
Core Facility supported by the National Science Council in
Taiwan. For lentiviral infection, 3A6 was seeded at 4 × 104 in
each well of 6 well tissue culture plates 1 d prior to infection. The
GM was replaced with fresh GM containing 8 µg mL–1 polybrene
(Sigma). The virus was added to cells at a multiplicity of infec-
tion (MOI) of 3. For improving lentiviral efficiency, the cells were
centrifuged at 1100×g in plates for 60 min at 32 °C, followed by
incubation at 37 °C in 5% CO2. To prevent the toxicity of poly-
brene, the medium was replaced with polybrene-free medium
6 h later.[32,33] The cells were further incubated overnight at 37 °C
in 5% CO2. 2 d after infection, the cells were replaced with GM
containing 3 µg mL–1 puromycin (Sigma, St. Louis, MO) for 2 d,
followed by culture with 0.3 µg mL–1 puromycin.
2.8. In Vivo Differentiation
The cell-gel mixtures were placed in the upper wells of 12 well
transwell plates and induced for differentiation in each defined
medium. After one week of induction for differentiation, the
media in the upper wells were carefully removed and the cell-
gel mixtures were aspirated with 18G syringes and each cell-
gel mixture (200 µL) was injected into a square piece of gelatin
sponge (8 × 8 × 10 mm, Spongostan®, Ferrosan, Soeborg, Den-
mark), followed by transplantation of the cell-containing gelatin
sponges under the skin of immunodeficient mice. The specimens
were harvested for evaluation four weeks later. Animal care and
experiments were approved by the Institutional Animal Care
and Utilization Committee of Academia Sinica (IACUC).
2.9. Histochemistry and Immunohistochemistry
2.5. Cell Proliferative Assay
Samples were recovered and fixed in 4% paraformaldehyde.
For osteogenic and chondrogenic differentiation, samples were
processed for paraffin sections. For adipogenic differentiation,
samples were processed for frozen sections in OCT. The cells
treated with OIM were assayed with 40 × 10–3 m Alizarin red
S staining (ARS; A-5533, Sigma), Mallory’s trichrome staining
(AABFB001, American MasterTech, Lodi, CA), and immuno-
histochemistry (IHC) using antibodies against alkaline phos-
phatase (1:500, ab108337, Abcam, Cambridge, MA). Immune
cell infiltration was detected by May Grünwald (MG1L, Sigma)
and Giemsa (GS500, Sigma) solutions and IHC using antibodies
against CD3(1:500, sc-1127, Santa Cruz, Dallas, CA). Vessel inva-
sion was detected by IHC using antibodies against CD31(1:50,
ab28364, Abcam): the cells treated with AIM were stained with
0.3% Oil red O (O0628, Sigma) and 0.7% Sudan black B (199664,
Sigma); and cells treated with CIM were assayed with 1% Alcian
blue staining (A5268, Sigma) and IHC using antibodies against
type II collagen (1:50, CP18, Calbiochem, Darmstadt, Germany),
type X collagen (1:500, ab58632, Abcam) and aggrecan (1:200,
GTX113122, GeneTex). For antigen retrieval, the slides used for
type II collagen detection were incubated in 2.5% hyaluroni-
dase (H2126, Sigma) and 1 mg mL–1 protease (Sigma); the slides
used for type X collagen detection were incubated in 0.1 unit
mL–1 chondroitinase ABC (C3667, Sigma) and 1 mg mL–1 papain
(P4762, Sigma); and the slides used for aggrecan detection were
To compare the proliferation rates of 3A6 with RFP in Fmoc-FF/
Fmoc-RGD and Fmoc-FF/Fmoc-RGE hydrogels, cell-gel mixture
at a density of 2.5 × 106/200 µL was plated in the upper wells of
12 well transwell plates in GM. To assay cell proliferation at 1, 5,
and 7 d, the cell-gel mixture was transferred to a chamber slide,
observed using a Leica confocal microscope, and analyzed with
Zen software (Zeiss).
2.6. Live-Dead Staining
The viability of 3A6 in Fmoc-FF/Fmoc-RGD and Fmoc-FF/Fmoc-
RGE hydrogels was tested by
a Live-Dead assay (Molecular
Probes, Invitrogen) performed 7 d post culture. 500 µL aliquots
of the assay solution containing 4 × 10–3 m EthD-1 (ethidium
homodimer-1) and 2 × 10–3 m calcein AM were pipetted onto
each cell-gel mixture. After 30 min of incubation at room tem-
perature, the cell-gel mixture was transferred to a chamber slide,
observed using a Leica confocal microscope and analyzed with
Zen software.
2.7. In Vitro Differentiation
The cell-gel mixtures were placed in the upper wells of 12-well
transwell plates and the defined induction medium was added
in the lower well with medium changes every 2 d. Cells were
induced in osteogenic induction medium [OIM, DMEM-LG supple-
mented with 10% FBS, 50 µg mL–1 ascorbate-2 phosphate (A8960,
heated in retrieval buffer (1 × 10–2
m sodium citrate, 0.05%
Tween 20, pH 6.0). Detection was performed using a Super Sen-
sitive Polymer-HRP IHC Detection System (BioGenex Laborato-
ries Inc., San Ramon, CA) and the sections were counterstained
with hematoxylin.
Sigma), 10−8 m dexamethasone (D1756, Sigma) and 1 × 10–2
m
β-glycerophosphate (50020, Fluka)], adipogenic induction medium
[AIM, DMEM-LG supplemented with 10% FBS, 50 µg mL–1
ascorbate-2 phosphate, 10−7 m dexamethasone, 50 µm indometh-
acin (I7378, Sigma), 0.45 × 10–3 m 3-isobutyl-1-methyl-xanthine
(I5879, Sigma), and 10 µg mL–1 insulin (A1138211, Gibco, Rock-
ville, MD)], and chondrogenic induction medium [CIM, DMEM-LG
2.10. Statistical Analysis
Data were analyzed with GraphPad Prism 5 (GraphPad Software,
San Diego) and expressed as mean SEM. Statistical significance
Macromol. Biosci. 2016, DOI: 10.1002/mabi.201600192
© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
4
www.MaterialsViews.com
Early View Publication; these are NOT the final page numbers, use DOI for citation !!

Macromolecular
Bioscience
Self-Assembled Peptide-Based Hydrogels as Scaffolds for Proliferation and Multi-Differentiation of Mesenchymal Stem Cells
www.mbs-journal.de
between groups was determined by the
two-tailed Student’s t-test. Comparisons
within three groups were analyzed with
ANOVA. Significance was determined at
P < 0.05.
3. Results and Discussion
3.1. Hydrogel Formation
The peptide derivatives of Fmoc-FF, Fmoc-
RGD, and Fmoc-RGE were synthesized
by solid phase peptide synthesis (SPPS;
see Experimental Section for details)
and their corresponding chemical struc-
tures are shown in Scheme 1. Hydrogels
of Fmoc-FF/Fmoc-RGD (1:1) were pre-
pared by utilizing a sequential change in
pH and a stable gel was obtained under
physiological pH. TEM was used to deter-
Scheme 1. The chemical structures of Fmoc-FF, Fmoc-RGD, Fmoc-RGE, and the negative
stained TEM image of Fmoc-FF/Fmoc-RGD gel. Scale bar: 100 nm.
mine the nanoscopic structure of the fibers. The TEM image
reveals, for a 1:1 blend of Fmoc-FF/Fmoc-RGD gel, that the
supramolecular nanofibers are ’flat ribbons’ and they com-
prise parallel-aligned fine fibrils across their width. The
gel stiffness can be determined by oscillatory rheology
where the storage modulus (G′) and loss modulus (G″) of
the hydrogel can be obtained. The G′ of the hydrogels was
found to be higher than their G″, implying that the Fmoc-
FF/Fmoc-RGD hydrogel is a stable elastic material rather
than a viscous one (Figure S2, Supporting Information).
hydrogel were dead (Figure S1C, Supporting Information),
suggesting the self-assembly process of Fmoc-FF/Fmoc-
RGD hydrogel (but not Fmoc-FF/Fmoc-RGE hydrogel) did
not affect cell viability. Together, these results suggest
MSCs delivered in the Fmoc-FF/Fmoc-RGD hydrogel attach
to, proliferate, and survive for up to 7 d.
3.3. In Vitro Multi-Differentiation in Fmoc-FF/Fmoc-RGD
Hydrogel
In order to get the best density for differentiation after
transplantation into mice, we compared the differentia-
tion potential of MSCs at a density of 2.5 × 105, 8 × 105, and
2.5 × 106 per 200 µL in the Fmoc-FF/Fmoc-RGD hydrogel but
not in the Fmoc-FF/Fmoc-RGE hydrogel, because most of cells
in the latter died 7 d after seeding. The cells suspended in the
gel were placed in the upper-well of a 12-well transwell plate
and induced in each defined medium for up to 21 d. Cells cul-
tured in GM for 3 d served as un-differentiated controls.
When MSCs were induced for osteogenic differen-
tiation, ARS staining was used to evaluate calcium-rich
deposits. The enrichment of cells with calcium-rich
deposits was more obvious in cells at higher cell density
and with longer duration of differentiation, when com-
pared with cells in GM or at lower cell density and with
shorter duration of differentiation (Figure 1A and 1C).
In addition, Mallory’s trichrome staining was used to
evaluate collagen fiber synthesis. The enrichment of cells
with synthesis of collagen fibers was also obvious in cells
at higher cell density and with longer duration of differ-
entiation, when compared with cells in GM or at lower
cell density and with shorter duration of differentiation
(Figure 1B and 1C).
3.2. Cell Attachment, Proliferation, and Viability in
Fmoc-FF/Fmoc-RGD and Fmoc-FF/Fmoc-RGE Hydrogel
To facilitate the observation of cells in the 3D hydrogel,
the human MSC cell line overexpressed with red fluo-
rescence protein (MSC-RFP) was prepared by Lentiviral
transduction. The MSC-RFP cell-gel mixture at a density
of 2.5 × 106/200 µL was seeded in the upper-well of a tran-
swell plate and cultured in GM. Attachment and prolifera-
tion of MSC-RFP in the Fmoc-FF/Fmoc-RGD and Fmoc-FF/
Fmoc-RGE hydrogels were observed by confocal micros-
copy 1, 5, and 7 d later. At 1 d post-culture, the cells in the
Fmoc-FF/Fmoc-RGD hydrogel have already attached to and
evenly distributed in the hydrogel (Figure S1B, Supporting
Information). The numbers of RFP+ cells increased along
with the culture period from day 1 to day 7, while the cell
number in the Fmoc-FF/Fmoc-RGE hydrogel did not change
significantly (Figure S1B, Supporting Information). More-
over, a Live-Dead assay performed 7 d post-culture revealed
most of the cells in the Fmoc-FF/Fmoc-RGD hydrogel were
alive, with only several dead cells observed throughout the
whole gel, while most of the cells in the Fmoc-FF/Fmoc-RGE
Macromol. Biosci. 2016, DOI: 10.1002/mabi.201600192
© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
5
www.MaterialsViews.com
Early View Publication; these are NOT the final page numbers, use DOI for citation !!

Macromolecular
Bioscience
Y.-L. Wang et al.
www.mbs-journal.de
Figure 1. MSCs were seeded at a density of 2.5 × 105, 8 × 105, and 2.5 × 106 per 200 µL in the Fmoc-FF/Fmoc-RGD hydrogels and cultured in
GM or OIM for the indicated periods. The samples were harvested for paraffin sections, followed by staining with (A) ARS or (B) Mallory’s
trichrome. (C) ARS+ and trichrome+ cells were quantified (cells per 1000× high power field). Data represent mean SEM. *P < 0.05, **P <
0.01, and ***P < 0.001 compared with cells cultured in GM as determined by ANOVA. Bar = 50 µm.
When MSCs were induced for adipogenic differentia-
tion, Oil red O staining was used to evaluate lipid droplet
accumulation. The number of cells with oil droplet accu-
mulation was greater at higher cell density and with
longer duration of differentiation, when compared with
that in GM or at lower cell density and with shorter dura-
tion of differentiation (Figure 2A and 2C). Besides, Sudan
black B staining also revealed the number of cells with oil
droplet accumulation was greater at higher cell density
and with longer duration of differentiation, when com-
pared with that in GM or at lower cell density and with
shorter duration of differentiation (Figure 2B and 2C).
When MSCs were induced for chondrogenic differ-
entiation, Alcian blue staining was used to evaluate
polysaccharide production. The number of cells with
staining of polysaccharides was greater at higher cell den-
sity and with longer duration of differentiation, when com-
pared with that in GM or at lower cell density and with
shorter duration of differentiation (Figure 3A and 3E). In
addition, IHC for chondrogenic proteins such as type II col-
lagen (Figure 3B and 3E), type X collagen (Figure 3C and 3E),
and aggrecan (Figure 3D and 3E) also revealed the enrich-
ment of cells positive for all the immunostainings was more
obvious at higher cell density and with longer duration of
differentiation, when compared with that in GM or at lower
cell density and with shorter duration of differentiation.
Quantitative RT-PCR analysis further revealed that
MSCs induced in OIM, AIM, and CIM for 7 d significantly
Macromol. Biosci. 2016, DOI: 10.1002/mabi.201600192
© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
6
www.MaterialsViews.com
Early View Publication; these are NOT the final page numbers, use DOI for citation !!

Macromolecular
Bioscience
Self-Assembled Peptide-Based Hydrogels as Scaffolds for Proliferation and Multi-Differentiation of Mesenchymal Stem Cells
www.mbs-journal.de
Figure 2. MSCs were seeded at a density of 2.5 × 105, 8 × 105, and 2.5×106 per 200 µL in the Fmoc-FF/Fmoc-RGD hydrogels and cultured in
GM or AIM for the indicated periods. The samples were harvested for frozen sections, followed by staining with (A) Oil red O or (B) Sudan
black B. (C) Oil red O+ and Sudan black B+ cells were quantified (cells per 1000× high power field). Data represent mean SEM. *P < 0.05,
**P < 0.01, and ***P < 0.001 compared with cells cultured in GM as determined by ANOVA. Bar = 50 µm.
increased the mRNA levels of osteogenic markers
(RUNX2, Osteocalcin), adipogenic markers (PPAR-γ, LPL),
and chondrogenic markers (Sox9, Col2a1, and Aggrecan),
respectively, compared to the non-differentiation controls
(Figure S3, Supporting Information). These data together
suggest MSCs delivered in the Fmoc-FF/Fmoc-RGD hydro-
gels and induced for multi-differentiation can undergo
differentiation into bone, fat, and cartilage lineages
in vitro.
3D hydrogel, we chose this density for the in vivo multi-
differentiation assay. Moreover, appropriate periods of in
vitro differentiation before transplantation are required
for successful differentiation upon transplantation under-
neath the dorsal skin of mice.[28] Because the cell-hydrogel
mixture is a semi-liquid form, we used spongostan® for
in vivo delivery, which was convenient for sample harvest
after transplantation and only induced limited immune
cell infiltration and vessel invasion when transplanted
alone (Figure S4A, S4B, and S5, Supporting Information).[26]
Thus, cells were delivered either in Fmoc-FF/Fmoc-RGD or
in Fmoc-FF/Fmoc-RGE and induced for each differentiation
for 7 d, followed by injection of the mixture into gelatinous
sponges and implantation underneath the dorsal skin of
immunodeficient mice for another 4 weeks.
3.4. In Vivo Multi-Differentiation in Fmoc-FF/Fmoc-RGD
and Fmoc-FF/Fmoc-RGE Hydrogels
Because the in vitro differentiation assay revealed 2.5 × 106
per 200 µL was the best density for differentiation in the
Macromol. Biosci. 2016, DOI: 10.1002/mabi.201600192
© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
7
www.MaterialsViews.com
Early View Publication; these are NOT the final page numbers, use DOI for citation !!

Macromolecular
Bioscience
Y.-L. Wang et al.
www.mbs-journal.de
Figure 3. MSCs were seeded at a density of 2.5 × 105, 8 × 105, and 2.5 × 106 per 200 µL in the Fmoc-FF/Fmoc-RGD hydrogels and cultured in
GM or CIM for the indicated periods. The samples were harvested for paraffin sections, followed by staining with (A) Alcian blue or assayed
with IHC for (B) collagen II (C) collagen X or (D) aggrecan. (E) Alcian blue +, collagen II+, collagen X+, and aggrecan+ cells were quantified
(cells per 1000× high power field). Data represent mean SEM. *P < 0.05, **P < 0.01, and ***P < 0.001 compared with cells cultured in GM
as determined by ANOVA. Bar = 50 µm.
Macromol. Biosci. 2016, DOI: 10.1002/mabi.201600192
© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
8
www.MaterialsViews.com
Early View Publication; these are NOT the final page numbers, use DOI for citation !!

Macromolecular
Bioscience
Self-Assembled Peptide-Based Hydrogels as Scaffolds for Proliferation and Multi-Differentiation of Mesenchymal Stem Cells
www.mbs-journal.de
Figure 4. MSCs were seeded at a density of 2.5 × 106 per 200 µL in the Fmoc-FF/Fmoc-RGD or Fmoc-FF/Fmoc-RGE hydrogels and cultured
in GM or OIM for 7 d. Then, the cell-gel mixtures were aspirated with syringes and each injected into a gelatin sponge, followed by trans-
plantation into immunodeficient mice for four weeks. The grafts were harvested for paraffin sections, followed by (A) H&E staining, (B) ARS
staining, (C) Mallory’s trichrome staining, and (D) IHC for alkaline phosphatase. (E) ARS+, trichrome+ cells and alkaline phosphatase+ cells
were quantified (cells per 1000× high power field). Data represent mean SEM. *P < 0.05, **P < 0.01, and ***P < 0.001 compared with cells
cultured in GM as determined by ANOVA. Bar = 50 µm.
Gross evaluation of specimens recovered from the mice
revealed that grafts with cells delivered in Fmoc-FF/Fmoc-
RGD had greater volume than those with cells delivered
in Fmoc-FF/Fmoc-RGE in GM, OIM, and AIM, but not in
CIM (Figure S3, Supporting Information). H&E staining
also showed cell numbers were greater in grafts with cells
delivered in Fmoc-FF/Fmoc-RGD than those delivered
in Fmoc-FF/Fmoc-RGE when examined for osteogenic
(Figure 4A), adipogenic (Figure 5A), and chondrogenic dif-
ferentiation (Figure 6A).
delivered in Fmoc-FF/Fmoc-RGD than those delivered
in Fmoc-FF/Fmoc-RGE or the undifferentiated controls
(Figure 4B and 4E). Mallory’s trichrome staining also
showed the number of cells with the synthesis of col-
lagen fibers was greater in cells in Fmoc-FF/Fmoc-RGD
than those in Fmoc-FF/Fmoc-RGE or the undifferentiated
controls (Figure 4C and 4E). IHC with anti-alkaline phos-
phatase antibodies further showed the number of cells
positive for alkaline phosphatase was greater in cells in
Fmoc-FF/Fmoc-RGD than those in Fmoc-FF/Fmoc-RGE or
the undifferentiated controls (Figure 4D and 4E).
For grafts assayed for osteogenic differentia-
tion, ARS staining revealed the enrichment of cells
with calcium-rich deposits was more obvious in cells
For grafts assayed for chondrogenic differentiation,
Alcian blue staining revealed the number of cells with the
Macromol. Biosci. 2016, DOI: 10.1002/mabi.201600192
© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
9
www.MaterialsViews.com
Early View Publication; these are NOT the final page numbers, use DOI for citation !!

Macromolecular
Bioscience
Y.-L. Wang et al.
www.mbs-journal.de
Figure 5. MSCs were seeded at a density of 2.5 × 106 per 200 µL in the Fmoc-FF/Fmoc-RGD or Fmoc-FF/Fmoc-RGE hydrogels and cultured in
GM or AIM for 7 d. Then, the cell-gel mixtures were aspirated with syringes and each injected into a gelatin sponge, followed by transplan-
tation into immunodeficient mice for four weeks. The grafts were harvested for frozen sections, followed by (A) H&E staining, (B) Oil red
O staining, and (C) Sudan black B staining. (D) Oil red O+ and Sudan black B+ cells were quantified (cells per 1000× high power field). Data
represent mean SEM. *P < 0.05, **P < 0.01, and ***P < 0.001 compared with cells cultured in GM as determined by ANOVA. Bar = 50 µm.
synthesis of chondrogenic polysaccharides was greater
in Fmoc-FF/Fmoc-RGD than those in Fmoc-FF/Fmoc-
RGE or the undifferentiated controls (Figure 6B and 6F).
IHC with antibodies against type II collagen (Figure 6C
and 6F), type X collagen (Figure 6D and 6F), or aggrecan
(Figures 6E and 6F) showed the number of cells with the
synthesis of these chondrogenic proteins was greater in
Fmoc-FF/Fmoc-RGD than those in Fmoc-FF/Fmoc-RGE or
the undifferentiated controls.
Together, these results suggest MSCs delivered in the
Fmoc-FF/Fmoc-RGD hydrogels and induced for multi-dif-
ferentiation can maintain the property in vitro and con-
tinuously differentiate into bone, fat, and cartilage line-
ages in vivo.
for tissue engineering using MSCs has not been demon-
strated. This is the first study to demonstrate the suita-
bility of Fmoc-FF/Fmoc-RGD hydrogel to serve as scaffolds
for multi-differentiation of MSCs both in vitro and in vivo.
We showed that MSCs delivered in Fmoc-FF/Fmoc-RGD
hydrogel had better survival and replication ability than
those delivered in Fmoc-FF/Fmoc-RGE hydrogel. Moreover,
we showed that MSCs delivered in Fmoc-FF/Fmoc-RGD
hydrogel at a density of 2.5 × 106 per 200 µL had better dif-
ferentiation ability than those at a density of 2.5 × 105 or
8 × 105 per 200 µL in terms of in vitro differentiation into
osteoblasts, adipocytes, and chondrocytes. Finally, we con-
firmed that Fmoc-FF/Fmoc-RGD hydrogel provides better
scaffolds for in vivo differentiation of MSCs into osteo-
blasts, adipocytes, and chondrocytes than Fmoc-FF/Fmoc-
RGE hydrogel. The results shown here are promising and
recognize the usefulness of Fmoc-FF/Fmoc-RGD hydrogel
for future applications in MSC-based tissue engineering.
There are several strengths contributing to the suc-
cesses in the current study. Primary cultures of human
MSCs vary and depend on the individuals who provide the
3.5. General Discussion
Many kinds of hydrogels have been explored as scaffolds for
tissue engineering and regeneration medicine. Although
Fmoc-FF/Fmoc-RGD hydrogel has been investigated for
delivering anchorage-dependent cells,[26] its application
Macromol. Biosci. 2016, DOI: 10.1002/mabi.201600192
© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
10
www.MaterialsViews.com
Early View Publication; these are NOT the final page numbers, use DOI for citation !!

Macromolecular
Bioscience
Self-Assembled Peptide-Based Hydrogels as Scaffolds for Proliferation and Multi-Differentiation of Mesenchymal Stem Cells
www.mbs-journal.de
Figure 6. MSCs were seeded at a density of 2.5 × 106 per 200 µL in the Fmoc-FF/Fmoc-RGD or Fmoc-FF/Fmoc-RGE hydrogels and cultured in
GM or CIM for 7 d. Then, the cell-gel mixtures were aspirated with syringes and each injected into a gelatin sponge, followed by transplan-
tation into immunodeficient mice for four weeks. The grafts were harvested for paraffin sections, followed by (A) H&E staining, (B) Alcian
blue staining, and IHC for (C) collagen II, (D) collagen X, and (E) aggrecan. (E) Alcian blue +, collagen II+, collagen X+ and aggrecan+ cells
were quantified (cells per 1000× high power field). Data represent mean SEM. *P < 0.05, **P < 0.01, and ***P < 0.001 compared with cells
cultured in GM as determined by ANOVA. Bar = 50 µm.
tissues for isolation of MSCs. We therefore used immor-
talized MSCs with the hTERT gene introduced, thereby
increasing stem-like properties and decreasing sponta-
neous differentiation.[31] These cells are also feasible for
fast and easy amplification in a short time, saving a lot of
time in culture. Cells delivered in hydrogels are difficult
for long-term microscopic observation. Here, MSCs were
lentivirally transduced with RFP fluorescence protein,
making microscopic observation using a fluorescence
microscope to evaluate cell number and morphology easy.
Although the cell-hydrogel mixture was aspirated from
the culture device and injected directly into the space
underneath the skin in mice, the mixture did not form
an integral and solid structure, which prevented future
analysis. As a result, we injected these MSCs-Fmoc-FF/
Fmoc-RGD mixtures into pieces of spongostan® and trans-
planted whole constructs underneath the dorsal skin
of mice. Note that the spongostan® did not significantly
affect the Fmoc-FF/Fmoc-RGD gel (Figure S6, Supporting
Information). These modifications led to the successful
Macromol. Biosci. 2016, DOI: 10.1002/mabi.201600192
© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
11
www.MaterialsViews.com
Early View Publication; these are NOT the final page numbers, use DOI for citation !!

Macromolecular
Bioscience
Y.-L. Wang et al.
www.mbs-journal.de
differentiation of these cells into osteogenic, adipogenic,
and chondrogenic lineages in vivo.
hydrogels was spherical, supporting our results.[37] They
demonstrated that RGD ligands enhanced cartilage-spe-
cific gene expression and matrix synthesis, but only when
mechanically stimulated, suggesting that cell-matrix
interactions mediate chondrocyte response to mechanical
stimulation.
Zhou and co-workers first applied Fmoc-FF/Fmoc-RGD
hydrogel for in vitro delivery of primary human dermal
fibroblasts.[26] This hydrogel provides a 3D nanofibrous
network that is highly hydrated and stiff and presents
bioactive ligands for adhesion, spreading, and prolifera-
tion of human dermal fibroblasts in vitro, while Fmoc-FF/
Fmoc-RGE hydrogel fails to promote adhesion, spreading,
and proliferation of these cells. However, RGD inhibition
has been shown to enhance in vitro chondrogenesis of
bovine MSCs in 3D alginate gel.[34] Moreover, the addi-
tion of an anti-α5 antibody reduced sGAG accumulation,
while an anti-αvβ3 integrin blocking antibody had no
effect on chondrogenesis. These results suggest that the
role of cell-matrix interactions in regulating cell prolif-
eration and differentiation is dependent on the cell type
or material used and requires extensive investigation.
The current study investigated the application of Fmoc-
FF/Fmoc-RGD hydrogel in tissue engineering in a robust
manner. We compared the effects of Fmoc-FF/Fmoc-RGD
hydrogel and Fmoc-FF/Fmoc-RGE hydrogel on the pro-
liferation, survival, and multi-differentiation of human
MSCs into osteoblasts, adipocytes, and chondrocytes. In
addition, some of the effects were investigated both in
vitro and in vivo. Our data suggest Fmoc-FF/Fmoc-RGD
hydrogel was suitable for MSC-based tissue engineering
for bone, fat, and cartilage regeneration. Similarly, RGD-
modified poly (ethylene glycol) (PEG) hydrogel has been
applied for stimulating chondrogenic and osteogenic dif-
ferentiation.[35] Together, these data suggest RGD-modifi-
cation of hydrogels can be applied for tissue engineering
using MSCs.
The discrepancy between our study and previous
studies in showing the roles of RGD interaction in chon-
drogenesis of MSCs may be dependent on the types of
MSCs and hydrogels used. The response to RGD modifi-
cation in bovine MSCs of the previous study[34] and that
in the human MSCs of our study may be different. A
previous study mentioned that a rounded morphology
plays an important role for chondrogenic differentia-
tion.[36] Consistent with our previous studies,[1,28] MSCs
encapsulated in alginate gel adopt a spherical mor-
phology and differentiate into chondrocytes in defined
chondrogenic medium.[34] However, MSCs encapsulated
in RGD-modified alginate gel spread and take a fibro-
blastic morphology, while failing to express chondro-
genic genes after 7 d of culture in defined chondrogenic
medium. However, MSCs delivered in Fmoc-FF/Fmoc-RGD
hydrogel spread in GM, and most of the cells in defined
chondrogenic medium underwent chondrogenic differ-
entiation and adopted a spherical morphology both in
vitro and in vivo. Villanueva and co-workers found that
the morphology of chondrocytes in RGD-modified PEG
We aspirated and injected these MSCs-Fmoc-FF/Fmoc-
RGD hydrogel mixtures into pieces of spongostan®, fol-
lowing by transplantation underneath the dorsal skin of
mice. The gel form of the cell-hydrogel mixture makes it
possible for injection into intervertebral discs[38] and the
drill hole defect in rat tibia[39] to enhance disc regenera-
tion and bone formation, respectively. There are limita-
tions of the study that need to be addressed. To prevent
immune rejection, small immunodeficient animals were
used for the investigation of the in vivo differentiation
ability of human MSCs. To have consistent results, immor-
talized MSCs rather than primary MSCs were used. Subcu-
taneous sites rather than orthotopic sites were used for
transplantation. To overcome these shortcomings, future
studies with the use of larger animals and their autolo-
gous or allogeneic MSCs are required.
With regard to 3D bio-printing (the process of creating
spatially-controlled cell patterns to reproduce the behavior
of biological tissues), MSCs-Fmoc-FF/Fmoc-RGD mixtures
after appropriate differentiation in vitro may serve as 3D
printing inks, such as adipo-ink, osteo-ink, and chondro-
ink.[40] In comparison with other bio-inks, such as decel-
lularized ECM,[41] DNA-based hydrogels,[42] and collagen
or alginate based bio-inks,[43] Fmoc-FF/Fmoc-RGD-based
bio-inks may be more available and convenient to pre-
pare. Besides, these cell-hydrogel mixtures might also be
applied for tissue engineering and regeneration medicine
especially in wound healing, cartilage repair, and recovery
after injuries such as broken bones in the future.
4. Conclusion
We demonstrated that MSCs when seeded in Fmoc-FF/
Fmoc-RGD hydrogel can be induced for differentiation into
osteoblasts, adipocytes, and chondrocytes both in vitro and
in vivo, while those seeded in Fmoc-FF/Fmoc-RGE hydrogel
failed to survive and decreased in their ability to differ-
entiate. These results suggest that Fmoc-FF/Fmoc-RGD
hydrogel is a suitable scaffold for MSC-based tissue engi-
neering and may be applied for cell therapies and regen-
erative medicine.
Supporting Information
Supporting Information is available from the Wiley Online
Library or from the author.
Macromol. Biosci. 2016, DOI: 10.1002/mabi.201600192
© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
12
www.MaterialsViews.com
Early View Publication; these are NOT the final page numbers, use DOI for citation !!

Macromolecular
Bioscience
Self-Assembled Peptide-Based Hydrogels as Scaffolds for Proliferation and Multi-Differentiation of Mesenchymal Stem Cells
www.mbs-journal.de
Acknowledgements: The study was supported by grants from
the Taipei Veterans General Hospital (V102E1-008), the Ministry
of Science and Technology (MOST 103-2321-B-010-024; MOST
103-2321-B-010-018), and the Academia Sinica Translational
Medicine Program (5202401023-2325). Space and administrative
support was received from the Institute of Biomedical Sciences,
Academia Sinica, and the Institute of Biomedical Science
Building, Taipei Veterans General Hospital.
[21] A. Khanlari, J. E. Schulteis, T. C. Suekama, M. S. Detamore,
S. H. Gehrke, J. Appl. Polym. Sci. 2015, 132.
[22] J. Yu, K. T. Du, Q. Fang, Y. Gu, S. S. Mihardja, R. E. Sievers,
J. C. Wu, R. J. Lee, Biomaterials 2010, 31, 7012.
[23] C. N. Salinas, K. S. Anseth, Biomaterials 2008, 29, 2370.
[24] J. C. Chang, S. H. Hsu, D. C. Chen, Biomaterials 2009, 30, 6265.
[25] V. Jayawarna, M. Ali, T. A. Jowitt, A. F. Miller, A. Saiani,
J. E. Gough, R. V. Ulijn, Adv. Mater. 2006, 18, 611.
[26] M. Zhou, A. M. Smith, A. K. Das, N. W. Hodson, R. F. Collins,
R. V. Ulijn, J. E. Gough, Biomaterials 2009, 30, 2523.
[27] M. Zhou, R. V. Ulijn, J. E. Gough, J. Tissue Eng. 2014, 5,
2041731414531593.
Received: May 14, 2016; Revised: September 27, 2016;
Published online: ; DOI: 10.1002/mabi.201600192
[28] H. L. Ma, T. H. Chen, L. Low-Tone Ho, S. C. Hung, J. Biomed.
Mater. Res. A 2005, 74, 439.
[29] R. Siddappa, R. Licht, C. van Blitterswijk, J. de Boer, J. Orthop.
Res. 2007, 25, 1029.
Keywords: differentiation; hydrogels; mesenchymal stem cells;
self-assembled peptides; supramolecular chemistry
[30] S. C. Mendes, J. M. Tibbe, M. Veenhof, K. Bakker,
S. Both, P. P. Platenburg, F. C. Oner, J. D. de Bruijn,
C. A. van Blitterswijk, Tissue Eng. 2002, 8, 911.
[31] C. C. Tsai, C. L. Chen, H. C. Liu, Y. T. Lee, H. W. Wang, L. T. Hou,
S. C. Hung, J. Biomed. Sci. 2010, 17, 64.
[1] S. Zhang, Nature Biotechnol. 2003, 21, 1171.
[2] W. Y. Seow, C. A. E. Hauser, Mater. Today 2014, 17, 381.
[3] E. Arslan, I. C. Garip, G. Gulseren, A. B. Tekinay, M. O. Guler,
Adv. Healthcare Mater. 2014, 3, 1357.
[4] C. A. Hauser, S. Zhang, Chem. Soc. Rev. 2010, 39, 2780.
[5] X. Zhao, S. Zhang, Macromol. Biosci. 2007, 7, 13.
[6] S. I. Stupp, Nano Lett. 2010, 10, 4783.
[32] P. Lin, D. Correa, Y. Lin, A. I. Caplan, PloS one 2011, 6, e23891.
[33] U. O’Doherty, W. J. Swiggard, M. H. Malim, J. Virology 2000,
74, 10074.
[7] P. W. Kopesky, E. J. Vanderploeg, J. S. Sandy, B. Kurz,
A. J. Grodzinsky, Tissue Eng. Part A 2010, 16, 465.
[8] J. M. Anderson, A. Andukuri, D. J. Lim, H. W. Jun, ACS Nano
2009, 3, 3447.
[9] D. L. Hern, J. A. Hubbell, J. Biomed. Mater. Res. 1998, 39, 266.
[10] P. X. Ma, Adv. Drug Delivery Rev. 2008, 60, 184.
[11] A. K. Silva, C. Richard, M. Bessodes, D. Scherman,
O. W. Merten, Biomacromolecules 2009, 10, 9.
[34] J. T. Connelly, A. J. Garcia, M. E. Levenston, Biomaterials
2007, 28, 1071.
[35] N. J. Steinmetz, S. J. Bryant, Acta Biomater. 2011, 7, 3829.
[36] L. Gao, R. McBeath, C. S. Chen, Stem Cells 2010, 28, 564.
[37] I. Villanueva, C. A. Weigel, S. J. Bryant, Acta Biomater. 2009,
5, 2832.
[38] A. H. Milani, A. J. Freemont, J. A. Hoyland, D. J. Adlam,
B. R. Saunders, Biomacromolecules 2012, 13, 2793.
[39] S. Dhivya, S. Saravanan, T. P. Sastry, N. Selvamurugan,
J. Nanobiotechnol. 2015, 13, 40.
[12] A. H. Zisch, M. P. Lutolf, J. A. Hubbell, Cardiovasc. Pathol.
2003, 12, 295.
[13] D. J. Prockop, Science 1997, 276, 71.
[14] C. C. Wei, A. B. Lin, S. C. Hung, Cell Transplant 2014, 23, 505.
[15] A. I. Caplan, Tissue Eng. 2005, 11, 1198.
[16] J. Gao, A. I. Caplan, Chir. Organic Mol. 2003, 88, 305.
[17] M. P. Lutolf, J. A. Hubbell, Nature Biotechnol. 2005, 23, 47.
[18] E. Ruoslahti, M. D. Pierschbacher, Science 1987, 238, 491.
[19] E. Ruoslahti, M. D. Pierschbacher, Cell 1986, 44, 517.
[20] D. A. Cheresh, J. W. Smith, H. M. Cooper, V. Quaranta, Cell
1989, 57, 59.
[40] F. Pati, J. Jang, D. H. Ha, S. Won Kim, J. W. Rhie, J. H. Shim,
D. H. Kim, D. W. Cho, Nature Commun. 2014, 5, 3935.
[41] J. P. Jung, D. B. Bhuiyan, B. M. Ogle, Biomater. Res. 2016, 20,
27.
[42] C. Li, A. Faulkner-Jones, A. R. Dun, J. Jin, P. Chen, Y. Xing,
Z. Yang, Z. Li, W. Shu, D. Liu, R. R. Duncan, Angew. Chem. Int.
Ed. Engl. 2015, 54, 3957.
[43] C. J. Ferris, K. G. Gilmore, G. G. Wallace, M. In het Panhuis,
Appl. Microbiol. Biotechnol. 2013, 97, 4243.
Macromol. Biosci. 2016, DOI: 10.1002/mabi.201600192
© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
13
www.MaterialsViews.com
Early View Publication; these are NOT the final page numbers, use DOI for citation !!