Cell adhesive hydrogels synthesised by copolymerisation of arg-protected Gly-Arg-Gly-Asp-Ser methacrylate monomers and enzymatic deprotection

Article abstract of DOI:10.1039/b813392a

This work reports the synthesis of protected Gly-Arg-Gly-Asp-Ser functionalised hydrogels, which are deprotected (and activated for cell adhesion) by reaction with glutathione-S-transferase. The Royal Society of Chemistry.

Full text of DOI:10.1039/b813392a

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Cell adhesive hydrogels synthesised by copolymerisation of arg-protected
Gly-Arg-Gly-Asp-Ser methacrylate monomers and enzymatic
deprotectionw
Lynne Perlin,ab Sheila MacNeil*b and Stephen Rimmer*a
Received (in Cambridge, UK) 1st August 2008, Accepted 8th September 2008
First published as an Advance Article on the web 9th October 2008
DOI: 10.1039/b813392a
This work reports the synthesis of protected Gly-Arg-Gly-Asp-Ser
functionalised hydrogels, which are deprotected (and activated for
cell adhesion) by reaction with glutathione-S-transferase.
to liberate amines from aryl sulfonamides in the presence of the
cosubstrate; reduced glutathione (GSH).6,7 In a previous report
Todd et al. used proteolytic enzymes to cleave a protected Phe
from an Arg-Gly-Asp sequence tethered to a hydrogel and
showed enhanced adhesion of osteoblasts after cleavage.8 Also,
previous enzymatic deprotections involving enzymes other
than GST are covered in ref. 9. Here we report the use of a
methacrylated GRGDS peptide with the Arg residue protected
with an aryl sulfonamide, which can be removed post-
polymerisation with GST, and a similar polymerisable peptide
containing a spacer arm between the methacrylate and peptide
sections.
Functionalisation with peptides based on the arginine-glycine-
aspartic acid (Arg-Gly-Asp) sequence is one of a range of key
technologies for improving the cell adhesive properties of
substrates for use as scaffolds in tissue engineering.1 Two
strategies are of general use for covalent attachment: copoly-
merisation with a suitable peptide monomer or direct reaction
with the peptide onto activated surfaces. However, both of
these strategies generally use the unprotected peptide so that it
is essential to carefully consider the correct choice of reaction
media including pH. Also, the Arg residue can bind metals,
which can give rise to complications in controlled radical
polymerisations. Arg-Gly-Asp containing methacrylate mono-
mers have been previously prepared and polymerized2–5 but
these strategies always require the use of aqueous or highly
polar reaction conditions. Side reactions involving Arg or other
nucleophilic/basic peptides during polymerisation are possible
due to the use of the non-protected amino acid. For example,
Hern and Hubbell noted the occurrence of side reactions
during acrylation of Arg-containing peptides.4 Arg also acts
as a basic catalyst and requires protection during solid-phase
peptide synthesis. This latter aspect led us to investigate the
possibility of polymerising Arg protected peptides with meth-
acrylate end groups and then deprotecting the peptide on the
potential cell-adhesive polymer. Deprotection of the peptide on
the hydrogel has the added advantage of a simpler peptide
purification procedure. Unfortunately, deprotection of the Arg
protected with tert-butoxy oxycarbonyl (tBOC), the group
usually used to protect Arg, requires reaction with high
concentrations of trifluoroacetic acid (TFA). This procedure
leads to hydrolysis of methacrylate esters on the polymers and
in the case of polymer hydrogels cross-linked with ethandiol
dimethacrylate led to undesirable cleavage of cross-links and
dissolution. Thus, our search for an alternative protection–
deprotection strategy in which the deprotection could be
delayed until after the polymer had been synthesised led us
to a previous report of using glutathione-S-transferase (GST)
In order to examine the feasibility, of deprotecting aryl
sulfonamides of Arg by application of GST we prepared Arg
derivatives with the guanidine function protected with the
4-bromobenzene sulfonamide (4-Bbs). N-carbobenzoxy-L-
arginine (N0-bromobenzene sulfonamide) (1) was synthesized
using the reaction of N-carbobenzoxy-L-arginine with
4-bromobenzylsulfonyl chloride in acetone and 3 M sodium
hydroxide solution. The syntheses of 1 were achieved with
yields of B50 and 89% purity as assessed by HPLC. The
product was used to assess the removal of the 4-Bbs group by
the GST. A range of concentrations (50, 100 and 200 mmol
cmꢀ3) of 1 was incubated with two different concentrations of
GST (0.05 and 0.5 mg cmꢀ3) at room temperature for 7 days.
However, even after this extended reaction time the highest
yield of the deprotected amino acid was only 40%
([1]= 200 mmol dmꢀ3, [GST] = 0.5 mg cm3). Increasing the
reaction temperature to 37 1C produced a large improvement
in the yield of deprotected amino acid so that at this tempera-
ture conversions of 85% were achieved in 24 hours.
([1] = 50–200 mmol dmꢀ3, [GST] = 0.05–0.5 mg cm3).
In general cells do not adhere to high water content hydro-
gels such as those based on poly(1,2-propandiol methacrylate)
(PGMA) or poly(N-vinyl pyrolidinone) unless they are mod-
ified with cell-adhesive peptides, alkyl amines or hydrophobic
sequences.1–5,10–12 Thus, hydrogels based on these monomer
units are a good choice for derivatisation with RGD as the
effects of the peptide should not be complicated by the effects
of non-specific protein (and cellular) adsorption. As far as we
are aware, PGMA hydrogels functionalised with cell adhesive
peptides have not been previously reported.
a Dept. of Chemistry, Univ. of Sheffield, Sheffield, South Yorkshire,
UK S3 7HF. E-mail: s.rimmer@sheffield.ac.uk
Two protected polymerisable peptides were prepared as
shown in Scheme 1, by solid-phase peptide synthesis. 2 was
used to directly attach the peptide to the polymer chain, and 3
provides for an alkyl spacer between the polymer chain and
the peptide sequence. Synthesis of 2 and 3 was achieved by
b Dept. of Engineering Materials, The Kroto Research Institute, Univ.
of Sheffield, Sheffield, South Yorkshire, UK.
E-mail: s.macneil@sheffield.ac.uk
w Electronic supplementary information (ESI) available: Further
experimental details. See DOI: 10.1039/b813392a
ꢁc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 5951–5953 | 5951

View Article Online
Scheme 1 The structure of the two peptide GRGDS monomers.
using the fluorenylmethoxy carbonyl (FMOC) derivative of
the Bbs-protected Arg, 4. 4 was synthesised in 86% yield and
99% purity (HPLC). The FMOC group is not stable to the
conditions required to add the 4-Bbs group. Therefore, the
synthesis of 4 required a three-step process using an alternative
strategy for protection of a-amine group, as shown in Scheme S1,
ESIw. In the first attempt of this approach the 4-Bbs derivative
of Arg with the a-amine protected with the carbobenzoxy
group was used. However, removal of the carbobenzoxy group
with the usual H2/Pd procedure also removed the bromine
from the 4-Bbs group.
Fig. 1 Cell viability data (MTT assay) of fibroblasts cultured in
serum-free media for 24 h on substrates containing various nominal
amounts of GRGDS: materials prepared by copolymerisation with (a)
2 and (b) 3. ANOVA showed a significant difference in both data sets
(p o 0.001), * denotes results significantly different from the others
(post hoc analysis using the Tukey procedure). TCP indicates cells
cultured on tissue culture plastic as an internal positive control.
Phase-contrast images of the cells on PGMA functionalised
with 2 are shown in Fig. 2. PGMA with the lowest nominal
concentration of peptide was a poor substrate for cell culture, as
shown in Fig. 2(a). Cells on non-functionalised PGMA do not
adhere to any useful extent as we have previously reported in ref.
10 and 12. Here cells were clearly rounded and appeared to be
poorly adhered to the substrate. In contrast, Fig. 2(b), (c) and (d)
confirmed the viability data shown in Fig. 1 showing that the
material produced with a nominal peptide concentration of 10
mmol gꢀ1 provided an excellent substrate and the cells were
observed to be well spread and were beginning to form cell–cell
contacts. However, the micrographs also showed a morphologi-
cal difference between the response of cells to material produced
with a nominal peptide concentration of 0.01 mmol gꢀ1 and the
materials with peptide concentrations of 0.1 and 1 mmol gꢀ1. The
viability data (Fig. 1) provide only a small indication that there is
a difference but it is clear from the micrographs that increasing
the amount of peptide from 0.01 to 0.1 mmol gꢀ1 had a
substantial effect on cell morphology.
Peptide monomers 2 and 3 contained identical peptide
sequences (GRGDS) but 3 contained an alkyl spacer, which
was designed to add some space between the integrin binding
sequence and the polymer chain, whereas 2 does not. The alkyl
chain was added to the peptide at the end of solid-phase
synthesis by adding 12-(fluorenylmethyl carbamate)dodecanoic
acid to the last amino acid. The final peptide monomers were
prepared by removal of the FMOC and reaction of the free
amine terminus with methacrylic acid.
Both peptide monomers were produced in 96% (HPLC)
purity after preparative HPLC. Copolymers (P(GMA-co-
EDMA) containing GMA, ethylene glycol dimethacrylate
(EDMA) (2.5 wt% of the GMA) and various concentrations
of each peptide were synthesised as films (60 mm) polymerised
onto a PET backing sheet by UV-initiated radical polymerisa-
tion. The films were washed with ethanol and sterile
phosphate-buffered saline, incubated with GST for 24 h at
37 1C and then primary human dermal fibroblasts were
cultured on the surface for 24 h. Cell viability (and indirectly
cell number) was assessed by the MTT assay and their
morphology was examined by optical microscopy (Fig. 2).
Fig. 1 shows the cell viability data from this set of experiments
for both peptide monomers. Two observations can be made
from these data. There is a substantial and significant increase
in the MTT values at the highest nominal peptide concentra-
tions, lower concentrations did not produce any improvement.
Secondly there was no significant difference in the perfor-
mance of the cells on the two peptides.z Cells on these
materials also performed better, in these serum-free cultures,
than on the tissue culture plastic (TCP) control wells.
The similarity in the performance of the polymers functio-
nalised with 2 or 3 was unexpected as in general spacer arms,
usually poly(ethylene glycol), are regarded as useful molecular
features in a cell adhesion promoting monomer. However, the
spacer used in this work was a hydrophobic moiety, which was
attached to a highly swollen system. It appears that in this
situation the use of a spacer does not provide any performance
benefits.
Fig. 2 Phase-contrast optical micrographs of fibroblasts cultured in
serum-free media for 24 h on PGMA hydrogels containing: (a) 0; 0.01;
(b) 0.1; (c) 1 and (d) 10 mmol gꢀ1 (amount of peptide per mass of
monomer feed).
ꢁc
This journal is The Royal Society of Chemistry 2008
5952 | Chem. Commun., 2008, 5951–5953

View Article Online
Fig. 4 MTT cell viability data and phase-contrast micrographs of
fibroblasts cultured on PGMA functionalised with 2 (10 mmol gꢀ1):
before application of soluble GRGDS and 60 min after the addition.
culture plastic (not shown) remained attached to the substrate
when this soluble peptide was added.
In conclusion we have shown that P(GMA-co-EDMA)
hydrogels, which are generally non-fouling and non-cell ad-
hesive, can be modified by copolymerisation of GMA, EDMA
and peptide containing methacrylate monomers. The Arg
amino acid residue of these peptide monomers can be pro-
tected with 4-Bbs, which can be removed by reaction with the
transferase, GST. Once deprotected these materials provide
excellent substrates for cell culture, and HDFs appear to
proliferate faster on the optimised materials than on TCP.
In this work GST was added as a separate step and is clearly
an effective methodology for deprotecting peptides without
any adverse effects on the subsequent interaction of cells with
these peptide-functionalised hydrogels.
Fig. 3 Micrographs of fibroblasts cultured for 24 h on: (a) TCP; (b),
(c) and (d) PGMA functionalised with 2 (10 mmol gꢀ1) and then used
without deprotection with GST (b) or following 1 day (c) or 3 days (d)
deprotection. Cell nuclei stained with DAPI (blue) and F-actin stained
with FITC-phallodin (green).
Fig. 3 shows micrographs of cells cultured on a polymer
functionalised with 2 (10 mmol gꢀ1). The nuclei are stained
with DAPI and the cytoskeleton can be seen via staining of
F-actin with FITC-phallodin. The control cells grown on TCP
(a) displayed a clear and extensive network of actin at 24 h. In
contrast, cells on (b)—no deprotection—were still quite
rounded with no real cytoskeleton organisation. Cells on
(c)—1 day deprotection—were more organised than cells on
(b) and cells on (d) (material functionalised with 2 that had
been reacted with GST for 3 days) were almost as organised as
the cells cultured on TCP. While these results are qualitative
an observer blind to the cell substrates would have had no
difficulty distinguishing micrographs (b) and (c) from (a) but
would not have reliably discriminated between (d) and (a).
Thus cells grown on materials that had undergone only 1 day
deprotection showed some F-actin formation but this was
substantially less than that achieved by cells on materials
deprotected for 3 days. The data provide good evidence for
the release of the GRGDS functionality following deprotection
of the guanidine moiety on the Arg unit.
Notes and references
z Using Student’s-T test, there was no significant difference between
the two TCP controls (p = 0.12) and no significant difference between
the materials prepared with 2 or 3 at 10 mmol gꢀ1 (p = 0.26).
1 L. Perlin, S. MacNeil and S. Rimmer, Soft Matter, 2008, DOI:
10.1039/b801646a.
2 L. M. Weber, K. N. Hayda, K. Haskinsb and K. S. Anseth,
Biomaterials, 2007, 28, 3004.
3 C. R. Nuttelman, M. C. Tripod and K. S. Anseth, Matrix Biol.,
2005, 24, 208.
4 D. L. Hern and J. A. Hubbell, J. Biomed. Mater. Res., 1998, 39,
266.
5 C. E. Kang, E. J. Gemeinhart and R. A. Gemeinhart, J. Biomed.
Mater. Res., 2004, 71A, 403.
6 Z. Zhao, K. A. Koeplinger, T. Paterson, R. A. Conradi, P. S.
Burton, A. Suarato, R. L. Heinrikson and A. G. Tomasselli, Drug
Metab. Dispos., 1999, 27, 992.
7 K. A. Koeplinger, Z. Zhao, T. Peterson, J. W. Leone, F. S.
Schwende, R. L. Heinrikson and A. G. Tomasselli, Drug Metab.
Dispos., 1999, 27, 986.
8 S. J. Todd, D. Farrar, J. E. Gough and R. V. Ulijn, Soft Matter,
2007, 3, 547.
9 D. Kadereit and H. Waldmann, Chem. Rev., 2001, 101, 3367.
10 Y. Sun, J. Collett, N. J. Fullwood, S. Mac Neil and S. Rimmer,
Biomaterials, 2007, 28, 661.
11 L. Smith, S. Rimmer and S. MacNeil, Biomaterials, 2006, 27, 2806.
12 R. Haigh, N. Fullwood and S. Rimmer, Biomaterials, 2002, 23, 3509.
13 A similar procedure is described in: M. Ebara, M. Yamato, T.
Aoyagi, A. Kikuchi, Ki. Sakai and T. Okano, Biomacromolecules,
2004, 5, 505.
Whilst the foregoing clearly show that these materials can
provide better substrates (in terms of cell viability) for serum-
free culture than TCP, the data do not provide direct evidence
that the effects are due to GRGDS binding to cell surface
integrins. In order to test this hypothesis GRGDS was added
to compete with the polymer-bound peptide.13 Fig. 4 shows
phase-contrast micrographs of HDFs cultured on P(GMA-co-
EDMA) functionalised with deprotected 2, before and after
addition of soluble GRGDS. The results, show the detach-
ment of cells strongly suggesting that cells adhere to these
materials by binding the polymer-bound peptide to cell surface
integrins. In control experiments the cells cultured on tissue
ꢁc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 5951–5953 | 5953