Light-activated hydrogel formation via the triggered folding and self-assembly of a designed peptide

Article abstract of DOI:10.1021/ja054719o

Photopolymerization can be used to construct materials with precise temporal and spatial resolution. Applications such as tissue engineering, drug delivery, the fabrication of microfluidic devices and the preparation of high-density cell arrays employ hyd

Full text of DOI:10.1021/ja054719o

Published on Web 11/10/2005
Light-Activated Hydrogel Formation via the Triggered Folding
and Self-Assembly of a Designed Peptide
Lisa A. Haines,† Karthikan Rajagopal,† Bulent Ozbas,‡ Daphne A. Salick,†
Darrin J. Pochan,*,‡ and Joel P. Schneider*,†
Contribution from the Department of Chemistry and Biochemistry and Materials Science and
Engineering, Delaware Biotechnology Institute, UniVersity of Delaware,
Newark, Delaware 19716-2522
Received July 14, 2005; E-mail: schneijp@udel.edu
Abstract: Photopolymerization can be used to construct materials with precise temporal and spatial
resolution. Applications such as tissue engineering, drug delivery, the fabrication of microfluidic devices
and the preparation of high-density cell arrays employ hydrogel materials that are often prepared by this
technique. Current photopolymerization strategies used to prepare hydrogels employ photoinitiators, many
of which are cytotoxic and require large macromolecular precursors that need to be functionalized with
moieties capable of undergoing radical cross-linking reactions. We have developed a simple light-activated
hydrogelation system that employs a designed peptide whose ability to self-assemble into hydrogel material
is dependent on its intramolecular folded conformational state. An iterative design strategy afforded
MAX7CNB, a photocaged peptide that, when dissolved in aqueous medium, remains unfolded and unable
to self-assemble; a 2 wt % solution of freely soluble unfolded peptide is stable to ambient light and has the
viscosity of water. Irradiation of the solution (260 < λ < 360 nm) releases the photocage and triggers
peptide folding to produce amphiphilic â-hairpins that self-assemble into viscoelastic hydrogel material.
Circular dichroic (CD) spectroscopy supports this folding and self-assembly mechanism, and oscillatory
rheology shows that the resulting hydrogel is mechanically rigid (G′ ) 1000 Pa). Laser scanning confocal
microscopy imaging of NIH 3T3 fibroblasts seeded onto the gel indicates that the gel surface is noncytotoxic,
conducive to cell adhesion, and allows cell migration. Lastly, thymidine incorporation assays show that
cells seeded onto decaged hydrogel proliferate at a rate equivalent to cells seeded onto a tissue culture-
treated polystyrene control surface.
Photopolymerization is extensively used in the fabrication
of a diverse array of materials that include industrial membranes
and coatings,1 dental adhesives,2 and optical and electronic
materials.1 The use of light to initiate polymerization is now
finding use in the construction of hydrogel materials, dilute
polymer networks capable of encapsulating a large volume of
water.3,4 Light-derived hydrogels are useful materials having
broad biomedical applications that include drug delivery,5-8
wound healing9,10 tissue engineering11-14 and construction of
high-density cell arrays.15-17 In addition, hydrogels are exten-
sively used in the fabrication of contact lenses18,19 and microf-
luidic devices serving as environmentally sensitive channel
dams.20-22 Irrespective of the final application, photopolymer-
ization allows hydrogel material to be formed with both temporal
and spatial resolution, whether in a targeted body cavity or the
strict confines of a microfluidic channel.
Typically, photopolymerized hydrogels are prepared via
radical chemistry in which a solution of macromolecular
precursor, a preformed water-soluble polymer containing reac-
† Department of Chemistry and Biochemistry.
‡ Materials Science and Engineering, Delaware Biotechnology Institute.
(1) Hughes, L. J.; Hughes, L. A. Handbook of polymer coatings for electronics,
2nd ed.; Noyes Publications: Park Ridge, NJ, 1990; pp 178-192.
(2) Moszner, N.; Salz, U. Prog. Polym. Sci. 2001, 26, 535-576.
(3) Fisher, J. P.; Dean, D.; Engel, P. S.; Mikos, A. G. Annu. ReV. Mater. Res.
2001, 31, 171-181.
(12) Elisseeff, J.; McIntosh, W.; Anseth, K.; Riley, S.; Ragan, P.; Langer, R. J.
Biomed. Mater. Res. 2000, 51, 164-171.
(13) Burdick, J. A.; Chung, C.; Jia, X. Q.; Randolph, M. A.; Langer, R.
Biomacromolecules 2005, 6, 386-391.
(14) Zhang, Z. Y.; Shum, P.; Yates, M.; Messersmith, P. B.; Thompson, D. H.
Bioconjugate Chem. 2002, 13, 640-646.
(4) Nguyen, K. T.; West, J. L. Biomaterials 2002, 23, 4307-4314.
(5) Hill-West, J. L.; Dunn, R. C.; Hubbell, J. A. J. Surl. Res. 1995, 59, 759-
763.
(6) An, Y. J.; Hubbell, J. A. J. Controlled Release 2000, 64, 205-215.
(7) Elisseeff, J.; Anseth, K.; Sims, D.; McIntosh, W.; Randolph, M.; Langer,
R. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 3104-3107.
(8) Burkoth, A. K.; Burdick, J.; Anseth, K. S. J. Biomed. Mater. Res. 2000,
51, 352-359.
(9) West, J. L.; Hubbell, J. A. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 13188-
13193.
(15) Revzin, A.; Sekine, K.; Sin, A.; Tompkins, R. G.; Toner, M. Lab-on-a-
Chip 2005, 5, 30-37.
(16) Revzin, A.; Tompkins, R. G.; Toner, M. Langmuir 2003, 19, 9855-9862.
(17) Suh, K. Y.; Seong, J.; Khademhosseini, A.; Laibinis, P. E.; Langer, R.
Biomaterials 2004, 25, 557-563.
(18) Wichterle, O.; Lim, D. Nature 1960, 185, 117-118.
(19) McMahon, T. T.; Zadnik, K. Cornea 2000, 19, 730-740.
(20) Eddington, D. T.; Beebe, D. J. AdV. Drug DeliVery ReV. 2004, 56, 199-
210.
(10) Hill-West, J. L.; Chowdhury, S. M.; Sawhney, A. S.; Pathak, C. P.; Dunn,
R. C.; Hubbell, J. A. Obstet. Gynecol. (N.Y.) 1994, 83, 59-64.
(11) Bryant, S. J.; Nuttelman, C. R.; Anseth, K. S. J. Biomater. Sci.sPolym.
Ed. 2000, 11, 439-457.
(21) Beebe, D. J.; Moore, J. S.; Bauer, J. M.; Yu, Q.; Liu, R. H.; Devadoss, C.;
Jo, B. H. Nature 2000, 404, 588-590.
(22) Burdick, J. A.; Khademhosseini, A.; Langer, R. Langmuir 2004, 20, 5153-
5156.
9
10.1021/ja054719o CCC: $30.25 © 2005 American Chemical Society
J. AM. CHEM. SOC. 2005, 127, 17025-17029
17025

A R T I C L E S
Haines et al.
Figure 1. (a) Light-induced material formation. UV illumination of MAX7CNB results in side-chain decaging and â-hairpin intramolecular folding. Subsequent
facial and lateral self-assembly ultimately affords hydrogel material. (b) Sequences of carboxyamidated peptides. R-Carboxy-2-nitrobenzyl protection of the
cysteine in MAX7 affords MAX7CNB.
tive groups that can be cross-linked, and a photoinitiator are
exposed to light. Radical generation and subsequent polymer
cross-linking produces hydrogel material. With respect to
biomedical applications, using macromolecular precursors to
form the gel is favored over the direct polymerization of small
molecule monomers to circumvent toxicity issues inherent to
many commercially available monomers.4 In microfluidic
devices, macromolecular precursors are used to better control
the mechanical properties of the final hydrogel since the extent
and type of cross-linking can be programmed at the level of
precursor before material formation is initiated. Precursors are
normally prepared by derivatizing an existing polymer with
functional groups capable of undergoing radical cross-linking
reactions. For example, acrylated poly(ethylene glycol)s16 and
poly(vinyl alcohol)s23 as well as cinnamated hyaluronic acids24
have been used. Not only is the design of the precursor important
for the success of a particular application but also the choice of
photoinitiator can be crucial. In vivo applications demand that
the initiator be cytocompatible; however, many common initia-
tors are cytotoxic at low concentrations or display cytotoxic
behavior that is highly dependent on concentration.11,25 In
addition, most initiators are only sparingly soluble in water and
must be dissolved in an organic solvent prior to use.11 However,
carefully assessing initiator cytoxicity as a function of concen-
tration, chemical additives, and light intensity can lead to useful
systems. For example, Anseth and co-workers11 have reported
that, although sparingly soluble in water, 2-hydroxy-1-[4-
(hydroxyethoxy)phenyl]-2-methyl-1-propanone and camphor-
quinone (UV and visible light initiators, respectively) show
efficient reactivity yet are cytocompatible under appropriate
environmental conditions. Although elegant work is ongoing,
there exists a large opportunity for alternate light-activated
strategies that produce cytocompatible hydrogels with controlled
mechanical and morphological properties.
or photoinitiators. This new system employs light to initiate the
self-assembly of water-soluble peptides into hydrogel material.
Specifically, light is used to trigger the folding of a small peptide
into an amphiphilic â-hairpin conformation, a structure highly
amenable to efficient self-assembly into mechanically rigid
hydrogel (Figure 1). Central to this hydrogelation strategy is
the design of MAX7CNB, a 20-residue photocaged peptide that
remains unfolded in aqueous solution under ambient light.
However, exposure to UV radiation results in an uncaged
peptide that intramolecularly folds. Folded hairpins are designed
to undergo both facial self-assembly, via the association of their
hydrophobic valine-rich faces and lateral assembly, via the
formation of intermolecular hydrogen bonds between neighbor-
ing hairpins (both intra- and intermolecular H-bonds are shown
in black in Figure 1). Together, these self-assembly events lead
to a hydrogel having a nanostructure composed of a network
of bilayer hairpin fibrils where noncovalent physical interfibril
cross-links are formed by the hydrophobic collapse of orthogonal
hairpins (Figure 1a). Extensive transmission electron microscopy
and small-angle neutron scattering studies of hydrogel networks
formed from similar â-hairpins that fold and self-assemble into
hydrogel when triggered by changes in pH,26 ionic strength,27
and temperature28 support this proposed model of nanostructure.
Results and Discussion
MAX7CNB was conceived through an iterative design
process that began with an earlier design, MAX1. This 20-
residue peptide is composed of high â-sheet propensity
valine and lysine residues29-31 flanking a central tetrapeptide
(-VDPPT-) designed to adopt type II′ turn structure (Figure
1b).32-34 The ability of MAX1 to assemble is dependent on its
(26) Schneider, J. P.; Pochan, D. J.; Ozbas, B.; Rajagopal, K.; Pakstis, L.;
Kretsinger, J. J. Am. Chem. Soc. 2002, 124, 15030-15037.
(27) Ozbas, B.; Kretsinger, J.; Rajagopal, K.; Schneider, J. P.; Pochan, D. J.
Macromolecules 2004, 37, 7331-7337.
(28) Pochan, D. J.; Schneider, J. P.; Kretsinger, J.; Ozbas, B.; Rajagopal, K.;
Haines, L. J. Am. Chem. Soc. 2003, 125, 11802-11803.
We sought to develop a simple light-activated hydrogel
system that did not involve the use of macromolecular precursors
(23) Martens, P.; Anseth, K. S. Polymer 2000, 41, 7715-7722.
(24) Miyamoto, K.; Sasaki, M.; Minamisawa, Y.; Kurahashi, Y.; Kano, H.;
Ishikawa, S. J. Biomed. Mater. Res. 2004, 70A, 550-559.
(25) Hanks, C. T.; Strawn, S. E.; Wataha, J. C.; Craig, R. G. J. Dent. Res. 1991,
70, 1450-1455.
(29) Brack, A.; Orgel, L. E. Nature 1975, 256, 383-387.
(30) Kim, C. A.; Berg, J. M. Nature 1993, 362, 267-70.
(31) Minor, D. L., Jr.; Kim, P. S. Nature 1994, 367, 660-3.
(32) Nair, C. M.; Vijayan, M.; Venkatachalapathi, Y. V.; Balaram, P. J. Chem.
Soc., Chem. Commun. 1979, 1183-1184.
9
17026 J. AM. CHEM. SOC. VOL. 127, NO. 48, 2005

Light-Activated Hydrogel Formation
A R T I C L E S
random coil signature, MAX6 is unfolded, indicating that the
inclusion of negative charge on the hydrophobic face success-
fully inhibits folding.
On the basis of this result, incorporation of a residue modified
with a charged, photosensitive group on its side chain at position
16 should prohibit folding, self-assembly, and ultimate material
formation. An additional design requirement is that, after
decaging, the resultant side chain must allow folding and self-
assembly to occur. Cysteine residues are attractive for modifica-
tion since their thiol side chains are not only relatively
hydrophobic but also sufficiently nucleophilic and therefore
easily functionalized. Although the unperturbed pKa of cysteine
is about 8.5 (resulting in a significant population of deprotonated
side chains at pH 9, which could actually hinder folding), we
envisioned that this residue’s pKa would be shifted to higher
values since its side chain resides in an extremely hydrophobic
environment in the folded and self-assembled state. In addition,
intermolecular disulfide bond formation after decaging is also
a possibility that could enhance self-assembly. MAX7 was
prepared to determine whether including a cysteine at position
16 would still allow the peptide to fold and self-assemble. The
CD spectrum of MAX7 (Figure 2), which is nearly identical to
that of MAX1 (not shown for clarity), shows a clear minimum
at 215 nm, consistent with â-sheet structure, demonstrating that
this peptide readily folds and self-assembles into â-sheet-rich
structures under the solution conditions where MAX6 remains
unfolded. At low peptide concentrations (micromolar) such as
those used for the CD studies, there is not enough peptide to
gel the water present in the aqueous buffer. However, peptide
folding and self-assembly still occurs, affording aggregates that
are not large enough to appreciably scatter light, making CD
measurements possible. Importantly, these data demonstrates
that cysteine’s side chain is conducive to folding and self-
assembly.
Figure 2. CD spectra of 150 µM solutions of MAX6 and MAX7 at pH
9.0 (125 mM borate and 10 mM NaCl, 37 °C).
unimolecular folded state. Under acidic conditions, the lysine
side chains are protonated, which prevents the peptide from
folding and, thus, self-assembling. Unfolded MAX1 is highly
soluble, and 1-2 wt % solutions display rheological properties
similar to those of water. Intramolecular folding and consequent
self-assembly can be triggered by either screening some of the
lysine-based charge with the addition of salt at physiological
pH27 or by neutralizing the charge by increasing the pH of low
ionic strength solutions to 9.26
Folded MAX1 hairpins are amphiphilic, where one face of
the hairpin is lined with hydrophobic valine residues and the
other face is lined with hydrophilic lysine residues, similar to
the folded hairpin shown in Figure 1a. In addition to electrostatic
interactions, previous studies indicated that the formation of both
intra- and intermolecular hydrophobic interactions between
valine residues is a key factor in promoting folding and self-
assembly, respectively.28,35 We envisioned that incorporating
an amino acid residue bearing a charged side chain on the
hydrophobic face would disrupt these hydrophobic interactions,
thereby inhibiting the folding and self-assembly of the peptide.
If so, a triggering mechanism could then be devised where this
unwanted charge could be removed by UV illumination,
allowing the peptide to fold and assemble into a hydrogel on
cue.
The final design element entails the choice of photocage.
Although a plethora of potential photoactive groups exists, the
R-carboxy-2-nitrobenzyl cage (Figure 1b) was employed be-
cause it bears a negative charge at neutral to basic pH, has a
respectable quantum yield for decaging,36,37 and forms a
noncytotoxic byproduct upon photolysis.38 The final peptide
MAX7CNB was prepared by reacting MAX7 with 2-bromo-
2-(2-nitrophenyl)acetic acid via a modified protocol originally
reported by Bayley and co-workers.39 Figure 3 shows CD spectra
of a 150 µM solution of MAX7CNB under basic conditions
after 1 and 13 h. The random coil signature remains nearly
invariant, demonstrating that the cage inhibits peptide folding
even after extended periods of time at 37 °C. Finally, exposure
of the same solution (contained within a 1 mm path length quartz
cell) to UV light (λ > 300 nm) for 30 min from a hand-held
lamp results in decaging and peptide folding/self-assembly as
evident by the â-sheet signature of the CD spectrum (Figure
3).
MAX6 was initially prepared to test whether a charged group
incorporated at the hydrophobic face would prevent folding.
MAX6 is identical in sequence to MAX1 except for the
alteration of valine at position 16 to glutamic acid, a carboxylate-
bearing residue (Figure 1b). In the folded state, the glutamic
acid side chain would be displayed in the center of the
hydrophobic face. Figure 2 shows the CD spectrum of MAX6
under solution conditions at which MAX1 readily folds and self-
assembles into hydrogel (pH 9, 37 °C).28 As evident by the
(36) Pan, P.; Bayley, H. FEBS Lett. 1997, 405, 81-85.
(37) McCray, J. A.; Trentham, D. R. Annu. ReV. Biophys. Biophys. Chem. 1989,
18, 239-270.
(33) Bean, J. W.; Kopple, K. D.; Peishoff, C. E. J. Am. Chem. Soc. 1992, 114,
(38) Walker, J. W.; Gilbert, S. H.; Drummond, R. M.; Yamada, M.; Sreekumar,
R.; Carraway, R. E.; Ikebe, M.; Fay, F. S. Proc. Natl. Acad. Sci. U.S.A.
1998, 95, 1568-1573.
5328-5334.
(34) Stanger, H. E.; Gellman, S. H. J. Am. Chem. Soc. 1998, 120, 4236-4237.
(35) Rajagopal, K.; Ozbas, B.; Pochan, D. J.; Schneider, J. P. Eur. Biophys. J.,
in press.
(39) Chang, C. Y.; Niblack, B.; Walker, B.; Bayley, H. Chem. Biol. 1995, 2,
391-400.
9
J. AM. CHEM. SOC. VOL. 127, NO. 48, 2005 17027

A R T I C L E S
Haines et al.
Figure 3. CD spectra of 150 µM solution of MAX7CNB (125 mM borate
and 10 mM NaCl, pH 9.0, 37 °C) as a function of time. Spectra of the
same solution after 1 and 13 h and after decaging by UV illumination are
shown.
Figure 4. Oscillatory rheology frequency sweep (1% strain) of 2 wt %
MAX7 and MAX7CNB (after UV illumination) hydrogels. (Inset) Self-
supporting MAX7 (left) and decaged MAX7CNB (right) gels.
the decaged sample is due to the benign photocage byproduct.
The combination of the CD results in Figure 3 and the rheology
data in Figure 4 indicates that peptide folding and consequent
self-assembly does not occur until UV radiation activates
decaging.
Quantitative decaging was verified by reverse-phase (RP) HPLC
and mass spectroscopic analysis of the CD sample (data not
shown).
At higher peptide concentrations (2 wt %, ∼6 mM), where
MAX1 is known to form a rigid hydrogel, an aqueous solution
of MAX7CNB is freely flowing, displaying the viscosity of
water at basic pH. However, exposure of this solution to UV
for 1 h results in the formation of a rigid hydrogel. In our
experiments, MAX7CNB solutions are prepared in simple glass
vials, resulting in rather large solution column heights (∼0.3
cm); as a result, efficient decaging takes about an hour.
Photoinduced hydrogelation within a microfluidic device should
be much more expedient since column heights on the order of
micrometers are normally encountered.21 Figure 4 compares
frequency sweep oscillatory rheology data obtained from 2 wt
% preparations of MAX7 (having never been caged) and
MAX7CNB after decaging. This experiment measures the
storage modulus (G′), a measure of a material’s elastic response
(rigidity) to varying frequencies of applied oscillatory stress.
The insensitivity of G′ to frequency for both MAX7 and decaged
MAX7CNB demonstrates that these gels are heavily cross-
linked.26 The cross-links are permanent in nature yet are formed
noncovalently via hydrophobic association of â-hairpins in the
self-assembled state. More importantly, the magnitude of the
observed storage modulus for the decaged hydrogel is ap-
proximately 1000 Pa, making it a significantly rigid hydrogel
material. The storage modulus of decaged MAX7CNB is slightly
lower than that observed for MAX7, probably due to an effect
of the 2-nitrosoglyoxylic byproduct formed during the decaging
event. The loss modulus (G′′), a measure of a material’s viscous
response, was also measured as a function of varying frequencies
of applied oscillatory stress. G′′ values were more than an order
of magnitude lower than the corresponding G′ values for each
gel, again indicating that the gels were quite rigid. The inset in
Figure 4 shows the self-supporting nature of 2 wt % MAX7
and decaged MAX7CNB hydrogels. The yellow coloration in
The oxidation state of the cysteine residue of decaged
MAX7CNB in the hydrogel was difficult to directly measure
since the 2-nitrosoglyoxylic byproduct formed during the
decaging event absorbs in the same region as many colorimetric
probes used to measure free thiol content. However, a hydrogel
formed from uncaged MAX7 that had been subjected to
decaging conditions (e.g., the clear hydrogel pictured in the inset
of Figure 4) was analyzed by a 2,2′-dithiodipyridine assay.40
The extent of oxidation was determined independently for three
samples, and in each case approximately 60% of the hairpins
were oxidized with very little variation among the samples. An
extensive polymeric network of disulfide bonds is not possible
since the peptide contains only one cysteine. In fact, the largest
possible disulfide-bonded aggregate is a dimer. Matrix-assisted
laser desorption ionization time-of-flight (MALDI-TOF) mass
spectrometry of MAX7 hydrogel shows exclusively monomer
and disulfided dimer (see Supporting Information). It is currently
unknown whether the oxidation state of MAX7 influences the
rheological properties of the hydrogel since, in our hands, it is
nearly impossible to keep the peptide reduced under the
conditions of the experiment.
Finally, future applications such as tissue engineering or the
fabrication of high-density cell arrays demand that the hydrogel
surface be cytocompatible. The cytocompatibility of the decaged
MAX7CNB hydrogel surface was assessed by employing a live/
dead assay where NIH 3T3 fibroblasts were seeded onto the
gel after light activation and equilibration with Dulbecco’s
modified Eagle cell culture medium (pH 7.4) containing serum.
Cells were incubated on the surface for 24 h, enough time to
allow the cells to attach to the surface and adopt healthy spread-
(40) Pedersen, A. O.; Jacobsen, J. Eur. J. Biochem. 1980, 106, 291-295.
9
17028 J. AM. CHEM. SOC. VOL. 127, NO. 48, 2005

Light-Activated Hydrogel Formation
A R T I C L E S
cells. Also evident in the image are cells that contain ruffled
edges; fibroblasts undergoing migration ruffle their membranes
as they pull themselves across a given substratum.41 Together,
the microscopic data demonstrate that the gel surface is
noncytotoxic and is conducive to cell adhesion and allows
migration. Last, cell proliferation was monitored over 72 h by
measuring [3H]thymidine uptake as a function of time (Figure
5d). In this experiment, fibroblasts (20 000 cells/well) were
seeded onto 2 wt % MAX7CNB hydrogels (after exposure to
light and equilibration with cell culture medium) and onto
control tissue culture-treated polystyrene surfaces. The data show
that cells proliferate on the hydrogel surface at nearly the same
rate as the control surface and that the photocage byproduct,
2-nitrosoglyoxylic acid, minimally influences proliferation.
For the cell-based experiments described here, hydrogelation
was initiated at pH 9 and cell culture medium was added to the
resulting gels, effectively adjusting the pH to 7.4. Although this
methodology is conducive to the fabrication of high-density cell
arrays, many tissue engineering applications necessitate that cells
be incorporated within the three-dimensional network of the
hydrogel with serum components present. For these applications,
a caged peptide capable of undergoing light-induced folding
and self-assembly at pH 7.4 is needed; we are currently
developing such a peptide.
Figure 5. (a) Live/dead assay of NIH3T3 fibroblasts seeded onto decaged
MAX7CNB hydrogel after 24 h employing LSCM. Live cells fluoresce
green and dead cells fluoresce red. (b) Identical assay performed on cells
seeded onto a sterile borosilicate control surface. (c) LSCM image of cells
at 8 h showing the formation of actin stress fibers (stained with a green
fluorescent dye) defining the lamellipodia (1) and filopodia (2) of cells as
indicated. The ruffled membrane of a migrating cell is also observed (3).
Cell nuclei are stained with a blue fluorescent dye for reference. (d) Rates
of cell proliferation measured by [3H]thymidine uptake of cells seeded onto
2 wt % MAX7CNB-derived gel compared to a tissue culture-treated
polystyrene control surface.
Conclusions
Our approach to hydrogel formation entails the triggered
folding and consequent self-assembly of â-hairpin peptides. We
had earlier demonstrated that peptides can be designed to
undergo triggered hydrogelation in response to changes in
solution pH,26 ionic strength,27 or temperature.28 Herein, we
show that light can be used as a trigger to initiate peptide folding
and consequent self-assembly. We achieve this by covalently
incorporating a cytocompatible photocleavable moiety that
electrostatically inhibits peptide folding and, therefore, the
intermolecular self-assembly process that leads to hydrogel
formation. However, when irradiated, the photocage is released
to initiate folding, self-assembly, and material formation to
produce a mechanically rigid, noncytotoxic hydrogel. The utility
of this novel photoinitiated hydrogelation mechanism is currently
being investigated in our lab for use in applications that demand
spatially and temporally resolved material formation.
out morphologies via the formation of actin stress fibers. Figure
5a shows a laser scanning confocal microscopy (LSCM) image
of cells maintained on 2 wt % decaged gel after fluorescent
dyes have been added that specifically dye living cells (green)
and dead cells (red). Also, shown for comparison are cells that
have been seeded onto a sterile borosilicate control surface under
identical conditions (Figure 5b). These reproducible images
demonstrate that the gel surface is noncytotoxic. Figure 5c shows
cells that have been stained with both a DNA-specific fluores-
cent dye (4′,6-diamidino-2-phenylindole, blue) and a fluorescent
dye specific for F-actin (Alexa fluor 488 phalloidin, green). Cells
undergoing surface adhesion events form stress fibers, lamel-
lipodia and filopodia, all of which contain F-actin. These
structures help define the cellular morphology and aid locomo-
tion.41 The formation of lamellipodia (flattened projections from
cells) and filopodia (thin extensions from cells) are clearly seen
in Figure 5c; the inset shows a magnified fibroblast where thin
filopodia are more clearly observed. Actin stress fibers (not
labeled for clarity) can be seen crossing over the nuclei of the
Acknowledgment. We thank the NIH (R01 DE016386-01)
for support. We also thank Ann Snellinger for collecting mass
spectral data and Professor John Koh for discussion concerning
cell culturing.
Supporting Information Available: Experimentals, analytical
HPLC and ESI and MALDI-TOF mass spectra of peptides.
This material is available free of charge via the Internet at
http://pubs.acs.org.
(41) Mitchison, T. J.; Cramer, L. P. Cell 1996, 84, 371-379.
JA054719O
9
J. AM. CHEM. SOC. VOL. 127, NO. 48, 2005 17029