Feature Article

Light-Responsive Biomaterials: Development

and Applications
Joshua S. Katz, Jason A. Burdick*

Novel biomaterials are beneficial to the growing fields of drug delivery, cell biology, microdevices, and tissue engineering. With recent advances in chemistry and materials science,
light is becoming an attractive option as a method to control biomaterial behavior and
properties. In this Feature Article, we explore some of the early and recent advances in the
design of light-responsive biomaterials. Particular
attention is paid to macromolecular assemblies for
drug delivery, multi-component surface patterning
for advanced cell assays, and polymer networks that
undergo chemical or shape changes upon light
exposure. We conclude with some remarks about
future directions of the field.

Introduction
The past couple of decades have seen a rapid evolution in
the development of novel materials for applications in drug
delivery, tissue engineering, microdevices, and cellular
biology.[1] Early efforts in these fields focused on materials
that were supportive of the desired applications, though
not necessarily direct participants. Following these initial
efforts, the focus turned to the design of functional
materials whose properties were more important in
achieving successful outcomes, such as degradation,
functional group presentation, or mechanical robustness.
As the development of these materials has advanced, there
has been further interest in control over material properties
in time and space. One approach to achieve this precise
control is through the employment of exogenous agents
that can act on the material, such as enzymes, acoustic
energy, magnetic fields, or light.[2]
Light is a particularly attractive source of energy for use
in controlling biomaterial behavior. Its intensity and
J. S. Katz, J. A. Burdick
University of Pennsylvania, Department of Bioengineering, 210
South 33rd Street, 240 Skirkanich Hall, Philadelphia, PA 19104, USA
Fax: 1 215 573 2071; E-mail: burdick2@seas.upenn.edu
Macromol. Biosci. 2010, 10, 339348
2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

wavelength can easily be controlled through the use of

filters, and photomasks or lasers allow for fabrication of
complex features and exposure areas with resolution as
small as approximately 1 mm. The advent of confocal
microscopy has enabled a further increase in resolution, in
addition to three-dimensional control.[3,4] The majority of
light-responsive chemical moieties are responsive in the UV
spectral range, which is generally not limiting in an in vitro
environment, and may be sufficient for many applications.
However, systems that would be desirable for translation to
an in vivo setting are limited by the narrow spectral range
in the near-infrared (NIR) region in which light has
sufficient penetration through tissue for utility.[5] However,
recent advances in two-photon microscopy and lasers and
the development of NIR-active molecules suggest that this
drawback can be overcome and has allowed for the
development of a range of useful materials.
Light-responsive polymeric materials can be designed
through the incorporation of specific chemical moieties into
the polymer network or through composite systems in
which a light-responsive group is suspended within the
network, though not covalently attached. In this article we
explore some of the recent advances in the control of
biomaterial behavior using light. Specifically, we focus
on the incorporation of light-responsive moieties into

DOI: 10.1002/mabi.200900297

339

J. S. Katz, J. A. Burdick

macromolecular assemblies for drug delivery, the development of bio-friendly methods for light-controlled patterning of two-dimensional cellular substrates and threedimensional gels, and the harnessing of light to induce
shape changes in biomaterials.

Macromolecular Assemblies
Macromolecular assemblies are formed by the self-assembly of polymers in aqueous solutions, driven by noncovalent interactions between the macromolecules.[6,7] The
classical example of a macromolecular assembly found in
nature is the phospholipid vesicle (Figure 1a), where a
membrane is formed through hydrophobic interactions
between the hydrocarbon tails of phospholipids and is
stabilized in solution by the polar head groups. The ratio of
hydrophobic to hydrophilic segments within a macromolecule determines its self-assembly behavior; molecules
that are largely hydrophobic tend to aggregate as
membranes (e.g., vesicles), while those that are more
hydrophilic form micelles (Figure 1b,c).[8] Micelles and
vesicles have shown significant promise for use as drug
delivery vehicles in vivo, and have been synthesized from a
wide variety of molecules ranging from lipids to di-and triblock copolymers.[9,10] Controlling the release of encapsulated agents by an external stimulus is of great interest for
the drug delivery field and has been extensively reviewed
for many different stimuli.[1113] The synthesis of micelles
and vesicles that release their content upon light exposure
could allow for improved site-specific delivery of cargo to
diseased tissues such as tumors, through selective irradiation of the target site. Many drugs and drug candidates are
currently limited by a narrow therapeutic index[14] and sitespecific delivery could enable larger dose release at a target
site while reducing systemic doses that lead to debilitating
side-effects.
Light-induced release of encapsulated contents within
macromolecular assemblies has been explored through
both chemical and physical changes to the assembly. An
early study explored phospholipid vesicles (liposomes) that
were sensitive to photo-oxidation induced by a sensitizer.[15] Phosphocholine was modified to include a vinyl
ether linkage between the polar head group and the
palmitoyl tail. Liposomes were formulated to encapsulate
glucose (for release) and zinc phthalocyanine, a photosensitizer. Exposure of these liposomes to >640 nm light
enhanced glucose release over unexposed controls. However, the vesicles were quite leaky, releasing nearly 30% of
their content within 1 h in the dark (compared to roughly
double that under light exposure), likely rendering these
vesicles unsuitable for clinical use.
A more recent example incorporated a photo-isomerizable azo group into the tails of a phosphocholine-based

340

Macromol. Biosci. 2010, 10, 339348

2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Joshua S. Katz is a Ph. D. candidate in the

Department of Bioengineering at the University
of Pennsylvania. He received his S. B. in Chemistry with a minor in Biomedical Engineering
from the Massachusetts Institute of Technology
in 2006. At MIT, he worked with Dr. Darrell Irvine
on the development of responsive materials for
immune system engineering. Currently, he
works in the labs of Dr. Jason Burdick and Dr.
Daniel Hammer at Penn. His research interests
are in the synthesis and design of smart polymeric biomaterials systems for tissue engineering and drug delivery. In 2006, he was awarded a
Graduate Research Fellowship from the National
Science Foundation.
Jason A. Burdick, Ph. D. is the Wilf Family Term
Assistant Professor of Bioengineering at the University of Pennsylvania in Philadelphia, PA, USA.
Jason has his Ph. D. in Chemical Engineering
from the University of Colorado working with
Dr. Kristi Anseth and was a postdoc at MIT under
Dr. Robert Langer. His research involves the
development of photopolymerizable and
degradable biomaterials for various biological
applications and his laboratory is specifically
interested in understanding and controlling
polymers on a molecular level to control overall
macroscopic properties. The applications of his
research range from controlling stem cell differentiation through material cues to fabricating
scaffolding for regenerative medicine. Jason currently has over 70 peer-reviewed publications
and is on the editorial boards of the J. Biomedical
Materials Research A and Tissue Engineering. He
has been awarded a K22 Scholar Development
and Career Transition Award through the
National Institutes of Health, an Early Career
Award through the Coulter Foundation, a
National Science Foundation CAREER award,
and a Packard Fellowship in Science and Engineering.

lipid.[16] These liposomes were formulated to accommodate

the active loading of drugs (aciridine orange and doxorubicin), allowing for high drug encapsulation. When in the
trans conformation, the lipid bilayer was well packed and
relatively stable with little release observed. However, UVinduced isomerization from trans to the cis conformation of
the azo group disrupted the packing of the membrane and
induced release of the two encapsulated model drugs
(Figure 1d). This same chemistry was exploited for gene
delivery using the small cationic surfactant azobenzene
trimethylammonium bromide.[17] In this case, isomerization from trans to cis induced a decrease in the hydrophobicity of the molecule, causing vesicle rupture. These
vesicles can complex with DNA and were used as a

DOI: 10.1002/mabi.200900297

Light-Responsive Biomaterials: Development and Applications

required wavelength for cleavage varied

by the charge on the head group; the
glutamate- and aspartate-based lipids
cleaved at 365 nm and the basic lysine
cleaved at 254 nm, presumably due to the
amine group on the lysine interacting
with the photocleavable group.
The nitrobenzyl group has also been
incorporated into micelle-forming block
copolymers.[25,26] In these systems, the
group protects a side-chain carboxylic
acid in a hydrophobic state; the hydroFigure 1. Self-assembled macromolecular structures. (a) Lipid molecules (hydrophilic phobicity within the protected block
head and hydrophobic tail) self-assemble into vesicle structures. Diblock copolymers of
drives micelle formation. Cleavage of
hydrophilic and hydrophobic blocks assemble into vesicles with a larger hydrophobic
the protecting group with UV light
content [polymersomes, (b)] and micelles with a greater hydrophilic content (c). These
irreversibly alters the hydrophobicity of
structures can be designed to respond to light. For example, (d) UV-light induced
the molecule (now a free carboxylic acid)
isomerization of azo-containing lipids interferes with their packing ability, leading to
and drives the rearrangement of the
the leakage of any loaded contents and (e) irreversible cleavage of a hydrophobic
protecting group changes the hydrophobic/hydrophilic balance of the macromolecule,
micelle and release of the contents
leading to micelle breakdown. In all cartoons, black is hydrophobic; gray is hydrophilic. (Figure 1e). Several other photo-sensitive
protecting groups have also been utilized
in micelle and vesicle systems. Jinqiang
Jiang and coworkers designed micelles from poly(ethylene
transfection vehicle for an enhanced green fluorescent
glycol)-block-poly(pyrene methacrylate).[27] The bulky
protein (eGFP) plasmid to fibroblasts. While the vesicles
themselves were successful transfection agents, UV light
hydrophobic pyrene drives micellization, but could be
roughly doubled the efficiency and greatly enhanced
cleaved from the methacrylate with exposure to 365 nm
endosomal escape of the DNA, an important issue in gene
light, yielding a fully water-soluble poly(ethylene glycol)therapy approaches.
block-poly(methacrylic acid) polymer. Yugui Jiang and
Photo-isomerizable groups, including the azobenzene
coworkers designed reversible vesicles using a poly(ethygroup discussed above,[18] azopyridine,[19] stilbenes,[20] and
lene glycol) (PEG)-terminated malachite green derivative.[28] This malachite green derivative has a terminal
spiropyrans[21,22] have also been used to control the
assembly and disassembly of micelles and polymeric
cyano group that was cleaved to yield a positive charge
vesicles. In one particularly interesting study, Mabrouk
when exposed to UV light. This positive charge was
and coworkers formed polymersomes with an asymmetric
sufficient to disassemble the vesicles. Removal of the light
membrane, the inner leaflet consisting of poly(butadiene)
source led to reattachment of the cyano group and vesicle
and the outer being a liquid crystalline diazobenzene-based
reassembly.
methacrylate.[23] UV light exposure induced the liquid
Two groups have recently reported micelle systems that
are responsive to two-photon illumination.[29,30] As discrystal to transform from a nematic to isotropic state,
inducing a small, unstable pore penetrating the membrane.
cussed earlier, two-photon exposure has several advanThe pores induced by this change in state caused a
tages over simple UV irradiation, including improved
curling instability of the outer leaflet of the polymersome,
spatial resolution and significantly deeper tissue penetraleading to outward spiraling of the leaflet and, ultimately,
tion in vivo. Goodwin and coworkers capped a PEG molecule
bursting of the vesicle.
with a 2-diazo-1,2-naphthoquinone moiety.[30] Upon expoSeveral aromatic light-sensitive protecting groups have
sure to a 350 nm or 795 nm laser in the presence of water,
been exploited to allow irreversible chemical changes to
the naphthoquinone underwent a Wolff rearrangement,
drug delivery assemblies such as liposomes, micelles, and
yielding an indene carboxylate which is significantly more
vesicles. Chandra and coworkers designed lipids where a
hydrophilic. The rearrangement drove disassembly of the
hydrophilic amino acid (aspartate, glutamate or lysine) was
micelles and release of their contents. Babin and coworkers
coupled to a saturated hydrocarbon through a nitrobenzyl
designed a PEG-poly(methacrylic acid) polymer, where the
linker.[24] UV light cleaves the molecules, separating the
acid groups were protected by a coumarin derivative.[29]
amino acid from the hydrophobic tail. Incorporation of
Similar to the naphthoquinone, the coumarin is also
these lipids at only 5% into distearoylphosphatidylcholine
sensitive to one- (
350 nm) or two-photon (
795 nm)
(DSPC) liposomes was sufficient to release encapsulated
light, in this case, cleaving from the methacrylate, yielding
contents upon exposure to UV light. Interestingly, the
the free acid. As with other systems, this drastic change in
Macromol. Biosci. 2010, 10, 339348
2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.mbs-journal.de

341

J. S. Katz, J. A. Burdick

hydrophilicity drove the micelles to disassemble in solution

and release their contents. The 2-nitrobenzyl groups
susceptibility to two-photon cleavage has been
explored,[31] but to the best of our knowledge, assembly
disruption using this protecting group with NIR light has
yet to be reported for a self-assembled system.
In addition to chemical changes affecting release from
macromolecular assemblies, several light-sensitive particles have been encapsulated within vesicles that cause a
physical change upon light exposure, leading to vesicle
disruption. Paasonen and coworkers encapsulated gold
nanoparticles within liposomes.[32] The nanoparticles, 1
3 nm in diameter, were loaded into the aqueous core,
hydrophobic membrane or tethered to the membrane,
depending on the surface chemistry of the nanoparticles.
Exposure to UV light (250 nm) induced heating of the gold
nanoparticles, consequently leading to permeabilization of
the liposome membrane and release of contents. One major
drawback to this system, however, is the extremely low
wavelength required to heat the particle and cause release.
Two other similar systems are significantly more likely to
be translatable to the clinic, as they absorb in the NIR
regime. Wu and coworkers encapsulated hollow gold
nanoshells, rather than nanoparticles, into and onto
liposomes.[33] These gold nanoshells rapidly heat upon
exposure to 800 nm light, creating small bubbles of vapor
that cavitate the membrane. As a result, the membranes are
not destroyed but do transiently (for the duration of laser
illumination) allow for release of encapsulated contents.
Troutman and coworkers developed a method to coat
liposomes with small (<10 nm) clusters of gold.[34] The
clusters, in close proximity to each other, produced plasmon
resonances that exist at wavelengths of approximately
1 000 nm. Upon NIR excitation, the plasmon bands
produced in the clusters were converted to thermal energy
which allowed for liposome destabilization and release of
encapsulated contents.
One recently-reported composite system is of particular
note,[35] as it combines NIR-induced release with polymersomes, which are significantly more stable than liposomes,[36] and avoids the use of potentially-toxic metals.
Robbins and coworkers designed a three component system
for inducing vesicle release. Polymersomes were fabricated
from polybutadiene-block-poly(ethylene glycol); a protein
(ferritin, myoglobin or albumin) was loaded into the
aqueous core, while a NIR porphyrin dye was loaded into
the membrane.[37] Upon excitation of the porphyrin, the
vesicles were observed to bud and collapse, releasing
encapsulated molecules, such as biotin. Fluorescence
recovery after photobleaching studies suggested that some
of the protein was intercalating with the membrane,
perhaps aiding in the light-sensitivity.
While the technology is clearly being advanced for lightresponsive macromolecular assembly delivery systems,

342

Macromol. Biosci. 2010, 10, 339348

2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

there have yet to be any reports translating these systems to

an in vivo setting. Non-responsive systems have already
been shown to have utility for in vivo imaging and disease
treatment,[3739] and such studies exploring drug delivery
and disease treatment using light-sensitive materials
should be forthcoming in the near future.

Two-Dimensional Protein Patterning

In the interest of understanding in vivo processes, there is a
growing need for fundamental biological studies to better
understand cell behavior in vitro. Though many cellular
processes are governed by mm-scale patterns of proteins,[40]
including immunological and neural synapse activation,[41,42] integrin clustering,[43,44] motility,[45] and differentiation,[46] relatively few cellular studies actually take
place on surfaces that present ligands in a spatially-defined
manner. Further advances in patterning technologies will
help to facilitate more studies that better mimic the in vivo
cellular environment and will help to better understand
cells in native tissues and during disease, as well as our
ability to control cells in therapy development.
Photochemical (using light to effect a change to
molecules attached to a surface) and photolithographic
(using light to change the solubility of a molecule protecting
a surface) technologies offer many benefits over other
patterning methods. A single mask can be used to quickly
generate many complex surfaces with high fidelity and
resolution. However, because of the need for harsh solvents,
intense UV light, and dehydration, many classical photobased techniques are not compatible with biological
systems to generate surface patterns of ligands.[4749] Only
recently have several groups reported the development of
bio-friendly photolithographic and photochemical patterning methods for patterning multiple ligands on a
surface.
Early work in the development of light-patterned
biological surfaces centered on the use of caged biotins,
which, through streptavidin bridges, could be used to
spatially pattern any biotinylated molecule (i.e. proteins,
antibodies, DNA, etc.). Sundberg and coworkers tethered a
nitrobenzyl-caged biotin to surfaces that, upon UV
exposure, cleaved the nitrobenzyl group and presented
free biotin.[50] By employing successive exposures to
different regions of the slide, different fluorescent biotinylated antibodies could be coupled. In this case, they avoided
exposure of protein-bound regions by masking alreadyexposed regions. However, as a result, this severely limits
the possible resolution to several hundred of mm, areas too
large for patterns at the cellular level. A similar approach
was employed by Blawas and coworkers who, rather than
covalently attaching a biotin to a glass substrate, coupled
caged biotin to bovine serum albumin which could then

DOI: 10.1002/mabi.200900297

Light-Responsive Biomaterials: Development and Applications

adsorb onto glass.[51] UV light released the cage, exposing

free biotin that could bind streptavidin and subsequently
other biotinylated molecules. As with the first case, though,
the spatial resolution of this system for patterns of multiple
proteins is too low for patterning at the cellular level.
Holden and Cremer addressed the issue of spatial resolution
of patterning by changing the method used to make biotin
available to proteins.[52] Rather than using an uncaging
procedure that requires UV light and could be damaging to
proteins, they attached different fluorophores to different
biological molecules, an immunoglobulin G (IgG) and a
biotin. Exposure of the fluorophores to their excitation
wavelength (in the visible spectrum) generated radicals
that could react with the surface, immobilizing attached
molecules. Because the excitation wavelengths are nondamaging to biological molecules, patterns, including
overlapping patterns, of different molecules could easily
be generated at high resolution.
The above examples are all instances of photochemical
patterning, rather than the use of photolithography. Only
recently have photoresist systems been developed that can
be used to pattern multiple biological ligands in a spatiallydefined manner. Doh and Irvine developed a carboxylic
acid-generating photoresist that could be used to pattern

multiple proteins in complementary regions of a surface

(Figure 2).[53] Their photoresist, a random terpolymer of 2nitrobenzyl methacrylate, methyl methacrylate, and
hydroxyl-terminated PEG methacrylate, exhibited pH
dependent water solubility following UV exposure.
Furthermore, the exposed resist is also a polyelectrolyte,
allowing a small (10 nm) layer of resist to remain behind on
a cationic surface. Biotinylation of the resist prior to casting,
therefore, created a biotinylated surface even after removal
of the exposed photoresist. Consequently, a first pattern
could be etched, and the remainder of the resist exposed.
Before removal of the second portion of the resist, a first
protein could be bound to the surface of the resist under
mild acidity, where the resist is not water soluble but
proteins are still highly stable, through a streptavidinbiotin bridge. Following this initial binding, the resist still
present after the first exposure could be removed by
increasing the pH, also removing any protein that bound in
that region. A new layer of resist is then exposed to which a
second protein could be bound by the same chemistry.
Interestingly, it was found that another formulation of this
same terpolymer could act as a negative-tone photoresist,
where an originally water-soluble photoresist is transiently
stabilized through intermolecular hydrogen bonds following UV exposure.[54] In a method similar
to the one above, two proteins could
again be patterned in spatially defined
regions without exposing either protein
to UV light or dehydration. In a beautiful
example of the applicability of this
technology, the authors demonstrated
how surface patterns of anti-CD3 and
ICAM-1 could with proper spatial
organization stimulate the formation
of immunological synapses with CD4 T
cells, leading to T cell activation
(Figure 2b).[55]
Several other recent advances are
worth noting, though they have yet to
be expanded to multicomponent patterning.
Alvarez
and
coworkers
employed two-photon techniques to
pattern carboxylic acids on a quartz
surface that could then be functionalized
with biotin or other biological molecules.[56] While the use of two photon
microscopy slows the process for produFigure 2. Light-controlled multi-component protein patterning. (a) Bio-friendly photo- cing patterns rapidly, there are two
resist terpolymer that is sensitive to light. Selective light exposure of the photoresist advantages compared to simple lithoalters the film solubility, allowing the spatial patterning of proteins selectively with graphic techniques. First, the authors
intermittent light exposure. (b) The patterns can be used to investigate T cell response
demonstrated the ability to generate
to patterns of ligands (anti-CD3 surrounded by ICAM-1). T cell surface receptor PKC-u and
intracellular signaling molecule LFA-1 localize well with the patterned ligand. Scale bar: different densities of ligand on the surface based on the intensity or duration of
5 mm. Figure adapted from ref.[55] with permission. Copyright 2006 National Academy of
Sciences, USA.
exposure. Second, confocal techniques

Macromol. Biosci. 2010, 10, 339348

2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.mbs-journal.de

343

J. S. Katz, J. A. Burdick

allow for resolutions that are finer than

the limitation that is usually imparted by
the wavelength of the light, meaning
smaller features can be patterned on a
surface. Similarly, towards the transfer to
a third dimension, Hahn and coworkers
used confocal microscopy to pattern
arginine-glycine-aspartic acid (RGD)
ligands on PEG hydrogels.[57] The ability
to pattern on the surface of gels is
promising as it will enable the generation
of surfaces that can be spatially tailored
for chemical functionality and be
designed to have specific physical properties, which are known to affect cell
behavior.[58,59] Their specific system has
a few limitations in that it relies on
incomplete polymerization of the gel in
order to couple the ligand. Edahiro and
coworkers expanded patterning from
proteins to actual cells.[60] Combining a
spirobenzopyran, which reversibly forms
a zwitterionic structure upon non-toxic
Figure 3. Patterning of chemistry and structure in hydrogels with light. (a) PhotocheUV irradiation, with N-isopropylacrylamical patterning of a hydrogel, where UV light activates protected moieties covalently
mide, a thermo-sensitive polymer, cells
attached to a gel. These activated moieties can then, in a second step, bind to
could be selectively removed from a
biologically active molecules. (b) Photo-degradation of a hydrogel. A hydrogel containsurface by a combination of UV exposure
ing crosslinks that contain UV-cleavable sites can be broken down upon UV exposure.
With both of these techniques, patterning is possible by spatially controlling light
and low-temperature washing. Perhaps
exposure.
due to the nature of cell growth, though
the patterns produced were coarse and
not well-defined. However, the field as a
whole is certainly progressing to achieve higher levels of
nitrobenzyl- or coumarin-derived protecting group as a
complexity to further enhance our ability to understand
thiol-protected cysteine, which they coupled to agarose
and control cellular behavior.
gels. Irradiation of specific regions of the gel site-specifically
cleaved the nitrobenzyl group, yielding free sulfhydryls
that could react with thiol-reactive compounds. In areas
where the gels were patterned with the cell-responsive
Three-Dimensional Patterning of Hydrogels
glycine-arginine-glycine-aspartic acid-serine (GRGDS)
ligand, neurites seeded on the surface of the gel were seen
While two-dimensional surfaces have been shown to be
to grow into the patterned regions.[64] The original work
quite useful for fundamental cell studies, any attempt at
engineering full tissues requires three-dimensional scafreported was limited by control of the pattern in only two
folds that support cellular growth and behavior. Recently,
dimensions, as there is little control over the pattern
several groups have reported methods for scaling cell
throughout the depth, and only patterns such as columns
could be fabricated. To improve on this, Wosnick and
patterning into the third dimension. A significant amount
of this work has focused on further crosslinking or
Shoichet expanded this work to pattern the gels using
interpenetrating networks of hydrogels as a method to
confocal microscopy, enabling complete freedom in pattern
spatially alter local mechanics or inhibit the spreading of
choice in all dimensions, with reported feature sizes as
encapsulated cells. While this is beyond the scope of our
small as
5 mm.[66] For this work, however, they focus only
focus, we reference several notable reports for the
on the pattern fabrication itself and do not report any
interested reader.[6163] The rest of the research in this field
studies with cells.
Only recently was the first photo-degradable hydrogel
can be approximately broken down into two categories
system reported for tissue engineering.[67] In groundligand presentation and scaffold degradation. The field of
ligand presentation has been pioneered by Shoichet and her
breaking work, Kloxin and coworkers modified PEG
lab (Figure 3a).[6466] In their work, they utilized the 2diacrylate to include a 2-nitrobenzyl linker between the

344

Macromol. Biosci. 2010, 10, 339348

2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

DOI: 10.1002/mabi.200900297

Light-Responsive Biomaterials: Development and Applications

PEG and acrylate end groups. This diacrylate acts as a

crosslinker in a radically-polymerized gel system. Upon UV
exposure, using either a UV lamp or confocal microscope,
portions of the gel could be eroded away by rearrangement
of the nitrobenzyl groups and cleavage of the crosslink
(Figure 3b). Erosion of channels within gels containing
fibronectin created regions that were permissive to cellular
movement. As a second example of the utility of this
chemistry in gel systems, an RGDS peptide was coupled to
the same nitrobenzyl-acrylate moiety. Copolymerizing the
photo-cleavable peptide into PEG gels created gels in which
the presence of RGDS ligand could be tailored in a spatial
and temporal manner. It was found that human mesenchymal stem cells undergo chondrogenesis when transiently
exposed to RGDS for a 10 d period followed by 11 d of no
ligand, whereas no chondrogenesis was observed if the
ligand was or was not consistently present for the entire
culture period. In addition to the interesting findings
observed in this paper, this work further opens the door for
the design of new biomaterials for tissue engineering that
can be spatially and temporally tailored for specific and
varying mechanical properties, crosslink densities, and
ligand concentrations through combinations of photocleavable groups and corresponding non-cleavable molecules. This example is just the first step towards added
complexity in designing three-dimensional systems with
added cellular control.

Shape-Memory and Shape-Changing

Biomaterials
In addition to the importance of spatial control of cell and
ligand populations in the development of advanced tissue
engineering scaffolds, control of the spatial dimensions and
shape of a biomaterial could provide improved versatility
for biomedical implants. Shape-memory polymers are
materials that can be set in temporary shape until a
stimulus allows it to return to a more permanent and
desired conformation. Shape-changing polymers are materials that undergo a temporary change in shape upon
suitable stimulation but may revert back to their original
shape when the stimulus is removed.[68] In this section, we
explore several recent advances in shape-memory and
shape-changing polymers controlled by exposure to light.
Similar to our discussion of macromolecular assemblies
above, shape transitions initiated by light can be induced
through either chemical changes to a polymer network or
through composite systems in which an absorbing species
causes a physical change to the network.
Photoinduced transitions in gels were first reported
nearly twenty years ago for poly(N-isopropyl acrylamide)
(NIPAm) gels.[69] In the first of these systems, a UV-ionizable
moiety,
bis[4-(dimethylamino)phenyl](4-vinylphenyl)Macromol. Biosci. 2010, 10, 339348
2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

methyl leucocyanide, was introduced into the polymer

gel backbone. Upon irradiation, cyanide is released,
inducing an osmotic force and driving water into the gel,
leading to increased swelling. As the ionization is reversible,
upon removal of the light stimulus, the gel reverted back to
its original swelling state. There are, however, several
drawbacks to this system. It was difficult to control the rate
of swelling and deswelling, as ion recombination is
governed by diffusion of the cyanide ions. Further, for
biomedical applications, cyanide ions would not be ideal,
because of their toxicity. Some of these issues were
addressed in another paper that appeared shortly after
this original publication.[70] In this study, the ionizable
moiety was replaced by a copper-containing chromophore
which absorbs visible light. The chromophore, upon
irradiation, locally heated the gel, leading to a variable
ionization state of the NIPAm. The new ionization state in
turn led to localized deswelling of the gel. As with the prior
system, upon removal of the light stimulus, the gel reverted
back to its original state.
A major limitation of both systems described above is
that light can only be used to control swelling of a material.
However, often times, there may be an interest in changing
the entire shape, rather than only the dimensions. This issue
was addressed by Lendlein and coworkers in a seminal
paper.[71] In their work, they grafted cinnamic acid (CA) into
the backbone of hydroxyethyl acrylate hydrogels. When
exposed to ultraviolet radiation of l >260 nm, the CA
molecules underwent a [2 2] cycloaddition, adding new
crosslinks into the material. These crosslinks are reversible,
though, as they were rapidly cleaved with light, l <260 nm.
To use this as a shape-memory material, prior to secondary
(UV) crosslinking, the gel was deformed into a temporary
(i.e. elongated) shape. Once exposed to l > 260 nm light, the
gel was permanently fixed in this new shape until
stimulated with light, l <260 nm, breaking the CA crosslinks, allowing the gel to revert to its original shape. As a
second demonstration of the utility of this system, the
authors also used the cinnamic moiety (cinnamylidene
acetic acid, CAA) to build a crosslinked interpenetrating
network around a preformed gel to hold it in a temporary
(corkscrew) shape until exposed to l <260 nm light,
breaking down the secondary network. Between these
two demonstrations, it is clear that the CA molecule could
potentially be used to control shape memory of nearly any
gel system, either through covalent incorporation of it into
the polymer backbone of the gel or through an interpenetrating polymer network, though there are obvious
limitations to biological systems resulting from the
required wavelengths for the chemical groups activity.
Recently, several groups have explored the use of
composite systems in order to harness light for gels that
exhibit thermo-responsive behavior. In these systems,
as with macromolecular assemblies, a light-absorbing

www.mbs-journal.de

345

J. S. Katz, J. A. Burdick

material is dispersed within the polymer

matrix, providing a route to convert light
into thermal energy, leading to a heatinduced phase change for the polymer.
Two independent groups have loaded
gold particles into NIPAm microgels.
Budhlall and coworkers loaded gold
nanoparticles into NIPAm microparticles
and observed their deswelling behavior
in response to light irradiation.[72] Rather
than dispersing the gold throughout the
microgels, these particles had a core/shell
structure, with the gold comprising the
Figure 4. Light-induced transitions in shape-memory polymers. (a) Schematic of polymer
core. However, light exposure still led to
network, where a material is heated above its Tg and set in a temporary shape and then
cooled. UV light can then be introduced to heat the polymer via entrapped nanorods to
significant deswelling of the gels, as the
again raise the temperature above the Tg to allow the material to return to its
polymer was heated above its lower
permanent shape. (b) Example of a coiled sample that returns to its rectangular shape
critical solution temperature (LCST). West
with NIR light exposure.
and coworkers loaded different formulations of gold nanoshells into NIPAm
microgels swollen with various probe easily translatable to other polymeric materials with
teins.[73,74] Upon irradiation in the NIR regime, the nanothermal-responsive behavior.
shells heated the NIPAm above its LCST causing the gels to
Several other composites have been developed to induce
collapse and release their contents in a time-dependent
swelling within polymeric materials. Fujigaya and cowmanner, where the rate of release was inversely correlated
orkers employed carbon nanotubes as a NIR light absorbing
to the molecular weight of the encapsulated agent. While
material in NIPAm gels.[83] Similar to the gold nanopartithis system offers much promise for localized delivery of
therapeutics using NIR light, more advanced formulations
cles, the carbon nanotubes could easily absorb NIR light and
dissipate the energy as localized heat in the gel, leading to
will likely be required to accommodate for the fairly
significant release (ca. 20% of amount released with
rapid deswelling of the material. The swelling/deswelling
irradiation) that was observed in the absence of light.
was repeatable, showing fidelity over 1 200 cycles of light/
While the examples cited above have again focused on
dark. Tatsuma and coworkers loaded titanium dioxide
simple swelling/deswelling of gels upon irradiation, we
nanoparticles into acrylic acid gels.[84] Titanium dioxide
recently reported a light-responsive composite shape
catalyzed the reduction of silver ions to silver nanoparticles
memory system (Figure 4).[75] Similar to the nanoshells
upon UV irradiation. The process is reversible with visible
light. The composite gels were soaked in silver nitrate to
employed by West and coworkers,[73,74] we used gold
load silver ions, which coordinated the acid side chains on
nanorods as a NIR absorbing material. Gold nanorods are an
the polymer, and shrank the gel. Upon exposure to UV light,
exciting material for use as NIR absorbers, as they can be
the silver ions were reduced to silver nanoparticles,
easily synthesized with a tunable absorbance and have the
liberating the acid side chains, allowing the gel to swell.
ability to be surface functionalized through gold surfaceVisible light reoxidized the silver, leading to collapse of the
thiol reactions.[7679] We loaded gold nanorods at a low (<1
gel. The use of photomasks enabled the creation of complex
vol.-%) concentration into tert-butyl acrylate polymer slabs,
swelling morphologies. One could imagine this material
crosslinked with a poly(b-amino ester) polymer.[80] This
being especially useful for applications that require
system could easily be tuned for a specific glass-transition
antibacterial properties, as the silver could serve both to
temperature by altering the monomer/crosslinker ratio,
control the material swelling properties as well as impart
and tert-butyl acrylate had previously been characterized
antibacterial qualities.[85]
for shape-memory applications.[81,82] The formulation was
shown to be non-toxic both in vitro and in vivo. Exposure to
NIR light (770 nm) led to rapid temperature changes of over
30 8C within the slabs. As a demonstration of shapeConclusions and Future Directions
memory ability, a heated polymer slab was shaped into a
coil and cooled back to room temperature. Upon irradiation
There have been many promising advances in the
with NIR light, the sample quickly heated and returned to
development of light-responsive biomaterials, but there
its original rectangular shape (Figure 4b). As with Lendleins
is still much that needs to be done to make these advances
system described above, this method is versatile and could
translatable to the clinic. The advent of two-photon systems

346

Macromol. Biosci. 2010, 10, 339348

2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

DOI: 10.1002/mabi.200900297

Light-Responsive Biomaterials: Development and Applications

and NIR-responsive materials has opened the field to many

new possibilities that were previously unattainable due to
poor in vivo light penetration. At present, though, most
confocal systems are likely too complicated to translate
easily for routine in vivo use, but the materials that have
been generated are quite useful for studies in vitro. As more
materials are developed in the field of molecular imaging
for in vivo use, we imagine that the applicable chemistries
will be translated for use also in materials for drug delivery
and tissue engineering. As the demand develops further
through more applicable advances in the materials field, we
are confident that the technology and systems to control
these materials will be developed as well. Indeed, while
more research is necessary to develop clinically useful
materials, the future holds much promise for employing
light to better control drug delivery and regenerative
medicine.

Acknowledgements: The authors acknowledge funding through a

National Science Foundation graduate research fellowship (J.S.K.),
MRSEC grant (DMR05-20020), and CAREER Award (JAB).

Received: August 15, 2009; Published online: December 14, 2009;

DOI: 10.1002/mabi.200900297
Keywords: biomaterials; drug delivery; light-responsive;
polymers; tissue engineering

[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]

[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]

R. Langer, D. A. Tirrell, Nature 2004, 428, 487.

J. F. Mano, Adv. Eng. Mater. 2008, 10, 515.
F. Helmchen, W. Denk, Nat. Meth. 2005, 2, 932.
W. R. Zipfel, R. M. Williams, W. W. Webb, Nat. Biotechnol.
2003, 21, 1368.
R. Weissleder, Nat. Biotechnol. 2001, 19, 316.
D. E. Discher, A. Eisenberg, Science 2002, 297, 967.
A. Blanazs, S. P. Armes, A. J. Ryan, Macromol. Rapid Commun.
2009, 30, 267.
D. E. Discher, F. Ahmed, Annu. Rev. Biomed. Eng. 2006, 8, 323.
R. Tong, D. A. Christian, L. Tang, H. Cabral, J. R. Baker, K.
Kataoka, D. E. Discher, J. J. Cheng, MRS Bull. 2009, 34, 422.
D. E. Discher, V. Ortiz, G. Srinivas, M. L. Klein, Y. Kim, C. A.
David, S. S. Cai, P. Photos, F. Ahmed, Prog. Polym. Sci. 2007, 32,
838.
Y. Wang, H. Xu, X. Zhang, Adv. Mater. 2009, 21, 2849.
F. Meng, Z. Zhong, J. Feijen, Biomacromolecules 2009, 10, 197.
M.-H. Li, P. Keller, Soft Matter 2009, 5, 927.
E. C. van Dalen, H. N. Caron, H. O. Dickinson, L. C. M. Kremer,
Cochrane Database Syst. Rev. 2008.
V. C. Anderson, D. H. Thompson, Biochim. Biophys. Acta 1992,
1109, 33.
R. H. Bisby, C. Mead, C. G. Morgan, Photochem. Photobiol.
2000, 72, 57.
Y. C. Liu, A. L. M. Le Ny, J. Schmidt, Y. Talmon, B. F. Chmelka,
C. T. Lee, Langmuir 2009, 25, 5713.
G. Wang, X. Tong, Y. Zhao, Macromolecules 2004, 37,
8911.

Macromol. Biosci. 2010, 10, 339348

2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

[19] L. Lin, Z. Yan, Z. J. S. Gu, Y. Y. Zhang, Z. Feng, Y. L. Yu, Macromol.

Rapid Commun. 2009, 30, 1089.
[20] J. Eastoe, M. S. Dominguez, P. Wyatt, A. Beeby, R. K. Heenan,
Langmuir 2002, 18, 7837.
[21] S. Tazuke, S. Kurihara, H. Yamaguchi, T. Ikeda, J. Phys. Chem.
1987, 91, 249.
[22] H. I. Lee, W. Wu, J. K. Oh, L. Mueller, G. Sherwood, L. Peteanu, T.
Kowalewski, K. Matyjaszewski, Angew. Chem. Int. Ed. 2007,
46, 2453.
[23] E. Mabrouk, D. Cuvelier, F. Brochard-Wyart, P. Nassoy, M. H. Li,
Proc. Natl. Acad. Sci. USA 2009, 106, 7294.
[24] B. Chandra, R. Subramaniam, S. Mallik, D. K. Srivastava, Org.
Biomol. Chem. 2006, 4, 1730.
[25] Z. Xie, X. Hu, X. Chen, G. Mo, J. Sun, X. Jing, Adv. Eng. Mater.
2009, 11, B7.
[26] M. Lepage, J. Q. Jiang, J. Babin, B. Qi, L. Tremblay, Y. Zhao, Phys.
Med. Biol. 2007, 52, N249.
[27] J. Q. Jiang, X. Tong, Y. Zhao, J. Am. Chem. Soc. 2005, 127,
8290.
[28] Y. G. Jiang, Y. P. Wang, N. Ma, Z. Q. Wang, M. Smet, X. Zhang,
Langmuir 2007, 23, 4029.
[29] J. Babin, M. Pelletier, M. Lepage, J. F. Allard, D. Morris, Y. Zhao,
Angew. Chem. Int. Ed. 2009, 48, 3329.
[30] A. P. Goodwin, J. L. Mynar, Y. Z. Ma, G. R. Fleming, J. M. J.
Frechet, J. Am. Chem. Soc. 2005, 127, 9952.
[31] I. Aujard, C. Benbrahim, M. Gouget, O. Ruel, J. B. Baudin, P.
Neveu, L. Jullien, Chem. Eur. J. 2006, 12, 6865.
[32] L. Paasonen, T. Laaksonen, C. Johans, M. Yliperttula, K. Kontturi, A. Urth, J. Controlled Release 2007, 122, 86.
[33] G. H. Wu, A. Milkhailovsky, H. A. Khant, C. Fu, W. Chiu, J. A.
Zasadzinski, J. Am. Chem. Soc. 2008, 130, 8175.
[34] T. S. Troutman, S. J. Leung, M. Romanowski, Adv. Mater. 2009,
21, 2334.
[35] G. P. Robbins, M. Jimbo, J. Swift, M. J. Therien, D. A. Hammer,
I. J. Dmochowski, J. Am. Chem. Soc. 2009, 131, 3872.
[36] B. M. Discher, Y. Y. Won, D. S. Ege, J. C. M. Lee, F. S. Bates, D. E.
Discher, D. A. Hammer, Science 1999, 284, 1143.
[37] P. P. Ghoroghchian, P. R. Frail, K. Susumu, D. Blessington, A. K.
Brannan, F. S. Bates, B. Chance, D. A. Hammer, M. J. Therien,
Proc. Natl. Acad. Sci. USA 2005, 102, 2922.
[38] D. H. Levine, P. P. Ghoroghchian, J. Freudenberg, G. Zhang, M. J.
Therien, M. I. Greene, D. A. Hammer, R. Murali, Methods 2008,
46, 25.
[39] F. Ahmed, R. I. Pakunlu, A. Brannan, F. Bates, T. Minko, D. E.
Discher, J. Controlled Release 2006, 116, 150.
[40] J. Y. Lim, H. J. Donahue, Tissue Eng. 2007, 13, 1879.
[41] M. L. Dustin, D. R. Colman, Science 2002, 298, 785.
[42] A. Grakoui, S. K. Bromley, C. Sumen, M. M. Davis, A. S. Shaw,
P. M. Allen, M. L. Dustin, Science 1999, 285, 221.
[43] M. Wu, D. Holowka, H. G. Craighead, B. Baird, Proc. Natl. Acad.
Sci. USA 2004, 101, 13798.
[44] F. G. Giancotti, E. Ruoslahti, Science 1999, 285, 1028.
[45] X. Y. Jiang, D. A. Bruzewicz, A. P. Wong, M. Piel, G. M.
Whitesides, Proc. Natl. Acad. Sci. USA 2005, 102, 975.
[46] J. Ericson, S. Norlin, T. M. Jessell, T. Edlund, Development 1998,
125, 1005.
[47] A. S. Blawas, W. M. Reichert, Biomaterials 1998, 19, 595.
[48] H. Sorribas, C. Padeste, L. Tiefenauer, Biomaterials 2002, 23,
893.
[49] J. Tien, C. M. Nelson, C. S. Chen, Proc. Natl. Acad. Sci. USA 2002,
99, 1758.
[50] S. A. Sandberg, R. W. Barrett, M. Pirrung, A. L. Lu, B.
Kiangsoontra, B. C. P. Holmes, J. Am. Chem. Soc. 1995, 117,
12050.

www.mbs-journal.de

347

J. S. Katz, J. A. Burdick

[51] A. S. Blawas, T. F. Oliver, M. C. Pirrung, W. M. Reichert,

Langmuir 1998, 14, 4243.
[52] M. A. Holden, P. S. Cremer, J. Am. Chem. Soc. 2003, 125, 8074.
[53] J. Doh, D. J. Irvine, J. Am. Chem. Soc. 2004, 126, 9170.
[54] J. S. Katz, J. Doh, D. J. Irvine, Langmuir 2006, 22, 353.
[55] J. Doh, D. J. Irvine, Proc. Natl. Acad. Sci. USA 2006, 103, 5700.
[56] M. Alvarez, A. Best, S. Pradhan-Kadam, K. Koynov, U. Jonas, M.
Kreiter, Adv. Mater. 2008, 20, 4563.
[57] M. S. Hahn, J. S. Miller, J. L. West, Adv. Mater. 2005, 17, 2939.
[58] D. S. W. Benoit, M. P. Schwartz, A. R. Durney, K. S. Anseth, Nat.
Mater. 2008, 7, 816.
[59] A. J. Engler, S. Sen, H. L. Sweeney, D. E. Discher, Cell 2006, 126,
677.
[60] J. Edahiro, K. Sumaru, Y. Tada, K. Ohi, T. Takagi, M. Kameda, T.
Shinbo, T. Kanamori, Y. Yoshimi, Biomacromolecules 2005, 6,
970.
[61] M. S. Hahn, J. S. Miller, J. L. West, Adv. Mater. 2006, 18, 2679.
[62] S. Khetan, J. S. Katz, J. A. Burdick, Soft Matter 2009, 5, 1601.
[63] D. Myung, W. Koh, A. Bakri, F. Zhang, A. Marshall, J. M. Ko, J.
Noolandi, M. Carrasco, J. R. Cochran, C. W. Frank, C. N. Ta,
Biomed. Microdev. 2007, 9, 911.
[64] Y. Luo, M. S. Shoichet, Nat. Mater. 2004, 3, 249.
[65] Y. Luo, M. S. Shoichet, Biomacromolecules 2004, 5, 2315.
[66] J. H. Wosnick, M. S. Shoichet, Chem. Mater. 2008, 20, 55.
[67] A. M. Kloxin, A. M. Kasko, C. N. Salinas, K. S. Anseth, Science
2009, 324, 59.
[68] M. Behl, A. Lendlein, Soft Matter 2007, 3, 58.
[69] A. Mamada, T. Tanaka, D. Kungwatchakun, M. Irie, Macromolecules 1990, 23, 1517.
[70] A. Suzuki, T. Tanaka, Nature 1990, 346, 345.

348

Macromol. Biosci. 2010, 10, 339348

2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

[71] A. Lendlein, H. Y. Jiang, O. Junger, R. Langer, Nature 2005, 434,

879.
[72] B. M. Budhlall, M. Marquez, O. D. Velev, Langmuir 2008, 24,
11959.
[73] M. Bikram, A. M. Gobin, R. E. Whitmire, J. L. West, J. Controlled
Release 2007, 123, 219.
[74] S. R. Sershen, S. L. Westcott, N. J. Halas, J. L. West, J. Biomed.
Mater. Res. 2000, 51, 293.
[75] K. C. Hribar, R. B. Metter, J. L. Ifkovits, T. Troxler, J. A. Burdick,
Small 2009, 5, 1830.
[76] B. Thierry, J. Ng, T. Krieg, H. J. Griesser, Chem. Commun. 2009,
1724.
[77] T. Niidome, M. Yamagata, Y. Okamoto, Y. Akiyama, H. Takahashi, T. Kawano, Y. Katayama, Y. Niidome, J. Controlled
Release 2006, 114, 343.
[78] H. Takahashi, Y. Niidome, T. Niidome, K. Kaneko, H. Kawasaki,
S. Yamada, Langmuir 2006, 22, 2.
[79] T. K. Sau, C. J. Murphy, Langmuir 2004, 20, 6414.
[80] D. G. Anderson, C. A. Tweedie, N. Hossain, S. M. Navarro, D. M.
Brey, K. J. Van Vliet, R. Langer, J. A. Burdick, Adv. Mater. 2006,
18, 2614.
[81] A. M. Ortega, S. E. Kasprzak, C. M. Yakacki, J. Diani, A. R.
Greenberg, K. Gall, J. Appl. Polym. Sci. 2008, 110, 1559.
[82] K. Gall, C. M. Yakacki, Y. P. Liu, R. Shandas, N. Willett, K. S.
Anseth, J. Biomed. Mater. Res. A. 2005, 73, 339.
[83] T. Fujigaya, T. Morimoto, Y. Niidome, N. Nakashima, Adv.
Mater. 2008, 20, 3610.
[84] T. Tatsuma, K. Takada, T. Miyazaki, Adv. Mater. 2007, 19, 1249.
[85] I. Sondi, B. Salopek-Sondi, J. Colloids Interface Sci. 2004, 275,
177.

DOI: 10.1002/mabi.200900297