Synthesis of photodegradable macromers for conjugation and release of bioactive molecules

Article abstract of DOI:10.1021/bm400169d

Hydrogel scaffolds are used in biomedicine to study cell differentiation and tissue evolution, where it is critical to control the delivery of chemical cues both spatially and temporally. While large molecules can be physically entrapped in a hydrogel, moderate molecular weight therapeutics must be tethered to the hydrogel network through a labile linkage to allow controlled release. We synthesized and characterized a library of polymerizable ortho-nitrobenzyl (o-NB) macromers with different functionalities at the benzylic position (alcohol, amine, BOC-amine, halide, acrylate, carboxylic acid, activated disulfide, N-hydroxysuccinyl ester, biotin). This library of polymerizable macromers containing o-NB groups should allow direct conjugation of nearly any type of therapeutic agent and its subsequent controlled photorelease from a hydrogel network. As proof-of-concept, we incorporated the N-hydroxysuccinyl ester macromer into hydrogels and then reacted phenylalanine with the NHS ester. Upon exposure to light (λ = 365 nm; 10 mW/cm², 10 min), 81.3% of the phenylalanine was released from the gel. Utilizing the photodegradable macromer incorporating an activated disulfide, we conjugated a cell-adhesive peptide (GCGYGRGDSPG), a protein that exhibits enzymatic activity (bovine serum albumin (BSA)), and a growth factor (transforming growth factor-β1 (TGF-β1)) into hydrogels, controlled their release with light (λ = 365 nm; 10 mW/cm², 0-20 min), and verified the bioactivity of the photoreleased molecules. The photoreleasable peptide allows real-time control over cell adhesion. BSA maintains full enzymatic activity upon sequestration and release from the hydrogel. Photoreleased TGF-β1 is able to induce chondrogenic differentiation of human mesenchymal stem cells comparable to native TGF-β1. Through this approach, we have demonstrated that photodegradable tethers can be used to sequester peptides and proteins into hydrogel depots and release them in an externally controlled, predictable manner without compromising biological function.

Full text of DOI:10.1021/bm400169d

Article
pubs.acs.org/Biomac
Synthesis of Photodegradable Macromers for Conjugation and
Release of Bioactive Molecules
Donald R. Griffin,†,‡ Jessica L. Schlosser,†,‡ Sandra F. Lam,‡ Thi H. Nguyen,§ Heather D. Maynard,§,∥
and Andrea M. Kasko*,‡,∥
‡Department of Bioengineering, University of California, Los Angeles, 410 Westwood Plaza, 5121 Eng V, Los Angeles, California
90095 United States
§Department of Chemistry and Biochemistry, University of California, Los Angeles, 607 Charles E. Young Drive East, Los Angeles,
California 90095-1569 United States
∥California Nanosystems Institute, 570 Westwood Plaza, Los Angeles, California 90095 United States
S
* Supporting Information
ABSTRACT: Hydrogel scaffolds are used in biomedicine to study cell differentiation
and tissue evolution, where it is critical to control the delivery of chemical cues both
spatially and temporally. While large molecules can be physically entrapped in a
hydrogel, moderate molecular weight therapeutics must be tethered to the hydrogel
network through a labile linkage to allow controlled release. We synthesized and
characterized a library of polymerizable ortho-nitrobenzyl (o-NB) macromers with
different functionalities at the benzylic position (alcohol, amine, BOC-amine, halide,
acrylate, carboxylic acid, activated disulfide, N-hydroxysuccinyl ester, biotin). This
library of polymerizable macromers containing o-NB groups should allow direct
conjugation of nearly any type of therapeutic agent and its subsequent controlled
photorelease from a hydrogel network. As proof-of-concept, we incorporated the N-
hydroxysuccinyl ester macromer into hydrogels and then reacted phenylalanine with
the NHS ester. Upon exposure to light (λ = 365 nm; 10 mW/cm2, 10 min), 81.3% of
the phenylalanine was released from the gel. Utilizing the photodegradable macromer incorporating an activated disulfide, we
conjugated a cell-adhesive peptide (GCGYGRGDSPG), a protein that exhibits enzymatic activity (bovine serum albumin
(BSA)), and a growth factor (transforming growth factor-β1 (TGF-β1)) into hydrogels, controlled their release with light (λ =
365 nm; 10 mW/cm2, 0−20 min), and verified the bioactivity of the photoreleased molecules. The photoreleasable peptide
allows real-time control over cell adhesion. BSA maintains full enzymatic activity upon sequestration and release from the
hydrogel. Photoreleased TGF-β1 is able to induce chondrogenic differentiation of human mesenchymal stem cells comparable to
native TGF-β1. Through this approach, we have demonstrated that photodegradable tethers can be used to sequester peptides
and proteins into hydrogel depots and release them in an externally controlled, predictable manner without compromising
biological function.
released via hydrolysis to induce the differentiation of
mesenchymal stem cells (MSCs) into osteoblasts. Growth
factors such as vascular endothelial growth factor (VEGF) can
be released via enzymatic degradation of an MMP-sensitive
tether to induce angiogenesis.4 Alternatively, affinity inter-
actions (such as ion interactions) can be used to sequester and
release biomolecules from hydrogels. Affinity interactions are
more transient than covalent bonds, but if sufficiently strong,
they can retard the diffusion of species out of the hydrogel. All
three methods typically result in a sustained release profile.
While this is desirable in many therapeutic settings, the ability
to externally control the release of the therapeutic may allow
the administration of a more complex dosing profile.
INTRODUCTION
■
Hydrogels are important biomaterials used in tissue engineering
and regenerative medicine, providing physical support for cells.
Additionally, soluble cues such as proteins or other
biomolecules can be sequestered within and released from
hydrogels.1 Three general techniques exist for controlling the
delivery of biomolecules from hydrogels: physical entrapment,
covalent tethering, and affinity-based sequestration. The
method used to control a biomolecule’s release from a hydrogel
is dictated, at least in part, by its size (molecular weight). Large
molecules such as proteins can be physically entrapped within
the mesh of the hydrogel, which impedes their diffusion. Lower
molecular weight species are typically covalently conjugated to
the network through degradable linkages (usually ones sensitive
to hydrolytic or enzymatic degradation) because their diffusion
is not significantly retarded by the hydrogel. For example,
therapeutic agents such as dexamethasone2 or statins3 can be
Received: February 1, 2013
Revised: February 19, 2013
Published: March 18, 2013
© 2013 American Chemical Society
1199
dx.doi.org/10.1021/bm400169d | Biomacromolecules 2013, 14, 1199−1207

Biomacromolecules
Article
in 1:1 DMSO/PBS) overnight. The hydrogels were then washed with
the 50% DMSO/PBS solution. All gels were placed in individual wells
of a 48-well plate and placed with 500 μL of the DMSO solution. Half
the gels (N = 3) were exposed (λ = 365 nm; 10 mW/cm2, 10 min),
while the remaining three remained unexposed. All gels were allowed
to leach on a shaker plate overnight, then tested for the presence of L-
Phe at 257 nm via standard UV/vis protocol. A standard curve of L-
Phe was prepared prior to testing.
While hydrolysis and enzymolysis are both effective methods
for sustained release of therapeutic agents, the release rate
cannot be adjusted or arrested after the hydrogel is fabricated,
and release is not spatially controlled. As an alternative to
hydrolytic and enzymatic degradation for controlled (sus-
tained) release, we have developed and optimized photo-
degradation as a mechanism for controlled drug release.
Photodegradation offers precise external temporal and spatial
control over drug release. Photodegradable groups have been
used in the presence of live cells to uncage neurotransmitters,5
to pattern physical voids within a hydrogel,6−9 and to spatially
pattern functional groups on and within10−13 hydrogels.
We previously reported coupling a photosensitive polymer-
izable ortho-nitrobenzyl (o-NB) group to fluorescein (model
drug) to generate a model photoreleasable therapeutic agent.14
We copolymerized this macromer into hydrogel depots and
quantified the release of fluorescein as a function of light
exposure at multiple wavelengths (365−436 nm), intensities
(5−20 mW/cm2), and durations (0−20 min) and correlated
the release profiles to a predictive model. Although these results
were promising, the conjugation was performed in organic
solvent, which would be unsuitable for many biomolecules, and
the site we chose for conjugation left the ortho-nitroso ketone
fragment attached to the model therapeutic. Furthermore, each
new therapeutic agent of interest would require independent
synthesis. We next reported a series of o-NB linkers with
different rates of photodegradation to allow the multistaged
release of cells15 and model therapeutics.16 Although these
reports resolved some of the issues noted above, the variety of
functional groups that could be incorporated was still limited.
Bioconjugation techniques take advantage of functional
groups commonly found on biomolecules such as amines,
carboxylic acids, alcohols, and thiols. In order to allow
conjugation of a wider variety of molecules, we are interested
in o-NB macromers with different reactive groups at the
benzylic position (release site) that allow easy incorporation
under mild conditions. Here we report the synthesis of
photodegradable o-NB macromers with a variety of functional
groups at the benzylic position. This will allow for covalent
conjugation of a wider variety of biomolecules and therapeutics
to the o-NB linker and their subsequent delivery from a
hydrogel, without having to resynthesize the macromer each
time. We demonstrate that amino acids, peptides, and proteins
can be quantitatively sequestered into hydrogels using a
photodegradable tether and subsequently released in an
externally controlled, predictable manner without compromis-
ing biological function.
Fabrication of Hydrogels Containing Cell Adhesive Peptide.
Stock solutions of PEG526-methacrylate-4-(2-methoxy-5-nitro-4-(1-
(4-oxo-4-(2-(pyridin-2-yldisulfanyl)ethoxy)butanoyl)oxy))butanoate
(10 mg/mL in DMSO), TEMED (10% by vol. in phosphate buffered
saline (PBS), pH 7.4, 1 mM), and APS (0.22 M, in PBS) were
prepared prior to addition. PEG 10000 DA hydrogel disks were
fabricated by dissolving PEG 10000 diacrylate (0.10 g, 9.9 μmol) in
PBS (0.35 mL) and DMSO (0.4 mL), followed by addition of
PEG526-methacrylate-4-(2-methoxy-5-nitro-4-(1-(4-oxo-4-(2-(pyri-
din-2-yldisulfanyl)ethoxy)butanoyl)oxy))butanoate (1.0 mg, 1.9 μmol,
0.1 mL stock). To initiate polymerization, APS (100 μL) and TEMED
(25 μL) were added sequentially, followed by immediate placement of
solution between two glass slides separated by rubber spacers (0.33
mm). The resulting hydrogels were cured for 90 min, cut into 5 mm
discs, and leached with 1:1 DMSO/PBS, ethanol, and PBS. The
hydrogels were divided into sets (10 gels/set, N = 3) and each set was
placed in a 1 mL loading solution of buffered aqueous
GCGYGRGDSPG (0.1 mM in PBS, 3 equiv total) overnight. The
loading solution was tested for the presence of released pyridine-2-
thione (8080 M−1 cm−1) at 1 and 24 h after exposure to check the
progress of the disulfide exchange by the standard UV−vis protocol.17
The hydrogels were then washed with PBS and either seeded with cells
(30000 cells per well), exposed (λ = 365 nm; 10 mW/cm2, 20 min)
and seeded with cells, or exposed to fluorescein-NHS (5 mol. equiv. in
1:1 DMSO/PBS) for 2 h, before washing repeatedly with 1:1 DMSO/
PBS to remove unconjugated fluorescein.
Fluorescence Calibration Curve. Fluorescein-NHS (4.8 mg, 10
μmol) was dissolved in DMSO (5.07 mL), isoleucine (6.6 mg, 51
μmol) was dissolved in PBS (5.07 mL), and the two solutions were
combined and stirred overnight. This stock solution (1 mM) was
diluted serially and tested on a Beckman Coulter DTX 880 Multimode
Detector (λex = 485 nm; λem = 535 nm) to create a calibration curve.
Cell-Adhesive Hydrogel Exposure and Release Measurement.
Each hydrogel was placed individually in the well of a 48-well plate,
exposed for a specified time to light (N = 3, 365 nm, 10 mW/cm2) at
21 °C. Following exposure, each hydrogel was leached with a 1:1
DMSO/PBS mixture (1 mL) overnight before testing on a Beckman
Coulter DTX 880 Multimode Detector (λex = 485 nm; λem = 535 nm).
Mesh Size Calculation. To calculate the mesh size of the
polymerized hydrogels, a separate hydrogel was polymerized between
glass slides separated by a larger spacer (1.66 mm) using identical
polymerization and leaching conditions to those stated above. The
complex modulus was measured using a TA Instruments Q800 DMA.
The hydrogel mass was measured before and after lyophilization, and
combined with the density of PEG 10K18 to determine the swelling
ratio (Q). The molecular weight between cross-links (Mc) was then
calculated using a modified equation from the literature (eq Eq. 1)19
and used to find the cross-linked network characteristic length of the
hydrogel (ξ) (eq Eq. 2).
EXPERIMENTAL SECTION
■
Release Experiments. Phenylalanine Release. Stock solutions of
PEG526-methacrylate-PDG NHS (10 mg/mL in DMSO), tetrame-
thylethylene diamine (TEMED, 10% by vol. in phosphate buffered
saline (PBS), pH 7.4, 1 mM), and ammonium persulfate (APS, 10 wt
%, in PBS) were prepared prior to addition. PEG 10000 DA hydrogel
disks were fabricated by dissolving PEG 10000 diacrylate (0.10 g, 9.9
μmol) in PBS (0.35 mL) and DMSO (0.4 mL), followed by addition
of PEG526-methacrylate-4-(4-(1-((4-((2,5-dioxopyrrolidin-1-yl)oxy)-
4-oxabutanoyl)oxy)ethyl)-2-methoxy-5-nitrophenoxybutanoate (1.0
mg, 1.9 μmol, 0.1 mL stock). To initiate polymerization, APS (100
μL) and TEMED (25 μL) were added sequentially, followed by
immediate placement of solution between two glass slides separated by
a glass slide (1 mm). The resulting hydrogels were cured for 90 min,
cut into 5 mm discs, and leached with 1:1 DMSO/PBS. All hydrogels
were placed in a 3 mL loading solution of L-phenylalanine (10 mg/mL
1
Mc
2
Mn
E
1
=
+
1/3
2(1 + v)
ρ RT(υ2)
p
(Eq. 1)
(Eq. 2)
ξ = υ2−1/3Cn1/2 ln1/2
BSA Loading and Diffusion. The 10 wt % PEG 10KDA hydrogels
(d = 5 mm, h = 1 mm) were placed in individual wells on a 48 well
plate and each well was loaded with 250 μL of fluorescein tagged BSA
(1 mg/mL in PBS) for 16 h. After equilibration, all solution was taken
out of each well, tested on a Beckman Coulter DTX 880 Multimode
Detector, λex = 485 nm; λem = 535 nm and replaced with fresh PBS
1200
dx.doi.org/10.1021/bm400169d | Biomacromolecules 2013, 14, 1199−1207

Biomacromolecules
Article
Scheme 1. Synthesis of PEG526-methacrylate-4-(4-(1-hydroxyethyl)-2-methoxy-5-nitrophenoxy)butanoate
every 5 min until diffusion of fluorescein out of the gel was no longer
detected.
Hydrogel Synthesis for Protein Conjugation after Polymerization
(Linker w/PEG 526MA). Hydrogels were made with PEG526-
methacrylate-4-(2-methoxy-5-nitro-4-(1-(4-oxo-4-(2-(pyridin-2-
yldisulfanyl)ethoxy)butanoyl)oxy))butanoate identical to the samples
made for RGD incorporation.
%) and TEMED (25 μL, 50 v/v%) were added sequentially to the
experimental solution. The solution was polymerized between two
glass slides (thickness = 1 mm) for 12 h then washed with ultrapure
water (4 × 30 min), ethanol (for sterilization; 1 × 1 h), 50:50 ethanol/
PBS (2 × 30 min), and PBS (2 × 30 min).
Hydrogel Exposure and Release Measurement. Each hydrogel was
placed individually in the well of a 48-well plate, exposed for a specified
time to light (N = 3, 365 nm, 10 mW/cm2) at 21 °C. Following
exposure each hydrogel was leached with PBS (0.25 mL) overnight
before testing each solution by micro-BCA analysis (Pierce).
BSA Activity Test. Assessment of BSA esterase activity was
performed following a literature procedure.20 Briefly, the concen-
tration of the released BSA solution (N = 3) was quantified by BCA.
Subsequently a solution of native BSA was created of equal
concentration. These solutions were combined, separately, with
solutions of p-nitrophenyl acetate. Following incubation, the change
in absorbance for each solution was measured by UV/vis at 348 nm
and compared.
hMSC Culture. Human mesenchymal stem cells (hMSCs) were
provided by the Texas A&M Health Science Center College of
Medicine. hMSCs were cultured in αMEM with 2 mM L-glutamine-
(Hyclone) supplemented with 16.5% fetal bovine serum (FBS, Atlanta
Biologicals) and 100 μg/mL penicillin−streptomycin (Hyclone) at 37
°C in a 5% CO2 environment. Growth media was exchanged every 2−
3 days.
Cell Differentiation. The hMSCs were cultured in monolayer at a
density of 5 × 103 cells/cm2 in 24 well plates for 8 h at 37 °C in 5%
CO2. TGF-β1 (Peprotech) was diluted to concentrations of 10 ng/mL
in serum-free medium and applied to hMSCs for the positive control.
Medium containing released TGF-β1 from the exposed hydrogels was
applied to hMSCs to verify its bioactivity in comparison to the positive
control. For the negative control, the hMSCs received fresh serum-free
medium that did not contain any TGF-β1. Cells were cultured for
three days with no medium changes and then fixed overnight at 4 °C
in 10% buffered formaldehyde and rinsed twice with PBS. The cells
were permeabilized using 0.1% Triton X-100 in PBS for 5 min at RT
and rinsed twice. Blocking solution (1% BSA in PBS) was applied for
30 min, and the cells were subsequently rinsed 3× with PBS. The cells
were incubated with toluidine blue (1:400 in blocking solution) at RT
for 1 h and rinsed 3× with PBS. Phase contrast images (Zeiss
AxioObserver Inverted Fluorescent Microscope) of the (stained)
hMSCs were taken.
Protein Infusion into PEG526-methacrylate-4-(2-methoxy-5-
nitro-4-(1-(4-oxo-4-(2-(pyridin-2-yldisulfanyl)ethoxy)butanoyl)-
oxy))butanoate Containing Hydrogels. Following polymerization
and leaching the hydrogels were infused with a BSA solution (1 mM).
Hydrogels with PEG526-methacrylate-4-(2-methoxy-5-nitro-4-(1-(4-
oxo-4-(2-(pyridin-2-yldisulfanyl)ethoxy)butanoyl)oxy))butanoate
were also infused with PBS only and glutathione (1 mM) solutions to
act as negative and positive controls, respectively. The pyridine-2-
thione release (8080 M−1 cm−1) was monitored at 342 nm for 48 h
using UV/vis spectroscopy. No change in absorbance was seen relative
to control hydrogels during this period.
Hydrogel Synthesis for Protein Conjugation after Polymerization
(Linker w/PEG 10KMA, 10 wt %). PEG 10K methacrylate 4-(2-
methoxy-5-nitro-4-(1-(4-oxo-4-(2-(pyridin-2-yldisulfanyl)ethoxy)-
butanamido)ethyl)phenoxy)butanoate/PEG 10KMA (4:96 mol %,
0.15 g) was dissolved in PBS (1.275 mL). Solutions of APS (150 μL,
10 w/v%) and TEMED (75 μL, 10 v/v%) were added sequentially,
and the hydrogels were polymerized between two glass slides
(thickness = 0.5 mm) for 1 h. The hydrogels were then cut into 5
mm discs using a biopsy punch. The discs were washed with PBS six
times to remove unreacted material (5 × 30 min and 1 × overnight
washes) and stored at 5 °C until use.
Protein Conjugation after Polymerization (Linker w/PEG 10KMA,
10 wt %). Following polymerization and leaching, the hydrogels were
infused with a BSA solution (1 mM). Two sets of hydrogels were also
infused with PBS only and glutathione (1 mM) solutions to act as
negative and positive controls, respectively. The pyridine-2-thione
release (8080 M−1cm−1) was monitored at 342 nm for 24 h using UV/
vis spectroscopy and compared to the expected exchange based on
complete incorporation of the o-NB linker during polymerization.
Prepolymerization Exchange with BSA and Subsequent Hydrogel
Synthesis (10 wt % PEG). Stock solutions of PEG 10KMA 4-(2-
methoxy-5-nitro-4-(1-(4-oxo-4-(2-(pyridin-2-yldisulfanyl)ethoxy)-
butanamido)ethyl)phenoxy)butanoate/PEG 10DKMA (4:96 mol %,
224 mg in 950 μL) and BSA (1 mM) were predissolved in PBS. A total
of 475 μL of each stock solution were combined to initiate exchange,
while 475 μL of each solution were also combined with PBS (475 μL)
to act as negative controls of exchange. After 4 h, aliquots (100 μL) of
all three solutions (two negatives, one experimental) were diluted
(1:10) with PBS and tested for (8080 M−1 cm−1) absorbance at 342
nm by UV/vis spectroscopy. APS (75 μL, 10 w/v%) and TEMED (75
μL, 10 v/v%) were added sequentially to the experimental solution.
The solution was polymerized between two glass slides (thickness =
0.5 mm) for 1 h and washed with PBS (5 × 30 min, 1 × overnight).
Prepolymerization Exchange with TGF-β1 and Subsequent
Hydrogel Synthesis (10 wt % PEG). Stock solutions of PEG
10KMA 4-(2-methoxy-5-nitro-4-(1-(4-oxo-4-(2-(pyridin-2-
yldisulfanyl)ethoxy)butanamido)ethyl)phenoxy)butanoate/PEG
10KDMA (4:96 mol %, 224 mg in 950 μL) and TGF-β1 (1 μg/mL)
were predissolved in PBS. A total of 100 μL of each stock solution was
combined to initiate exchange and was tested for (8080 M−1 cm−1)
absorbance at 342 nm by UV/vis spectroscopy at t = 0 and t = 4 h.
PEG 10,000 diacrylate (90 mg in 750 μL) was dissolved in PBS and
combined with the exchanged TGF-β1 solution. APS (25 μL, 50 w/v
Histology. Cells were stained with toluidine blue (Acros Organics)
to visualize sulfated glycosaminoglycan (GAG) deposition. Following
standard protocol,21 a 5 mg/mL solution of toluidine blue was used to
stain the cells for 15 min and then washed three times with PBS for 5
min each.
GAG Measurement. After culturing the cells for 3 days, GAG
content was quantitatively measured spectrophotometrically using the
dimethylmethylene blue (DMMB; Polysciences, Inc.) assay with slight
modifications.22 Briefly, cells were digested with 1 mL of papain
solution (Acros Organics) for 16 h at 60 °C. The cell solution was
then passed through a syringe filter and a DMMB solution was applied
to the sample. Absorbance was measured at 650 nm and compared to
a chondroitin sulfate solution standard (Sigma-Aldrich).
TGF-β1 Quantification. The PBS leach solutions surrounding the
hydrogels were diluted 1:100 with PBS, then tested for TGF-β1
presence using a sandwich ELISA (TGF-β Emax ImmunoAssay System,
Promega).
1201
dx.doi.org/10.1021/bm400169d | Biomacromolecules 2013, 14, 1199−1207

Biomacromolecules
Article
Scheme 2. Synthesis of Polymerizable o-NB Modified with an Activated Ester and an Activated Disulfide
Scheme 3. Synthesis of PEG526-methacrylate-4-(4-(1-aminoethyl)-2-methoxy-5-nitrophenoxy)butanoate
Statistics. Data are presented as mean standard deviation with
three samples averaged for each data point.
uncharacteristically low, as a significant amount of product was
lost during purification via gradient chromatography. The NHS
ester should allow for direct conjugation of proteins to the
photodegradable group through any free amines,25 while the
activated pyridyl disulfide reacts with free thiols via disulfide
exchange.17
RESULTS AND DISCUSSION
■
The main building block for the photodegradable macromers in
this report is 4-(4-(1-hydroxyethyl)-2-methoxy-5-
nitrophenoxy)butanoic acid, the synthesis of which has been
previously reported.6,14,23 This o-NB group contains both a
carboxylic acid and a benzylic alcohol, allowing for separate
functionalization of these two moieties. In order to obtain a
functional group reactive in the radical polymerizations typically
used to fabricate poly(ethylene glycol) hydrogels, we first
esterified the carboxylic acid group using tosylated PEG 526
methacrylate and potassium fluoride in DMF24 (Scheme 1).
Unlike carbodiimide couplings or acid chloride mediated
esterifications, this nucleophilic substitution leaves the benzylic
alcohol unaffected. While the yield of this reaction is modest
(52%), this is in part due to the difficulty of isolating the
product, which is a viscous oil.
In order to functionalize the o-NB linker with an amine at the
benzylic position, we first converted the benzyl alcohol of 4-(4-
(1-hydroxyethyl)-2-methoxy-5-nitrophenoxy)butanoic acid to a
bromide using phosphorus tribromide. We then reacted the
benzyl bromide with ammonium hydroxide to yield the benzyl
amine, which we then protected with tert-butyl carbonate. We
coupled PEG-526 methacrylate to the carboxylic acid to yield a
macromer containing a protected amine (Scheme 3).
Deprotection under standard acidic conditions (trifluoroacetic
acid) simultaneously cleaves ester linkages in the macromer,
and deprotection using tetrabutylammonium fluoride was also
unsuccessful. However, the t-BOC can be selectively removed
using bismuth(III) trichloride in a mixture of acetonitrile and
water, with all other functionalities remaining intact.26
4-(4-(1-Bromoethyl)-2-methoxy-5-nitrophenoxy)butanoic
acid can be converted to the acid chloride using thionyl
chloride or phosphorus pentachloride and used to esterify
PEG-526 methacrylate; however, some halogen exchange
occurs in the process, producing a mixture of benzyl bromide
and benzyl chloride macromers (Supporting Information,
Scheme S2).
The benzylic alcohol can be reacted with succinic anhydride
to produce a carboxylic acid (Scheme 2). The carboxylic acid is
easily esterified with N-hydroxysuccinimide (NHS) or with 2-
(pyridin-2-yldisulfanyl)ethanol via 1-ethyl-3-(3-dimethylamino-
propyl) carbodiimide (EDC) coupling (Scheme 2). Biotin can
be converted to an acid chloride and used to esterify the
benzylic alcohol of the o-NB group, or it can be conjugated
directly to the alcohol via EDC coupling (Supporting
Information, Scheme S1). The yield of this reaction was
1202
dx.doi.org/10.1021/bm400169d | Biomacromolecules 2013, 14, 1199−1207

Biomacromolecules
Article
Scheme 4. Synthesis of PEG526-methacrylate-4-(4-(1-acryloyloxyethyl)-2-methoxy-5-nitrophenoxy)butanoate
The final macromer we synthesized contained both an
acrylate and a methacrylate functionality; free thiols (such as
those found on cysteine) react rapidly with acrylates through a
base-catalyzed Michael addition, while reaction with the
methacrylate is slow.27 4-(4-(1-Hydoxyethyl)-2-methoxy-5-
nitrophenoxy)butanoic acid is acrylated, and the carboxylic
acid is subsequently converted to an ester by EDC coupling to
PEG-526 methacrylate (Scheme 4).
Chart 1 summarizes the reactivity of each of the o-NB
macromers in this report. This modular library of o-NB linkers
allows conjugation to a wide variety of functional groups found
on biomolecules and therapeutic agents. Depending on the
linker chosen, a small molecular fragment may remain attached
to the therapeutic agent after photorelease. For the o-NB
linkers with alcohol, alkyl halide, or amine at the benzylic
position, depending on how the therapeutic agent is
conjugated, it may be released in its unaltered state.
Conjugation of a therapeutic agent to o-NB linkers with either
the carboxylic acid, NHS ester, or pyridyl disulfide results in an
additional small molecular fragment attached to the therapeutic
agent (i.e., succinic acid), which may or may not affect the
therapeutic activity of the drug.
To demonstrate the utility of these linkers for releasing
therapeutic agents, we first copolymerized PEG 10K diacrylate
and the NHS-functionalized macromer, PEG526-methacrylate-
4-(4-(1-((4-((2,5-dioxopyrrolidin-1-yl)oxy)-4-oxabutanoyl)-
oxy) ethyl)-2-methoxy-5-nitrophenoxybutanoate (abbreviated
PEG-526MA-o-NB-NHS), using ammonium persulfate (APS)
and N,N,N′,N′-tetramethylethylene diamine (TEMED) as the
redox initiating system. The resultant hydrogels were leached to
remove any unreacted macromer or initiator and then
incubated with a solution of L-phenylalanine. The free amine
should react with the NHS ester to produce an amide linkage
and release N-hydroxysuccinimide, analogous to the standard
bioconjugation technique that utilizes amines in proteins to
react with NHS-functionalized molecules. The in-gel reaction
was allowed to proceed overnight before any unreacted
phenylalanine was leached from the gels through successive
washing. One set of gels was then exposed to light (λ = 365
nm; 10 mW/cm2, 10 min), and the amount of phenylalanine
released was quantified via UV−vis spectroscopy. Assuming (a)
100% reactive incorporation of PEG-526MA-o-NB-NHS into
the hydrogel, (b) none of the NHS esters hydrolyzed during
polymerization or exchange, and (c) all of the o-NB groups
photolyzed, 81.3% of the succinyl amide of phenylalanine was
released from the gel. Although these results indicate that PEG-
526MA-o-NB-NHS can be used to conjugate molecules
containing free amines into the gel, there is no easy way to
quantify the amount of amino acid or other amine-containing
molecule within the gel prior to release.
synthesized with cysteine residues, we next investigated the
photodegradable macromer containing an activated disulfide
linkage, poly(ethylene glycol)(PEG)-526-methacrylate-4-(2-
methoxy-5-nitro-4-(1-((4-oxo-4-(2-(pyridin-2-yldisulfanyl)-
ethoxy)butanoyl)oxy)butanoate (abbreviated as PEG-526MA-
o-NB-SSpyr). The pyridine disulfide moiety undergoes disulfide
exchange with free thiols,17 releasing pyridine-2-thione, which is
quantified via absorbance spectroscopy (Scheme 5). This
approach allows conjugation of thiol-containing biomolecules
to the photodegradable macromer either before (Scheme 5a) or
after (Scheme 5b) formation of the hydrogel. Not only can the
amount of incorporated biomolecule be easily quantified (by
measuring pyridine-2-thione release), but biomolecules sensi-
tive to hydrogel formation conditions can be introduced
postfabrication.
To demonstrate the utility of this linker for sequestering and
releasing peptides we copolymerized PEG 10K diacrylate and
PEG-526MA-o-NB-SSpyr using APS and TEMED. Hydrogels
containing 1 mM activated disulfide were incubated with a
solution of the cell-adhesive peptide GCGYGRGDSPG. In
solution, disulfide exchange is complete within 5 min at pH 6−
8, however, release of pyridine-2-thione is somewhat slower
from the hydrogel (likely due to sterics28), so gels were allowed
to react overnight at 4 °C. Based on pyridine-2-thione release,
the gels were found to incorporate 0.34 mM RGD via exchange.
Although this concentration is lower than the concentration of
the pyridine disulfide groups available within the gel, the RGD
concentration is sufficient to promote cell adhesion. In order to
quantify release of RGD and determine the exposure time
required to fully release the adhesive peptide, a set of hydrogels
were incubated with NHS-FITC, which reacts with the N-
terminus of the peptide. The unreacted FITC was washed from
the hydrogels, which were subsequently exposed to 365 nm
light (I0 = 10 mW/cm2). The amount of released peptide was
quantified via fluorescence. Complete release occurs in less than
10 min (Figure 1a), indicating that these exposure conditions
are sufficient to release all of the cell-adhesive peptide from the
gels.
To test the activity of the peptide and confirm its release
from the gel, fibroblasts were seeded onto gels containing the
photoreleasable RGD peptide and onto gels that had been
exposed to light (λ = 365 nm, I0 = 10 mW/cm2, t = 20 min)
and washed multiple times to remove the photoreleased
peptide. Cells adhere to gels containing the RGD, and begin to
spread within 60 min, while cells seeded onto gels from which
the peptide was photoreleased round up (Figure 1b) and are
washed away (data not shown). Photodegradation can
therefore be used as a tool to control cell adhesion to these
biomaterials.
Low molecular weight compounds diffuse freely into and out
of hydrogels; however, the diffusion of larger species is retarded
by the gel, and, above a certain molecular weight, prevented.
Because many proteins either contain free thiols or are easily
functionalized with a thiol group, and peptides are easily
1203
dx.doi.org/10.1021/bm400169d | Biomacromolecules 2013, 14, 1199−1207

Biomacromolecules
Article
to determine the effect of the gel structure (ξ=143.5 Å) on the
diffusion of larger biomolecules in the gel19 and determine the
approximate size of biomolecules that could be effectively
introduced into and released from the hydrogel. For this
hydrogel system, where ξ = 143.5 Å and v2 = 0.05, Dg/D0
decreases from 0.88 to 0.62 when Rs increases from 10 to 50 Å,
a relevant size range for macromolecular species such as
proteins. Practically, this means that any macromolecular agent
loaded into or released from these hydrogel depots requires
extended equilibration time (on the order of a few hours) to
account for retarded diffusion through the gel.
Chart 1. Reactivity of o-NB Conjugates towards Different
Functional Groups Found on Biomolecules
⎛
⎝
⎞
⎠
⎛
⎞
D
⎛
⎞
⎟
Rs
ξ ⎠
v2
1 − v
g
⎜
⎟
⎟
= 1 −
exp −Y
⎜
⎝
⎟
⎠
⎜
⎜
D0
⎝
2
(Eq. 3)
To experimentally verify the effect of the gel on protein
diffusion out of the network, we prepared a set of hydrogels
that did not contain the activated disulfide and incubated these
gels in a solution of FITC-labeled bovine serum albumin (BSA,
Mn ∼ 66500) overnight. We monitored the diffusion of BSA
out of the gels and found that the BSA is completely released
within 3 h (Figure 2a). Therefore, proteins and peptides of the
same or smaller size should be able to diffuse into and out of
these hydrogels completely within a few hours.
To test the utility of this system for sequestering proteins,
hydrogels containing the activated disulfide were incubated
with a solution of BSA (which contains a free thiol29), but no
disulfide exchange occurred, even under extended incubation
(>48 h). Because BSA diffuses into and out of the gel within a
few hours, we presume the photodegradable tether is sterically
inaccessible to larger proteins. To confirm, we synthesized a
new linker, PEG-10K-methacrylate-4-(2-methoxy-5-nitro-4-(1-
(4-oxo-4-(2-(pyridin-2-yldisulfanyl)ethoxy)butanamido)ethyl)-
phenoxy)butanoate (abbreviated PEG-10K-MA-o-NB-SSpyr).
The PEG chain in this macromer is significantly longer (Mn =
10,000 vs Mn = 536 Da), which allows greater distance between
the network cross-link site and the activated disulfide (227
ethylene oxide repeat units vs 11). We copolymerized PEG-
10K-MA-o-NB-SSpyr with PEG 10K dimethacrylate and
infused the hydrogels with a solution of BSA. Pyridine-2-thione
was released, confirming that sterics were likely limiting the
interaction of protein with the photodegradable linker. Despite
the significantly longer tether, only approximately 10% of the
disulfide groups underwent exchange, reinforcing our hypoth-
esis that sterics play an important role in conjugating proteins
to these hydrogels postfabrication.30
If a protein is stable to the polymerization conditions, it can
undergo disulfide exchange with PEG-10K-MA-o-NB-SSpyr
prior to incorporation into the hydrogel (Scheme 5a). We
incubated BSA in a buffered solution of PEG-10K-MA-o-NB-
SSpyr at 4 °C overnight; pyridine-2-thione release indicates
complete exchange occurred. The PEG-10K-MA-o-NB-S-BSA
conjugate was copolymerized with PEG10K dimethacrylate
into a hydrogel. After washing to remove any unreacted
materials, hydrogels were exposed to 365 nm light (I0 = 10
mW/cm2), allowed to equilibrate in buffered solution overnight
at 4 °C, and protein release was quantified via UV−vis
spectroscopy (λ = 280 nm). The release profile of BSA was
exponential (Figure 2b). The actual concentration of BSA
The diffusion coefficient for a molecule in the gel, Dg, relative to
its diffusion coefficient in free solution, D0, is a function of the
radius of that molecule, Rs, the mesh size of the hydrogel (ξ),
and the polymer volume fraction in the gel (v2; (Eq. 3; Y is the
ratio of critical volume needed for translational movement of
the molecule to average free volume per liquid molecule,
usually approximated to equal one). We characterized the
physical properties of the hydrogel (E* = 32.75 kPa, Q = 20),
released after complete degradation (126
8 μg/mL) was
slightly lower than expected (155 μg/mL); this difference may
be due to hydrolysis of the tether prior to fabrication,
incomplete reactive incorporation of the tethered protein
1204
dx.doi.org/10.1021/bm400169d | Biomacromolecules 2013, 14, 1199−1207

Biomacromolecules
Article
a
Scheme 5. Incorporation of Thiol-Containing Biomolecules into Hydrogels
a
(a) Disulfide exchange of PEG-10K-methacrylate- 4-(2-methoxy-5-nitro-4-(1-(4-oxo-4-(2-(pyridin-2-yldisulfanyl)ethoxy)butanamido)ethyl)-
phenoxy) butanoate, followed by copolymerization into a hydrogel network, and (b) copolymerization of PEG526-methacrylate-4-(2-methoxy-5-
nitro-4-(1-((4-oxo-4-(2-(pyridin-2-yldisulfanyl) ethoxy)butanoyl)oxy)butanoate or PEG-10K-methacrylate- 4-(2-methoxy-5-nitro-4-(1-(4-oxo-4-(2-
(pyridin-2-yldisulfanyl) ethoxy)butanamido)ethyl)phenoxy)butanoate into a hydrogel, followed by disulfide exchange.
BSA was quantified using p-nitrophenyl acetate as the substrate.
The released BSA exhibits identical esterase activity compared
to the native BSA that did not experience sequestration and
release (λ = 405 nm; native, A = 0.185 0.006; released, A =
0.196
0.006). These results demonstrate that moderate
molecular weight proteins can be sequestered and released
from hydrogels using light while maintaining their enzymatic
activity.
These results are encouraging, but in order to use this system
to deliver chemical cues to cells, we need the ability to
incorporate more sensitive biomolecules such as growth factors.
TGF-β1 is a growth factor important in wound healing and
implicated in many diseases such as fibrosis and cancer. It has a
moderate molecular weight (∼25 kDa) and contains nine
cysteine residues; eight form disulfide bonds, while one is free,
allowing its facile exchange with the activated disulfide.31,32
TGF-β1was incubated with PEG-10K-MA-o-NB-SS-Pyr for 12
Figure 1. (a) Fractional release of GCGYGRGDSPG as a function of
exposure time (λ = 365 nm, I0 = 10.0 0.2 mW/cm2); (b) changes in
morphology of fibroblasts seeded onto an unexposed gel (left) and an
exposed gel (right; λ = 365 nm, I0 = 10 mW/cm2, t = 20 min; images
acquired 60 min after seeding).
during polymerization, or slight sequestration of the released
BSA into the hydrogel. The enzymatic activity of the released
Figure 2. (a) Diffusion of FITC-labeled BSA out of hydrogels as a function of time; (b) BSA release as a function of exposure time at 365 nm (λ =
365 nm, I0 = 10.0 0.2 mW/cm2).
1205
dx.doi.org/10.1021/bm400169d | Biomacromolecules 2013, 14, 1199−1207

Biomacromolecules
Article
h at 4 °C and pyridine-2-thione release was monitored. The
TGF-β1 photodegradable macromer conjugate was copoly-
merized with PEG10K dimethacrylate into hydrogels. After
washing to remove any unreacted materials, the gels were
exposed to 365 nm light (I0 = 10 mW/cm2, t = 10 min) and
allowed to equilibrate in buffer for 2 h, to release a final
concentration of 5.2 ng/mL TGF-β1 (quantified by ELISA).
The solutions were applied without dilution to plated hMSCs,
which undergo chondrogenesis in the presence of TGF-β1.33,34
Glycosaminoglycan (GAG) production was visualized via
toluidine blue staining (Figure 3a−c). After three days,
free thiols either before or after incorporation (respectively) of
the macromer into a hydrogel depot. The NHS-ester allows
conjugation of any protein through lysine residues or N-
terminal amines. While conjugation prior to hydrogel
fabrication is more efficient, NHS-esters can survive radical
polymerizations and thus should allow postfabrication incor-
poration (as demonstrated using phenylalanine as a model
compound). The carboxylic acid functionality will allow
conjugation to alcohols and amines via ester and amide
formation. The alcohol functionality provides conjugation to
carboxylic acids via ester formation, or conjugation to
molecules with good leaving groups via nucleophilic sub-
stitution (Chart 1). Only the acrylate and the benzyl bromide
should be sensitive to standard free radical polymerization
conditions, requiring their conjugation to biomolecules prior to
hydrogel fabrication. All other groups allow postfabrication
incorporation of biomolecules into the hydrogel.
CONCLUSIONS
■
Here we report the synthesis of a library of o-NB macromers
containing different functionalities at the benzylic position. As
proof-of-concept, the N-hydroxysuccinyl ester macromer was
incorporated into hydrogels and then reacted with phenyl-
alanine. Upon exposure to light (λ = 365 nm, 10 mW/cm2, 10
min), 81.3% of theoretical load of phenylalanine was released
from the gel, demonstrating the utility of these linkers for
incorporating and releasing therapeutics such as peptides and
proteins. We successfully demonstrated the quantifiable
conjugation of a bioactive peptide (GCGYGRGDSPG), an
enzymatically active protein (BSA) and a bioactive growth
factor (TGF-β1) into hydrogels via disulfide exchange and
demonstrated that these biomolecules can be released
controllably from the hydrogels using light. Neither the
incorporation process nor photorelease has any apparent effect
on their bioactivity. This platform provides researchers with an
array of chemistries that should allow for direct conjugation of
nearly any type of therapeutic agent to the linker and its
subsequent controlled release using light. Because light is an
externally controlled trigger, this approach allows precise spatial
and temporal patterning of biological signal within a hydrogel
matrix. Precise control over the delivery of therapeutics is
critical to recapture the complex signaling cascades found in
nature. External control of the temporal and spatial distribution
of different signals may introduce a pathway to engineering
complex tissues.
Figure 3. Expression of GAGs visualized by toluidine blue staining
after three days of culture. (a) Untreated hMSCs only show nuclear
staining (negative control), while (b) hMSCs treated with 10 ng/mL
TGF-β1 (positive control) or (c) 5.2 ng/mL photoreleased TGF-β1
exhibit dark granules in the cell cytoplasm, indicating GAG
production. (d) GAG production per cell, normalized to untreated
hMSCs (- control).
hMSCs treated with the released TGF-β1 produce GAGs
(Figure 3c, observed as dark granules in the cytoplasm) and
appear similar to the positive control (Figure 3b, hMSCs
treated with 10 ng/mL TGF-β1 for three days), while the
untreated hMSCs do not stain with toluidine blue (Figure 3a,
except for the cell nucleus). GAG production was also
measured via dimethylmethylene blue (DMMB) assay and
normalized to the number of cells (measured via PicoGreen
assay; Figure 3d). Despite relatively large error in the
measurements, it is clear that GAG production is greater in
both the positive control and the cells treated with photo-
released TGF-β1. The combination of the differences in
toluidine blue staining and the qualitative differences in GAG
production demonstrate that the sequestered and released
TGF-β1 retains its biological activity and is able to induce
differentiation of hMSCs.
Through examples above, we have demonstrated that this
platform can be used to incorporate and release biomolecules
and therapeutics of various sizes predictably and controllably.
This library of o-NB-containing macromers should allow direct
conjugation of many different functional groups to the
macromer, either before or after hydrogel fabrication. The
acrylate and pyridyldisulfide moieties should react directly with
ASSOCIATED CONTENT
■
S
* Supporting Information
Experimental details for the synthesis of all macromers,
Schemes S1 and S2. This material is available free of charge
via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*Tel.: +1 310 794 6341. Fax: +1 310 794 5956. E-mail:
akasko@ucla.edu.
Author Contributions
†These authors contributed equally to this work.
Notes
The authors declare the following competing financial
interest(s): A.M.K. is a co-inventor on U.S. Patent Application
1206
dx.doi.org/10.1021/bm400169d | Biomacromolecules 2013, 14, 1199−1207

Biomacromolecules
Article
(29) Suttiprasit, P.; Krisdhasima, V.; McGuire, J. J. Colloid Interface
Sci. 1992, 154, 316.
(30) Matsumoto, N. M.; Gonzalez-Toro, D. C.; Chacko, R. T.;
Maynard, H. D.; Thayumanavan, S. Polym. Chem. 2013,
DOI: 10.1039/c3py00085k.
(31) Roberts, A. B.; Sporn, M. B. Peptide Growth Factors and Their
Receptors; Springer: Berlin, 1991.
(32) Daopin, S.; Piez, K. A.; Ogawa, Y.; Davies, D. R. Science 1992,
257, 369.
(33) Johnstone, B.; Hering, T. M.; Caplan, A. I.; Goldberg, V. M.;
Yoo, J. U. Exp. Cell Res. 1998, 238, 265.
No. 11/374,471, which includes compounds described in this
report.
ACKNOWLEDGMENTS
■
Funding for A.M.K. for this work was provided by UCLA
HSSEAS Start-up funds, UCLA/CNSI IRG Seed funding,
Millipore Corporation and the National Institutes of Health
through the NIH Director’s New Innovator Award Program, 1-
DP2-OD008533. H.D.M. thanks the NIH (NIBIB R01 EB
136774-01A1) for funding. Cells employed in this work were
provided by the Texas A&M Health Science Center College of
Medicine Institute for Regenerative Medicine at Scott & White
through a grant from NCRR of the NIH, Grant
#P40RR017447.
(34) Mackay, A. M.; Beck, S. C.; Murphy, J. M.; Barry, F. P.;
Chichester, C. O.; Pittenger, M. F. Tissue Eng. 1998, 4, 415.
REFERENCES
■
(1) Vermonden, T.; Censi, R.; Hennink, W. E. Chem. Rev. 2012, 112,
2853.
(2) Kim, H.; Kim, H. W.; Suh, H. Biomaterials 2003, 24, 4671.
(3) Benoit, D. S. W.; Nuttelman, C. R.; Collins, S. D.; Anseth, K. S.
Biomaterials 2006, 27, 6102.
(4) Zisch, A. H.; Lutolf, M. P.; Ehrbar, M.; Raeber, G. P.; Rizzi, S. C.;
Davies, N.; Schmokel, H.; Bezuidenhout, D.; Djonov, V.; Zilla, P.;
Hubbell, J. A. FASEB J. 2003, 17, 2260.
(5) Wilcox, M.; Viola, R. W.; Johnson, K. W.; Billington, a. P.;
Carpenter, B. K.; Mccray, J. a.; Guzikowski, a. P.; Hess, G. P. J. Org.
Chem. 1990, 55, 1585.
(6) Kloxin, A. M.; Kasko, A. M.; Salinas, C. N.; Anseth, K. S. Science
2009, 324, 59.
(7) Kloxin, A. M.; Tibbitt, M. W.; Kasko, A. M.; Fairbairn, J. A.;
Anseth, K. S. Adv. Mater. 2010, 22, 61.
(8) Sarig-Nadir, O.; Livnat, N.; Zajdman, R.; Shoham, S.; Seliktar, D.
Biophys. J. 2009, 96, 4743.
(9) Tibbitt, M. W.; Kloxin, A. M.; Dyamenahalli, K. U.; Anseth, K. S.
Soft Matter 2010, 6, 5100.
(10) Hahn, M. S.; Miller, J. S.; West, J. L. Adv. Mater. 2006, 18, 2679.
(11) Hahn, M. S.; Taite, L. J.; Moon, J. J.; Rowland, M. C.; Ruffino,
K. A.; West, J. L. Biomaterials 2006, 27, 2519.
(12) Luo, Y.; Shoichet, M. S. Nat. Mater. 2004, 3, 249.
(13) Luo, Y.; Shoichet, M. S. Biomacromolecules 2004, 5, 2315.
(14) Griffin, D. R.; Patterson, J.; Kasko, A. M. Biotechnol. Bioeng.
2010, 107, 1012.
(15) Griffin, D. R.; Kasko, A. M. J. Am. Chem. Soc. 2012, 134, 13103.
(16) Griffin, D. R.; Kasko, A. M. ACS Macro Lett. 2012, 1, 1330.
(17) Bontempo, D.; Heredia, K. L.; Fish, B. A.; Maynard, H. D. J. Am.
Chem. Soc. 2004, 126, 15372.
(18) Mohsen-Nia, M.; Modarress, H.; Rasa, H. J. Chem. Eng. Data
2005, 50, 1662.
(19) Bryant, S. J.; Anseth, K. S. Scaffolding in Tissue Engineering; CRC
Press: Boca Raton, FL, 2006.
(20) Grover, G. N.; Alconcel, S. N. S.; Matsumoto, N. M.; Maynard,
H. D. Macromolecules 2009, 42, 7657.
(21) Zhu, H.; Guo, Z. K.; Jiang, X. X.; Li, H.; Wang, X. Y.; Yao, H. Y.;
Zhang, Y.; Mao, N. Nat. Protoc. 2010, 5, 550.
(22) Hwang, N. S.; Varghese, S.; Elisseef, J. Methods Mol. Biol. 2007,
407, 351.
(23) Holmes, C. P.; Jones, D. G. J. Org. Chem. 1995, 60, 2318.
(24) Burakowska, E.; Zimmerman, S. C.; Haag, R. Small 2009, 5,
2199.
(25) Hermanson, G. T. Bioconjugate Techniques; Academic Press: San
Diego, CA, 2008.
(26) Navath, R. S.; Pabbisetty, K. B.; Hu, L. Q. Tetrahedron Lett.
2006, 47, 389.
(27) Mather, B. D.; Viswanathan, K.; Miller, K. M.; Long, T. E. Prog.
Polym. Sci. 2006, 31, 487.
(28) Schlosser, J. L.; Kerr, J.; Wong, D.; Kasko, A. M. 2013,
manuscript in preparation.
1207
dx.doi.org/10.1021/bm400169d | Biomacromolecules 2013, 14, 1199−1207