Poly(ethylene glycol) (PEG) is a hydrophilic, non-immunogenic
and non-cytotoxic polymer that has found wide-spread application
in the design of biomaterials for e.g. controlled release
of therapeutics and tissue regeneration.1–4 The use of PEG is
particularly attractive as this polymer significantly reduces
protein adsorption and consequently cell adhesion, imparting
“stealth” capability to mask any underlying biomaterial (e.g.
nanoparticles,5–7 core–shell micelles,8,9 polymeric surfaces10
or even hydrogels1,11) from the host’s immune system.5,12,13
From a controlled release perspective, PEG hydrogels have
emerged as potential matrices for release of both small molecule
and macromolecular therapeutics given these inherent advantages
of PEG-based materials in vivo.1,11 However, the use of
PEG hydrogels in such applications has been limited by their
high degree of swelling (and associated limited mechanical
strength) and weak drug-hydrogel interactions that result in
either fast drug release (in the case of hydrophilic drugs) or
poor drug loading (in the case of hydrophobic drugs). Given
that conventional PEG hydrogels are prepared from stepgrowth
polymerization of α,ω functionalized PEG macromonomers
that cross-link via chain ends,14–30 chemical modification
of the hydrogels to, for example, limit swelling or introduce
drug affinity groups to enhance drug-hydrogel interactions is
synthetically challenging, at least without sacrificing potential
cross-linking sites within the hydrogel that can further exacerbate
the challenge of controlling hydrogel swelling.27,29 Most
of the cross-linking reactions used also result in the formation
of non-degradable bonds, making clearance of the hydrogel
following use problematic.31 As such, while some successful
examples of the use of PEG-based hydrogels for deliveringproteins have been reported,32–34 the full potential of using
PEG-based materials for drug delivery has yet to be unlocked.
The weaknesses of PEG in terms of controlled release applications
(i.e. degradability and poor bioavailability of hydrophobic
therapeutics) can be addressed by combining PEG with
hydrophobic, biocompatible, and bioresorbable polymers such
as poly(lactic acid) (PLA), poly(glycolic acid), (PGA) or their
copolymer poly(lactic acid-co-glycolic acid) (PLGA).35,36 The
design of nanoparticle drug delivery vehicles in particular has
benefitted from this approach, wherein PEG-PLA or poly(oligoethylene
glycol methacrylate)-PLA (POEGMA-PLA) block
copolymers can be assembled into micelles or vesicles that can
carry a hydrophobic payload in the hydrophobic PLA core
while evading the host’s immune system via the hydrophilic
PEG corona.37 This approach has also been extended to PEG
hydrogels through the use of diacrylated PLA-b-PEG-b-PLA
cross-linkers38–44 and stereocomplexation between PEG-poly-
(L-lactic acid) (PEG-PLLA) and PEG-poly(D-lactic acid) (PEG-PDLA)
block-copolymers.45–48 Recently, Fan and co-workers combined
both approaches, using stereocomplexed PLLA and PDLA
macromonomers as cross-linkers for hydrogel synthesis.49 As a
result of their controllable physicochemical properties such as
the hydrogel permeability, drug loading, and degradation
rate,39,41 PEG-PLA hydrogels have been investigated as matrices
for controlled release40,50 as well as temporary scaffolds for
tissue engineering.51 However, given that the hydrophobic
PLA/PGA phase often serves as both the hydrophobic drug
depot and the cross-linking site in such hydrogels, independent
tuning of cross-link density, drug affinity, and hydrogel
degradation in such systems is inherently challenging.
Recently, we have reported the preparation of injectable, in
situ covalently cross-linked POEGMA hydrogels that display all
the desired biointerfacial properties of PEG (i.e. protein and
cell repellency, non-toxicity, and minimal inflammatory
responses in vivo).52,53 Hydrogel formation occurs through the
formation of dynamic covalent hydrazone bonds,54,55 which
allows for in vivo gelation as well as hydrolytic degradation and
ultimate clearance of the POEGMA precursors.52 Copolymerization
of oligo(ethylene glycol methacrylate) monomers
(OEGMA) of varying ethylene oxide side chain lengths (n) and/
or (meth)acrylate monomers with various side chain functionalities
allows for facile control over the lower-critical solution
temperature (LCST)56–58 as well as the functionality of the
POEGMA precursors, giving access to POEGMA hydrogels with
a broad range of physiochemical properties and drug affinities
via simple free radical copolymerization.52,53
While these injectable POEGMA hydrogels address many of
the challenges associated with PEG hydrogels (degradability,
independent control over swelling and mechanical properties,
and facile polymer functionalization), hydrogels based on
POEGMA have analogous swelling and interfacial properties to
PEG hydrogels, making them unlikely candidates to address
the issues of fast release of proteins or low uptake of hydrophobic
drugs associated with PEG hydrogels.
Herein, we aim to improve the capacity of POEGMA hydrogels
for drug delivery by functionalizing hydrogel precursor
polymers with PLA via copolymerization of pre-synthesized
oligo (D,L-lactide) macromonomers (OLA)59 with OEGMA
during the polymer precursor synthesis (Scheme 1). Our
approach differs from most found in the literature given that
we do not explicitly use the OLA grafts for the purpose of
cross-linking; instead, cross-linking is driven primarily by
hydrazone bond formation between the hydrazide and aldehyde-
functionalized polymer precursors. As such, the PLA residues
will be (at least partially, within the context of the crosslinked
network formed) free to self-assemble during gelation
via hydrophobic association to form a nanostructured hydrogel
with nanodomains governed by the mole fraction and sidechain
length of the OLA co-monomers. The results show that
the incorporated OLA co-monomers significantly alter the physiochemical
properties (i.e. hydrogel swelling, mechanical
strength and degradation) of the POEGMA hydrogels. Furthermore,
loading and release of bovine serum albumin (BSA), a
model protein which associates with hydrophobic domains,60
showed a strong dependence on the mole fraction of PLA in
the hydrogel, suggesting that functionalized poly(OEGMA-co-
OLA) precursors may offer a versatile route towards the synthesis
of injectable hydrogels with the potential for sustained
release.