hyaluronan-methylcellulose hydrogels for cell and drug ... · recent neuroregenerative approaches...
TRANSCRIPT
Hyaluronan-Methylcellulose Hydrogels for Cell and Drug Delivery to the Injured Central Nervous System
by
Matthew John Caicco
A thesis submitted in conformity with the requirements for the degree of Master’s of Applied Science
Department of Chemical Engineering & Applied Chemistry Institute of Biomaterials & Biomedical Engineering
University of Toronto
© Copyright by Matthew J. Caicco 2012
ii
Hyaluronan-Methylcellulose Hydrogels for Cell and Drug Delivery
to the Injured Central Nervous System
Matthew J. Caicco
Master’s of Applied Science
Department of Chemical Engineering & Applied Chemistry Institute of Biomaterials & Biomedical Engineering
University of Toronto
2012
Abstract
Spinal cord injury and stroke are two devastating neurological events that lack effective clinical
treatments. Recent neuroregenerative approaches involving the delivery of cells or drugs to the
injured tissue have shown promise, but face critical challenges to clinical translation. Herein,
hyaluronan-methylcellulose (HAMC) hydrogels were investigated as a versatile means of
overcoming the challenges facing central nervous system cell and drug delivery. HAMC was
shown to support the viability of encapsulated human umbilical tissue-derived cells,
demonstrating utility as a scaffold for therapeutic cell delivery to the injured spinal cord. In a
drug delivery context, release of the neuroregenerative drug cyclosporin A from the hydrogel
was tunable over 2-28 days and the drug diffused to the stem cell niche in the brain and persisted
for up to 24 days at a stable concentration when the HAMC-based system was implanted epi-
cortically. HAMC is thus a promising tool for emerging neuroregenerative therapies.
iii
Acknowledgments
First and foremost I would like to thank Professor Molly Shoichet for giving me the opportunity
to work in her research group. Her guidance and support was outstanding and I feel quite
privileged to have been a part of her laboratory. I am also extremely grateful for the extensive
assistance I received from Dr. Michael Cooke, Yunafei Wang, and Anup Tuladhar, who
dedicated many an hour in the surgery room and on the cryostat on my behalf. Special thanks are
also extended to Dr. Tasneem Zahir, who got me started in the right direction when I first joined
the lab. Many thanks to Dr. Shawn Owen and Michelle Young for their considerable help with
developing the CsA quantitation method. Finally, I would like to thank the remaining co-authors
on my papers, committee members, and all researchers of the Shoichet Lab for their wide-
ranging assistance with this thesis.
iv
Table of Contents
Acknowledgments .......................................................................................................................... iii
Table of Contents ........................................................................................................................... iv
List of Abbreviations .................................................................................................................... vii
List of Figures .............................................................................................................................. viii
Declaration of Co-Authorship ........................................................................................................ xi
Abstracts of Articles Appearing in Thesis .................................................................................... xii
1 Introduction ................................................................................................................................ 1
1.1 Rationale ............................................................................................................................. 1
1.2 Goal & Hypotheses ............................................................................................................. 2
1.3 Spinal Cord Injury & Stroke: Pathology & Treatment ....................................................... 3
1.3.1 Spinal Cord Injury ................................................................................................... 3
1.3.2 Stroke ...................................................................................................................... 4
1.4 Cell delivery to the injured spinal cord ............................................................................... 5
1.4.1 Biomaterials for cell transplant survival ................................................................. 6
1.5 Drug delivery to stroke-injured brain ................................................................................. 7
1.5.1 Regenerative potential of Cyclosporin A ................................................................ 7
1.5.2 Challenge of drug delivery to the brain .................................................................. 8
1.6 Hydrogels ............................................................................................................................ 9
1.6.1 Hydrogel properties ................................................................................................ 9
1.6.2 Hyaluronan-methylcellulose hydrogels ................................................................ 10
1.6.2.1 Controlling Drug Release from HAMC Hydrogels ................................ 13
1.7 Scope of thesis .................................................................................................................. 14
2 Characterization of hyaluronan-methylcellulose hydrogels for cell delivery to the injured spinal cord ................................................................................................................................ 15
2.1 Introduction ....................................................................................................................... 15
v
2.2 Materials & Methods ........................................................................................................ 16
2.2.1 Material preparation .............................................................................................. 16
2.2.2 Rheological characterization of HAMC blends .................................................... 17
2.2.3 In vitro characterization of hUTC viability in HAMC ......................................... 17
2.2.4 Statistics ................................................................................................................ 18
2.3 Results ............................................................................................................................... 18
2.3.1 Rheological characterization of HAMC without hUTC ....................................... 18
2.3.2 HAMC rheology with hUTC ................................................................................ 20
2.3.3 hUTC viability in HAMC ..................................................................................... 22
2.4 Discussion ......................................................................................................................... 24
2.5 Conclusions ....................................................................................................................... 27
3 A hydrogel composite system for sustained epi-cortical delivery of Cyclosporin A to the brain for the treatment of stroke ............................................................................................... 27
3.1 Introduction ....................................................................................................................... 27
3.2 Materials & Methods ........................................................................................................ 30
3.2.1 Materials ............................................................................................................... 30
3.2.2 Hydrogel preparation ............................................................................................ 30
3.2.3 PLGA microsphere preparation and characterization ........................................... 31
3.2.4 In vitro CsA release from HAMC ......................................................................... 31
3.2.5 Neurosphere assay for CsA activity ...................................................................... 32
3.2.6 Drug delivery device implantation surgeries ........................................................ 33
3.2.7 Analysis of in vivo CsA penetration ..................................................................... 33
3.2.8 CsA detection by LC-MS/MS ............................................................................... 34
3.2.9 Statistics ................................................................................................................ 34
3.3 Results & Discussion ........................................................................................................ 35
3.3.1 CsA release from HAMC in vitro ......................................................................... 35
vi
3.3.2 In vitro bioactivity of CsA released from PLGA microspheres dispersed in HAMC ................................................................................................................... 40
3.3.3 In vivo brain tissue penetration of CsA delivered from composite HAMC system ................................................................................................................... 41
3.4 Conclusions ....................................................................................................................... 44
3.5 Supplemental material ...................................................................................................... 44
4 Discussion & Recommendations for future work .................................................................... 46
4.1 HAMC as a cell delivery vehicle ...................................................................................... 46
4.2 HAMC as a drug delivery vehicle .................................................................................... 50
5 Conclusions .............................................................................................................................. 52
6 References ................................................................................................................................ 54
vii
List of Abbreviations
τy yield stress
aCSF artificial cerebrospinal fluid
BBB blood-brain barrier
bFGF basic fibroblast growth factor
CFSE carboxyfluorescein diacetate succinimidyl ester
CNS central nervous system
CsA cyclosporin A
EGF epidermal growth factor
ELISA enzyme-linked immunosorbent assay
EthD1 ethidium homodimer-1
G’ elastic (storage) modulus
G” viscous (loss) modulus
HA hyaluronan
HAMC hyaluronan-methylcellulose
hUTC human umbilical tissue-derived cells
MC methylcellulose
LC-MS/MS liquid chromatography tandem mass spectrometry
NSPC neural stem/progenitor cell
PLGA poly(lactic-co-glycolic) acid
ROS reactive oxygen species
rPDGF -A recombinant platlet-derived growth factor-A
RSPC retinal stem/progenitor cell
SCI spinal cord injury
SEM scanning electron microscopy
SFM serum-free media
SVZ subventricular zone (subependyma of the lateral ventricles)
tPA tissue plasminogen activator
viii
List of Figures
Figure 1: (A) Cellulose is a linear polymer of glucose where R=H. Alkylation using methylene
chloride yields methylcellulose where R=H and CH3. (B) The repeating disaccharide structure of
hyaluronan. .................................................................................................................................... 11
Figure 2: Shear stress vs. shear rate relationships for five HAMC blends without cells
(0.25/0.25, 0.5/0.5, 0.75/0.75, 1.0/1.0, 1.0/0.75) demonstrate that yield stress (τy) increases with
total polymer content. ................................................................................................................... 19
Figure 3: Gelation point of (A) 0.5/0.5, (B) 0.75/0.75, (C) 1.0/1.0 and (D) 1.0/0.75 HAMC.
Storage (G’) and loss (G”) moduli were measured over time after temperature adjustment from 4
to 37 °C at time zero, simulating in vivo injection. Gelation time and moduli at the gelation point
tended to increase with total polymer content, but all blends gelled in five minutes or less. ....... 20
Figure 4: Comparison of shear stress vs. shear rate relationship for (A) 0.5/0.5, (B) 0.75/0.75,
(C) 1.0/1.0 and (D) 1.0/0.75 HAMC without cells and with 10 million cells per mL. For all
blends, the presence of cells reduces, but does not eliminate, the yield stress. Note that the x-
axes of (C) and (D) are different from those in (A) and (B). ........................................................ 21
Figure 5: Gelation point of 0.75/0.75 HAMC without cells and with 10 million cells per mL.
Storage (G’) and loss (G”) moduli were measured over time after temperature adjustment from 4
to 37 °C at time zero, simulating in vivo injection. The presence of cells slows gelation by
roughly 1.4 minutes. ..................................................................................................................... 22
Figure 6: Confocal reconstructions of CFSE+ hUTC suspensions immediately (day 0) and 3 days
after seeding in 0.5/0.5, 0.75/0.75, 1.0/1.0 and 1.0/0.75 HAMC illustrating random cellular
distribution and inhibition of cellular aggregation and settling. Cells assume a more extended
morphology after 3 days in the gel. Boxed region is 1.7x1.7x1.7 mm. ........................................ 23
Figure 7: Percent live hUTC immediately (day 0) and 3 days after seeding in 0.5/0.5, 0.75/0.75,
1.0/1.0 and 1.0/0.75 HAMC. Panel (A) depicts statistics comparing day 0 and 3 for each blend,
while panel (B) depicts statistics comparing different blends on each day. (n=6 per group, mean
± standard deviation). .................................................................................................................... 24
ix
Figure 8: Schematic of the three methods investigated for the controlled in vitro release of CsA
from HAMC into aCSF. (A) Solubulized CsA, (B) particulate CsA, and (C) PLGA-encapsulated
CsA. .............................................................................................................................................. 35
Figure 9: In vitro cumulative release profiles of CsA from HAMC. (A) Comparison of (•)
solubilized, (▲) particulate, and (■) PLGA-encapsulated CsA release. Dispersion of CsA
particulates into the gel extends release to 7-8 days, while PLGA encapsulation provides
sustained release for 21-25 days. (B) Release of solubilized CsA fits a diffusion-controlled
release model and (C) release of particulate CsA fits a Hixson-Crowell release model. (mean ±
standard deviation, n=3 per release study). ................................................................................... 37
Figure 10: CsA released over 21 days from PLGA microspheres dispersed in HAMC had
equivalent bioactivity to stock CsA as measured by the neurosphere assay. Both conditions were
tested at a CsA concentration of 100 ng/mL and showed significantly greater numbers of
neurospheres than controls in which there was no CsA. (mean ± standard deviation, n=4 trials per
condition, 6 wells per trial). .......................................................................................................... 41
Figure 11: Penetration profiles of CsA in uninjured mouse brain tissue at (A) 6 days, (B) 12
days, (C) 18 days and (D) 24 days post-implant. Data is plotted at midpoint of tissue section (e.g.
the section spanning 500 to 1000 µm is plotted at 750 µm). (E) A constant CsA concentration
was detected in the SVZ region up to 24 days post-implant. (F) CsA remaining in HAMC
decreased over time. Percentages are relative to initial CsA amount in HAMC. (mean ± standard
deviation, n=3 animals per time point). ........................................................................................ 43
Figure 12: Solid CsA particulates (100 µm in size by laser diffraction) were dispersed in HAMC
and the concentration of dissolved drug in the hydrogel was measured over time by absorbance at
229 nm. The dissolved CsA concentration reached a plateau at approximately 45 µg/mL
(compared to 6.6 µg/mL in water [91]), which was interpreted as its solubility limit. The mass
transfer coefficient of dissolution, km, was estimated to be 8×10-5 cm/s via [58]: ........................ 44
Figure 13: (A) CsA-loaded PLGA microspheres had a mean diameter of 25±7 µm by laser
diffraction (Malvern Mastersizer 2000, Worcestershire, UK). (B) SEM image (10 kV
acceleration voltage, 1200X magnification; Hitachi S-2500, Tokyo, Japan) of microspheres
shows smooth surface morphology and spherical shape. ............................................................. 45
x
Figure 14: Schematic for localized and sustained delivery of CsA to the brain. (A) Sagittal and
(B) coronal view of mouse brain with drug delivery system. (C) Drug delivery system in
expanded view. ............................................................................................................................. 46
xi
Declaration of Co-Authorship
The original scientific content of the thesis is comprised of two articles that are submitted to
peer-reviewed internationally recognized journals. In both cases these contributions were
primarily the work of Matthew J. Caicco. The contributions of the co-authors are declared in the
following section in conformity with the requirements for the degree of Master’s of Applied
Science.
xii
Abstracts of Articles Appearing in Thesis
Characterization of hyaluronan-methylcellulose hydrogels for cell delivery to the injured spinal cord
Matthew J. Caicco, Tasneem Zahir, Andrea J. Mothe, Brian G. Ballios, Anthony J. Kihm, Charles H. Tator, Molly S. Shoichet
No effective clinical treatment currently exists for traumatic spinal cord injury. Cell replacement
therapy holds promise for attaining functional repair. Cells may be delivered directly or near to
the injury site; however this strategy requires a delivery vehicle to maintain cell viability. We
have identified an injectable, biocompatible and biodegradable hydrogel scaffold composed of
hyaluronan (HA) and methylcellulose (MC) that may be an effective scaffold for therapeutic cell
delivery. The purpose of the present study was to determine the effects of polymer concentration
on HAMC mechanical strength, gelation time, and cell viability. The yield stress of HAMC, a
measure of mechanical stiffness, was tunable via manipulation of MC and HA content.
Measurement of the elastic and storage moduli as functions of time revealed that HAMC gels in
less than 5 minutes at physiological temperatures. Human umbilical tissue-derived cells
encapsulated in HAMC were homogenously and stably distributed over 3 days in culture and
extended processes into the scaffold. Cell viability was stable over this period in all but the most
concentrated HAMC formulation. Due to its strength-tunability, rapid gelation, and ability to
maintain cell viability, HAMC is promising vehicle for cell delivery and is being tested in
ongoing in vivo studies.
MJC conceived, designed, and executed the experiments and wrote the manuscript. TZ and AJM
aided with hUTC culture. BGB assisted with microscopy. AJK and CHT aided in project
conception. MSS conceived the project and edited the manuscript.
xiii
A hydrogel composite system for sustained epi-cortical delivery of Cyclosporin A to the brain for the treatment of stroke
Matthew J. Caicco, Michael J. Cooke, Yuanfei Wang, Anup Tuladhar, Cindi M. Morshead, Molly S. Shoichet
Stimulation of endogenous neural stem/progenitor cells (NSPCs) with therapeutic factors holds
potential for the treatment of stroke. Cyclosporin A (CsA) is a particularly promising candidate
molecule as it has been shown to act as a survival factor for these cells over a period of weeks
both in vitro and in vivo; however, systemically-delivered CsA compromises the entire immune
system, necessitating sustained localized delivery. Herein we describe a local delivery strategy
for CsA using an epi-cortical hydrogel of hyaluronan-methylcellulose (HAMC) as the drug
reservoir. Three methods of incorporating the drug into the hydrogel (solubilized, particulate, and
poly(lactic-co-glycolic) acid (PLGA) microsphere-encapsulated) resulted in tunable release,
spanning a period of hours to weeks. Importantly, PLGA-encapsulated CsA released from the
hydrogel had equivalent bioactivity to fresh drug as measured by the neurosphere assay.
Moreover, when CsA was released from the PLGA/HAMC composite that was injected on the
cortex of adult mice, CsA was detected in the NSPC niche at a relatively constant concentration
for at least 24 days post-implant. This suggests that this hydrogel composite system may be
promising for the treatment of stroke.
MJCa conceived, designed, and executed the experiments and wrote the manuscript. MJCo, YW,
AT performed surgeries and assisted with preparation of brain tissue for analysis. CMM aided in
project conception. MSS conceived the project and edited the manuscript.
1
1 Introduction
1.1 Rationale
Traumatic spinal cord injury (SCI) and stroke are two debilitating central nervous system (CNS)
insults that have no effective clinical treatments [1, 2]. Therapies aimed at stimulating
regeneration of the damaged CNS tissue hold particular promise as they purport to reverse
effects of injury rather than simply treating its symptoms. Emerging neuroregenerative strategies
center on the concept of using stem cells to repopulate the injured tissue with functional neurons
and glial cells [3]. Stem cells can be used in two ways: 1) exogenous stem cells or stem cell-
derived cell populations can be transplanted at or near the injury site and directly repopulate the
injured tissue; and 2) endogenous stem cells can be stimulated with exogenously delivered
factors to repair the damaged tissue. There are two approaches to the second method, either the
factors can be secreted by a transplanted cell population (a cell delivery approach) or they can be
delivered directly (a drug delivery approach). Both cell and drug delivery-based strategies for
endogenous stem cell stimulation face significant challenges to their success. Cell delivery is
limited by poor cell survival following transplantation into the injured tissue [4], while drug
delivery is limited by poor permeability of systemically delivered drugs across the blood-brain
barrier (BBB) [5].
A potential means of addressing the obstacles to both cell and drug delivery is the use of
hyaluronan-methylcellulose (HAMC) hydrogels. In a cell delivery context, HAMC can act as a
scaffold for the maintenance of cell viability upon transplantation [6, 7]. In a drug delivery
context, HAMC can act as a localized reservoir for the sustained release of therapeutics directly
into the tissue [8-14]. The key properties that make HAMC attractive for both of these uses are
its biocompatibility, biodegradability, injectability through a fine-gauge needle, and in situ
2
gelation. These characteristics allow HAMC to be delivered into or onto CNS tissues in a
minimally invasive manner and persist at the injection site for a defined period of time without
immune reaction before being naturally resorbed by the body. The overall goal of this thesis was
to investigate and adapt HAMC for two specific cell and drug delivery applications: 1) the
delivery of human umbilical tissue derived cells (hUTC) to the injured spinal cord; and 2) the
delivery of cyclosporin A (CsA) to the stroke-injured brain. hUTC are a promising cell
population for SCI cell therapy because they secrete a variety of trophic factors that could
stimulate endogenous stem cells to repair the injured tissue [15]. CsA is attractive because it
enhances the survival of endogenous neural stem/progenitor cells (NSPCs) [16, 17], potentially
increasing the numbers available for regeneration of the stroke-injured brain. Together, this work
will advance cell and drug delivery-based therapies for two devastating CNS injuries.
1.2 Goal & Hypotheses
The overall goal of this thesis is:
To optimize HAMC hydrogels for local delivery of hUTC and CsA to the CNS.
This goal was pursued as follows:
Hypothesis 1: hUTCs can survive in HAMC hydrogels that are optimized for tissue injection.
Objectives to test this hypothesis:
1. Optimize HAMC mechanical properties for tissue injection.
2. Assess hUTC viability in HAMC.
Hypothesis 2: CsA release from HAMC is tunable and delivery to the brain is sustainable.
Objectives to test this hypothesis:
1. Design and test a tunable CsA release strategy from HAMC in vitro.
3
2. Measure drug penetration into brain tissue from optimal formulation in vivo using a
mouse model.
1.3 Spinal Cord Injury & Stroke: Pathology & Treatment
1.3.1 Spinal Cord Injury
SCI results from compression or laceration of the spinal cord and affects 130,000 people each
year worldwide [18]. Compression injuries comprise 70% of clinical cases [19] and are typically
caused by dislocation of a vertebra and pinching of the cord between the anterior and posterior
faces of the vertebral foramen in adjacent vertebrae. The initial primary injury causes
uncontrolled necrotic cell death and a local inflammatory response as neutrophils infiltrate the
damaged tissue through the ruptured blood-spinal cord barrier. CNS microglia become activated
to macrophages and this cell-mediated inflammatory response sees increased production of
cytotoxic proteins, free radicals, and nitric oxides [20]. Ischemia, hypoxia, the release of toxins
by necrotic cells, and free radical formation drives neurons and oligodendrocytes to apoptosis.
These secondary processes persist for weeks, increasing the volume of injured tissue and forming
a cystic cavity. This lesion is eventually isolated from healthy tissue by the glial scar [20].
There is no treatment for spinal cord injury that has shown significant functional recovery.
Delivery of methylprednisolone sodium succinate offers modest functional benefit, but its use is
contentious because of potential serious side effects [21]. In addition, surgical decompression of
the spinal cord post-injury followed by physical rehabilitation is commonly pursued, but
outcomes vary widely and recovery is often limited [22]. Neuroprotection has been the goal of
various treatments currently in development. Examples of neuroprotective strategies include
attenuating the excitotoxic environment, mitigating the inflammatory response, and promoting
neuronal survival. Completed or in-progress trials of neuroprotective molecules include
4
erythropoietin and the ion channel blockers Riluzole and nimodipine [21]. As they target the
acute phases of injury, neuroprotective strategies are limited to new SCI victims and are
generally not applicable to the existing SCI population. In contrast, neuroregenerative strategies
are relevant to all SCI patients as they aim to promote the repair of damaged tissue and
reestablishment of functional connections.
1.3.2 Stroke
Stroke is caused by an occlusion (ischemic stroke) or rupture (hemorrhagic stroke) of cerebral
arteries and permanently disables approximately 5 million people each year worldwide [23]. As
hemorrhagic stroke typically causes immediate death [24] and ischemic stroke comprises the
majority of stroke cases [25], ischemic stroke is a more significant therapeutic target. Following
the primary oxygen shortage and necrotic cell death, programmed cell death occurs in the region
surrounding the core stroke site (penumbra), in a series of events termed the secondary injury.
There are two main mechanisms involved in secondary stroke injury: excitotoxicity and
oxidative stress [26]. The dramatic reduction in available energy caused by the interrupted blood
flow prevents nerve cells from recycling the neurotransmitter glutamate. Abnormally high
glutamate levels trigger a series of intracellular apoptotic signaling events leading to excitotoxic
cell death [27]. Oxidative stress is applied by reactive oxygen species (ROS) that are generated
during reperfusion of the injury site with blood [28]. ROS damage cellular proteins, lipids, and
nucleic acids, which triggers apoptosis. The net pathological outcome of the primary and
secondary injuries depends on the size and location of the stroke lesion, but typically involves
some form of functional deficit.
The only drug approved for the treatment of stroke is recombinant tissue plasminogen activator
(tPA). tPA acts by breaking down blood clots occluding the cerebrovasculature, thereby
5
recovering cerebral blood flow. However, tPA is only effective if administered within three
hours of primary injury [29]. Due to this short therapeutic time window, the number of patients
who might benefit from tPA treatment is small. Furthermore, although effective at reducing
disability, it does not improve mortality [30]. Similar to SCI, recent pre-clinical research efforts
have focused on neuroprotective approaches to the treatment of stroke. One such strategy
involves the delivery of agonists against the glutamate receptors responsible for excitotoxic cell
death [29]. Despite promising pre-clinical data, the approach failed in clinical studies as
functional benefit was not observed. Similarly, delivery of the antioxidant Edaravone displayed
some pre-clinical benefit via inhibition of lipid peroxidation and vascular endothelial cell
damage, but was not successful in the clinic [29]. As is true for SCI, a major inadequacy inherent
to neuroprotective strategies is that they do not combat the widespread neuronal loss caused by
the primary injury. Regeneration of this lost tissue is a required component of any potential cure
for stroke.
1.4 Cell delivery to the injured spinal cord
Cell transplantation therapy has become one of the favored strategies to induce regeneration after
SCI. Either through directly replacing damaged tissue or providing trophic support, exogenous
cells can stimulate neuroregeneration. Examples of cells that have been transplanted in animal
models of SCI include microglia, activated macrophages, olfactory ensheathing glia, Schwann
cells, bone marrow stem cells, hematopoietic stem cells, mesenchymal stem cells, umbilical cord
blood stem cells, embryonic stem cells, adult NSPCs, and glial restricted precursors [31, 32].
hUTC, a proprietary cell type of Advanced Technologies & Regenerative Medicine, hold
particular promise for SCI cell therapy due to the plethora of potentially regeneration-stimulating
growth factors they secrete, such as hepatocyte growth factor, basic fibroblast growth factor,
6
monocyte chemotactic protein 1 and interleukin 8, as well as the neurotrophic factors brain-
derived neurotrophic factor and interleukin 6 [15]. These cells are isolated from post-natal
umbilical cord tissue based upon a specific cell surface marker profile and can be passaged up to
40 times before senescence is observed. However, cells transplanted into or near the lesion site
are subjected to a hostile environment and undergo cell death via multiple mechanisms. This
leads to low levels of cell survival, ranging between 0.2 and 10% [33-35]. For cell
transplantation to be successful, it is essential that cell survival be maintained.
1.4.1 Biomaterials for cell transplant survival
To increase survival, cells have been delivered in biomaterial scaffolds designed to provide a
permissive microenvironment for sustained viability. Hydrogels, which have physical and
chemical properties similar to the natural extracellular matrix, are frequently used as cellular
scaffolds. Natural materials that form hydrogels such as collagen [36], agarose [36], fibrin [37],
chitosan [38], dextran [39], methylcellulose [40], and hyaluronan [41] have been investigated as
cell delivery vehicles. In some cases, the material was modified with a cell-adhesive peptide such
as RGDS, YIGSR, or IKVAV to enhance cell viability. Survival factors such as brain-derived
neurotrophic factor and nerve growth factor have also been added to the matrix. Typically, cells
are found to survive better in the scaffold than in traditional two-dimensional culture conditions.
In addition to the inclusion of cell adhesion molecules and survival factors, the mechanical
properties of the scaffold are critical to its performance. For example, human fetal NSPCs were
found to survive in a 0.25 wt% Puramatrix (a peptide hydrogel) scaffold, but completely die
when the gel concentration was increased to 1 wt% [42]. Consequently, it is vital that hydrogels
for cell delivery have tunable viscoelastic moduli, as the optimal moduli may vary for different
cell types or even the same cell type from different species. The scarcity of strength-tunable
biodegradable hydrogels for cell delivery and the general lack of knowledge surrounding the
7
relationship between mechanical properties and survival motivated the investigation of HAMC
for hUTC delivery.
1.5 Drug delivery to stroke-injured brain
1.5.1 Regenerative potential of Cyclosporin A
CsA is a cyclic undecapeptide that was discovered in the early 1970s by workers at Sandoz
Limited through routine screening of soil extracts for potential pharmacological activity [43]. In
a major breakthrough for immunopharmacology, CsA was found to selectively inhibit antibody
and T-lymphocyte mediated immune responses [44]. Marketed under the brand names
Sandimmune and later Neoral (a microemulsion formulation), CsA significantly improved the
survival of organ allografts in transplant patients.
Research showing that inflammation and degenerative damage in the brain changes NSPC
proliferation, migration, and differentiation led to interest into the effects of immunomodulatory
molecules like CsA on NSPC behaviour [45, 46]. Hunt et al. [16] observed a 1.7-fold increase in
NSPC numbers when cultured for 7 days in the presence of 100 ng/mL CsA compared to CsA-
free controls. Using live cell imaging, the group established that this increase was due to an
enhancement of NSPC survival and not proliferation rate. Furthermore, they observed a 2.6-fold
increase in the numbers of NSPCs derived from adult mice treated for 14 days with 15
mg/kg/day of CsA via a subcutaneously implanted osmotic mini-pump. The group subsequently
reported similar survival benefits for NSPCs derived from the spinal cord [47]. Concurrently
with the Hunt work, Erlandsson et al. [17] treated stroke-injured mice for 18 days with daily
injections of 15 mg/kg CsA and observed a significantly smaller lesion volume in those animals
compared to untreated controls. In addition, stroke-injured animals treated with systemic CsA
(osmotic minipump loaded with 100 µL of 25 mg/mL CsA with replacement every 7 days)
8
displayed some functional recovery as measured by the foot fault test 32 days post-injury.
Consequently, CsA is a promising neuroregenerative molecule for the treatment of stroke.
The exact molecular mechanism through which CsA enhances the survival of NSPCs is currently
unknown. However, a few potential routes have been postulated [47]. These can be divided into
calcineurin-dependent and calcineurin-independent pathways. CsA is known to block the
phosphatase activity of calcineurin via binding to Cyclophilin A, which prevents 1) Bcl-2
Associated Death promoter inhibition of bcl-xL, a pro-survival protein found in mitochondria
and 2) production of free radicals by neuronal Nitric Oxide Synthase. Independent of calcineurin,
CsA blocks the opening of mitochondrial permeability transition pores by binding to Cyclophilin
D, and therefore preventing the release of pro-apoptotic proteins and enhancing cell survival.
Ongoing studies are examining which of these potential pathways dominate the observed pro-
survival effect on NSPCs.
1.5.2 Challenge of drug delivery to the brain
As the brain is isolated from the body by the BBB, drug delivery to the brain is particularly
challenging. The endothelial cells lining the cerebral vessels form tight junctions that permit
small hydrophobic molecules (<600 Da) and nutrients to enter the brain at low levels, but limit
the transport of most foreign substances and large molecules [48]. Administering large systemic
doses will result in a small percentage of drug diffusion across the BBB, but will also likely lead
to toxic or unwanted systemic side effects. For CsA, these include renal dysfunction, tremor,
hirsutism, hypertension, gum hyperplasia and global immunosuppression [49]. Temporary
disruption of the BBB, either mechanically through the use of electrical stimuli or chemically by
the injection of hyperosmotic sugar solutions such as mannitol, has been investigated, but proven
unsuitable for the clinic due to side effects [50]. Therefore, the currently available systemic
9
delivery methods are incapable of delivering drugs like CsA to the brain in a safe and effective
manner.
To overcome the issues inherent to systemic delivery, localized delivery strategies have been
pursued. The two most common local delivery strategies are intracerebroventricular or
intracranial delivery via minipump/catheter infusion and bolus injection. However, both of these
have limitations that render them poorly suited to clinical translation. Bolus injection is a one-
time injection of a drug solution. For therapeutics like CsA, prolonged delivery with a constant
delivery rate is required. Although mini-pump/catheter infusion can provide a constant delivery
rate, it is highly invasive and prone to local inflammation and infection [51]. The lack of a
suitable local delivery strategy for CsA led to the development of the epi-cortical hydrogel
reservoir-based system presented in this work.
1.6 Hydrogels
1.6.1 Hydrogel properties
Hydrogels are polymers that are cross-linked to form a network that is rendered insoluble, but
absorbs large amounts of water and swells [52]. Hydrogels can be classified by their chemistry
(e.g. natural vs. synthetic precursors) or the mechanism of the crosslinking reaction. In hydrogel
synthesis, chain interactions and subsequent network formation can occur via physical, ionic, or
covalent crosslinking [53]. Physical networks form through chain entanglements, hydrogen
bonding, or hydrophobic interactions. Ionic hydrogels are created via multivalent interactions
between macromolecular polymer chains and can be altered by changes in the ionic strength
and/or pH of the system. For biomedical applications, physical and ionic hydrogels are preferred
to covalent hydrogels, as they do not require the addition of a chemical crosslinking agent that
may be cytotoxic.
10
The crosslink density in a hydrogel affects the water content, as measured through the
equilibrium swelling ratio, the mechanical properties such as stiffness [e.g. compressive modulus
or storage modulus (G’)], and the transport of macromolecules, which is related to the mesh size
and ultimate diffusivity [54]. For example, decreasing the number of crosslinks increases average
mesh size, resulting in a higher degree of swelling, a decreased equilibrium modulus, and a
shorter time scale for diffusion of molecules into and out of the gel. Controlling these properties
is essential in both cell and drug delivery applications, as mesh size can limit drug or nutrient
diffusion, while the stiffness of the gel effects the survival of encapsulated cells, as discussed in
section 1.4.1.
1.6.2 Hyaluronan-methylcellulose hydrogels
HAMC (Figure 1) is a hydrogel composed of two biological polymers: hyaluronan (HA) and
methylcellulose (MC). HA is an anionic copolymer of D-glucuronic acid and D-N-
acetylglucosamine, linked via alternating β-1,4 and β-1,3 glycosidic bonds. HA is an ubiquitous
extracellular matrix component found throughout connective, epithelial, and neural tissues in
mammals [55]. The intrinsic biocompatibility of HA makes it particularly well suited to
biomedical applications and is manufactured by microbial fermentation for this purpose. In
contrast to HA, MC, although biocompatible, does not occur naturally and must be synthetically
produced by heating cellulose with a caustic solution followed by treatment with methyl
chloride. In this reaction, some hydroxyl residues on the β(1→4) linked D-glucose units that
compose cellulose are replaced by methoxy groups. Out of a theoretical maximum of three
methoxy substitutions per glucose unit, water soluble MC has a degree of substitution of 1.2-1.8
[56].
11
Figure 1: (A) Cellulose is a linear polymer of glucose where R=H. Alkylation using methylene chloride yields methylcellulose where R=H and CH3. (B) The repeating disaccharide structure of hyaluronan.
MC dissolved in water has inverse thermal gelling properties [57]. As the temperature increases,
hydrogen bonds between the polymer and surrounding solvent break, and hydrophobic junctions
form between polymer chains to produce a gel. Importantly, the gelation temperature decreases
as the concentration of salt in the solution increases. Salt acts to draw water molecules away
from the polymer chains, facilitating the formation of hydrophobic junctions. As mentioned, HA
(manufactured as a sodium salt) is anionic and will thus produce this salting-out effect when
added to an MC solution. Unlike a simple salt, HA also increases the viscosity of the MC
solution, which further facilitates gel formation through an increased number of molecular
entanglements. The net result is a polymer blend that can gel at or below normal body
temperature.
HAMC was originally developed by Gupta, Tator, and Shoichet [8] as a vehicle for localized
delivery of therapeutic agents to the injured spinal cord. The blend they proposed was 2% 1500
kDa HA and 7% 13 kDa MC. It was designed to gel upon injection into the intrathecal space
adjacent to the injury site, providing local and sustained drug release. They found that HAMC
12
was well tolerated in the intrathecal space of rats and in fact provided mild neuroprotection on its
own as evidenced by a reduced inflammatory response and improved functional behaviour
relative to controls. Kang et al. [11] reported that this 2/7 HAMC blend degraded within 4-7 days
in vivo, making it suitable for neuroprotective delivery strategies, but unsuitable for drug
delivery over the 2-4 weeks necessary for neuroregeneration. To address this issue, Baumann et
al. [9] developed a new HAMC formulation composed of 1% 2600 kDa HA and 3% 300 kDa
MC with dispersed drug-loaded poly(lactic-co-glycolic) acid (PLGA) nanoparticles. The higher
molecular weight HA and MC were intended to slow gel degradation while drug encapsulation
within PLGA nanoparticles was intended to slow release. This new composite HAMC blend was
found to be injectable and low swelling, provide satisfactory diffusivity of molecules up to 150
kg/mol, and have significantly slower in vitro degradation suitable for both neuroprotective and
neuroregenerative therapy. It was subsequently shown by Baumann et al. [10] that the composite
HAMC formulation was biocompatible with the intrathecal space of rats. HAMC was first
adapted for the treatment of stroke by Cooke et al. [13] who showed that soluble poly(ethylene
glycol)-modified epidermal growth factor could diffuse out of epi-cortically placed HAMC (1%
1500 kDa HA, 2% 300 kDa MC) and into the brain tissue of stroke-injured mice and elicit
biological effects. Similar results were subsequently observed for erythropoietin [14].
HAMC was first used as a cell delivery vehicle by Ballios et al. [6]. The group screened a series
of low molecular weight HA (1500 kDa), high molecular weight MC (300 kDa) blends for
retinal stem/progenitor cell (RSPC) delivery to the sub-retinal space of the eye. A blend
composed of 0.5% HA and 0.5% MC was selected for in vivo studies based on its injectability
through a 34G needle, its satisfactory gelation speed, and its homogenous and stable cell
distribution. After injection into the sub-retinal space of mice, the HA in the gel was found to
degrade to 10% of initial levels within 3 days, falling to a minimum of 3% after 1 week. The MC
13
showed more persistence within the sub-retinal space and remained at approximately 20% of its
initial value after 7 days. Hsieh et al. [7] adapted this 0.5/0.5 HAMC blend for NSPCs. They
added electrospun fibres to the hydrogel to act as substrates for cell adhesion and to influence the
differentiation profile of the NSPCs cultured within the fiber/hydrogel composite. The inclusion
of these fibers promoted cell survival in vitro and guided cell differentiation similar to HAMC
alone, yet different from media controls, demonstrating the importance of 3D culture to NSPC
behaviour.
1.6.2.1 Controlling Drug Release from HAMC Hydrogels
Although HAMC hydrogels provide spatially controlled release, they do not provide temporally
controlled release on their own. When incorporated into HAMC in a soluble form, drug release is
completed quickly because the polymer chains present a minimal physical barrier. The resulting
drug diffusivity is similar to what is observed in water and can be described by a simple Fickian
diffusion model [9, 11, 13, 14, 58]. One method for extending release only applicable to
hydrophobic drugs like CsA is to disperse solid drug particulates into the hydrogel [58]. The
drug must be hydrophobic so that it does not immediately dissolve into the gel and diffuse into
the release medium. The hydrophobicity of the drug causes its dissolution to be slow, resulting in
sustained release. The rate of release is determined by the solubility limit of the drug in HAMC
and the size of the dispersed drug particulates [58]. Lower solubility and larger particles (less
exposed surface area per unit volume) result in slower release. Another method of extending
release is to encapsulate the drug within PLGA micro- or nanoparticles before dispersion into the
gel. Drug-loaded PLGA particles are one of the few biodegradable polymers approved for
therapeutic use by the FDA and so are widely used in the field of controlled drug delivery [59].
The hydrolytic formation of interconnected pores within the PLGA matrix slowly exposes the
encapsulated drug to the hydrogel, where it dissolves and is released. Drug-loaded PLGA
14
particles embedded in HAMC have been shown to extend the release of protein therapeutics to
up to 54 days [9, 60, 61]
1.7 Scope of thesis
The body of work reported here describes the development of novel cell and drug delivery
approaches for the treatment of SCI and stroke, respectively. The unifying component of these
two strategies is their use of biodegradable, injectable HAMC hydrogels. HAMC was
investigated as a cell delivery scaffold for hUTC and a drug delivery vehicle for CsA. These
original contributions are divided into two chapters:
Chapter 2. Five cell delivery-specific HAMC hydrogels (0.25/0.25, 0.5/0.5, 0.75/0.75, 1.0/1.0,
and 1.0/0.75 HA/MC weight percent) were characterized in terms of their gelation time, yield
stress, and viscoelastic moduli with and without the inclusion of hUTC. The purpose was to
assess the moduli and effect of hUTC on the material. Hydrogels with dispersed hUTC were
imaged for qualitative assessment of cell morphological characteristics as functions of both time
and HA/MC weight percent. Viability of entrapped cells over time was quantified to determine
the optimal HAMC formulation.
Chapter 3. With the aim of sustaining CsA release from a drug delivery formulation of HAMC
(1.4/3.0 HA/MC weight percent) over a period of 3-4 weeks, CsA was dispersed into HAMC in
solubilized, particulate, and PLGA-capsulated forms and drug release from the hydrogel was
quantified as a function of time in vitro. The bioactivity of PLGA-encapsulated CsA released
from HAMC was confirmed. HAMC containing PLGA-encapsulated CsA was implanted on the
cortex of adult mice and the drug concentration in the tissue was quantified as a function of time
and depth from the cortical surface.
15
2 Characterization of hyaluronan-methylcellulose hydrogels for cell delivery to the injured spinal cord
2.1 Introduction
Traumatic compression of the spinal cord is a devastating injury, resulting in significant neural
tissue damage and a dramatic loss of locomotor and sensory function. Unlike the peripheral
nervous system, where injured axons can regenerate and reestablish functional connections,
repair in the central nervous system is very limited. Current treatment options for spinal cord
injury (SCI) are restricted to systemic delivery of methylprednisolone, decompressive surgery,
and physical rehabilitation, all of which result in only minimal functional recovery [62].
An emerging approach for achieving functional repair after SCI is exogenous cell
transplantation. Transplanted cells can replace damaged tissue and provide trophic or cell-contact
mediated support for neuroprotection and regeneration [31, 32]. However, some recent reports
have indicated that neural stem/progenitor cells showed significant cell death after bolus
injection into the spinal cord [35, 63]. Regardless of the cell therapy tested for spinal cord repair,
the delivery vehicle must be selected carefully in order to support extended cell viability and
therapeutic activity. We have identified a novel a biodegradable and injectable hydrogel scaffold
that may be used to deliver and encapsulate cells for spinal cord delivery [6, 7]. Composed of a
physical blend of hyaluronan (HA) and methylcellulose (MC), this HAMC hydrogel provides the
cells with a three-dimensional microenvironment, which is an important factor in enhancing cell
viability [4, 64]. HA is a natural extracellular matrix polysaccharide that has demonstrated
wound-healing properties [65], while MC results in gel formation via thermally-induced physical
crosslinks [57]. Retinal stem/progenitor cells delivered to the sub-retinal space in HAMC were
more evenly distributed than those delivered in traditional saline solutions [6]. Similar results
16
were observed in vitro for neural stem/progenitor cells [7]. Consequently, HAMC possesses
considerable potential as a cell delivery vehicle.
Previous work with a drug delivery formulation of HAMC revealed that the mechanical
properties of the material are strongly dependent upon the concentration of HA and MC used to
formulate the hydrogel [9]. However, cell delivery applications of HAMC have focused almost
exclusively on a single formulation composed of 0.5 wt% MC and 0.5 wt% HA. Since cell
viability in three-dimensional culture conditions is known to be dependent upon scaffold
stiffness [64, 66], our goal was to investigate how cells respond to various concentrations of HA
and MC in the hydrogel. In addition, we were interested in understanding how the presence of
cells impacts the mechanical properties of the material. Human umbilical tissue-derived cells
(hUTC) were studied as they are known to secrete a variety of trophic factors such as hepatocyte
growth factor, basic fibroblast growth factor, monocyte chemotactic protein 1 and interleukin 8,
as well as the neurotrophic factors brain-derived neurotrophic factor and interleukin 6 [15]. In
addition, a small population of hUTC can differentiate to form neurons (TuJ1+ cells) [15]. For
these reasons, transplanted hUTC have the potential to stimulate recovery in the injured spinal
cord.
2.2 Materials & Methods
2.2.1 Material preparation
HA was purchased from Novamatrix (1500 kDa; Drammen, Norway) and MC was purchased
from Shin-Etsu (300 kDa; Tokyo). HA and MC were sterilized via dissolution in ddH2O,
filtration through a 0.22 µm poly(ether sulfone) PES membrane, and lyophilized to recover the
solid polymer. Sterile HAMC was prepared by dissolving HA and MC in hUTC media
(Dulbecco’s Modified Eagle Medium (Gibco) with penicillin-streptomycin (PenStrep, Sigma-
17
Aldrich) and GlutaMAXTM (Gibco)) overnight at 4°C. hUTC (provided by Advanced
Technologies and Regenerative Medicine LLC, ATRM) in a media suspension (or an equivalent
volume of media alone for non-cell controls) were physically mixed into the hydrogel at 1/9 cell
suspension/hydrogel ratio. HAMC blends with the following HA/MC weight percent ratios were
produced: 0.25/0.25, 0.50/0.50, 0.75/0.75, 1.0/0.75, and 1.0/1.0.
2.2.2 Rheological characterization of HAMC blends
All rheological data were collected using a TA Instruments AR1000 rheometer (New Castle, DE,
USA) equipped with a 60 mm, 1° acrylic cone. Temperature was controlled using an integrated
Peltier plate and sample evaporation was minimized using a solvent trap. HAMC yield stress (τy)
was characterized via stress-controlled steady state experiments at 37 °C. To allow for thermal
equilibration, samples were conditioned for 20 minutes at 37 °C prior to shear. Shear rates were
then recorded for shear stresses ranging between 0.01 and 20 Pa. The gelation points of the
HAMC blends were characterized via measurement of the storage (G’) and loss (G”) moduli as
functions of time. To simulate in vivo injection, the temperature of the Peltier plate was changed
from 4 to 37 °C at time zero and the moduli were recorded periodically for 40 minutes at an
angular frequency of 1 Hz and 1% strain (confirmed to lie within the linear viscoelastic regions
of the HAMC blends).
2.2.3 In vitro characterization of hUTC viability in HAMC
Viability of hUTC was studied in the four HAMC blends immediately (day 0) and 3 days after
seeding. Cells were fluorescently labelled using CellTraceTM CFSE dye (Invitrogen) and
ethidium homodimer-1 (EthD1, Invitrogen). The labeled cells were trypsinized and re-suspended
at a concentration of 1×104 cells/500 mL of HAMC. Viability was assayed using confocal
18
imaging (Olympus Fluoview FV1000) and single cell counting, where CFSE+ EthD1- cells were
classified as live and CFSE+ EthD1+ were classified as dead.
2.2.4 Statistics
All statistical analyses were performed using Prism 5.0 (GraphPad Software Inc.). Differences
between two groups were assessed by paired t-tests, while differences between three or more
groups were assessed by one-way ANOVA with Bonferonni correction where appropriate.
Significance levels were indicated by p < 0.05 (*), 0.01 (**), and 0.001 (***).
2.3 Results
2.3.1 Rheological characterization of HAMC without hUTC
Rheological testing was used to characterize the yield stress of five HAMC blends (0.25/0.25,
0.5/0.5, 0.75/0.75, 1.0/0.75, and 1.0/1.0) without the inclusion of cells. Previous work has shown
that 0.5/0.5 HAMC possesses a non-zero yield stress [7], meaning that it will not deform in
response to shear until a certain minimum amount of stress is applied. Yield stress magnitude
was used in this study as a measure of overall hydrogel strength. Figure 2 displays shear stress
vs. shear rate traces for the five HAMC blends without hUTC where the yield stress is given by
the vertical intercept. With the exception of 0.25/0.25 HAMC, yield stress increased with total
polymer content in the hydrogel, ranging from 1.6 Pa for 0.5/0.5 to 4.3 Pa for 1.0/1.0. This
demonstrates that the gel is strengthened upon addition of both MC, which comprises the
physical gel-forming crosslinks, and HA, which enhances gelation via viscosity and salting-out
effects [8]. The zero yield stress of 0.25/0.25 HAMC signified that it cannot resist deformation in
response to shear and thus does not form a gel. Consequently, it was not examined further.
19
Figure 2: Shear stress vs. shear rate relationships for five HAMC blends without cells (0.25/0.25, 0.5/0.5, 0.75/0.75, 1.0/1.0, 1.0/0.75) demonstrate that yield stress (τy) increases with total polymer content.
The gelation points of the HAMC blends without the inclusion of hUTC were
characterized via measurement of the storage (G’) and loss (G”) moduli as functions of time.
The gelation point is defined as the time in which G’ becomes equal to G”. As shown in Figure
3, gelation time and moduli at the gelation point tended to increase with total polymer content.
This means that the gel, although it takes longer to form, is stronger when there is more MC and
HA in the blend, which is in agreement with the yield stress data presented in Figure 2.
Significantly, all blends formed a gel rapidly, as the slowest gelling blend required only 5
minutes to reach its gelation point.
20
Figure 3: Gelation point of (A) 0.5/0.5, (B) 0.75/0.75, (C) 1.0/1.0 and (D) 1.0/0.75 HAMC. Storage (G’) and loss (G”) moduli were measured over time after temperature adjustment from 4 to 37 °C at time zero, simulating in vivo injection. Gelation time and moduli at the gelation point tended to increase with total polymer content, but all blends gelled in five minutes or less.
2.3.2 HAMC rheology with hUTC
As shown in Figure 4, the addition of hUTC (at a loading of 10 million cells per mL) reduced
the yield stress of all four blends. This indicates that dispersion of cells throughout the hydrogel
matrix reduces its strength. In addition, the presence of cells slows gelation, as displayed in
Figure 5 for the 0.75/0.75 formulation. Specifically, it takes approximately 1.4 more minutes for
G’ to intercept G” when hUTC are included in the hydrogel. Although the equilibrium values of
G’ are similar with and without cells, the difference between the equilibrium G’ and G” values is
smaller with the inclusion of cells, which is indicative of a weaker gel and thus corroborates the
yield stress data in Figure 4.
21
Figure 4: Comparison of shear stress vs. shear rate relationship for (A) 0.5/0.5, (B) 0.75/0.75, (C) 1.0/1.0 and (D) 1.0/0.75 HAMC without cells and with 10 million cells per mL. For all blends, the presence of cells reduces, but does not eliminate, the yield stress. Note that the x-axes of (C) and (D) are different from those in (A) and (B).
22
Figure 5: Gelation point of 0.75/0.75 HAMC without cells and with 10 million cells per mL. Storage (G’) and loss (G”) moduli were measured over time after temperature adjustment from 4 to 37 °C at time zero, simulating in vivo injection. The presence of cells slows gelation by roughly 1.4 minutes.
2.3.3 hUTC viability in HAMC
CFSE-labelled hUTC were dispersed in each of the four HAMC blends and their distribution was
studied using confocal reconstructive imaging (Figure 6). hUTC were homogenously distributed
within the HAMC matrix immediately after mixing (day 0) and this distribution was stably
maintained after 3 days of culture in all four blends. Interestingly, the initial rounded
morphology of the cells observed on day 0 transitioned to a more extended morphology after 3
days and the extent of this cellular extension tended to decrease with total polymer content in the
scaffold. As shown in Figure 7, the population of live cells (CFSE+ EthD1-) was similar across
all formulations immediately after seeding (day 0). On day 3, the only significant decrease in live
cells was observed in 1.0/1.0 HAMC both in comparison to 1.0/1.0 on day 0 (Figure 7A) and all
other blends on day 3 (Figure 7B). The maintenance of live cell numbers in 0.5/0.5, 0.75/0.75,
23
and 1.0/0.75 HAMC after 3 days of culture demonstrates their suitability as a scaffold for the
delivery of hUTC.
Figure 6: Confocal reconstructions of CFSE+ hUTC suspensions immediately (day 0) and 3 days after seeding in 0.5/0.5, 0.75/0.75, 1.0/1.0 and 1.0/0.75 HAMC illustrating random cellular distribution and inhibition of cellular aggregation and settling. Cells assume a more extended morphology after 3 days in the gel. Boxed region is 1.7x1.7x1.7 mm.
24
Figure 7: Percent live hUTC immediately (day 0) and 3 days after seeding in 0.5/0.5, 0.75/0.75, 1.0/1.0 and 1.0/0.75 HAMC. Panel (A) depicts statistics comparing day 0 and 3 for each blend, while panel (B) depicts statistics comparing different blends on each day. (n=6 per group, mean ± standard deviation).
2.4 Discussion
The efficacy of therapeutic cell delivery to the injured spinal cord requires an appropriate
delivery vehicle or scaffold to support maximal cell viability and persistence in the injured
tissue. HAMC, a physical hydrogel that is injectable and biodegradable, has been shown to
enhance the survival and distribution of retinal stem/progenitor cells [6] and neural
stem/progenitor cells [7]. Consequently, HAMC is a promising vehicle for the delivery of hUTC
to the spinal cord. However, previous studies have been limited to a single HAMC blend with a
0.5/0.5 HA/MC ratio by weight. Accordingly, the aim of this study was to analyze the effect of
polymer composition on gel mechanical properties and cell survival. The five HA/MC weight
percentages examined were selected because they surround the previously successful 0.5/0.5
wt% blend [6]. Higher polymer concentrations were postulated to increase gel strength, but
possibly hinder cell survival, while less concentrated blends were expected to be more
25
permissive to cellular growth, but weaker mechanically. All gels matched the modulus estimated
for the spinal cord (<300 Pa) [67].
It was shown that gel strength could be tuned through simple adjustment of the MC and HA
contents in the gel. Specifically, increasing the total polymer content in the scaffold resulted in
an increase in yield stress and equilibrium storage modulus. This tunability is significant as the
mechanical properties for optimal cell viability are dependent upon the particular cell population
of interest [68, 69]. Interestingly, dispersion of hUTC into the hydrogels caused a reduction in
the yield stress compared to non-cell controls. One possible explanation is that cells scattered
throughout the polymer matrix physically impede the formation of hydrophobic junctions
between MC chains. However, it should be emphasized that the strength reduction is modest, as
even the weakest HAMC blend remains a gel (i.e., it has a non-zero yield stress) upon the
addition of cells.
In addition to gel strength, gelation time is important to the success of HAMC as a cell delivery
scaffold. Due to the inverse thermal-gelling properties of MC, HAMC acts like a viscous liquid
at ambient temperature (G’ < G”) but gels upon exposure to physiological temperatures (G’ >
G”). Rapid gelation upon injection into the body is thought to positively contribute the longevity
of the scaffold. Although gelation time upon simulated in vivo injection was observed to increase
with total polymer content, all blends were confirmed to gel in five minutes or less, which is
sufficiently fast for hUTC delivery. Addition of cells to the 0.75/0.75 hydrogel delayed gelation,
but only by roughly 1.4 minutes. It should be noted that a difference in testing methodology
resulted in the HAMC gelation times reported herein to be much faster than those reported
previously [6]. In contrast to the observation-based inverted tube test method used in previous
work, the G’/G” time sweep method uses precise quantitation of viscoelastic behaviour to
26
determine the point in which a gel network has formed and so is considered a more accurate
technique [70].
Relevant to the ultimate application of HAMC for cell delivery, we tested whether the difference
in rheological properties of the four HAMC blends impacted the morphology and survival of
encapsulated hUTC. Cells were evenly distributed throughout the gel immediately after
formulation and this was maintained for 3 days in culture. HAMC thus prevents cellular
aggregation and allows the cells to exist in a more natural three-dimensional arrangement.
Another feature important to the viability of anchorage-dependent cells like hUTC is the ability
to extend processes into the scaffold. Adhesion to the substrate in this manner prevents anoikis
and so enhances the survival of transplanted cells [64, 71]. The presence of cell processes
extending into the matrix after 3 days in all HAMC blends reflects positively on the utility of the
hydrogel as a cell delivery scaffold. The mechanism of cell adhesion to the material is undefined;
however, we postulate that HA is mediating the process. HA interacts with cells via the CD44
cell-surface glycoprotein, which is expressed in the majority of mammalian cells, including
hUTC [15]. However, the length and abundance of these extensions tended to decrease with total
polymer content. This could be due to a reduction in gel permeability limiting molecular
transport, but it is known that HAMC formulations as high as 1.0/2.0 wt% permit the rapid
Fickian diffusion of large proteins [13, 14]. Consequently, limitations in waste removal and
nutrient provision are not likely the cause of the reduction in cell extensions at higher polymer
concentrations. It is more likely that the increased stiffness of the hydrogel acts as a physical
barrier to cellular elongation. The consequences of this impediment to the extension of processes
were observed most acutely in 1.0/1.0 HAMC, as live cells (as a percent of total cells on day 0)
dropped from 90.4±8.2% on day 0 to 38.5±9.0% on day 3. Importantly, a significant decrease in
27
live cells on day 3 was not observed in the three other blends, meaning that the stiffness of these
hydrogels was appropriate for the maintenance of cell viability.
2.5 Conclusions
HAMC hydrogels designed for localized, minimally invasive cell delivery to the injured spinal
cord were characterized in terms of mechanical strength, gelation time, and cell viability.
Mechanical strength of the scaffolds as measured through yield stress and elastic modulus was
tunable through simple adjustment to the concentration of constituent polymers and viscous
HAMC solutions gelled rapidly upon heating to physiological temperatures. hUTC cultured in
HAMC were homogenously and stably distributed throughout the scaffold and were able to
adopt an extended, adherent morphology. Live cell numbers were stable over three days in all
blends except the most concentrated, 1.0/1.0 HAMC. Consequently, HAMC holds considerable
potential as a scaffold for cell transplantation therapy. Ongoing studies are examining the
efficacy of 0.5/0.5 HAMC for the delivery of hUTC to the injured rat spinal cord.
3 A hydrogel composite system for sustained epi-cortical delivery of Cyclosporin A to the brain for the treatment of stroke
3.1 Introduction
Stroke is a traumatic neurological event caused by occluded or ruptured cerebral blood vessels
that permanently disables approximately 5 million people every year [23]. Currently, stroke is
treated with either tissue plasminogen activator or an endovascular mechanical device to promote
revascularization [72]; however, there are no clinical therapies capable of repairing damaged
brain tissue and restoring lost function, except through rehabilitation, which has limited benefits,
relying on endogenous repair and plasticity. To further enhance repair, two main strategies have
28
been investigated: stem cell transplantation [73] and endogenous stem cell stimulation. The latter
requires delivery of exogenous factors to stimulate the neural stem/progenitor cells (NSPCs) in
the subventricular zone (SVZ) of the brain to migrate to the injury site and differentiate into
mature cell phenotypes, thereby restoring the lost tissue. Therapeutic factors investigated for this
purpose include epidermal growth factor [74, 75], erythropoietin [74], nerve growth factor [76],
colony stimulating factor [77], basic fibroblast growth factor [76], and cyclosporin A (CsA) [16,
17].
CsA holds particular promise as a neuroprotective and neuroregenerative agent as it has been
shown to act directly on NSPCs to enhance their survival both in vitro and in vivo. When NSPCs
were cultured with CsA in vitro, the number of neurospheres increased compared to controls
[16]. Interestingly, the number of neurospheres derived from adult mice treated for 14 days with
CsA via subcutaneously implanted osmotic mini-pumps was 2.6-fold higher than untreated
controls [16]. Stroke-injured mice treated for 18 days with daily injections of CsA possessed a
significantly smaller lesion volume compared to untreated controls [17]. Moreover, stroke-
injured animals treated with systemic CsA for 32 days displayed some functional recovery, as
measured by the foot fault test [17]. Taken together, these results demonstrate that CsA has
considerable potential as a tissue-regenerative molecule for stroke treatment.
The delivery of drugs to the brain poses a unique challenge due to the blood-brain barrier (BBB),
which is comprised of tight junctions formed by endothelial cells lining the cerebrovasculature
that limit the transport of molecules. While CsA can cross the BBB, its diffusion is attenuated
[78], requiring very high doses of CsA to achieve therapeutically relevant quantities in the brain
via systemic administration. Due to the potentially deleterious side effects associated with these
high doses, including undesirable global immunosuppression, CsA requires localized delivery if
29
it is to be used to treat stroke. However, currently available local delivery strategies, such as
intracerebroventricular or intracranial delivery via minipump/catheter infusion or bolus injection,
are either highly invasive or incapable of delivering a sustained drug dose. Since CsA requires
sustained delivery over a period of weeks to act on NSPCs [16, 17] and existing systems are
either highly invasive or unsustained, a novel, local delivery strategy is required.
To circumvent the blood-brain barrier and sustain delivery in a minimally invasive manner, a
drug-loaded hydrogel composite has been proposed [8, 13, 14]. Comprised of polymeric particles
dispersed in a physically cross-linked blend of hyaluronan (HA) and methylcellulose (MC), the
HAMC composite is injected on the cortical surface and acts as a reservoir for the controlled
release of therapeutics. The hydrogel is bioresorbable, injectable through a fine-gauge needle,
and fast gelling at physiological temperatures [8]. Soluble epidermal growth factor modified with
poly(ethylene glycol) [13] and erythropoietin [14] were shown to penetrate through the ischemic
cortex and reach with SVZ when delivered epi-cortically from HAMC. However, delivery of a
soluble drug from HAMC is governed by Fickian diffusion and so sustained release is unlikely to
be achieved from the hydrogel alone. For hydrophobic drugs, dispersion of solid drug
particulates into the gel that slowly dissolve can yield sustained release, typically over 7 days
[58]. In addition, encapsulation of the drug within poly(lactic-co-glycolic acid) (PLGA) particles
prior to dispersion into the gel has been shown to increase the duration of release to a period of
weeks to months [9, 60, 61].
Here we designed a HAMC hydrogel capable of releasing bioactive CsA for period of 3-4 weeks.
The in vitro release profiles of soluble, particulate, and PLGA-encapsulated CsA were compared
and the PLGA-encapsulation method was found to yield the longest duration of release.
Importantly, the bioactivity of the PLGA-encapsulated CsA was equivalent to free drug as
30
measured by the neurospehere assay. HAMC containing PLGA-encapsulated CsA was injected
onto the cortical surface of mice and the drug was detected at the SVZ up to 24 days post-
implant. This novel biomaterial provides local, sustained, and minimally invasive release of a
promising molecule for the treatment of stroke.
3.2 Materials & Methods
3.2.1 Materials
1.4-1.8×106 g/mol sodium hyaluronate (HA) was purchased from NovaMatrix (Sandvika,
Norway). 3.4×105 g/mol methylcellulose (MC) was obtained from Shin Etsu (Chiyoda-ku,
Tokyo, Japan). Cyclosporin A (CsA) (>99% purity) and internal standard tacrolimus (>99%
purity) were purchased from LC Laboratories (Woburn, USA). HPLC grade dichloromethane
(DCM), acetonitrile, and ethanol were supplied by Caledon Labs (Georgetown, CA). Artificial
cerebrospinal fluid (aCSF) was formulated as previously described [8] with distilled and
deionized water (ddH2O) prepared from a Millipore Milli-RO 10 Plus and Milli-Q UF Plus at 10
MΩ resistivity (Millipore, Bedford, USA). Acid-terminated poly(D,L-lactic-co-glycolic) acid
(PLGA) 50:50 of inherent viscosity 0.16-0.24 dL/g, poly(vinyl alcohol) (PVA) (Mn 30,000-
70,000), ammonium acetate (>99% purity), and all other reagents were purchased from Sigma-
Aldrich (Oakville, CA) and used as received unless specified otherwise.
3.2.2 Hydrogel preparation
HAMC hydrogels were prepared through the physical blending of hyaluronan and methyl
cellulose in aCSF for a final composition of 1.4 wt% HA and 3 wt% MC. MC and HA were
sequentially dispersed in aCSF using a dual asymmetric centrifugal mixer (Flacktek Inc.,
Landrum, USA) and left to dissolve overnight at 4 °C. For sterile hydrogels used in animal
studies, MC and HA were dissolved in ddH2O, sterile filtered, and lyophilized (Labconco,
31
Kansas City, USA) under sterile conditions. The resulting sterile polymers were kept at 4 °C
until use.
3.2.3 PLGA microsphere preparation and characterization
CsA-loaded PLGA microspheres were prepared using an oil/water emulsion solvent evaporation
technique. An organic phase consisting of 0.9 mL DCM, 120 mg PLGA and 12 mg CsA was
added to an 18 mL aqueous phase containing 10 mg/mL PVA. The emulsion was formed through
homogenization (Kinematica, Bohemia, USA) on ice for 60s at 4300 rpm. The emulsion was
then added to 150 mL of 1 mg/mL PVA in water and stirred gently for 3 h at room temperature.
The hardened microspheres were collected and washed by centrifugation, lyophilized, and stored
at -20°C until use. Microspheres used in vivo were sterilized by gamma irradiation.
Microsphere size was measured using laser diffraction (Malvern Mastersizer 2000,
Worcestershire, UK) and surface morphology was examined using scanning electron microscopy
(SEM). Drug loading was defined as the CsA mass per mg of particles, while encapsulation
efficiency is the measured drug loading of the particles divided by the theoretical maximum drug
loading. To determine CsA encapsulation efficiency, a known mass of particles was dissolved in
1 mL of acetonitrile and the resulting solution was analyzed for CsA content.
3.2.4 In vitro CsA release from HAMC
CsA release was quantified from three types of HAMC formulation: (1) HAMC containing
solubilized CsA; (2) HAMC containing solid CsA particulates; and (3) HAMC containing CsA-
loaded PLGA microspheres. For (1), an initial CsA solution in acetonitrile was prepared. For (2),
an initial particulate dispersion was produced by mixing CsA powder into 0.5 wt% MC solution.
For (3), CsA-loaded PLGA microspheres were added to aCSF and dispersed via sonication for 1
min at 26 W and 20 kHz. 10 µL of the solubilized CsA solution, particulate CsA dispersion, or
32
PLGA particle dispersion was added to the bottom of 2 mL eppendorf tube. 90 µL of HAMC
was then added to the tube and mixed into the CsA solution/dispersion using the dual
asymmetric centrifugal mixer, resulting in a 100 µL flat drug-loaded HAMC disk at the bottom
of the tube. The HAMC was allowed to gel for 30 min at 37 °C. At time zero, 900 µL of aCSF
was added to the tube. The aCSF was removed and replaced at various time points and analyzed
for CsA content. All release studies were performed in triplicate and the cumulative release is
expressed as mean ± standard deviation.
3.2.5 Neurosphere assay for CsA activity
All experiments were carried out in accordance with the Guide to the Care and Use of
Experimental Animals developed by the Canadian Council on Animal Care and approved by the
Animal Care Committee at the University of Toronto. NSPCs were isolated by dissection of the
forebrain subependyma of adult male C57BL/6 mice (9-11 weeks old, 25-30 g; Charles River,
CA) as previously described [79]. Briefly, tissue was digested with enzymes (1.33 mg/mL
trypsin, 0.67 mg/mL hyal- uronidase, and 0.2 mg/mL kynurenic acid; all from Sigma-Aldrich)
for 40 min at 37°C. Enzyme activity was inhibited with 0.67 mg/mL trypsin inhibitor (Roche
Diagnostics), and the tissue was mechanically dissociated into a single-cell suspension. For all
conditions cells were plated at clonal density (5 cells/µL) [80] in 24-well polystyrene plates
(VWR Scientific) with serum-free medium (SFM) supplemented with epidermal growth factor
(EGF) (20 ng/mL; Sigma-Aldrich), basic fibroblast growth factor (bFGF) (10 ng/mL; Sigma-
Aldrich), heparin (7.35 ng/mL; Sigma-Aldrich), and 1% penicillin/streptomycin (Invitrogen).
CsA dissolved in 1:1 ethanol:EGF, bFGF, and heparin-supplemented SFM was added to the
cultures for a final concentration of 100 ng/mL [16]. The CsA source was either stock CsA
powder or CsA released from the drug delivery system. To prepare the latter source, release
samples spanning a 21-day period were pooled and the combined sample was analyzed for CsA
33
content. The solution was then lyophilized and re-dissolved in 1:1 ethanol:growth factor-
supplemented SFM for use in the neurosphere assay.
3.2.6 Drug delivery device implantation surgeries
The drug delivery system was spatially localized on the brain cortex of 9-11 week old C57BL/6
mice as previously described [13, 14]. Briefly, anesthetized animals had a burr hole drilled into
the skull at the coordinates 2.25 lateral to the midline and 0.6 anterior to Bregma and the exposed
dura was pierced using a 26 G needle. A polycarbonate disk with a 2 mm opening was fixed over
the burr hole and 3 µL of HAMC containing CsA-loaded PLGA microspheres was placed in the
central opening in direct contact with the brain cortical surface. A second disk without an
opening was fixed above the first disk and the skin was sutured over the disk system (Figure 14).
3.2.7 Analysis of in vivo CsA penetration
Animals were sacrificed 1, 6, 12, 18, and 24 days post-implantation and the drug delivery device
containing HAMC was retrieved. The device was placed into 1.5 mL of acetonitrile and agitated
overnight to extract any remaining CsA. Brains were snap frozen with CO2(s) cooled isopentane
and stored at -80 °C. Three 1 mm coronal slices were prepared using a McIlwain tissue chopper
(Mickle Laboratory Engineering Company, Surrey, UK) at the implant site and rostral and
caudal to the implant site. Dorsal-ventral sections (0.5 mm) were then obtained from each
coronal slice using a Leica CM3050S cryostat system operating at -18 °C. Sections at the same
depth from the cortical surface were combined in 2 mL polystyrene microtubes and
homogenized with 1.0 mm diameter zirconia beads in 120 µL of ethanol for analysis of CsA
content.
34
3.2.8 CsA detection by LC-MS/MS
Liquid chromatography tandem mass spectrometry (LC-MS/MS) was used to quantitate CsA in
release study samples and tissue homogenates. To prepare samples for analysis, 100 µL of
release sample or homogenate was first mixed with 200 µL of acetonitrile containing internal
standard. The resulting mixture was centrifuged at 14,000 rpm for 12 min to remove precipitated
proteins. A Sciex API4000 triple quadrupole mass spectrometer (Ottawa, CA) fitted with an
electrospray ionization interface was used for all analyses. The instrument was operated in
electrospray positive ionization mode and was coupled to an Agilent 1100 capillary LC system
(Mississauga, CA). The separation of CsA and internal standard was performed using a
Spherisorb CN column (30 mm x 4.6 mm, 5 µm) (Waters, Milford, USA) with a mobile phase
composed of 65% aqueous acetonitrile containing 2 mM ammonium acetate and 0.1% (v/v)
formic acid operating at a flow rate of 1 mL/min and a sample injection volume of 10 µL. Both
compounds eluted in less than 1 min and a total cycle time of 2.5 minutes was achieved.
Quantitation was performed using multiple reaction monitoring of the ammonium-adduct
transition masses of CsA (m/z 1220 à 1202) and internal standard tacrolimus (m/z 822 à 768).
Instrument parameters were optimized for the simultaneous detection of both the drug and
internal standard. Calibration curves were established using standard samples at CsA
concentrations ranging from 100 ng/mL to 1 ng/mL with an internal standard concentration of 10
ng/mL. The coefficient of determination (r2) from a 1/x-weighted least squares linear regression
was found to be 0.999.
3.2.9 Statistics
All statistical analyses were performed using Prism 5.0 (GraphPad Software Inc.). Differences
between three or more groups were assessed by one-way ANOVA with Bonferonni correction.
35
Significance levels were indicated by Significance levels were indicated by p < 0.05 (*), p < 0.01
(**), and p < 0.001 (***).
3.3 Results & Discussion
3.3.1 CsA release from HAMC in vitro
To control the release of CsA from HAMC, the drug was incorporated in the hydrogel matrix in
three forms (Figure 8): solubilized, particulate, and PLGA-encapsulated.
Figure 8: Schematic of the three methods investigated for the controlled in vitro release of CsA from HAMC into aCSF. (A) Solubulized CsA, (B) particulate CsA, and (C) PLGA-encapsulated CsA.
Solubilized CsA (Figure 8A) was predicted to diffuse out of the gel rapidly [13, 14, 58], while
particulate (Figure 8B) and PLGA-encapsulated (Figure 8C) CsA were expected to result in
extended release profiles [9, 58, 60, 61]. To test these hypotheses, HAMC containing CsA was
injected onto the bottom of microcentrifuge tubes forming cylindrical discs (0.37 cm in
thickness) with a planar surface. aCSF was placed on top of the hydrogel and removed and
replaced for analysis of CsA content at various time points. As shown in Figure 9A, solubilized
CsA was confirmed to diffuse out of the gel relatively quickly, reaching 100% cumulative
36
release in 2 days. Diffusion-controlled release of a drug source from a planar geometry can be
estimated by the following analytical approximation [81]:
Mt
M!
=2L
DA
!" t0.5 (1)
where Mt/M∞ is the fraction of drug molecules released from the hydrogel at time t, DA is the
diffusivity of the drug in the matrix, and L is the scaffold thickness. Using a fitted DA value of
2.6×10-6 cm2/s, which is characteristic of the diffusion of small molecules, the proportionality to
the square root of time was maintained for the first 50-55% of release (Figure 9B). This
indicates that diffusion was the dominant release mechanism, as was found for other soluble
drugs in previous work with HAMC [11, 58]. Depletion of drug in the hydrogel results in a
diminished concentration gradient and driving force that slows the latter stages of release
compared to the predictions of Eq. (1), a result that is also consistent with previous findings [58].
37
Figure 9: In vitro cumulative release profiles of CsA from HAMC. (A) Comparison of (•) solubilized, (▲) particulate, and (■) PLGA-encapsulated CsA release. Dispersion of CsA particulates into the gel extends release to 7-8 days, while PLGA encapsulation provides sustained release for 21-25 days. (B) Release of solubilized CsA fits a diffusion-controlled release model and (C) release of particulate CsA fits a Hixson-Crowell release model. (mean ± standard deviation, n=3 per release study).
When CsA was dispersed into HAMC in particulate form, its release was significantly slower
than the solubilized formulation. The particulate dispersion resulted in sustained release for 7-10
days (Figure 9A), as only a fraction of the total CsA is dissolved and thus able to diffuse out of
38
the gel at given time [58]. Since the dispersed CsA particulates were relatively large (100 µm in
diameter by laser diffraction), it was postulated that the release profile was governed by the slow
dissolution of drug particulates and not diffusion of the drug out of the gel matrix. The following
dimensionless number (ξ) represents the ratio of the characteristic times of these two processes
[82]:
! =kmnpRi
2L2
DA
(2)
where km is the mass transfer coefficient of CsA dissolution in HAMC (8×10-5 cm/s, see
Supplementary Fig. 1S), np is the number of particulates per unit volume in the gel (382 cm-3),
and Ri is the initial particulate radius (50 µm). Since the calculated ξ value of 0.042 is less than
one, diffusion is indeed much faster than dissolution. Consequently, it can be assumed that the
concentration of CsA in the gel matrix at any given time is negligible compared to the saturation
concentration of the drug and release should follow a Hixson-Crowell profile [58, 83]:
Mt
M!
=1" 1" kmCsat
!Rit
#
$%
&
'(
3
(3)
where Csat is the saturation concentration of CsA in HAMC (45 µg/mL, see Figure 12) and ρ is
the density of the drug particulates (~1 g/cm3). As shown in Figure 9C, the experimental data is
in close agreement with the Hixson-Crowell prediction, further indicating that drug release from
the dispersed CsA particulates is dissolution-controlled. Interestingly, this effect of particulate
dissolution controlling release is enhanced by the presence of methylcellulose (MC) in HAMC,
which promotes the solubilization of hydrophobic molecules [58].
To extend release beyond the 7-10 days of the particulate dispersion, CsA was encapsulated
within PLGA microspheres prior to incorporation into HAMC. Drug-loaded PLGA microspheres
39
are widely used in the field of controlled drug delivery because they are one of the few
biodegradable polymers approved for therapeutic use by the FDA [59]. In these systems,
sustained release results from drug diffusion through pores in the polymer matrix formed by
degradation of PLGA and dissolution of entrapped drug. CsA-loaded PLGA microspheres were
synthesized with a mean diameter of 25±7 µm (see Figure 13) and measured drug loading of 71
µg CsA per mg microspheres. As shown in Figure 9A, sustained release of CsA from PLGA
microspheres in HAMC was achieved for 21-25 days. Interestingly, the initial burst release
characteristic of PLGA particles [84, 85] was nearly non-existent. This attenuation of burst
release from PLGA particles when dispersed in HAMC was previously reported for encapsulated
α-chymotrypsin [9], anti-NogoA [60], and neurotrophin-3 [61]. It was postulated previously [61]
that two possible mechanisms could be causing this behaviour: (1) a reduced degradation rate of
the PLGA in the particles when embedded in HAMC, resulting in an altered release profile; or
(2) absorption of MC to the surface of the particles, resulting in reduced diffusion across the
PLGA-hydrogel boundary and an altered release profile. However, it was found via gel
permeation chromatography studies that PLGA degradation was unaffected by the presence of
HAMC [61]. Consequently, it was suggested that the formation of a diffusive barrier via
interaction between hydrophobic MC and PLGA at the hydrogel–particle interface is responsible
for the low burst and sustained release. In addition, mathematical modeling of release as a
sequential process whereby drug first diffuses of the bulk-eroding PLGA particles then diffuses
out of the HAMC in a Fickian manner failed to accurately predict release from the composite
system [60, 61], suggesting that diffusion through the PLGA particles and HAMC are not
distinct processes and further supporting the MC-PLGA diffusive barrier mechanism.
Through incorporation of CsA into the HAMC matrix in three distinct forms, a spectrum of
release profiles were obtained spanning a period of hours to a period of weeks. Solubilized CsA
40
was released from the gel rapidly in a diffusion-controlled manner, particulate CsA resulted in
slower dissolution-controlled release, and PLGA-encapsulated CsA extended release even
further out to 3-4 weeks. As CsA must be delivered over this longer timescale to have a potential
therapeutic benefit for the treatment of stroke [16, 17], the PLGA encapsulation system was the
only formulation examined further in terms of bioactivity and brain tissue penetration.
3.3.2 In vitro bioactivity of CsA released from PLGA microspheres dispersed in HAMC
The bioactivity of CsA released from PLGA microspheres dispersed in HAMC was assessed
using the neurosphere assay [16, 86]. In this assay, single cells isolated from the forebrain
subependyma of adult mice were cultured for 7 days in vitro and NSPCs formed structures
termed neurospheres during this period. These neurospheres were dissociated and re-plated in the
presence or absence of CsA. When cultured in the presence of CsA, the total numbers of
neurospheres that form is enhanced, representing the pro-survival effect that CsA exerts on
NSPCs (Figure 10), which is consistent with previous reports [16]. The fold-increase in the
number of neurospheres is the same when cells are cultured in the presence of stock CsA and
CsA released from PLGA microspheres dispersed in HAMC. This indicates that the PLGA-
encapsulated CsA had equivalent bioactivity to stock CsA. Although CsA is a polypeptide, it
contains no secondary or tertiary structure and so is not susceptible to the degradation often
experienced by proteins during PLGA particle synthesis [87]. Consequently, maintenance of
bioactivity post-encapsulation and release was expected.
41
Figure 10: CsA released over 21 days from PLGA microspheres dispersed in HAMC had equivalent bioactivity to stock CsA as measured by the neurosphere assay. Both conditions were tested at a CsA concentration of 100 ng/mL and showed significantly greater numbers of neurospheres than controls in which there was no CsA. (mean ± standard deviation, n=4 trials per condition, 6 wells per trial).
3.3.3 In vivo brain tissue penetration of CsA delivered from composite HAMC system
The ability of the drug delivery system to deliver a sustained dose that penetrates the brain tissue
was investigated using a mouse model. HAMC containing PLGA-encapsulated CsA was injected
on the cortex of adult mice using the device [13, 14] depicted in Figure 14. In contrast to other
local delivery systems such as catheter/minipumps and bolus injection [88, 89], this epi-cortical
delivery strategy avoids the trauma and infection observed when inserting cannulas and needles
directly into the brain tissue. Moreover, local delivery circumvents the blood-brain barrier and
avoids the large systemic doses required to get even a small amount of CsA into the brain;
however, drug penetration into brain tissue from the cortical surface can be limited by rapid
elimination [88]. Consequently, it was critical to determine if CsA could diffuse out of the
42
composite delivery vehicle and penetrate the brain tissue to the neural stem cells in the SVZ
(>1500 µm below the cortical surface at the chosen coordinates [90]).
To this end, tissue penetration profiles were quantified at 6, 12, 18, and 24 days post-implant. Six
500 µm sequential tissue sections were prepared ventral to the cortical surface and the
concentration of CsA in the tissue was measured by LC-MS/MS. As shown in Figure 11A-D,
the CsA concentration was highest closest to the cortical surface and decreased with depth.
Importantly, CsA was detectable out to 3000 µm and at all time points tested post-implantation.
Summation of the CsA content at the depth the SVZ or below revealed a relatively constant drug
concentration over the 24-day time period (Figure 11E). This is critical for the therapeutic
benefit of the system, as a constant concentration of CsA over a prolonged period of time is
required to stimulate NSPCs [16, 17]. Additionally, the HAMC implant was extracted at each
time point and analyzed for CsA content. The amount of CsA remaining decreased over time
(Figure 11F), as expected from diffusion of drug out of the hydrogel and into the tissue. The
total amount of CsA detected at each time point represented a very small fraction (<0.01%) of
the total drug initially loaded into the implant, which is consistent with rapid clearance of small
molecule and protein drugs from the brain [88].
43
18-Day
0 1000 2000 30000
100
200
300
Depth Below Cortical Surface (µm)
ng C
sA p
er m
L tis
sue
6-Day
0 1000 2000 30000
100
200
300
Depth Below Cortical Surface (µm)
ng C
sA p
er m
L tis
sue
24-Day
0 1000 2000 30000
100
200
300
Depth Below Cortical Surface (µm)
ng C
sA p
er m
L tis
sue
12-Day
0 1000 2000 30000
100
200
300
Depth Below Cortical Surface (µm)
ng C
sA p
er m
L tis
sue
CsA >1500 µm below cortical surface
0 10 20 300
20
40
60
80
100
Time since Implantation (days)
ng C
sA p
er m
L tis
sue
CsA Remaining in HAMC
0 10 20 300
5000
10000
15000
Time since Implantation (days)
CsA
(ng)
59%52%
34%28%
100%
A
C
B
D
E F
Figure 11: Penetration profiles of CsA in uninjured mouse brain tissue at (A) 6 days, (B) 12 days, (C) 18 days and (D) 24 days post-implant. Data is plotted at midpoint of tissue section (e.g. the section spanning 500 to 1000 µm is plotted at 750 µm). (E) A constant CsA concentration was detected in the SVZ region up to 24 days post-implant. (F) CsA remaining in HAMC decreased over time. Percentages are relative to initial CsA amount in HAMC. (mean ± standard deviation, n=3 animals per time point).
44
3.4 Conclusions
Herein we developed a novel method for the localized and sustained epi-cortical delivery of CsA
for the potential treatment of stroke. CsA release from the HAMC hydrogel system was tunable
via the manner in which the drug was incorporated into the gel. Solubilized CsA yielded release
on the order of hours, while particulate CsA extended release to days and PLGA microsphere-
encapsulated CsA sustained release for a period of 3-4 weeks, a clinically relevant timescale.
PLGA-encapsulated CsA released from the system was found to be bioactive and capable of
penetrating to the SVZ of mice at a stable concentration over a 24-day period. Thus, this
hydrogel composite system may be useful for the treatment of stroke.
3.5 Supplemental material
0 2 4 6 80
20
40
60
Time (days)
CsA
Con
cent
ratio
n (µ
g/m
L)
Figure 12: Solid CsA particulates (100 µm in size by laser diffraction) were dispersed in HAMC and the concentration of dissolved drug in the hydrogel was measured over time by absorbance at 229 nm. The dissolved CsA concentration reached a plateau at approximately 45 µg/mL (compared to 6.6 µg/mL in water [91]), which was interpreted as its solubility limit. The mass transfer coefficient of dissolution, km, was estimated to be 8×10-5 cm/s via [58]:
dCA
dt~ kmaV
Csat (1S)
45
where a is the total surface area of the 100-µm CsA particulates (0.06 cm2), V is the HAMC volume (1 cm3), CA
Sat is the saturation concentration of CsA in MC (45 µg/mL), and dCA/dt is the approximate slope of the dissolution plot prior to the plateau (2.2×10-4 µg/(cm3 s)).
Figure 13: (A) CsA-loaded PLGA microspheres had a mean diameter of 25±7 µm by laser diffraction (Malvern Mastersizer 2000, Worcestershire, UK). (B) SEM image (10 kV acceleration voltage, 1200X magnification; Hitachi S-2500, Tokyo, Japan) of microspheres shows smooth surface morphology and spherical shape.
46
Figure 14: Schematic for localized and sustained delivery of CsA to the brain. (A) Sagittal and (B) coronal view of mouse brain with drug delivery system. (C) Drug delivery system in expanded view.
4 Discussion & Recommendations for future work
4.1 HAMC as a cell delivery vehicle
Therapeutic cell delivery to the injured spinal cord is currently limited by poor survival of the
transplanted cells [92]. To increase survival, cells can be delivered in a biomaterial scaffold
designed to provide them with a microenvironment that is more permissive to their viability.
Bioresorabable, injectable HAMC hydrogels have been shown to enhance the survival and
distribution of retinal stem/progenitor cells [6] and neural stem/progenitor cells [7]. However,
very little was previously known about the interplay between the cells and the mechanical
properties of the material. Herein, it was shown that yield stress, elastic modulus, and gelation
time was tunable via adjustment of the concentration of HA and MC used to formulate the gel.
Moreover, the addition of hUTC to the hydrogel, a promising cell type for SCI cell therapy, was
47
shown to only slightly attenuate the mechanical properties and gelation speed of the hydrogel.
Importantly, the morphology and survival of encapsulated hUTC were examined in relation to
the mechanical properties of the HAMC blends. Cells were evenly distributed throughout all gels
immediately after formulation and this was maintained for 3 days in culture. Additionally, on day
3, the only significant decrease in live cells was observed in 1.0/1.0 HAMC both in comparison
to 1.0/1.0 on day 0 and all other blends on day 3. The maintenance of live cell numbers in
0.5/0.5, 0.75/0.75, and 1.0/0.75 HAMC after 3 days of culture demonstrated their suitability as a
scaffold for the delivery of hUTC.
CFSE (carboxyfluorescein diacetate succinimidyl ester) was used to visualize the hUTC within
the HAMC matrix. CFSE passively diffuses into cells and is colorless and non-fluorescent until
the acetate groups are cleaved by intracellular esterases to yield highly fluorescent
carboxyfluorescein succinimidyl ester. The succinimidyl ester group reacts with intracellular
amines forming fluorescent conjugates (492 nm excitation peak, 517 nm emission peak) that are
retained in the cell throughout development and meiosis and are inherited by daughter cells. To
visualize dead or dying cells, EthD1 (ethidium homodimer-1) labeling was used. EthD1 can only
enter cells with compromised plasma membranes, which only occurs when a cell is dead or
dying. Once inside the cell, EthD1 binds to DNA and becomes strongly fluorescent (528 nm
excitation peak, 617 nm emission peak). Consequently, live cells were identified as CFSE+
EthD1- (green) and dead cells were identified as CFSE+ EthD1+ (green-orange overlap). The
percentage of live hUTC (relative to total hUTC on day 0) was unchanged over 3 days in culture
for all blends (except 1.0/1.0). Importantly, the hUTC were deprived of serum during culture in
the HAMC hydrogels and so the doubling time of the cells was much longer than the 3-day study
period [15]. As a result, the maintenance of live cell percentage relative to total cells on day 0
48
was most likely due to the survival of the initial cell population and not replacement of dead cells
with new cells. That is, survival was the dominant mechanism rather than proliferation.
Previous work with HAMC for cell delivery was focused on the 0.5/0.5 formulation. An
exception is the gelation time data reported by Ballios et al. [6], who published the first work on
HAMC for cell delivery. In addition to 0.5/0.5 HAMC, they examined the gelation time of
0.25/0.25, 0.75/0.75, and 1.0/1.0 formulations. In contrast to the G’/G” time sweep method used
herein, which utilizes precise quantitation of viscoelastic behaviour to determine the point in
which a gel network has formed, Ballios et al. used the observation-based inverted tube test
method. Consequently, the gelation times reported by Ballios et al. were considerably longer
than those reported here and only the 0.5/0.5 and 0.75/0.75 blends met their 10-60 minute
criteria. The viability of RSPCs was examined in two these blends and live cell numbers were
found to be unchanged over 3 days, similar to what was found for hUTCs in HAMC. Citing
easier injectability, only the 0.5/0.5 blend was investigated in vivo. Interestingly, tissue analysis
at 4 weeks following injection into the mouse sub-retinal space revealed that RSPCs delivered in
HAMC were evenly distributed across the retinal pigmented epithelium, while those delivered in
saline were distributed poorly. Recently, Tam et al. [93] improved the cellular microenvironment
provided by 0.5/0.5 HAMC through immobilization of the cell adhesive peptide GRGDS and the
oligodendrocyte-differentiating factor recombinant platelet-derived growth factor A (rPDGF-A)
to the MC component of the gel. NSPCs cultured in the co-immobilized (GRGDS + rPDGF-A)
HAMC gels differentiated to more oligodendrocytes after 7 days compared to unmodified
HAMC. This demonstrated that HAMC (modified with the appropriate chemical cues) is capable
of directing cell fate in addition to maintaining cell viability. The inclusion of cell adhesion
motifs such as GRGDS will likely be useful in improving the viability hUTCs cultured in
HAMC.
49
There are a variety of issues that must be addressed in future studies to further develop HAMC
for cell delivery. Firstly, cell viability in the hydrogels was examined only out to 3 days post-
encapsulation. Stimulating neuroregeneration likely requires hUTC trophic factor secretion for a
period of weeks to elicit effects on endogenous stem cell populations and so it is desirable for the
material to maintain cell viability over this period. Consequently, live cells should be quantified
at 7 days post-encapsulation and beyond. Furthermore, the profile of trophic factors that the
hUTC secrete when encapsulated within HAMC is unknown. The performance of enzyme-linked
immunosorbent assays (ELISAs) on the cell-loaded HAMC at various time points (after cooling
to return it to a liquid state) would be one method of quantifying the trophic factor profile.
ELISAs can detect the presence of biological proteins at pico-molar levels. Another unknown is
the required concentration of hUTC within the HAMC scaffold to evoke neuroregeneration in
vivo. Consequently, various hUTC concentrations should be investigated in in vivo studies. The
optimal injection site of the in vivo implant is also unknown. Specifically, is it more efficacious
to inject the scaffold at the injury site (to maximize local trophic factor concentration) or at some
distance rostral/caudal to the injury site (to avoid the hostile injury microenvironment)? Also for
the reason of avoiding the unfavorable tissue conditions that exist immediately post-injury,
should application of the cells be delayed from injury onset? Lastly, as mentioned, the provision
of chemical cues within a cell scaffold can aid in the maintenance of cell viability [68]. Future
studies should examine the immobilization of adhesion motifs and growth factors to the
constituent polymers for the purpose of enhancing the cellular microenvironment. As previously
described, some progress has recently been achieved to this end, as the MC component of
HAMC was modified using thiol-maleimide and biotin–streptavidin chemistry to covalently
conjugate the cell adhesive peptide GRGDS and the oligodendrocyte-differentiating factor
rPDGF-A [93].
50
4.2 HAMC as a drug delivery vehicle
Currently, there are no clinically acceptable means of delivering drugs to the brain that have poor
permeability across the BBB and require sustained administration for therapeutic effect. An ideal
drug delivery system would combine localized delivery to bypass the BBB with minimal
invasiveness for patient safety during long-term treatment. Due to its known biocompatibility,
biodegradability, and injectability through a fine-gauge needle, drug-loaded HAMC hydrogels
were postulated to satisfy these design criteria. Contained on the cortical surface of the brain via
a small burr hole drilled in the skull, drug-loaded HAMC would act as reservoir for sustained,
minimally invasive delivery of therapeutics directly to the tissue. Previous work has shown that
epi-cortically placed HAMC loaded with soluble poly(ethylene glycol)-modified epidermal
growth factor was capable of releasing the drug into the tissue as deep as the SVZ and eliciting
biological effect without adverse immunological tissue effects [13]. Similar results were
subsequently observed for soluble erythropoietin in HAMC [14]. In both cases, the drug of
interest was a water-soluble protein and release was diffusion-controlled over a period of 2 days.
CsA, the promising neuroregenerative molecule investigated for HAMC-based delivery in this
thesis, differed from these drugs in two ways: 1) CsA has poor solubility in water (6.6 µg/mL
[91]); and 2) delivery over a period of 3-4 weeks was desired. Interestingly, the former point had
potential to be of use in satisfying the latter point, as slow dissolution of hydrophobic drug
particulates dispersed in HAMC was known to be an effective means of extending release
beyond the normal 2 days achievable for soluble drugs [58]. As predicted, dispersion of solid
CsA particulates in HAMC resulted in sustained release for 7-8 days. To extend release further to
the 3-4 week goal, CsA was encapsulated within PLGA microspheres using a well-established
single emulsion process. The slow formation of interconnected pores within the polymer matrix
51
via hydrolytic cleavage of polymer chains was predicted to delay exposure of CsA to the
dissolution environment of the HAMC hydrogel. Using this PLGA-encapsulation method,
release of CsA from HAMC was extended to 21-28 days. This released CsA was found to have
equivalent bioactivity to fresh drug as measured by a neurosphere assay, a prerequisite finding
for in vivo studies. Once in vivo, the PLGA-encapsulated CsA released from HAMC was found
to diffuse through the tissue to the NSPC niche and persist at a stable concentration for at least
24 days post-implant.
The major question that must be addressed in future work is whether the drug delivery system
can deliver a therapeutically relevant amount of CsA to the tissue. Based on a maximum
injection volume of 3 µL, a maximum PLGA microsphere loading in HAMC of 10 wt%, and a
drug loading in the microspheres of 10 wt%, the maximum theoretical dose of CsA that can be
contained on the cortical surface within HAMC is 30 µg. These parameters were used in the
tissue penetration studies described in section 3.3.3 and, as shown in Figure 11, the actual
amount detected in the implant at time zero was roughly 13 µg. This considerable difference
between theoretical and actual initial dose can be attributed to less than 100% drug encapsulation
efficiency within the PLGA microspheres (encapsulation efficiencies in the 70-75% range were
common) and microsphere aggregation during dispersion into HAMC (aggregates cannot be
injected and thus reduce the effective microsphere loading). As shown in Figure 11, this
maximal loading of CsA in HAMC resulted in SVZ CsA concentrations of 8-21 ng/mL over the
24-day study period. Currently, it is unknown whether these concentrations are sufficient to
stimulate NSPCs in vivo, as previous studies using systemically administered or ventricle-infused
CsA for NSPC stimulation have not reported the resulting drug concentrations in the brain [16,
17]. Determining these concentrations is an essential step to answering the question of whether
the drug delivery system can provide therapeutic levels of CsA. If the concentrations prove to be
52
higher than the 8-21 ng/mL current maximum, then changes to the drug delivery system will be
required. Increasing the drug loading within the microspheres and reducing aggregation during
HAMC formulation will result in greater amounts of drug available for diffusion into the brain
tissue. Drug loading can be enhanced in two ways. Firstly, the absolute amount of CsA added to
the organic phase during synthesis can be increased. Drug loadings as high as 20 wt% (compared
to 10 wt% in this work) have been reported for hydrophobic drugs in PLGA particles [94].
Secondly, the synthesis parameters can be modified in an effort to improve drug encapsulation
efficiency beyond the 70-75% range reported here. PLGA concentration in the organic phase,
organic phase to aqueous phase ratio, and shear rate during homogenization can all impact the
resultant encapsulation efficiency [95]. Reducing aggregation during HAMC formulation can be
achieved through changes to the method in which the microspheres are dispersed into the
hydrogel. Currently, microspheres are dispersed into the aCSF before the dry MC and HA are
added. It may be advantageous to disperse the microspheres after the MC has been dissolved, as
the mild hydrophobicity of MC may help prevent the hydrophobic PLGA microspheres from
aggregating.
5 Conclusions The primary contributions of this thesis have been:
1. The characterization of HAMC hydrogels for delivery of hUTC:
a. Strength of HAMC hydrogels was tunable by changing HA/MC concentration.
b. Live hUTC numbers were stable over three days in all blends except the most
concentrated, 1.0/1.0 HAMC.
c. hUTC cultured in HAMC were homogeneously and stably distributed throughout
the polymer matrix.
53
2. The development of a HAMC-based system for the epi-cortical delivery of CsA to the
stroke injured brain:
a. Release of CsA from HAMC in vitro was tunable via the method of incorporation
into the hydrogel, ranging from a period of 2 days (solubilized CsA) to 21-25
days (PLGA-encapsulated CsA).
b. PLGA-encapsulated CsA was bioactive upon release from HAMC in vitro.
c. PLGA-encapsulated CsA was capable of diffusing out of epi-cortically contained
HAMC and into the NSPC niche of adult mice. The concentration of drug in the
tissue was stable over a 24-day period.
54
6 References [1] A. Blesch, M.H. Tuszynski, Spinal cord injury: plasticity, regeneration and the challenge of translational drug development, Trends in Neurosciences, 32 (2009) 41-47.
[2] P. Langhorne, F. Coupar, A. Pollock, Motor recovery after stroke: a systematic review, Lancet Neurology, 8 (2009) 741-754.
[3] A. Orlacchio, G. Bernardi, A. Orlacchio, S. Martino, Stem Cells: An Overview of the Current Status of Therapies for Central and Peripheral Nervous System Diseases, Current Medicinal Chemistry, 17 (2010) 595-608.
[4] M.S. Shoichet, S.R. Winn, Cell delivery to the central nervous system, Adv. Drug Deliv. Rev., 42 (2000) 81-102.
[5] S. Pasha, K. Gupta, Various drug delivery approaches to the central nervous system, Expert Opinion on Drug Delivery, 7 (2010) 113-135.
[6] B.G. Ballios, M.J. Cooke, D. van der Kooy, M.S. Shoichet, A hydrogel-based stem cell delivery system to treat retinal degenerative diseases, Biomaterials, 31 (2010) 2555-2564.
[7] A. Hsieh, T. Zahir, Y. Lapitsky, B. Amsden, W.K. Wan, M.S. Shoichet, Hydrogel/electrospun fiber composites influence neural stem/progenitor cell fate, Soft Matter, 6 (2010) 2227-2237.
[8] D. Gupta, C.H. Tator, M.S. Shoichet, Fast-gelling injectable blend of hyaluronan and methylcellulose for intrathecal, localized delivery to the injured spinal cord, Biomaterials, 27 (2006) 2370-2379.
[9] M.D. Baumann, C.E. Kang, J.C. Stanwick, Y.F. Wang, H. Kim, Y. Lapitsky, M.S. Shoichet, An injectable drug delivery platform for sustained combination therapy, J. Control. Release, 138 (2009) 205-213.
[10] M.D. Baumann, C.E. Kang, C.H. Tator, M.S. Shoichet, Intrathecal delivery of a polymeric nanocomposite hydrogel after spinal cord injury, Biomaterials, 31 (2010) 7631-7639.
[11] C.E. Kang, P.C. Poon, C.H. Tator, M.S. Shoichet, A New Paradigm for Local and Sustained Release of Therapeutic Molecules to the Injured Spinal Cord for Neuroprotection and Tissue Repair, Tissue Engineering Part A, 15 (2009) 595-604.
[12] Y.F. Wang, M.J. Cooke, Y. Lapitsky, R.G. Wylie, N. Sachewsky, D. Corbett, C.M. Morshead, M.S. Shoichet, Transport of epidermal growth factor in the stroke-injured brain, J. Control. Release, 149 (2011) 225-235.
[13] M.J. Cooke, Y.F. Wang, C.M. Morshead, M.S. Shoichet, Controlled epi-cortical delivery of epidermal growth factor for the stimulation of endogenous neural stem cell proliferation in stroke-injured brain, Biomaterials, 32 (2011) 5688-5697.
55
[14] Y. Wang, M.J. Cooke, C.M. Morshead, M.S. Shoichet, Hydrogel delivery of erythropoietin to the brain for endogenous stem cell stimulation after stroke injury, Biomaterials, 33 (2012) 2681-2692.
[15] I.R. Harris, D.J. Messina, A.J. Kihm, A. Seyda, D.C. Colter, Postpartum cells derived from umbilical cord tissue, and methods of making and using the same, in, United States, 2005.
[16] J. Hunt, A. Cheng, A. Hoyles, E. Jervis, C.M. Morshead, Cyclosporin A Has Direct Effects on Adult Neural Precursor Cells, J. Neurosci., 30 (2010) 2888-2896.
[17] A. Erlandsson, C.H.A. Lin, F.G. Yu, C.M. Morshead, Immunosuppression promotes endogenous neural stem and progenitor cell migration and tissue regeneration after ischemic injury, Experimental Neurology, 230 (2011) 48-57.
[18] S. Thuret, L.D.F. Moon, F.H. Gage, Therapeutic interventions after spinal cord injury, Nature Reviews Neuroscience, 7 (2006) 628-643.
[19] M.D. Norenberg, J. Smith, A. Marcillo, The pathology of human spinal cord injury: Defining the problems, J. Neurotrauma, 21 (2004) 429-440.
[20] O.N. Hausmann, Post-traumatic inflammation following spinal cord injury, Spinal Cord, 41 (2003) 369-378.
[21] G.W.J. Hawryluk, J. Rowland, B.K. Kwon, M.G. Fehlings, Protection and repair of the injured spinal cord: a review of completed, ongoing, and planned clinical trials for acute spinal cord injury, Neurosurgical Focus, 25 (2008).
[22] H. Barbeau, S. Rossignol, Enhancement of locomotor recovery following spinal cord injury, Current Opinion in Neurology, 7 (1994) 517-524.
[23] M.M. Osman, D. Lulic, L. Glover, C.E. Stahl, T. Lau, H. van Loveren, C.V. Borlongan, Cyclosporine-A as a neuroprotective agent against stroke: Its translation from laboratory research to clinical application, Neuropeptides, 45 (2011) 359-368.
[24] D.A. Godoy, G. Pinero, M. Di Napoli, Predicting mortality in spontaneous intracerebral hemorrhage: can modification to original score improve the prediction?, Stroke, 37 (2006) 1038-1044.
[25] V. Guiraud, M. Ben Amor, J.L. Mas, E. Touze, Triggers of Ischemic Stroke A Systematic Review, Stroke, 41 (2010) 2669-2677.
[26] E.H. Lo, M.A. Moskowitz, T.P. Jacobs, Exciting, radical, suicidal - How brain cells die after stroke, Stroke, 36 (2005) 189-192.
[27] G.A. Ford, C.A. Bryant, A.A. Mangoni, S.H.D. Jackson, Stroke, dementia, and drug delivery, Br. J. Clin. Pharmacol., 57 (2004) 15-26.
[28] M.A. Hilliard, Axonal degeneration and regeneration: a mechanistic tug-of-war, Journal of Neurochemistry, 108 (2009) 23-32.
56
[29] C. Culmsee, J. Krieglstein, Ischaemic brain damage after stroke: new insights into efficient therapeutic strategies - International Symposium on Neurodegeneration and Neuroprotection, Embo Reports, 8 (2007) 129-133.
[30] W. Hacke, G. Donnan, C. Fieschi, M. Kaste, R. von Kummer, J.P. Broderick, T. Brott, M. Frankel, J.C. Grotta, E.C. Haley, T. Kwiatkowski, S.R. Levine, C. Lewandowski, M. Lu, P. Lyden, J.R. Marler, S. Patel, B.C. Tilley, G. Albers, T. Brott, J. Grotta, P.E. Bluhmki, M. Wilhelm, S. Hamilton, A.S.G. Investigators, E.S.G. Investigators, N.R.-P.S.G. Investigato, Association of outcome with early stroke treatment: pooled analysis of ATLANTIS, ECASS, and NINDS rt-PA stroke trials, Lancet, 363 (2004) 768-774.
[31] P.J. Reir, Cellular transplantation strategies for spinal cord injury and translational neurobiology, NeuroRx, 1 (2004) 424-451.
[32] G.U. Enzmann, R.L. Benton, J.F. Talbott, Q.L. Cao, S.R. Whittemore, Functional considerations of stem cell transplantation therapy for spinal cord repair, J. Neurotrauma, 23 (2006) 479-495.
[33] T. Kallur, V. Darsalia, O. Lindvall, Z. Kokaia, Human fetal cortical and striatal neural stem cells generate region-specific neurons in vitro and differentiate extensively to neurons after intrastriatal transplantation in neonatal rats, Journal of Neuroscience Research, 84 (2006) 1630-1644.
[34] A. Bakshi, C.A. Keck, V.S. Koshkin, D.G. LeBold, R. Siman, E.Y. Snyder, T.K. McIntosh, Caspase-mediated cell death predominates following engraftment of neural progenitor cells into traumatically injured rat brain, Brain Research, 1065 (2005) 8-19.
[35] A.M. Parr, I. Kulbatski, C.H. Tator, Transplantation of adult rat spinal cord stem/progenitor cells for spinal cord injury, J. Neurotrauma, 24 (2007) 835-845.
[36] S.M. O'Connor, D.A. Stenger, K.M. Shaffer, W. Ma, Survival and neurite outgrowth of rat cortical neurons in three-dimensional agarose and collagen gel matrices, Neuroscience Letters, 304 (2001) 189-193.
[37] H. Itosaka, S. Kuroda, H. Shichinohe, H. Yasuda, S. Yano, S. Kamei, R. Kawamura, K. Hida, Y. Iwasaki, Fibrin matrix provides a suitable scaffold for bone marrow stromal cells transplanted into injured spinal cord: A novel material for CNS tissue engineering, Neuropathology, 29 (2009) 248-257.
[38] K.E. Crompton, J.D. Goud, R.V. Bellamkonda, T.R. Gengenbach, D.I. Finkelstein, M.K. Horne, J.S. Forsythe, Polylysine-functionalised thermoresponsive chitosan hydrogel for neural tissue engineering, Biomaterials, 28 (2007) 441-449.
[39] S.G. Levesque, M.S. Shoichet, Synthesis of cell-adhesive dextran hydrogels and macroporous scaffolds, Biomaterials, 27 (2006) 5277-5285.
[40] V. Patel, G. Joseph, A. Patel, S. Patel, D. Bustin, D. Mawson, L.M. Tuesta, R. Puentes, M. Ghosh, D.D. Pearse, Suspension matrices for improved Schwann-cell survival after implantation into the injured rat spinal cord, J Neurotrauma, 27 (2010) 789-801.
57
[41] M. Hiraoka, K. Kato, T. Nakaji-Hirabayashi, H. Iwata, Enhanced Survival of Neural Cells Embedded in Hydrogels Composed of Collagen and Laminin-Derived Cell Adhesive Peptide, Bioconjugate Chemistry, 20 (2009) 976-983.
[42] J.R. Thonhoff, D.I. Lou, P.M. Jordan, X. Zhao, P. Wu, Compatibility of human fetal neural stem cells with hydrogel biomaterials in vitro, Brain Research, 1187 (2008) 42-51.
[43] M. Dreyfuss, E. Harri, H. Hofmann, H. Kobel, W. Pache, H. Tscherter, CYCLOSPORIN-A AND C NEW METABOLITES FROM TRICHODERMA-POLYSPORUM (LINK EX PERS) RIFAI, European Journal of Applied Microbiology, 3 (1976) 125-133.
[44] J.F. Borel, C. Feurer, C. Magnee, H. Stahelin, Effects of the new anti-lymphocytic peptide cyclosporin A in animals, Immunology, 32 (1977) 1017-1025.
[45] O. Butovsky, Y. Ziv, A. Schwartz, G. Landa, A.E. Talpalar, S. Pluchino, G. Martino, M. Schwartz, Microglia activated by IL-4 or IFN-gamma differentially induce neurogenesis and oligodendrogenesis from adult stem/progenitor cells, Molecular and Cellular Neuroscience, 31 (2006) 149-160.
[46] S. Pluchino, G. Martino, Neural stem cell-mediated immunomodulation: repairing the haemorrhagic brain, Brain, 131 (2008) 604-605.
[47] J. Hunt, C. Morshead, Cyclosporin A enhances cell survival in neural precursor populations in the adult central nervous system, Molecular and Cellular Pharmacology, 2 (2010) 81-88.
[48] K.E. Sandoval, K.A. Witt, Blood-brain barrier tight junction permeability and ischemic stroke, Neurobiology of Disease, 32 (2008) 200-219.
[49] Novartis, Sandimmune Prescrbing Information in, 2011.
[50] A.G. de Boer, P.J. Gaillard, Strategies to improve drug delivery across the blood-brain barrier, Clinical Pharmacokinetics, 46 (2007) 553-576.
[51] L. Jasmin, P.T. Ohara, Long-term intrathecal catheterization in the rat, Journal of Neuroscience Methods, 110 (2001) 81-89.
[52] N.A. Peppas, J.Z. Hilt, A. Khademhosseini, R. Langer, Hydrogels in biology and medicine: From molecular principles to bionanotechnology, Advanced Materials, 18 (2006) 1345-1360.
[53] G.D. Prestwich, D.M. Marecak, J.F. Marecek, K.P. Vercruysse, M.R. Ziebell, Controlled chemical modification of hyaluronic acid: synthesis, applications, and biodegradation of hydrazide derivatives, J. Control. Release, 53 (1998) 93-103.
[54] C.N. Salinas, K.S. Anseth, Mesenchymal Stem Cells for Craniofacial Tissue Regeneration: Designing Hydrogel Delivery Vehicles, Journal of Dental Research, 88 (2009) 681-692.
[55] E.A. Balazs, J.L. Denlinger, Clinical uses of hyaluronan, Ciba Foundation symposium, 143 (1989) 265-275; discussion 275-280, 281-265.
58
[56] M.D. Baumann, A Composite Polymeric Drug Delivery System for Treatment of Spinal Cord Injury, in: Department of Chemical Engineering & Applied Chemistry University of Toronto, 2010.
[57] N. Sarkar, Thermal gelation properties of methyl and hydroxypropyl methylcellulose, J. Appl. Polym. Sci., 24 (1979) 1073-1087.
[58] Y.F. Wang, Y. Lapitsky, C.E. Kang, M.S. Shoichet, Accelerated release of a sparingly soluble drug from an injectable hyaluronan-methylcellulose hydrogel, J. Control. Release, 140 (2009) 218-223.
[59] R.C. Mundargi, V.R. Babu, V. Rangaswamy, P. Patel, T.M. Aminabhavi, Nano/micro technologies for delivering macromolecular therapeutics using poly(D,L-lactide-co-glycolide) and its derivatives, J. Control. Release, 125 (2008) 193-209.
[60] J.C. Stanwick, M.D. Baumann, M.S. Shoichet, In vitro sustained release of bioactive anti-NogoA, a molecule in clinical development for treatment of spinal cord injury, International Journal of Pharmaceutics, 426 (2012) 284-290.
[61] J.C. Stanwick, M.D. Baumann, M.S. Shoichet, Enhanced neurotrophin-3 bioactivity and release from a nanoparticle-loaded composite hydrogel, J. Control. Release, 160 (2012) 666-675.
[62] C.H. Tator, Review of treatment trials in human spinal cord injury: Issues, difficulties, and recommendations, Neurosurgery, 59 (2006) 957-982.
[63] S. Karimi-Abdolrezaee, E. Eftekharpour, J. Wang, C.M. Morshead, M.G. Fehlings, Delayed transplantation of adult neural precursor cells promotes remyelination and functional neurological recovery after spinal cord injury, J. Neurosci., 26 (2006) 3377-3389.
[64] D.J. Mooney, H. Vandenburgh, Cell delivery mechanisms for tissue repair, Cell Stem Cell, 2 (2008) 205-213.
[65] D.D. Allison, K.J. Grande-Allen, Review. Hyaluronan: A powerful tissue engineering tool, Tissue Eng., 12 (2006) 2131-2140.
[66] C.M. Wang, R.R. Varshney, D.A. Wang, Therapeutic cell delivery and fate control in hydrogels and hydrogel hybrids, Adv. Drug Deliv. Rev., 62 (2010) 699-710.
[67] G.L. Chang, T.K. Hung, W.W. Feng, An in-vivo measurement and analysis of viscoelastic properties of the spinal cord of cats, Journal of biomechanical engineering, 110 (1988) 115-122.
[68] M.S. Shoichet, Polymer Scaffolds for Biomaterials Applications, Macromolecules, 43 (2010) 581-591.
[69] D. Dado, S. Levenberg, Cell-scaffold mechanical interplay within engineered tissue, Semin. Cell Dev. Biol., 20 (2009) 656-664.
59
[70] T. Joh, Section 2 - Evaluation of gel point, in: O. Yoshihito, K. Kanji, F. Takao, I. Okihiko, H. Yoshitsugu, M. Tsutomu, S. Tadao, W. Lin, I. Hatsuo (Eds.) Gels Handbook, Academic Press, Burlington, 2001, pp. 51-64.
[71] S.M. Frisch, E1A induces the expression of epithelial characteristics, J. Cell Biol., 127 (1994) 1085-1096.
[72] H.P. Adams, G. del Zoppo, M.J. Alberts, D.L. Bhatt, L. Brass, A. Furlan, R.L. Grubb, R.T. Higashida, E.C. Jauch, C. Kidwell, P.D. Lyden, L.B. Morgenstern, A.I. Qureshi, R.H. Rosenwasser, P.A. Scott, E.F.M. Wijdicks, Guidelines for the early management of adults with ischemic stroke - A guideline from the American Heart Association/American Stroke Association Stroke Council, Clinical Cardiology Council, Cardiovascular Radiology and Intervention Council, and the atherosclerotic peripheral vascular disease and quality of care outcomes in research interdisciplinary working groups, Stroke, 38 (2007) 1655-1711.
[73] S.I. Savitz, D.M. Rosenbaum, J.H. Dinsmore, L.R. Wechsler, L.R. Caplan, Cell transplantation for stroke, Ann. Neurol., 52 (2002) 266-275.
[74] B. Kolb, C. Morshead, C. Gonzalez, M. Kim, C. Gregg, T. Shingo, S. Weiss, Growth factor-stimulated generation of new cortical tissue and functional recovery after stroke damage to the motor cortex of rats, Journal of Cerebral Blood Flow and Metabolism, 27 (2007) 983-997.
[75] T. Teramoto, J.H. Qu, J.C. Plumier, M.A. Moskowitz, EGF amplifies the replacement of parvalbumin-expressing striatal interneurons after ischemia, Journal of Clinical Investigation, 111 (2003) 1125-1132.
[76] H.S.G. Kalluri, R.J. Dempsey, Growth factors, stem cells, and stroke, Neurosurgical Focus, 24 (2008).
[77] N. Sprigg, P.M.W. Bath, Colony stimulating factors (blood growth factors) are promising but unproven for treating stroke, Stroke, 38 (2007) 1997-1998.
[78] D. Lulic, J. Burns, E.C. Bae, H. van Loveren, C.V. Borlongan, A Review of Laboratory and Clinical Data Supporting the Safety and Efficacy of Cyclosporin A in Traumatic Brain Injury, Neurosurgery, 68 (2011) 1172-1185.
[79] C.M. Morshead, A.D. Garcia, M.V. Sofroniew, D. van der Kooy, The ablation of glial fibrillary acidic protein-positive cells from the adult central nervous system results in the loss of forebrain neural stem cells but not retinal stem cells, European Journal of Neuroscience, 18 (2003) 76-84.
[80] B.L.K. Coles-Takabe, I. Brain, K.A. Purpura, P. Karpowicz, P.W. Zandstra, C.M. Morshead, D. Van der Kooy, Don't Look: Growing Clonal Versus Nonclonal Neural Stem Cell Colonies, Stem Cells, 26 (2008) 2938-2944.
[81] C.S. Brazel, N.A. Peppas, Modeling of drug release from swellable polymers, European Journal of Pharmaceutics and Biopharmaceutics, 49 (2000) 47-58.
60
[82] B.N. Nalluri, C. Milligan, J.H. Chen, P.A. Crooks, A.L. Stinchcomb, In vitro release studies on matrix type transdermal drug delivery system of naltrexone and its acetyl prodrug, Drug Development and Industrial Pharmacy, 31 (2005) 871-877.
[83] A.W. Hixson, J.H. Crowell, Dependence of reaction velocity upon surface and agitation I - Theoretical consideration, Industrial and Engineering Chemistry, 23 (1931) 923-931.
[84] V.R. Sinha, A. Trehan, Biodegradable microspheres for protein delivery, J. Control. Release, 90 (2003) 261-280.
[85] K.S. Soppimath, T.M. Aminabhavi, A.R. Kulkarni, W.E. Rudzinski, Biodegradable polymeric nanoparticles as drug delivery devices, J. Control. Release, 70 (2001) 1-20.
[86] C.M. Morshead, B.A. Reynolds, C.G. Craig, M.W. McBurney, W.A. Staines, D. Morassutti, S. Weiss, D. Vanderkooy, Neural stem-cells in the adult mamnalian forebrain - A relatively quiescent subpopulation of subependymal cells, Neuron, 13 (1994) 1071-1082.
[87] U. Bilati, E. Allemann, E. Doelker, Strategic approaches for overcoming peptide and protein instability within biodegradable nano- and microparticles, European journal of pharmaceutics and biopharmaceutics : official journal of Arbeitsgemeinschaft fur Pharmazeutische Verfahrenstechnik e.V, 59 (2005) 375-388.
[88] W.M. Saltzman, M.L. Radomsky, Drugs released from polymers - diffusion and elimination in bran-tissue, Chemical Engineering Science, 46 (1991) 2429-2444.
[89] W.M. Saltzman, Biomaterials for drug delivery to the brain, Protein Engineering, 10 (1997) 78-78.
[90] K.B.J. Franklin, G. Paxinos, The Mouse Brain in Stereotaxic Coordinates, 3rd ed., Academic Press, 2007.
[91] Z.Q. Yang, J. Xu, P. Pan, X.N. Zhang, Preparation of an alternative freeze-dried pH-sensitive cyclosporine A loaded nanoparticles formulation and its pharmacokinetic profile in rats, Pharmazie, 64 (2009) 26-31.
[92] M.J. Cooke, K. Vulic, M.S. Shoichet, Design of biomaterials to enhance stem cell survival when transplanted into the damaged central nervous system, Soft Matter, 6 (2010) 4988-4998.
[93] R.Y. Tam, M.J. Cooke, M.S. Shoichet, A covalently modified hydrogel blend of hyaluronan-methyl cellulose with peptides and growth factors influences neural stem/progenitor cell fate, Journal of Materials Chemistry, 22 (2012) 19402-19411.
[94] C. Wischke, S.P. Schwendeman, Principles of encapsulating hydrophobic drugs in PLA/PLGA microparticles, International Journal of Pharmaceutics, 364 (2008) 298-327.
[95] F. Danhier, E. Ansorena, J.M. Silva, R. Coco, A. Le Breton, V. Preat, PLGA-based nanoparticles: An overview of biomedical applications, J. Control. Release, 161 (2012) 505-522.