Citation
A biomimetic reverse thermal gel for spinal cord injury

Material Information

Title:
A biomimetic reverse thermal gel for spinal cord injury
Creator:
Bardill, James ( author )
Language:
English
Physical Description:
1 electronic file (103 pages) : ;

Subjects

Subjects / Keywords:
Spinal cord -- Wounds and injuries ( lcsh )
Drug delivery systems ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Review:
No effective treatment for paralysis after spinal cord injury (SCI) exists. To meet this unmet medical need, researchers have developed various molecular and cell therapies to repair or replace damaged spinal tissue. Despite the promise these treatments hold, their therapeutic effect remains limited due to inefficient delivery methods. To overcome this obstacle, biomimetic polymer scaffolding systems have been developed to control the delivery of these new treatments to injured spinal cord tissue. These polymers are designed to integrate into surrounding tissues to provide support to damaged tissue, while also providing an environment within the SCI lesion that encourages tissue regeneration. ( , )
Review:
An injectable, polyethylene glycol-based reverse thermal gel was developed to deliver a minimally invasive treatment directly into the lesion site of a compressed SCI. To mimic the extracellular environment, the polymer was conjugated to arginylglycylaspartic acid, a cellular adhesive peptide that promotes neural attachment, survival, and growth. In-vitro tests revealed the polymer had no cytotoxic effects on PC12 cellular proliferation and could support the survival and growth of a primary culture of retinal ganglion cells in a 3D polymer matrix. The polymer was then injected directly into the compression injury site of a rat SCI model to assess the polymer’s capability to provide axonal support within the SCI lesion site. Although functional and histological analysis were unable to demonstrate definitive evidence of the polymer providing axonal support within the SCI lesion, regenerating axons were detected within the SCI site where the polymer was injected. These preliminary results demonstrate this novel polymer has the capability to act as a promising neural scaffold to promote a regenerative environment within a SCI lesion site. For future studies, we predict incorporating cellular and molecular therapies into the polymer will provide an effective delivery vehicle to enhance therapeutic efficiency of SCI treatment approaches.
Thesis:
Thesis (M.S.) - University of Colorado Denver.
Bibliography:
Includes bibliographic references
Additional Physical Form:
System requirements: Adobe Reader.
General Note:
Department of Bioengineering
Statement of Responsibility:
by James Bardill.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
945637918 ( OCLC )
ocn945637918
Classification:
LD1193.E56 2015m B37 ( lcc )

Downloads

This item has the following downloads:


Full Text
A BIOMIMETIC REVERSE THERMAL GEL FOR SPINAL CORD INJURY
by
JAMES BARDILL
B.S., University of Minnesota-Twin Cities, 2008
A thesis submitted to the
Faculty of the Graduate School of the
University of Colorado in partial fulfillment
of the requirements for the degree of
Master of Science
Bioengineering
2015


This thesis for the Master of Science degree by
James Bardill
has been approved for the
Bioengineering Program
by
Daewon Park, Chair
Vikas Patel
Rony Marwan
November 5, 2015
11


Bardill, James (M.S., Bioengineering)
A Biomimetic Reverse Thermal Gel for Spinal Cord Injury
Thesis directed by Assistant Professor Daewon Park
ABSTRACT
No effective treatment for paralysis after spinal cord injury (SCI) exists. To meet
this unmet medical need, researchers have developed various molecular and cell therapies
to repair or replace damaged spinal tissue. Despite the promise these treatments hold,
their therapeutic effect remains limited due to inefficient delivery methods. To overcome
this obstacle, biomimetic polymer scaffolding systems have been developed to control the
delivery of these new treatments to injured spinal cord tissue. These polymers are
designed to integrate into surrounding tissues to provide support to damaged tissue, while
also providing an environment within the SCI lesion that encourages tissue regeneration.
An injectable, polyethylene glycol-based reverse thermal gel was developed to
deliver a minimally invasive treatment directly into the lesion site of a compressed SCI.
To mimic the extracellular environment, the polymer was conjugated to
arginylglycylaspartic acid, a cellular adhesive peptide that promotes neural attachment,
survival, and growth. In-vitro tests revealed the polymer had no cytotoxic effects on
PC 12 cellular proliferation and could support the survival and growth of a primary
culture of retinal ganglion cells in a 3D polymer matrix. The polymer was then injected
directly into the compression injury site of a rat SCI model to assess the polymers
capability to provide axonal support within the SCI lesion site. Although functional and
histological analysis were unable to demonstrate definitive evidence of the polymer
iii


providing axonal support within the SCI lesion, regenerating axons were detected within
the SCI site where the polymer was injected. These preliminary results demonstrate this
novel polymer has the capability to act as a promising neural scaffold to promote a
regenerative environment within a SCI lesion site. For future studies, we predict
incorporating cellular and molecular therapies into the polymer will provide an effective
delivery vehicle to enhance therapeutic efficiency of SCI treatment approaches.
The form and content of this abstract are approved. I recommend its publication.
Approved: Daewon Park
IV


ACKNOWLEDGEMENTS
First and foremost, I would like to thank my advisor, Dr. Daewon Park for giving
me the opportunity to gain insight into exciting tissue engineering approaches with
biomaterials. His guidance and instruction have been invaluable and I look forward to
future work with him. Dr. Vikas Patel took time out of his busy clinical/ surgery
schedule to provide me vital insight for surgical techniques for the in-vivo study that
were instrumental in completing this project. Dr. Rony Marwan provided great insight
into spinal cord development and clinical approaches that were crucial for developing a
practical therapeutic approach. I look forward to continuing to work with him on a future
project related to spina bifida defects.
One of the most important aspects of my experience was enjoying the work
environment of the Translational Biomaterials Research Laboratory. Finding a workplace
with as much enthusiasm and encouragement as this lab provided is truly rare, and I am
fortunate to have shared this experience with current and past lab members: Melissa
Laughter, David Lee, Anna-Laura Nelson, Madia Stein, Adam Rocker, Ryan Brody,
Maria Bortot, Krishna Madhaven, Matt Taylor, and Lindsey Hockensmith.
Finally, I would like to thank my family. My dad, mom, and three sisters have
been nothing but supportive of everything I have ventured into during my life.


Declaration of original work
by
James Bardill
This masters thesis was independently composed and authored by myself, using the
support from my advisor, committee members, lab members, fellow students, and the
Department of Bioengineering. The research and ideas presented in this document
originated from the Translational Biomaterials Research Laboratory under the guidance
of Dr. Daewon Park. All resources and funds were provided by Dr. Daewon Park and the
Department of Bioengineering.
James Bardill
vi


TABLE OF CONTENTS
Chapter
1 Introduction....................................................................1
1.1 Spinal cord injury background information..................................1
1.2 SCI epidemiology............................................................2
1.3 SCI pathophysiology.........................................................2
1.4 Current clinical SCI treatments.............................................4
1.4.1 Surgical intervention...................................................4
1.4.2 Pharmacological intervention............................................5
1.5 Emerging therapies in SCI treatment.........................................5
1.5.1 Molecular approaches....................................................6
1.5.2 Cellular approaches.....................................................7
1.5.3 Challenges of emerging therapies........................................9
1.6 Goal of this study.........................................................10
2 Literature review..............................................................12
2.1 Polymers scaffolds for SCI.................................................12
2.1.1 Biomedical applications of PEG.........................................13
2.1.2 SCI scaffold designs...................................................16
2.1.3 Surgically implanted scaffold designs..................................16
vii


2.1.4 Injectable scaffolds
17
2.1.5 Reverse thermal gels (RTGs)........................................19
2.1.6 Previous research with RTGs........................................20
2.1.7 RGD peptide........................................................21
2.2 Animal Models of SCI...................................................24
2.2.1 Rat models of SCI..................................................25
2.2.2 Functional testing.................................................27
2.2.3 Immunohi stochemi stry (IHC).......................................28
3 Research objectives and experimental approach...............................30
3.1 Hypothesis.............................................................30
3.2 Specific aims..........................................................30
4 Materials and equipment.....................................................33
4.1 Materials..............................................................33
4.2 Equipment..............................................................34
5 Methods.....................................................................36
5.1 Synthesis of PEGS A....................................................36
5.2 Synthesis of carboxylic acid terminated PNIPAm.........................37
5.3 Synthesis of PEGS A-PNIP Am............................................37
viii
5.3.1 Molar Reaction
37


5.3.2 Gram base conjugation..............................................38
5.4 RGD conjugation........................................................38
5.4.1 RGD conjugation to PEGS A..........................................38
5.4.2 RGD conjugation to PEGSA-PNIPAm (mol base).........................39
5.4.3 RGD conjugation to PEGSA-PNIPAm (gram base)........................39
5.5 Polymer biocompatibility testing.......................................40
5.5.1 PC12 cell culture..................................................41
5.5.2 Retinal ganglion cell (RGC) cell culture...........................41
5.5.3 MTT cytotoxi city..................................................41
5.5.4 Alamar blue cell viability assay...................................42
5.5.5 RGC attachment and growth..........................................42
5.6 Compression SCI rat model..............................................43
5.6.1 Ethics and surgical approval.......................................43
5.6.2 Surgical procedures................................................44
5.6.3 Post-surgery procedure.............................................45
5.7 Functional Assessment..................................................46
5.8 Histology..............................................................46
5.8.1 H&E staining.......................................................46
5.8.2 Immunohi stochemi stry.............................................46
IX


5.9 Statistical analysis
47
6 Results and Discussion.....................................................48
6.1 Polymer synthesis.......................................................48
6.2 Polymer characterization................................................50
6.2.1 PEGSA FT-IR characterization........................................50
6.2.2 PEGSA-PNIPAm characterization.......................................51
6.2.3 PEGSA-PNIPAm gelling properties.....................................52
6.2.4 PEGSA-PNIPAm morphological characterization.........................52
6.2.5 PEGSA-RGD FTTR characterization.....................................54
6.2.6 PEGSA-PNIPAm-RGD FTIR characterization..............................55
6.3 In-vitro assessment.....................................................56
6.3.1 MTT cytotoxi city...................................................56
6.3.2 Alamar blue cell viability..........................................57
6.3.3 RGC attachment and growth...........................................58
6.4 Compression SCI rat model...............................................59
6.5 Functional assessment...................................................62
6.6 Histology...............................................................63
6.6.1 H&E staining........................................................63
6.6.2 GFAP IHC............................................................64
x


6.6.3 GAP-43 MC..............................................................66
7 Conclusion...................................................................68
8 Study limitations and future work............................................72
8.1 Modify control and experimental groups...................................72
8.2 Modify SCI compression...................................................72
8.3 Modify PEGSA backbone to prevent gel shrinking..........................73
8.4 Increase study duration..................................................74
8.5 Monitor polymer presence in spinal tissue...............................74
8.6 Long term goal of RTG SCI application....................................74
References........................................................................77
xi


LIST OF FIGURES
Figure
1-1. SCI pathophysiology depicting the primary injury stage caused by impact
and the secondary injury cascade forming the glial scar 7.....................3
1-2. Signaling pathways activated by neurotrophic factors.....................7
1- 3. NSC therapy for SCI (B) aims to reduce glial scarring and improve axon
myelination to restore normal function (A) 32.................................9
2- 1. Desirable properties of polymer scaffolds for tissue engineering applications
34.........................................................................12
2-2. A 10% RTG solution (mg/mL) at room temperature (left) followed by 10
seconds of exposure to 37C water shows formation of solid gel (right).....19
2-3. Schematic of RGD integrin binding to trigger multiple signaling pathways.22
2-4. Polymer scaffold with conjugated RGD showed substantial differentiation of
human NSCs into motor neurons. Inset is a schematic of the polymer. Green
represents axon extensions via piII-Tubulin staining. Red represents motor
neuron markers (Islet-1, HB9) 71...........................................23
6-1. Reaction of PEGDGE with SA to produce PEGS A. The PEGS A repeating
unit contains two hydroxyl groups that will be used for further chemical
conjugation................................................................48
6-2. Conjugation of PNIPAm to PEGS A backbone..............................49
6-3. PEGSA-PNIPAm-RGD reaction sequence. GRGDS is a modified RGD
peptide....................................................................50
6-4. I I-IR spectrum of PEGDGE and PEGSA......................................51
6-5. I I-IR spectrum of PEGSA, PNIPAm, PEGSA-PMPAm............................51
6-6. LCST of PEGSA-PNIPAm conjugation reactions and PNIPAm.................52
6-7. SEM micrograph (xl80) showing porosity of PEGSA-PNIPAm with
consistent porous size throughout the structure. Scale bar 100pm..............53
6-8. SEM micrograph (x950) showing an average pore size of 5pm-20pm...........54
6-9. I I-IR spectrum of PEGSA, RGD, and PEGSA-RGD.............................55
xii


6-10. FT-IR spectrum of PEGS A-PNIP Am, RGD. The conjugation of RGD to
PEGSA-PNIPAm could not be confirmed because the PNIPAm peaks
overshadow the RGD peaks..............................
56
6-11. FT-IR spectrum of region a (in Figure 6-8) of PEGSA-PNIPAm, RGD, and
PEGS A-PNIP Am-RGD. The PEGSA-PNIPAm peaks are located in the same
chemical shift location as the RGD peaks. The RGD peaks have a weaker signal
than the large PNIPAm peaks...............................................56
6-12. PC12 cells showed no cytotoxicity when exposed to PEGSA-PNIPAm
media extracts, per ISO 10993-5. Statistical analysis by ANOVA demonstrated
no difference between the positive control (Media+cells) sample and experimental
samples (p=0.36, n=5). Experimental samples are normalized to cells exposed to
pure media. Media only is the negative control. Error bars represent standard error
of the mean...............................................................57
6-13. PC12 cells showed no change in viability when exposed to PEGSA-
PNIPAm media extracts, per ISO 10993-5. Statistical analysis by ANOVA
demonstrated no difference between the positive control (Media+cells) sample
and experimental samples (p=0.13, n=5). Experimental samples are normalized to
cells exposed to pure media. Media only is the negative control. Error bars
represent standard error of the mean.....................................58
6-14. RGC axon extensions within PEGSA-PNIPAm-RGD 3D polymer matrix.
Axons stained with pill-tubulin, nuclei with DAPI, dead cells appear red.
Confocal microscopy z-stack, scale bar = 100 pm..........................59
6-15. Compression spinal cord injury. (A) Exposed thoracic vertebrae (T9-T11),
(B) exposed spinal cord tissue after laminectomy, (C) spinal cord compression
with 30 g aneurysm clip..................................................60
6-16. A 10 pi injection of PEGSA-PNIPAm dyed with toluidine blue confirmed
localization of the polymer into the spinal cord injection site..........62
6-17. Vertical spinal cord sections stained with H&E showing the progression of
glial scarring after compression SCI with polymer injection. (A) Healthy,
uninjured tissue, (B) compressed spinal cord 4 hours after surgery with polymer
injection, (C) 4 weeks after surgery. Scale bar = 500 pm..........................64
6-18. Glial scar formation of longitudinal compressed spinal cord after 8 weeks
with polymer injection. The decreased cellular density at the SCI site could be
due to the presence of the polymer, which does not stain with H&E. Scale bar =
1000 pm...........................................................................64
6-19. GFAP immunostaining to detect astrocyte accumulation around SCI site in
vertical spinal cord sections from (A) representative 4 week polymer injection,
(B) 8 week polymer injection. Images acquired with Nikon confocal microscope,
20X air objective. Scale bar = 500 pm.............................................65
xm


6-20. GAP-43 immunostaining to detect regenerating axons. Vertical SCI lesion
site from representative 4 week polymer injection (top to bottom representing
rostral and caudal ends of the spinal cord). Scale bar = 500 pm. Confocal
microscopy SCI image, 10X air objective...................................66
6-21. GAP-43 immunostaining to detect regenerating axons. Longitudinal spinal
cord section from 8 week polymer injection (left and right sides representing
rostral and caudal ends of the spinal cord). Scale bar = 500 pm. Confocal
microscopy SCI image, 10X air objective...................................67
8-1. After several minutes, PEGSA-PNIPAm shrinks, resulting in a gel that no
longer holds the shape of the vial........................................73
xiv


LIST OF TABLES
Table
1. Surgical outcomes and complications of saline injected rats........................61
2. Surgical outcomes and complications of polymer injected rats.......................61
3. Summary of BBB scores for 4 week rats. Only 1 saline injection rat survived
the entire 4 week duration of the study, and the polymer injection rats have a large
variability (3.4 3.7, n = 6).........................................................63
xv


LIST OF ABBREVIATIONS
3D 3-dimensional
ANOVA analysis of variance
BBB Basso, Beattie, and Bresnahan
BDNF brain-derived neurotrophic factor
BSA Bovine serum albumin
chABC chondroitinase ABC
CNS central nervous system
CNTF ciliary neurotrophic factor
CSF Cerebral spinal fluid
CSPG chondroitin sulfate proteoglycans
DAPI 4',6-diamidino-2-phenylindole
DCC Di cy cl ohexyl carb odiimi de
DCM dichloromethane
DMAP 4-Dimethylaminopyridine
DMF N,N-dimethylformamide
DMSO dimethyl sulfoxide
ECM extracellular matrix
EDI Anti-CD68 macrophage
EDC N-(3-dimethylamino-propyl)-N-ethylcarbodiimide hydrochloride
ESC embryonic stem cell
FBS Fetal bovine serum
FDA Food and Drug Administration
FT-IR Fourier transform infrared spectroscopy
GAP-43 Growth associated protein-43
GFAP Glial fibrillary acidic protein
RGD Gly-Arg-Gly-Asp-Ser
HAMC hyaluronic acid and methylcellulose
H&E Hematoxylin and Eosin
hNSC human neural stem cell
HPMA N-(2-hydroxyproplyl) methacrylamide
HS Horse serum
IACUC institutional animal car and use committee
me immunohi stochemi stry
IL-10 Interleukin-10
IP Intraperitoneal
IR infrared
LCST Lower critical solution temperature
xvi


LFB Luxol fast blue
MP methylprednisolone
MSC mesenchymal stem cell
MTT 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide
MW Molecular weight
NGF nerve growth factor
NHS N-hydroxysuccinimide
NIPAAm N-Isopropylacrylamide
NSC neural stem cell
NT-3 neurotrophin-3
NTF Neurotrophic factors
OB Olfactory bulb
OCT optimal cutting temperature
OEC Olfactory ensheathing cell
OM Olfactory mucosa
PBS phosphate buffered saline
PC12 pheochromocytoma cell line
PCL Poly(caprolactone)
PEG Poly(ethylene glycol)
PEGDGE Poly(ethylene glycol diglycidyl ether)
PEGSA Poly(ethylene glycol diglycidyl ether) succinic acid
PEGSA- PNIPAm PEGSA conjugated to PNIPAm
PEGSA- PNIPAm- RGD PEGSA-PNIPAm conjugated to RGD peptide
PFA Paraformaldehyde (4% solution in PBS)
PGA polyglycolic acid
PHEMA poly(2-hydroxyethylmethacrylate
PLGA poly(lactic-co-glycolic acid)
PNI peripheral nerve injury
PNIPAm poly(N-isopropyl acrylamide)
PNS peripheral nervous system
PS Penicillin streptomycin
PSHU poly(serinol hexamethylene urea)
PSHU-RGD arginylglycylaspartic acid conjugated poly(serinol hexamethylene urea)
PVA Poly(vinyl alcohol)
RGC Retinal ganglion cell
RGD arginylglycylaspartic acid, Arg-Gly-Asp
XVII


RTG Reverse thermal gel
SA Succinic acid
SC subcutaneous
SCI spinal cord injury
SEM scanning electron microscope
SNI sciatic nerve injury
UV-VIS Ultravi ol et-vi sibl e
TPP triphenylphosphine
XV111


1 Introduction
1.1 Spinal cord injury background information
Spinal cord injuries (SCI) result in devastating neurological deficits that lead to
long term health and lifestyle complications. Paralysis (quadriplegia or paraplegia) is the
most common disability caused by SCI, with other serious disabilities including loss of
bladder/ bowel control, increased risks for urinary tract infections, and cardiopulmonary
distress.
The severity of a SCI depends on the location of the injury to the spinal column
and corresponds to the degree of motor and sensory loss experienced by the patient. For
patients with cervical level injuries, ailments include quadriplegia, loss of urinary/ bowel
control, and respiratory ailments. Higher level cervical injuries often require constant
personal care assistance, while lower level cervical injuries may be managed with
assistive technologies for more independent patient activity. Thoracic level SCI ailments
include paraplegia and loss of bladder/ bowel control, though control over respiration,
arms, and hands is usually normal, allowing for more patient independence. Lumbar and
sacral level injury ailments include partial loss of hip/ leg control and loss of bladder/
bowel control, with the lower level injuries allowing for more control over walking
ability.
The severity of a SCI also depends on whether the injury is complete or
incomplete. Complete injuries result in total loss of motor and sensory function below the
injury site, while incomplete injuries result in partial loss of motor and sensory function.
1


Despite improvements in clinical care and physical therapy that have led to longer
patient survival rates, increased patient independence, and overall improved quality of
life, there is currently no treatment capable of restoring all lost functions.
1.2 SCI epidemiology
An estimated 250,000 people in the United States live with SCI, with 11,000 new
cases reported each year 1. The overall leading causes of SCI are falls and motor vehicle
accidents, followed by acts of violence and sports injuries. Injuries at the cervical level
are the most common, followed by thoracic/lumbosacral injuries. The lifetime costs of
SCI can reach up to $5,000,000 for quadriplegics and up to $2,000,000 for paraplegics,
not taking into account loss of wages or occupational benefits 2 SCI related mortality
rates are highest within the first 12-18 months of injury, with cervical injury related
deaths having the highest mortality rates 3.
1.3 SCI pathophysiology
Damage to the central nervous system (CNS) causes devastating neurological and
functional deficits that are difficult to overcome. The longstanding belief for decades
was that CNS neurons possessed no intrinsic repair mechanisms. However, a key
discovery to promote CNS neural repair was to provide damaged neurons a permissive
cellular environment that resembles the environment of neurons within the peripheral
nervous system (PNS) 4,5. These findings revealed that damaged CNS did not lack
intrinsic regenerative capability. Rather, CNS injury generates a toxic cellular
environment that fails to support damaged neuron tissue and inhibits regenerative
capability 6. This harsh environment has become one of the key therapeutic targets in SCI
research to promote repair of damaged CNS neurons.
2


The pathophysiology of SCI is highly complex process that is not well understood
at the molecular level due to the varying degrees of injury seen from patient to patient.
Instead, a general understanding of SCI is broken down into different phases of the injury
and the effects each phase has on the spinal cord. SCIs are classified into a primary and
secondary phase (Figure 1-1)7 Most traumatic SCI are caused by forces to the vertebral
column that result in compression, contusion, displacement, and/or laceration of the
spinal cord. This primary mechanical damage results in axon severing and substantial
damage to neurons, intemeurons, and glia within the spinal cord.
-x
-x
Acute stage:
1. Direct impact
2. Lacerated axons
3. Contused axons
4. Influx of inflammatory cells
Chronic stage:
1. Central cavity
2. Scar tissue
3. inhibition of axonal regeneration
4. Demyelination
5. Axonal dieback
6. Influx of inflammatory cells
Figure 1-1. SCI pathophysiology depicting the primary injury stage
caused by impact and the secondary injury cascade forming the glial scar
7
Treatment options:
1. Cell-, tissue transplantation
2. Blocking inhibition of axonal
regrowth
3. Enhancing axonal regeneration
4. Modulating inflammatory
responses
Within minutes of the primary damage, a cascade of secondary damage occurs at
the molecular level that leads to chronic neurological deficits. This cascade of secondary
events includes: (a.) vascular bleeding and ischemia that lead to necrosis of spinal tissue,
3


breakdown of the blood spinal cord barrier, and an influx of inflammatory cells (b.) the
formation of free radicals that cause oxidative damage to neurons, (c.) disruption of ionic
and amino acid balances (ex. glutamate excitotoxicity) that lead to apoptosis of neurons 1.
Ultimately, axons and gray matter are damaged and replaced with connective tissue
deposits that lead to the formation of a glial scar at the site of the injury. The glial scar
consists of reactive astrocytes, microglia, macrophages, and chondroitin sulfate
proteoglycans 1. This scarring acts as a barrier to the extension of axons through the
injury, preventing regeneration and replacement of damaged spinal tissue.
Overall, a general understanding of the direct effects of SCI is available, but
knowledge of specific biological mechanisms and pathways remains elusive.
1.4 Current clinical SCI treatments
The complexity and overall lack of understanding of SCI pathophysiology has
made new treatment approaches difficult to take to the clinical level. Surgical
intervention and methylprednisolone (MP) are two of the few treatment approaches
available to patients with SCI, however, even these approaches are now marred in
controversy between physicians questioning their overall efficacy.
1.4.1 Surgical intervention
The priority for patients with an acute SCI is immediate spinal immobilization to
prevent further primary damage to the spinal cord. Even the slightest movements to an
injured spinal cord can exacerbate any existing primary damage. In addition to the
primary damage, prolonged compression of the spinal cord contributes to the long term
cascade of secondary damage to spinal cord tissue at the molecular level. Surgical
4


decompression and realignment of the injured spinal cord helps stabilize spinal blood
flow, reducing the effects of long term ischemia. Patients that have undergone spinal
decompression surgeries have shown significantly improved neurological outcomes, but
the timing of performing these surgeries remains controversial8.
1.4.2 Pharmacological intervention
The administration of the corticosteroid MP is currently the only approved
treatment for patients with acute SCI that specifically targets the secondary injury
cascade. MP was found to be effective in combatting the damaging free radical oxidation
of neural lipid membranes that occurs during the secondary effects of SCI 9 Despite early
reports that administration of MP improved neurological outcomes when given 8 hours
after acute SCI, further studies have shown that there may be no short term or long term
benefits 8. In fact, the immune suppressive effects of MP has been found to have severe
side effects, including gastrointestinal bleeding, infections (wound, respiratory, and
urinary), sepsis, peptic ulcer disease, and hyperglycemia 10. Despite MP being one of the
few SCI treatments around, it is no longer prescribed by many physicians due to these
recent findings 9 In addition, some physicians that do prescribe MP but are uncertain of
the risks often do so out of fear of litigation 10, u. Nevertheless, the administration of MP
remains a highly controversial subject for SCI treatment12.
1.5 Emerging therapies in SCI treatment
Despite the complexity of SCI pathophysiology, advances in cellular and
molecular research have advanced the understanding of SCI mechanisms, opening up
new avenues for treatment strategies. Each new approach is designed to target the
5


secondary SCI cascade to help create an environment within the injury site to promote
neural regeneration.
1.5.1 Molecular approaches
1.5.1.1 Neuroprotective approaches
The secondary cascade following SCI involves a complex inflammatory response
that has both protective and damaging effects on spinal tissue 13. Researchers are
exploring molecular techniques that exploit the beneficial effects of this inflammation
and combat the detrimental effects. Interleukin-10 (IL-10) is a known anti-inflammatory
cytokine that is capable of inhibiting monocyte and macrophage production in PNI.
Systemic administration of IL-10 into rats with a contusion SCI showed significant
reduction of glial scar size and also showed improved hind limb motor function 14
In addition to the inflammatory response, the secondary injury cascade also
results in the formation of a glial scar. This scar formation acts as a physical barrier to
axonal regeneration through the SCI site, contributing to long term neurological and
functional deficits. Following injury, astrocytes enhance the expression of extracellular
matrix (ECM) proteins, particularly chondroitin sulfate proteoglycans (CSPGs) 15.
Enzymatic degradation of CSPGs by chondroitinase ABC (ChABC) promotes a more
accommodating environment within the lesion for axonal regeneration and enhanced
motor recovery l.
1.5.1.2 Neurotrophic factors
Neurotrophic factors (NTFs) activate signaling pathways in developing neurons to
induce growth, guidance, differentiation, and survival of neural cells (Figure 1-2).
6


NT-3
NGF

BDNF NT-4/5
0[> #>
Ras-Rat-MAPK
PI3K. PLC
Pro-NGF Pro-BDNF
4>
JNK
NFkB
Cell survival. Cell death
differentiation,
synaptic plasticity
Figure 1-2. Signaling pathways activated by neurotrophic factors.
Previous studies of peripheral nerve injury (PNI) showed NTFs play a key role in axon
growth and regeneration, making NTFs a promising therapeutic approach for SCI1.
Administration of brain-derived neurotrophic factor (BDNF) has demonstrated extensive
axonal growth at SCI sites in a rat model16, regeneration of severed axons, axonal
myelination, and enhanced neural plasticity 17. Ciliary neurotrophic factor (CNTF)
treatment in a rat hemisection SCI model showed significant neuron regeneration and
axonal growth, further enhancing the promising application of NTFs to promote recovery
after SCI18. A combination of these and other NTFs will likely activate more NTF
signaling pathways to promote an environment in the SCI lesion capable of repair and
regeneration 19
1.5.2 Cellular approaches
Patients with severe SCI are often left with limited treatment options from their
physicians. Until recently, there was no known regenerative capability of CNS tissue.
Fortunately, recent cellular research has shown that several different types of neural cells
7


and stem cells could provide injured spinal cord tissue the ability to repair and provide
significant functional recovery.
1.5.2.1 Olfactory ensheathing cells (OECs)
One of the first CNS regions to demonstrate continuous regenerative capability
was observed in the olfactory nerve. Olfactory axons are peripheral nerves located in the
olfactory mucosa (OM) that undergo constant regeneration. As these axons grow, they
extend into the olfactory bulb (OB), located in the brain 1. OECs are believed to be the
driving force to support and guide this neural regeneration, making OECs a potential
therapeutic approach for SCI 20 21. A spinal cord compression injury in rats with
transplanted OECs into the injury site demonstrated reduced glial scar formation and
axon regeneration across the lesion site 22, while a transection SCI model in rats with
transplanted OECs into the injury site also showed axonal regeneration 23. Human trials
have also shown sensory and motor improvements in multiple patients with OEC
transplantation 24. Concerns about the long term effects of OEC transplants have also
been reported. One patient received an OEC transplant into the injury site three years
after the injury and experienced back pain 8 years after the transplant, likely caused by
the development of a cystic mass at the site of the transplantation, which required
removal 25.
1.5.2.2 Mesenchymal stem cells (MSCs)
MSCs are an attractive option for SCI treatment because they are easily
accessible, plentiful, and have an immunosuppressant effect1. Bone marrow MSCs were
found to be effective in improving functional scores in animal models of traumatic SCI
26. Human trials using MSCs reported significant bladder and bowel control
8


improvements, in addition to motor improvements 27 However, further patient studies
with MSC transplantation have reported significant side effects, including urinary tract
infections, respiratory distress, and pain 28.
1.5.2.3 Neural stem cells (NSCs)
NSCs are multipotent cells that can differentiate into any cell type in the central
nervous system, which has made them a promising therapeutic candidate for SCI 28. The
aim of NSCs for SCI treatment is transplantation into the injury site to allow the cells to
differentiate into oligodendrocytes and other neurons that can replace damaged tissue
(Figure 1-3) 29. When grafted into SCI sites, NSCs were able to differentiate into
multiple neuronal phenotypes and also exhibited long distance axon extension 30. NSCs
also have beneficial effects after SCI within the glial scar, acting to preserve tissue
integrity and supply neurotrophic factors, essentially protecting the spinal cord from
further damage 31.
Figure 1-3. NSC therapy for SCI (B) aims to reduce glial scarring and
improve axon myelination to restore normal function (A)32
1.5.3 Challenges of emerging therapies
The results of cellular and molecular therapy approaches have demonstrated
promising capability to combat the challenges of SCI tissue repair. The success of these
9


new approaches, however, has been highly variable and limited. Most molecular or
cellular agents are delivered by direct injection into the injury site, which have problems
with localized delivery to targeted regions. Delivering cellular therapies present a
particular challenge for cell viability. Direct injections of cellular therapies into injured
spinal tissue often results in substantial cell death due to the harsh environment of injury
site. This environment lacks the necessary ligands for transplanted cells to attach to,
causing cell death.33.
Biomaterial scaffolding systems have been developed to create more effective
delivery methods for cells, drugs, and molecular agents for many different tissue
engineering applications. These scaffold designs mimic the natural ECM environment to
support cellular adhesion, differentiation, and proliferation. Administration of cellular
and molecular agents within a biomaterial scaffold will be crucial to enhance therapeutic
efficacy to promote neuron support, guidance, and repair across the SCI site.
1.6 Goal of this study
Treatments for SCI remain ineffective, however, the development of new cellular
and molecular therapies bring promising potential for significant repair of damaged
spinal tissue. Without a scaffolding system to act as a supportive substrate, the efficacy
of these potential therapies will remain limited and inconsistent. Synthetic polymers are
practical biomaterials that can be deployed to act as scaffolding devices for these
potential therapies to help ensure the most effective means to deliver the treatment.
Polymer scaffolds properties can be modulated for different delivery applications to help
support nerve regenerative capability. Many types of polymer scaffolds have been
10


developed but also present limitations with delivery due to invasive procedures needed
for implantation.
The overall objective of this study is to develop a novel, polyethylene glycol
(PEG) based injectable polymer scaffold that can act as a supportive substrate for injured
spinal cord tissue. This polymer surface will be functionalized with a cellular-adhesive
laminin peptide to allow the polymer scaffold to mimic the ECM. This surface
modification will allow the polymer to interact with and cue neural integrin receptors to
induce cellular activities, such as growth, proliferation, and differentiation. The
injectable properties of this polymer provides a minimally invasive delivery method to
targeted tissue while also providing a gelled substrate to support therapeutic agents. This
injectable polymer, known as a reverse thermal gel (RTG), will be evaluated for neuron
support in an in-vitro retinal ganglion cell model and will be assessed for regenerative
capability in a rat model of compression SCI.
11


2 Literature review
2.1 Polymers scaffolds for SCI
A variety of different materials have been explored as potential tissue engineering
scaffolds for SCI. Scaffold designs for SCI need to take into account biocompatibility,
biodegradability, porosity, mechanical stability, and cellular adhesion (Figure 2-1) 34. As
SCI pathophysiology is complex and generally not well understood, sorting out the best
approaches for scaffold designs has become a difficult endeavor.
Btod*g(dalion
Figure 2-1. Desirable properties of polymer scaffolds for SCI tissue
engineering applications 34
Polymer scaffolds are made of natural or synthetic materials, or often a
combination of the two. Naturally derived biopolymers have shown excellent potential
for regenerative medicine approaches. These materials include collagen, fibrin,
hyaluronic acid, chitosan, polysaccharides, and peptides 35. These natural biomaterials
12


offer excellent biocompatibility, minimal cytotoxic effects, and mimic the ECM for
optimal tissue support and cellular attachment36. Naturally derived polymers, however,
are limited by generally poor mechanical properties, are often costly to obtain, and have
limited chemical modification sites 37.
To overcome the disadvantages of naturally derived polymers, synthetic materials
have been explored as tissue engineering scaffolds. Synthetic biodegradable polymers
include polyvinyl alcohol (PVA), polycaprolactone (PCL), poly (lactic acid) (PLA), poly
(glycolic acid) (PGA), poly(lactic acid-co-glycolic acid) (PLGA), and PEG37. Synthetic
materials offer the ability to control desired scaffold properties for optimal integration
into tissue, including mechanical properties, porosity, degradation, and biomolecule
chemical conjugations.
2.1.1 Biomedical applications of PEG
PEG has become one of the most researched biomaterials for tissue engineering
scaffold devices. PEG exists in linear or branched forms with a variety of molecular
weights (MWs) and a wide variety of chemical properties. Most biomedical applications
of PEG use MWs of several hundred to approximately 20,000 38. An excellent reference
for the biomedical applications of PEG can be found in a textbook written by J. Milton
Harris, Poly (ethylene glycol) Chemistry Biotechnical and Biomedical Applications.
From a chemical reaction perspective, PEG has become a popular selection for
biomedical applications with its wide range of solubilities in different organic solvents.
Possessing a wide range of solubilities in organic solvents increases the number of
different chemical conjugation techniques that can be applied to a reaction, thereby
13


increasing PEGs flexibility for tunable biomedical properties and applications. Similarly,
PEG is also insoluble in organic solvents like ether and hexane. This makes PEG a user
friendly polymer for the precipitation and the purification of final reaction products.
In addition to its excellent solubility properties, the terminal hydroxyl end groups
of PEG can be covalently bonded to other polymers or molecules. PEG is often attached
to hydrophobic polymers, such as polyurethanes, to enhance the polymers support for
physiological applications 38 39 40. Furthermore, modifying polymers with PEG also
increases their overall size, a property that has been utilized to retain biomolecules within
systemic circulation for longer periods of time. This approach has been used in the
pharmaceutical industry to prolong the half-life of drugs in circulation by decreasing their
clearance by the renal system 41.
From a biochemical perspective, the hydrophilic properties of PEG are what make
it a promising substrate for tissue engineering applications. PEG is water soluble,
nonimmunogenic, and mostly inert when exposed to biological materials. The Food and
Drug Administration (FDA) has also classified PEG as non-toxic 40. A common
application in tissue engineering for PEG is the formation of 3-dimensional (3D) cross-
linked scaffolds. A 3D, cross-linked polymer scaffold results from long polymer chains
covalently binding to one another, forming a network of polymer chains that are insoluble
in water. These cross-linked polymer networks, often referred to as hydrogels, are able to
absorb large quantities of water. The properties of the hydrogel (mechanical and
degradation) can be tuned by the amount of chemical crosslinks in the network, which are
controlled by various reaction mechanisms 42. The large volume of water retained by
hydrogels makes them attractive scaffold designs in tissue engineering applications.
14


Specifically, the high water content environment is able to mimic physiological
conditions and allow exchange of nutrients and wastes within the polymer matrix 43.
These properties of PEG make it an attractive biomaterial for tissue engineering
applications. In-vitro models using 3D PEG hydrogels encapsulated with neural cells
showed survival, proliferation, and differentiation into functional neurons that can
respond to neurotransmitters 44 Injectable PEG scaffolds with poly (N-
isopropylacrylamide) (PNIPAm) have also been investigated for mechanical properties
and neurotrophin release. In vitro analysis revealed that a controlled release of BDNF
and NT-3 from the polymer scaffold for four weeks. This PEG-PNIPAm scaffold was
also designed to have similar mechanical compressive strength to spinal tissue 45. These
results show that a PEG-PNIPAm scaffold can serve as a potential treatment approach for
SCI by effectively delivery therapeutic agents and withstanding mechanical forces within
the spinal cord.
PEG has also shown promising potential to inhibit the damaging inflammatory
effects that occur after SCI. Part of the secondary cascade of SCI involves lipid
peroxidation reactions that degrade neural membranes and axonal components. An in-
vitro guinea pig SCI model showed that PEG can act to restore cell membrane structural
integrity and decrease free radial lipid peroxidation of neurons 46. The authors found that
PEG itself is not capable of decreasing reactive oxygen species, but instead, may be able
to block oxidative damage to neuron cell membranes by supporting membrane integrity
after SCI47 Further studies using an in-vivo compression SCI model in guinea pigs
showed that a subcutaneous injection of a PEG solution 6 hours after the injury promoted
significant improvements in somatosensory and reflex tests 48. This ability to provide a
15


neurons protection from the oxidative stresses of SCI, even without the use of additional
biomolecules, has made PEG an attractive biomaterial for SCI applications.
2.1.2 SCI scaffold designs
The flexibility and control over the properties of polymer scaffold designs has
made them promising biomaterials to act as supportive substrates for SCI repair. The
different stages of SCI pathophysiology have led to the development of a variety of
scaffold designs with aims to repair damaged/lost tissue, support undamaged tissue, and
promote axon growth across the lesion site 49. Incorporating cellular and molecular
therapies into tissue engineered scaffolds has been the most effective approach for
significant axon regeneration and motor recovery from SCI1.
2.1.3 Surgically implanted scaffold designs
When designing a scaffolding system for SCI, the delivery method of the scaffold
to the injured spinal cord is an important consideration to take into account. Surgically
implantable scaffolds are a common approach to delivering therapeutic agents. The main
advantages of these scaffold designs are the fine control of the mechanical properties of
the scaffold to match the spinal tissue 33 and control over the microstructure desired. One
implantable scaffold design using poly (2-hydroxyethylmethacrylate) (PHEMA) was
tailored to match the mechanical modulus of spinal tissue. This hydrogel was then loaded
with BDNF and implanted into a transection SCI model in rats, results showing axon
regeneration through the hydrogel50. A fibrin based scaffold design implanted into a
similar rat model two weeks after the injury promoted neural sprouting and reduced
astrogliosis at the lesion site 51. Implantable designs are also capable of supporting long
term survival of implanted cells. A N-(2-hydroxyproplyl) methacrylamide (HPMA)
16


hydrogel conjugated to RGD peptide seeded with MSCs showed significant functional
improvements, infiltration of axons into the hydrogel, and survival of MSCs 5 months
after implantation 52
Many implantable treatment approaches are able to show axon growth but the
arrangement can often be disorganized, resulting in ineffective functional recovery 53.
Implantable polymer scaffold designs with microporous channels have shown superior
axon growth by providing axons linear guidance through the scaffold. These designs are
also capable of supporting cellular grafts and delivery of therapeutic agents to cue neural
extensions, such as BDNF 53,54,55.
These and many other studies demonstrate that implantable polymer scaffold
designs are capable of supporting cellular and molecular therapies that promote
significant spinal cord tissue regeneration and motor improvements.
2.1.4 Injectable scaffolds
Despite the advantages implantable scaffold designs hold, invasive surgical
techniques are required to implant the scaffold into the injury site. These techniques
often require excising spinal tissue that may result in long term complications. For
instance, a polymer channel scaffold implanted into the SCI site of rats showed limited
axon growth and limited functional improvements due to the development of
syringomyelia 56. Surgically implanted scaffolds will also cause disruption of the dura
mater, resulting in significant fibrotic tissue scaring and possible long term cerebral
spinal fluid (CSF) leaking 57.
17


Injectable scaffolding systems have been developed to counter the invasiveness of
implantable scaffolds. These injectable designs form scaffolds in situ, filling and
conforming to the lesion site to help reestablish tissue support58. Most importantly,
injectable scaffolds do not require removal or laceration of spinal tissue. Like
implantable scaffolds, injectable scaffolds can also be designed to have a porous
microstructure and can act as delivery agents for cellular, pharmaceutical, or molecular
therapies. Unlike implantable scaffolds, injectable scaffolds are more challenging
designs to exhibit control over mechanical properties. Scaffolds without good
mechanical properties could act to further damage within the SCI lesion, collapse under
the stresses of other tissue, or become solubilized and lost.
Immediate transition from a soluble solution to a stable gel is critical for injection
applications into spinal tissue. Different chemical designs have been fabricated for this
important scaffold property. Photopolymerization of an injected polymer matrix is a
common approach to form stable gels. One study showed that a PLA-PEG-PLA
hydrogel loaded with Neurotrophin-3 (NT-3) showed controlled release of the trophic
factor that aided in significant axon growth and functional improvements in a rat SCI
model59 Though excellent results, this method required 60 seconds of
photopolymerization for gelation to occur. This causes tissue to be exposed to UV light
and could be damaging to neurons survival. One alternative to photopolymerization
methods for in situ gel formation was attempted with an agarose gel encapsulated with
BNDF loaded lipid microtubules. The solution was injected into injured rat spinal cord
tissue and gelled by applying cooled nitrogen gas over the spinal cord for 30 seconds.
18


These gels promoted neurite extensions within the scaffold and reduced the size of the
glial scar at the lesion site 60.
2.1.5 Reverse thermal gels (RTGs)
Another approach to developing an effective injectable polymer scaffold that does
not require an outside source for gel formation is to create a temperature responsive
system. Polymers have been designed that contain both hydrophilic and hydrophobic
components. The properties of these polymers change dramatically with slight changes
in temperature. At room temperature, these polymers exists as aqueous solutions that are
injectable through small gauge needles. When these polymers are exposed to increased
temperatures, the hydrophobic components begin to aggregate, allowing for the formation
of solid gel structures (Figure 2-2).
Room Temperature 37C
Figure 2-2. A 10% RTG solution (mg/mL) at room temperature (left)
followed by 10 seconds of exposure to 37C water shows formation of
solid gel (right).
These polymers undergo reversible physical transitions from aqueous solution to
physical gels in response to temperature changes, and are known as reverse thermal gels
(RTGs) (also called temperature sensitive or thermal responsive hydrogels). Like
19


implantable scaffolds, these formed gels can act as scaffolding devices for tissue support
and repair61. The main advantage RTG designs have over other injectable scaffold
designs is fast gelation upon exposure to physiological temperature, without the need for
photopolymerization, organic solvents, or chemical cross linking techniques that could be
toxic to tissue.
A mixture of a hyaluronic acid and methylcellulose (HAMC) was injected into the
intrathecal space of compressed rat spinal cord tissue. HAMC is unique because it exists
as an injectable gel at room temperature but the gel strength increases upon exposure to
increased temperature. After a one month study, these gels were found to be compatible
with spinal cord tissue and could also provide functional benefits 62. Similarly, a PEG-
based polyurethane RTG was loaded with bone marrow stem cells and injected into a
contusion SCI model. The cells had significantly higher survival when encapsulated
within the RTG and no significant difference in immune response when compared to a
PBS control. Rats with the RTG injection also showed improvements in openfield
locomotion 63. These results demonstrated how RTGs are capable of improving the
survival and delivery of cellular treatments to SCIs.
2.1.6 Previous research with RTGs
Our laboratory has recently developed RTGs for a variety of different applications
to act as tissue engineering scaffolds. PSHU-PNIPAm is a novel RTG composed of a
poly (serinol hexamethylene urea) backbone (PSHU) conjugated to a water soluble RTG,
PNIPAm. PSHU (by itself) is a biocompatible, hydrophobic, linear polymer with
multiple urea molecules in each repeating unit that allows the backbone to resemble a
protein structure. Each repeating unit of the PSHU backbone has a reactive primary
20


amine group that is used to conjugate to biomolecules or other polymers. In order to
make PSHU a hydrophilic and injectable solution, it needs to be conjugated to PNIPAm.
PNIPAm is a prevalent, water soluble RTG used in many polymer applications due to its
remarkably sharp lower critical solution temperature (LCST) of 32C. This LCST is
advantageous for tissue engineering applications because it is well below the
physiological temperature of 37C, allowing for fast gelation upon exposure to tissue
temperatures. The primary amine groups of PSHU can also be used to conjugate
biomolecules, drugs, growth factors, and/or other therapeutic agents to create a
biomimetic polymer that will directly interact with and affect cellular mechanisms.
PSHU conjugated to RGD (discussed next) and PSHU-PNIPAm conjugated to RGD
demonstrated enhanced cell viability and differentiation while investigating axon
sprouting of PC12 cells 64 PSHU-PNIPAm can be loaded into a syringe and injected by
a small gauge needle into tissues or other physiological specimens, allowing for the
formation of a stable gel within tissues that can serve as a biomimetic polymer scaffold.
The scaffold will degrade over time and release its contents in a controlled manner,
allowing for sustained delivery of its therapeutic contents to tissues 65.
2.1.7 RGD peptide
A key component all polymer scaffolding designs require for viable interactions
with host cells are cellular adhesive peptides. Cellular adhesive peptides are critical for
cellular anchorage, migration, proliferation, differentiation, and apoptosis pathways 66.
Without these adhesion molecules, polymer scaffolds will not mimic the ECM
environment, which is necessary to allow the polymer to interact with cellular integrin
receptors and influence signaling pathways 67. Most polymer scaffolds by themselves do
21


not contain ligands capable of attaching to integrin receptors, therefore, conjugation of
adhesion peptides to the polymer structure is a necessity. One of the most common
peptides used to encourage cellular attachment for biomimetic scaffolds is arginine-
glycine-aspartic acid (RGD). This specific peptide sequence is found in many ECM
proteins, including fibronectin, laminin, and some collagens. This peptide sequence acts
as ligand that binds to cell receptors, influencing cellular activities through signaling
pathway activations 68. Integrin receptors are of particular interest for RGD binding 66.
These adhesion proteins are expressed on the surfaces of cells and are responsible for
interactions between the cell and the ECM. When RGD binds to the integrin recognition
site, a series of signaling pathways become activated that influence cellular behaviors
(Figure 2-3).
t
IMBEI
Extracellular
,lgrtn
oT p
<
s:g^>ay

Figure 2-3. Schematic of RGD integrin binding to trigger multiple
signaling pathways. Conjugation of RGD to the polymer backbone allows
the polymer to mimic the ECM environment and bind to integrin proteins.
In order to model properties of the ECM, polymer scaffolds can be designed to
include the RGD sequence by using chemical reactions. Functional groups on the
polymer backbone, such as amine, hydroxyl, or carboxyl groups, are able to be modified
22


by chemical means to allow stable, covalent bonding of the RGD sequence to the
polymer backbone 67. In-vitro studies have successfully integrated RGD into the
backbone of the polymer scaffolds to encourage cell spreading and survival 70. Our lab
demonstrated that PSHU-RGD and PSHU-PNIPAm-RGD were able to support
proliferation and differentiation of PC12 cells, while the same polymers without RGD
saw limited to no cell viability 64 PSHU-RGD was also able to promote substantial
differentiation of human NSCs into motor neurons (Figure 2-4) 11.
Figure 2-4. Polymer scaffold with conjugated RGD showed substantial
differentiation of human NSCs into motor neurons. Inset is a schematic of
the polymer. Green represents axon extensions via piII-Tubulin staining.
Red represents motor neuron markers (Islet-1, HB9) 71.
These results demonstrated the critical importance of incorporating RGD into the
polymer backbone to enhance neural survival, growth, and differentiation. Development
an injectable, PEG-based RTG functionalized with RGD for SCI would introduce a
minimally invasive therapeutic approach that provides not only structural support to
damaged tissue, but also the ability to mimic the ECM environment and directly
influence cellular activities.
23


2.2 Animal Models of SCI
New tissue engineering approaches to treat SCI will require animal models that
resemble the effects of SCI in humans. As this literature review has shown, rat models
of SCI are the most commonly used. Rat models are inexpensive, allow for large sample
numbers, and have good anatomical size for surgical manipulations 72. Murine SCI
models offer the capability to alter genetics to study specific molecular pathways of SCI
73. However, murine models are less convenient because of the small anatomical size of
the spinal cord, which make surgical techniques and therapy placement more difficult.
Larger animal models, such as dogs, cats, pigs, sheep, and primates, are less
common in SCI research because of higher expenses, animal welfare concerns, and
limited understanding of when these models will be necessary for good translation to
human subjects 74 Larger animal models have been used in the development of new SCI
methods and the assessment of current SCI treatments. A canine model was used to
develop a new compression injury using a novel balloon technique, offering the
advantage of an SCI model requiring no laminectomy 15. A canine model was also used
to compare MP treatment to surgical decompression after SCI, results showing that
surgical decompression surgery with or without MP improved neurological function
when compared to MP treatment alone 76, further questioning therapeutic effects of MP.
Studies using new tissue engineering approaches are less common in large animal
models, but have been attempted. The degradation of a PLGA scaffold seeded with
NSCs was evaluated with four African green monkeys using a hemisection SCI model.
Though the study had a small sample size, researchers were able to demonstrate
successful integration of NSCs onto the PLGA scaffold for 82 days 77 Another study
24


with a canine model used Matrigel loaded with a neurotrophic factor and MSCs that
demonstrated neuronal extension, reduced glial scarring, reduced inflammation, and
functional improvements 78. Despite the importance of establishing larger animal SCI
models to further innovative therapeutic approaches, studies for tissue engineering will
continue to use rodent models until a better understanding and reproducibility of results
are obtained.
2.2.1 Rat models of SCI
Different SCI models have been developed to simulate the primary and secondary
pathophysiological effects of the injury. The three most common models used by
researchers involve contusion, compression, or transection injury. Each of these three
methods are able to target different severities of SCI and may also be suited for different
therapeutic goals. For instance, implanting solid scaffolds into spinal tissue will require a
transection model. If the goal of the research is to understand the pathophysiology of an
injury and the response to a treatment, contusion or compressive injuries are used. Each
of these three SCI models have benefits and drawbacks, and will be discussed below.
2.2.1.1 Contusion SCI animal models
Falls and motor vehicle accidents are the most common causes of SCI. These
events cause high impact forces to the spinal column, resulting in vertebral column
fractures, spinal cord displacement, and contusion injury to the spinal tissue. The most
common animal model of SCI used by researchers is the contusion model, as it may best
reproduce the acute effects of high impact trauma experienced by patients 72. This
method involves performing a laminectomy to expose the spinal cord tissue, followed by
subjecting the spinal tissue to a high impact force, such as a dropped weight or a
25


calibrated impactor device 79,80,81 Contusion models offer precision and flexibility in
the severity of contusion applied to simulate different degrees traumatic SCI seen in
patients and generally have good reproducibility. However, these models are often result
in incomplete damage of neuron cells and axons, which makes assessment of neural
regeneration more difficult 72.
2.2.1.2 Compression SCI models
To better mimic the chronic stages of SCI, compression animal models have been
developed to evaluate neuroprotective approaches and to gain a better understanding of
the effects of long term SCI pathophysiology 82 This method requires laminectomy to
expose spinal tissue, followed by compression of the spinal cord by either a calibrated
clip (often an aneurysm clip) or forceps. The force applied by the compressing object
and the length of compression time can easily be changed to vary the severity of the
injury. This method is believed to be the most clinically relevant SCI model because of
the similar chronic pathological deficits and motor impairments 82 However, application
of the compressive object can be difficult. This model requires the compression device to
completely surround the lateral sides of the spinal cord. Removing enough lateral
vertebrae by laminectomy to expose this much spinal cord requires more surgical skill
than simple dorsal laminectomy used in contusion models.
2.2.1.3 Transection SCI models
Compression and contusion models often do not cause a complete severing of
neural tissues after SCI, making it difficult to distinguish between axons that are
regenerating and axons that are still intact despite being injured. In order to better assess
axon regeneration, a model of complete SCI severance was developed. The advantage
26


transection SCI models have over contusion and compression models is the ability to
sever specific spinal tracts for injury without having to take into account the possibility of
damaged but intact tracts 82 However, complete or partial severing of the spinal cord is
rare in SCI patients, making this model the least clinically relevant1. Transection models
also do not reflect the pathophysiology of clinical SCIs that contusion and compression
models are better able to mimic 83.
2.2.2 Functional testing
To assess the efficacy of an SCI treatment, functional tests have been developed
in animal models that allow for both qualitative and quantitative results. Different animal
SCI models result in varying degrees of motor loss that change over time. Therefore,
choosing functional assessment will depend on the severity of the injury. There is no
consensus on which test provides the most meaningful recovery results and there is no
standard functional test to use for a specific injury type. The most commonly used test in
many SCI animal studies is the 21-point open field locomotion score (BBB rating). This
test assesses the hindlimb movement and trunk stability of rats in a large open area.
Monitored movements include hip, knee, and ankle movements, toe clearance, paw
position, forelimb-hindlimb coordination, trunk position, and tail position 84 Normal
locomotion receives scores of 21, while rats after SCI receive a score of 0, as most SCI
models result in several days of complete loss of hindlimb function. This test is generally
believed to be sensitive and reproducible for mild and moderate SCI injury models, but is
less reliable for more severe SCI models 85. In addition to open field testing, another
useful analysis of gait coordination is the walking track analysis (also known as footprint
analysis), which quantifies the placement of the hindlimb feet. This is a common method
27


to evaluate PNI regeneration in sciatic nerve injury models and also has been used in SCI
models 86~88. Functional tests have also been developed that assess a rats balance
capabilities while maintaining good gait coordination. Grid walk and latter walk tests
monitor how well the rat can place and grip their forelimbs and hindlimbs on rungs that
are spaced apart. The number of footfalls through the rung gaps are counted for a general
assessment, while more detailed examinations involve hindlimb recovery and angle of
paw placements 87,89 Many other functional tests exist, including kinematic analysis,
narrow beam walking, incline walk, and contact placing response. As the SCI model
becomes more severe, the ability of the animal to perform more complicated motor tests
becomes limited. The best approach to evaluate motor improvements of animals after
SCI is to use a variety of functional tests with varying difficulty 87
2.2.3 Immunohistochemistry (IHC)
After functional tests are completed, the spinal tissue is examined by IHC to study
the distribution and the localization of specific markers of healthy and damaged neurons.
Chemical and immunofluorescent staining techniques are often used together in SCI
models and are good complements to functional testing. Histological assessment using
Hematoxylin and Eosin (H&E), a chemical stain, shows robust structural information
such as the location and size of the SCI site, as well as the presence of scar tissue
formation. Luxol Fast Blue (LFB) is a more specific chemical stain that identifies myelin
and allows for differentiation between damaged and undamaged axons. To identify more
specific markers of nerve damage and regeneration, immunofluorescence methods are
used. Many different neural markers have been developed that are capable of assessing
different stages of SCI that have allowed for a better understanding of the short and long
28


term pathophysiology. These markers have become invaluable tools to evaluate the
efficacy of new SCI therapies. The glial scar is believed to be one of the main barriers to
axonal growth through the lesion site. To assess the formation and size of the SCI glial
scar, an antibody called glial fibrillary acid protein (GFAP) is used to assess the
localization of astrocytes 90. Targeting the glial scar for size reduction and axonal
penetration through the injury site has made GFAP one of the most common fluorescent
stains in SCI research 223057. Another common stain used to study the inflammatory
response after implementing a treatment for SCI is anti-CD68 (EDI). As mentioned
earlier, the inflammatory cascade after SCI contributes to significant neuron death and
axonal damage. The EDI stain is useful to evaluate the effect a therapy has on acute and
chronic stages of the SCI immune response 39 Specific neural components can also be
stained with immunofluorescence to evaluate the presence of axons that are both growing
and functional. The antibodies to neurofilament and beta-tubulin are useful for detecting
the presence of axons in SCI lesion areas to evaluate a therapies ability to penetrate the
glial scar30. Growth associated protein 43 (GAP-43) is an even more specific marker for
axon regeneration. The expression of this protein is a major component of elongating
axons that is an indication of favorable growth environments for injured CNS axons 91.
This stain can pinpoint areas where damaged axons and regenerating axons are located,
which makes it an attractive option to evaluate new SCI treatments 35.
29


3 Research objectives and experimental approach
3.1 Hypothesis
As previous studies have shown, polymer scaffolds present a promising approach
to deliver therapeutic agents to promote a regenerative environment in the injured spinal
cord. We hypothesize that a novel, PEG-based injectable RTG functionalized with RGD
will promote neural support and axon regeneration in a compressive SCI rat model.
3.2 Specific aims
(1) Synthesize and characterize a novel PEG polymer backbone
The first step for this work is to create a novel PEG-based polymer to serve as the
backbone for further chemical conjugations. As previously mentioned, PEG can provide
protective support to damaged neural tissues and has biocompatible and biodegradable
properties in multiple in-vitro and in-vivo studies. Many developed polymer backbones
have few chemical functional groups to promote conjugation to biomolecules, thus,
limiting the polymers capacity to act as an ECM scaffold to support cellular activity.
This novel PEG polymer backbone will have two functional hydroxyl groups per
repeating unit to provide greater biomolecule conjugation capability. To characterize the
structure, Fourier transform infrared spectroscopy (FT-IR) will be used to verify the
appearance of distinct hydroxyl and carbonyl groups in the polymer backbone.
(2) Conjugate PNIPAm to PEG backbone to create injectable RTG
After verifying the structure of the PEG backbone, PNIPAm will be chemically
conjugated to the free hydroxyl groups. PNIPAm conjugation confers temperature
sensitivity to the new polymer structure that allows it to exist as an aqueous solution at
30


room temperature and form a physical gel when exposed to physiological temperature.
PNIPAm provides a sharp transition at 37C to promote the fast transition from solution
to gel, which provides a practical injectable property to the polymer. The gelled polymer
forms the scaffold that will support cellular activity. This novel RTG structure will be
characterized by FTIR and the gelling properties will be verified using ultraviolet-visible
light (UV-VIS) spectroscopy and scanning electron microscopy (SEM).
(3) Conjugate RGD to PEG backbone to synthesize biomimetic RTG
In order for this new RTG to act as a biomimetic ECM scaffold, the polymer
backbone needs to have cellular adhesion properties. As discussed earlier, RGD is a
common laminin peptide used to confer cellular attachment to synthetic scaffolds. RGD
will be conjugated to the hydroxyl groups of the PEG backbone to create an injectable,
biomimetic RTG capable of promoting cellular activity. The large number of free
hydroxyl groups available on the PEG backbone allows for greater RGD conjugation and
therefore, more influence on cellular activity. Conjugation of RGD to the PEG backbone
will be verified using FTIR.
(4) In-vitro testing of biomimetic RTG for cytotoxicity and CNS axonal support
After synthesis of the novel biomimetic RTG, the cytotoxicity will be tested with
Alamar Blue and MTT assays to verify biocompatibility of the polymer. Retinal
ganglion cells (RGCs) will then be seeded within the polymer solution and gelled to
assess the polymers ability to support CNS axon growth and support.
Immunohistochemistry will verify axon extensions throughout the polymer network.
31


(5) Inject biomimetic RTG into compressed rat spinal cords to evaluate axon regenerative
capability
Finally, to assess this RTGs ability to act as a biomimetic scaffold for SCI, this
RTG will be injected into compressed rat spinal cords. As previously discussed, PEG
and RTGs have promising potential to form supportive cellular scaffolds for injured
neural tissues. There are two essential properties of this novel PEG based RTG that make
it an attractive therapy for SCI: (1) its ability to be injected into damaged tissues without
invasive implantation procedures, (2) to deliver RGD adhesive peptides to promote
cellular attachment and survival of damaged spinal neurons. To study the effect this
polymer has on damaged spinal cord tissue, a compression SCI model in rats will be
utilized to provide a clinically relevant models to study the chronic effects of SCI. We
believe the biomimetic properties of this injectable RTG will promote a supportive ECM
environment to induce axonal regeneration and function recovery after SCI.
32


4 Materials and equipment
4.1 Materials
Poly(ethylene glycol) diglycidyl ether (PEGDGE, Mn 526), triphenylphosphine
(TPP), 4, 4-azobis (4-cyanovaleric acid) (ACA), paraformaldehyde (PFA), sucrose,
nerve growth factor (NGF), dimethyl sulfoxide (DMSO), bovine serum albumin (BSA),
and heparin were purchased from Sigma-Aldrich (St. Louis, MO, USA). Succinic acid,
tetrahydrofuran (THF), anhydrous methanol, 4-Dimethylaminopyridine (DMAP), N, N-
Dicyclohexylcarbodiimide (DCC), N-(3 Dimethylamino-propyl)-N'-ethylcarbodiimide
hydrochloride (EDC), chloroform-d, and N-hydroxysuccinimide (NHS) were purchased
from Alfa Aesar (Ward Hill, MA, USA). Anhydrous N, N-Dimethylformamide (DMF)
was purchased from EMD Millipore (Billerica, MA, USA). Dichloromethane (DCM),
Acetone, and 30% hydrogen peroxide were purchased from BDH Chemicals through
VWR (Radnor, PA, USA). N-Isopropylacrylamide (NIPAAm) was purchased from TCI
Chemicals (Portland, OR, USA). Anhydrous diethyl ether was purchased from Fisher
Scientific (Pittsburgh, PA, USA). Gly-Arg-Gly-Asp-Ser (RGD) was purchased from
Biomatik (Wilmington, DE, USA) and Cellmano Biotech Limited (Hefei, AnHui, China).
Sprague Dawley rats were purchased from Charles River Laboratories (Wilmington, MA,
USA). Sterile saline, isoflurane, ketoprofen, ketamine, buprenorphine, gentamicin, and
bupivacaine (0.5 % Marcaine) were purchased from MWI Veterinary Supply (Boise, ID,
USA). Vicryl 4-0 sutures were purchased from Ethicon (Somerville, NJ, USA). Optimal
cutting temperature (OCT) compound was purchased from Sakura (Torrance, CA, USA).
MTT Cell Proliferation Assay Kit was purchased from Invitrogen-Molecular Probes
(Carlsbad, CA, USA). Alamar blue cell viability reagent, Phosphate Buffered Saline
33


(PBS) and RPMI-1640 Medium was purchased from Thermo Scientific. Laminin coated
cell culture dishes and Laminin solution (1.75 mg/mL) were purchased from Discovery
Labware, Inc (Bedford, MA, USA). Goat serum, GFAP (mouse IgGl), GAP-43 (Rabbit
IgG), Alexa Fluor 488 (goat anti-mouse IgG), Alexa Fluor 594 (goat anti-rabbit IgG), and
SlowFade Diamond antifade mountant with DAPI were purchased from Life
Technologies (Carlsbad, CA, USA), pill-tubulin (goat anti-rabbit, IgG) was purchased
from abeam (Cambridge, MA, USA). Bm-3a (goat anti-mouse, IgG) was purchased from
Santa Cruz Biotechnology (Dallas, TX, USA). Fluoromount G with DAPI was
purchased from Electron Microscopy Sciences (Hatfield, PA, USA). Triton X-100 was
purchased from MP Biomedicals. Retinal Ganglion Cell Isolation Kits (rat) were
obtained from MACS Miltenyl Biotec (San Diego, CA, USA).
4.2 Equipment
Fourier transform infrared spectroscopy (FT-IR) was performed on a Nicolet 6700
FT-IR spectrometer. Samples were run on polyethylene infrared (IR) sample cards.
Polymer purification was accomplished using dialysis tubing (Spectrum Labs, Rancho
Dominguez, CA, USA). Polymer morphology was imaged using a JEOL JSAM-60101a
analytical scanning electron microscope (SEM) (Peabody, MA, USA). MTT assay and
Alamar Blue assays were performed using absorbance values on a BioTek microplate
reader (Winooski, VT, USA). Glass bottom culture dishes for RGC growth were
purchased from MatTek Corporation (Ashland, MA, USA). Compression aneurysm clips
(30 g) and surgical tools were purchased from Kent Scientific (Torrington, CT, USA).
Treatment injections into spinal tissue were performed using a 10 pi Hamilton syringe
(Reno, Nevada, USA). A 32 gauge needle was used for the injection (TSK Laboratory
34


(Tochigi-shi, Tochigi-ken, Japan). Tissue was sectioned using a CryoStar NX70 Cryostat.
Confocal images were taken with Nikon microscope.
35


5 Methods
5.1 Synthesis of PEGSA
PEGDGE (5 g, 9.51 mmol) was mixed with succinic acid (SA) (1.35 g, 11.432
mmol) for a 1: 1.2 molar ratio of PEGDGE: SA. TPP (0.25 g, 0.953 mmol) was added to
the PEGDGE, SA mixture in a 10% catalytic amount to total mol of PEGDGE. This
catalyst acts to abstract the protons on the carboxylic acids on succinic acid. No solvent
was used in this reaction. The mixture was placed onto an oil bath set at 120C with
stirring set at 150 rpm. This temperature was necessary for the carboxylic acid groups
on the succinic acid to open the epoxide rings on the PEGDGE. All reactants dissolved
at this temperature after several minutes. This mixture was reacted for 48 hours, with the
stir speed being decreased to 60-80 rpm after 24 hours. The mixture becomes viscous
after 24 hours due to polymerization and the stir speed needs to be decreased. Otherwise,
the stir bar may become stuck, halting mixing of reactants which will lead to crosslinking
of polymer chains.
After 48 hours, the orange-brown viscous fluid is dissolved in DCM to make a
pippetable solution, and then precipitated three times in ether. Each ether precipitation is
followed by rotary evaporation. The product is then dissolved in a minimal amount of
Milli-Q water and lyophilized for 48 hours. For optimal purification, after precipitation,
the fluid is dissolved in water and added to 3500 molecular weight (MW) cutoff dialysis
tubing and dialyzed against 1 liter of water for two days, with water changes every 24
hours. The product is removed from dialysis tubing and lyophilized for 48 hours,
resulting in a final product that is a dark orange, viscous fluid.
36


5.2 Synthesis of carboxylic acid terminated PNIPAm
NIPAAm (5 g, 44 mmol) and ACA (0.062 g, 0.2 mmol) were dissolved in 25 mL
of anhydrous methanol and then purged with nitrogen gas for 30 minutes. After purging,
the polymerization reaction was carried out at 68C for 3 hours with stirring. The
reaction mixture was precipitated dropwise into warm water (60C) and washed with
fresh, warm Milli-Q water. The polymer was dissolved at 4C overnight in Milli-Q water
and then purified in dialysis tubing (MWCO: 12,000 14,000 Da) against 1 liter of Milli-
Q water for 3 days at room temperature, with water changes every 24 hours. The purified
PNIPAm-COOH was lyophilized for 2 days and stored at room temperature.
5.3 Synthesis of PEGSA-PNIPAm
Gel permeation chromatography results revealed that the PNIPAm procedure we
used was making much larger MW PNIPAm than originally anticipated. Large MW
PNIPAm allowed us to tweak the conjugation procedure in later reactions. Therefore, two
approaches were to conjugate PEGS A to PNIPAm.
5.3.1 Molar Reaction
The original reaction used a 25% conjugation of PNIPAm to PEGS A. Since
PEGSA has 2 hydroxyl groups, 25% of these hydroxyl groups were calculated to be
conjugated to PNIPAm. The first step of the reaction was to activate PNIPAm. PNIPAm
(4.045 g, 0.4045 mmol), DCC (0.1 g, 0.485 mmol), and DMAP (0.01 g, 0.082 mmol)
were dissolved in anhydrous THF. DCC was added in 1.2 molar excess to PNIPAm, and
DMAP was added in a 0.2 molar catalytic amount. This mixture was reacted with
stirring at 55C for 24 hours under nitrogen atmosphere. After 24 hours, PEGSA (0.5 g,
0.81 mmol) was dissolved in anhydrous THF and stirred at 55C. The activated PNIPAm
37


was added dropwise to the PEGSA solution and reacted 24 hours with nitrogen
atmosphere at 55C.
After 24 hours, the reaction was added dropwise into excess ether for precipitation.
Solvent was removed by rotary evaporation. Milli-Q water was added to the white
product and dissolved at 4C. To purify, the product was placed in dialysis tubing
(MWCO: 12,000-14,000) and dialyzed against 1 liter of MilliQ-water for 48 hours, with
water changes every 24 hours. After dialysis, the solution in the dialysis tubing is filtered
through a 0.2 micron filter to remove insoluble reaction byproducts and to sterilize the
product. After filtering, the product is lyophilized, resulting in a thick, flaky white
precipitate.
5.3.2 Gram base conjugation
To decrease the amount of PNIPAm used and increase RGD conjugation to the
PEGSA backbone, gram: gram reactions (PEGSA: PNIPAm) were attempted. PNIPAm
(0.5 g, 0.05 mmol) was activated with DCC (0.0124 g, 0.06 mmol) and DMAP (0.0012 g,
0.01 mmol) by dissolving in anhydrous THF and reacted for 24 hours at 55C with
stirring. After 24 hours, PEGSA (0.5 g, 0.81 mmol) was dissolved in anhydrous THF and
heated to 55C with stirring. The activated PNIPAm was then added dropwise to the
PEGSA solution and reacted 24 hours. Precipitation and purification of the product was
carried out identically as described above for the mole base reaction.
5.4 RGD conjugation
5.4.1 RGD conjugation to PEGSA
38


To begin the reaction between the PEGS A hydroxyl groups with RGD amine
groups, RGD (0.025 g, 0.051 mmol) EDC (0.016 g, 0.103 mmol), NHS (0.012 g, 0.104
mmol), and were dissolved in Milli Q-water. This solution was reacted at room
temperature for 2 hours, protected from light. PEGS A (0.021 g, 0.034 mmol), was
dissolved in Milli Q-water and added dropwise to the activated RGD solution and reacted
for 24 hours at room temperature, protected from light. This reaction was a 100%
conjugation of RGD to each of the 2 hydroxyl groups per repeating unit on PEGS A.
After 24 hours, the reaction was placed in dialysis tubing (MWCO: 3500) and dialyzed
against 1 liter of Milli Q-water for 24 hours, with one water change. After dialysis, the
solution was freeze dried, resulting in a brownish, elastic precipitate.
5.4.2 RGD conjugation to PEGSA-PNIPAm (mol base)
To begin the reaction between the remaining free PEGS A hydroxyl groups (on
PEGSA-PNIPAm, mol base reaction) with RGD amine groups, RGD (0.025 g, 0.051
mmol), EDC (0.016 g, 0.103 mmol), and NHS (0.012 g, 0.104 mmol) were dissolved in
Milli Q-water. This solution was reacted at room temperature for 2 hours. PEGSA-
PNIPAm (0.267 g, 0.432 mmol) was dissolved in Milli Q-water and added dropwise to
the activated RGD solution and reacted for 24 hours at room temperature, protected from
light. After 24 hours, the reaction was placed in dialysis tubing (MWCO: 3500) and
dialyzed against 1 liter of Milli Q-water for 24 hours, with one water change. After
dialysis, the solution was freeze dried, resulting in a white, flaky precipitate.
5.4.3 RGD conjugation to PEGSA-PNIPAm (gram base)
To begin the reaction between the remaining free PEGS A hydroxyl groups (on
PEGSA-PNIPAm, gram base reaction) with RGD amine groups, RGD (0.08 g, 0.1631
39


mmol), EDC (0.0753 g, 0.485 mmol), andNHS (0.056 g, 0.487 mmol) were dissolved in
1 mL of Milli Q-water. This reaction was activated at room temperature for 2 hours.
PEGSA-PNIPAm (0.1 g, 0.01 mmol) was dissolved in 2 mL of Milli Q-water and added
dropwise to the activated RGD solution and reacted for 24 hours at room temperature,
protected from light. After 24 hours, the reaction was placed in dialysis tubing (MWCO:
3500) and dialyzed against 1 liter of Milli Q-water for 24 hours, with one water change.
After dialysis, the solution was freeze dried, resulting in a white, flaky precipitate.
5.5 Polymer biocompatibility testing
Rat pheochromocytoma cell line (PC 12) was selected for initial in-vitro studies to
assess the polymers biocompatibility. PC12 cells are not a true neural cell, as they are
derived from adrenal chromaffin cells. However, when exposed to nerve growth factor
(NGF), PC12 cells differentiate to resemble sympathetic neurons with similar
morphology and functionality 92. The use of this cell line has been used for biomaterial
optimization and for cytotoxicity testing 93,94
Retinal ganglion cells (RGCs) were used to test neural attachment and growth to
PEGSA-PNIPAm-RGD to assess the polymers ability to support CNS neurons. Retinas
from day 5 rat pups were dissected and isolated according to instructions from Neural
Tissue Dissociation Kits and Retinal Ganglion Cell Isolation Kit, MACS Miltenyl
Biotec. Briefly, after dissection, retinal tissue is dissociated to single cell suspensions
through a series of enzymatic and mechanical digestions of ECM adhesion proteins.
RGCs are then labeled with magnetic beads and undergo a series of magnetic separations
for compete RGC isolation.
40


5.5.1 PC12 cell culture
PC12 cells (1.0 x 106 cells, ATCC CRL-1721) were cultured with RPMI-1640
Medium (2.05 mM L-glutamine) supplemented with 10% horse serum (HS), 5% fetal
bovine serum (FBS), and 1% penicillin-streptomycin (will be referred to as proliferation
media). To induce differentiation, PC12 cells were cultured in RPMI-1640 Medium
(2.05mM L-glutamine) supplemented with 1% HS, 0.5% FBS, and 100 ng/mL NGF.
Cells were cultured on 60mm laminin coated dishes to facilitate attachment.
5.5.2 Retinal ganglion cell (RGC) cell culture
Retinal ganglion cells (RGCs) were used to test neural attachment and growth to
PEGSA-PNIPAm-RGD to assess the polymers ability to support CNS neurons. Retinas
from day 5 rat pups were dissected and isolated according to instructions from Neural
Tissue Dissociation Kits and Retinal Ganglion Cell Isolation Kit, MACS Miltenyl
Biotec. Briefly, after dissection, retinal tissue is dissociated to single cell suspensions
through a series of enzymatic and mechanical digestions of ECM adhesion proteins.
RGCs are then labeled with magnetic beads and undergo a series of magnetic separations
for compete RGC isolation.
5.5.3 MTT cytotoxicity
The cytotoxicity of PEGSA-PNIPAm was tested with proliferating PC 12 cells
with ISO/EN 10993-5 guidelines. PEGSA-PNIPAm gels were prepared in 10%, 5%,
2.5%, 1.25%, and 0.625% solutions (mg/mL). To a 96-well plate, 50 pL of each solution
was added and then gelled in an incubator for 15 minutes. To the top of each gel, 200 pL
of warm PC 12 proliferating media was added and allowed to incubate overnight. To a
separate 96-well plate, laminin solution was added to for cellular attachment (1 pg/cm2).
41


PC12 cells were seeded onto laminin coated 96-well plates at a density of 1 x 104
cells/well and incubated for 24 hours with proliferating medium. After 24 hours, medium
was removed from the PC12 cells and replaced with 200 pL of the PEGSA-PNIPAm
media extract. Cells were allowed to culture for an additional 72 hours. MTT assays
were conducted following supplier instructions for Quick Protocol, with absorbance
values read at 540 nm. Experimental samples were normalized to control samples
(proliferating media + PC 12 cells).
5.5.4 Alamar blue cell viability assay
The cell viability of PEGS A-PNIPAm was tested with proliferating PC 12 cells with
ISO/EN 10993-5 guidelines. Polymer extract solutions and cells were prepared
identically to MTT cytotoxicity testing. After PC 12 cells cultured in PEGSA-PNIPAm
extract media for 72 hours, Alamar Blue cell viability assay was performed according to
manufacturers instructions, with absorbance read at 570 nm. Experimental samples
were normalized to control samples (proliferating media + PC12 cells).
5.5.5 RGC attachment and growth
To study RGC survival and axon extension in a 3D polymer matrix, isolated RGCs
were mixed with PEGSA-PNIPAm-RGD solutions and gelled. PEGSA-PNIPAm-RGD
was dissolved in RGC media to create a 15% polymer solution. After isolation, RGCs
were mixed with PEGSA-PNIPAm-RGD solutions to create a final polymer
concentration of 10%, with a cell density of 1 x 104 cells per 50 pL of polymer-media
solution. To a 35mm glass bottom culture dish, 50 pL of polymer-cell solution was
added. The culture dish was placed in an incubator for 10 minutes to allow polymer
gelation and RGC encapsulation within the gel. After incubation, 1 mL of warm, RGC
42


media was added to the culture dish, using a hotplate set at 37C to maintain gel stability
when removed from the incubator. Cells were cultured for 3 days, with media changes
each day. Samples were then fixed in 4% PFA for 15 minutes and washed with PBS
overnight. Cells were permeablized with 1% Triton-X (in PBS) for 90 minutes, followed
by PBS wash overnight. A blocking buffer composed of 2% bovine serum albumin
(BSA) in PBS was added to the cells for 90 minutes. After the blocking step, cells were
incubated overnight with primary antibody Brn-3a (anti-mouse, 1:200 dilution). Cells
were then washed with 1% Triton-X, 3x for 3 minutes each. The secondary antibody,
anti-rabbit Alexa 594 (1:500), was added to each sample and incubated for 45 minutes.
Cells were washed with PBS-Tween (0.002% in PBS) for 3 minutes and washed twice
with PBS, 3 minutes each. The next primary antibody pill-tubulin (anti-rabbit, 1:100
dilution, prepared in blocking buffer) was added and incubated overnight. Cells were
then washed with 1% Triton-X, 3x for 3 minutes each. The secondary antibody, anti-
rabbit Alexa 488 (1:500), was added to each sample and incubated for 45 minutes. Cells
were washed with PBS-Tween (0.002% in PBS) for 3 minutes and washed twice with
PBS, 3 minutes each. Hoechst 33342 (1:2000, in PBS), a DAPI stain, was added to each
sample and incubated for 5-10 minutes, followed by 3 washes in PBS, 3 minutes each.
Samples were incubated in PBS until imaged with confocal microscope.
5.6 Compression SCI rat model
5.6.1 Ethics and surgical approval
All animal experiments were performed under a protocol approved by the
Institutional Animal Care and Use Committee (IACUC) at the University of Colorado
Anschutz Medical Campus.
43


5.6.2 Surgical procedures
Female adult Sprague Dawley rats (225-300g, n=15) were allowed to acclimate
for 1 week prior to surgical procedures. Rats were maintained on a 14/10-hour light/dark
cycle with a continuous supply of fresh air and access to food and water ad libitum.
A compression model was chosen for this study to test the polymers effects on a
chronic SCI model. Rats were anesthetized with 5% isoflurane in oxygen and maintained
on 0.5-1% isoflurane in oxygen for the remainder of the surgery. To maintain body
temperature, rats were placed on a warm recirculating water blanket for the duration of
the procedure. Artificial tears ointment was applied to the eyes to prevent corneal
abrasion and drying. After removing hair with electric clippers, the surgical site was
administered with bupivacaine (0.5% Marcaine) at 2mg/kg by dermal block injections to
decrease incision pain after surgery. Buprenorphine (0.2mg/kg) was administered
subcutaneously (SQ) to minimize post-operative pain. The incision site was sterilized
with 70% ethanol and then betadine, and this process was repeated 3 times. Rats were
placed in a prone position and a midline skin incision was made around the T7-T13
vertebrae (T11, T12 vertebrae are more pronounced and can be palpated by hand).
Paravertebral muscles were dissected to expose vertebral spinous processes and lamina.
Using a micro bone rongeurs, a double laminectomy at the T9-T11 level was performed
to expose the spinal cord.
To simulate a chronic compression SCI model, a 30g aneurysm clip was selected
to act as the compression device. To properly apply the compression clip, the lateral
44


aspects of the lamina and vertebrae around the cord needed to be removed as much as
possible. The compression clip was applied to the spinal cord for 1 minute and then
removed.
Spinal cord injection treatments were assigned randomly. As this was a small
pilot study, only two treatment groups were used. A saline injection was used as the
treatment control and PEGSA-PNIPAm-RGD was used as the experimental treatment.
The polymer was dissolved in sterile saline and filtered (0.2 microns) before injection.
After compression, 10 pL of saline or polymer was injected into the injury site using a
Hamilton syringe with a 32 gauge, 4 mm needle. When injected, the needle was held in
the injection site for 30 seconds to ensure complete delivery and then removed slowly.
The muscle layers were sutured together with 4-0 absorbable suture with an interrupted
pattern. The skin layer was sutured using a subcuticular method with 4-0 absorbable
suture and the incision site was wiped with betadine. Rats recovered from anesthesia on
a warm water blanket and were single housed in a new, clean cage. Single housing for
several days post-surgery was necessary to prevent other rats from biting the sutured
incisions. When administration of analgesics stopped, rats were placed back with their
original cage mates.
5.6.3 Post-surgery procedure
Buprenorphine (0.2mg/kg) was administered 12 hours after surgery for post-
operative pain. Ketoprofen (5mg/kg) was injected SQ 24 hours after surgery to continue
pain alleviation, and this continued once every 24 hours for 72 hours. Gentamicin
(lmg/mL) was added to the drinking water to prevent urinary tract infections and
remained in the water for the duration of the study. Food was crushed into pieces, soaked
45


in water, and placed in the cage for the first week after surgery. Manual bladder
expression was performed at least twice per day until the rats reestablished this reflex.
5.7 Functional Assessment
The BBB locomotor score was used to assess open-field locomotion. Before
surgery, rats were assessed and given the highest BBB score of 21 to act as baseline
score. After surgery, rats were scored once a week.
5.8 Histology
Four weeks after surgery, rats were given an IP overdose of ketamine to induce
deep anesthesia and were transcardially perfused with heparinized saline (50 mL, 10
Units Heparin/mL) followed by 4% PFA (200 mL). After perfusion, a 1 cm segment of
the spinal cord was harvested (with the lesion site centered) and placed in fresh 4% PFA
overnight. The spinal cord segments were then immersed in a 30% sucrose solution for
48 hours or until the tissue sank for cryoprotection. The sucrose solution was removed
and the spinal tissue was embedded in OCT compound and then frozen at -80C. The
OCT blocks were cut on a cryostat into 10 pm longitudinal sections and placed onto glass
slides. Glass slides were stored at -20C.
5.8.1 H&E staining
5.8.2 Immunohistochemistry
Glass slides with spinal tissue sections were fixed in acetone for 10 min and
washed 3 times in PBS for 3 min each. The sections were then blocked in 3% hydrogen
peroxide (in PBS) for 10 min to block endogenous peroxidase activity and washed 3
times in PBS for 3 min each. Blocking buffer (5% goat serum, 0.4% Triton X-100, PBS)
46


was used to block non-specific binding sites of the sections for 30 min. Antibodies were
prepared by diluting in the blocking buffer. Sections were then stained with the primary
antibody GFAP (1:100) or GAP-43 (1:300) for 60 min (room temperature) and washed 3
times in PBS for 3 min each. GFAP stained slides received Alexa Fluor 488 (1:500)
secondary antibody for 30 min and GAP-43 stained slides received Alexa Fluor 594
(1:500) secondary antibody for 30 minutes. After 30 minutes, slides were washed 3 times
in PBS for 3 min each. Fluoromount G with DAPI mounting medium was used to
coverslip the slides. Slides were stored at -4C.
5.9 Statistical analysis
All results are expressed as means standard error of the mean. Analysis of variance
(ANOVA) was used to determine significant differences between groups. Statistical
significance was considered when p < 0.05.
47


6 Results and Discussion
6.1 Polymer synthesis
PEGSA was successfully synthesized using a novel reaction between an epoxide
based PEG monomer and succinic acid. The PEGSA backbone contains 2-hydroxyl
groups per repeating unit that are used to conjugate PNIPAm and RGD to the backbone
(Figure 6-1).
Poly(ethylne glycol diglycidyl ether)
(PEGDGE)
120 C
48 hours
no solvent
O
o
Succinic Acid (SA)
i *
PEGSA
Figure 6-1. Reaction of PEGDGE with SA to produce PEGSA. The
PEGSA repeating unit contains two hydroxyl groups that will be used for
further chemical conjugation.
Following PEGSA synthesis, PNIPAm was conjugated to the PEGSA backbone.
PNIPAm conjugation allows PEGSA to attain RTG properties for injectability and fast
gelation upon exposure to physiological temperatures (Figure 6-2).
48


PEGSA-PNIPAm
Figure 6-2. Conjugation of PNIPAm to PEGS A backbone.
To allow PEGSA-PNIPAm to mimic the ECM and promote cellular adhesion to
integrin receptors, RGD is conjugated to the remaining hydroxyl groups to promote
biomimetic properties (Figure 6-3).
49


PEGSA-PNIPAm
o
RNIPAm
GRGDS-NH2
1. Activate GRGDS-NH2
EDC/NHS
Room Temp, 2 hours
2. Conjugate to PEGSA-PNIPAm
Room Temp, 24 hours
O
HN
O
RNIPAm
ngrgds
PEGSA-PN IPAm-RGD
Figure 6-3. PEGSA-PNIPAm-RGD reaction sequence. GRGDS is a
modified RGD peptide.
6.2 Polymer characterization
6.2.1 PEGSA FT-IR characterization
The main backbone polymer was successfully synthesized using a reaction
between a PEG-based monomer with epoxide groups (PEGDGE) and succinic acid (an
intermediate in the Citric Acid Cycle). FT-IR spectra verified the successful reaction
between PEGDGE and SA. As shown in Figure 6-4, region a, the PEGSA product has a
distinct peak that corresponds to the hydroxyl functional groups. Region b also shows a
peak corresponding to the formation of carbonyl groups from the succinic acid ester
groups. Region c is associated with ether groups, which are present in both the PEGDGE
monomer and maintained in the PEGSA product.
50


3.0
PEGDGE
Figure 6-4. FT-IR spectrum of PEGDGE and PEGSA
6.2.2 PEGSA-PNIPAm characterization
After verifying the PEGSA backbone structure, PNIPAm was conjugated to a
small percentage of the hydroxyl functional groups. The conjugation reaction between
PEGSA and PNIPAm was verified with FT-IR spectra. Figure 6-5 provides a
comparison between the PEGSA, PNIPAm, and the final product PEGSA-PNIPAm.
Region a shows the distinct PEGSA carbonyl peak that is preserved in the PEGSA-
PNIPAm product. This carbonyl peak in PEGSA-PNIPAm is also accompanied by
distinct PNIPAm peaks.
Figure 6-5. FT-IR spectrum of PEGSA, PNIPAm, PEGSA-PNIPAm
51


6.2.3 PEGSA-PNIPAm gelling properties
To create an injectable polymer scaffold, the polymer needs to be soluble in
aqueous solutions. Conjugating PNIPAm to the PEGSA backbone allows the polymer to
remain soluble in aqueous solutions at room temperature while forming a physical gel
when exposed to physiological temperatures. To test the LCST of PEGSA-PNIPAm, a
UV-VIS spectrometer was used to measure the % transmittance of the polymer when
subjected to changes in temperature. The sharp LCST of PEGSA-PNIPAm begins
around 35C, which is just below physiological temperature of 37C, as shown in Figure
6-6. Increasing the amount of PNIPAm causes the gel to become more opaque, which is
why the % transmittance decreases for increasing PNIPAm conjugations. This LCST test
verifies the RTG properties of PEGSA-PNIPAm.
120
PEGSA-PNI PAm_5%
100 conjugation
Figure 6-6. LCST of PEGSA-PNIPAm conjugation reactions and
PNIPAm.
6.2.4 PEGSA-PNIPAm morphological characterization
52


A key feature for polymer scaffolding systems is a porous structure to mimic the
ECM environment for nutrient exchange, gas exchange, and cellular growth. A highly
porous structure is particularly important for neurons to promote axon extensions. A
scanning electron micrograph verified the porous structure of a freeze dried sample of
PEGSA-PNIPAm (Figure 6-7). This porous structure was homogenous throughout the
entire sample. Figure 6-8 shows a higher magnification of the same sample that shows
the average pore size to be 5pm-20pm. Increasing the PNIPAm concentration or
increasing the polymer solution concentration will result in smaller pore sizes, and vice
versa.
Figure 6-7. SEM micrograph (180X) showing porosity of PEGSA-
PNIPAm with consistent porous size throughout the structure. Scale bar =
100pm.
53


Figure 6-8. SEM micrograph (950X) showing an average pore size of
5pm-20pm.
6.2.5 PEGSA-RGD FTIR characterization
PNIPAm is a large polymer that may present steric hindrance to further
conjugation reactions between PEGSA hydroxyl groups and other molecules. Before
conjugating RGD to PEGSA-PNIPAm, the reaction between PEGSA hydroxyl groups
and RGD was attempted. RGD was activated with EDC/NHS chemistry, and PEGSA
was added slowly to allow the hydroxyl groups to act as nucleophiles to the activated
RGD. Normally, amine groups are used as nucleophiles with EDC/NHS activated
chemicals, so this activation reaction with hydroxyl groups needed to be verified. The
conjugation of RGD to the PEGSA backbone was verified with FT-IR (Figure 6-9).
Region a shows PEGSA with RGD carbonyl and secondary amine peaks.
54


Figure 6-9 FT-IR spectrum of PEGS A, RGD, and PEGSA-RGD
6.2.6 PEGSA-PNIPAm-RGD FTIR characterization
When the reaction between PEGSA hydroxyl groups with RGD was confirmed,
RGD was conjugated to PEGSA-PNIPAm. FT-IR spectra could not definitively confirm
the conjugation reaction (Figure 6-10). This is likely because of the PNIPAm carbonyl
peaks and amine peaks (region a) that dominate the signal and block out the RGD
carbonyl and amine peaks (Figure 6-11). However, this does not mean the conjugation
did not occur. RGD is a small molecule and is likely able to overcome any steric
hindrance effects from the large PNIPAm polymer. Confirming the reaction between
PEGSA and RGD (Figure 6-9) demonstrates the PEGSA hydroxyl groups can act as
effective nucleophiles to activated RGD. Therefore, we believe the conjugation of RGD
to PEGSA-PNIPAm was successful, but it is not definitive on FT-IR spectra because of
signal dominance by PNIPAm peaks.
55


Wavenumber [cm-1]
Figure 6-10. FT-IR spectrum of PEGSA-PNIPAm-RGD. The
conjugation of RGD to PEGSA-PNIPAm could not be confirmed because
the PNIPAm peaks overshadow the RGD peaks.
1800 1750 1700 16S0 1600 1550 1500
Wavenumber[cm-1]
Figure 6-11. FT-IR spectrum of region a (in Figure 6-10) of PEGSA-
PNIPAm, RGD, and PEGSA-PNIPAm-RGD. The PEGSA-PNIPAm
peaks are located in the same chemical shift location as the RGD peaks.
The RGD peaks have a weaker signal than the large PNIPAm peaks.
6.3 In-vitro assessment
6.3.1 MTT cytotoxicity
The MTT assay measures mitochondrial reduction potential in cells to assess
metabolic activity. These enzymes reduce the MTT dye to formazan, which has a purple
color. Confluent PC 12 cells were exposed to diluted media extracts from PEGSA-
PNIPAm gels for 72 hours, and then tested with the MTT dye. No statistical significance
56


between the experimental groups and the positive control group was found
Media+cells 10% polymer solution 5% polymer solution 2.5%polymer 1.25%polymer 0.625% polymer Mediaonly
solution solution solution
Figure 6-12). Therefore, PC12 cell metabolic activity was not affected by
exposure to the diluted polymer extract solutions. These results confirmed PEGSA-
PNIPAm was biocompatible and non-toxic to PC 12 cells.
Media+cells 10% polymer solution 5% polymer solution 2.5% polymer 1.25% polymer 0.625% polymer Mediaonly
solution solution solution
Figure 6-12. PC 12 cells showed no cytotoxicity when exposed to
PEGSA-PNIPAm media extracts, per ISO 10993-5. Statistical analysis by
ANOVA demonstrated no difference between the positive control
(Media+cells) sample and experimental samples (p=0.36, n=5).
Experimental samples are normalized to cells exposed to pure media.
Media only is the negative control. Error bars represent standard error of
the mean.
6.3.2 Alamar blue cell viability
57


The Alamar Blue assay measures similar reduction potential of cells to assess
cellular viability. The Alamar Blue dye contains Resazurin, which is non-toxic and
permeable to cells. When exposed to reduction enzymes within the cytosol, viable cells
reduce resazurin to resorufin, which fluoresces red. With a similar protocol to the MTT
assay, PC 12 cells were cultured in polymer extract solutions for 72 hours, followed by
addition of the Alamar Blue dye. No statistical difference was found between the
positive control sample and samples exposed to PEGSA-PNIPAm media extracts. These
results further confirmed the biocompatibility of PEGSA-PNIPAm and its ability to
support viable cells (Figure 6-13).
too * *
Media+cells 10% polymer 5% polymer 2.5% polymer 1.25% polymer 0.625% polymer Media only
solution solution solution solution solution
Figure 6-13. PC 12 cells showed no change in viability when exposed to
PEGSA-PNIPAm media extracts, per ISO 10993-5. Statistical analysis by
ANOVA demonstrated no difference between the positive control
(Media+cells) samples and experimental samples (p=0.13, n=5).
Experimental samples are normalized to cells exposed to pure media.
Media only is the negative control. Error bars represent standard error of
the mean.
6.3.3 RGC attachment and growth
After 3 days in culture, PEGSA-PNIPAm-RGD was able to support the
survival and growth of primary RGCs. Axon extensions within the 3D polymer
58


matrix provided evidence of the polymers capability to successfully attach and
support neurite outgrowth (Figure 6-14).
Figure 6-14. RGC axon extensions within PEGSA-PNIPAm-RGD 3D
polymer matrix. Axons stained with pni-tubulin (green), nuclei stained
with DAPI (blue), RGCs stained with Bm3a (red). Maximum intensity
confocal microscope z-stack image, scale bar =100 pm.
6.4 Compression SCI rat model
The laminectomy surgery was able to successfully expose spinal cord tissue.
However, proper placement of the compression clip required more lateral vertebrae to be
removed, which was difficult. If too little lamina was removed, the clip often would not
fit around the cord for good compression. The general surgical procedure is
demonstrated in Figure 6-15, focusing on the isolation of thoracic vertebrae, laminectomy
to expose the spinal cord, and clip compression.
Following compression SCI, each rat received a 10 pi injection of either saline or a
2.5% solution of PEGSA-PNIPAm-RGD. The polymer solution was dissolved in sterile
59


saline and filtered (0.2pm) before use. The saline injection served as a control group to
compare to the PEGSA-PNIPAm-RGD experimental group. As no effective SCI
treatment exists, there was no positive control used to compare to the polymer injection.
A total of 6 rats received saline injections and 8 rats received polymer injections. Of the
14 rats used for this study, 13 were successfully compressed for 60 seconds. One rat did
not receive a successful compression because of a forceps that slipped and cut the cord.
Bleeding occurred in about half of the rats during and after the clip compression. The
severity of the bleeding was usually minimal but a couple had profuse bleeding. This
bleeding was managed with gauze and saline washes. The compression clip was difficult
to keep secured to the spinal cord during the full 60 seconds in some of the rats and as a
result, the clip would often slip off the cord. This required clip repositioning and usually
resulted in more bleeding. As a result of the compression SCI, each rat had paraplegia
for at least two weeks before hindlimb movements started to return.
Figure 6-15. Compression spinal cord injury. (A) Exposed thoracic
vertebrae (T9-T11), (B) exposed spinal cord tissue after laminectomy, (C)
spinal cord compression with 30 g aneurysm clip.
Post-surgery complications occurred in many rats, particularly in the saline
injection rats. Hematuria was the predominant complication that often would be
60


followed by death before the 4 week time point. Of the 6 rats injected with saline, only 2
rats survived to 4 weeks, with 1 of those rats having the incomplete compression injury
mentioned earlier. Of the 4 saline rats that died, 3 died about 10-14 days after surgery
and 1 died 23 days after surgery. Each of these rats had successfully recovered and
appeared healthy 1 week after surgery, but this was followed by hematuria that became
progressively worse. Therefore, only 1 saline injected rat with complete compression
successfully made it through the 4 week study (Table 1).
Table 1. Surgical outcomes and complications of saline injected rats.
Saline Rat ID Full Compression Surgery Date Complications Survival Period (Days)
2 Yes 06/18/15 Hematuria 10
7 Yes 07/01/15 None 30
8 No 07/06/15 Hematuria 60
10 Yes 08/05/15 Hematuria 9
14 Yes 08/05/15 Hematuria 12
15 Yes 08/05/15 Hematuria 23
16 Yes 08/22/15 Hematuria 13 ,
A summary of surgical outcomes for polymer injected rats is provided in Table 2.
Of the 8 rats injected with polymer, 1 had to be euthanized 6 days after surgery. This rat
did not show signs of good health after surgery and was mostly immobile in its cage,
leading us to believe there was an unknown surgical complication that resulted in its poor
health and eventual euthanasia. Hematuria was also observed in several of the polymer
injected rats after surgery but resolved as the rats drank more of the gentamicin water.
One week after surgery, these rats generally appeared to be healthy and did not display
signs of pain.
Table 2. Surgical outcomes and complications of polymer injected rats
61


Polymer Rat ID Full Compression Surgery Date Complications Survival Period (Days)
6 Yes 06/29/15 minor hematuria 30
9 Yes 07/06/15 None 60
17 Yes 08/06/15 minor hematuria 30
18 Yes 08/06/15 Did not recover 5
19 Yes 08/06/15 None 30
20 Yes 08/07/15 None 30
21 Yes 08/07/15 None 30
22 Yes 08/07/15 None 30
To verify localization of the polymer, PEGSA-PNIPAm was dyed with toluidine
blue and injected in a separate rats spinal cord. The polymer was easily visualized
gelling into a specific region of the spinal cord. The surgical area was then washed twice
with PBS to confirm the polymer remained at the injection site, as shown in Figure 6-16.
Figure 6-16. A 10 pi injection of PEGSA-PNIPAm dyed with toluidine
blue confirmed localization of the polymer into the spinal cord injection
site.
6.5 Functional assessment
To assess motor recovery, each rat underwent BBB locomotion testing, walking
track analysis, and latter walk testing. Unfortunately, the only test most rats were able to
successfully complete was the BBB open field test. The highest score of 21 was given to
each rat before surgery and the lowest score of 0 was given to each rat 1 week after
surgery. A summary of the BBB scores for rats that survived 4 weeks is provided in
62


Table 3. Unfortunately, only 1 saline rat survived the 4 week study duration,
making the sample size too small to perform statistical analysis. The polymer injection
rats showed large variability in motor recovery, with a standard deviation of 3.7. Overall,
no statistical conclusions can be made with this data due to the high variability of
polymer injected rat scores and small sample size of saline injected rats.
Table 3. Summary of BBB scores for 4 week rats. Only 1 saline injection
rat survived the entire 4 week duration of the study. The polymer
injection rats have a large variability (3.4 3.7, n = 6).
Treatment BBB Score: 4 weeks
Saline Rat 1 15
Polymer Rat 1 10
Polymer Rat 2 1
Polymer Rat 3 9
Polymer Rat 4 3
Polymer Rat 5 3
Polymer Rat 6 3
6.6 Histology
6.6.1 H&E staining
The location of the SCI lesion was confirmed with H&E staining. Compared to
healthy, uninjured tissue, compressed spinal tissue (4 hours post-surgery) shows gray
63


matter dislocation and compression. After 4 weeks and 8 weeks, the characteristic glial
scar is observed (Figure 6-17, Figure 6-18). After 8 weeks, a defined region within the
glial scar shows gaps in the tissue that have less cellular density than the saline injected
rat (Figure 6-18). This could be an indication of polymer localization into the SCI site, as
the polymer will not stain with H&E.
Figure 6-17. Vertical spinal cord sections stained with H&E showing the
progression of glial scarring after compression SCI with polymer
injection. (A) Healthy, uninjured tissue, (B) compressed spinal cord 4
hours after surgery, (C) 4 weeks after surgery. Scale bar = 500 pm.
Figure 6-18. Glial scar formation of longitudinal compressed spinal cord
after 8 weeks with polymer injection. The decreased cellular density at
the SCI site could be due to the presence of the polymer, which does not
stain with H&E. Scale bar = 1000 pm.
6.6.2 GFAPIHC
64


GFAP staining compared the accumulation of astrocytes at the lesion site between
the two treatment groups. One of the goals of the polymer injection directly into the SCI
site was to decrease the formation of the glial scar to encourage neural repair. We expect
a higher presence of astrocytes in areas where the treatment is less effective at preventing
glial scar formation. After 4 weeks, astrocytes were observed within and around the
compression site with the polymer treatment (Figure 6-19 A). Unfortunately there are not
enough 4 week saline rats to compare to the polymer treatment at 4 weeks, making it
difficult to draw any conclusions based on these images alone. After 8 weeks, astrocytes
appear slightly more distributed throughout the lesion site (Figure 6-19 B). This could be
an indication that the polymer begins to accumulate and support astrocytic activity as the
glial scar forms, but more rats are also needed at this time point to make definitive
conclusions.
Figure 6-19. GFAP immunostaining to detect astrocyte accumulation
around SCI site in vertical spinal cord sections from (A) representative 4
week polymer injection, (B) 8 week polymer injection. Images acquired
with Nikon confocal microscope, 20X air objective. Scale bar = 500 pm.
65


6.6.3 GAP-43 IHC
The SCI lesion site forms a barrier to axonal growth and repair. The goal of the
polymer injection into the lesion site was to decrease glial scarring and provide a cellular
attachment scaffold for neural growth and axonal extensions into the lesion site. GAP-43
staining identifies proteins of growth cones from elongating axons to help distinguish
between regenerating axons and non-regenerating axons. Polymer injected rats after 4
showed the slight appearance of regenerating axons within the lesion site (Figure 6-20).
GAP-43 staining appeared more specific for the 8 week polymer injected rat (Figure
6-21). These results were encouraging, as the presence of regenerating axons within the
polymer injection site was a sign of a more growth permissive SCI environment.
Figure 6-20. GAP-43 immunostaining to detect regenerating axons.
Vertical SCI lesion site from representative 4 week polymer injection (top
to bottom representing rostral and caudal ends of the spinal cord). Scale
bar = 500 pm. Maximum intensity confocal microscopy image, 10X air
objective.
66


#, . y v> S''V
> >
*'- 'tr ^
\ - ''f-Tr % % 1 _.
' L^*, £-J_ * w- ;v
V:& " - -> ^

*v r-*r.-\
. < v.-~-
4^ * s A

' Vi' £t/' - .' c : ------
>. . .. '1 -> -_v
, sV^v-' X^-sO,
f
Figure 6-21. GAP-43 immunostaining to detect regenerating axons.
Longitudinal spinal cord section from 8 week polymer injection (left and
right sides representing rostral and caudal ends of the spinal cord). Scale
bar = 500 pm. Maximum intensity confocal microscopy image, 10X air
objective.
67


7 Conclusion
The objective of this study was to develop an injectable RTG capable of acting as a
cellular scaffold for SCI repair. Current SCI treatments lack effective regenerative
treatments, but cellular and molecular therapy approaches present promising new avenues
to permit repair from the injury. To enhance the therapeutic efficacy of these new
treatment options, polymer scaffolds have been developed to act as supportive substrates
and delivery vehicles.
A novel RTG with a PEG-based backbone was successfully synthesized and
characterized. Conjugation of RGD to PEGSA-PNIPAm was not able to be verified by
FT-IR, as PNIPAm has signal dominance over specific RGD peaks. However, RGD
conjugation to the PEGSA backbone (without PNIPAm) was confirmed with FT-IR. By
proving RGD conjugation to the PEGSA backbone, this enhanced the confidence that
RGD was successfully conjugated to PEGSA-PNIPAm.
The gelling properties of PEGS A-PNIP Am was confirmed with gelling tests at 37C.
A sharp LCST was demonstrated with temperature controlled UV-VIS spectroscopy,
verifying the RTG properties necessary for injectability and in-situ forming gels. Within
5-10 minutes of exposure to 37C, the gels would begin to shrink by roughly 20-30%,
which is an indication of gel instability. This contrasts with the gel stability of PSHU-
PNIPAm (mentioned in literature review) which is able to maintain its conformation over
prolonged periods exposed to 37C. The primary difference between the two polymers is
the properties of the backbone. PEGSA is a hydrophilic backbone, while PSHU is a
hydrophobic backbone with longer alkyl chains. PEGSA-PNIPAm may not have long
enough alkyl chains to maintain complete gel stability over prolonged periods.
68


Preliminary in-vitro studies with PC 12 cells showed that PEGSA-PNIPAm RTG
promoted a permissive environment for cell survival. MTT and Alamar Blue assays
demonstrated that PEGSA-PNIPAm gels were non-cytotoxic and were capable of
supporting the proliferation of PC12 cells. In-vitro studies with a primary RGC culture
further demonstrated the polymers potential as a cellular support scaffold. Brn-3a and
piII-Tubulin staining showed the gelled polymer was capable of supporting viable RGCs
with axon extensions.
The RTG was investigated for axon regeneration in a compression rat SCI model.
Polymer or saline solutions were injected into the SCI site and assessed for neural
regeneration and motor repair. Saline injected rats experienced health complications
(hematuria) that resulted death before the 30 day study duration concluded for all but one
rat. However, polymer injected rats that experienced similar health complications were
able to recover quickly, and all but one survived the entire duration of the study.
Bleeding during clip compression is a likely cause for hematuria, but the exact reason for
the high death rate in saline injected rats is unknown. It is possible that formed blood
clots are dislodging over time, causing embolisms in saline injected rats. The polymer
may be acting as a barrier to prevent these clots from circulating. It is also possible the
polymer is acting to provide pressure that allows bleeding to clot. Therefore, the saline
rats may never have properly healed the bleeding that occurred after compression,
resulting in death.
Motor repair was investigated with three different functional tests: BBB score,
walking track, and latter walk. After 4 weeks, none of the rats were able to successfully
navigate the latter walk test. The hindlimbs of each rat were unable grip any of the latter
69


rungs, resulting in complete misses for each step taken. This loss of hindlimb control
also resulted in poor results with walking track analysis. Rats were unable to plant their
painted hindlimbs onto the paper for readable paw prints required for analysis. Finally,
BBB openfield locomotion scores showed high variability in the polymer treatment
group, while the small sample size for the saline injected rats prevented statistical
analysis. Therefore, functional testing was unable to provide conclusive data regarding
each treatments motor repair capability.
H&E staining revealed characteristic scar formation at the SCI site. A comparison
between the two treatment groups after 8 weeks revealed possible localization of the
polymer within the injury site.
IHC staining with GFAP confirmed the accumulation of astrocytes around the SCI
site, but no difference between treatment groups could be discerned. GAP-43 staining
was able to confirm the presence of regenerating axons in polymer injected rats after 4
and 8 weeks. This was the most encouraging result, as it demonstrated that the polymer
injection site was able to overcome toxic effects of the SCI environment to promote
regeneration.
In closing, a novel, biomimetic RTG was able to support the in-vitro survival of
primary retinal ganglion cells, proving it to be a promising scaffold for CNS neural
tissue. Despite variability in the compression SCI model and inconclusive functional
tests, after 4 and 8 weeks, histology revealed the presence of regenerating axons within
the SCI lesion where the RTG was injected. These preliminary results demonstrated this
novel, injectable RTG is a promising, noninvasive approach for delivering a biomimetic
70


scaffold within the SCI site to help overcome regenerative barriers and promote neural
repair.
71


8 Study limitations and future work
8.1 Modify control and experimental groups
The animal study was designed to be a small pilot study to assess spinal cord axon
regeneration after injecting a new biomimetic RTG. This SCI study only included a
control saline injection and the PEGSA-PNIPAm-RGD experimental injection. Future
studies should include a surgery only group (no compression) and also a compression
only group (no injection) to help determine if surgical or injection errors are contributing
to large variability in SCI lesions. Including a PEGSA-PNIPAm experimental group
with no conjugated RGD would allow the polymer by itself to be assessed as a supportive
cellular scaffold. Including this group would also allow for a good comparison to
PEGSA-PNIPAm-RGD to verify its ability to deliver and support the cellular actions of
the RGD peptide.
8.2 Modify SCI compression
As mentioned earlier, the source of the large variability in survival and functional
assessment between animals in the SCI model is likely due to bleeding during clip
compression. The aneurysm clip used for compression may be too sharp for the delicate
spinal tissue, leaving little room for error while applying the clip. Bleeding seemed to
occur after full compression and not during the initial application of the clip. This likely
occurred because the clip either punctured or lacerated the spinal cord vasculature. Using
a clip with smoother edges or a forceps will be a better alternative.
Problems were also encountered after initial application of the compression clip. The
clip had a tendency to slip off the spinal cord after initial application, resulting in possible
72


incomplete compression, requiring reapplication. Clips that appeared to remain in place
may not have been compressing the entire spinal cord. In the future, further laminectomy
should be performed to better verify complete cord compression.
The compression time will also be modified. A one minute compression may be too
long and could result in more complications and inconsistencies between rats. Further
studies will use a brief compression (1-5 seconds) to minimize these errors.
8.3 Modify PEGSA backbone to prevent gel shrinking
As discussed earlier, the PEGSA-PNIPAm gel begins to shrink when exposed to
physiological temperatures after several minutes (Figure 8-1).
Figure 8-1. After several minutes, PEGSA-PNIPAm shrinks, resulting in
a gel that no longer holds the shape of the vial.
A similar polyurethane based polymer, PSHU-PNIPAm, does not have these
shrinking problems 95. The aggregation of hydrophobic groups contributes to a more
stable gel over prolonged exposures to physiological temperatures. PEGSA is a more
hydrophilic polymer backbone compared to PSHU, which results in fewer hydrophobic
group aggregations during gelling. This is an indication that the PEGSA backbone needs
to be modified to promote more stability while gelled. Replacing succinic acid with
73


longer chained di-carboxylic acids, such as sebacic acid, could allow more hydrophobic
groups in the backbone to promote enhanced gel stability. Experimenting with different
PEG chain lengths is also another option.
8.4 Increase study duration
We observed slight visual improvements (not statistical) in motor recovery with the
two rats (polymer injection and saline injection) extended to 8 weeks. Extending the
study to 8 weeks and 12 weeks is important to see if functional improvements and axonal
regeneration eventually level off over time.
8.5 Monitor polymer presence in spinal tissue
The polymer was successfully injected into the spinal tissue. However, a method
needs to be developed to monitor long term presence and localization of the polymer
within the spinal tissue over the course of the study. This can be achieved by conjugating
a fluorophore to the PEGS A hydroxyl groups. After injecting into spinal tissue, the rats
can be euthanized at different time points to conduct IHC. A similar method was used to
observe the localization of a natural polymer injected into spinal tissue, with successful
visualization of the fluorescently labeled polymer62. This study did not attempt long
term visualization of the polymer, but it would be worth trying various time points to
assess not only the presence of the polymer over time but also possible in-vivo
degradation of the polymer.
8.6 Long term goal of RTG SCI application
The long-term goal of this RTG is not only to incorporate cellular adhesive
properties through the use of RGD peptide, but to also integrate NTFs and cells into the
74


RTG system to promote SCI repair. NTFs can be integrated into polymer through
electrostatic interactions to provide essential biochemical cues for neurite growth. The
administration of NTFs will not only be made within the RTG, but also should be made
upstream and downstream of the SCI site. This type of NTF administration will provide
regenerating neurons cues to not only grow within the SCI lesion but could also allow
regenerating axons to bridge the entire SCI lesion.
Replacing damaged tissue in the SCI site is also a priority, and this can be
accomplished through the use stem cell therapies. As the primary RGC experiment
demonstrated, cells can be successfully encapsulated into the gelled polymer scaffold to
promote neurite growth and axon extension. Our lab was able to successfully grow
human NSCs onto a polyurethane biomimetic scaffold that promoted differentiation into
motor neurons. In-vitro studies with PEGSA-PNIPAm encapsulated with human NSCs
would be the next step for the future direction of this polymers cellular applications. If
the human NSCs can successfully survive and grow within a 3D polymer matrix (as was
seen with the RGC experiments), the next in-vivo study would incorporate human NSCs
into the polymer for SCI injection.
Overall, incorporating NTFs and stem cells within the RTG design should provide a
growth permissive, biomimetic ECM environment within the toxic SCI lesion site to
promote neural repair. If neurons are able to grow and differentiate within the SCI lesion
and extend beyond the injury site, this could allow the body to provide natural cues to
enhance meaningful axon rewiring to downstream tissues. As research continues to
discover different methods to overcome the toxic SCI environment to promote neural
75


regeneration, injectable RTGs will continue to prove to be effective delivery devices to
enhance the therapeutic effects of each of these promising treatment approaches.
76


References
1. Silva N a., Sousa N, Reis RL, Salgado AJ. From basics to clinical: A
comprehensive review on spinal cord injury. ProgNeurobiol. 2014;114:25-57.
doi: 10.1016/j.pneurobio.2013.11.002.
2. Nscisc. Spinal cord injury facts and figures at a glance. J Spinal Cord Med.
2013;36(1): 1-2. doi:10.1179/1079026813Z.000000000136.
3. Selvarajah S, Hammond ER, Haider AH, et al. The burden of acute traumatic
spinal cord injury among adults in the united states: an update. J Neurotrauma.
2014;31(3):228-238. doi:10.1089/neu.2013.3098.
4. Richardson PM, McGuinness UM, Aguayo AJ. Axons from CNS neurones
regenerate into PNS grafts. Nature. 1980;284:264-265.
5. Cajal RY. Degeneration and Regeneration of the Nervous System. Oxford
University Press. 1928.
6. Ziegler MG. Nervous System. Prim Auton Nerv Syst. 2012:291-294.
doi: 10.1016/B978-0-12-386525-0.00061-5.
7. Ronsyn MW, Berneman ZN, Van Tendeloo VFI, Jorens PG, Ponsaerts P. Can cell
therapy heal a spinal cord injury? Spinal Cord. 2008;46(8):532-539.
doi: 10.1038/sc.2008.13.
8. Li Y, Walker CL, Zhang YP, Shields CB, Xu X-M. Surgical decompression in
acute spinal cord injury: A review of clinical evidence, animal model studies, and
potential future directions of investigation. Front Biol (Beijing). 2014;9(2):127-
136. http://download.springer.eom/static/pdf/558/art%3A10.1007%2Fsll515-014-
1297-z.pdf? originUrl=http://link. springer, com/arti cl e/10.1007/sl 1515-014-1297-
z&token2=exp=1438393533~acl=/static/pdf/558/art%253A10.1007%252Fsl 1515-
014-129.
9. Hurlbert RJ. Methylprednisolone for the treatment of acute spinal cord injury:
point. Neurosurgery. 2014;61 Suppl l(l):32-35.
doi: 10.1227/NEU.0000000000000393.
10. Hurlbert RJ, Hadley MN, Walters BC, et al. Pharmacological therapy for acute
spinal cord injury. Neurosurgery. 2013;72(SUPPL.2):93-105.
doi: 10.1227/NEU.0b013e31827765c6.
11. Liu JC, Patel A, Vaccaro AR, Lammertse DP, Chen D. Methylprednisolone after
traumatic spinal cord injury: yes or no? PMR. 2009;l(7):669-673.
doi: 10.1016/j.pmrj .2009.06.002.
77


12. Bydon M, Lin J, Macki M, Gokaslan ZL, Bydon A. The Current Role of Steroids
in Acute Spinal Cord Injury. WorldNeurosurg. 2013;82(June):l-7.
doi: 10.1016/j.wneu.2013.02.062.
13. Bethea JR. Spinal cord injury-induced inflammation: a dual-edged sword. Prog
Brain Res. 2000;128:33-42. doi:10.1016/S0079-6123(00)28005-9.
14. BETHEA JR, NAGASHIMA H, ACOSTA MC, et al. Systemically Administered
Interleukin-10 Reduces Tumor Necrosis Factor-Alpha Production and
Significantly Improves Functional Recovery Following Traumatic Spinal Cord
Injury in Rats. J Neurotrauma. 1999; 16(10): 851 -863.
doi: 10.1089/neu. 1999.16.851.
15. Zuo J, Neubauer D, Dyess K, Ferguson TA, Muir D. Degradation of chondroitin
sulfate proteoglycan enhances the neurite-promoting potential of spinal cord tissue.
Exp Neurol. 1998;154(2):654-662. doi: 10.1006/exnr. 1998.6951.
16. Lu P, Jones LL, Tuszynski MH. BDNF-expressing marrow stromal cells support
extensive axonal growth at sites of spinal cord injury. Exp Neurol.
2005;191(2):344-360. doi:10.1016/j.expneurol.2004.09.018.
17. Weishaupt N, Blesch A, Fouad K. BDNF: the career of a multifaceted
neurotrophin in spinal cord injury. Exp Neurol. 2012;238(2):254-264.
doi: 10.1016/j.expneurol.2012.09.001.
18. Ye JH, Houle JD. Treatment of the chronically injured spinal cord with
neurotrophic factors can promote axonal regeneration from supraspinal neurons.
Exp Neurol. 1997; 143(l):70-81. doi: 10.1006/exnr. 1996.6353.
19. McCall J, Weidner N, Blesch A. Neurotrophic factors in combinatorial approaches
for spinal cord regeneration. Cell Tissue Res. 2012;349(l):27-37.
doi: 10.1007/s00441-012-1388-6.
20. Doucette R. PNS-CNS transitional zone of the first cranial nerve. J Comp Neurol.
1991 ;312(3):451-466. doi: 10.1002/cne.903120311.
21. Li L, Adnan H, Xu B, et al. Effects of transplantation of olfactory ensheathing
cells in chronic spinal cord injury: a systematic review and meta-analysis. Eur
Spine J. 2015;24(5):919-930. doi:10.1007/s00586-014-3416-6.
22. Ramer LM, Au E, Richter MW, Liu J, Tetzlaff W, Roskams a. J. Peripheral
Olfactory Ensheathing Cells Reduce Scar and Cavity Formation and Promote
Regeneration after Spinal Cord Injury. J Comp Neurol. 2004;473(1): 1-15.
doi: 10.1002/cne.20049.
78


23. Imaizumi T, Lankford KL, Burton W V, Fodor WL, Kocsis JD.
Xenotransplantation of transgenic pig olfactory ensheathing cells promotes axonal
regeneration in rat spinal cord. NatBiotechnol. 2000;18(9):949-953.
doi: 10.1038/79432.
24. Huang H, Xi H, Chen L, Zhang F, Liu Y. Long-term outcome of olfactory
ensheathing cell therapy for patients with complete chronic spinal cord injury. Cell
Transplant. 2012;21(SUPPL. 1):S23-S31. doi: 10.3727/096368912X633734.
25. Dlouhy BJ, Awe O, Rao RC, Kirby P a, Hitchon PW. Autograft-derived spinal
cord mass following olfactory mucosal cell transplantation in a spinal cord injury
patient. JNeurosurgSpine. 2014;21 (October): 1-5.
doi: 10.3171/2014.5.SPINE13992.
26. Tetzlaff W, Okon EB, Karimi-Abdolrezaee S, et al. A systematic review of cellular
transplantation therapies for spinal cord injury. J Neurotrauma. 2011 ;28(8): 1611-
1682. doi: 10.1089/neu.2009.1177.
27. Pal R, Venkataramana NK, Bansal A, et al. Ex vivo-expanded autologous bone
marrow-derived mesenchymal stromal cells in human spinal cord
injury/paraplegia: a pilot clinical study. Cytotherapy. May 2015.
http://www.tandfonline.eom/doi/abs/10.3109/14653240903253857#.Vc6xHvlViko
. Accessed August 15, 2015.
28. Mariano ED, Batista CM, Barbosa BJAP, et al. Current perspectives in stem cell
therapy for spinal cord repair in humans: a review of work from the past 10 years.
ArqNeuropsiquiatr. 2014;72(6):451-456. doi:10.1590/0004-282X20140051.
29. Willerth SM, Sakiyama-Elbert SE. Cell therapy for spinal cord regeneration. A civ
DrugDelivRev. 2008;60(2):263-276. doi:10.1016/j.addr.2007.08.028.
30. Lu P, Wang Y, Graham L, et al. Long-distance growth and connectivity of neural
stem cells after severe spinal cord injury. Cell. 2012;150(6):1264-1273.
doi: 10.1016/j.cell.2012.08.020.
31. Stenudd M, Sabelstrom H, Frisen J. Role of Endogenous Neural Stem Cells in
Spinal Cord Injury and Repair. JAMA Neurol. 2015;72(2):235.
doi: 10.1001/jamaneurol.2014.2927.
32. Lowry NA, Temple S. Making Human Neurons from Stem Cells after Spinal Cord
Injury. PLoSMed. 2007;4(2):e48. doi: 10.1371/journal.pmed.0040048.
33. Aurand ER, Lampe KJ, Bjugstad KB. Defining and designing polymers and
hydrogels for neural tissue engineering. Neurosci Res. 2012;72(3): 199-213.
doi: 10.1016/j .neures.2011.12.005.
79


34. Straley KS, Foo CWP, Heilshom SC. Biomaterial design strategies for the
treatment of spinal cord injuries. JNeurotrauma. 2010;27(1): 1-19.
doi: 10.1089/neu.2009.0948.
35. Pego AP, Kubinova S, Cizkova D, et al. Regenerative medicine for the treatment
of spinal cord injury: more than just promises? J Cell Mol Med. 2012; 16(11):2564-
2582. doi: 10.1111/j. 1582-4934.2012.01603.x.
36. Lutolf MP, Hubbell J a. Synthetic biomaterials as instructive extracellular
microenvironments for morphogenesis in tissue engineering. Nat Biotechnol.
2005;23(l):47-55. doi:10.1038/nbtl055.
37. Okamoto M, John B. Synthetic biopolymer nanocomposites for tissue engineering
scaffolds. ProgPolym Sci. 2013;38(10-11):1487-1503.
doi: 10.1016/j .progpolymsci .2013.06.001.
38. Harris JM. Poly (ethylene Glycol) Chemistry: Biotechnical and Biomedical
Applications. Springer Science & Business Media; 2013.
https://books.google.com/books?hl=en&lr=&id=lU30BwAAQBAJ&oi=fnd&pg=P
Al&dq=polyethylene+glycol+biological+applications&ots=eENOq2tZD5&sig=Iv
M6PfsmsVRODfqJCttMVmKeFeY#v=onepage&q&f=false.
39. Park D, Wu W, Wang Y. A functionalizable reverse thermal gel based on a
polyurethane/PEG block copolymer. Biomaterials. 2011;32(3):777-786.
doi: 10.1016/j.biomaterials.2010.09.044.
40. Zhu J. Bioactive modification of poly(ethylene glycol) hydrogels for tissue
engineering. Biomaterials. 2010;31(17):4639-4656.
doi: 10.1016/j .biomaterials.2010.02.044.
41. Yamaoka T, Yasuhiko T, Yoshito I. Distribution and tissue uptake of
poly(ethylene glycol) with different molecular weights after intravenous
administration to mice. JPharm Sci. 1994;83(4):601-606.
http://onlinelibrary.wiley.eom/doi/10.1002/jps.2600830432/abstract.
42. Pashuck ET, Stevens MM. Designing regenerative biomaterial therapies for the
clinic. Sci TranslMed. 2012;4(160):160sr4. doi: 10.1126/scitranslmed.3002717.
43. Drury JL, Mooney DJ. Hydrogels for tissue engineering: Scaffold design variables
and applications. Biomaterials. 2003;24(24):4337-4351. doi:10.1016/S0142-
9612(03)00340-5.
44. Mahoney MJ, Anseth KS. Three-dimensional growth and function of neural tissue
in degradable polyethylene glycol hydrogels. Biomaterials. 2006;27(10):2265-
2274. doi: 10.1016/j.biomaterials.2005.11.007.
80


45. Comolli N, Neuhuber B, Fischer I, Lowman A. In vitro analysis of PNIPAAm-
PEG, a novel, injectable scaffold for spinal cord repair. Acta Biomater.
2009;5(4): 1046-1055. doi: 10.1016/j.actbio.2008.10.008.
46. Luo J, Borgens R, Shi R. Polyethylene glycol immediately repairs neuronal
membranes and inhibits free radical production after acute spinal cord injury. J
Neurochem. 2002;83(2):471-480. doi: 10.1046/j. 1471-4159.2002.01160.x.
47. Luo J, Shi R. Diffusive oxidative stress following acute spinal cord injury in
guinea pigs and its inhibition by polyethylene glycol. Neurosci Lett.
2004;359(3): 167-170. doi:10.1016/j.neulet.2004.02.027.
48. Borgens RB, Bohnert D. Rapid recovery from spinal cord injury after
subcutaneously administered polyethylene glycol. J Neurosci Res.
2001;66(6):1179-1186. http://www.ncbi.nlm.nih.gOv/pubmed/l 1746451. Accessed
August 31, 2015.
49. Nomura H, Tator CH, Shoichet MS. Bioengineered strategies for spinal cord
repair. J Neurotrauma. 2006;23(3-4):496-507. doi:10.1089/neu.2006.23.496.
50. Bakshi A, Fisher O, Dagci T, Himes BT, Fischer I, Lowman A. Mechanically
engineered hydrogel scaffolds for axonal growth and angiogenesis after
transplantation in spinal cord injury. JNeurosurg Spine. 2004;l(3):322-329.
doi: 10.317 l/spi.2004.1.3.0322.
51. Johnson PJ, Parker SR, Sakiyama-Elbert SE. Fibrin-based tissue engineering
scaffolds enhance neural fiber sprouting and delay the accumulation of reactive
astrocytes at the lesion in a subacute model of spinal cord injury. JBiomedMater
Res A. 2010;92(1): 152-163. doi:10.1002/jbm.a.32343.
52. Hejcl A, Sedy J, Kapcalova M, et al. HPMA-RGD hydrogels seeded with
mesenchymal stem cells improve functional outcome in chronic spinal cord injury.
Stem Cells Dev. 2010; 19(10): 1535-1546. doi:10.1089/scd.2009.0378.
53. Stokols S, Tuszynski MH. Freeze-dried agarose scaffolds with uniaxial channels
stimulate and guide linear axonal growth following spinal cord injury.
Biomaterials. 2006;27(3):443-451. doi: 10.1016/j.biomaterials.2005.06.039.
54. Moore MJ, Friedman J a., Lewellyn EB, et al. Multiple-channel scaffolds to
promote spinal cord axon regeneration. Biomaterials. 2006;27(3):419-429.
doi: 10.1016/j .biomaterials.2005.07.045.
55. Hurtado A, Moon LDF, Maquet V, Blits B, Jerome R, Oudega M. Poly (D,L-lactic
acid) macroporous guidance scaffolds seeded with Schwann cells genetically
modified to secrete a bi-functional neurotrophin implanted in the completely
81


transected adult rat thoracic spinal cord. Biomaterials. 2006;27(3):430-442.
doi: 10.1016/j. biomaterials.2005.07.014.
56. Nomura H. Complete spinal cord transection treatedby implantation of a
reinforced synthetic hydrogel channel results in syringomyelia and caudal
migration of the rostral stump. Neurosurgery. 2006;59(1): 183-192.
http://journals.lww.eom/neurosurgery/Abstract/2006/07000/COMPLETE_SPINA
LCORDTRAN SECTIONTREATEDB Y.22. aspx.
57. Iannotti C, Zhang YP, Shields LBE, et al. Dural repair reduces connective tissue
scar invasion and cystic cavity formation after acute spinal cord laceration injury
in adult rats. JNeurotrauma. 2006;23(6):853-865. doi:10.1089/neu.2006.23.853.
58. Macaya D, Spector M. Injectable hydrogel materials for spinal cord regeneration: a
review. BiomedMater. 2012;7(1):012001. doi: 10.1088/1748-6041/7/1/012001.
59. Piantino J, Burdick J a., Goldberg D, Langer R, Benowitz LI. An injectable,
biodegradable hydrogel for trophic factor delivery enhances axonal rewiring and
improves performance after spinal cord injury. Exp Neurol. 2006;201(2):359-367.
doi: 10.1016/j. expneurol .2006.04.020.
60. Jain A, Kim Y-T, McKeon RJ, Bellamkonda R V. In situ gelling hydrogels for
conformal repair of spinal cord defects, and local delivery of BDNF after spinal
cord injury. Biomaterials. 2006;27(3):497-504.
doi: 10.1016/j .biomaterials.2005.07.008.
61. Ruel-Gariepy E, Leroux J-C. In situ-forming hydrogelsreview of temperature-
sensitive systems. Eur JPharm Biopharm. 2004;58(2):409-426.
doi: 10.1016/j.ejpb.2004.03.019.
62. Gupta D, Tator CH, Shoichet MS. Fast-gelling injectable blend of hyaluronan and
methylcellulose for intrathecal, localized delivery to the injured spinal cord.
Biomaterials. 2006;27(11):2370-2379. doi: 10.1016/j .biomaterials.2005.11.015.
63. Ritfeld GJ, Rauck BM, Novosat TL, et al. The effect of a polyurethane-based
reverse thermal gel on bone marrow stromal cell transplant survival and spinal
cord repair. Biomaterials. 2014;35(6): 1924-1931.
doi: 10.1016/j .biomaterials.2013.11.062.
64. Engineering NT. A Biomimetic Reverse Thermal Gel for 3-Dimensional Neural
Tissue Engineering. 2014; 1(4): 1-5.
65. Castellanos BP, Park D, Long CS, et al. Temperature-Responsive Materials May
Act As Extracellular Matrix and Cell Vehicle for Cardiac Tissue Engineering. J
Am Coll Cardiol. 2015;65(10):A960. doi:10.1016/S0735-1097(15)60960-3.
82


Full Text

PAGE 1

A BIOMIMETIC REVERSE THERMAL GEL FOR SPINAL CORD INJURY b y JAMES BARDILL B.S., University of Minnesota Twin Cities, 2008 A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Master of Science Bioengineering 2015

PAGE 2

ii This thesis for the Master of Science degree by James Bardill has been approved for the Bioengineering Program by Daewon Park, Chair Vikas Patel Rony Marwan November 5, 2015

PAGE 3

iii Bardill, James (M.S., Bioengineering) A Biomimetic Reverse Thermal Gel for Spinal Cord I njury Thesis directed by Assistant Professor Daewon Park ABSTRACT No effective treatment for paralysis after spinal cord injury (SCI) exists. To meet this unmet medical need, researchers have developed various molecular and cell therapi es to repair or replace damaged spinal tissue. Despite the promise these treatments hold, their therapeutic effect remains limited due to inefficient delivery methods. To overcome this obstacle, biomimetic polymer scaffolding systems have been d eveloped to control the delivery of these new treatments to injured spinal cord tissue. These polymers are designed to integrate into surrounding tissues to provide support to damaged tissue while also provi ding an environment within the SCI lesion that encoura ges tissue regeneration. An injectable polyethylene glycol based reverse ther mal gel was developed to deliver a minimally invasive treatment directly into the lesion si te of a compressed SCI. To mimic the extracellular environment, the polymer was conju gated to arginylglycylaspartic acid, a cellular adhesive peptide that promotes neural attachment, survival, and growth. In vitro tests revealed the polymer had no cytotoxic effects on PC12 cellular proliferation and could support the survival and growth o f a primary culture of retinal ganglion cells in a 3D polymer matrix. The polymer was then injected capability to provide axonal support within the SCI lesion site. Although functional and histological analysis were unable to demonstrate definitive evidence of the polymer

PAGE 4

iv providing axonal support within the SCI lesion, regenerating ax ons were detected within the SCI site where the polymer was injected These prelim inary results demonstrate this novel polymer has the capability to act as a promising neural scaff old to promote a regenerative environment with in a SCI lesion site For future studi e s, we predict incorporating cellul ar and molecular therapies into the p olymer will provide an effective delivery vehicle to enhance therapeutic efficiency of SCI treatment approaches. The form and content of this abstract are approved. I recommend its publication. Approved: Daewon Park

PAGE 5

ACKNOWLEDGEMENTS First and foremost, I would like to thank my advisor, Dr. Daewon Park fo r giving me the opportunity to g ain insight into exciting tissue engineering approaches with biomaterials His guidance and instruction have been invaluable and I look forward to future wor k with him. Dr. Vikas Patel took time out of his busy clinical/ surgery schedule to provide me vital insight f or surgical techniques for the in vivo study that were instr umental in completing this project Dr. Rony Marwan provided great insight into spina l cord development and clinical approaches that were crucial for developing a practical therapeutic approach. I look forward to continuing to work with him on a future project related to spina bifida defects. One of the most important aspects of my experi ence was enjoying the work environment of the Translational Biomaterials Research Laborat ory. Finding a workplace with as much enthusiasm and encouragement as this lab provided is truly rare, and I am fortunate to have shared this experien ce with current a nd past lab members : Melissa Laughter, David Lee, Anna Laura Nelson, Madia Stein, Adam Rocker, Ryan Brody, Maria Bortot, Krishna Madhaven, Matt Taylor, and Lindsey Hockensmith. Finally, I would like to thank my family. My dad, mom, and three sisters have been nothing but supportive of everything I have ventured into during my life.

PAGE 6

vi Declaration of original work by James Bardill support from my advisor, committee memb ers, lab members, fellow students, and the Department of Bioengineering. The research and ideas presented in this document originated from the Translational Biomaterials Research Laboratory under the guidance of Dr. Daewon Park. All resources and funds wer e provided by Dr. Daewon Park and the Department of Bioengineering. James Bardill

PAGE 7

vii TABLE OF CONTENTS Chapter 1 Introduction ................................ ................................ ................................ ................. 1 1.1 Spinal cord injury background information ................................ ......................... 1 1.2 SCI epidemiology ................................ ................................ ................................ 2 1.3 SCI pathophysiology ................................ ................................ ............................ 2 1.4 Current clinical SCI treatments ................................ ................................ ............ 4 1.4.1 Surgical intervention ................................ ................................ ..................... 4 1.4.2 Pharmacological intervention ................................ ................................ ....... 5 1.5 Emerging therapies in SCI treatment ................................ ................................ ... 5 1.5.1 Molecular approaches ................................ ................................ ................... 6 1.5.2 Cellular approaches ................................ ................................ ....................... 7 1.5.3 Challenges of emerging therapies ................................ ................................ 9 1.6 Goal of this study ................................ ................................ ............................... 10 2 Literature review ................................ ................................ ................................ ....... 12 2.1 Polymers scaffolds for SCI ................................ ................................ ................ 12 2.1.1 Biomedical applications of PEG ................................ ................................ 13 2.1.2 SCI scaffold designs ................................ ................................ ................... 16 2.1.3 Surgically implanted scaffold designs ................................ ........................ 16

PAGE 8

viii 2.1.4 Injectable scaffolds ................................ ................................ ..................... 17 2.1.5 Reverse thermal gels (RTGs) ................................ ................................ ...... 19 2.1.6 Previous research with RTGs ................................ ................................ ...... 20 2.1.7 RGD peptide ................................ ................................ ............................... 21 2.2 Animal Models of SCI ................................ ................................ ....................... 24 2.2.1 Rat models of SCI ................................ ................................ ....................... 25 2.2. 2 Functional testing ................................ ................................ ........................ 27 2.2.3 Immunohistochemistry (IHC) ................................ ................................ ..... 28 3 Research objectives and experimental approach ................................ ...................... 30 3.1 Hypothesis ................................ ................................ ................................ .......... 30 3.2 Specific a ims ................................ ................................ ................................ ...... 30 4 Materials and equipment ................................ ................................ ........................... 33 4.1 Materials ................................ ................................ ................................ ............. 33 4.2 Equipment ................................ ................................ ................................ .......... 34 5 Methods ................................ ................................ ................................ ..................... 36 5.1 Synthesis of PEGSA ................................ ................................ .......................... 36 5.2 Synthesis of carboxylic acid terminated PNIPAm ................................ ............. 37 5.3 Synthesis of PEGSA PNIPAm ................................ ................................ .......... 37 5.3.1 Molar Reaction ................................ ................................ ............................ 37

PAGE 9

ix 5.3.2 Gram base conjugation ................................ ................................ ............... 38 5.4 RGD conjugation ................................ ................................ ................................ 38 5.4.1 RGD conjugation to PEGSA ................................ ................................ ...... 38 5.4.2 RGD conjugation to PEGSA PNIPAm (mol base) ................................ .... 39 5.4.3 RGD conjugation to PEGSA PNIPAm (gram base) ................................ .. 39 5.5 Polymer biocompatibility testing ................................ ................................ ....... 40 5.5.1 PC12 cell culture ................................ ................................ ......................... 41 5.5. 2 Retinal ganglion cell (RGC) cell culture ................................ .................... 41 5.5.3 MTT cytotoxicity ................................ ................................ ........................ 41 5.5.4 Alamar blue cell viability assay ................................ ................................ .. 42 5.5.5 RGC attachment and growth ................................ ................................ ....... 42 5.6 Compression SCI rat model ................................ ................................ ............... 43 5.6.1 Ethics and surgical approval ................................ ................................ ....... 43 5.6.2 Surgical procedures ................................ ................................ ..................... 44 5.6.3 Post surgery procedure ................................ ................................ ............... 45 5.7 Functional Assessment ................................ ................................ ....................... 46 5.8 Histology ................................ ................................ ................................ ............ 46 5.8.1 H&E staining ................................ ................................ .............................. 46 5.8.2 Immunohistochemistry ................................ ................................ ............... 46

PAGE 10

x 5. 9 Statistical analysis ................................ ................................ .............................. 47 6 Results and Discussion ................................ ................................ ............................. 48 6.1 Polymer synthesis ................................ ................................ ............................... 48 6.2 Polymer characterization ................................ ................................ .................... 50 6.2.1 PEGSA FT IR characterization ................................ ................................ .. 50 6.2.2 PEGSA PNIPAm characterization ................................ ............................. 51 6.2.3 PEGSA PNIPAm gelling properties ................................ ........................... 52 6.2.4 PEGSA PNIPAm morphological characterization ................................ ..... 52 6.2.5 PEGSA RGD FTIR characterization ................................ .......................... 54 6.2.6 PEGSA PNIPAm RGD FTIR characterization ................................ .......... 55 6.3 In vitro assessment ................................ ................................ ............................. 56 6.3.1 MTT cytotoxicity ................................ ................................ ........................ 56 6.3.2 Alamar blue cell viability ................................ ................................ ............ 57 6.3.3 RGC attachment and growth ................................ ................................ ....... 58 6.4 Compression SCI rat model ................................ ................................ ............... 59 6.5 Functional assessment ................................ ................................ ........................ 6 2 6. 6 Histology ................................ ................................ ................................ ............ 63 6.6.1 H&E staining ................................ ................................ .............................. 63 6.6.2 GFAP IHC ................................ ................................ ................................ .. 64

PAGE 11

xi 6.6.3 GAP 43 IHC ................................ ................................ ............................... 66 7 Conclusion ................................ ................................ ................................ ................ 68 8 Study limitations and future work ................................ ................................ ............. 72 8.1 Modify control and experimental groups ................................ ........................... 72 8.2 Modify SCI compression ................................ ................................ ................... 72 8.3 Modify PEGSA backbone to prevent gel shrinking ................................ ........... 73 8.4 Increase study duration ................................ ................................ ....................... 74 8.5 Monitor polymer presence in spinal tissue ................................ ......................... 74 8.6 Long term goal of RTG SCI application ................................ ............................ 74 References ................................ ................................ ................................ ........................ 77

PAGE 12

xii LIST OF FIGURES Figure 1 1. SCI pathophysiology depicting the primary injury stage caused by impact and the secondary injury cascade forming the glial scar 7 ................................ .................. 3 1 2. Signaling pathways activated by neurotrophic factors. ................................ ............... 7 1 3. NSC therapy for SCI (B) aims t o reduce glial scarring and improve axon myelination to restore normal function (A) 32 ................................ ................................ ..... 9 2 1. Desirable properties of polymer sc affolds for tissue engineering applications 34 ................................ ................................ ................................ ................................ ........ 12 2 2. A 10% RTG solution (mg/mL) at room temperature (left) followed by 10 seconds of exposure to 37C water shows formation of solid gel (right). ......................... 19 2 3. Schematic of RGD integrin binding to trigger multiple signaling pathways. ........... 22 2 4. Polymer scaffold with conjugated RGD showed substantial differentiation of human NSCs into motor neurons. Inset is a schematic of the polymer. Green represents axon extension Tubulin staining. Red represents motor neuron markers (Islet 1, HB9) 71 ................................ ................................ ...................... 23 6 1. Reaction of PEGDGE with SA to produce PEGSA. The PEGSA repeating unit contains two hydroxyl groups that will be used for further chemical conjugation. ................................ ................................ ................................ ........................ 48 6 2. Conjugation of PNIPAm to PEGSA backbone. ................................ ........................ 49 6 3. PEGSA PNIPAm RGD reaction sequence. GRGDS is a modified RGD peptide. ................................ ................................ ................................ ............................... 50 6 4. FT IR spectrum of PEGDGE and PEGSA ................................ ................................ 51 6 5. FT IR spectrum of PEGSA, PNIPAm, PEGSA PNIPAm ................................ ........ 51 6 6. LCST of PEGSA PNIPAm conjugation reactions and PNIPAm. ............................. 52 6 7. SEM micrograph (x180) showing porosity of PEGSA PNIPAm with cons istent porous size throughout the structure. Scale bar 100m ................................ .... 53 6 8. SEM micrograph (x950) showing an average pore size of 5m 20 m .................... 54 6 9. FT IR spectrum of PEGSA, RGD, and PEGSA RGD ................................ .............. 55

PAGE 13

xiii 6 10. FT IR spectrum of PEGSA PNIPAm, RGD. The conjugation of RGD to PEGSA PNIPAm could not be confirmed because the PNIPAm peaks overshadow the RGD peaks. ................................ ................................ .............................. 56 6 11. FT IR spectrum of region a (in Figure 6 8) of PEGSA PNIPAm, RGD, and PEGSA PNIPAm RGD. The PEGSA PNIPAm peaks are located in the same chemical shift location as the RGD peaks. The RGD peaks have a weaker signal than the large PNIPAm peaks. ................................ ................................ ........................... 56 6 12. PC12 cells showed no cytotoxicity when exposed to PEGSA PNIPAm media extracts, per ISO 10993 5. Statistical analysis by ANOVA demonstrated no difference between the positive control (Media+cells) sample and experimental samples (p=0.36, n=5). Experi mental samples are normalized to cells exposed to pure media. Media only is the negative control. Error bars represent standard error of the mean. ................................ ................................ ................................ ........................ 57 6 13. PC12 cells showed no change in viability when exposed to PEGSA PNIPAm media extracts, per ISO 10993 5. Statistical analysis by ANOVA demonstrated no difference between the positive control (Media+cells) sample and experimental s amples (p=0.13, n=5). Experimental samples are normalized to cells exposed to pure media. Media only is the negative control. Error bars represent standard error of the mean. ................................ ................................ ................. 58 6 14. RGC axon extensions within PEGSA PNIPAm RGD 3D polymer matrix. tubulin, nuclei with DAPI, dead cells appear red. Confocal microscopy z stack, scale bar = 100 m. ................................ ........................... 59 6 15. Compression spinal cord injury. (A) Exposed thoracic vertebrae (T9 T11), (B) exposed spinal cord tissue after laminecto my, (C) spinal cord compression with 30 g aneurysm clip. ................................ ................................ ................................ .... 60 6 16. A 10 l injection of PEGSA PNIPAm dyed with toluidine blue conf irmed localization of the polymer into the spinal cord injection site. ................................ .......... 62 6 17. Vertical spinal cord sections stained with H&E sh owing the progression of glial scarring after compression SCI with polymer injection. (A) Healthy, uninjured tissue, (B) compressed spinal cord 4 hours after surgery with polymer injection, (C) 4 weeks after surgery. Scale bar = 500 m ................................ ................ 64 6 18. Glial scar formation of longitudinal compressed spinal cord after 8 weeks with polymer injection. The decreased cellular density at the SCI site could be due to the presence of the polymer, which does not stain with H&E. Scale bar = 1000 m ................................ ................................ ................................ ............................. 64 6 19. GFAP immunostaining to detect astrocyte accumulation around SCI site in vertical spinal cord sections from (A) representative 4 week polymer injection, (B) 8 week polymer injection. Images acquired with Nikon confocal microscope, 20X air objective. Scale bar = 500 m ................................ ................................ .............. 65

PAGE 14

xiv 6 20. GAP 43 immunostaining to detect regenerating axons. Vertical SCI lesion site from representative 4 week polymer injection (top to bott om representing rostral and caudal ends of the spinal cord). Scale bar = 500 m. Confocal microscopy SCI image, 10X air objective. ................................ ................................ ........ 66 6 21. GAP 43 immunostaining to detect regenerating axons. Longitudinal spinal cord section from 8 week polymer injection (left and right sides representing rostral and caudal ends of the spinal cord). Scale bar = 500 m. Confocal microscopy SCI image, 10X air objective. ................................ ................................ ........ 67 8 1. After several minutes, PEGSA PNIPAm shrinks, resulting in a gel that no longer holds the shape of the vial. ................................ ................................ ..................... 73

PAGE 15

xv LIST OF TABLES Table 1. Surgical outcomes and complications of saline injected rats. ................................ ....... 61 2 Surgical outcomes and complications of polymer injected rats ................................ .... 61 3 Summary of BBB scores for 4 week rats. Only 1 saline injection rat survived the entire 4 week duration of the study, and the polymer injection rats have a large variability ( 3.4 3.7, n = 6). ................................ ................................ .............................. 63

PAGE 16

xvi LIST OF ABBREVIATIONS 3D 3 dimensional ANOVA analysis of variance BBB Basso, Beattie, and Bresnahan BDNF brain derived neurotrophic factor BSA Bovine serum albumin chABC chondroitinase ABC CNS central nervous system CNTF ciliary neurotrophic factor CSF Cerebral spinal fluid CSPG chondroitin sulfate proteoglycans DAPI diamidino 2 phenylindole DCC Dicyclohexylcarbodiimide DCM dichloromethane DMAP 4 Dimethylaminopyridine DMF N,N dimethylformamide DMSO dimethyl sulfoxide ECM extracellular matrix ED1 Anti CD68 macrophage EDC N (3 dimethylamino propyl) N ethylcarbodiimide hydrochloride ESC embryonic stem cell FBS Fetal bovine serum FDA Food and Drug Administration FT IR Fourier transform infrared spectroscopy GAP 43 Growth associated protein 43 GFAP Glial fibrillary acidic protein RGD Gly Arg Gly Asp Ser HAMC hyaluronic acid and methylcellulose H&E Hema toxylin and Eosin hNSC human neural stem cell HPMA N (2 hydroxyproplyl) methacrylamide HS Horse serum IACUC institutional animal car and use committee IHC immunohistochemistry IL 10 Interleukin 10 IP Intraperitoneal IR infrared LCST Lower critical solution temperature

PAGE 17

xvii LFB Luxol fast blue MP methylprednisolone MSC mesenchymal stem cell MTT 3 (4,5 dimethylthiazol 2 yl) 2,5 diphenyltetrazolium bromide MW Molecular weight NGF nerve growth factor NHS N hydroxysuccinimide NIPAAm N Isopropylacrylamide NSC neural stem cell NT 3 neurotrophin 3 NTF Neurotrophic factors OB Olfactory bulb OCT optimal cutting temperature OEC Olfactory ensheathing cell OM Olfactory mucosa PBS phosphate buffered saline PC12 pheochromocytoma cell line PCL Poly(caprolactone) PEG Poly(ethylene glycol) PEGDGE Poly(ethylene glycol diglycidyl ether) PEGSA Poly(ethylene glycol diglycidyl ether) succinic acid PEGSA PNIPAm PEGSA conjugated to PNIPAm PEGSA PNIPAm RGD PEGSA PNIPAm conjugated to RGD peptide PFA Paraformaldehyde ( 4% solution in PBS) PGA polyglycolic acid PHEMA poly(2 hydroxyethylmethacrylate PLGA poly(lactic co glycolic acid) PNI peripheral nerve injury PNIPAm poly(N isopropyl acrylamide) PNS peripheral nervous system PS Penicillin streptomycin PSHU poly(serinol hexamethylene urea) PSHU RGD arginylglycylaspartic acid conjugated poly(serinol hexamethylene urea) PVA Poly(vinyl alcohol) RGC Retinal ganglion cell RGD arginylglycylaspartic acid, Arg Gly Asp

PAGE 18

xviii RTG Reverse thermal gel SA Succinic acid S C subcutaneous SCI spinal cord injury SEM scanning electron microscope SNI sciatic nerve injury UV VIS Ultraviolet visible TPP triphenylphosphine

PAGE 19

1 1 Introduction 1.1 Spinal cord injury background information Spinal cord injuries (SCI) result in devastating neurological deficits that lead to long term health and lifestyle complications. Paralysis (quadriplegia or paraple gia) is the most common disability caused by SCI, with other serious disabilities including loss of bladder / bowel control, i ncreased risks for urinary tract i nfections, and cardiopulmonary distress. The severity of a SCI depends on the location of the injury to the spinal column and corresponds to the degree of motor and sensory loss experienced by the patient. For patients with cervical level injuries, ailments include quadriplegia, loss of urinary/ bowel control, and respiratory ailments. Higher level cervical injuries often require constant personal care assistance, while lower level cervical injur ies may be managed with assistive technologies for more independent patient activity. Thoracic level SCI ailments include pa raplegia and loss of bladder/ bowel control, though control over respiration, arms, and hands is usually normal, allowing for more patient independence. L umbar and sacral level injury ailments include partial loss o f hip/ leg control and loss of bladder/ bowel control, with the lower level injuries allowing for more control over walking ability. The severity of a SCI also depends on whether the injury is complete or incomplete. Complete injuries result in total loss of motor and sensory function below the injury site, while incomplete injuries result in partial loss of motor and sensory function.

PAGE 20

2 Despite improvements in clinical care and physical therapy t hat have led to longer patient survival rates, increased patient independence, and overall improved quality of life, there is currently no treatment capable of restoring all lost functions. 1.2 SCI epidemiology An estimated 250,000 people in the United States live with SCI, with 11,000 new cases reported each year 1 The overall leading cause s of SCI are falls and motor vehicle accidents, followed by acts of violence and sports injuries. Injuries at the cervical level are the most common, followed by thoracic/lumbosacral injuries. The lifetime costs of SCI can reach up to $5,000,000 for quadr iplegics and up to $2,000,000 for paraplegics, not taking into account loss of wages or occupational benefits 2 SCI related mortality rates are highest within the first 12 18 months of injury, with cervical inj ury related deaths having the highest m ortality rates 3 1.3 SCI patho physiology Damage to the central nervous system ( CNS ) causes devastating neurological and functional deficits that are difficult to overcome. The longstanding belief for decades was th at CNS neurons possessed no intrinsic repair mechanisms. Howeve r, a key discovery to promote CNS neural r epair was to provide damaged neurons a permissive cellular environment that resembles the environment of neurons within the peripheral nervous system ( PNS ) 4,5 These findings revealed that damaged CNS did not lack intrinsic regenerative capability. Rather, CNS injury generates a toxic cellular environment that fails to support damaged neuron tissue and inhibits regenerative capability 6 T his harsh environment has become one of the key therapeutic targets in SCI research to promote repair of damaged CNS neurons.

PAGE 21

3 The pathophysiology of SCI is highly complex process that is not well understood at the molecular level due to the varying degree s of injury seen from patient to patient. Instead, a general understanding of SCI is broken down into different phases of the injury and the effects each phase has on the spinal cord. SCI s are classified into a primary and secondary ph ase ( Figure 1 1 ) 7 Most traumatic SCI are caused by forces to the vertebral column that result in compression, contusion, displacement, and/or laceration of the spinal cord. This primary mechanical damage results in axon severing and substantial damage to neurons, interneurons, and glia within the spinal cord. Figure 1 1 SCI pathophysiology depicting the primary injury stage caused by impact and the secondary injury cascade forming the glial scar 7 Within minutes of the primary damage, a cascade of secondary damage occurs at the molecular level that lead s to chronic neurological deficits. This cascade of secondary events includes: (a.) vascular bleeding and ischemia that lead to necrosis of spinal tissue,

PAGE 22

4 breakdown of the blood spinal cord barrier, and an influx of inflammatory cells (b.) the form ation of free radicals that cause oxidative damage to neurons (c.) disruption of ionic and amino acid balances (ex. glutamate excitotoxicity) that lead to apoptosis of neurons 1 Ultimately, axons and gray matter are damaged and replace d with connective tissue deposits that lead to the formation of a glial scar at the site of the injury. The glial scar consists of reactive astrocytes, microglia, macrophages, and chondroitin sulfate proteoglycans 1 This scarring acts as a barrier to the extension of axons through the injury, preventing regeneration and replacement of damaged spinal tissue. Overall, a general understanding of the direct effects of SCI is available, but knowledge of specific biological mechanisms and p athways remains elusive. 1.4 Current clinical SCI t reatments The complexity and overall lack of understanding of SCI pathophysiology has made new treatment approaches difficult to take to the clinical level. Surgical intervention and methylprednisolone (MP) are two of the few treatment approaches available to patients with SCI, however, even these approaches are now marred in controversy between physicians questioning their overall efficacy. 1.4.1 Surgical i ntervention The priority for patients with an acute SCI is immediate spinal immobilization to prevent further primary damage to the spinal cord. Even the slightest movements to an injured spinal cord can exacerbate any existing primary damage. In addition to the primary damage, p rolonged compression of the spin al cord contributes to the long term cascade of secondary damage to spinal cord tissue at the molecular level. Surgical

PAGE 23

5 decompression and realignment of the injured spinal cord helps stabilize spinal blood flow, reducing the effects of long term ischemia. Patients that have undergone spinal decompression surgeries have shown significantly improved neurological outcomes, but the timing of performing these surgeries remains controversial 8 1.4.2 Pharmacological i ntervention The administration of the cortico steroid MP is cur rently the only approved treatment for patients with acute SCI that specifically targets the secondary injury cascade MP was found to be effective in combatting the damaging free radical oxidation of neural lipid membranes that occurs during the secondary effects of SCI 9 Despite early reports that administration of MP improved neurological outcomes when given 8 hours after acute SCI, further studies have shown that there may be no shor t term or long term benefits 8 In fact, the immune suppressive effects of MP has been found to have severe side effects, including gastrointestinal bleeding, infections (wound, respiratory, and urinary), sepsis, peptic ulcer disease, and hyperglycemia 10 Des pite MP being one of the few SCI treatments around it is no longer prescribed by many physicians due to these recent findings 9 In addition, some physicians that do prescribe MP but are uncertain of the risks often do so out of fear of litigation 10 11 Nevertheless, the administrat ion of MP remains a highly controversial subject for SCI treatment 12 1.5 Emerging therapies in SCI treatment Despite the complexity of SCI pathophysiology, advances in cel lular and molecular research have advanced the understanding of SCI mechanisms, opening up new avenues for treatment strategies. Each new approach is designed to target the

PAGE 24

6 secondary SCI cascade to help create an environment within the injury site to promote neural regeneration. 1.5.1 Molecula r approaches 1.5.1.1 Neuroprotective a pproaches The secondary cascade following SCI involves a complex i nflammatory response that has both protective and damaging effects on spinal tissue 13 Researchers are exploring molecular techniques that exploit the beneficial effects of this inflammation and combat the detrimenta l effects. Interleukin 10 (IL 10) is a known anti inflammatory cytokine that is capable of inhibiting monocyte and macrophage production in PNI. Systemic administration of IL 10 into rats with a contusion SCI showed significant reduction of glial scar s ize and also showed improved hind limb motor function 14 In additio n to the inflammatory response, the secondary injury cascade also results in the formation of a glial scar. This scar formation acts as a physical barrier to axonal regeneration through the SCI site, contributing to long term neurological and functional d eficits. Following injury, astrocytes enhance the expression of extracellular matrix (ECM) proteins, particularly chondroitin sulfate proteoglycans (CSPGs) 15 Enzymatic degradation of CSPGs by chondroitinase ABC (ChABC) promotes a more accommodating environment within the lesion for axonal regeneration and enhanced motor recovery 1 1.5.1.2 Neurotrophic factors Neurotroph ic factors (N TFs) activate signaling pathways in developing neurons to induce growth, guidance, differentiation, and survival of neural cells ( Figure 1 2 ).

PAGE 25

7 Figure 1 2 Signaling pat hways activated by neurotrophic factors Previous studies of peripheral nerve injury (PNI) showed NTFs play a key role in axon growth and regeneration, making NTFs a promising therapeutic approach for SCI 1 Administration of b rain derived neur otrophic factor (BDNF) has demonstrated exten sive axonal growth at SCI sites in a rat model 16 regeneration of severed axons, axonal myelination, and enhanced neural plasticity 17 Ciliary neurotrophic factor (CNTF) treatment in a rat hemisection SCI model showed significant neuron regeneration and axonal growth, further enhancing the promising application of NTFs to promote recovery after SCI 18 A combination of these and ot her NTFs will likely activate more NTF signaling pathways to promote an environment in the SCI lesion capable of repair and regeneration 19 1.5.2 Cellular approaches Patients with severe SCI are often left with limited treatment options from their physicians. Until recently, there was no known regenerative capability of CNS tissue. Fortunately, recent cellular research has shown tha t several different types of neural cells

PAGE 26

8 and stem cells could provide injured spinal cord tissue the ability to repair and provide significant functional recovery. 1.5.2.1 Olfactory ensheathing c ells (OECs) One of the first CNS regions to demonstrate continuous r egenerative capability was observed in the olfactory nerve. Olfactory axons are peripheral nerves located in the olfactory mucosa (OM) that undergo constant regeneration. As these axons grow, they extend into the olfactory bulb (OB), located in the brain 1 OECs are beli eved to be the driving force to support and guide this neural regeneration, making OECs a potential therapeutic approach for SCI 20 21 A spinal cord compression injury in rats with transplanted OECs into the injury site demonstrated reduced glial scar format ion and axon regeneration across the lesion site 22 while a transection SCI model in rats with transplanted OECs into the injury site also showed axonal re generation 23 Human trials ha ve also shown sensory and motor improvements in multiple patients with OEC transplantation 24 Concerns about the long term effects of OEC transplants have also been reported. One patient received an OEC transplant into the injury site three year s after the injury and experienced back pai n 8 years after the transplant, likely caused by the development of a cystic mass at the site of the transplantation, which required removal 25 1.5.2.2 Mesenchymal stem c ells (MSCs) MSCs are an attractive option for SCI treatment because they are easily accessible, plentiful, and have an immunosuppressant effect 1 Bone marrow MSCs were found to be effective in improving functional scores in animal models of traumatic SCI 26 Human trials using MSCs reported significant bladder and bowel control

PAGE 27

9 improvements, in addition to motor improvements 27 However, further patient studies with MSC transplantation have reported significant side effects, i ncluding urinary tract infections, respiratory distress, and pain 28 1.5.2.3 Neural stem c ells (NSCs) NSCs are multipotent cells that can differentiate into any cell type in the central nervous system, which has made them a promising therapeutic candidate for SCI 28 The aim of NSCs for SCI treatment is transplantation into the injury site to allow the cells to differentiate into oligodendro cytes and other neurons that can replace damaged tissue ( Figure 1 3 ) 29 When grafted into SCI sites, NSCs were able to differentiate into multiple neuronal phenotypes and also exhibited long distance axon extension 30 NSCs also ha ve beneficial effects after SCI within the glial scar, acting to preserve tissue integrity and supply neurotrophic factors, essentially protecting the spinal cord from further damage 31 Figure 1 3 NSC therapy for SCI (B) aims to reduce glial scarring and improve axon myelination to restore normal function (A) 32 1.5.3 Challenges of emerging therapies The results of cellular and molecular therapy approaches have demonstrated promising capability to combat the challenges of SCI tissue rep air. The success of these

PAGE 28

10 new approaches, however, has been highly variable and limited. Most molecular or cellular agents are delivered by direct injection into the injury site, which have problems with localized delivery to targeted regions. Deliverin g cellular therapies present a particular challenge for cell viability. Direct injections of cellular therapies into injured spinal tissue often results in substantial cell death due to the harsh environment of injury site. This environment lacks the nec essary ligands for transplanted cells to attach to, causing cell death. 33 Biomaterial scaffolding systems have been developed to create more effective delivery methods for cells, drugs, and molecular agents for many different tissue engineering applications. These scaffold designs mimic the natural ECM envi ronment to support cellular adhesion, differentiation, and proliferatio n. A dministration of cellular and molecular agents within a biomaterial scaffold will be c rucial to enh ance therapeutic efficacy to promote neuron support, guidance, and repair across the SCI site. 1.6 Goal of this study Treatments for SCI remain ineffective, however, the development of new cellular and molecular therapies bring promising potential for significant repair of damaged spinal tissue. Without a scaffolding system to act as a supportive substrate, the efficacy of these potential therapies will remain limited and inconsistent. Synthetic polymers are practical biomaterials that can be deployed to act as scaffolding devices for these potential therapies to help ensure the most e ffective means to deliver the treatment. Polymer scaffolds p roperties can be modulated for different delivery applications to help support nerve regenerative capability. Many types of polymer scaffolds have been

PAGE 29

11 developed but also present limitations with delivery due to invasive procedures needed for implantation. The overall objective of this study is to develop a novel, polyethylene glycol (PEG) based injectable polymer scaffold that can act as a supportive substrate for injured spinal cord tissue. This polymer surface will be functionalized with a cellular adhesive laminin peptide to al low the polymer scaffold to mimic the ECM. This surface modification will allow the polymer to interact with and cue neural integrin receptors to induce cellular a ctivities, such as growth, proliferation, and differentiation. The injectable properties of this polymer provides a minimally invasive delivery method to targeted tissue while also providing a gelled substrate to support therapeutic agents. This injectab le polymer, known as a reverse thermal gel (RTG), will be evaluated for neuron support in an in vitro retinal ganglion cell model and will be assessed for regenerative capability in a rat model of compression SCI.

PAGE 30

12 2 Literature r eview 2.1 Polymers scaffolds for SCI A variety of different materials have been explored as potential tissue engineering scaffolds for SCI. Scaffold designs for SCI need to take into account biocompatibility, biodegradability, porosity, mechanical stability, and cellular adhesion ( Figure 2 1 ) 34 As SCI pathophysiology is complex and generally not well understood sorting out the best approaches for scaffold designs has become a difficult endeavor. Figure 2 1 Desirable properties of polymer scaffolds for SCI tissue engineering applications 34 Polymer scaffolds are made of natural or synthetic materials, or often a combination of the two. Naturally derived biopolymers have shown excellent potential for regenerativ e medicine approaches. These m aterials include collagen, fibrin, hyaluronic acid, chitosan, polysaccharides and peptides 35 These natural biomaterials

PAGE 31

13 offer excellent biocompatibility, minimal cytotoxic effec ts, and mimic the ECM for optimal tissue support and cellular attachment 36 Naturally derived polymers, however, are limited by generally poor mechanical properties, are often costly to obtain, and have limited chemical modification sites 37 To overcome the disadvantages of naturally derived polymers, synthetic materials have be en explored as t issue engineering scaffolds. Synthetic biodegradable polymers include polyvinyl alcohol (PVA), polycaprolactone (PCL), poly (lactic acid) (PLA), poly (glycolic acid) (PGA), poly(lactic acid co glycolic acid) (PLGA), and PEG 37 S ynthetic materials offer the ability to control desired scaffold properties for optimal integ ration into tissue, including mechanical properties, porosity, degradation, and biomolecule chemical conjugations. 2.1.1 Biomedical applications of PEG PEG has become one of the most researched biomaterials for tissue engineering scaffold devices. PEG exists in linear or branched forms with a variety of molecul ar weights (MWs) and a wide variety of chemical properties. Most biomedical applications of PEG use MWs of several hundred to approximately 20,000 38 An excellent reference for the biomedical applications of PEG can be found in a textbook written by J. Milton Harris, From a chemical react ion perspective, PEG has become a po pular selectio n for biomedical applications with its wide range of solubilities in different organic solvents. Posse ssing a wide range of solubilities in organic solvents increases the number of different chemical conjugation techniques that can be appli ed to a reaction, thereby

PAGE 32

14 increasing PEGs flexibility for tunable biomedical properties and applications. Similarly PEG is also insoluble in organic solvents like ether and hexane. This makes PEG a user friendly polymer for the precipitation and the puri fication of final reaction products. In addition to its excellent solubility properties, the terminal hydroxyl end groups of PEG can be covalently bonded to other polymers or molecules. PEG is often attached to hydrophobic polymers, such as polyurethanes, to enhance the polymers support for physiological applications 38 39 40 Furthermore, modi fying polymers with PEG also increases their overall size, a property that has been utilized to re tain biomolecules within systemic circulation for longer periods of time. This approach has been used in the pharmaceutical industry to prolong the half life of drugs in circulation by decreasing their clearance by the renal system 41 From a bio chemical perspective, the hydrophilic properties of PEG are what make it a promising substrate for tissue engineering applications. PEG is water soluble, nonimmunogenic, and mostly inert when expo sed to biological materials. The Food and Drug Administrat ion (FDA) has also classified PEG as non toxic 40 A common application in tissue engineering for PEG is the formation of 3 dimensional (3D) cross linked scaffolds. A 3D, cross linked polymer scaf fold results from long polymer chains covalentl y binding to one another, forming a network of polymer chains that ar e insoluble in water. T hese cross linked polymer networks often referred to as hydrogels, are able to absorb large quantities of water The properties of the hydrogel (mechanical and de gradation) can be tuned by the amount of chemical crosslinks in the network, which are controlled by various reaction mechanisms 4 2 The large volume of water retained by hydrogels makes them attractive scaffold designs in tissue engineering applications.

PAGE 33

15 Specifically, the high water content environment is able to mimic physiological conditions and allow exchange of nutrients and wastes within the polymer matrix 43 These properties of PEG make it an attractive biomaterial for tissue engineering applications. In vitro models using 3D PEG hydrogels encapsul ated with neural cells showed survival proliferation, and differentiation into functional neurons that can respond to neurotransmitters 44 Injectable PEG scaffolds with poly (N isopropylacrylamide) ( PNIPAm ) have also been investigated for mechanical properties and neurotrophin release. In vitro analysis revealed that a controlled release of BDNF and NT 3 from the polymer scaffold for four weeks. This PEG PNIPAm scaffold was also designed to have similar mechanical compressive strength to spinal tiss ue 45 These results show that a PEG PNIPAm scaffold can serve as a potential treatmen t approach for SCI by effectively delivery therapeutic agents and withstanding mechanical forces within the spinal cord. PEG has also shown promising potential to inhibit the damaging inflammatory effects that occur after SCI. Part of the secondary cascade of SCI involves lipid peroxidation reactions that degrade neural membranes and axonal components An in vitro guinea pig SCI model showed that PEG can act to restore cell membrane structural integrity and decrease free radial lipid peroxidation of neuron s 46 The authors found that PEG itself is not capable of decreasing reactive oxygen species, but instead, may be able to block oxidative damage to neuron cell membranes by supporting membrane integrity after SCI 47 Further studies using a n in vivo compression SCI model in guinea pigs showed that a subcutaneous injection of a PEG solut ion 6 hours after the injury promoted significant improvements in somatosensory and reflex tests 48 This ability to provide a

PAGE 34

16 neurons protection from the oxidative stresses of SCI, even without the use of additional biomolecules, has made PEG an attractive biomaterial for SCI applications. 2.1.2 SCI s caffold designs The flexibility and control over the properties of polymer scaffold designs has made them promising biomaterials to act as supportive substrates for SCI repair. The different stages of SCI pathophysiology have led to the development of a v ariety of scaffold designs with aims to repair damaged/lost tissue, support undamaged tissue, and promote axon growth across the lesion site 49 Incorporating c ellular and molecular therapies into tissue engineered scaffol ds has been the most effective approach for significant axon regeneration and motor recovery from SCI 1 2.1.3 Surgically implanted scaffold designs When designi ng a scaffolding system for SCI, the delivery method of the scaffold to the injured spinal cord is an important consideration to take into account. Surgically implantable scaffolds are a common approach to delivering therapeutic agents. The main advantag es of these scaffold designs are the fine control of the mechanical properties of the scaffold to match the spinal tissue 33 and control over the microstructure desired. One implantable scaffold design using poly ( 2 hydroxyethylmethacrylate) (PHEMA) was tailored to match the mechanical modulus of spinal tissue. This hydrogel was then loaded with BDNF and implanted into a transection SCI model in rats, results showing axon regeneration through th e hydrogel 50 A fibrin based scaffold design implanted into a similar rat m odel two weeks after the injury promoted neural sprouting and reduced astrogliosis at the lesion site 51 Implantable designs are also capable of supporting long term survival of implanted cells. A N (2 hydroxyproplyl) methacrylamide (HPMA)

PAGE 35

17 hydrogel conjugated to RGD peptide seeded with MSCs showed significant functi onal improvements, infiltration of axons into the hydrogel, and survival of MSCs 5 months after implantation 52 Many implantable treatment approaches are able to show axon growth but the arrangement can often be disorganized, resulting in ineffective functional recovery 53 Implantable polymer scaffold designs with microporous channels have shown superior axon growt h by providing axons linear guidance through the scaffold. These designs are also capable of supporting cellular grafts and delivery of therapeutic agents to cue neural extensions, such as BDNF 53 54, 55 T hese and many other studies demonstrate that implantable polymer scaffold designs are capable of supporting cellular and molecular therapies that promote significant spinal cord tissue regeneration and motor improvements. 2.1.4 Injectable scaffolds Despite the advantages implantable scaffold designs hold, invasive surgical techniques are required to implant the scaffold into the injury site These techniques often require excising spinal tissue that may result in long term complications. For instance, a polymer channel scaffold implanted into the SCI site of rats showed limited axon growth and limited functional improvements due to the develo pment of syringomyelia 56 Surgically implanted scaffolds will also cause disruption of the dura mater, resulting in significant fibrotic tissue scaring and possible long term cerebral spinal fluid (CSF) leaking 57

PAGE 36

18 Injectable scaffolding systems have been developed to counter the invasiveness of implantable scaffolds. These injectable designs form scaffolds in situ, filling and conforming to the lesion site to help reestablish tissue support 58 Most importantly, injectable scaffolds do not require removal or laceration of spinal tissue. Like implantable scaffolds, injectable scaffolds can also be designed to have a porous microstructure and can a ct as delivery agents for cellular, pharmaceutical, or molecular therapies. Unlike implantable scaffolds, injectable scaffolds are more challenging designs to exhibit control over mechanical properties. Scaffolds without good mechanical properties could act to further damage within the SCI lesion, collapse under the stresses of other tissue, or become solubilized and lost. Immediate transition from a soluble solution to a stable gel is critical for injection applications into spinal tissue. Different che mical designs have been fabricated for this important scaffold property. Photopolymerization of an injected polymer matrix is a common approach to form stable gels. One study showed that a PLA PEG PLA hydrogel loaded with Neurotrophin 3 (NT 3) showed con trolled release of the trophic factor that aided in significant axon growth and functional improvements in a rat SCI model 59 Though excellent results, this method required 60 seconds of photopolymerization for gelation to occur. This causes tissue to be exposed to UV light and could be damaging to neurons survival. One alternative to photopolymerization methods for in situ gel formation was attempted with an agarose gel encapsulated with BNDF loaded li pid microtubules. The solution was injected into injured rat spinal cord tissue and gelled by applying cooled nitrogen gas over the spinal cord for 30 seconds.

PAGE 37

19 These gels promoted neurite extensions within the scaffold and reduced the size of the glial scar at the lesion site 60 2.1.5 Reverse thermal gels (RTGs) Another approach to developing an effective injectable polymer scaffold that does not require an outside source for gel formation is to create a temperature responsive system. Polymers have been designed that contain both hydrophilic a nd hydrophobic components. T he properti es of these polymers change dramatically with slight changes in temperature. At room temperature, these polymers exists as aqueous solutions that are injectable through small gauge needles. When these polymers are exposed to increased temperatures, the h ydrophobic components begin to aggregate, allowing for the formation of solid gel structures ( Figure 2 2 ). Figure 2 2 A 10% RTG solution (mg/mL) at room temperature (left) followed by 10 seconds of exposure to 37C water shows formation of solid gel (right). These polymers undergo reversible physical transitions from aqueous solution to physical gels in response to temperature changes, and are known as reverse thermal gels (RTGs) (also called temperature sensitive or thermal responsive hydrogels). Like

PAGE 38

20 implantable scaffolds, these formed gels can act as scaffolding devices for tissue sup port and repair 61 The main advantage RTG designs have over other injectable scaffold designs is fast gelation upon exposure to ph ysiological temperature, without the need for photopolymerization, o rganic solvents, or chemical cross linking techniques that could be toxic to tissue. A mixture of a hyaluronic aci d and methylcellulose (HAMC) was injected into the intrathecal space of compressed rat spinal cord tissue. HAMC is unique because it exists a s an injectable gel at room temperature but the gel strength increases upon exposure to increased temperature. After a one month study, these gels were found to be compatible with spinal cord tissue and could also provide functional benefits 62 Similarly, a PEG based polyurethane RTG was loaded with bone marrow stem cells and injected into a contusion SCI model. The cells had significantly higher survival when encapsulated within the RTG and no significant difference in immune response when compared to a PBS control. Rats with the RTG injection also showed improvements in openfield locomotion 63 These results demonstrated how RTGs are capable of improving the survival and delivery of cellular treatments to SCIs. 2.1.6 Previous research with RTGs Our laboratory has recently developed RTGs for a variety of different applications to act as ti ssue engineering scaffolds. PSHU PNIPAm is a novel RTG composed of a poly ( serinol hexamethylene urea) backbone (PSHU) conjugated to a water solu ble RTG, PNIPAm. PSHU (by itself) is a biocompatible, hydrophobic, linear polymer with multiple urea molecule s in each re peating unit that allows the backbone to resemble a protein structure. Each repeating unit of the PSHU backbone has a reactive primary

PAGE 39

21 amine group that is used to conjugate to biomolecules or other polymers. In order to make PSHU a hydrophilic and injectable solution, it needs to be conjugated to PNIPAm. PNIPAm is a prevalent water soluble RTG used in many polymer applications due to its remarkably sharp lower critical solution temperature (LCST) of 32C. This LCST is advantageous for tiss ue engineering applications because it is well below the physiological temperature of 37C, allowing for fast g elation upon exposure to tissue temperatures The primary amine groups of PSHU can also be used to conjugate biomolecules, drugs, growth factors and/or other therap eutic agents to create a biomimetic polymer that will directly interact with and affect cellular mechanisms. PSHU co njugated to RGD (discussed next ) and PSHU PNIPAm conjugated to RGD demonstrated enhanced cell viability and differenti ation whil e investigating axon sprouting of PC12 cells 64 PSHU PNIPAm can be loaded into a syringe and injected by a small gauge needle into tissues or other physiological specimens, allowing for the formation of a stable gel within tissues that can serve as a biomimetic polymer scaffold. The scaffold will deg rade over time and release its contents in a controlled manner, allowing for sustained delivery of its therapeutic contents to tissues 65 2.1.7 RGD peptide A key component all polymer scaffolding desig ns require for viable interactions with host cells are cellular adhesive peptides. Cellular adhesive peptides are critical for cellular anchorage, migration, proliferation, differentiation, and apoptosis pathways 66 Without these adhesion molecules, polymer scaffolds will not mimic the ECM environment, which is necessary to allow the polymer to interact with cellular integrin receptors and influence signaling pathways 67 Most polymer scaffolds by themselves do

PAGE 40

22 not contain ligands capable of attaching to integrin receptors therefore, conjugation of adhesion p eptides to the polymer structure is a necessity. One of the most common peptides used to encourage cellular attachment for biomimetic scaffolds is arginine glycine aspartic acid (RGD). This specific peptide sequence is found in many ECM proteins includi ng fibronectin, laminin, and some collagens. This peptide sequence acts as ligand that binds to cell receptors, influencing cellular activities through signaling pathway activations 68 Integrin receptors are of part icular interest for RGD binding 66 These adhesion proteins are expressed on the surface s of cells and are resp onsible for interactions between the cell and the ECM. When RGD binds to the int egrin recognition site a series of signaling pathways become activated tha t influence cellular behaviors ( Figure 2 3 ). Figure 2 3 Schematic of RGD integrin binding to trigger multiple signaling pathways. Conjugation of RGD to the polymer backbone allows the polym er to mimic the ECM environment and bind to integrin proteins. In order to model properties of the ECM, polymer scaffolds can be designed to include the RGD sequence by using chemical reactions. Functional groups on the polymer backbone, such as amine, hydroxyl, or carboxyl groups, are able to be modified

PAGE 41

23 by chemical means to allow stable, covalent bonding of the RGD sequence to the polymer backbone 67 In vitro studies have successfully integrated RGD into the backbone of the polymer scaffolds to encourage cell spreading and survival 70 Our lab demonstrated that PSHU RGD and PSHU PNIPAm RGD were able to support proliferation and differentiation of PC12 cells, while the same polymers without RGD saw limited to no cell viability 64 PSHU RGD was also able to promote substantial differentiation of human NSCs into motor neurons ( Figure 2 4 ) 71 Figure 2 4 Polymer scaffold with conjugated RGD showed substantial differentiation of human NSCs into motor neurons. Inset is a schematic of Tubulin staining. Red represents motor neuron markers (Islet 1, H B9) 71 These results demonstrated the critical im portance of incorporating RGD into the polymer backbone to enhance neural survival, growth, and differentiation. Development an injectable PEG based RTG functionalized with RGD for SCI would introduce a minimally invasive therapeutic approach that provid es not only structural support to damaged tissue, but also the ability to mimic the ECM environment and directly influence cell ular activities

PAGE 42

24 2.2 Animal Models of SCI New tissue engineering approaches to treat SCI will require a nimal models that resemble the effects of SCI in humans. As this literature review has shown, r at models of SCI are the most commonly used. Rat models are inexpensive, allow for large sample numbers, and have good anatomical size for surgical manipulations 72 Murine SCI models offer the capability to alter genetics to study specific molecular pathways of SCI 73 However, murine models are less convenient because of the small anatomical size of the spinal cord, which make surgical techniques and therapy placement more difficult. Larger animal models, such as dogs, cats, pigs, sheep, and primates, are less common in SCI research because of higher expenses, animal welfare concerns, and limited understanding of when these models will be necessary for good translation to human subject s 74 Larger animal models have been used in the development of new SCI methods and the assessment of current SCI treatments. A canine model was used to develop a new compression injury using a novel ball oon technique, offering the advantage of an SCI model requiring no laminectomy 75 A canine model was also used to compare MP treatment to surgical decompression after SCI, results showing that surgical decompression surgery with or without MP improved neurological function when compared to MP treatment alone 76 further questioning therapeu tic effects of MP. Studies using new tissue engineering approaches are less common in large animal models, but have been attempted. The degradation of a PLGA scaffold seeded with NSCs was evaluated with four African green monkeys using a hemisection SCI model. Though the study had a small sample size, researchers were able to demonstrate successful integration of NSCs onto the PLGA scaffold for 82 days 77 Another study

PAGE 43

25 with a canine model used Matrigel loaded with a neurotrophic factor and MSCs that demonstrated neuronal extension, reduced glial scarring, reduced inflammation, and functional improvements 78 Despite the importance of establishing larger animal SCI models to further innovative therapeutic approaches, studies for tissue engineering will continue to use rodent models until a better understanding and reproducibility of results are obtained. 2.2.1 Rat models of SCI Different SCI models have been developed to simulate the primary and secondary pathophysiological effects of the injury. The three most common models used by researchers involve contusion, compression, or transection injury. Each of these three methods are able to target different severities of SCI and may also be suited for different therapeutic goals. For instance, implanting solid scaffolds into spinal tissue will require a transection model. If the goal of the research is to understand the pathophysiology of an injury and the response to a treatment, contusion or compressive injuries are used. Each of these three SCI models have benefits and drawbacks, and will be discussed below. 2.2.1.1 Contusion SCI animal models Falls and motor vehicle accidents are the most common causes of SCI. These events cause high impact forces to the spinal column, resulting in vertebral column fractures, spinal cord displacement, and contusion injury to the spinal tissue. The most common animal model of SCI used by researchers is the contusion model, as it may best reproduce the acute effects of high impact trauma experienced by patients 72 This method involves performing a laminectomy to expose the spinal cord tissue, followed by subjecting the spinal tissue to a high impact force, such as a dropped weight or a

PAGE 44

26 calibr ated impactor device 79,80 81 Contusion models offer precision and flexibility in the severity of contusion applied to simulate different degrees traumatic SCI seen in patients and generally have good reproducibility. However, these models are oft en result in incomplete damage of neuron cells and axons, which makes assessment of neural regeneration more difficult 72 2.2.1.2 Compression SCI models To better mimic the chronic stages of SCI, compression animal models have been developed to evaluate neuroprotective approaches and to gain a better understanding of the effects of long term SCI pathophysiology 82 This method requires laminectom y to expose spinal tissue, followed by compression of the spinal cord by either a calibrated clip (often an aneurysm clip) or forceps. The force applied by the compressing object and the length of compres sion time can easily be changed to vary the severity of the injury. This method is believed to be the most clinically relevant SCI model because of the similar chronic pat hological deficits and motor impairments 82 However, application of the compressive object can be difficult. This model requires the compression device to completely surround the lateral sides of the spinal cord. Removing enough lateral vertebrae by laminectomy to ex pose this much spinal cord requires more surgical skill than simple dorsal laminectomy used in contusion models 2.2.1.3 Transection SCI models Compression and contusion models often do not cause a complete severing of neural tissues after SCI, making it difficul t to distinguish between axons that are regenerating and axons that are still intact despite being injured. In order to better assess axon regeneration, a model of complete SCI severance was developed. The advantage

PAGE 45

27 transection SCI models have over contu sion and compression models is the ability to sever specific spinal tracts for injury without having to take into account the possibility of damaged but intact tracts 82 However, compl ete or partial severing of the spinal cord is rare in SCI patients, making this model the least clinically relevant 1 Transection models also do not reflect the pathophysiology of clinical SCI s that contusion and compression models ar e better able to mimic 83 2.2.2 Functional testing To assess the efficacy of an SCI treatment, function al tests have been developed in animal models that allow for both qualitative and quantitative results. Different animal SCI models result in varying degrees of motor loss that change over time. Therefore, choosing functional assessment will depend on th e severity of the injury. There is no consensus on which test provides the most meaningful recovery results and there is no standard functional test to use for a specific injury type. The most commonly used test in many SCI animal studies is the 21 point open field locomotion score (BBB rating). This test assesses the hindlimb movement and trunk stability of rats in a large open area. Monitored movements include hip, knee, and ankle movement s toe clearance, paw position, forelimb hindlimb coordination, trunk position, and tail position 84 Normal locomotion rece ives scores of 21, while rats after SCI receive a score of 0, as most SCI models result in several days of complete loss of hindlimb function. This test is generally believed to be sensitive and reproducible for mild and moderate SCI injury models, but is less reliable for more severe SCI models 85 In addition to open field testing, another useful analysis of gait coordination is the walking track analysis (also known as footprint analysis), which quantifies the placement of the hindlimb feet. This is a common method

PAGE 46

28 to evaluate PNI regeneration in sciatic nerve injury models and also has been used in SCI models 86 88 Functional t ests have also been developed that assess a rats balance capabilities while maintaining good gait coordination. Grid walk and latter walk tests monitor how well the rat can place an d grip their forelimbs and hindlimbs on rungs that are spaced apart. T he number of footfalls th rough the rung gaps are counted for a general assessment, while more detailed examinations involve hindlimb recovery and angle of paw placements 87,89 Many other functional tests exist, including kinematic analysis, narrow beam walking, incline walk, and cont act placing response. As the SCI model becomes more severe, the ability of the animal to perform more complicated motor tests becomes limited. The best approach to evaluate motor improvements of animals after SCI is to use a variety of functional tests w ith varying difficulty 87 2.2.3 Immunohistochemistry (IHC) After functional tests are completed, the spinal tissue is examined by IHC to study the distribution and the localization of specific marke rs of healthy and damaged neurons. Chemical and immunofluorescent staining techniques are often used together in SCI models and are good complements to functional testing. Histological assessment using Hematoxylin and Eosin (H&E), a chemical stain, shows robust structural information such as the location and size of the SCI site, as well as the presence of scar tissue formation. Luxol Fast Blue (LFB) is a more specific chemical stain that identifies myelin and allows for differentiation between damaged and undamaged axons. To identify more specific markers of nerve damage and regeneration, immunofluorescen ce methods are use d Many different neural markers have been developed that are capable of assessing different stages of SCI that have allowed for a better understanding of the short and long

PAGE 47

29 term pathophysiology. These markers have become invaluable tool s to evaluate the efficacy of new SCI therapies. The glial scar is believed to be one of the main barriers to axonal growth through the lesion site. To assess the formation and size of the SCI glial scar, an antibody called glial fibrillary acid pr otein (GFAP) is used to assess the localization of astrocytes 90 Targeting the glial scar for size reduction and axonal penetration through the injury site has made GFAP one of the most common fluorescent stains in SCI res earch 22,30,57 Another common stain used to st udy the inflammatory response after implementing a treatment for SCI is anti CD68 (ED1). As mentioned earlier, the inflammatory cascade after SCI contributes to significant neuron death and axonal damage. The ED1 stain is useful t o evaluate the effect a therapy has on acute and chronic stages of the SCI immune response 39 Specific neural components can also be stained with immunofluorescence to evaluate the presence of axons that a re both growing and functional. The antibodies to neurofilament and beta tu bulin are useful for detecting the presence of axons in SCI lesion areas to evaluate a therapies ability to penetrate the glial scar 30 Growth associated protein 43 (GAP 43) is an even more specific marker for axon regeneration. The expression of this protein is a major component of elongati ng axons that is an indication of favorable growth environments for injured CNS axons 91 This stain can pinpoint areas where damaged axons and regenerating axons are located, which makes it an attractive option to evaluate new SCI treatments 35

PAGE 48

30 3 Research objectives and experimental approach 3.1 Hypothesis As previous studies have sho wn, polymer scaffolds present a promising approach to deliver therapeutic agents to promote a regenerative environm ent in the injured spinal cord. W e hypothesize that a novel, PEG based injecta ble RTG functionalized with RGD will promote neural support and axon regeneration in a compressive SCI rat model. 3.2 Specific a ims (1) Synthesize and characterize a novel PEG polymer backbone The first step for this work is to create a novel PEG based polym er to serve as the backbone for further chemical conjugations. As previously mentioned, PEG can provide protective support to damaged neural tissues and has biocompatible and biodegradable properties in multiple in vitro and in vivo studies. Many develop ed polymer backbones have few chemical functional groups to promote conjugation to biomolecules, thus, limiting the polymers capacity to act as an ECM scaffold to support cellular activity. This novel PEG polymer backbone will have two functional hydroxyl groups per repeating unit to provide greater biomolecule conjugation capability. To characterize the structure, Fourier transform infrared spectroscopy (FT IR) will be used to verify the appearance of distinct hydroxyl and carbonyl groups in the polymer backbone. (2) Conjugate PNIPAm to PEG backbone to create injectable RTG After verifying the structure of the PEG backbone, PNIPAm will be chemically conjugated to the free hydroxyl groups. PNIPAm conjugation confers temperature sensitivity to the new pol ymer structure that allows it to exist as an aqueous solution at

PAGE 49

31 room temperature and form a physical gel when exposed to physiological temperature. PNIPAm provides a sharp transition at 37C to promote the fast transition from solution to gel, which prov ides a practical injectable property to the polymer. The gelled polymer forms the scaffold that will support cellular activity. This novel RTG structure will be characterized by FTIR and the gelling properties will be verified using ultraviolet visible lig ht (UV VIS) spectroscopy and scanning electron microscopy (SEM) (3) Conjugate RGD to PEG backbone to synthesize biomimetic RTG In order for this new RTG to act as a biomimetic ECM scaffold, the polymer backbone needs to have cellular adhesion properties. As discussed earlier, RGD is a common laminin peptide used to confer cellular attachment to synthetic scaffolds. RGD will be c onjugated to the hydroxyl groups of the PEG backbone to create an injectable, biomimetic RTG capable of promoting cellular activity. The large number of free hydroxyl groups available on the PEG backbone allows for greater RGD conjugation and therefore, m ore influence on cellular activity Conjugation of RGD to the PEG backbone will be verified using FTIR. (4) In vitro testing of biomimetic RTG for cytotoxicity and CNS axonal support After synthesis of the novel biomimetic RTG, the cytotoxicity will be tested with Alamar Blue and MTT assays to verify biocompatibility of the polymer. Retinal ganglion cells (RGCs) will then be seeded within the polymer solution and gelled to CNS axon growth and support. Immunohistochemistry will verify axon extensions throughout the polymer network.

PAGE 50

3 2 (5 ) Inject biomimetic RTG into compressed rat spinal cords to evaluate axon regenerative capability Finally, to assess this RTGs ability to act as a biomimetic scaffold for SCI, this RTG will be injected into compressed rat spinal cords. As previously discussed, PEG and RTGs have promising potential to form supportive cellular scaffolds for injured neural tissues. There are two essential proper ties of this novel PEG based RTG that make it an attractive therapy for SCI: (1) its ability to be injected into damaged tissues without invasive implantation procedures, (2) to deliver RGD adhesive peptides to promote cellular attachment and survival of d amaged spinal neurons. To study the effect this polymer has on damaged spinal cord tissue, a compression SCI model in rats will be utilized to provide a clinically relevant models to study the chronic effects of SCI. We believe the biomimetic properties of this injectable RTG will promote a supportive ECM environment to induce axonal regeneration and function recovery after SCI.

PAGE 51

33 4 Materials and e quipment 4.1 Materials Poly(ethylene glycol) diglycidyl ether (PEGDGE, Mn 526), triphenylphosphine azo bis (4 cyanovaleric acid) (ACA), paraformaldehyde (PFA), sucrose, nerve growth factor (NGF), dimethyl sulfoxide (DMSO), bovine serum albumin (BSA), and heparin were purchased from Sigma Aldrich (St. Louis, MO, USA). Succinic acid, tetrahydrofuran (THF), anhydrous methanol, 4 Dimethylaminopyridine (DMAP), N Dicyclohexylcarbodiimide (DCC), N (3 Dimethylamino propyl) N' ethylcarbodiimide hydrochlori de (EDC), chloroform d, and N hydroxysuccinimide (NHS) were purchased from Alfa Aesar (Ward Hill, MA, USA). Anhydrous N, N Dimethylformamide (DMF) was purchased from EMD Millipore (Billerica, MA, USA). Dichloromethane (DCM), Acetone, and 30% hydrogen per oxide were purchased from BDH Chemicals through VWR (Radnor, PA, USA). N Isopropylacrylamide (NIPAAm) was purchased from TCI Chemicals (Portland, OR, USA). Anhydrous diethyl ether was purchased from Fisher Scientific (Pittsburgh, PA, USA). Gly Arg Gly Asp Ser ( RGD ) was purchased from Biomatik (Wilmington, DE, USA) and Cellmano Biotech Limited (Hefei, AnHui, China). Sprague Dawley rats were purchased from Charles River Labo ratories (Wilmington, MA, USA). Sterile saline, isoflurane, ketoprofen, ketamine buprenorphine, gentamicin, and bupivacaine (0.5 % Marcaine) were purchased from MWI Veterinary Supply (Boise, ID, USA). Vicryl 4 0 sutures were purchased from Ethicon (Somerville, NJ, USA). Optimal cutting temperature (OCT) compound was purchased from S akura (Torrance, CA, USA). MTT Cell Proliferation Assay Kit was purchased from Invitrogen Molecular Probes (Carlsbad, CA, USA). Alamar blue cell viability reagent, Phosphate Buffered Saline

PAGE 52

34 (PBS) and RPMI 1640 Medium was purchased from Thermo Scientific. Laminin coated cell culture dishes and Laminin solution (1.75 mg/mL) were purchased from Discovery Labware, Inc (Bedford, MA, USA). Goat serum, GFAP (mouse IgG1), GAP 43 (Rabbit IgG), Alexa Fluor 488 (goat anti mouse IgG), Alexa Fluor 594 (goat anti rabbit IgG), and SlowFade Diamond antifade mountant with DAPI were purchased from Life Technologies (Carlsbad, CA, USA). tubulin (g oat anti rabbit, IgG) was purchased from abcam (Cambridge, MA, USA). Brn 3a (goat anti mouse, IgG) was purchased from Santa Cruz Biotechnology (Dallas, TX, USA). Fluoromount G with DAPI was purchased from Electron Microscopy Sciences (Hatfield, PA, USA). Triton X 100 was purchased from MP Biomedicals. Retinal Ganglion Cell Isolation Kits (rat) were obtained from MACS Miltenyl Biotec (San Diego, CA, USA). 4.2 Equipment Fourier transform infrared spectroscopy (FT IR) was performed on a Nicolet 6700 FT IR s pect rometer S amples were run on polyethylene infrared (IR) sample cards. Polymer purification was accomplished using dialysis tubing (Spectrum Labs, Rancho Dominguez, CA, USA). Polymer mo rphology was imaged using a JEOL JSAM 6010la analytical scanning elect ron microscope (SEM) (Peabody, MA, USA). MTT assay and Alamar Blue assays were performed using absorbance values on a BioTek microplate reader ( Winooski, VT, USA). Glass bottom culture dishes for RGC growth were purchased from MatTek Corporation (Ashland MA, USA). Compression aneurysm clips (30 g) and surgical tools were purchased from Kent Scientific (Torrington, CT, USA). Treatment injections into spinal tissue were performed using a 10 l Hamilton syringe (Reno, Nevada, USA). A 32 gauge needle was used for the injection (TSK Laboratory

PAGE 53

35 ( Tochigi shi, Tochigi ken, Japan). Tissue was sectioned using a CryoStar NX70 Cryostat. Confocal images were taken with Nikon microscope.

PAGE 54

36 5 Methods 5.1 Synthesis of PEGSA PEGDGE (5 g, 9.51 mmol) was mixed with succinic acid ( SA) (1.35 g, 11.432 mmol) for a 1: 1.2 molar ratio of PEGDGE: SA. TPP (0.25 g, 0.953 mmol) was added to the PEGDGE, SA mixture in a 10% catalytic amount to total mol of PEGDGE. This catalyst acts to abstract the protons on the carboxylic acids on succ inic acid. No solvent was used in this reaction. The mixture was placed on to an oil bath set at 120C with stirring set at 150 rpm. This temperature was ne cessary for the carboxylic acid groups on the succinic acid to open the e poxide rings on the PEGD GE. All reactants dissolved at this temperature after several minutes. This mixture was reacted for 48 hours, with the stir speed being decreased to 60 80 rpm after 24 hours. The mixture becomes viscous after 24 hours due to polymerization and the stir speed needs to be decreased. Otherwise, the stir bar may become stuck, halting mixing of reactants which will lead to crosslinking of polymer chains. After 48 hours, the orange brown viscous fluid is dissolved in DCM to make a pip pet able solution, and the n precipitated three times in ether. Each ether precipitation is followed by rotary evaporation. The product is then dissolved in a minimal amount of Milli Q water and lyophilized for 48 hours. For optimal purification, after precipitation, the fluid is dissolved in water and added to 3500 molecular weight (MW) cutoff dialysis tubing and dialyzed against 1 liter of water for two days, with water changes every 24 hours. The product is removed from dialysis tubing and lyophilized for 48 hours, resulting in a final product that is a dark orange, viscous fluid.

PAGE 55

37 5.2 Synthesis of carboxylic acid terminated PNIPAm NIPAAm (5 g, 44 mmol) and ACA (0.062 g, 0.2 mmol) were dissolved i n 25 mL of anhydrous methanol and then purged with nitrogen gas for 30 minutes. After pu rging, t he polymerization reaction was carried out at 68C for 3 hours with stirring. The reaction mixture was precipitated dropwise into warm water (60C) and washed with fresh, warm Milli Q water. The polymer was dissolved at 4C overnight in Milli Q w ater and then purified in dialysis tubing (MWCO: 12,000 14,000 Da) against 1 liter of Milli Q water for 3 days at room temperature, with water changes every 24 hours. The purified PNIPAm COOH was lyophilized for 2 days and stored at room temperature. 5.3 Sy nthesis of PEGSA PNIPAm Gel permeation chromatography results revealed that the PNIPAm procedure we used was making much larger MW PNIPAm than originally anticipated. Large MW PNIPAm allowed us to tweak the conjugation procedure in later re ac tions. Therefore, two approaches were to conjugate PEGSA to PNIPAm 5.3.1 Molar Reaction The original reaction used a 25% conjugation of PNIPAm to PEGSA. Since PEGSA has 2 hydroxyl groups, 25% of these hydroxyl groups were calculated to be conjugated to PNIPA m The first step of the reaction was to activate PNIPAm PNIPAm (4.045 g, 0.4045 mmol), DCC (0.1 g, 0.485 mmol), and DMAP (0.01 g, 0.082 mmol) were dissolved in anhydrous THF. DCC was added in 1.2 mola r excess to PNIPAm and DMAP was added in a 0.2 molar catalytic amount. This mixture was reacted with stirring at 55C for 24 hours under nitrogen atmosphere. After 24 hours, PEGSA (0.5 g, 0.81 mmol) was dissolved in anhydrous THF and stirred at 55C. The activated PNIPAm

PAGE 56

38 was added dropwise to th e PEGSA solution and reacted 24 hours with nitrogen atmosphere at 55C. After 24 hours, the reaction was added dropwise in to excess ether for precipitation. Solvent was removed by rotary evaporation. Milli Q water was added to the white product an d disso lved at 4C. To purify, the product was placed in dialysis tubing (MWCO: 12,000 14,000) and dialyzed against 1 liter of MilliQ water for 48 hours, with water changes every 24 hours. After dialysis, the solution in the dialysis tubing is filtered through a 0.2 micron filter to remove insoluble reaction byproducts and to sterilize the product. After filtering, the product is lyophilized, resulting in a thick, flaky white precipitate. 5.3.2 Gram base conjugation To decrease the amount of PNIPAm used and increase R GD conjugation to the PEGSA backbone, gram: gram reactions (PEGSA : PNIPAm ) were attempted. PNIPAm (0.5 g, 0.05 mmol) was activated with DCC (0.0124 g, 0.06 mmol) and DMAP (0.0012 g, 0.01 mmol) by dissolving in anhydrous THF and reacted for 24 hours at 55 C with stirring. After 24 hours, PEGSA (0.5 g, 0.81 mmol) was dissolved in anhydrous THF and heated to 55C with stirring. The activated PNIPAm was then added dropwise to the PEGSA solution and reacted 24 hours. Precipitation and purification of the prod uct was carried out identically as described above for the mole base reaction. 5.4 RGD conjugation 5.4.1 RGD conjugation to PEGSA

PAGE 57

39 To begin the reaction between the PEGSA hydroxyl groups with RGD amine groups, RGD (0.025 g, 0.051 mmol) EDC (0.016 g, 0.103 mmol), NHS (0.012 g, 0.104 mmol), and were dissolved in Milli Q water. This solution was reacted at room temperature for 2 hours, protected from light. PEGSA (0.021 g, 0.034 mmol), was dissolved in Milli Q water and added dropwise to the activated RGD solution and reacted for 24 hours at room temperature, protected from light. This reaction was a 100% conjugation of RGD to each of the 2 hydroxyl groups per repeating unit on PEGSA. After 24 hours, the reaction was placed in dialysis tubing (MWCO: 3500) and dialyze d against 1 liter of Milli Q water for 24 hours, with one water change. After dialysis, the solution was freeze dried, resulting in a brownish, elastic precipitate. 5.4.2 RGD conjugation to PEGSA PNIPAm (mol base) To begin the reaction between the remaining fre e PEGSA hydroxyl groups (on PEGSA PNIPAm mol base reaction) with RGD amine groups, RGD (0.025 g, 0.051 mmol), EDC (0.016 g, 0.103 mmol), and NHS (0.012 g, 0.104 mmol) were dissolved in Milli Q water. This solution was reacted at room temperature for 2 ho urs. PEGSA PNIPAm (0.267 g, 0.432 mmol) was dissolved in Milli Q water and added dropwise to the activated RGD solution and reacted for 24 hours at room temperature, protected from light. After 24 hours, the reaction was placed in dialysis tubing (MWCO: 3500) and dialyzed against 1 liter of Milli Q water for 24 hours, with one water change. After dialysis, the solution was freeze dried, resulting in a white, flaky precipitate. 5.4.3 RGD conjugation to PEGSA PNIPAm (gram base) To begin the reaction between the remaining free PEGSA hydroxyl groups (on PEGSA PNIPAm gram base reaction) with RGD amine groups, RGD (0.08 g, 0.1631

PAGE 58

40 mmol), EDC (0.0753 g, 0.485 mmol), and NHS (0.056 g, 0.487 mmol) were dissolved in 1 mL of Milli Q water. This reaction was activated at room temperature for 2 hours. PEGSA PNIPAm (0.1 g, 0.01 mmol) was dissolved in 2 mL of Milli Q water and added dropwise to the activated RGD solution and reacted for 24 hours at room temperature, protected from light. After 24 hours, the reaction was pla ced in dialysis tubing (MWCO: 3500) and dialyzed against 1 liter of Milli Q water for 24 hours, with one water change. After dialysis, the solution was freeze dried, resulting in a white, flaky precipitate. 5.5 Polymer biocompatibility testing Rat pheochromoc ytoma cell line (PC12) was selected for initial in vitro studies to assess the polymers biocompatibility. PC12 cells are not a true neural cell, as they are derived from adrenal chromaffin cells. Howeve r, when exposed to nerve growth factor (NGF), PC12 cel ls differentiate to resemble sympathetic neurons with similar morphology and functionality 92 The use of this cell line has been used for bio material optimization and for cytotoxicity testing 93,94 Retinal ganglion cells (RGCs) were used to test neural attachment and growth to PEGSA PNIPAm RGD to assess the polymers ability to support CNS neurons. Retinas from day 5 rat pups were dissected and isolated according Biotec. Briefly, after dissection, retinal tissue is dissociated to single cell suspensions through a series of enzymatic and mechanical digesti ons of ECM adhesion proteins. RGCs are then labeled with magnetic beads and undergo a series of magnetic separa tions for compete RGC isolation.

PAGE 59

41 5.5.1 PC12 c ell culture PC12 cells (1.0 x 10 6 cells, ATCC CRL 1721 ) were cultured with RPMI 1640 M edium (2.05 mM L glutamine) supplemented with 10% horse serum (HS), 5% fetal bovine serum (FBS), and 1% penicillin streptomycin (will be referred to as proliferation media). To induce differentiation, PC12 cells were cultured in RPMI 1640 Medium (2.05mM L glutamine) sup plemented with 1% HS, 0.5% FBS, and 100 ng/mL NGF. Cells were cultured on 60mm laminin coated dishes to facilitate attachment. 5.5.2 Retinal ganglion cell (RGC) cell culture Retinal ganglion cells (RGCs) were used to test neural attachment and growth to PEGSA P NIPAm RGD to assess the polymers ability to support CNS neurons. Retinas Biotec. Brie fly, after dissection, retinal tissue is dissociated to single cell suspensions through a series of enzymatic and mechanical digestions of ECM adhesion proteins. RGCs are then labeled with magnetic beads and undergo a series of magnetic separations for co mpete RGC isolation. 5.5.3 MTT c ytotoxicity The cytotoxicity of PEGSA PNIPAm was tested with proliferating PC12 cells with ISO/EN 10993 5 guidelines. PEGSA PNIPAm gels were prepared in 10%, 5%, 2.5%, 1.25%, and 0.6 25% solutions (mg/mL ). To a 96 well plate, 50 L of each solution was added and then gelled in an incubator for 15 minutes. To the top of each gel, 200 L of warm PC12 proliferating media was added and allowed to incubate overnight. To a separate 96 well plate, laminin solution was added to for cellular attachment (1 g/cm 2 ).

PAGE 60

42 PC12 cells were seeded onto laminin coated 96 well plates at a density of 1 x 10 4 cells/well and incubated for 24 hours with proliferating medium. After 24 hours, medium was removed from the PC12 cells and rep laced with 200 L of the PEGSA PNIPAm media extract. Cells were allowed to culture for an additional 72 hours. MTT assays values read at 540 nm. Experimental samples we re normalized to control samples (proliferating media + PC12 ce lls). 5.5.4 Alamar blue cell viability assay The cell viability of PEGSA PNIPAm was tested with proliferating PC12 cells with ISO/ EN 10993 5 guidelines. Polymer extract solutions and cells were pr epared identically to MTT cytotoxicity testing. After PC12 cells cultured in PEGS A PNIPAm extract media for 72 hours, Alamar Blue cell viab ility assay was performed according to instructions, with absorbance read at 570 nm. Experimental sa mples were normalized to control samples (proliferating media + PC12 cells). 5.5.5 RGC attachment and growth T o study RGC su rvival and axon extension in a 3D polymer matrix i solated RGCs were mixed with PEGSA PNIPAm RGD solutions and gelled. PEGSA PNIPAm RGD was dissolved in RGC media to create a 15% polymer solution. After isolation, RGCs were mixed with PEGSA PNIPAm RGD solutions to create a final polymer concentration of 10%, with a cell density of 1 x 10 4 cells per 50 L of polymer media solution. To a 35 mm glass bottom culture dish, 50 L of polymer cell solution was added. The culture dish was placed in an incubator for 10 minutes to allow polymer gelation and RGC encapsulation within the gel After incubation, 1 mL of warm, RGC

PAGE 61

43 media was added to the culture dish using a hotplate set at 37C to maintain gel stability when re moved from the incubator. Cells were cultured for 3 days, with media changes each day. Samples were then fixed in 4% PFA for 15 minutes and washed with PBS over night. Cells were permeablized with 1% Triton X (in PBS) for 90 minutes, followed by PBS wash overnight. A blocking buffer composed of 2% bovine serum albumin (BSA) in PBS was added to the cells for 90 minutes. After the blocking step, cells were incuba ted overnight with primary antibody Brn 3a (anti mouse, 1:200 dilution). Cells were then washed with 1% Trit on X, 3x for 3 minutes each. The secondary antibody, anti rabbit Alexa 594 (1:500), was added to each sample and incubated for 45 minutes. Cells were washed with PBS Tween (0.002% in PBS) for 3 minutes and washed t wice with PBS, 3 minutes each. The next tubulin ( anti rabbit, 1:100 dilution prepared in bl ocking buffer) was added and incubated overnight Cells were then washed with 1% Triton X, 3x for 3 minutes each. The secondary antibody, anti rabbit Alexa 488 (1:500), was added to each sample and incubated for 45 minutes. Cells were washed with PBS Tw een (0.002% in PBS) for 3 minutes and washed twice with PBS, 3 minutes each. Hoechst 33342 (1:2000 in PBS ), a DAPI stain, was added to each sample and incubated for 5 10 minutes, followed by 3 washes in PBS, 3 minutes each. Samples were incubated in P BS until imaged with confocal microscope. 5.6 Compression SCI rat model 5.6.1 Ethics and surgical approval All animal experiments were performed under a protocol approved by the Institutional Animal Care and Use Committee (IACUC) at the University of Colorado Anschu tz Medical Campus.

PAGE 62

44 5.6.2 Surgical procedures Female adult S pr ague Dawley rats (225 300g, n=15 ) were allowed to acclimate for 1 week prior to surgical procedures. Rats were maintained on a 14/10 hour light/dark cycle with a continuous supply of fresh air and access to food and water ad libitum. A compression model was chosen for t his study to test the polymers effects on a chronic SCI model Rats were anesthetized with 5% isoflurane in oxygen and maintained on 0.5 1% isoflurane in oxygen for the remainder o f the surgery. To maintain body temperature, rats were placed on a warm recirculating water blanket for the duration of the procedure Artificial tears ointment was applied to the eyes to prevent corneal abrasion and drying. After removing hair with electric clippers, the surgical site was administered with bupivacaine (0.5% Marcaine) at 2mg/kg by dermal block injections to decrease i ncision pain after surgery. Buprenorphine (0.2mg/kg) was administered subcutaneously (SQ) to minimize post operative pain. The incision site was sterilized with 70% ethanol and then betadine and this process was repeated 3 times. Rats were placed in a prone position and a midline sk in incision was made around the T7 T13 vertebrae (T11, T12 vertebrae are more pronounced and can be palpated by hand). Paravertebral muscles were dissected to expose vertebral spinous processes and lamina. Using a micro bon e rongeurs, a double laminectomy at the T9 T11 level was perfo rmed to expose the spinal cord. To simulate a chro nic compression SCI model, a 30 g aneurysm clip was selected to act as the compression device. To properly apply the compression clip, the latera l

PAGE 63

45 aspects of the lamina and vertebrae around the cord needed to be removed as much as possible. The compression clip was applied to the spinal cord for 1 minute and then removed. Spinal cord injection treatments were assigned randomly. As this was a smal l pilot study, only two treatment groups were used. A saline injection was used as the treatment control and PEGSA PNIPAm RGD was used as the experimental treatment. The polymer was dissolved in sterile saline and filtered (0.2 microns) before injection. After compression, 10 L of saline or polymer was injected into the injury site using a Hamilton syringe with a 32 gauge, 4 mm needle. When injected, the needle was held in the injection site for 30 seconds to ensure complete delivery and then removed slowly. The muscle layers were sutured together with 4 0 absorbable suture with an interrupted pattern. The skin layer was sutured using a subcuticular method with 4 0 absorbable suture and the incision site was wiped with betadine. Rats recovered from anesthesia on a warm water blanket and were single housed in a new, clean cage. Single housing for several days post surgery was necessary to prevent other rats from biting the sutured incisions. When administration of analgesics stopped, rats were place d back with their original cage mates. 5.6.3 Post surgery procedure Buprenorphine (0.2mg/kg) was administered 12 hours after surgery for post operative pain. K etoprofen (5mg/kg) was injected SQ 24 hours after surgery to continue pain alleviation and this conti nued once every 24 hours for 72 hours. Gentamicin (1mg/mL) was added to the dri nking water to prevent urinary tract infections and remained in the water for the duration of the study. Food was crushed into pieces, soaked

PAGE 64

46 in water, and placed in the cage for the first week after surgery. Manual bladder expression was performed at least twice per day until the rats reestablished this reflex. 5.7 Functional Assessment The BBB locomotor score was used to assess open field locomotion. Before surgery, rats were assessed and given the highest BBB score of 21 to act as baseline score. After surgery, rats were scored once a week. 5.8 Histology Four weeks aft er surgery, rats were given an IP overdose of ketamine to induce deep anesthesia and were transcardially perfuse d with heparinized saline (50 mL, 10 Units Heparin/mL) followed by 4% PFA (200 mL) After perfusion, a 1 cm segment of the spinal cord was harvested (with the lesion site centered) and placed in fresh 4% PFA overnight. The spinal cord segments were then immersed in a 30% sucrose solution for 48 hours or until the tissue sank for cryoprotection The sucrose solution was removed and the spinal tissue was embedded in OCT compound and then frozen at 80C. The OCT blocks were cut on a cryostat into 10 m lo ngitudinal sections and placed onto glass slides. Glass slides were stored at 20C. 5.8.1 H&E staining 5.8.2 Immunohistochemistry Glass slides with spinal tissue sections were fixed in acetone for 10 min and washed 3 times in PBS for 3 min each. The sections were then blocked in 3 % hydrogen peroxide (in PBS) for 10 min to block endogenous peroxidase activity and washed 3 times in PBS for 3 min each. Blocking buffer (5% goat serum, 0.4% Triton X 100, PBS)

PAGE 65

47 was used to block non specific binding sites of the sections for 30 min. Antibodies were prepared by diluting in the blocking buffer. Sections were then stained with the primary antibody GFAP (1:1 00) or GAP 43 (1:300) for 60 min (room temperature) and washed 3 times in PBS for 3 min each. GFAP stained slides recei ved Alexa Fluor 488 (1:500) secondary antibody for 30 min and GAP 43 stained slides received A lexa Flu or 594 (1:500) secondary antibody for 30 minutes. After 30 minutes, slides were washed 3 times in PBS for 3 min each. Fluoromount G with DAPI mounting m edium was used to coverslip the slides. Slides were stored at 4C. 5.9 Statistical analysis All results are expressed as means standard error of the mean. Analysis of variance (ANOVA) was used to determine significant differences between groups. Statistic al significance was considered when p < 0.05.

PAGE 66

48 6 Results and Discussion 6.1 Polymer synthesis PEGSA was successfully synthesized using a novel reaction between an epoxide based PEG monomer and succinic acid. The PEGSA backbone contains 2 hydroxyl groups per repeating unit that are used to conjugate PNIPAm and RGD to the backbone ( Figure 6 1 ) Figure 6 1 Reaction of PEGDGE with SA to produce PEGSA. The PEGSA repeating unit contains two hydroxyl groups that will be used for further chemical conjugation. Following PEGSA synthesis, PNIPAm was conjugated to the PEGSA backbone PNIPAm conjugation allows PEGSA to attain RTG properties for injectability and fast gelation upon exposure to physiological temperatures ( Figure 6 2 )

PAGE 67

49 Figure 6 2 Conjugation of PNIPAm to PEGSA backbone. To allow PEGSA PNIPAm to mimic the ECM and promote cellular adhesion to integrin receptors RGD is conjugated to the remaining hydroxyl groups to promote biomimetic properties ( Figure 6 3 )

PAGE 68

50 Figure 6 3 PEGSA PNIPAm RGD reaction sequence GRGDS is a modified RGD peptide. 6.2 Polymer characterization 6.2.1 PEGSA FT IR characterization The main backbone polymer was successfully synthesized using a reaction between a PEG based monomer with epoxide groups (PEGDGE) and succinic aci d (an intermediate in the Citric Acid Cycle). FT IR spectra verified the successful reaction between PEGDGE and SA. As shown in Figure 6 4 region a the PEGSA product has a distinct peak that corresponds to the hydroxyl functional groups. Region b also shows a peak corresponding to the formation of carbonyl groups from the suc cinic a cid ester groups Region c is associated with ether groups, which are present in both the PEGDGE monomer and maintained in the PEGSA product.

PAGE 69

51 Figure 6 4 FT IR spectrum of PEGDGE and PEGSA 6.2.2 PEGSA PNIPAm characterization After verifying the PEGSA backbone structure, PNIPAm was conjugated to a small percentage of the hydroxyl functional groups. The conjugation reaction between PEGSA and PNIPAm was verified with FT IR spectra. Figure 6 5 provides a comparison between the PEGSA, PNIPAm, and th e final product PEGSA PNIPAm. Region a shows the distinct PEGSA carbonyl peak that is preserved in the PEGSA PNIPAm product. This carbonyl peak in PEGSA PNIPAm is also accompanied by distinct PNIPAm peaks. Figure 6 5 FT IR spectrum of PEGSA, PNIPAm, PEGSA PNIPAm a b c a

PAGE 70

52 6.2.3 PEGSA PNIPAm gelling properties T o create an injectable polymer scaffold, the polymer needs to be soluble in aqueous solutions. Conjugating PNIPAm to the PEGSA backbone allows the polymer to remain soluble in aqueous solutions at room temperature while forming a physical gel when exposed to physiological temperatures. To test the LCST of PEGSA PNIPAm, a UV VIS spectrometer was used to measure the % transmittance of the polymer when subjected to changes in temperature. The sharp LCST of PEGSA PNIPAm begins around 35C, which is just belo w physiological temperature of 37C, as shown in Figure 6 6 Increasing the amount of P NIPAm causes the gel to become more opaque, which is why the % transmittance decreases for increasing PNIPAm conjugations. This LCST test verifies the RTG propert ies of PEGSA PNIPAm. Figure 6 6 LCST of PEGSA PNIPAm conjugation reactions and PNIPAm. 6.2.4 PEGSA PNIPAm morphological characterization 0 20 40 60 80 100 120 26 28 29 30 31 33 34 35 36 38 39 40 42 43 44 45 %Transmittance Temperature C PEGSA-PNIPAm_5% conjugation PEGSA-PNIPAm_25% conjugation PNIPAm Each polymer was prepared in a 1% solution in water

PAGE 71

53 A key feature for polymer scaffolding systems is a porous structure to mimi c the ECM environment for nutri ent exchange, gas ex change, and cellular growth. A hig hly porous structure is particularly impor t ant for neuron s to promote axon extension s A scannin g electron micrograph verified the porous structure of a freeze dried sample of PEGSA PNIPAm ( Figure 6 7 ) This porous structure was homogenous throughout the entire sample. Figure 6 8 sho ws a higher ma gnificat ion of the same sample that shows the average pore size to be 5m 20m. Increasing the PNIPAm concentration or increasing the polymer solution concentration will result in smaller pore sizes, and vice versa. Figure 6 7 SEM micrograph (180X ) showing porosity of PEGSA PNIPAm with consistent porous size throughout the structure. Scale bar = 100 m

PAGE 72

54 Figure 6 8 SEM micrograph (950X ) showing an average pore size of 5m 20m 6.2.5 PEGSA RGD FTIR characterization PNIPAm is a large polymer that may present steric hindrance to further conjugation reactions between PEGSA hydroxyl groups and other molecules. Before conjugating RGD to PEGSA PNIPAm the reaction between PEGSA hydroxyl groups and RGD was attempted RGD was activated with EDC/NHS chemistry, and PEGSA was added slowly to allow the hydroxyl groups to act as nucleophiles to the activated RGD. Normally, amine groups are used as nucleoph iles w ith EDC/NHS activated chemicals, so this activation reaction with hydroxyl groups needed to be verified. The conjugation of RGD to the PEGSA backbone was verified with FT IR ( Figure 6 9 ) Region a shows PEGSA with RGD carbonyl and secondary amine peaks.

PAGE 73

55 Figure 6 9 FT IR spectrum of PEGSA, RGD, and PEGSA RGD 6.2.6 PEGSA PNIPAm RGD FTIR characterization When the reaction between PEGSA hydroxyl groups with RGD was confirmed, RGD was conjugated to PEGSA PNIPAm. FT IR spectra could not definitively confirm the conjugation reaction ( Figure 6 10 ) This is likely because of the PNIPAm carbonyl peaks and amine peaks (region a ) that dominate the si gnal and block out the RGD carbonyl and amine peaks ( Figure 6 11 ). However, this does not me an the conjugation did not occu r. RGD is a small molecule and is likely able to overcome any steric hindrance effects from the large PNIPAm polymer. Confirming the re action between PEGSA and RGD ( Figure 6 9 ) demonstrates the PEGSA hydroxyl groups can act as effective nucleophiles t o activated RGD. Therefore, we believe the conjugation o f RGD to PEGSA PNIPAm was successful, but it is not definitive on FT IR spectra because of signal dominance by PNIPAm peaks. a

PAGE 74

56 Figure 6 10 FT IR spectrum of PEGSA PNIPAm RGD The conjugation of RGD to PEGSA PNIPAm could not be confirmed because the PNIPAm peaks overshadow the RGD peaks. Figure 6 11 FT IR spectrum of region a (in Figure 6 10 ) of PEGSA PNIPAm, RGD and PEGSA PNIPAm RGD. The PEGSA PNIPAm peaks are located in the same chemical shift location as the RGD peaks. The RGD peaks have a weaker signal than the large PNIPAm peaks. 6.3 In vitro assessment 6.3.1 MTT cytotoxicity The MTT assay measures mitocho ndrial reduction potential in cel ls to as sess metabolic activity These enzymes reduce the MTT dye to formazan, which has a purple color. Confluent PC12 cells were exposed to diluted media extracts from PEGSA PNIPAm gels for 72 hours, and then tested with the MTT dye No statistical si gnificance a

PAGE 75

57 between the experimental groups and the positive control group was found Figure 6 12 ). Therefore, PC12 cell metabolic activity was not affected by exposure to the diluted polymer extract solutions These results confirmed PEGSA PNIPAm was biocompatible an d non toxic to PC12 cells. Figure 6 12 PC12 cells showed no cytotoxicity when exposed to PEGSA PNIPAm media extracts, per ISO 10993 5 Statistical analysis by ANOVA demonstrated no difference between the positive control (M edia+cells) sample and experimental samples (p=0.36, n=5). Experimental samples are normalized to cells exposed to pure media. Media only is the negative control. Error bars represent standard error of the mean. 6.3.2 Alamar b lue cell viability

PAGE 76

58 The Alamar Blue assay measures similar reduction potential of cells to a ssess cellular viability. The A lamar Blue dye contains Resazurin, which is non toxic and permeable to cells. When exposed to reduction enzymes within t he cytosol, viable cells red uce r esazurin to re sorufin, which fluoresces red. With a similar protocol to the MTT assay, PC12 cells were cultured in polymer extract solutions for 72 hour s, followed by addition of the A lamar B lue dye. No statistical difference was found between the positive control sam ple and samples exposed to PEGSA PNIPAm media extracts. These results further confirmed the biocompatibility of PEGSA PNIPAm and its ability to support viable cells ( Figure 6 13 ) Figure 6 13 PC12 cells showed no change in viability when exposed to PEGSA PNIPAm media extracts, per ISO 10993 5. Statistical analysis by ANOVA demonstrated no difference between the positive control (Media+cells) sample s and experimental samples (p=0.13, n=5). E xperimental samples are normalized to cells exposed to pure media. Media only is the negative control. Error bars represent standard error of the mean. 6.3.3 RGC attachment and growth After 3 days in culture, PEGSA PNIPAm RGD was able to support the survival and growth of primary RGCs. Axon extensions within the 3D polymer

PAGE 77

59 matrix provided evidence of the polymers capability to successfully attac h and support neurite outgrowth ( Figure 6 14 ) Figure 6 14 RGC axon extensions within PEGSA PNIPAm RGD 3D polymer matrix tubulin (green), nuclei stained with DAPI (blue), RGCs stained with Brn3a (red) Maximum intensity c onfocal microscope z stack image scale bar = 100 m 6.4 Compression SCI rat model The laminectomy surgery was able to successfully expose spinal cord tissue. However, proper placement of the compr ession clip required more lateral vertebrae to be removed, which was difficult. If too little lamina was removed, the clip often would not fit around the cord for good compression. The gene ral surgical procedure is demonstrated in Figure 6 15 focusing on the isolation of thoracic vertebrae, laminectomy to expose the spinal cord, and clip compression. Following compre ssion SCI, each rat received a 10 l injection of either saline or a 2.5% solution of PEGSA PNIPAm RGD. The polymer solution was dissolved in sterile

PAGE 78

60 saline and filtered (0.2m) before use The saline injection served as a co ntrol group to compare to the PEGSA PNIPAm RGD experimental group. As no effective SCI treatment exists, there was no positive control used to compare to the polymer injection. A total of 6 rats received saline injections and 8 rats received polymer inje ctions. Of the 14 rats used for this study, 13 were successfully compressed for 60 seconds. One rat did not receive a successful compression because of a forceps that slipped and cut the cord. Bleeding occurred in about half of the rats during and after the clip compression. The severity of the ble eding was usually minimal but a couple had profuse bleeding. This bleeding was managed with gauze and saline washes. The compression clip was difficult to keep secured to the spinal cord during the full 60 seconds in some of the rats and as a result, the clip would often slip off the cord. This re quired clip repositioning and usually resulted in more bleeding. As a result of the compression SCI, each rat had paraplegia for at least two weeks before hindlimb movements started to return. Figure 6 15 Compression spinal cord injury. (A) Exposed thoracic vertebrae (T9 T11), (B) exposed spinal cord tissue after laminectomy, (C) spinal cord compression with 30 g aneurysm clip. Post surgery complications occurred in many rats, particularl y in the saline injection rats. Hematuria wa s the predominant complication that often would be

PAGE 79

61 followed by death before the 4 week time point. Of the 6 rats injected with saline, only 2 rats su rvived to 4 weeks, with 1 of those rats having the incomplete compression injury mentioned earlier Of the 4 saline rats that died, 3 died about 10 14 days after surgery and 1 died 23 days after surgery. Each of these rats had successfully recovered and appeared healthy 1 week after surgery, but this was followed by hematuria that became progressively worse. Therefore, only 1 saline injected rat with complete compression successfully made it through the 4 week study ( Table 1 ) Table 1 Surgical outcomes and complications of saline injected rats. A summary of surgical outcomes for polymer injected rats is provided in Table 2 Of the 8 rats injected with polymer, 1 had to be euthanized 6 days after surgery. This rat did not show signs of good health after surgery and was mostly immobile in its cage, leading us to believe there was an unknown surgical complication that resulted in its poor health and eventual euthanasia. Hematuria was also observed in several o f the polymer injected rats after surgery but resolved as the rats dra nk more of the gentamicin water. One week after surgery, these rats generally appeared to be healthy and did not display signs of pain. Table 2 Surgical outcomes and complications of polymer injected rats

PAGE 80

62 To verify localization of the polymer, PEGSA PNIPAm was dyed with toluidine gelling into a specific region of the spinal cord. The surgical area was then washed twice with PBS to confirm the polymer r emained at the injection site, as shown in Figure 6 16 Figure 6 16 A 10 l injection of PEGSA PNIPAm dyed with toluidine blue confirmed localization of the polymer into the spinal cord injection site. 6.5 Functional assessment To assess motor recovery, each rat underwent BBB l ocomotion testing walking track analysis, and l atter walk testing. Unfortunately the only test most rats were able to successfully complete was the BBB open field test. The highest score of 21 was given to each rat before surgery and the lowest score of 0 was given to each rat 1 week after surgery. A summary of the BBB scores for rats that survived 4 weeks is provided in

PAGE 81

63 Table 3 Unfortunately, only 1 saline rat survived the 4 week study duration, making the sample size too small to perform statistical analysis. The polymer injection rats showed large variability i n motor recovery, with a standard deviation of 3.7. Overall, no statistical conclusions can be made with this data due to the high variability of polymer injected rat scores and small sample size of saline injected rats. Table 3 Summary of BBB scores for 4 week rats. Only 1 saline injection rat survived the entire 4 w eek duration of the study. T he polymer injection rats have a large variability ( 3.4 3.7, n = 6). Treatment BBB Score: 4 weeks Saline Rat 1 15 Polymer Rat 1 10 Polymer Rat 2 1 Polymer Rat 3 9 Polymer Rat 4 3 Polymer Rat 5 3 Polymer Rat 6 3 6.6 Histology 6.6.1 H&E staining The location of the SCI lesion was confirmed with H&E staining. Compared to healthy, uninjured tissue, compressed spinal tissue (4 hours post surgery) shows gray

PAGE 82

64 matter dislocation and compression. After 4 weeks and 8 weeks the characteristic glial scar is observed ( Figure 6 17 Figure 6 18 ). A fter 8 weeks, a defined region within th e glial scar shows gaps in th e tissue that have less cellular density than the saline injected rat ( Figure 6 18 ) This could be an indication of polymer localization into the SCI site, as the polymer will not stain with H&E Figure 6 17 Vertical spinal cord sections stained with H&E showing the progression of glial scarring after compression SCI with polymer injection (A) Healthy, uninjured tissue, (B) compressed spinal cord 4 hours after surgery, (C) 4 weeks after surgery. Scale bar = 5 00 m Figure 6 18 G lial scar formation of longitudinal compressed spinal cord after 8 weeks wi th polymer injection. The decreased cellular density at the SCI site could be due to the presence of the polymer which does not stain with H&E Scale bar = 1000 m 6.6.2 GFAP IHC

PAGE 83

65 GFAP staining compared the accumulation of astrocytes at the lesion site between the two treatment groups. One of the goals of the polymer in jection directly into the SCI site was to decrease the formation of the glial scar to encourage neural repair. We expect a higher presence of astrocytes in areas where the treatment is less effective at preventing glial scar formation. After 4 weeks astrocytes were observed within a nd around the compression site with the polymer treatment ( Figure 6 19 A ). Unfortunately there are not enough 4 week saline rats to compare to the po lymer treatment at 4 weeks, making it difficult to draw any conclusions based on these images alone. After 8 weeks, astrocytes appear slightly more distributed throughout the lesion site ( Figure 6 19 B ) This could be an indication that the polymer begins to accumulate and support astrocytic ac tivity as the glial scar forms, but more rats are also needed at this time point to make definit ive conclusions. Figure 6 19 GFAP immunostaining to detect astrocyte accumulation around SCI site in vertical spinal cord sections from (A) representative 4 week polymer injection, (B) 8 week polymer injection. Images acquired with Nikon confocal microscope 2 0X air objective. Scale bar = 500 m

PAGE 84

66 6.6.3 GAP 43 IHC The SCI lesion site forms a barrier to axonal growth an d repair. The goal of the polymer injection into the lesion site was to decrease glial scarring and provide a cellular attachment scaffold for neural growth and axonal extensions into the lesion site. GAP 43 staining identifies proteins of growth cones f rom elongating axons to help distinguish between regenerating axons and non regenerating axons. Polymer injected rats after 4 showed the slight appearance of regenerating axons within the lesion site ( Figure 6 20 ) GAP 43 staining appeared more specific for the 8 week po lymer injected rat ( Figure 6 21 ). These results were encouraging, as the presence of regenera ting axons within the polymer injection site was a sign of a more growth permissive SCI environment. Figure 6 20 GAP 43 immunostaining to detect r egenerating axons. Vertical SCI lesion site from representative 4 week polymer injection ( top to bottom representing rostral and caudal ends of the spinal cord ). Scale bar = 500 m Maximum intensity confocal microscopy image, 10X air objective.

PAGE 85

67 Figure 6 21 GAP 43 immunostaining to detect regenerating axons. Longitudinal spinal cord section from 8 week polymer injection (left and right sides represent ing rostral and caudal ends of the spinal cord). Scale bar = 500 m Maximum intensity confocal microscopy image, 10X air objective.

PAGE 86

68 7 Conclusion The objective of this study was to develop an injectable RTG capable of acting as a cellular scaffold for SCI repair. Current SCI treatments lack effective regenerative treatments, but cellular and molecular therapy approaches present promising new avenues to permit repair from the injury. To enhance the therapeu tic efficacy of these new treatment options, polymer scaffolds have been developed to act as supportive substrates and delivery vehicles. A novel RTG with a PEG based backbone was successfully synthesized and characterized. Conj ugation of RGD to PEGSA PNIP Am w as not able to be verified by FT IR as PNIPAm has signal dominance over specific RGD peaks However, RGD conjugation to the PEGSA backbone (without PNIPAm) was confirmed with FT IR. By proving RGD conjugation to the PEGSA backbone, this enhanced the confidence that RGD was suc cessfully conjugated to PE GSA PNIPAm. The gelling properties of PEGSA PNIPAm was confirmed with gelling tests at 37C. A sharp LCST was demo nstrated with temperature contro lled UV VIS spectroscopy, verifying the RTG properties necessary for injectability and in situ forming gels. Within 5 10 minutes of exposure to 37C, the gels would begin to shrink by roughly 20 30% which is an indication of gel instability. This contra sts with t he gel stability of PSHU PNIPAm (mentioned in literature review) which is able to maintain its conformation over prolonged periods exposed to 37C. The primary difference between the two polymers is the properties of the backbone. PEGSA is a hy drophilic backbone, while PSHU is a hydrophobic backbone with long er alkyl chains. PEGSA PNIPAm may not have long enough alkyl chains to maintain complete gel stability over prolonged periods.

PAGE 87

69 Preliminary in vitro studies wi th PC12 cells showed that PEGSA PNIPAm RTG promoted a permissive environment for cell survival. MTT and Alamar Blue assays demonst rated that P EGSA PNIPAm gels were non cytotoxic and wer e capable of supporting the proliferation of PC12 cells In vitro studies with a primary RGC cu lture scaffold. Brn 3a and III Tubulin staining showed the gelled polymer was capable of supporting viable RGCs with axon extension s The RTG was investigated for axon regeneration in a c ompression rat SCI model. Polymer or saline solutions were injected int o the SCI site and assessed for neural regeneration and motor repair. Saline injected rats experienced health complications (hematuria) that resulted death before the 30 day study duration concluded for all but one rat. However, polymer injected rats that experienced similar health complications were able to recover quickly and all but one survived the entire duration of the study. Bleeding during clip compression is a likely cau se for hematuria, but the exact reason for the high death rate in saline injected rats is unknown. It is possible that formed blood clots are dislodging over time, causing embolisms in saline injected rats. The polymer may be acting as a barrier to prevent these clots from circulating. It is also possible the polymer is acting to provide pressure that allows bleeding to clot. Therefore, the saline rats may never have properly healed the bleeding that occurred after compression, resulting in death. Motor re pair was investigated with three different functional tests: BBB score, walking track, and latter walk. After 4 weeks, none of the rats were able to successfully n avigate the latter walk test The hindlimbs of each rat were unable grip any of the latter

PAGE 88

70 r ungs, resulting in complete misses for each step taken This loss of hindlimb control also resulted in poor results with walking track analysis. Rats were unable to plant their painted hindlimbs onto the paper for readable paw prints required for analysi s. Finally, BBB openfield locomotion scores showed high variability in the polymer treatment group while the small sample size for the saline injected rats prevented statistical analysis Therefore, functional tes ting was unable to provide conclusive dat a regarding H&E staining revealed characteristic s car formation at the SCI site. A comparison between the two treatment groups after 8 weeks revealed possible localization of the polymer within the injury site. IHC staining with GFAP confirmed the accumulation of astrocytes around the SCI site, but no difference between treatment groups could be discerned. GAP 43 staining was able to confirm the presence of regenerati ng axons in polymer injected rats after 4 and 8 weeks This was the most encouraging result, as it demonstrated that the poly mer injection site was able to overcome toxic effects of the SCI environment to promote regeneration. In closing a novel, biomimetic RTG was able to support the in vitro surv i val of primary retinal ganglion cells proving it to be a promising scaffold for CNS neural tissue. Despite variability in the compression SCI model and inconclusive functional tests, after 4 and 8 weeks, histology revealed the presence of regen erating axo ns within the SCI lesion where the RTG was injected These preliminary results demonstrate d this novel injectable RTG is a promising noninvasive approach for delivering a biomimetic

PAGE 89

71 scaffold within the SCI site to help overcome regenerative barriers and promote neural repair.

PAGE 90

72 8 Study limitations and f uture work 8.1 Modify control and experimental groups The animal study was designed to be a small pilot st udy to assess spinal cord axon regeneration after injecting a new biomimetic RTG. This SCI study only included a control saline injection and the PEGSA PNIPAm RGD experimental injection. Future studies should include a surgery only group (no compression) and also a compression only group (no injection) to help determine if surgical or injection erro rs are contributing to large variability in SCI lesions. Including a PEGSA PNIPAm experimental group with no conjugated RGD would allow the polymer by itself to be assessed as a supportive cellular scaffold. Including this group would also allow for a go od comparison to PEGSA PNIPAm RGD to verify its ability to deliver and support the cellular actions of the RGD peptide. 8.2 Modify SCI compression As mentioned earlier, the source of the large variability in survival and functional assessment between animals in the SCI model is l ikely due to bleeding during clip compression. The aneurysm clip used for compression may be too sharp for the delicate spinal tissue, leaving little room for error while applying the clip Bleeding seemed to occur a fter full compression and not during the initial application of the clip. This likely occurred because the clip either punctured or lacerated the spinal cord vasculature. Using a clip with smoother edge s or a forceps will be a better alternative. Problem s were also encountered after initial application of the compression clip. The clip had a tendency to slip off the spinal cord afte r initial application, resulting in possible

PAGE 91

73 incomplete compression, requiring reapplication. C lips that appeared to remain i n place may not have been compressing the entire spinal cord. In the future, further laminectomy should be performed to better verify complete cord compression The compression time will also be modified. A one minute compression may be too long and could result in more complications and inconsistencies between rats Further studies will use a brief compressi on (1 5 seconds) to minimize these errors 8.3 Modify PEGSA backbone to prevent gel shrinking As discussed earlier, the PEGSA PNIPAm gel begins to shrink when exposed to physiological temperatures after several minutes ( Figure 8 1 ) Figure 8 1 After several minutes, PEGSA PNIPAm shrinks, resulting in a gel that no longer holds the shape of the vial. A similar polyurethane based polymer, PSHU PNIPAm, does not have these shrinking problems 95 The aggregation of hydrophobic groups contributes to a more stable g el over prolonged exposures to physiological temperatures. PEGSA is a more hydrophilic polymer backbone compared to PSHU, which results in fewer hydrophobic group aggregations during gelling. This is an indication that the PEGSA backbone needs to be modi fied to promot e more stability while gelled. Replacing succinic acid with

PAGE 92

74 longer chained di carboxylic acid s, such as sebacic acid, could allow more hydrophobic groups in the backbone to promote enhanced gel stability. Experimenting with different PEG cha in lengths is also another option. 8.4 Increase study duration We observed slight visual improvements (not statistical) in motor recovery with the two rats (polymer injection and saline injection) extended to 8 weeks. Extending the study to 8 weeks and 12 wee ks is important to see if functional improvements and axonal regeneration eventually level off over time. 8.5 Monitor polymer presence in spinal tissue The polymer was successfully injected into the spinal tissue. However, a method needs to be developed to monitor long term presence and localization of the polymer within the spinal tissue over the course of the study. This can be achieved by conjugating a fluorophore to the PEGSA hydroxyl groups. After injecting into spinal tissue, the rats can be euthanize d at different time points to conduct IHC. A similar method was used to observe the localization of a natural polymer injected into spinal tissue, with successful visualization of the fluorescently labeled polymer 62 This study did not atte mpt long term visualization of the polymer but it would be worth trying various time points to assess not only the presence of the polymer over time but also possible in v ivo degradation of the polymer. 8.6 Long term goal of RTG SCI application The long term goal of this RTG is not only to incorporate cellular adhesive properties through the use of RGD peptide, but to also integrate NT Fs and cells into the

PAGE 93

75 RTG system to promote SCI repair. NTFs can be integrated into polymer through electrostatic interaction s to provide essential biochemical cues for neurite growth The administration of NTFs will not only be made within the RTG, but also should be made upstream and downstream of the SCI site. This type of NTF administration will provide regenerating neuron s cues to not only grow with in the SCI lesion but could also allow regenerating axons to bridge the entire SCI lesion. Replacing damaged tissue in the SCI site is also a priority, and this can be accomplished through the use stem cel l therapies. As the primary RGC experiment demonstrated, cells can be successfully encapsulated into the gelled polymer scaffold to promote neu rite growt h and axon extension. O ur lab was able to successfully grow human NSCs onto a polyurethane biomimetic scaffold that promoted differentiation into motor neurons. In vitro studies with PEGSA PNIPAm encapsulated with human NSCs would be the next step for the future direction of this polymers cellular applications. If the human NSCs can successfully survive and grow withi n a 3D polymer matrix (as was seen with the RGC experiments), the next in vivo study would incorporate human NSCs into the polymer for SCI injection. Overall, incorporating NTFs and stem cells within the RT G design should provide a growth permissive bio mimetic ECM environment within the toxic SCI lesion site to promote neural repair. If neurons are able to grow and differentiate within the SCI lesion and extend beyond the injury site, this could allow the body to provide natural cues to enhance meaningf ul axon rewiring to downstream tissues. As research continues to discover different methods to overcome the toxic SCI environment to promote neural

PAGE 94

76 regeneration, injectable RTGs will continue to prove to be effective delivery devices to enhance the therap eutic effects of each of these promising treatment approaches.

PAGE 95

77 References 1. Silva N a., Sousa N, Reis RL, Salgado AJ. From basics to clinical: A comprehensive review on spinal cord injury. Prog Neurobiol 2014 ;114:25 57. doi:10.1016/j.pneurobio.2013.11.002. 2. Nscisc. Spinal cord injury facts and figures at a glance. J Spinal Cord Med 2013;36(1):1 2. doi:10.1179/1079026813Z.000000000136. 3. Selvarajah S, Hammond ER, Haider AH, et al. The burden of acute traumatic spinal cord injury among adults in the united states: an update. J Neurotrauma 2014;31(3):228 238. doi:10.1089/neu.2013.3098. 4. Richardson PM, McGuinness UM, Aguayo AJ. Axons from CNS neurones regenerate into PNS grafts. Nature 1980;284:264 265. 5. Cajal RY. Degeneration and Regeneration of the Nervous System. Oxford University Press 1928. 6. Ziegler MG. Nervous System. Prim Auton Nerv Syst 2012:291 294. doi:10.1016/B978 0 12 386525 0.00061 5. 7. Ronsyn MW, Berneman ZN, Van Tendeloo VFI, Jorens PG, Ponsaerts P. Can cell therapy heal a spinal cord injury? Spinal Cord 2008;46(8):532 539. doi:10.1038/sc.2008.13. 8. Li Y, Walker CL, Zhang YP, Shields CB, Xu X M. Surgical decompression in acute spinal cord injury: A review of clinical evidence, animal model studies, and potential future directions of investigation. Front Biol (Beijing) 2014;9(2):127 136. http://download.springer.com/static/pdf/558/art%3A10.1 007%2Fs11515 014 1297 z.pdf?originUrl=http://link.springer.com/article/10.1007/s11515 014 1297 z&token2=exp=1438393533~acl=/static/pdf/558/art%253A10.1007%252Fs11515 014 129. 9. Hurlbert RJ. Methylprednisolone for the treatment of acute spinal cord injury : point. Neurosurgery 2014;61 Suppl 1(1):32 35. doi:10.1227/NEU.0000000000000393. 10. Hurlbert RJ, Hadley MN, Walters BC, et al. Pharmacological therapy for acute spinal cord injury. Neurosurgery 2013;72(SUPPL.2):93 105. doi:10.1227/NEU.0b013e31827765c6 11. Liu JC, Patel A, Vaccaro AR, Lammertse DP, Chen D. Methylprednisolone after traumatic spinal cord injury: yes or no? PM R 2009;1(7):669 673. doi:10.1016/j.pmrj.2009.06.002.

PAGE 96

78 12. Bydon M, Lin J, Macki M, Gokaslan ZL, Bydon A. The Current Role of Ste roids in Acute Spinal Cord Injury. World Neurosurg 2013;82(June):1 7. doi:10.1016/j.wneu.2013.02.062. 13. Bethea JR. Spinal cord injury induced inflammation: a dual edged sword. Prog Brain Res 2000;128:33 42. doi:10.1016/S0079 6123(00)28005 9. 14. BETH EA JR, NAGASHIMA H, ACOSTA MC, et al. Systemically Administered Interleukin 10 Reduces Tumor Necrosis Factor Alpha Production and Significantly Improves Functional Recovery Following Traumatic Spinal Cord Injury in Rats. J Neurotrauma 1999;16(10):851 863. doi:10.1089/neu.1999.16.851. 15. Zuo J, Neubauer D, Dyess K, Ferguson TA, Muir D. Degradation of chondroitin sulfate proteoglycan enhances the neurite promoting potential of spinal cord tissue. Exp Neurol 1998;154(2):654 662. doi:10.1006/exnr.1998.6951. 16. Lu P, Jones LL, Tuszynski MH. BDNF expressing marrow stromal cells support extensive axonal growth at sites of spinal cord injury. Exp Neurol 2005;191(2):344 360. doi:10.1016/j.expneurol.2004.09.018. 17. Weishaupt N, Blesch A, Fouad K. BDNF: the ca reer of a multifaceted neurotrophin in spinal cord injury. Exp Neurol 2012;238(2):254 264. doi:10.1016/j.expneurol.2012.09.001. 18. Ye JH, Houle JD. Treatment of the chronically injured spinal cord with neurotrophic factors can promote axonal regeneratio n from supraspinal neurons. Exp Neurol 1997;143(1):70 81. doi:10.1006/exnr.1996.6353. 19. McCall J, Weidner N, Blesch A. Neurotrophic factors in combinatorial approaches for spinal cord regeneration. Cell Tissue Res 2012;349(1):27 37. doi:10.1007/s00441 012 1388 6. 20. Doucette R. PNS CNS transitional zone of the first cranial nerve. J Comp Neurol 1991;312(3):451 466. doi:10.1002/cne.903120311. 21. Li L, Adnan H, Xu B, et al. Effects of transplantation of olfactory ensheathing cells in chronic spinal cord injury: a systematic review and meta analysis. Eur Spine J 2015;24(5):919 930. doi:10.1007/s00586 014 3416 6. 22. Ramer LM, Au E, Richter MW, Liu J, Tetzlaff W, Roskams a. J. Peripheral Olfactory Ensheathing Cells Reduce Scar and Cavity Formation an d Promote Regeneration after Spinal Cord Injury. J Comp Neurol 2004;473(1):1 15. doi:10.1002/cne.20049.

PAGE 97

79 23. Imaizumi T, Lankford KL, Burton W V, Fodor WL, Kocsis JD. Xenotransplantation of transgenic pig olfactory ensheathing cells promotes axonal regene ration in rat spinal cord. Nat Biotechnol 2000;18(9):949 953. doi:10.1038/79432. 24. Huang H, Xi H, Chen L, Zhang F, Liu Y. Long term outcome of olfactory ensheathing cell therapy for patients with complete chronic spinal cord injury. Cell Transplant 20 12;21(SUPPL. 1):S23 S31. doi:10.3727/096368912X633734. 25. Dlouhy BJ, Awe O, Rao RC, Kirby P a, Hitchon PW. Autograft derived spinal cord mass following olfactory mucosal cell transplantation in a spinal cord injury patient. J Neurosurg Spine 2014;21(Oct ober):1 5. doi:10.3171/2014.5.SPINE13992. 26. Tetzlaff W, Okon EB, Karimi Abdolrezaee S, et al. A systematic review of cellular transplantation therapies for spinal cord injury. J Neurotrauma 2011;28(8):1611 1682. doi:10.1089/neu.2009.1177. 27. Pal R, V enkataramana NK, Bansal A, et al. Ex vivo expanded autologous bone marrow derived mesenchymal stromal cells in human spinal cord injury/paraplegia: a pilot clinical study. Cytotherapy May 2015. http://www.tandfonline.com/doi/abs/10.3109/14653240903253857# .Vc6xHvlViko Accessed August 15, 2015. 28. Mariano ED, Batista CM, Barbosa BJAP, et al. Current perspectives in stem cell therapy for spinal cord repair in humans: a review of work from the past 10 years. Arq Neuropsiquiatr 2014;72(6):451 456. doi:10.1590/0004 282X20140051. 29. Willerth SM, Sakiyama Elbert SE. Cell therapy for spinal cord regeneration. Adv Drug Deliv Rev 2008;60(2):263 276. doi:10.1016/j.addr.2007.08.028. 30. Lu P, Wang Y, Graham L, et al. Long distance growth and connectivit y of neural stem cells after severe spinal cord injury. Cell 2012;150(6):1264 1273. doi:10.1016/j.cell.2012.08.020. 31. Stenudd M, Sabelstrm H, Frisn J. Role of Endogenous Neural Stem Cells in Spinal Cord Injury and Repair. JAMA Neurol 2015;72(2):235. doi:10.1001/jamaneurol.2014.2927. 32. Lowry NA, Temple S. Making Human Neurons from Stem Cells after Spinal Cord Injury. PLoS Med 2007;4(2):e48. doi:10.1371/journal.pmed.0040048. 33. Aurand ER, Lampe KJ, Bjugstad KB. Defining and designing polymers and hydrogels for neural tissue engineering. Neurosci Res 2012;72(3):199 213. doi:10.1016/j.neures.2011.12.005.

PAGE 98

80 34. Straley KS, Foo CWP, Heilshorn SC. Biomaterial design strategies for the treatment of spinal cord injuries. J Neurotrauma 2010;27(1):1 19. d oi:10.1089/neu.2009.0948. 35. Pgo AP, Kubinova S, Cizkova D, et al. Regenerative medicine for the treatment of spinal cord injury: more than just promises? J Cell Mol Med 2012;16(11):2564 2582. doi:10.1111/j.1582 4934.2012.01603.x. 36. Lutolf MP, Hubbe ll J a. Synthetic biomaterials as instructive extracellular microenvironments for morphogenesis in tissue engineering. Nat Biotechnol 2005;23(1):47 55. doi:10.1038/nbt1055. 37. Okamoto M, John B. Synthetic biopolymer nanocomposites for tissue engineering scaffolds. Prog Polym Sci 2013;38(10 11):1487 1503. doi:10.1016/j.progpolymsci.2013.06.001. 38. Harris JM. Poly(ethylene Glycol) Chemistry: Biotechnical and Biomedical Applications Springer Science & Business Media; 2013. https://books.google.com/books ?hl=en&lr=&id=lU30BwAAQBAJ&oi=fnd&pg=P A1&dq=polyethylene+glycol+biological+applications&ots=eENOq2tZD5&sig=Iv M6PfsmsVRODfqJCttMVmKeFeY#v=onepage&q&f=false. 39. Park D, Wu W, Wang Y. A functionalizable reverse thermal gel based on a polyurethane/PEG block copolymer. Biomaterials 2011;32(3):777 786. doi:10.1016/j.biomaterials.2010.09.044. 40. Zhu J. Bioactive modification of poly(ethylene glycol) hydrogels for tissue engineering. Biomaterials 2010;31(17):4639 4656. doi:10.1016/j.biomaterials.2010.02.044. 41. Yamaoka T, Yasuhiko T, Yoshito I. Distribution and tissue uptake of poly(ethylene glycol) with different molecular weights after intravenous administration to mice. J Pharm Sci 1994;83(4):601 606. http://onlinelibrary.wiley.com/doi/10.1002/jps.260083 0432/abstract. 42. Pashuck ET, Stevens MM. Designing regenerative biomaterial therapies for the clinic. Sci Transl Med 2012;4(160):160sr4. doi:10.1126/scitranslmed.3002717. 43. Drury JL, Mooney DJ. Hydrogels for tissue engineering: Scaffold design varia bles and applications. Biomaterials 2003;24(24):4337 4351. doi:10.1016/S0142 9612(03)00340 5. 44. Mahoney MJ, Anseth KS. Three dimensional growth and function of neural tissue in degradable polyethylene glycol hydrogels. Biomaterials 2006;27(10):2265 22 74. doi:10.1016/j.biomaterials.2005.11.007.

PAGE 99

81 45. Comolli N, Neuhuber B, Fischer I, Lowman A. In vitro analysis of PNIPAAm PEG, a novel, injectable scaffold for spinal cord repair. Acta Biomater 2009;5(4):1046 1055. doi:10.1016/j.actbio.2008.10.008. 46. L uo J, Borgens R, Shi R. Polyethylene glycol immediately repairs neuronal membranes and inhibits free radical production after acute spinal cord injury. J Neurochem 2002;83(2):471 480. doi:10.1046/j.1471 4159.2002.01160.x. 47. Luo J, Shi R. Diffusive oxid ative stress following acute spinal cord injury in guinea pigs and its inhibition by polyethylene glycol. Neurosci Lett 2004;359(3):167 170. doi:10.1016/j.neulet.2004.02.027. 48. Borgens RB, Bohnert D. Rapid recovery from spinal cord injury after subcuta neously administered polyethylene glycol. J Neurosci Res 2001;66(6):1179 1186. http://www.ncbi.nlm.nih.gov/pubmed/11746451. Accessed August 31, 2015. 49. Nomura H, Tator CH, Shoichet MS. Bioengineered strategies for spinal cord repair. J Neurotrauma 200 6;23(3 4):496 507. doi:10.1089/neu.2006.23.496. 50. Bakshi A, Fisher O, Dagci T, Himes BT, Fischer I, Lowman A. Mechanically engineered hydrogel scaffolds for axonal growth and angiogenesis after transplantation in spinal cord injury. J Neurosurg Spine 2004;1(3):322 329. doi:10.3171/spi.2004.1.3.0322. 51. Johnson PJ, Parker SR, Sakiyama Elbert SE. Fibrin based tissue engineering scaffolds enhance neural fiber sprouting and delay the accumulation of reactive astrocytes at the lesion in a subacute model o f spinal cord injury. J Biomed Mater Res A 2010;92(1):152 163. doi:10.1002/jbm.a.32343. 52. Hejcl A, Sed J, Kapcalov M, et al. HPMA RGD hydrogels seeded with mesenchymal stem cells improve functional outcome in chronic spinal cord injury. Stem Cells De v 2010;19(10):1535 1546. doi:10.1089/scd.2009.0378. 53. Stokols S, Tuszynski MH. Freeze dried agarose scaffolds with uniaxial channels stimulate and guide linear axonal growth following spinal cord injury. Biomaterials 2006;27(3):443 451. doi:10.1016/j. biomaterials.2005.06.039. 54. Moore MJ, Friedman J a., Lewellyn EB, et al. Multiple channel scaffolds to promote spinal cord axon regeneration. Biomaterials 2006;27(3):419 429. doi:10.1016/j.biomaterials.2005.07.045. 55. Hurtado A, Moon LDF, Maquet V, B lits B, Jrme R, Oudega M. Poly (D,L lactic acid) macroporous guidance scaffolds seeded with Schwann cells genetically modified to secrete a bi functional neurotrophin implanted in the completely

PAGE 100

82 transected adult rat thoracic spinal cord. Biomaterials 20 06;27(3):430 442. doi:10.1016/j.biomaterials.2005.07.014. 56. Nomura H. Complete spinal cord transection treatedby implantation of a reinforced synthetic hydrogel channel results in syringomyelia and caudal migration of the rostral stump. Neurosurgery 20 06;59(1):183 192. http://journals.lww.com/neurosurgery/Abstract/2006/07000/COMPLETE_SPINA L_CORD_TRANSECTION_TREATEDBY.22.aspx. 57. Iannotti C, Zhang YP, Shields LBE, et al. Dural repair reduces connective tissue scar invasion and cystic cavity formation a fter acute spinal cord laceration injury in adult rats. J Neurotrauma 2006;23(6):853 865. doi:10.1089/neu.2006.23.853. 58. Macaya D, Spector M. Injectable hydrogel materials for spinal cord regeneration: a review. Biomed Mater 2012;7(1):012001. doi:10.1 088/1748 6041/7/1/012001. 59. Piantino J, Burdick J a., Goldberg D, Langer R, Benowitz LI. An injectable, biodegradable hydrogel for trophic factor delivery enhances axonal rewiring and improves performance after spinal cord injury. Exp Neurol 2006;201(2):359 367. doi:10.1016/j.expneurol.2006.04.020. 60. Jain A, Kim Y T, McKeon RJ, Bellamkonda R V. In situ gelling hydrogels for conformal repair of spinal cord defects, and local delivery of BDNF after spinal cord injury. Biomaterials 2006;27(3) :497 504. doi:10.1016/j.biomaterials.2005.07.008. 61. Ruel Garipy E, Leroux J C. In situ forming hydrogels -review of temperature sensitive systems. Eur J Pharm Biopharm 2004;58(2):409 426. doi:10.1016/j.ejpb.2004.03.019. 62. Gupta D, Tator CH, Shoiche t MS. Fast gelling injectable blend of hyaluronan and methylcellulose for intrathecal, localized delivery to the injured spinal cord. Biomaterials 2006;27(11):2370 2379. doi:10.1016/j.biomaterials.2005.11.015. 63. Ritfeld GJ, Rauck BM, Novosat TL, et al. The effect of a polyurethane based reverse thermal gel on bone marrow stromal cell transplant survival and spinal cord repair. Biomaterials 2014;35(6):1924 1931. doi:10.1016/j.biomaterials.2013.11.062. 64. Engineering NT. A Biomimetic Reverse Thermal Ge l for 3 Dimensional Neural Tissue Engineering. 2014;1(4):1 5. 65. Castellanos BP, Park D, Long CS, et al. Temperature Responsive Materials May Act As Extracellular Matrix and Cell Vehicle for Cardiac Tissue Engineering. J Am Coll Cardiol 2015;65(10):A960 doi:10.1016/S0735 1097(15)60960 3.

PAGE 101

83 66. Ruoslahti E. RGD and other recognition sequences for integrins. Annu Rev Cell Dev Biol 1996;12:697 715. doi:10.1146/annurev.cellbio.12.1.697. 67. Niu X, Wang Y, Luo Y, Xin J. Arg Gly Asp (RGD) modified biomimetic polymeric materials. J Mater Sci 2005;21(4):571 576. 68. Ruoslahti E. Integrins. J Clin Invest 1991;87(1):1 5. doi:10.1172/JCI114957. 69. Schaffner P, Dard MM. Structure and function of RGD peptides involved in bone biology. Cell Mol Life Sci 2003;60 (1):119 132. doi:10.1007/s000180300008. 70. Cook AD, Hrkach JS, Gao NN, et al. Characterization and development of RGD peptide modified poly(lactic acid co lysine) as an interactive, resorbable biomaterial. J Biomed Mater Res 1997;35(4):513 523. doi:10.1 002/(SICI)1097 4636(19970615)35:4<513::AID JBM11>3.0.CO;2 C. 71. Yun D, Lee YM, Laughter MR, Freed CR, Park D. Substantial Differentiation of Human Neural Stem Cells Into Motor Neurons on a Biomimetic Polyurea. Macromol Biosci 2015:n/a n/a. doi:10.1002 /mabi.201500066. 72. Talac R, Friedman JA, Moore MJ, et al. Animal models of spinal cord injury for evaluation of tissue engineering treatment strategies. Biomaterials 2004;25(9):1505 1510. doi:10.1016/S0142 9612(03)00497 6. 73. Stokes BT, Jakeman LB. Experimental modelling of human spinal cord injury: a model that crosses the species barrier and mimics the spectrum of human cytopathology. Spinal cord Off J Int Med Soc Paraplegia 2002;40(3):101 109. doi:10.1038/sj.sc.3101254. 74 Kwon BK, Streijger F, Hill CE, et al. Large animal and primate models of spinal cord injury for the testing of novel therapies. Exp Neurol 2015;269:154 168. doi:10.1016/j.expneurol.2015.04.008. 75. Fukuda S, Nakamura T, Kishigami Y, et al. New canine spinal cord injury model free from laminectomy. Brain Res Protoc 2005;14(3):171 180. doi:10.1016/j.brainresprot.2005.01.001. 76. Rabinowitz RS, Eck JC, Jr CMH, et al. Urgent surgical decompression compared to methylprednisolone for the treatment of acute spinal cord injury: a randomized prospective study in beagle dogs. Spine (Phila Pa 1976) 2008;33(21):2260 2268. http://journals.lww.com/spinejournal/Abstract/2008/10010/Urgent_Surgical_Deco mpression_Compared_to.3.aspx. 77. Pritchard CD, Slotkin JR, Yu D et al. Establishing a model spinal cord injury in the African green monkey for the preclinical evaluation of biodegradable polymer

PAGE 102

84 scaffolds seeded with human neural stem cells. J Neurosci Methods 2010;188(2):258 269. doi:10.1016/j.jneumeth.2010.02.019. 78. Park S S, Lee YJ, Lee SH, et al. Functional recovery after spinal cord injury in dogs treated with a combination of Matrigel and neural induced adipose derived mesenchymal Stem cells. Cytotherapy 2012;14(5):584 597. doi:10.3109/14653249.2012.658913. 79. METZ GAS, CURT A, van de MEENT H, KLUSMAN I, SCHWAB ME, DIETZ V. Validation of the Weight Drop Contusion Model in Rats: A Comparative Study of Human Spinal Cord Injury. J Neurotrauma 2000;17(1):1 17. doi:10.1089/neu.2000.17.1. 80. STOKES BT, NOYES DH, BEHRMANN DL. An Electromechanical Spinal Injury Technique with Dynamic Sensitivity. J Neurotrauma 1992;9(3):187 195. doi:10.1089/neu.1992.9.187. 81. Gruner J a. A monitored contusion model of spinal cord injury in the rat. J Neurotrauma 1992;9(2):12 3 126; discussion 126 128. 82. Robins SL, Fehlings MG. Models of experimental spinal cord injury: Translational relevance and impact. Drug Discov Today Dis Model 2008;5(1):5 11. doi:10.1016/j.ddmod.2008.07.001. 83. Blight AR. Animal Models of Spinal Cor d Injury. Top Spinal Cord Inj Rehabil 2000;6(2):1 13. doi:10.1310/2XNY A824 UCTF EN4D. 84. Basso DM, Beattie MS, Bresnahan JC. A sensitive and reliable locomotor rating scale for open field testing in rats. J Neurotrauma 1995;12(1):1 21. doi:10.1089/neu .1995.12.1. 85. Barros Filho TEP De, Molina AEIS. Analysis of the sensitivity and reproducibility of the Basso, Beattie, Bresnahan (BBB) scale in Wistar rats. Clinics (Sao Paulo) 2008;63(1):103 108. 86. Dijkstra JR, Meek MF, Robinson PH, Gramsbergen A. Methods to evaluate functional nerve recovery in adult rats: Walking track analysis, video analysis and the withdrawal reflex. J Neurosci Methods 2000;96(2):89 96. doi:10.1016/S0165 0270(99)00174 0. 87. Metz G a, Merkler D, Dietz V, Schwab ME, Fouad K. Efficient testing of motor function in spinal cord injured rats. Brain Res 2000;883(2):165 177. doi:10.1016/S0006 8993(00)02778 5.

PAGE 103

85 88. Anand P, Mathangi DC, Jeraud M, Namasivayam a., Babu RS. Behavioral a nalysis after sciatic nerve compression in albino rats. Ann Neurosci 2011;18(2):37 43. doi:10.5214/ans.0972.7531.1118203. 89. Metz G a, Whishaw IQ. The ladder rung walking task: a scoring system and its practical application. J Vis Exp 2009;(28):4 7. do i:10.3791/1204. 90. Bignami a, Eng LF, Dahl D, Uyeda CT. Localization of the glial fibrillary acidic protein in astrocytes by immunofluorescence. Brain Res 1972;43(2):429 435. doi:10.1016/0006 8993(72)90398 8. 91. Doster SK, Lozano a M, Aguayo a J, Will ard MB. Expression of the growth associated protein GAP 43 in adult rat retinal ganglion cells following axon injury. Neuron 1991;6(4):635 647. doi:10.1016/0896 6273(91)90066 9. 92. Shafer T, Atcison W. Transmitter, ion channel and receptor properties of pheochromocytoma (PC12) cells: a model for neurotoxicological studies. Neurotoxicology 1991;12(3):473 492. http://www.ncbi.nlm.nih.gov/pubmed/1720882. 93. Zhang Y, Ali SF, Dervishi E, et al. Cytotoxicity effects of graphene and single wall carbon nanotu bes in neural phaeochromocytoma derived pc12 cells. ACS Nano 2010;4(6):3181 3186. doi:10.1021/nn1007176. 94. Yun D, Famili A, Lee YM, Jenkins PM, Freed CR, Park D. Biomimetic poly(serinol hexamethylene urea) for promotion of neurite outgrowth and guidanc e. J Biomater Sci Polym Ed 2014;25(4):354 369. doi:10.1080/09205063.2013.861170. 95. Pea B, Shandas R, Park D. A heparin mimicking reverse thermal gel for controlled delivery of positively charged proteins. J Biomed Mater Res Part A 2015;103(6):2102 2108. doi:10.1002/jbm.a.35345.