Electrospun Cartilage Glycosaminoglycans for Biomimetic Scaffolds
A Thesis
Submitted to the Faculty
of
Drexel University
by
Sean Michael Dowd
in partial fulfillment of the
requirements for the degree
of
Master of Science in Materials Science and Engineering
May 2012
© Copyright 2012
Sean M. Dowd. All Rights Reserved.
ii
Dedications
To my family and friends; those still with us, and those who have passed on
iii
Acknowledgments
To begin, I must thank my advisor Professor Caroline Schauer who’s help,
support, and above all patience have been nothing short of superhuman and really helped
me to understand, and gain an appreciation of, the world of polymer science. Secondly,
I’d like to thank my mentors Keith Fahnestock and Laura Toth. These people were there
from the very beginning to aid me with my infinite amount of questions, and frequent
corrections of my errors. I would also like to thank Jennifer Atchison and Marjorie
Austero for their help in obtaining SEMs, FTIRs, and in the beginning for making sure
that what I was doing actually made sense. I must also thank Dr. Giuseppe Palmese for
the use of the equipment in his lab. I would also like to thank the rest of the Natural
Polymers & Photonics Group for their support. Last, but not least, I’d like to thank my
committee members Professor Caroline Schauer, Professor Michele Marcolongo, and
Professor Wei-Heng Shih.
iv
Table of Contents
List of Tables .................................................................................................................... vii
List of Figures .................................................................................................................. viii
Abstract .............................................................................................................................. xi
Chapter 1: Introduction ....................................................................................................... 1
1.1 Materials .................................................................................................................... 1
1.1.1 Overview of Glycosaminoglycans ...................................................................... 1
1.1.2 Challenges of Using Glycosaminoglycans ......................................................... 1
1.2 Applications .............................................................................................................. 2
1.2.1 Epidemiology of Knee Cartilage and Meniscus ................................................. 2
1.2.2 Cartilage Tissue Degradation ............................................................................. 3
1.3 Current Non-Tissue Engineered Treatments for OA ................................................ 4
1.4 Tissue Engineered Treatments .................................................................................. 6
1.5 Overview of Knee Cartilage and Meniscus Anatomy ............................................... 9
1.5.1 Articular Cartilage .............................................................................................. 9
1.5.2 Meniscus ........................................................................................................... 11
1.6 Overall Goal and Objectives ................................................................................... 13
1.6.1 Overall Goal ..................................................................................................... 13
1.6.2 Objectives ......................................................................................................... 13
Chapter 2: Background ..................................................................................................... 15
2.1 Materials .................................................................................................................. 15
2.1.1 Chondroitin Sulfate........................................................................................... 15
2.1.2 Hyaluronic Acid ............................................................................................... 17
2.1.3 Solvents ............................................................................................................ 18
2.1.4 Crosslinking ...................................................................................................... 19
v
2.2 Electrospinning........................................................................................................ 20
2.2.1 Overview of Electrospinning ............................................................................ 21
2.2.2 Parameters for Electrospinning ........................................................................ 22
2.3 Applications of Electrospun Fibers and Membranes .............................................. 27
Chapter 3: Experimental Design and Procedure ............................................................... 29
3.1 Materials .................................................................................................................. 29
3.1.1 Dimethylformamide Solvent System ................................................................ 29
3.1.2 Acetic Acid Solvent System ............................................................................. 30
3.1.3 Divinyl Sulfone Crosslinking ........................................................................... 30
3.2 Electrospinning........................................................................................................ 31
3.2.1 Dimethylformamide Solvent System ................................................................ 32
3.2.2 Acetic Acid Solvent System ............................................................................. 33
3.3 Measuring pH .......................................................................................................... 35
3.4 Viscosity .................................................................................................................. 36
3.5 Attenuated Total Reflectance Fourier Transform Infrared Spectroscopy ............... 38
3.6 Field Emission Scanning Electron Microscopy ...................................................... 39
3.7 Energy-Dispersive X-ray Spectroscopy .................................................................. 40
3.8 Solubility ................................................................................................................. 40
Chapter 4: Results and Discussion .................................................................................... 42
4.1 Dimethylformamide Solvent System ...................................................................... 42
4.1.1 Viscosity ........................................................................................................... 42
4.1.2 Attenuated Total Reflectance Fourier Transform Infrared Spectroscopy ........ 44
4.1.3 Field Emission Scanning Electron Microscopy ................................................ 47
4.2 Acetic Acid Solvent System .................................................................................... 56
4.2.1 Viscosity ........................................................................................................... 57
vi
4.2.2 Attenuated Total Reflectance Fourier Transform Infrared Spectroscopy ........ 58
4.2.3 Field Emission Scanning Electron Microscopy ................................................ 61
4.2.4 Energy-Dispersive X-ray Spectroscopy ........................................................... 65
4.3 Chemical Stability ................................................................................................... 67
4.3.1 Field Emission Scanning Electron Microscopy ................................................ 68
4.3.2 Attenuated Total Reflectance Fourier Transform Infrared Spectroscopy ........ 70
4.3.3 Solubility .......................................................................................................... 72
Chapter 5: Conclusions ..................................................................................................... 73
Chapter 6: Future Work .................................................................................................... 75
6.1 Collagen .................................................................................................................. 75
6.2 Mechanical Testing ................................................................................................. 75
6.3 Cell and Clinical Studies ......................................................................................... 75
References and Works Cited ............................................................................................. 76
Appendices ........................................................................................................................ 84
Appendix A: Equations used to calculate the optimal amount of DVS to be used ....... 84
Appendix B: Previous Attempts to Electrospin the AA solvent system ....................... 86
vii
List of Tables
Table 1: Scaffolds that have been developed for tissue engineering applications for
human cartilage. Compiled from (19). ................................................................................ 8
Table 2: Mass and weight percent of ChS used. All other masses were held constant. .. 30
Table 3: The recorded electrospinning parameters and variables for each sample. The
accelerating voltage, distance between the syringe tip and collector plate was recorded,
pump rate, and electric field were recorded for all 13 samples. ....................................... 33
Table 4: Summary of the AA solvent system solutions that were spun and the
atmospheric conditions on that day. .................................................................................. 34
Table 5: Summary of the pH of the DMF solutions examined ......................................... 42
Table 6: Key peaks for both ChS and HA. ....................................................................... 44
Table 7: Summary table of SEMs and Histograms of fiber diameters for the DMF
optimization study. ............................................................................................................ 47
Table 8: A summary table of the fiber diameter statistical data from Figure 24, the
statistical sample size was 50 except for Sample 9 which had a sample size of 10. ........ 54
Table 9: The pH for each different wt% of ChS. .............................................................. 56
Table 10: Shared peaks between ChS and HA. ................................................................ 58
Table 11: Summary table of SEM images and histograms. .............................................. 61
Table 12: Summary data for the fiber diameters for the AA solvent system. Same data as
found in Figure 32. ............................................................................................................ 63
Table 13: The pH of the DVS crosslinker solutions. ........................................................ 67
Table 14: IR Peaks for Figure 38. ..................................................................................... 71
Table 15: Results of the solubility test. ............................................................................. 72
Table 16: Results of Eq.1. ................................................................................................. 84
Table 17: Results of the solver tool. As can be seen, the mass of each chemical correctly
correspond to the moles used for each sample. ................................................................. 85
Table 18: The same procedure used for the DMF solvent system was used. The only
solution that worked was at 15 kV, highlighted in yellow. The fiber diameter is recorded
for that sample in the far right column.............................................................................. 86
viii
List of Figures
Figure 1: Material components of human meniscus. Cartilage has a very similiar
composition. Image from (12). .......................................................................................... 4
Figure 2: Techniques for cartilage tissue engineering. Image taken from (19). ................ 7
Figure 3: The location of articular cartilage in the knee. The most prominent area is seen
in this image. The white area on the femur is an area of articular cartilage. Taken from
(11). ..................................................................................................................................... 9
Figure 4: Sketch of the layers of hyaline cartilage. The image is arranged such that the
superficial zone is at the top, the middle zone is in the middle, and the deep zone is on the
bottom. The image on the left containing the circular shapes indicates the orientation of
the cells. The middle image shows the alignment of the collagen fibers. The right image
is a false color image of human hyaline cartilage. Image taken from (1). ....................... 10
Figure 5: Image of the meniscus. The colored areas refer to the amount of
vascularization in the tissue. The reddest are the most vascularized, while the white areas
are avascularized. Image taken from (12). ....................................................................... 11
Figure 6: Arrangement of the fiber orientations within the meniscus, and their location.
Image taken from (12). ..................................................................................................... 12
Figure 7: Image of chondroitin-4-sulfate. ......................................................................... 15
Figure 8: Structure of aggrecan. This image contains keratin sulfate in addition to ChS.
The “G” numbers refer to globular domains of the protein. Taken from (6). .................. 16
Figure 9: Structure of hyaluronic acid. ............................................................................. 17
Figure 10: Structures of dimethylformamide (right) and acetic acid (left). ...................... 18
Figure 11: Mechanism of reaction for DVS. .................................................................... 20
Figure 12: Basic electrospinning set-ups. Image from (54). ............................................ 21
Figure 13: Common electrospinning parameters (58). ..................................................... 23
Figure 14: The effects of voltage on fiber morphology. Image taken from (64). ............ 26
Figure 15: Photograph of the electrospinning set-up that was used in this experiment. On
the lower left is the syringe pump. Loaded on to the syringe pump is a 10 mL syringe
with a 16 gauge needle attached. The needle is connected the high voltage supply (top
right). ................................................................................................................................. 32
Figure 16: The pH paper used in this experiment. ............................................................ 35
ix
Figure 17: The instruments used to measure viscosity. Top) The UGV, Bottom) the
rheometer. ......................................................................................................................... 37
Figure 18: Image of the AT-FTIR that was used in this study. ........................................ 38
Figure 19: Image of the type of SEM that was used in this experiment (75). .................. 39
Figure 20: Image of the reflectance spectrometer set-up used to characterize the solubility
of the samples. .................................................................................................................. 41
Figure 21: Viscosity of the DMF solvent systems with each polymer. ............................ 43
Figure 22: IR spectra of ChS+HA fibers for the DMF solvent system as compared to the
bulk polymers ChS and HA. ............................................................................................. 45
Figure 23: IR spectra for the DMF solvent system as compared to the bulk polymers;
focused from 2500 cm-1
to 600 cm-1
. ................................................................................ 46
Figure 24: Summary of all fiber diameters. Purple indicates no NaPO4, Yellow indicates
a constant field, Blue represents a change in voltage, Red indicates a change in pump
rate, and Green represents a change in distance. .............................................................. 51
Figure 25: Graph comparing the effect of voltage on fiber diameter. .............................. 51
Figure 26: Graph comparing the effects of distance on fiber diameter. ........................... 52
Figure 27: Graph comparing the effects of the electric field on the fiber diameters. ....... 52
Figure 28: Graph comparing the effects of pump rate on fiber diameter. ........................ 53
Figure 29: Graph showing the viscosity of the lower weight percents for the AA solvent
system. Please note the mass of HA was held constant throughout. ............................... 57
Figure 30: IR of the bulk ChS, bulk HA, and 30% mass ChS mat. .................................. 59
Figure 31: IR spectra for the range of 2500-600 showing the key peaks for the polymers.
........................................................................................................................................... 60
Figure 32: Summary of fiber diameters for the AA solvent system. ................................ 64
Figure 33: Fiber diameters for the samples demonstrate a linear trend in conjunction with
an increase of ChS. ........................................................................................................... 64
Figure 34: EDS scan of fibers found in the 30% mass ChS sample with background
elements removed. ............................................................................................................ 66
Figure 35: Image used in the EDX scan. Image is another view of the 30 wt% ChS
solution. ............................................................................................................................. 66
x
Figure 36: SEM image of the fibers produced from this solution (top) and a histogram
with table containing the statistical analysis of the fiber diameters (bottom). .................. 68
Figure 37: SEM image of the fibers produced from this solution (top) and a histogram
with table containing the statistical analysis of the fiber diameters (bottom). .................. 69
Figure 38: IR spectra for the excess HA fibers. Before and after crosslinking are shown.
........................................................................................................................................... 71
Figure 39: Screen shot of the solver tool in use. ............................................................... 85
xi
Abstract
Electrospun Cartilage Glycosaminoglycans for Biomimetic Scaffolds
Sean Michael Dowd
Caroline L. Schauer, Ph.D.
Osteoarthritis is a condition through which cartilage is broken down by the human
body (1). Osteoarthritis of the knee affects approximately 84.5 million people over the
age of 45 in the US alone, or over 27% of the entire population (1). In order to help
combat such a debilitating condition, this study investigated electrospinning natural
polymers that are found in human cartilage: hyaluronic acid and chondroitin sulfate.
Polymer fibers were created using two different solvent systems: aqueous N,N -
dimethylformamide and acetic acid. Optimization tests were performed on each sample to
attempt to control factors such as fiber diameter and polymer concentration. In order to
understand which system would be better for the human body, crosslinking and solubility
studies were performed. When attempting to spin, it was discovered that only the
solution containing aqueous N,N – dimethylformamide and an excess of HA could be
successfully be electrospun and crosslinked.
1
Chapter 1: Introduction
1.1 Materials
1.1.1 Overview of Glycosaminoglycans
Glycosaminoglycans (GAGs), linear polymers consisting of two repeating
disaccharide units, are an important part of human physiology (2). These units are
always consist of an amino-sugar and uronic acid. Occasionally, a galactose unit will be
seen in place of one of the other two (2). In mammals, virtually every cell produces at
least one GAG in order to provide support to the extra-cellular matrix (ECM) allowing
the cells to hold together (2). To use an analogy to everyday life, GAGs are the glue that
holds the body’s cells together.
Both human cartilage and meniscus have large quantities of two GAGs:
chondroitin sulfate (ChS) and hyaluronic acid, also known as hyaluronan or hyaluronate
(HA) (1, 3) . These GAGs are essential parts of the cartilage and meniscus. Specifically,
ChS and HA are responsible for most of the chemical bonding that takes place within the
tissue (1, 4–6).
1.1.2 Challenges of Using Glycosaminoglycans
As GAGs are a form of biopolymers, there are several complications that must be
dealt with in order for them to be used. For example, GAGs, and other biopolymers can
have different molecular weights (MW), purity, charge distribution, and even structure
depending on the source (7, 8). A good example of this is chondroitin sulfate.
Although more detail about chondroitin sulfate will be discussed later, it is a
perfect example of how structural differences can change the behavior of the polymer.
Chondroitin sulfate can exist in two structural forms: chondroitin-4-sulfate or
2
chondroitin-6- sulfate (7, 9, 10). The only difference between the two is where the
sulfate is located (7, 9, 10). However, this change means the loss of a primary hydroxyl
group which can make chemical modification difficult.
1.2 Applications
One of the primary applications for GAGs involves the repair of the major soft
tissues of the human knee. In order to fully understand these applications, however, a
basic overview of human anatomy is required.
1.2.1 Epidemiology of Knee Cartilage and Meniscus
Around the world, both cartilage and meniscus injuries are a huge problem.
Many of these injuries are sports related, but not all. What is particularly alarming is that
a problem in either the meniscus, or the cartilage, can lead to damage of the other tissue
as well.
Meniscus injuries involve a large percentage of patients. In the United States,
approximately 200,000 people suffer from meniscus injury (11). Most of these are sports
related injuries (11). Of these, 92% require a meniscectomy (11). A meniscectomy is a
surgical procedure where the patients’ own meniscus is removed (11, 12). Furthermore,
over many years a patient that had their meniscus removed will suffer from articular
cartilage degradation as well at an average rate of 4% per year (1, 11, 12).
To compound this problem is osteoarthritis (OA). OA is a condition through
which cartilage tissue (regardless of type) is broken down by the body (1). OA of the
knee affects approximately 84.5 million people over the age of 45 in the US alone, or
over 27% of the entire population (1). What are more horrifying are the age
demographics. Of the US population, 1/5 people over the age of 45 will suffer from OA,
3
as compared to 1/2 of the population over 65 that will have OA (1). With statistics like
these, it is clear that some sort of biomimetic material or regenerative template is needed.
1.2.2 Cartilage Tissue Degradation
Cartilage due to OA is due to several issues associated with the tissue and its
environment. The most important factor of cartilage loss due to OA is age (13). The age
of the tissue allows for degradation from mechanical loading and water loss in the tissue
(13). As the tissues age, the collagen and a proteoglycan named aggrecan (see Chapter
2.1.1) in the tissue does not retain as much water as before (13, 14). See Figure 1 for
the material components of human cartilage and meniscus tissue. This change in water
retention is due to changes in molecular weight, concentration, and structure that will
lead to a change in the ECM and result in lowered mechanical strength of the tissue (1,
13, 14). This decrease in mechanical strength and stability could allow for the tissue to
become permanently damaged if too much stress is placed on the joint (13). This will
lead to a cascade effect in which more water is lost and the mechanical strength is
decreased even more until physiologically normal stresses can cause the tissue to
mechanically fail (13).
4
Figure 1: Material components of human meniscus. Cartilage has a very similiar
composition. Image from (12).
1.3 Current Non-Tissue Engineered Treatments for OA
As of the time of this writing, there are several clinically available or FDA
approved methods for the treatment of OA. As with all medical procedures, some
treatments are for the early stages of OA, while others are intended for late-stage OA.
Three of the major treatment options will be discussed here. They are the injection of
GAGs into the intra-articular cavity, a meniscectomy, and the use of an on the market
biomimetic material.
One of the more common treatments on the market is the direct injection of HA
into the OA site. This treatment works on the principle that since HA is a natural
component of the cartilage in this area, a direct injection should help to rejuvenate the
amount of HA that is available for the cartilage to use to heal itself (15, 16). As one
author noted:
5
“HA has a variety of effects on cells in vitro that may relate to its reported
effects on joint disease. For example, it inhibits prostaglandin E2 synthesis
induced by interleukin-1 (IL-1); protects against proteoglycan depletion and
cytotoxicity induced by oxygen-derived free radicals, IL-1, and mononuclear cell–
conditioned medium; and affects leukocyte adherence, proliferation, migration,
and phagocytosis. HA has been shown to have a direct effect on control
mechanisms of cell activation in monocytes, e.g., it increases messenger RNA
expression for IL-1b, tumor necrosis factor a (TNFa), and insulin-like growth
factor 1 (IGF-1). Monoclonal antibodies against the HA receptor, CD44, block
the effect of HA on expression of IL-1b, TNFa, and IGF-1, indicating direct
interaction of HA with the cell. In cartilage, HA has been shown to suppress
cartilage matrix degradation by fibronectin fragments (15).”
Additionally, since the treatment was made clinically available, it has been discovered
that direct injections of HA may help to inhibit the progress of OA all together (15).
This treatment has been in use for several decades, but there are still numerous
issues with this treatment. To begin, many studies that compared the injection of HA to
the injection of a placebo have shown that there is little statistical difference between the
two (15, 16). Furthermore, there is still little evidence that the treatment is effective in
humans (15, 16). Both of these issues are huge weaknesses that must be addressed.
One of the oldest, and still most common, treatments for OA is a meniscectomy.
This is the surgical removal of the meniscus tissue (11, 12). A meniscectomy is
performed when a surgeon has determined that the meniscus has taken too much damage,
and that the removal of the meniscus is the only way to help alleviate some pain (11, 12).
6
This treatment has existed for nearly a century, but is beginning to die out (12). As early
as 1948, it was discovered that the complete removal of the meniscus would lead to
adhervse, and possibly worse conditions in the patient’s knee (12). As a result, most
surgeons have begun to move away from this treatment in favor only partially removed
the meniscus tissue or attempting to repair the tissue (11, 12).
Needless to say, the complete removal of a patient’s meniscus is a completely
unacceptable treatment. The meniscus is an essential component of the knee, and it must
be preserved as much as possible. Without it, the knee joint would not be able to
withstand the compressive forces that it experiences on a daily basis.
Finally, a relatively new treatment that was released on the market that may yet
alter the way meniscus and cartilage conditions are treated in the knee. The product
name is Menaflex®. Menaflex® was a collagen based implant that was either a
biomimetic that served in place of a functional meniscus, or a platform for tissue
regeneration (17, 18). This ambiguity is exactly what caused the US Food and Drug
Administration (FDA) to rescind its previous clearance for the device, and ban its use in
the US (17, 18). As a result, the company that sold the product went bankrupt and,
unfortunately, little data is available in regard to Menaflex® aside from information listed
in the legal briefs surrounding the lawsuits concerning the FDA’s actions (17, 18).
1.4 Tissue Engineered Treatments
One of the most heavily researched areas for the treatment of OA is tissue
engineering. Tissue engineering is the development of a bio-compatible scaffold that is
loaded with the appropriate cells types for the target tissue (19). The cartilage types in
the knee are ideal for tissue engineered devices because they have poor natural
7
regenerative ability (19). See Figure 2 for a basic schematic for tissue engineered
cartilage.
Figure 2: Techniques for cartilage tissue engineering. Image taken from (19).
The tissue engineering approach to treating OA is a comparatively new
development. As mentioned above, tissue engineering requires the development of a
biocompatible scaffold in addition to the correct cells types for the targeted tissue. As
can be seen in Table 1, several materials have already been evaluated for use for tissue
engineered cartilage, included both HA and ChS.
8
Table 1: Scaffolds that have been developed for tissue engineering applications for
human cartilage. Compiled from (19).
Scaffolds Hydrogel Available
Proteic scaffolds
Collagen Yes
Gelatin No
Fibrin Yes
Polysaccharidic scaffolds
Alginate Yes
Cellulose Yes
Chitosan Yes
Hyaluronic Acid Yes
Chondroitin Sulfate Yes
Artificial scaffolds
Carbon fiber No
Calcium phosphate No
Dacron® No
Polyethylmethacrylate Yes
Polyglycolic acid Yes
Polylactic acid No
Teflon ® No
9
1.5 Overview of Knee Cartilage and Meniscus Anatomy
1.5.1 Articular Cartilage
Articular cartilage is defined as the type of hyaline cartilage that is present on the
ends of all of the bones in the knee (see Figure 3) (1). This tissue provides cushioning for
the bone from mechanical forces (1).
Figure 3: The location of articular cartilage in the knee. The most prominent area is
seen in this image. The white area on the femur is an area of articular cartilage. Taken
from (11).
Hyaline cartilage has very unique properties due to its structure. The structure is
comprised of three layers of organization: the superficial zone, the middle zone, and the
10
deep zone (1). The superficial zone is comprised of collage fibers that are aligned
parallel to the surface of the cartilage (1). This zone occupies approximately 10-20% of
the total tissue (1). The middle zone occupies 40-60% of the tissue is is comprised of
collagen fibers in a random formation (1). Finally, the deep zone consists of the
remaining tissue and contains fibers that are parallel to the surface of the tissue (1). This
behavior can be seen in Figure 4. What really affects the mechanical properties of the
hyaline cartilage is the collagen. The collagen types that are present are type II (over
50% dry weight of the tissue),V, VI, IX, and XI (1). These collagen types have
extremely high mechanically strength (1).
Figure 4: Sketch of the layers of hyaline cartilage. The image is arranged such that the
superficial zone is at the top, the middle zone is in the middle, and the deep zone is on the
bottom. The image on the left containing the circular shapes indicates the orientation of
the cells. The middle image shows the alignment of the collagen fibers. The right image
is a false color image of human hyaline cartilage. Image taken from (1).
11
1.5.2 Meniscus
The meniscus is a U-shaped tissue that is located between the femur and tibia (see
Figure 5) (3, 11, 12). The meniscus provides balance in the knee and transfer mechanical
loads in the joint (3, 11, 12). There are two menisci in each knee in the medial and lateral
directions (3, 11, 12). Given its mechanical function, the meniscus must be able to
withstand many mechanical forces.
Figure 5: Image of the meniscus. The colored areas refer to the amount of
vascularization in the tissue. The reddest are the most vascularized, while the white
areas are avascularized. Image taken from (12).
Not surprisingly, meniscus tissue is not very different from the articular cartilage.
In fact, the meniscus is a a type of cartilage known as fibrocartilage (3, 11).
Fibrocartilage has similar properties to hyaline, but it has a different structure. As the
name suggests, fibrocartilage consists of a fibrous structure (12). Due to its similarity to
hyaline cartilage, the meniscus can support a lot of force.
12
The mechanical properties of the meniscus are also due to its structure and
materials. The meniscus is composed of three layers: superficial, lamellar, and deep (3,
11, 12). Collagen fibers in the superficial and lamellar regions are randomly orientated
(3, 12). The deep zone of the meniscus is comprised of ordered fibers that run parallel to
the surface of the meniscus (3, 12). See Figure 6 for a pictorial example of the fiber
orientation and locations. The top two layers of the meniscus are comprised of elastic like
collagen type I, while the deep zone is comprised of mostly collagen type II (12). This
mix of flexible and rigid collagen types explain how the menisci can handle both the
weight of the individual, and endure all the physical activity that the individual performs.
Figure 6: Arrangement of the fiber orientations within the meniscus, and their location.
Image taken from (12).
13
1.6 Overall Goal and Objectives
1.6.1 Overall Goal
The overall goal of this study was to electrospin chondroitin sulfate and
hyaluronic acid in such a manner that the natural anatomy of cartilage tissue was
mimicked. This will lay the groundwork for a device that could be used in an
implantable device to replace either articular cartilage or meniscus.
For articular cartilage, the device will have varying dimensions. This is due to the
amount of degradation that can be variable from patient to patient (1). However, one
constant is that the thickness of the mimicked articular cartilage should be 2.26 ± 0.49
mm (20).
For the meniscus tissue, the dimensions are more quantifiable. The dimensions
do vary depending on which of the two meniscus locations is being targeted, the sex of
the patient, the age, and the activity level (12). Targeted dimensions for a device for the
medial meniscus should be 40.5 – 45.5 mm in length, and 27 mm in width (12). For the
lateral meniscus, length should be 32.4 – 35.7 mm and the width should be 26.6 – 29.3
mm (12). Furthermore, the circumferential dimension for the medial meniscus should be
approximately 90 – 110 mm, but the circumferential dimension for the lateral meniscus it
should be 80 – 100 mm (12).
1.6.2 Objectives
The objectives of this study are three-fold. First, an optimization study of an
aqueous N,N - dimethylformamide system will be performed (see Chapter 2.1.3). This
study was based on work from previous researchers (21, 22). Secondly, the use of a
solvent system that is considered to be safer to the human body, acetic acid, will be will
be used to create electrospun fiber. Finally, the chemical stability of the aforementioned
14
electrospun mats will be examined through the crosslinking of the mats and solubility
testing.
In short, it is believed that the electrospun and crosslinked GAGs will form a
biomimetic or regeneration template that can be used by orthopedic surgeons.
Furthermore, this device should be simple enough that future researchers will be able to
use this device as a starting point for the attachment of collagen and growth factors to
encourage tissue regeneration.
15
Chapter 2: Background
2.1 Materials
2.1.1 Chondroitin Sulfate
ChS is the most important GAG in both the cartilage and the meniscus. There are
two types of ChS found in these tissues; they are chondroitin-4-sulfate and chondroitin-6-
sulfate (see Figure 7 for structure) (1, 6, 9). Both forms play a huge role within aggrecan
in cartilage (1, 6, 23). The molecular weight of chondroitin sulfate is 50,000 (24).
Figure 7: Image of chondroitin-4-sulfate.
16
ChS is part of a proteoglycan called aggrecan (1, 6, 9, 23). Aggrecan, like most
proteoglycans, is comprised of two parts: one part is a protein, and the other is a
polysaccride (see Figure 8 for structure), and its main role is to act as a structural protein
(1, 6). Over 90% of aggrecan’s mass is comprised of chondroitin sulfate and its close
relative keratin sulfate (6). Aggrecan is essentially a brittle-brush style polymer, with the
ChS acting as the brush (6). There can be up to 100 individual ChS chains on each
aggrecan molecule (20 kDa each) (6). The ChS mainly acts an attachment site. Collagen
and HA attach to the ChS, and this allows the aggrecan to perform its role as a structural
proteoglycan. More importantly, aggrecan can absorb up to 50 times its own weight in
water gives the cartilage tissue mechanical stability (6, 14).
Figure 8: Structure of aggrecan. This image contains keratin sulfate in addition to ChS.
The “G” numbers refer to globular domains of the protein. Taken from (6).
As briefly described earlier, aggrecan plays a key role in OA. As previously
mentioned, water loss is one of the key factors of OA (13). As the body ages, aggrecan
17
will undergo changes in its molecular weight due to the native cells not being able to
produce the same quality and quantity aggrecan as before (1, 13, 14). Since the aggrecan
changes, the amount of water that aggrecan can hold decreases (6, 13, 14). This in turn
leads to the progression of OA (13).
2.1.2 Hyaluronic Acid
Figure 9: Structure of hyaluronic acid.
Hyaluronic acid (HA) is another essential GAG (Figure 9). Unlike ChS, HA acts
as an individual molecule, and not part of a proteoglycan. Its main role and structure are
very similar to ChS. In hyaline cartilage, HA acts as a structural component connecting
the aggrecan and collagen together (4, 6, 25). Additionally, HA is responsible for cellular
18
signaling and wound healing in this tissue (4, 21, 25). Needless to say, it is not a part of
an essential proteoglycan, but it is essential for cartilage tissue.
2.1.3 Solvents
Figure 10: Structures of dimethylformamide (right) and acetic acid (left).
GAGs such as HA and ChS are fairly easy to solubilize. Both ChS and HA are
extremely hydrophilic and will dissolve in water rapidly (5, 21, 25). However, water is
not the only solvent that can dissolve these GAGs. Many aqueous organic solvents can
also dissolve ChS and HA. Several studies have shown that both HA and ChS can be
dissolved using a combination of distilled water and N,N - dimethylformamide (DMF)
(21, 22, 26–30). DMF can be toxic to the environment, but a study has shown that the
DMF can evaporate out of electrospun fibers (see Chapter 2.2) (31). Furthermore, both
ChS and HA have been shown to be able to dissolve in acetic acid (AA) (32, 33). AA
can also be toxic, but it has a high vapor pressure and will evaporate at room temperature
19
and low pressure (34–36). The chemical structures for both DMF and AA can be seen in
Figure 10.
2.1.4 Crosslinking
Crosslinking is the process of chemically bonding at least two polymer chains
together (37). For GAGs and similar biopolymers, there are numerous crosslinking
agents that could be used. These include, but are not limited to genipin, epichlorohydrin,
glutaraldehyde, and hexamethylene-1,6-diaminocarboxysulphonate (38–44). Ideally, a
crosslinker should be water soluble, nontoxic, and cost-effective (45). One crosslinking
agent that comes comparatively close to this description is divinyl sulfone (DVS). DVS
has been used many times in the past to crosslink polymers that contain primary hydroxyl
groups (45–47). In addition to the above qualities, one upside to using DVS is that its
unique sulfone group can be detected using sulfur microanalysis (45). In short, compared
to other crosslinkers, DVS is very easy to identify.
The mechanism of reaction for DVS is simple. DVS crosslinks via a modified
form of a 1,4 addition, also known as a Michael addition (45). In the most general form,
a Michael addition is a reaction in which a carbanion undergoes a conjugate addition to a
carbonyl compound (48). The general form, however, is not what describes the DVS
crosslinking reaction. As can be seen in Figure 11, the vinyl groups on the DVS
molecule prefer to attack primary hydroxyl groups in alkaline environments (45).
20
Figure 11: Mechanism of reaction for DVS.
2.2 Electrospinning
Electrospinning was chosen as the primary processing technique for the proposed
device. This was chosen for several reasons. The first is that electrospinning can
produce fibrous structures similar to natural cartilage tissue, and produce the fibers in a
low-cost way (21, 49, 50). Secondly, electrospinning can produce fibers that are an ideal
21
diameter for a biomimetic device. Specifically, a fiber diameter of 0.5 – 2 μm would be
ideal for such an application (51–53).
2.2.1 Overview of Electrospinning
Electrospinning was first developed in the early part of the twentieth century, with
the first US patent being awarded in 1934 to Anton Formhals (54, 55). Electrospinning is
a process where electrostatic forces are applied to a polymeric solution or melt for the
fabrication of nanofibers (fibers with diameters less than 100 nm) (54).
This simple process is flexible, scalable and cost-effective (54). The basic
components of an electrospinning set-up are a high-voltage electrical supply, an infusion
pump, a collector, and of course a container for the polymeric solution with a thin
capillary to serve as a spinneret (8, 54, 56). An example of how a typical electrospinning
set-up appears can be seen in Figure 12.
Figure 12: Basic electrospinning set-ups. Image from (54).
The process is fairly simple. First, the polymer solution or melt is loaded into a
syringe pump with a small capillary at one end (54, 56, 57). Next, an electrical charge is
22
applied to the solution via an electrode that is connected to the voltage supply (54, 56,
57). At the same time, the collector (usually a flat plate) is charged, but with the opposite
charge of the solution (54, 56, 57). This collector is placed at a pre-determined distance
from the edge of the capillary (54, 56, 57). As both the polymer and collection plate are
oppositely charged, and since only the polymer is capable of movement, it is the natural
tendency of the now electrically charged polymer to attempt to complete the created
electric circuit (54, 56, 57). The charged polymer will attempt to travel toward the plate,
and as a result, the polymer cone at the end of the capillary form a Taylor cone from
which jets of polymer will begin to eject from tip of the capillary (54, 56, 57).
The jets will spin in the air until they have reached the plate (54). All the while,
the polymer jet will continue to become thinner and thinner due to the increased strain on
the jet, and the solvents that were in solution with the polymer will evaporate off (54).
When the polymer jets finally land on the plate they are extremely small in diameter (54).
2.2.2 Parameters for Electrospinning
As with most manufactured items, there are several parameters that must be met
in order to use the process effectively. As can be seen in Figure 13, these parameters can
be broadly broken up into three different categories. For the purposes of this study,
however, the parameters will be re-categorized. Specifically, the parameters of the
solution, the electrospinning machinery, and environmental conditions will be examined.
23
Figure 13: Common electrospinning parameters (58).
2.2.2.1 Solution Parameters
There are several properties of the solution that can affect the quality and quantity
of fibers produced. Several of these properties can be adjusted, but most should be taken
into consideration before attempting a spin. They are viscosity, conductivity, and
concentration.
Viscosity, and closely related elasticity, is one of the most important
electrospinning parameters. In terms of solution properties, it is the viscosity of the
solution that determines whether or not a jet will form off of the droplet at the end of the
capillary (8, 56, 59). Specifically, the viscosity parameter affects the critical
entanglement, or concentration, that will determine if a solution will electrospin
successfully (60, 61). Critical entanglement is the concentration where the polymer
chains in solution begin to overlap each other and become more entwined (60–62). As a
result, this entanglement will cause a bending instability in the fibers, and this is what
causes the polymer jet to whip around during the electrospinning process (60, 61).
24
Needless to say, the viscosity is an essential property to understand. The type of
viscosity that must be known is the relative viscosity. This is because the relative
viscosity that determines the fiber diameter and the morphology of electrospun fibers
(63). If the viscosity is too high then the solution will not be able to spin (56).
Interestingly enough, if the viscosity is too low then it still will not spin (56). Rather, a
balance must be reached in order to optimize fiber production for a solution to spin.
Unfortunately, given that every polymer solution is different, there are no set guidelines
or recommendations in terms of desired viscosity for electrospinning
Another parameter of note is the electrical conductivity of the solution. As the
name “electrospinning” implies, the solution must be able to carry an electric current if it
is to spin successfully (54–57). However, much like viscosity, there is no real guideline
for what is to be considered a “good” conductivity reading for electrospinning a solution
(56). What is certain is that the solution must be conductive, or else electrospinning will
not occur.
Finally, the concentration of a solution can affect how well a solution will spin.
Concentration generally ties in closely with viscosity, in that an increased concentration
usually means an increased viscosity (56, 57, 59). This is not always the case however.
Several studies have shown that the concentration of the individual components of a
solution can be modified and have various effects on the viscosity (56, 57). More
specifically, there appears to be an inverse relationship between concetration and fiber
diameter size. One study showed that a decrease in concentration led to an increase in
fiber diameters (57). This study was performed while examining the effects of viscosity,
but instead the researchers stumbled across an interesting behavior concerning the
25
concentration. This further implies the underlying importance of taking concentration
into consideration when attempting to electrospin a solution.
2.2.2.2 Electrospinning Parameters
It comes as no surprise that the electrospinning equipment can affect how well a
solution will spin, or if it will spin at all. As alluded to previously, there are three
parameters that can be changed easily by just adjusting the various components of the
electrospinning set-up. Those parameters are target distance, accelerating voltage, and
pump rate.
The distance between the capillary and collector is an important part of
electrospinning. Distances in the centimeter to millimeter range usually yield more
oriented fibers (56). However, once again there are no set guidelines for how close a
sample should be (56). Furthermore, the distance can actually be determined by the
accelerating voltage because a higher voltage at a close distance could result in electrical
damage to the sample (56).
As it has been made clear, the accelerating voltage is one of the most important
parts that can be controlled via the electrospinning set-up. The accelerating voltage
directly effects the fluid flow from the capillary (56, 64). As one researcher noted, “the
changes in the applied voltage will be reflected on the shape of the suspending droplet at
the nozzle of the spinneret, its surface charge, dripping rate, velocity of the flowing fluid,
and hence on the structural morphology of electrospun fibers (64).” The same
researchers, however, did notice that the accelerating voltage can affect the morphology
of the fibers that were formed. This can be seen in Figure 14.
26
Figure 14: The effects of voltage on fiber morphology. Image taken from (64).
Lastly, the pump rate can also affect the electrospinning process. Generally, the
pump rate can determine the quality of the fibers that are formed (56, 57, 65). The pump
rate, more commonly referred to as the feed rate, affects the quality by altering the
droplet that forms at the end of the capillary. This is respect, it is similar to viscosity, but
the pump rate does not usually affect if fibers are formed or not. It usually affects only
the morphology, but there are several examples where this is not the case (56, 57, 65).
27
2.2.2.3 Environmental Parameters
Not surprisingly, the environment in which a solution is spun can affect the entire
electrospinning process. Of all the parameters discussed, these can be some of the
hardest to control or alleviate the effects of. The primary environmental factors that
affect electrospinning are temperature and humidity.
Temperature affects several other parameters in the electrospining process. The
temperature mostly affects if a reaction will take place within the spinning solution, the
vapor pressure of the solvents in the solution, and also the degradation of any temperature
sensitive polymers (56, 57). Obviously, temperature plays a huge role in determining
which polymers and solvents must be used, and how they should be used. Without it, the
solution could evaporate before it makes contact with the collector.
Finally, the humidity of the environment plays a dramatic role in the
electrospining process. The humidity can affect anything from the fiber morphology, the
pore size of the fibers, or if the fibers (56). The exact mechanisms behind how humidity
generates such varying effects are still unknown, and are still being investigated.
2.3 Applications of Electrospun Fibers and Membranes
Needless to say, electrospun fibers have a large amount of possible applications.
Electrospun fibers and membranes can be used in filtration, clothing, drug delivery,
biomaterials, tissue engineering, nanotubes, and nanoelectronics (66–72). Within the
scope of this study, focus was placed on biomaterials and tissue engineering.
Electrospinning has been used in the field of biomaterials for many years. One of
the natural advtanges of this process is that the fibrous membrane naturally resembles the
ECM of most tissues in both size and mechanical properties (67). As a result,
28
electrospun fibers and mats have become fairly popular, but the most common
application is in the overlapping field of tissue engineering.
Tissue engineering has used electrospinning for almost as long as biomaterials,
and for the same reason (67). Electrospun mats and membranes have seen use in
attempts to tissue engineer bone, nerve cells, and many other tissues (67, 69, 73). The
reason why electrospinning was used was because the mats offered a way for the cells to
attach to a mimicked environment and behave physiologically normal (67, 69, 73).
29
Chapter 3: Experimental Design and Procedure
3.1 Materials
The basic materials that were used in this study were:
Solids:
Chondroitin sulfate sodium salt, 90+% purity (Indofine Chemical Company,
Hillsborough, NJ, 5 gram container)
Hyaluronic acid sodium salt (Dali Hyaluronic Acid Company, Liuzhou City,
Guangxi, China, 1000 gram package)
Disodium phosphate, Anhydrous, ≥99.999% (metals basis), (Fluka, St. Louis,
MO, 100 gram container)
Liquids:
N,N – Dimethylformamide, Anhydrous, 99.8% (Sigma-Aldrich, St. Louis, MO, 2
Liter bottle)
Acetic acid 99.7% (Sigma-Aldrich, St. Louis, MO, 2.5 Liter bottle)
Deionized water (Milli-pore Filtration System, Ultra-pure water filter, water
resistivity: 18.2Ω at 25°C)
Divinyl sulfone 97% (Sigma-Aldrich, St. Louis, MO, 25 gram container)
3.1.1 Dimethylformamide Solvent System
The first solvent system to be used during this study contained N,N - dimethyl
formaldehyde (DMF) as one of the primary solvents for ChS and HA. The procedure for
creating this solution was taken and modified from a previous study (21, 22). First 0.6 g
of HA sodium, 0.2 g ChS sodium salt, and 0.6 g of disodium phosphate were mixed
together in a glass beaker with a glass stirring rod. Next, the solids were mixed in a
30
20mL volume of 1:1 DMF and distilled water. This solution was stirred with a glass
stirring rod until the polymer completely dissolved into solution.
3.1.2 Acetic Acid Solvent System
The acetic acid (AA) solvent system was made in a similar fashion to the DMF
system. For each sample, 0.3 g of HA was used and a varying amount of ChS was used.
Unlike the DMF samples, the materials in the spun solution were altered depending on
what was to be examined. Specifically, the mass of the ChS was varied for each
experiment. This data can be seen in Table 2. Once the polymers were thoroughly mixed
with a glass stirring rod, the polymers were then added to a 6 mL, 1:5 solution of AA:
distilled water. Once again, the solution was stirred with a glass stirring rod until the
polymer completely dissolved into solution.
Table 2: Mass and weight percent of ChS used. All other masses were held constant.
Mass of ChS (g) Weight Percent (M/M) of ChS
0.33 5
0.69 10
1.10 15
1.56 20
2.68 30
3.1.3 Divinyl Sulfone Crosslinking
DVS crosslinking was attempted on both the DMF and AA solvent systems in
order to compare them on a one-for-one basis. Such a comparison would determine
31
which solution would be the best for a biomimetic type device. A low concentration for
both polymers was chosen so as to be more economical. Calculations were performed to
determine the ideal amounts of polymer and DVS that would be needed using the number
of moles of each substance. It was shown, however, that given the different molecular
weights of the polymers, mathematically there would always be an excess of one
polymer. For solutions with an excess of ChS the amount of moles used was 4.30x10-4
moles, and for solutions with an excess of HA the amount of moles used was 5.30x10-4
moles. These calculations can be seen in Appendix A. Both solvent systems were mixed
in the same ways as previously, but DVS was added simultaneously with the aqueous
solvents attempting to electrospin. Once again, the solution was stirred with a glass
stirring rod until the polymer completely dissolved into solution.
3.2 Electrospinning
The electrospinning apparatus consisted of a syringe pump (Harvard Apparatus,
70-2209, Holliston, MA), a high voltage power supply (Gamma High Voltage Research,
ES60P-10W, Ormond Beach, FL), a 10mL syringe (Becton, Dickinson and Company ,
REF309604, Franklin Lakes, NJ), and a 21 or 16 guage Luer-lock needle (both Becton,
Dickinson and Company , 305165 and 305198 respectively, Franklin Lakes, NJ). An
image of the apparatus can be seen in Figure 15.
32
Figure 15: Photograph of the electrospinning set-up that was used in this experiment.
On the lower left is the syringe pump. Loaded on to the syringe pump is a 10 mL syringe
with a 16 gauge needle attached. The needle is connected the high voltage supply (top
right).
3.2.1 Dimethylformamide Solvent System
The polymer solution was electrospun with several parameters being studied for
each trial. The control for this study was based on the previous work from (21, 22).
Specifically, the control solution was electrospun at a distance of 7 cm, with an
accelerating voltage of 17 kV, and a pump rate of 5 μL/min. All samples were allowed to
spin for 15 minutes. Table 3 shows the variables that were changed for each sample in
each trial. Please note that both the temperature, and relative humidity in the
33
environment during electrospinning were recorded using a digital thermohygrometer
(Cole-Palmer, RH520A, Vernon Hills, IL).
Table 3: The recorded electrospinning parameters and variables for each sample. The
accelerating voltage, distance between the syringe tip and collector plate was recorded,
pump rate, and electric field were recorded for all 13 samples.
Sample Conditions Voltage
(kV)
Distance
(cm)
Rate
(µL/min)
Electric
Field
(kV/cm)
Temp
(°C)
RH
(%)
0 No NaPO4 17 7 5 2.4 23 22
1
Control
Constant
Field 17 7 5 2.4 22 22
2 Constant
Field 21 9 5 2.4 22 22
3 Constant
Field 27 11 5 2.4 22 22
4 Varied
Voltage 15 7 5 2.1 22 22
5 Varied
Voltage 12 7 5 1.7 22 22
6 Varied
Voltage 9 7 5 1.3 22 22
7 Varied Pump
Rate 17 7 7 2.4 22 22
8 Varied Pump
Rate 17 7 3 2.4 22 22
9 Varied Pump
Rate 17 7 1 2.4 22 22
10 Varied
Distance 17 5 5 3.4 22 22
11 Varied
Distance 17 9 5 1.9 22 22
12 Varied
Distance 17 11 5 1.5 22 22
3.2.2 Acetic Acid Solvent System
Similar to the DMF system, all AA systems were electrospun at a distance of 7
cm, with an accelerating voltage of 17 kV, and a pump rate of 5 μL/min. All samples
34
were allowed to spin for 30 minutes in order to allow for a large quantity of fibers. See
Table 4 for a breakdown of the spin conditions at the time of experiment.
Table 4: Summary of the AA solvent system solutions that were spun and the atmospheric
conditions on that day. Weight Percent of ChS
(wt% [m/m])
Temp
(°C) RH (%)
1 20.0 24
5 22.2 22
10 22.2 74*
15 22.2 69*
20 22.2 22
30 22.8 26
35
3.3 Measuring pH
The pH was taken using a pH paper (VWR North American, BDH35309.606).
Images of the equipment can be seen in Figure 16.
Figure 16: The pH paper used in this experiment.
36
3.4 Viscosity
The viscosity of all samples was determined using either an Ubbelohde glass
viscometer (UGV) (Fisher Scientific, 13-614A, Pittsburgh, PA), or a rheometer (TA
Instruments AR Rheometer) as can be seen in Figure 17. The UGV measures the
kinematic viscosity of the polymers. The kinematic viscosity was calculated using
ASTM D 445 standards. The kinematic viscosity was then converted to relative
viscosity. The capillary diameter of the UGV was 1 mm. Each solution was measured
three times in order to get an average kinematic viscosity at 22 – 23 °C (the operating
limit for the UGV was 22 ± 2°C). The UGV could only measure kinematic viscosity
from 2 cSt to 10 cSt.
Additionally, a rheometer (TA Instruments AR Rheometer) was used for solutions
that were either too viscous, or not viscous enough, to be measured in the UGV. The
conditions used were a temperature of 25 °C and a 40 mm circular parallel plate, with a
shear rates ranging from 0 – 300 s-1
. The rheometer recorded the dynamic viscosity in
Pa*s. This was converted to kinematic viscosity (cSt) so as to allow for a comparison
with the DMF solutions, and then to relative viscosity for the same reason. Unlike the
kinematic viscosity, the dynamic viscosity does not take the density of the solution into
consideration, thus a conversion was needed in order to perform a comparison. The
viscosity that was recorded was averaged from several points according to when the shear
stress was near zero in order to obtain a near-zero viscosity, which in theory remains
constant (8). This procedure was adapted from a previous study (8).
37
Figure 17: The instruments used to measure viscosity. Top) The UGV, Bottom) the
rheometer.
38
3.5 Attenuated Total Reflectance Fourier Transform Infrared Spectroscopy
Attenuated total reflectance Fourier transform infrared spectroscopy (ATR-FTIR)
was performed to determine if any mats contained both HA and ChS. Specifically, ATR-
FTIR (Thermo Scientific, Nicolet Nexus 870 FT-IR Spectrometer, Waltham, MA.) had a
temperature controller that ranged from 22-110°C (22°C was used), could measure near
and mid infrared, and also contained the single bounce ATR accessory. The ATR-FTIR
was used to examine the percent absorption peaks from 3500 cm-1
to 500 cm-1
. The
device can be seen in Figure 18.
Figure 18: Image of the AT-FTIR that was used in this study.
39
3.6 Field Emission Scanning Electron Microscopy
Field emission scanning electron microscopy (FESEM, Zeiss Supra 50VP) was
used to determine if fibers were successfully electrospun, and to determine the diameters
of the spun fibers. All samples were mounted on a stub using a double-sided carbon tape
then coated with platinum/palladium for 35 s, 40 mA using Denton vacuum desk II
sputtering machine. This was done to improve the charging on the samples. See Chapter
4 for a detailed breakdown for each SEM images. See Figure 19 for an image of the
SEM.
The FESEM images were analyzed using ImageJ software from the US National
Institutes of Health. This program uses Java programming in order to perform image
enhancement and measurement (74). This software is open source and available to the
public. ImageJ was used to measure the diameter of the fibers. For each sample, 50
fibers were measured. This was done to ensure a large statistical population.
Figure 19: Image of the type of SEM that was used in this experiment (75).
40
3.7 Energy-Dispersive X-ray Spectroscopy
Energy-dispersive x-ray spectroscopy (EDX) was used to determine if both ChS
and HA were in specific electrospun fibers. Specifically, since ChS contained a sulfate
group, the EDX would confirm that ChS was in a fiber because it is the only component
that would contain such a group. EDX was done in conjunction with, and in support of,
FTIR. The EDX operates as an attachment to the SEM. The EDX device was from
Oxford Energy-Dispersive X-ray Microanalysis.
3.8 Solubility
Solubility testing was performed to determine how well an electrospun mat would
survive in an aqueous environment. This was performed using three different aqueous
baths: one bath contained 20 mL of 1.0M of acetic acid, another contained 20 mL of
1.0M sodium hydroxide, and the last bath contained 20 mL of distilled water (38). Six
samples (each sample was 1 mm2 in area) were prepared, such that each bath would
contain two samples. One of each sample was immersed in the baths for 15 minutes and
72 hours respectively (38). After the appropriate amount of time, the solution was
examined for any change in transmittance. Transmittance at 600 nm as taken using the
reflectance spectrometer seen in Figure 20 (Ocean Optics, Inc., USB2000 Miniature Fiber
Optic Spectrometer, and DH-2000 Deuterium Tungsten Halogen Light Sources, Dunedin,
FL) (38).
The solubility was determined by the transmittance of the bath liquids. Cuvettes
filled with 1.5 mL of sample from each bath were used. The transmittance of the samples
was compared to that of control baths. If the sample had a transmittance from 80 - 90%,
the sample was dissolving (38). If the sample had a transmittance greater than 90%, and
there was no visual confirmation, the sample has dissolved (38). If the sample’s
41
transmittance was above 90%, and there was visual confirmation of the sample, then the
sample was still dissolving (38). If any samples were below 80%, they were considered
to be not dissolving (38).
Figure 20: Image of the reflectance spectrometer set-up used to characterize the
solubility of the samples.
42
Chapter 4: Results and Discussion
4.1 Dimethylformamide Solvent System
The DMF systems had multiple successful electrospin mats. As mentioned
previously, two different solutions were tested. The general parameters are seen below in
Table 5.
Table 5: Summary of the pH of the DMF solutions examined Solution pH
NaPO4 7
No NaPO4 7
4.1.1 Viscosity
The viscosity profiles of the solutions were unsurprising. On average, there was
an increase in viscosity with increasing amounts of polymer solution. This is consistent
with literature (76, 77). The 1 wt% ChS with 2 wt% HA in DMF co-polymer system was
not tested because it was not observed to flow once placed in the UGV. Furthermore,
time constraints prevented a rheological exam.
43
Figure 21: Viscosity of the DMF solvent systems with each polymer.
What is interesting, however, is the difference between the addition of ChS and
HA individually to the solvent system. As can be seen in Figure 21, the relative viscosity
of ChS was only 1.04 ± 0.01. The average kinematic viscosity was 0.10 ± 0.01 cSt. This
viscosity for the ChS is to be as expected. A previous study determined that for the
weight percent of ChS that was used, the average kinematic viscosity was approximately
0.09 cSt (78). This value is very close to what was expected.
Another trend was observed with the HA sample. Although data were recorded, it
was beyond the viscometer’s accuracy. The UGV used could only measure up to 10 cSt,
while the lowest HA value was 55.38 cSt. It is safe to say that the solution with only HA
is more viscous than the ChS solution. The kinematic viscosity recorded for the HA
sample is close to the lower end of the reported range of 50 – 1000 cSt that was been
recorded in other studies (79, 80). What data that was able to be obtained from this
0
10
20
30
40
50
60
70
DMF + Water
1 wt% ChS
2 wt% HA
Rel
ativ
e V
isco
sity
Polymer Concentration
44
experiment suggests that the HA being used is of lower purity due to ultrapure HA
having a kinematic viscosity of approximately 1000 cSt (79, 80).
4.1.2 Attenuated Total Reflectance Fourier Transform Infrared Spectroscopy
The IR spectroscopy data of the DMF solvent system fibers reveal that ChS is
present in the fibers. As can be seen in both Figure 22 and Figure 23, the IR spectrum for
the ChS and HA mat contain the same key peaks that are found in the bulk HA and ChS
peaks. The key characteristic peak for ChS, specifically Ch4S, is found at 848 cm-1
(9,
81). These can be seen below in Table 6.
All IR spectra for this solution were taken using the Na2PO4 solution. This was
done because it had the largest fibers overall, and as a result would most yield a more
defined spectrum. Furthermore, it was used because the majority of the mats used during
this part of the study were those containing Na2PO4.
Table 6: Key peaks for both ChS and HA.
Peak Wavenumber (cm-1
) Species
848 ChS Characteristic Peak: C–O–S (9, 81)
1396 C-O (21)
1633 C=O/C-N (21)
2800-2900 C-H (21)
45
Figure 22: IR spectra of ChS+HA fibers for the DMF solvent system as compared to the
bulk polymers ChS and HA.
46
Figure 23: IR spectra for the DMF solvent system as compared to the bulk polymers;
focused from 2500 cm-1
to 600 cm-1
.
47
4.1.3 Field Emission Scanning Electron Microscopy
As can be seen in both Table 7 and Table 8, fibers were produced using the DMF
solvent system. The diameters of the fibers varied greatly depending on the spin
conditions. In both Figure 24 and Figure 25, there appears to be a general trend with the
data. For example, all samples that had a constant field showed an increasing linear trend
despite both the voltage and the distance were increased.
Table 7: Summary table of SEMs and Histograms of fiber diameters for the DMF
optimization study. Sample SEM Image Histogram of Fiber Diameters
0
1
48
2
3
4
5
49
6
7
8
9
50
10
11
12
51
Figure 24: Summary of all fiber diameters. Purple indicates no NaPO4, Yellow indicates
a constant field, Blue represents a change in voltage, Red indicates a change in pump
rate, and Green represents a change in distance.
Figure 25: Graph comparing the effect of voltage on fiber diameter.
y = -0.0343x + 0.1996 R² = 0.7931
0
0.05
0.1
0.15
0.2
0.25
0.3
9 12 15 17
Fib
er
Dia
me
ter
(µm
)
Voltage (kV)
52
Figure 26: Graph comparing the effects of distance on fiber diameter.
Figure 27: Graph comparing the effects of the electric field on the fiber diameters.
y = 0.0367x + 0.0243 R² = 0.8385
0
0.05
0.1
0.15
0.2
0.25
0.3
7 5 9 11
Fib
er
Dia
me
ter
(µm
)
Distance (cm)
y = 0.0237x + 0.0454 R² = 0.9967
0
0.05
0.1
0.15
0.2
1 2 3
Fib
er
Dia
me
ter
(µm
)
Sample Number
53
Figure 28: Graph comparing the effects of pump rate on fiber diameter.
y = -0.028x + 0.2015 R² = 0.4311
0
0.05
0.1
0.15
0.2
0.25
0.3
1 3 5 7
Fib
er
Dia
me
ter
(µm
)
Pump Rate (µL/min)
54
Table 8: A summary table of the fiber diameter statistical data from Figure 24, the
statistical sample size was 50 except for Sample 9 which had a sample size of 10.
Sample Conditions Mean (µm) St. Dev.
(±µm) Max (µm) Min (µm)
0 No NaPO4 0.046 0.015 0.083 0.013
1 Constant
Field 0.068 0.033 0.187 0.024
2 Constant
Field 0.094 0.039 0.184 0.030
3 Constant
Field 0.116 0.040 0.218 0.043
4 Varied
Voltage 0.104 0.039 0.227 0.048
5 Varied
Voltage 0.098 0.034 0.182 0.039
6 Varied
Voltage 0.185 0.077 0.377 0.074
7 Varied
Pump Rate 0.129 0.044 0.236 0.048
8 Varied
Pump Rate 0.126 0.040 0.218 0.068
9 Varied
Pump Rate 0.203 0.069 0.307 0.103
10 Varied
Distance 0.102 0.044 0.240 0.044
11 Varied
Distance 0.105 0.042 0.220 0.031
12 Varied
Distance 0.190 0.086 0.373 0.056
55
The reasons for this are because the field was maintained constant. To put it
simply, the electric field aids in creating the initial instability in the polymer solution that
eventually results in the jets that create the polymer fibers (50, 82, 83). In the case of this
experiment, since the field was kept constant, fibers would always be produced.
The increasing fiber diameter is the result of the increasing distance (65). When
the voltage of a system is kept constant, but when the distance is not, it is possible that
the polymer solution will not be stretched as thinly (65). Although this previous study
was not examining the effect of the electric field on the fiber diameter, the same principle
applies. To summarize, the further away from a target the solution is, the more voltage is
required to spin successfully. This also helps to explain some of the linear behavior seen
in the samples where the distance was increased, but with no change in voltage. These
conclusions are supported by the data shown in both Figure 26 and Figure 27.
As the name would suggest, the behavior seen in the sample set where the voltage
was changed was due to the change in voltage. As noted earlier, a change in voltage can
severely affect the size and morphology of the fibers (56, 64, 65). For Sample 6
specifically, the drastic change in fiber diameters suggests that the voltage change created
more instability in the fibers than was necessary, and as a result led to the fibers being
larger than the others. This conclusion is supported by data shown in Figure 25.
Furthermore, the change in pump rate was the main reason for the behavior in the
sample set where the pump rate was varied. Although pump rate usually does not
drastically affect fiber morphology, this was not the case in this experiment (59, 65, 84).
Under certain circumstances, the pump rate can affect fiber diameter and morphology
(56, 64, 65). As the rate was lowered, less polymer solution was available on the needle
56
tip to spin which explains the overall low yield of this series. As can be seen in Figure
28, there does not appear to be a trend associated with altering the pump rate.
4.2 Acetic Acid Solvent System
The pH data for each sample can be seen in Table 9. The data showed that on
average the solutions were acidic in nature. However, it appeared that ChS did make the
solution more basic in the higher concentrations. This solvent system was spun at a
voltage of 15 kV, 7 cm distance, and with an infusion rate of 5 uL/min. Other distances
and voltages were attempted, but these yielded no fibers. Data for those attempts can be
seen in Appendix B.
Table 9: The pH for each different wt% of ChS.
ChS (wt %) pH
5 3
10 3
15 4
20 4
30 4
57
4.2.1 Viscosity
The viscosity measurements of the AA solutions follow the expected trend for
such a system. As can be seen in Figure 29, the viscosity of the solution decreased with
increasing amounts of ChS. This is to be expected as it is known that at room
temperature in a non-alkaline solution, the viscosity number of ChS is 41 (85). This
indicates that beyond a certain mass, the system will become oversaturated and the
viscosity will drop (85). This is due to less entanglement from the polymers (78, 85).
The data supports this assertion. The relative viscosities of the solutions were 253,029 ±
8.08 for the HA sample, 198,478 ± 5.94 for the 1wt% ChS sample, and 32,129 ± 0.36 for
the 5wt% ChS sample. Compared to the DMF solutions, these solutions were generally
more viscous due to the larger relative viscosity that is exhibited in these solutions.
Figure 29: Graph showing the viscosity of the lower weight percents for the AA solvent
system. Please note the mass of HA was held constant throughout.
0
50000
100000
150000
200000
250000
300000
Control 0 wt% ChS (0.25wt% HA)
1 wt% ChS 5 wt% ChS
Re
lati
ve V
isco
sity
Concentration of the polymer systems
58
4.2.2 Attenuated Total Reflectance Fourier Transform Infrared Spectroscopy
FTIR of the electrospun fibers confirmed that both HA and ChS were found in the
mat. As can be seen in both Figure 30 and Figure 31, the IR spectrum for the ChS and
HA mat contain the same key peaks that are found in the bulk HA and ChS peaks. The
key characteristic peak for ChS, specifically Ch4S, is found at 854 cm-1
(9). Given the
similarities in functional groups, both HA and ChS have similar peaks. These can be
seen below in Table 10.
All IR spectra for this solution were taken using the 30 wt% ChS solution. This
was done because it had the largest fibers overall, and would most yield a more defined
spectrum.
Table 10: Shared peaks between ChS and HA. Peak Wavenumber (cm
-1) Species
848 ChS Characteristic Peak: C–O–S (9, 81)
1396 C-O (21)
1633 C=O/C-N (21)
2800-2900 C-H (21)
59
Figure 30: IR of the bulk ChS, bulk HA, and 30% mass ChS mat.
60
Figure 31: IR spectra for the range of 2500-600 showing the key peaks for the polymers.
61
4.2.3 Field Emission Scanning Electron Microscopy
The SEM images of the electrospun fibers yielded fascinating results. Only the
samples that were subjected to the concentration study were examined using the SEM.
To begin, all polymer solutions that were tested successfully spun fibers. The images and
statistics of each sample can be seen in Tables 11 & 12.
Table 11: Summary table of SEM images and histograms.
Sample SEM Image Histogram of Fiber Diameters
5
10
62
15
20
30
63
Table 12: Summary data for the fiber diameters for the AA solvent system. Same data as
found in Figure 32.
Percent % of
ChS
Average
(μm)
St. Dev.
(±μm)
Max
(μm)
Min
(μm)
5 0.08 0.03 0.16 0.012
10 2.04 0.89 4.63 0.61
15 3.75 2.74 17.47 0.75
20 6.31 2.75 14.42 2.00
30 11.97 14.89 82.96 3.86
The SEM images for the AA solution systems revealed a very surprising trend.
As can be seen in both Figure 32 and Figure 33, there was a general increase of fiber
diameter of nearly 2-3 times for each increase in the weight percent of ChS. Although,
this would seem to be in defiance of the general rule of, “increased concentration,
decreased fiber diameters,” this is not the case. As noted earlier, there is evidence to
suggest that the increase in concentration will lead to an increase in fiber diameter (56,
64, 65). As the only variation between these samples was the concentration of ChS, it is
very likely that the concentration of ChS caused an increase in the fiber diameters.
64
Figure 32: Summary of fiber diameters for the AA solvent system.
Figure 33: Fiber diameters for the samples demonstrate a linear trend in conjunction
with an increase of ChS.
y = 0.4657x - 2.5546 R² = 0.9831
p = 0.01, α = 0.05
0
5
10
15
20
25
30
0 5 10 15 20 25 30
Ave
rage
Fib
er
Dia
me
ter
(mic
ron
s)
wt% ChS
65
Additionally, as can be seen in Figure 33, the relationship between polymer
concentration and fiber diameter is very linear. A linear regression was performed on the
data that confirmed this suspicion. This behavior lends further support to the theory that
an increase in concentration will lead to increased fiber diameters.
Finally, the viscosity and concentration of these solutions could explain the
increased fiber diameters. Previous studies noticed that there was a correlation between
fiber diameters, concentration, and viscosity (86, 87). One researcher did notice that
viscosity had a small effect on fiber size due to a longer polymer jet (87). The
researchers speculated that because the polymer jet was more elongated in higher
viscosity solutions which would lead to smaller diameter polymers (87). The researchers
concluded that concentration had a greater effect on electrospin fibers (87).
Furthermore, as mentioned earlier, an increase in the concentration of ChS can
lead to less entanglement of the polymer chains due to oversaturation (78, 85). This
could also explain the larger fiber diameters since the polymers were not as entangled as
others, but were beyond the critical entanglement point.
4.2.4 Energy-Dispersive X-ray Spectroscopy
EDX scans of the 30wt% ChS sample further confirmed that ChS was in the
sample fiber in addition to HA. As HA has been known to spin very well, it was assumed
that HA was already present in the fibers, or was the only compound to spin (21, 22).
FTIR did confirm that ChS was present in mat, but it could not determine if ChS was
actually in the fiber. As can be seen in Figure 34, EDX did show that sulfur was in the
fibers. The image used to acquire these scans can be seen in Figure 35.
66
Figure 34: EDS scan of fibers found in the 30% mass ChS sample with background
elements removed.
Figure 35: Image used in the EDX scan. Image is another view of the 30 wt% ChS
solution.
67
4.3 Chemical Stability
The DVS crosslinked solutions yielded some surprising results. As the DMF
solution without Na2PO4 was found to be among the best of the DMF solutions due to the
fewer materials required, and the smallest fibers, this was used as basis for the testing
solution for the DMF species. The 5 wt% ChS AA solution was used as the starting point
for the AA solutions with DVS. As noted previously, four solutions were used in this
part of the study. The pH of these solutions can be seen below in Table 13.
Table 13: The pH of the DVS crosslinker solutions.
Solution pH
DMF Excess HA 6
DMF with Excess ChS 6
AA with Excess HA 4
AA with Excess ChS 4
68
4.3.1 Field Emission Scanning Electron Microscopy
The DMF-DVS systems contained one of the best SEM images. The best image
was from the sample that contained excess HA. The solution containing excess ChS
failed to spin properly and as a result no SEM was performed.
Figure 36: SEM image of the fibers produced from this solution (top) and a histogram
with table containing the statistical analysis of the fiber diameters (bottom).
69
The fibers that were spun prior to the cross-linking step were very good. As can
be seen in Figure 36 above, the fibers have excellent morphology and have an average
fiber diameter of 88.89 ± 32.32 nm. The largest fiber diameter was 187.92 nm, while the
smallest fiber diameter was 25.00 nm.
Figure 37: SEM image of the fibers produced from this solution (top) and a histogram
with table containing the statistical analysis of the fiber diameters (bottom).
70
The fibers that were spun after the cross-linking step were drastically altered. As
can be seen in Figure 37 above, the fibers appear to have either bonded together, or have
become swelled due to the water bath that was used to remove excess salt. Given that
the average diameter of the remaining fibrous structures was 6.78 ± 2.61 µm, and the
largest fiber diameter was 11.97 µm, while the smallest fiber diameter was 2.22 µm, the
swelling choice appears to be correct. The fibers grew approximately 7000% from the
start of the cross-linking process to the end.
Despite this large jump, this is not a bad thing. The crosslinked “fibers” very
closely match the natural structure of human cartilage. In fact, this accidentally
discovered structure is ideal for a biomimetic device.
4.3.2 Attenuated Total Reflectance Fourier Transform Infrared Spectroscopy
The IR spectra of the excess HA fibers, as seen in Figure 38, show a successful
crosslinking. There is a clear change in the peaks at 2912 and 2848 which indicates a
change in the C-H bonds (88). Furthermore, the change in C=C appears at 1643 could
indicate that the vinyl groups on DVS have become less prominent, and suggests that the
crosslinking mechanism may have worked (88). Although it does not show definitive
proof of a successful crosslinking step, the IR does offer some support to the theory that
the polymers were successfully crosslinked. Table 14 contains the peak wavenumbers.
71
Table 14: IR Peaks for Figure 38. Peak Wavenumber (cm
-1) Species
848 ChS Characteristic Peak: C–O–S (9, 81)
1396 C-O (21)
1633 C=O/C-N (21)
2800-2900 C-H (21, 88)
Figure 38: IR spectra for the excess HA fibers. Before and after crosslinking are shown.
72
4.3.3 Solubility
The solubility test performed on the DMF-HA sample yielded some surprising
results. After 15 minutes, the mats were almost completely dissolved in the acidic and
water baths. The transmittance values at the 15 minute mark confirm that the mats were
dissolving at that time (Table 15). At the 72 hour mark, there was no visible trace of any
mat. All transmittance values indicated that the mats had dissolved into solution.
The results of this experiment concur with findings in literature. Although, it
should be pointed that there is little data available that relates to the solubility of
crosslinked ChS and HA in the same material; there is a plethora of information,
however, for each polymer crosslinked individually. The data that most closely matches
the behavior observed in this experiment is seen in several studies that examine the
solubility of crosslinked HA. The researchers found that crosslinked HA will dissolve
rapidly when exposed to hydroxyl radicals (89, 90). In one study, the crosslinked HA
was dissolved within minutes of being exposed to hydroxyl radicals (90).
Table 15: Results of the solubility test.
15 minutes 72 hours
Solution Visual
Observation T600 Comments
Visual
Observation T600 Comments
1.0 M AA
Reference N/A 99.17 ± 0.39 Control N/A 98.46 ± 1.97 Control
1.0 M AA No 99.33 ± 1.15 Dissolved No 95.56 ± 0.72 Dissolved
1.0 M NaOH
Reference N/A 98.53 ± 1.63 Control N/A 98.63 ± 1.43 Control
1.0 M NaOH Yes 88.38 ± 0.65 Partially
Dissolved No 92.11 ± 0.05 Dissolved
DI Water Reference N/A 98.44 ± 0.68 Control N/A 100.13 ± 0.09 Control
DI Water Yes 101.47 ± 1.20 Partially
Dissolved No 101.30 ± 0.47 Dissolved
73
Chapter 5: Conclusions
First, ChS-HA nanofibers were successfully electrospun using the previously
known method to electrospin HA. Secondly, a predominantly ChS mat was successfully
electrospun using an AA solvent system which may be better for the body than a DMF
based system. Finally, it was shown that despite its promise, the AA system did not spin
with the added crosslinking agent, but the DMF system was spun and crosslinked
successfully.
The DMF based solvent system operated far better than anticipated. The solution
spun extremely well in the ideal conditions of room temperature and low humidity.
Additionally, it was discovered that the DMF solvent system could be spun without
disodium phosphate, an additive that is required in the spinning of HA. In an industrial
application, this would be an excellent advantage as it would drive down the cost of
manufacture. Once this solution was identified, the viscosity of the solution, and its pH,
were determined in order to better understand the solution parameters.
The ability to spin the AA system in a lower volume solution was another result
that was unexpected. As a result, an attempt was made to discover the most ChS that
could be spun successfully and it was discovered that larger concentrations of ChS
produces larger fiber diameters.
After some calculations, it was decided to test each system against each other
using DVS to crosslink the fibers. This would create a common ground for both solvent
systems and allow for one on one comparison. The optimal amounts of polymer were
added to their respective solvents, and electrospun. The only solvent system that
successfully electrospun was the DMF based solution that contained an excess mass of
74
HA which was subsequently tested using solubility tests. In conclusion, this study
successfully laid the groundwork for the development of an electrospun biomimetic
scaffold for both human cartilage and knee meniscus. There is much more work to be
done in regard to these systems, but this study has shown that with some refinement, two
different systems can work can be used to accomplish this goal.
75
Chapter 6: Future Work
6.1 Collagen
As noted previously, collagen is the primary polymeric material by mass in both
the knee and cartilage. Future direction will be to include collagen into the CS-HA
crosslinked fibrous mats. The reason that collagen was not spun in this study was because
it is known to cause scar tissue formation, and possibly cause an immune system reaction
(21). Collagen should provide mechanical strength that is needed for this biomimetic
scaffold to function properly in the human body.
6.2 Mechanical Testing
The mechanical testing of this biomimetic scaffold should be attempted in the
future. One of the primary reasons why mechanical testing was not performed during
this study was that collagen was missing from the scaffold. Without collagen, it is
unlikely that the electrospun fibers would have matched the natural tissue in any form of
mechanical testing. Given the discovered structural properties of the electrospun and
crosslinked fibers, however, this should not be a challenge to future researchers.
6.3 Cell and Clinical Studies
As with all devices that will be inserted into the human body, cellular studies
must be performed on the device. Such a study will be able to definitively prove which
solvent system will be the best for a biomimetic scaffold. Hopefully, such tests will lead
to a new medical device on the market for the treatment of OA.
76
References and Works Cited
1. Athanasiou, K., Darling, E., and Hu, J. (2010) Articular cartilage tissue
engineering, on Tissue Engineering (Athanasiou, K., Ed.), pp 1-168, Morgan &
Claypool.
2. Varki, A., Cummings, R. D., Esko, J. D., Freeze, H. H., Stanley, P., Bertozzi, C.
R., Hart, G. W., Etzler, M. E., Kimata, K., Lindahl, U., and others. (2009)
Proteoglycans and Sulfated Glycosaminoglycans. In Essentials of Glycobiology
(Varki, A., Cummings, R. D., Esko, J. D., Freeze, H. H., Stanley, P., Bertozzi, C.
R., Hart, G. W., Etzler, M. E., Kimata, K., Lindahl, U., and others, Eds.) 2nd ed.,
Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
3. Fithian, D. C., Kelly, M. a, and Mow, V. C. (1990) Material properties and
structure-function relationships in the menisci., Clinical orthopaedics and related
research 19-31.
4. Holmes, M. W., Bayliss, M. T., and Muir, H. (1988) Hyaluronic acid in human
articular cartilage. Age-related changes in content and size., The Biochemical
journal 250, 435-41.
5. Nishimura, M., Yan, W., Mukudai, Y., Nakamura, S., Nakamasu, K., Kawata, M.,
Kawamoto, T., Noshiro, M., Hamada, T., and Kato, Y. (1998) Role of chondroitin
sulfate-hyaluronan interactions in the viscoelastic properties of extracellular
matrices and fluids, Biochimica et Biophysica Acta (BBA)-General Subjects,
Elsevier 1380, 1–9.
6. Kiani, C., Chen, L., Wu, Y. J., Yee, A. J., and Yang, B. B. (2002) Structure and
function of aggrecan., Cell research 12, 19-32.
7. Zaia, J., and Costello, C. E. (2001) Compositional Analysis of
Glycosaminoglycans by Electrospray Mass Spectrometry, Analytical Chemistry
73, 233-239.
8. Schiffman, J. D. (2009) Determination of the electrospinning parameters for
biopolyelectrolytes and their modifications, Drexel University, Philadelphia, PA.
9. Garnjanagoonchorn, W., Wongekalak, L., and Engkagul, A. (2007) Determination
of chondroitin sulfate from different sources of cartilage, Chemical Engineering
and Processing 46, 465-471.
10. Lu, H., McDowell, L., and Studelska, D. (2010) Glycosaminoglycans in Human
and Bovine Serum: Detection of Twenty-Four Heparan Sulfate and Chondroitin
Sulfate Motifs Including a Novel Sialic Acid-modified Chondroitin Sulfate
Linkage Hexasaccharide, Glycobiology insights 2010, 13-28.
77
11. Makris, E. a, Hadidi, P., and Athanasiou, K. a. (2011) The knee meniscus:
Structure-function, pathophysiology, current repair techniques, and prospects for
regeneration., Biomaterials, Elsevier Ltd 32, 7411-7431.
12. Athanasiou, K. A., and Sanchez-Adams, J. (2009) Engineering the Knee Meniscus,
Synthesis Lectures on Tissue Engineering, pp 1–97, Morgan & Claypool
Publishers.
13. Goldring, M. B., and Goldring, S. R. (2007) Osteoarthritis, Journal of Cellular
Physiology 213, 626-634.
14. Rizkalla, G., Reiner, A., and Bogoch, E. (1992) of the articular cartilage
proteoglycan aggrecan in health and osteoarthritis. Evidence for molecular
heterogeneity and extensive molecular changes in disease., Journal of Clinical 90,
2268-77.
15. Brandt, K. D., Smith, G. N., and Simon, L. S. (2000) Intraarticular injection of
hyaluronan as treatment for knee osteoarthritis: what is the evidence?, Arthritis
and rheumatism 43, 1192-203.
16. Lohmander, L. S., Dalen, N., Englund, G., Hamalainen, M., Martin, E., Karlsson,
K., Odensten, M., Ryd, L., and Sembo, I. (1996) Intra-articular hyaluronan
injections in the treatment of osteoarthritis of the knee : a randomised, double
blind, placebo controlled multicentre trial, Arthritis & Rheumatism 424-431.
17. (2010, December 28) FDA will rescind the controversial Menaflex 510(k),
Biomedical Business & Technology.
18. (2011, June) ReGen sues FDA over Menaflex reversal, Medical Product
Outsourcing.
19. Clouet, J., Vinatier, C., Merceron, C., Pot-vaucel, M., Maugars, Y., Weiss, P.,
Grimandi, G., and Guicheux, J. (2009) From osteoarthritis treatments to future
regenerative therapies for cartilage, Drug Discovery Today 14, 913-925.
20. Stockwell, R. a. (1971) The interrelationship of cell density and cartilage thickness
in mammalian articular cartilage., Journal of anatomy 109, 411-21.
21. Brenner, E. (2009) Investigation into the Electrospinning of Hyaluronic Acid,
Drexel University.
22. Brenner, E. K., Schiffman, J. D., Thompson, E. a., Toth, L. J., and Schauer, C. L.
(2012) Electrospinning of hyaluronic acid nanofibers from aqueous ammonium
solutions, Carbohydrate Polymers, Elsevier Ltd. 87, 926-929.
78
23. Watanabe, H., Yamada, Y., and Kimata, K. (1998) Roles of aggrecan, a large
chondroitin sulfate proteoglycan, in cartilage structure and function., Journal of
biochemistry 124, 687-93.
24. Mathews, M. B. (1956) The molecular weight of sodium chondroitin sulfate by
light scattering, Archives of Biochemistry and Biophysics 61, 367-377.
25. Kim, I. L., Mauck, R. L., and Burdick, J. a. (2011) Hydrogel design for cartilage
tissue engineering: a case study with hyaluronic acid., Biomaterials, Elsevier Ltd
32, 8771-82.
26. Saravanakumar, G., Choi, K. Y., Yoon, H. Y., Kim, K., Park, J. H., Kwon, I. C.,
and Park, K. (2010) Hydrotropic hyaluronic acid conjugates: Synthesis,
characterization, and implications as a carrier of paclitaxel., International journal
of pharmaceutics, Elsevier B.V. 394, 154-61.
27. Jin, R., Teixeira, L. S. M., Dijkstra, P. J., van Blitterswijk, C. a, Karperien, M., and
Feijen, J. (2010) Enzymatically-crosslinked injectable hydrogels based on
biomimetic dextran-hyaluronic acid conjugates for cartilage tissue engineering.,
Biomaterials, Elsevier Ltd 31, 3103-13.
28. Bairstow, S., McKee, J., Nordhaus, M., and Johnson, R. (2009) Identification of a
simple and sensitive microplate method for the detection of oversulfated
chondroitin sulfate in heparin products., Analytical biochemistry, Elsevier Inc. 388,
317-21.
29. Manuscript, A. (2010) NIH Public Access, Journal of Pharmacological Sciences
78, 292-300.
30. Guerrini, M., Beccati, D., Shriver, Z., Naggi, A., Viswanathan, K., Bisio, A.,
Capila, I., Lansing, J. C., Guglieri, S., Fraser, B., Al-Hakim, A., Gunay, N. S.,
Zhang, Z., Robinson, L., Buhse, L., Nasr, M., Woodcock, J., Langer, R.,
Venkataraman, G., Linhardt, R. J., Casu, B., Torri, G., and Sasisekharan, R. (2008)
Oversulfated chondroitin sulfate is a contaminant in heparin associated with
adverse clinical events., Nature biotechnology 26, 669-75.
31. Toth, L. J. (2012) Electrospun Crosslinked Hyaluronic Acid Fibers for Tissue
Engineering Applications, Drexel University.
32. Yagci, A., Murk, W., and Stronk, J. (2010) Spermatozoa bound to solid state
hyaluronic acid show chromatin structure with high DNA chain integrity: an
acridine orange fluorescence study, Journal of andrology.
33. Ko, C.-S., Huang, J.-P., Huang, C.-W., and Chu, I.-M. (2009) Type II collagen-
chondroitin sulfate-hyaluronan scaffold cross-linked by genipin for cartilage tissue
79
engineering., Journal of bioscience and bioengineering, The Society for
Biotechnology, Japan 107, 177-82.
34. Wisniak, J., and Tamir, A. (1977) Vapor-liquid equilibriums in the ternary systems
water-formic acid-acetic acid and water-acetic acid-propionic acid, Journal of
Chemical & Engineering Data, American Chemical Society 22, 253-260.
35. Potter, A. E., and Ritter, H. L. (1954) The Vapor Pressure of Acetic Acid and
Acetic-d3 Acid-d. The Liquid Density of Acetic-d3 Acid-d, The Journal of
Physical Chemistry, American Chemical Society 58, 1040-1042.
36. Bamberger, A., Sieder, G., and Maurer, G. (2000) High-pressure (vapor+liquid)
equilibrium in binary mixtures of (carbon dioxide+water or acetic acid) at
temperatures from 313 to 353 K, The Journal of Supercritical Fluids 17, 97-110.
37. Park, J., and Lakes, R. S. (2010) Biomaterials: An Introduction 3rd ed., pp 1-561,
Springer, New York, NY.
38. Austero, M. S., Donius, A. E., Wegst, U. G. K., and Schauer, C. L. (2012) New
crosslinkers for electrospun chitosan fibre mats. I. Chemical analysis, Journal of
the Royal Society, Interface / the Royal Society 1-13.
39. Ko, C.-S., Huang, J.-P., Huang, C.-W., and Chu, I.-M. (2009) Type II collagen-
chondroitin sulfate-hyaluronan scaffold cross-linked by genipin for cartilage tissue
engineering, Journal of Bioscience and Bioengineering 107, 177-182.
40. Ko, C.-shen, Wu, C., and Huang, H. (2007) Genipin cross-linking of type II
collagen-chondroitin sulfate-hyaluronan scaffold for articular cartilage therapy,
Journal of Medical and Biological Engineering 27, 7-14.
41. Tomihata, K., and Ikada, Y. (1997) Crosslinking of Hyaluronic Acid with
Glutaraldehyde 3553-3559.
42. Lee, C.-T., Kung, P.-H., and Lee, Y.-D. (2005) Preparation of poly(vinyl alcohol)-
chondroitin sulfate hydrogel as matrices in tissue engineering, Carbohydrate
Polymers 61, 348-354.
43. Oprea, A., Ciolacu, D., and Neamtu, A. (2010) Cellulose/chondroitin sulfate
hydrogels: Synthesis, drug loading/release properties and biocompatibility,
Cellulose 44, 369-378.
44. Fahnestock, K. J. (2011) Electrospinning and rheological characterization of
quaternized chitosan, Drexel University, Philadelphia, Pa U6 - ctx_ver=Z39.88-
2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-
8&rfr_id=info:sid/summon.serialssolutions.com&rft_val_fmt=info:ofi/fmt:kev:mt
80
x:dissertation&rft.genre=dissertation&rft.title=Electrospinning+and+rheological+c
haracterization+of+quate.
45. Liu, S., Weaver, J. V. M., Save, M., and Armes, S. P. (2002) Synthesis of pH-
responsive shell cross-linked micelles and their use as nanoreactors for the
preparation of gold nanoparticles, Langmuir, ACS Publications 18, 8350–8357.
46. Lu, X., Hu, Z., and Gao, J. (2000) Synthesis and Light Scattering Study of
Hydroxypropyl Cellulose Microgels, Macromolecules 33, 8698-8702.
47. Anbergen, U. (1990) Elasticity and swelling behaviour of chemically crosslinked
cellulose ethers in aqueous systems, Polymer 2-6.
48. Wade Jr., L. G. (2010) Organic Chemistry 7th ed., pp 1-1243, Pearson Prentice
Hall, Upper Saddle River, NJ.
49. Schiffman, J. D. (2009) Determination of the electrospinning parameters for
biopolyelectrolytes and their modifications, Office, p 329, Drexel University,
Philadelphia, PA.
50. Doshi, J. (1995) Electrospinning process and applications of electrospun fibers,
Journal of Electrostatics 35, 151-160.
51. Li, M., Mondrinos, M. J., Gandhi, M. R., Ko, F. K., Weiss, A. S., and Lelkes, P. I.
(2005) Electrospun protein fibers as matrices for tissue engineering., Biomaterials
26, 5999-6008.
52. Keun Kwon, I., Kidoaki, S., and Matsuda, T. (2005) Electrospun nano- to
microfiber fabrics made of biodegradable copolyesters: structural characteristics,
mechanical properties and cell adhesion potential, Biomaterials 26, 3929-3939.
53. Badami, A. S., Kreke, M. R., Thompson, M. S., Riffle, J. S., and Goldstein, A. S.
(2006) Effect of fiber diameter on spreading, proliferation, and differentiation of
osteoblastic cells on electrospun poly(lactic acid) substrates, Biomaterials 27, 596-
606.
54. Subbiah, T., Bhat, G. S., Tock, R. W., Parameswaran, S., and Ramkumar, S. S.
(2005) Electrospinning of nanofibers, Journal of Applied Polymer Science 96, 557-
569.
55. Formhals, A. (1934) Process and apparatus for preparing artificial threads.
56. Greiner, A., and Wendorff, J. H. (2007) Electrospinning: a fascinating method for
the preparation of ultrathin fibers., Angewandte Chemie (International ed. in
English) 46, 5670-703.
81
57. Schiffman, J. D., and Schauer, C. L. (2008) A Review: Electrospinning of
Biopolymer Nanofibers and their Applications, Polymer Reviews 48, 317-352.
58. Ramakrishna, S., Fujihara, K., Teo, W.-E., Lim, T.-C., and Ma, Z. (2005) An
Introduction to Electrospinning and Nanofibers, pp 1-396, World Scientific
Publishing Co., Singapore.
59. Gupta, P., Elkins, C., Long, T. E., and Wilkes, G. L. (2005) Electrospinning of
linear homopolymers of poly(methyl methacrylate): exploring relationships
between fiber formation, viscosity, molecular weight and concentration in a good
solvent, Polymer 46, 4799-4810.
60. McKee, M., Hunley, M., and Layman, J. (2006) Solution rheological behavior and
electrospinning of cationic polyelectrolytes, Macromolecules 39, 575-583.
61. McKee, M. G., Wilkes, G. L., Colby, R. H., and Long, T. E. (2004) Correlations of
Solution Rheology with Electrospun Fiber Formation of Linear and Branched
Polyesters, Macromolecules 37, 1760-1767.
62. Graessley, W. (1980) Polymer chain dimensions and the dependence of
viscoelastic properties on concentration, molecular weight and solvent power,
Polymer 21, 258-262.
63. Son, B., Yeom, B., and Song, S. (2009) Antibacterial electrospun chitosan/poly
(vinyl alcohol) nanofibers containing silver nitrate and titanium dioxide, Journal of
Applied 3-10.
64. Jacobs, V., and Anandjiwala, R. (2010) The influence of electrospinning
parameters on the structural morphology and diameter of electrospun nanofibers,
Journal of Applied 115, 3130-3136.
65. Deitzel, J., Kleinmeyer, J., and Harris, D. (2001) The effect of processing variables
on the morphology of electrospun nanofibers and textiles, Polymer 42, 261-272.
66. Gu, S.-Y., Wu, Q.-L., Ren, J., and Vancso, G. J. (2005) Mechanical Properties of a
Single Electrospun Fiber and Its Structures, Macromolecular Rapid
Communications 26, 716-720.
67. Sill, T. J., and von Recum, H. a. (2008) Electrospinning: applications in drug
delivery and tissue engineering, Biomaterials 29, 1989-2006.
68. Hou, H., Jun, Z., Reuning, A., Schaper, A., Wendorff, J. H., and Greiner, A.
(2002) Poly( p -xylylene) Nanotubes by Coating and Removal of Ultrathin
Polymer Template Fibers, Macromolecules 35, 2429-2431.
82
69. Yoshimoto, H., Shin, Y. M., Terai, H., and Vacanti, J. P. (2003) A biodegradable
nanofiber scaffold by electrospinning and its potential for bone tissue engineering,
Biomaterials 24, 2077-2082.
70. Ryu, Y. J., Kim, H. Y., Lee, K. H., Park, H. C., and Lee, D. R. (2003) Transport
properties of electrospun nylon 6 nonwoven mats, European Polymer Journal 39,
1883-1889.
71. Gibson, P., Schreuder-Gibson, H., and Rivin, D. (2001) Transport properties of
porous membranes based on electrospun nanofibers, Colloids and Surfaces A:
Physicochemical and Engineering Aspects 187-188, 469-481.
72. Gibson, P., and Schreuder‐Gibson, H. (1999) Electrospun fiber mats: transport
properties, AIChE journal 45, 190-195.
73. Yang, F., Murugan, R., Wang, S., and Ramakrishna, S. (2005) Electrospinning of
nano/micro scale poly(L-lactic acid) aligned fibers and their potential in neural
tissue engineering, Biomaterials 26, 2603-10.
74. (2012) ImageJ Features, National Institutes of Health.
75. Carl Zeiss. SUPRA® Series - Ultra High Resolution FE-SEM for Versatile
Analytics, Scanning Electron Microscopes.
76. Painter, P. C., and Coleman, M. M. (1997) Fundamentals of Polymer Science: An
Introductory Text 2nd ed., pp 1-478, CRC Press, Boca Raton, FL.
77. Young, R. J. (1991) Introduction to Polymers 2nd ed., pp 1-456, Nelson Thornes
Ltd, London, UK.
78. Arias, C., ag e, A., ueda, C., and Garc a lanco, F. (1998) Intrinsic viscosity
calculated out of single point measurements for chondroitin-4-sulfate and
chondroitin-6-sulfate solutions, Biophysical Chemistry 72, 307-312.
79. Balazs, E. A. (1979) Ultrapure hyaluronic acid and the use thereof, United States
Trademark and Patent Office.
80. Weiss, C., Balazs, E., and Onge, R. S. (1981) Clinical studies of the intraarticular
injection of HealonR (sodium hyaluronate) in the treatment of osteoarthritis of
human knees, Seminars in arthritis and 143-144.
81. Uchisawa, H., Okuzaki, B.-ichi, Ichita, J., and Matsue, H. (2001) Binding between
calcium ions and chondroitin sulfate chains of salmon nasal cartilage
glycosaminoglycan, International Congress Series 1223, 205-220.
83
82. Shin, Y. M., Hohman, M. M., Brenner, M. P., and Rutledge, G. C. (2001)
Electrospinning: A whipping fluid jet generates submicron polymer fibers, Applied
Physics Letters 78, 1149.
83. Fridrikh, S., Yu, J., Brenner, M., and Rutledge, G. (2003) Controlling the Fiber
Diameter during Electrospinning, Physical Review Letters 90, 1-4.
84. Schiffman, J. D., and Schauer, C. L. (2008) A Review: Electrospinning of
Biopolymer Nanofibers and their Applications, Polymer Reviews 48, 317-352.
85. Cleland, R. (1979) Effect of temperature on the limiting viscosity number of
hyaluronic acid and chondroitin 4 sulfate, Biopolymers 18, 1821-1828.
86. Fong, H., Chun, I., and Reneker, D. H. (1999) Beaded nanofibers formed during
electrospinning, Polymer 40, 4585-4592.
87. Buchko, C., Chen, L., and Shen, Y. (1999) Processing and microstructural
characterization of porous biocompatible protein polymer thin films, Polymer 40,
7397-7407.
88. Yan, D. (2000) Hyperbranched polymers made from A2 and ’2 type monomers.
1. Polyaddition of 1-(2-aminoethyl) piperazine to divinyl sulfone, Macromolecules
7693-7699.
89. Nihira, J. (1993) Regulated release of drug microspheres from inflammation
responsive degradable matrices of crosslinked hyaluronic acid 25, 133-143.
90. Yui, N., Okano, T., and Sakurai, Y. (1992) Inflammation responsive degradation
of crosslinked hyaluronic acid gels 22.
84
Appendices
Appendix A: Equations used to calculate the optimal amount of DVS to be used
The moles for each polymer were calculated using the following equation:
Eq.1
The results of this calculation are seen below in Table 16. The mass of all components
was 0.2 g to start off. Any mass could have been used, but 0.2 g was used because it
would translate to a low weight percent.
Table 16: Results of Eq.1.
Chemical Mass (g) Density (g/mL) Mw (g/mol) Moles
ChS 0.2 - 463 0.00043
HA 0.2 - 379 0.00053
DVS 0.2 1.17 118.1 0.00169
Once the number of moles had been determined, it was realized that there would always
be an excess of one polymer compared to the other. With this in mind, the Solver tool in
Microsoft Excel 2010 was used to back calculate the ideal mass for a polymer and DVS,
using the other polymer as the reference. The Solver tool used a GRG nonlinear
mathematical modeling to return the correct mass that would be needed to crosslink all
moles of polymer with DVS. A screenshot of Solver can be seen in Figure 39. The
results of Solver can be seen in Table 17.
85
Table 17: Results of the solver tool. As can be seen, the mass of each chemical correctly
correspond to the moles used for each sample.
Moles ChS as the Starting Point
Chemical Mass (g) Density (g/mL) Mw (g/mol) Moles
ChS 0.20 - 463 0.00043
HA 0.16 - 379 0.00043
DVS 0.03 1.17 118.1 0.00043
Moles HA as the Starting Point
Chemical Mass (g) Density (g/mL) Mw (g/mol) Moles
ChS 0.24 - 463 0.00053
HA 0.20 - 379 0.00053
DVS 0.03 1.17 118.1 0.00053
Figure 39: Screen shot of the solver tool in use.
86
Appendix B: Previous Attempts to Electrospin the AA solvent system
Table 18: The same procedure used for the DMF solvent system was used. The only
solution that worked was at 15 kV, highlighted in yellow. The fiber diameter is recorded
for that sample in the far right column. Conditions Voltage
(kV)
Distance
(cm)
Rate
(µL/min)
Electric
Field
(kV/cm)
Temp
(°C)
RH
(%)
Fiber Diameter
(µm)
Constant
Field
17 7 5 2.4 24 22 - -
Constant
Field
21 9 5 2.4 24 22 - -
Constant
Field
27 11 5 2.4 24 22 - -
Varied
Voltage
15 7 5 2.1 24 22 0.398 ± 0.147
Varied
Voltage
12 7 5 1.7 24 22
Varied
Voltage
9 7 5 1.3 24 22 - -
Varied
Pump Rate
17 7 7 2.4 24 22 - -
Varied
Pump Rate
17 7 3 2.4 24 22 - -
Varied
Pump Rate
17 7 1 2.4 24 22 - -
Varied
Distance
17 5 5 3.4 24 22 - -
Varied
Distance
17 9 5 1.9 24 22 - -
Varied
Distance
17 11 5 1.5 24 22 - -