 

 

2:559 LIBRARY
Michigan State

University

 

 

 

Thisistccertifythatthe
thesisentitied

INVESTIGATIONS ON THE EFFECTS OF CY CLIC LOADING AND
THERAPEUTIC TREATMENTS FOLLOWING TRAUMATIC INJURY
TO ARTICULAR CARTILAGE

presented by

NURIT GOLENBERG

has been accepted towards fulﬁllment
of the requirements for the

Master of Science degree in Mechanical Egineeririg—

aw mag/—

’ Major Professor's Signature

Wazoo?
/

Date

 

 

mummmywm

.—-L-h-I-I-0-0-0-0-0-D-I-l-‘-.-.-I-.-I-.-l-.-C-I-U-I-'-.-.-.-D-I. -

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5’08 KzlProleocﬂtPres/CIRCIDateDuo.indd

 

INVESTIGATIONS ON THE EFFECTS OF CYCLIC LOADING AND
THERAPEUTIC TREATMENTS FOLLOWING TRAUMATIC INJURY To
ARTICULAR CARTILAGE
By

Nurit Golenberg

A THESIS

Submitted to
Michigan State University
In partial fulﬁllment of the requirements
For the degree of

MASTER OF SCIENCE

Mechanical Engineering

2009

ABSTRACT

Investigations on the effects of cyclic loading and therapeutic treatments following
traumatic injury to articular cartilage

By
Nurit Golenberg

Osteoarthritis (0A) is a disabling disease of synovial joints, such as the hips and
knees, that results in loss of joint function and a reduced quality of life. Joint injuries
during sports related activities, speciﬁcally anterior cruciate ligament (ACL) tears,
greatly increase the risk of developing post-traumatic 0A. Since most patients suffering
ACL tears develop 0A with or without surgical reconstruction of the ligament, the
traumatic loads generated on articular cartilage during the acute injury have been
suggested to be a major cause of post-traumatic 0A. This thesis describes the mechanical
and histological properties of the rabbit tibial plateau using the ﬁbril-reinforced biphasic
model of cartilage. Additionally, investigation of the mechanical properties and
proteoglycan content of bovine chondral explants following two levels of unconﬁned
compressive loading and treatment with a nutraceutical, glucosamine-chondroitin sulfate,
has also been described in this thesis. The effect of cyclic loading was evaluated with and
without glucosamine chondroitin sulfate treatment following two levels of unconﬁned
compression using bovine chondral explants. The rabbit tibial plateau was again utilized
to investigate the long-term effects ofP188, a tn'block copolymer known to acutely repair
damaged cell membranes, following blunt impact to the rabbit tibio-femoral joint. The
data presented in this thesis may be used to investigate the progression of the chronic
joint disease and introduces possible intervention methods for the prevention of

developing CA in the more long term.

Dedication

I would like to thank my parents for their endless love and support throughout my life.

Without their guidance, I would never have dreamed this possible.

iii

Acknowledgments

I would like to thank my advisor and mentor, Dr. Roger Haut, for his guidance
and support throughout my research at the Orthopaedics Biomechanics Laboratories. I am
extremely grateful to Dr. Wright and Dr. Orth for serving on my cormnittee. I would like
to acknowledge Cliff Beckett for his all his technical support. I would also like to
acknowledge Jean Atkinson and Jane Walsh for their endless support, hard work and
dedication. Finally, I would like to thank my fellow graduate students; Dan Isaac, Tim
Baumer, Brian Powell, Eric Meyer, Mark Villwock, Feng Wei, and Jerrod Bramen, for

all their help, support, and most importantly their ﬁ'iendship.

iv

Table of Contents

List of Tables ................................................................................................................. vii

List of Figures ............................................................................................................... viii

Chapter 1

Background and Literature Review ................................................................................. 1

Chapter 2

Histomorphological and mechanical property correlations in rabbit tibial plateau cartilage

based on a ﬁbril-reinforced biphasic model ................................................................... 16
Abstract ............................................................................................................... 16
Introduction ........................................................................................................ 17
Materials and Methods ....................................................................................... 20
Results ................................................................................................................ 27
Discussion ........................................................................................................... 35
References .......................................................................................................... 41

Chapter 3

High levels of glucosamine-chondroitin sulfate can alter articular cartilage stiﬂiress and
up-regulate proteoglycan content of bovine chondral explants following unconﬁned

compression injury ......................................................................................................... 46
Abstract ............................................................................................................... 46
Introduction ........................................................................................................ 47
Materials and Methods ....................................................................................... 49
Results ................................................................................................................ 54
Discussion ........................................................................................................... 60
References .......................................................................................................... 63

Chapter 4

Investigation of low level cyclic loading following high levels of unconﬁned

compression with and without glucosamine chondroitin sulfate treatment ................... 66
Abstract ............................................................................................................... 66
Introduction ........................................................................................................ 67
Materials and Methods ....................................................................................... 69
Results ................................................................................................................ 74
Discussion ........................................................................................................... 76
References .......................................................................................................... 80

Chapter 5

Effects of acute repair of chondrocytes in the rabbit tibio-femoral joint 6 weeks following

blunt impact using P188 surfactant ................................................................................ 82
Abstract ............................................................................................................... 82
Introduction ........................................................................................................ 83
Materials and Methods ....................................................................................... 85

Results ................................................................................................................ 88

Discussion ........................................................................................................... 91
References .......................................................................................................... 96

Chapter 6

Conclusions and Recommendations for Future Work .................................................... 99

APPENDICES .............................................................................................................. 102
Appendix A: Raw data from chapter 1 ............................................................. 102
Appendix B: Raw data from chapter 2 ............................................................. 107
Appendix C: Raw data from chapter 3 ............................................................. 111
Appendix D: Raw data ﬁorn chapter 4 ............................................................. 113

List of Tables

Table A.1 Mechanical properties of the control rabbits in site 1 ............................... 103
Table A2 Mechanical properties of the control rabbits in site 2. .............................. 104
Table A3 Mechanical properties of the control rabbits in site 3 ............................... 105
Table A.4 Mechanical properties of the control rabbits in site 4. .............................. 106
Table 3.1 Mechanical properties following 10 MPa of unconﬁned compression. Matrix
Modulus (a) Fiber Modulus (b) Permeability (c). ..................................... 108
Table 3.2 Mechanical properties following 25 MPa of unconﬁned compression. Matrix
Modulus (a) Fiber Modulus (b) Permeability (c). ..................................... 109
Table 3.3 Matrix Modulus following 10 MPa (a) and 25 MPa of unconﬁned
compression with glcN-CS treatment. ...................................................... 110
Table 3.4 Proteoglycan content with and without glcN-CS treatment. ..................... 110
Table C.l Matrix modulus following unconﬁned compression and exercise. ........... 112

Table C.2 Matrix modulus following unconﬁned compression and exercise with glcN-

CS treatment .............................................................................................. 112
Table C.3 Proteoglycan content with and without glcN-CS treatment ...................... 112
Table D.l Live cell density analysis of the P188 treated rabbits (cells/mmZ). ........ 114
Table D2 Live cell density analysis of the untreated rabbits (cells/mmz) ................ 114

vii

Figure 1.1

Figure 1.2

Figure 2.1

Figure 2.2

Figure 2.3

Figure 2.4

Figure 2.5

Figure 2.6

Figure 2.7

Figure 2.8

List of Figures

Cartilage is comprised of water and a matrix of collagen ﬁbers, chondrocytes,
and a proteoglycan complex. ........................................................................ 2

The collagen ﬁber orientation is depth dependent and can be divided into 3
zones. The superﬁcial zone is roughly the top 20% of the cartilage and
contains collagen ﬁbers that are tangential to the cartilage surface. The
middle zone is roughly the middle 50% of the cartilage and the ﬁbers are
oriented randomly. The collagen ﬁbers in the deep zone, bottom 30% are
perpendicular to the subchondral bone surface. ............................................ 6

Indentation relaxation test sites located on the rabbit tibial plateau. Sites 1
and 3 correspond to the areas not covered by the meniscus on the medial and
lateral facets, respectively. Sites 2 (medial) and 4 (lateral) are the areas
covered by the meniscus ............................................................................. 21

Photograph of the indentation test ﬁxture. The X-Y mounting plate allows for
left/right or forward/backward placement, and the Z plate allows for up.down
placement. The camera mount allowed for rotation of the sample to set the
surface perpendicular to the indenter .......................................................... 22

Histomorphometric scoring system used to quantify the characteristics for
cartilage across the tibial plateau. ............................................................... 26

Gross photographs of the tibial surface, stained with India ink to highlight
ﬁssures. Surface irregularities were analyzed, comparing the medial and
lateral facets. ............................................................................................... 27

Theoretical curves were calculated to closely ﬁt the data collected dming
experimental testing. ................................................................................... 28

a) Sensitivity graphs for the mechanical parameters were developed. b) The
permeability and M were further analyzed by plateau location .................. 30

Media] and lateral cartilage thickness and mechanical parameter results ﬁ'om

the indentation relaxation test. The mean (bolded) and standard deviation (in
parentheses) are given in this table ............................................................. 33

Histological score results from the medial and lateral facets. Results are
given as the mean and range of data in parentheses. .................................. 33

viii

Figure 2.9

Figure 3.1

Figure 3.2

Figure 3.3

Figure 3.4

Figure 3.5

Figure 3.6

Histological images of the articular cartilage and underlying bone show
signiﬁcant differences between the medial and lateral facets. The zone of
calciﬁed cartilage showed more intense staining for tissue proteoglycans
medial (a) versus the lateral (b) facet, using a Safranin O-Fast Green protocol.
The cartilage on the medial facet (c) showed more disruptions than the lateral
facet (d). The tibial plateau showed more surface disruptions, including
ﬁssuring, on the medial (e) versus the lateral facet (f) ................................ 34

Explant indentation test system and ﬁxture (A). The explants were placed in
a hole in the bottom magnet on a ﬂat steel surface (B). A top magnet was
lowered over the top of the explant to hold down the edges (C). The spherical
indentor tip was lowered to a preload of 0.05 N (D) .................................. 52

Cartilage explants were loaded (10 MPa or 25 MPa) in unconﬁned
compression between two polished stainless steel plates ........................... 53

Gross analysis of the cartilage explants revealed surface lesions on both the
10 MPa (A) and 25 MPa (B) samples with more ﬁssures on the 25 MPa
samples. These samples also appeared elliptical in shape. ........................ 54

The matrix modulus was determined from the stress-relaxation curves
documented during each indentation test. The matrix modulus presented here
is given as a percent of the original property prior to unconﬁned compression.
A decrease, compared to the initial property, in the matrix modulus was
documented following both 10 MPa and 25 MP3 of unconﬁned compression.
Signiﬁcantly lower matrix modulus was documented following 25 MPa of
unconﬁned compression versus 10 MPa of unconﬁned compression. ‘*’
represents statistical signiﬁcance compared to 10 MPa samples ................ 55

The permeability was determined from the stress-relaxation curves
documented during each indentation test. The permeability presented here is
given as a percent of the original property prior to compression. A signiﬁcant
increase in permeability was documented following 25 MPa unconﬁned
compression compared to 10 MPa unconﬁned compression. ‘*’ represents
statistical signiﬁcance compared to 10 MPa samples. ................................ 56

The ﬁber modulus was determined from the stress-relaxation curves
documented during each indentation test. The ﬁber presented here is given as
a percent of the original property prior to compression. An increase in the
ﬁber modulus was documented following 25 MPa unconﬁned compression
compared to original property and this increase was signiﬁcantly higher than
samples following 10 MPa of unconﬁned compression ‘*’ represents
statistical signiﬁcance compared to the 10 MPa samples ........................... 57

ix

Figure 3.7 Proteoglycan content in samples was determined using DMB assay. Results

Figure 3.8

are shown here as pg PG per mg wet weight. No statistical difference was
documented in the PG content between 10 MPa and 25 MPa of unconﬁned
compression ................................................................................................ 58

The matrix modulus of the 10 MPa samples with and without glcN-CS was
determined ﬁ'om stress-relaxation curves. A signiﬁcant increase in the matrix
modulus was documented following treatment with glcN-CS compared to

untreated samples. ‘*’ denotes statistical signiﬁcance compared to samples
treated with glcN-CS ................................................................................... 59

Figure 3.9 The matrix modulus of the 25 MPa samples with and without glcN-CS was

determined from stress-relaxation curves. An increase in the matrix modulus
was documented following treatment with glcN-CS compared to untreated
samples; however, statistical signiﬁcance was not reached ........................ 59

Figure 3.10 Proteoglycan content in all samples was determined using DMB assay.

Figure 4.1

Figure 4.2

Figure 4.3

Figure 4.4

Results are shown here as ug PG per mg wet weight. A signiﬁcant increase
in PG content was documented in samples treated with glcN-CS compared to
untreated samples. ‘*’ denotes statistical signiﬁcance compared to glcN-CS
treated samples. ........................................................................................... 60

Explant indentation test system and ﬁxture (A). The explants were placed in
a hold of the bottom magnet on a ﬂat steel surface (B). A top magnet was
lowered over the top of the explant to hold down the edges (C). The indentor
tip was lowered to a preload of 0.05 N (D). ................................................ 71

Cartilage explants were loaded in unconﬁned compression at either 10 MPa
or 25 MPa between two polished stainless steel plates ............................... 72

The “cartilage exerciser” mechanical loading device applied compressive
loads to the cartilage explants in 12 separate loading chambers in a 24 well
plate. The samples were cyclically loaded 10 times with a peak stress of 0.5
MPa followed by 3600 seconds of rest. This protocol was then repeated for
the duration of the test. ................................................................................ 73

The matrix modulus following 10 MPa (a) and 25 MPa (b) of unconﬁned
compression and low level cyclic loading with and without treatment with
glcN-CS supplement. No differences were documented with the treatment of
glcN-CS following 10 MPa of unconﬁned compression, while an increase
was documented following 25 MPa of unconﬁned compression and cyclic
loading with glcN-CS treatment ................................................................. 75

Figure 4.5

Figure 4.6

Figure 4.7

Figure 5.1

Figure 5.2

Figure 5.3

Figure 5.4

PG content of samples with and without glcN-CS supplement. A signiﬁcant
increase in the tissue PG content was documented with the treatment of glcN-
CS. ‘*’ denotes a statistically signiﬁcant difference compared to samples
treated with glcN-CS ................................................................................... 76

The matrix modulus following 10 MPa (a) and 25 MPa (b) of unconﬁned
compression with and without exercise (from Chapter 2). No differences
were documented due to cyclic loading in either of the loading groups ..... 78

The PG content following 10 MPa and 25 MPa of unconﬁned compression
with and without cyclic loading. NO signiﬁcant differences were documented
between the samples with or without cyclic loading. ................................. 79

Impact experiments were conducted by dropping a gravity-accelerated mass
onto the ﬂexed knee so that impacts were isolated on the TF joint ............ 86

A custom cartilage cutting device was used to prepare slices of tissue for cell
viability. ...................................................................................................... 88

No signiﬁcant differences were documented in the density of viable cells
between the impacted and unimpacted limbs in the lateral femoral chondyle
(LFC), medial femoral chondyle (MFC), lateral tibial plateau (LTP) or
medial tibial plateau (MTP) of the treated (a) and untreated (b) rabbits ....90

Analysis of the impacted and unimpacted limbs revealed a signiﬁcant
difference in the density of viable cells in the ‘no P1 88’ group. No signiﬁcant
difference, however, was noted in the ‘P188’ group. This suggests that P188
had a beneﬁt in preventing the long term degeneration of acutely injured
chondrocytes in the TF joint. ‘*’ indicates a statistically signiﬁcant difference
between the impacted and the unimpacted limbs ........................................ 91

xi

Chapter 1

Background and Literature Review

Osteoarthritis (0A) is a disabling disease of joints, such as the hips and knees,
that results in loss of joint function and a reduced quality of life. OA affects almost 27
million Americans each year (Helrnick et al. 2008) with a total annual cost per OA
patient set at nearly $5700 per year (Maetzel et al., 2004). Injuries during sports related
activities, speciﬁcally anterior cruciate ligament (ACL) tears, increase the risk of
developing OA (Gelber et al., 2000). ACL tears are associated with large compressive
forces passing through the joint (Fang et al., 2001). These forces are associated with a
signiﬁcant frequency of occult bone lesions and changes to the overlying articular
cartilage homeostasis within 6 months of injury (V ellet et al., 1991; Johnson et al., 1998).
Since nearly all patients suffering ACL tears develop OA within 15 years of trauma with
or without ACL reconstruction (Myklebust and Bahr, 2005), the traumatic loads to the
articular cartilage during injury have been suggested to be the cause of post-traumatic
0A.

0A is characterized by increases in cartilage hydration, subchondral bone
changes, altered chondrocyte activity, and changes in the structure and composition of
proteoglycans and collagen (Mow and Setton, 1998). The process of joint OA involves
softening of articular cartilage followed by ﬁbrillation and subsequently, a total loss of
the tissue. Because articular cartilage is responsible for the distribution of loads applied to

a joint as well as providing a low friction surface for joint movement (Mow et al., 1980),

the loss of articular cartilage ultimately leads to failure of synovial joints (Badley, 2005;
Sangha, 2000).

The loss of articular cartilage function can be related to changes in the individual
constituents. Articular cartilage is comprised of water (60-80% of tissue’s wet weight)
and a matrix (20-40% of tissue’s wet weight) of cartilage cells or chondrocytes (less than
2%), proteoglycans (PG) (40%) and collagen ﬁbers (60%) (Mow et al., 1980) (Figure

1.1).

Collagen
Chondrocyte

  
   

Proteoglycan
Interstitial Water

Figure 1.1. Cartilage is comprised of water and a matrix of collagen ﬁbers, chondrocytes,
and a proteoglycan complex.

Chondrocytes are responsible for the production and maintenance of the cartilage
matrix. Changes in the cell membrane may be responsible for changes in the cellular
function. The cell membrane is comprised of a lipid bilayer and protein channels. The
membrane and protein channels, such as the Ca+ and Na+/K+ pumps, regulate the ionic
and osmotic environment inside the cell. Damage to the membrane or ion channels ﬁ'om
excessive compressive loads may cause the cell to lose its ionic and osmotic equilibrium.
This results in swelling and overall lysis, a deﬁning feature of necrotic cell death.
Traumatic loading to articular cartilage results in apoptotic cell death. Apoptosis results

in the fragmentation of the nucleus and eventually cell death. An apoptotic cell contains

ap0ptotic enzymes that, when released, signal surrounding cells and initiate apoptosis in
these cells (Levin et al. 2001). Previous studies have documented cell loss due to
traumatic compressive loads and clinically, chondrocyte loss is documented in the
cartilage of OA patients; therefore, cell death has been a focus of OA research. Colwell et
al. (2001) suggest that it may be possible to limit degeneration and promote repair of the
cartilage if cell viability is maintained.

While traumatic loading results in cell death, low level forces can cause changes
in the ion channels resulting in increased cellular biosynthesis. Mechanical loads are
converted into signals through the matrix which can alter cellular behavior by changing
the ionic or osmotic environment of the chondrocyte (Wilkins et al., 2000). The increase
in biosynthesis increases the synthesis of matrix proteins, speciﬁcally proteoglycans, and
therefore increases the mechanical pTOperties of cartilage and its resistance to traumatic
overloads (Wei and Haut, 2009).

The matrix, comprised of PGs and collagen, provides cartilage with its
compressive resistance to stress and strain overloads. The proteoglycans are comprised of
a core protein with glycosarninoglycan (GAG) side chains (Mow and Setton, 1998).
These GAG chains have a ﬁxed negative charge that generates a repulsive force causing
tension in the cartilage. However, at equilibrium, the swelling pressure due to the PGs is
balanced by the forces generated by the collagen. Collagen is a triple helix of polypeptide
chains. The orientation of collagen ﬁbers in cartilage is depth dependent. The cartilage is
comprised of three layers: the deep zone, middle zone, and superﬁcial zone. The deep
zone is roughly the bottom 30% of its thickness. The ﬁbers in this region are

perpendicular to the subchondral bone surface. The superﬁcial zone is roughly comprised

 

of the top 20% of the cartilage thickness. The collagen ﬁbers are tangential to the surface
of the cartilage and provide a majority of its tensile properties (Mow et al., 1980) (Figure
1.2). The ﬁbers in the middle or transitional zone, about the middle 50% of the cartilage,
are randomly oriented since the ﬁbers are in transition from the deep zone to the

superﬁcial zone.

Zones

Superﬁcial tangential _[
(1 0-20%)

 
 
      
 

Middle (40—60%)
Deep (30%) .. Tide mark
. . - ,j‘. =--’~:;;.___ Subchondral bone
Calcrﬁed CaItllage 6253K Cancellous bone

Figure 1.2. The collagen ﬁber orientation is depth dependent and can be divided into 3
zones. The superﬁcial zone is roughly the top 20% of the cartilage and contains collagen
ﬁbers that are tangential to the cartilage surface. The middle zone is roughly the middle
50% of the cartilage and the ﬁbers are oriented randomly. The collagen ﬁbers in the deep
zone, bottom 30% are perpendicular to the subchondral bone surface.

Morphological changes, such as ﬁbrillation and tissue swelling, are associated
with the initiation of OA. Speciﬁcally, ﬁbrillation is due to the unwinding of the collagen
ﬁbers and surface fraying. In studies using cartilage explants collagen damage has been
Observed by surface ﬁssuring (Repo and Finlay, 1997) and the cartilage becomes
ﬂattened and elliptical in shape (Jeffrey et al., 1995; Kaab et al., 2000). Additionally, the

collagen ﬁbers deform under the compressive loads and, under traumatic loading

conditions, the ﬁbers remain in their deformed shape (McCall, 1969). Damage to the

collagen ﬁbers allows the tissue to swell. Swelling of the tissue has been associated with
the softening of the tissue. This tissue damage can also be associated with a decrease in
the PG content of the tissue (Patwari et al., 2000; Huser and Davies, 2006). This
softening, due to ﬁbrillation and swelling, leads to increased pressure on the underlying
subchondral bone (Radin et al., 1986; Ewers et al., 2001).

In vivo animal models have been used to study the changes in articular cartilage
following both injury and exercise. Our laboratory has previously developed a small
animal model to study articular cartilage in vivo using Flemish Giant rabbits (Haut et a1.
1995). In these previous studies, an impact to the rabbit knee joint results in cartilage
surface ﬁssures and decreases in cell viability within 4 days of traumatic injury (Rundell
et al., 2005; Isaac et al., 2008) and thickening of the underlying subchondral bone and
thinning of the overlying articular cartilage compared to the unimpacted limb 3 years
post-trauma (Ewers et al., 2001). On the other hand, exercise increases matrix stiffness
and PG content in this tissue. Previous studies using a canine model document an
increase in the tissue PG content (Kiveranta et al., 2005), and an increase in its
compressive stiffness (Jurvelin et al., 1986) speciﬁcally in regions Of high loading
following regular exercise. A study using a Flemish Giant rabbit model documents that
post-traumatic exercise can help prevent the loss of matrix PGs (Weaver and Haut, 2005),
but the mechanism of its action was not investigated.

Cartilage explants have also been used to document the changes in the cartilage
following compressive loads. A decrease in the cell viability and PG content was
documented following traumatic loading and was found to be dependent on load levels

and strain rates (Torzilli et al., 1999). Previous studies document that loading above a

critical threshold of 15-20 MPa (Torzilli et al., 1999) causes increased cell death (Phillips
and Haut, 2004; Baars et al., 2006; Natoli and Athanasiou, 2008) and permanent damage
to the collagen network (Torzilli et al., 1999). Thibault et al. (2002) documents that 40
cycles of 2-8 MPa causes decreased mechanical properties, denatured collagen and loss
of PG without producing ﬁssures and cell loss. Cartilage explants have also been used to
document the effects of low level loading, below this critical threshold, on both the
mechanical and biological properties of the tissue.

Low level mechanical forces applied to cartilage may also cause changes in ion
channels of cell membranes. Deformation of the membrane channels may open pathways
allowing for changes in the ionic or osmotic environment of the cell, causing changes in
cellular metabolism (Wilkins et al., 2000). Single (Torzilli et al., 1999) and cyclic (Wei et
al., 2008) low level compression increase PG synthesis in some previous studies. This
increase in tissue PG correlates with an increase in tissue stiffness and its resistance to
traumatic compression (Wei et al., 2008). Cyclic loading increases the incorporation of
nutraceuticals, such as glucosamine-chondroitin sulfate, in the cells (Wei and Haut, 2009,
Sharma et al., 2008). Since excessive unconﬁned compression decreases the PG content
of cartilage and decrease its mechanical stiffness pos-trauma, intervention strategies have
become a focus of post-traumatic osteoarthritis studies.

Glucosamine and chondroitin sulfate (glcN-CS) have been used in attempts to
maintain mechanical properties of injured cartilage in recent years (Wei and Haut, 2009;
Shanna et al., 2008, Tiraloche et al., 2005). Glucosamine is a basic building block of
proteoglycans (Dodge and Jimenez, 2003), and chondroitin sulfate (CS) increased the

synthesis of proteoglycans and hylauronic acid (Morreale et al., 1996). Previous studies,

by others, have shown that bathing cartilage explants in a supplement of glcN-CS can up-
regulate the synthesis of tissue PG’s (Lippiello, 2003). Therefore, the recovery or
prevention of PG loss after trauma with the treatment of glcN-CS has been suggested to
maintain the mechanical stifﬁress of the cartilage and prevent overall loss of this tissue in
the joint.

Other pharmaceutical treatments have been investigated to target cell damage and
repair. Treatment with poloxamer 188 (P188) has been investigated as a means of
repairing the damaged chondrocytes following traumatic injury in hopes of preventing
the degeneration of joint cartilage. P188 is an 8400-dalton triblock copolymer containing
both hydrophobic and hydrophilic regions, which inserts into only the damaged areas of
cell membranes (Marks et al., 2001). Because of its low toxicity, P188 has been used
clinically in recent studies following brain trauma. The studies suggest that P188 can help
‘save’ neurons from developing early necrotic death following severe mechanical loading
(Barbee et al., 1992; Marks et al., 2001). Acute studies have been conducted recently to
determine the efﬁcacy ofP188 in repairing damaged cell membranes following traumatic
injury to articular cartilage. Speciﬁcally, Rundell et al. (2005), using the patello-femoral
joint, and Isaac et al. (In Review), using the tibio-femoral joint, have documented
increases in the percentage of viable cells in impacted limbs treated with P188 up to 4
days following a traumatic overloading of the joint. Similarly, in vitro studies have
documented the ability of P188 to acutely repair damaged membranes in chondrocytes 7
days after a traumatic overload (Baars et al., 2006; Phillips and Haut, 2004; Natoli and

Athanasiou, 2008). These results suggest that P188 may provide a possible treatment for

 

the prevention of post-trauma induced OA of the joint. However, the long term efﬁcacy
ofP188 and the effect of saving these cells is yet unknown in the current literature.

To study the effects of the degenerative changes in joint cartilage, various
computational models, such as the linear elastic and linear biphasic models of cartilage,
have been developed to extract the mechanical properties from creep and indentation-
relaxation tests. These tests have been used to extract properties initially and at
equilibrium. Speciﬁcally, the linear elastic model has been used to determine the
instantaneous and equilibrium modulii after assuming Poisson’s ratio of the tissue
(Parsons and Black, 1977). More recently, Jin and Lewis (2004) have developed a two-
punch test to extract the effective Poisson’s ratio of the tissue. The linear elastic model,
however, is limited in evaluating the mechanical properties only under small loads and
small strains. The model also does not incorporate the ﬂuid ﬂow characteristics of the
tissue. On the other hand, the linear biphasic model has been used to study changes in the
linear elastic equilibrium modulus, Poisson’s ratio, and the tissue permeability of
cartilage (Mow et al. 1980). Yet, it is unable to evaluate the instantaneous response of the
cartilage to a compressive load. The limitations of these models have resulted in
numerous inconsistencies in the data For example, using the linear biphasic model on the
rabbit tibial plateau, a previous study documents a lower aggregate modulus, extracted
ﬁom the equilibrium data, on the medial than lateral facet in areas covered by the
meniscus (Roemhildt et al. 2006). In contrast to these ﬁndings, a previous study using the
linear elastic model, suggested a higher creep modulus in medial compared to lateral
facet in regions partially covered by the meniscus in the rabbit tibial plateau (Rasanen

and Messner, 1996; Wei, Rasanen, and Messner, 1998). The instantaneous properties

correlate with the integrity of the collagen matrix while the equilibrium properties
correlate with the network of proteoglycans (J ulkunen et al., 2009). The content of
proteoglycans is inversely proportional to tissue permeability (Maroudas, 1979). The
correlations between morphological and mechanical properties of cartilage have often
been confusing and contradictory possible because of limitations in the computational
models. For example, using the linear biphasic model the equilibrium modulus and
Poisson’s ratio were found to be lower in the medial versus lateral compartment of the
tibial plateau (Roemhildt et al., 2006). The equilibrium modulus is consistent with
previous ﬁndings that suggest the medial compartment to be more degraded with slightly
rougher surfaces and/or the presence of superﬁcial or deep splits. These morphological
results suggest that the collagen network in cartilage from the medial compartment of the
rabbit tibial plateau may be more degraded than in the lateral compartment (Pelletier et
al., 1983; Bank et al., 2000) and that Poisson’s ratio should be larger in that compartment
(Kiviranta et al., 2006). Recent studies suggest that a more complex model of cartilage,
which incorporates a collagen ﬁber network within the matrix, may more accurately ﬁt
the creep and relaxation response of cartilage under unconﬁned and conﬁned
compression as well as indentation testing (Wilson et al. 2005), and therefore better
correlate with histological and morphological data. The ﬁbril-reinforced biphasic model
incorporates both the instantaneous and equilibrium parameters as well as the tissue
permeability and can evaluate these parameters using a single indentation test. However,
little information using this model has currently been documented.

This thesis describes investigations on articular cartilage in both in vivo and in

vitro scenarios using a previously established rabbit model and bovine chondral explants,

 

respectively. Using the ﬁbril-reinforced biphasic model of cartilage, Chapter 2 will
document the mechanical properties of the rabbit tibial plateau in the medial and lateral
facets in areas covered and uncovered by the meniscus. Histological and morphological
characteristics of the tibial plateau cartilage were documented and correlated with the
mechanical properties generated in the model. Chapter 3 will investigate the mechanical
properties of bovine chondral explants following 10 MPa and 25 MPa of unconﬁned
compression. Additionally, treatment of the explants with glcN-CS following unconﬁned
compression will be investigated. Using this model Chapter 4 will investigate the effect
of low level cyclic loading post-trauma on the chondral explants. GlcN-CS incorporation
will also be investigated following low level cyclic loading on the injured cartilage
explants. In Chapter 5, cell viability analysis is used to investigate the long term efﬁcacy
of a single injection of P1 88 to ‘save’ cells in the rabbit TF joint following a single

traumatic impact to the joint.

10

References

Baars, D.C., Rundell, S.A., Haut, RC. (2006) Treatment with the non-ionic surfactant
poloxamer P188 reduces DNA fragmentation in cells from bovine chondral explants
exposed to injurious unconﬁned compression. Biomech Model Mechanobiol 5, 133-139.

Badley, EM. (2005) The impact of disabling arthritis. Arthritis and Rheumatism 8, 221-
228.

Bank, R.A., Soudry, M., Maroudas, A., Mizrahi, J., and TeKoppele, J.M. (2000) The
increased swelling and instantaneous deformation of osteoarthritic cartilage is highly
correlated with collagen degradation. Arthritis and Rheumatism 43, 2202-2210.

Barbee, K., Ford, C., Blackrnan, B., Thibault, L. (1992) Neural cell injury:
characterization and treatment strategy In 2nd Injury Prevention Through Biomechanics
Symposium Proceeding, CDC. Edited by Yang K.

Colwell, C.W., D’Lirna, D.D., Hoenecke, H.R., Fronek, J ., Pulido, P., Monis, B.A.,
Chung, C., Resnick, D., Lotz, M. (2001) In vivo changes after mechanical injury. Clin.
Orthop. Rel. Res. 3918, 8116-8123.

Dodge, GR. and Jimenez, SA. (2003) Glucosamine sulfate modulates the levels of
aggrecan and matrix metalloproteinase-3 synthesized by cultured human osteoarthritis
articular chondrocytes. Osteoarthritis and Cartilage 11, 424-432.

Ewers, B.J., Weaver, B.T., Sevensma, B.T., Haut, R.C., (2001) Chronic changes in rabbit
retro-patellar cartilage and subchondral bone after blunt impact loading of the
patellofemoral joint. Journal of Orthopaedic Research 20, 545-550.

Fang, C., Johnson, D., Leslie, M.P., Carlson, C.S., Robbins, M., Di Cesare, PE. (2001)
Tissue distribution and measurement of cartilage oligomeric matrix protein in patients
with magnetic resonance imaging-detected bone bruises after acute anterior cruiate
ligament tears. Journal of Orthopaedic Research 19, 634-641.

Gelber, A.C., Hochberg, M.C., Mead, L.A., Wang, N-Y., Wigley, F.M., Klag, M.J.,
(2000) Joint injury in young adults and risk for subsequent knee and hip osteoarthritis.
Annals of Internal Medicine 133, 321-328.

Haut, R.C., Ide, T.M., and DeCamp, CE. (1995) Mechanical responses of the rabbit
petello-femoral joint to blunt impact. Journal of Biomechanical Engineering 117, 402-
408.

Helrnick, C.G., F elson, D.T, Lawrence, R.C., Gabriel, S., Hirch, R., Kwoh, C.K., Liang,
M.H., Kremers, H.M., Mayes, M.D., Merkel, P.A., Pillemer, S.R., Reveille, J.D., Stone,
J.H. (2008) Estimates of the prevalence of arthritis and other rheumatic conditions in the
United States. Arthritis and Rheumatism 58, 15-25.

11

Huser, C.A.M., Davies, ME. (2006) Validation of an in vitro single-impact load model
of the initiation of osteoarthritis-like changes in articular cartilage. Journal of
Orthopaedic Research 24, 725-732.

Isaac, D.I., Golenberg, N., Haut, R.C. Acute repair of chondrocytes in the rabbit
tibiofemoral joint following blunt impact using P188 surfactant and a preliminary
investigation of its long-term efﬁcacy. Journal of Orthopaedic Research (In Review).

Isaac, D.I., Meyer, E.G., Haut, RC. (2008) Chondrocyte damage and contact pressures
following impact on the rabbit tibiofemoral joint. J Biomech Eng 130, 0410181-5.

Jeffrey, J.E., Gregory, D.W., Aspden, RM. (1995) Matrix damage and chondrocyte
viability following a single impact load on articular cartilage. Archives of Biochemistry
and Biophysics 322, 87-96.

Jin, H. and Lewis, IL (2004) Determination of Poisson’s ratio of articular cartilage by
indentation using different-sized indenters. Journal of Biomechanical Engineering 126,
1 3 8-145.

Johnson, D.L., Urban, W.P., Caborn, D.N.M., Vanarthos, W.J., Carlson, CS (1998)
Articular cartilage changes seen with magnetic resonance imagine-detected bone bruises
associated with acute anterior cruciate ligament rupture. The American Journal of Sports
Medicine 26, 409-414.

Julkunen, P., Harjula, T., Marjanen, J., Helminen, H.J., Jurvelin, J .S. (2009) Comparison
of single-phase isotropic elastic and ﬁbril-reinforced poroelastic models for indentation
of rabbit articular cartilage. Journal of Biomechanics 42, 652-656.

Jurvelin, J., Kiviranta, I., Tammi, M., Helminen, H.J., (1986) Effect of physical exercise
on indentation stiffness of articular cartilage in the canine knee. Int. J. Sports Med. 7,
106-1 10.

Kaab M.J., Ito, K., Rahn, B., Clark, J .M., Notzli, HP. (2000) Effect of mechanical load
on articular cartilage collagen structure: a scanning electron-microscope study. Cells
Tissues Organs 167, 106-120.

Kiviranta, P., Rieppo, R.K., Julunen, P., Toyras, J ., and Jurvelin, J.S. (2006) Collagen
network primarily controls Poisson’s ratio of bovine articular cartilage in compression.
Journal of Orthopaedic Research 24, 690-699.

Levin, A., Burton-Wurster, N., Chen, C.T., Lust, G. (2001) Intercellular signaling as a

cause of cell death in cyclically impacted cartilage explants. Osteoarthritis and Cartilage
9, 702-711.

12

 

Lippiello, L. (2003) Glucosamine and chondroitin sulfate: biological response modiﬁers
of chondrocytes under simulated conditions of joint stress. OsteoArthritis and Cartilage
11, 335-342.

Maetzel, A., Li, L.C., Pencharz, J ., Tomlinson, G., Bombardier, C. (2004) The economic
burden associated with osteoarthritis, rheumatoid arthritis, and hypertension: a
comparative study. Annals of the Rheumatic Diseases 63, 395-401.

Marks, J .D., Pan, C.Y., Bushell, T., Cromie, W., Lee, RC. (2001) Amphiphilic tri-block
copolymers provide potent membrane-targeted neuroprotection. FASEB 15, 1107-1 109.

Maroudas, A. (1979) A physical properties of articular cartilage. In: Freeman, M.A.R.
(ed) Adult articular cartilage. Kent, NY: Pitrnan medical, 215-290.

McCall, J .G. (1969) Load-deformation studies of articular cartilage. Journal of Anatomy
105, 212-214.

Morreale, P., Manopulo, R., Galati, M., Boccanera, L., Saponati, G., Bocchi, L. (1996)
Comparison of anti-inﬂammatory efﬁcacy of chondroitin sulfate and diclofenac sodium
in patients with knee osteoarthritis. Journal of Rheumatology 23, 1385-1391.

Mow, V.C., Roth, V., Armstrong, CG. (1980) Biomechanics of joint cartilage. In:
Frankel, V.H., Nordin, M. (eds) Basic Biomechanics of the Skeletal System. Philadelphia,
PA: Lea & Febiger, 63-85.

Mow, V.C. and Setton, LA. (1998) Mechanical properties of normal and osteoarthiric
cartilage. In: Brandt, K.D., Doherty, M., Lohmander, L.S., (eds). Osteoarthritis. Oxford:
Oxford University Press, 108—122.

Myklebust, G. and Bahr, R. (2005) Return to play guidelines after anterior cruciate
ligament surgery. British Journal of Sports Medicine 39, 127-131.

Natoli, R.M., Athanasiou, K.A. (2008) P188 reduces cell death and IGF-1 reduces GAG
release following single-impact loading of articular cartilage. J Biomech Eng 130,
041012-1-9.

Parsons, JR. and Black, J. (1977) The viscoelastic shear behavior of normal rabbit
articular cartilage. Journal of Biomechanics 10, 21-29.

Patwari, P., Kurz, B., Sandy, J.D., Grodzinsky, A.J., (2000) Mannosarnine inhibits

aggrecanase-mediated changes in physical properties and biochemical composition of
articular cartilage. Archives of Biochemistry and Biophysics 374, 79-85.

13

Pelletier, J.P., Martel-Palletier, J., Altman, R.D., Ghandur/Mnaymneh, L., Hower, D.S.,
and Woessner, J.P., (1983) Collagenolytic activity and collagen matrix breakdown of the
articular cartilage in de Pond-Nuki dog model of osteoarthritis. Arthritis and Rheumatism
26, 866-874.

Phillips, D.M. and Haut, RC. (2004) The use of non-ionic surfactant (P188) to save
chondrocytes ﬁ'om necrosis following impact loading of chondral explants. Journal of
Orthopaedic Research 22, 1135-1142.

Radin, E., Rose, R. (1986) Role of Subchondral Bone in the Initiation and Progression of
Cartilage Damage. Clin Orthop Rel Res 213, 34-40.

Rasanen, T. and Messner, K. (1996) Regional variations of indentation stiffness and
thickness of normal rabbit knee articular cartilage. Journal of Biomedical Materials
Research 31, 519-524.

Repo, R.U., Finlay, J .B. (1977) Survival of articular cartilage after controlled impact.
Journal of Bone and Joint Surgery-American Volume 59, 1068-1076.

Roemhildt, M.L., Coughlin, K.M., Peura, G.D., Fleming, BC, and Beynnon, ED.
(2006) Material properties of articular cartilage in the rabbit tibial plateau. Journal of
Biomechanics 39, 2331-2337.

Rundell, S.A., Baars, D.C, Phillips, D.M., Haut, RC. (2005) The limitation of acute
necrosis in retro-patellar cartilage after a severe blunt impact to the in vivo rabbit
patellofemoral joint. Journal of Orthopaedic Research 23, 1363-1369.

Sangha, O. (2000) Epidemiology of rheumatic diseases. Rheumatology 39, 3-12.

Sharma, G., Sazena, R.K., Mishra, P. (2008) Synergistic effect of chondroitin sulfate and
cyclic pressure on biochemical and morphological properties of chondrocytes from
articular cartilage. Osteoarthritis and Cartilage 16, 1387-1394.

Thibault, M., Poole, A.R., Buschmann, MD. (2002) Cyclic compression of
cartilage/bone explants in vitro leads to physical weakening, mechanical breakdown of

collagen and release of matrix ﬁagments. Journal of Orthopaedic Research 20, 1265-
1273.

Torzilli, P.A., Grigiene, R., Borrelli, J. Jr., Helfet, D.L. (1999) Effect of impact load on
articular cartilage: cell metabolism and viability, and matrix water content. Journal of
Biomechanical Engineering 121, 433-441.

Vellet, A.D., Marks, P.H., Fowler, P.J., Munro, TC. (1991) Occult posttraumatic
osteochondral lesions of the knee: prevalence, classiﬁcation and short-term sequelae
evaluated with MR imagining. Radiology 178, 271-276.

14

 

Weaver, B.T. and Haut, R.C. (2005) Enforced Exercise after blunt trauma signiﬁcantly
affects biomechanical and histological changes in rabbit retro-patellar cartilage. Journal
of Biomechanics 38, 1 177-1 183.

Wei, X., Rasanen, T., and Messner, K. (1998) Maturation-related compressive properties
of rabbit knee articular cartilage and volume ﬁaction of subchondral tissue. Osteoarthritis
and Cartilage 6, 400-409.

Wei, F., Golenberg, N., Kepich, B.T., Haut, R.C. (2008) Effect of intermittent cyclic
preloads on the response of articular cartilage explants to an excessive level of
unconﬁned compression. Journal of Orthopaedic Research 26, 1636-1642.

Wei, F. and Haut, R.C. (2009) High levels of glucosamine-chondroitin sulfate can alter
the cyclic preload and acute overload responses of chondral explants. Journal of
Orthopaedic Research 27, 353-359. -

Wilkins, R.J., Browning, J .A., Urban, J PG. (2000) Chondrocyte regulation by
mechanical load. Bioheology 37, 67-74.

Wilson, W., van Donkelaar, C.C., van Rietbergen, B., and Huiskes, R. (2005) ‘A ﬁbril-

reinforced poroviscoelastic swelling model for articular cartilage’, Journal of
Biomechanics 38, 1 195-1204.

15

Chapter 2
Histomorphological and mechanical property correlations in rabbit tibial plateau
cartilage based on a fibril-reinforced biphasic model
Abstract
The rabbit knee joint has ﬁequently been used to study the mechanical properties

of articular cartilage using various computational models. These models, however, vary
in their ability to extract comparable material properties of cartilage from experimental
data, resulting in inconsistencies in the data ﬁ'om various laboratories. A more complex,
ﬁbril-reinforced biphasic model more accurately ﬁts the response of the cartilage by
incorporating a collagen ﬁber network within the matrix. Indentation-relaxation tests
were conducted on the medial and lateral facets in areas covered and uncovered by the
meniscus on the rabbit tibial plateau. Gross and histological data were analyzed and
correlated with the mechanical properties. The ﬁbril-reinforced biphasic model
accurately ﬁt the entire experimental curve extracting both the instantaneous (ﬁber ‘
modulus) and equilibrium (matrix modulus) responses, and tissue permeability. The data
was found signiﬁcant at all sites. Signiﬁcant correlations were also documented between

the mechanical and the histological and morphological data.

16

 

Introduction

The rabbit knee joint has ﬁequently been used as a model to study the
mechanical properties of articular cartilage (Wei et al., 1998; Roemhildt et al., 2006).
The model has also been used to better understand the development of joint disease, such
as osteoarthritis (OA), resulting from traumatic injury (Ewers et al., 2002), knee joint
instability (Sah et al., 1997; Vignon et al., 1987; Mansour et al., 1998), and
meniscectomy (Hoch et al., 1983).

To deterrrrine the changes that occur in the mechanical properties of articular
cartilage in the above studies, various computational models of cartilage have been
utilized to analyze the experimental data. In situ indentation tests have, for the most
part, been the experiment of choice to extract the material characteristics of cartilage,
largely because the method preserves the structural integrity of the tissue on the joint
surface (Sah et al. 1997 ; Mow et al. 1980). Various types of computational models may
vary, however, in their ability to extract comparable material properties Of cartilage from
experimental data. Consequently, correlations between mechanical properties and the
histological or morphological characteristics of cartilage tissue from different
laboratories can be confusing or even contradictory.

The linear elastic model of articular cartilage was one of the ﬁrst to be used in
order to reduce experimental indentation test data (Hayes et al., 1972). Theoretically,
this model is appropriate under small loads associated with small strain assumptions, but
various studies have utilized the model for 20% strain (Haut et al., 1995). The model
analysis has also been modiﬁed to account for non-linear geometric effects (Zhang et al.,
1997). Typically this model is used in relaxation and creep tests to extract the

compressive moduli of cartilage either instantly by assuming Poisson’s ratio to be 0.5, or

17

at equilibrium where the ratio is assumed to be 0.4 (Parsons and Black, 1977). More
recently, a 2-punch test has been developed in an attempt to extract the effective
Poisson’s ratio from this elastic model (J in and Lewis, 2004). The model has also been
used to estimate the relaxation and creep characteristics of cartilage by using an
associated linear viscoelastic solution (Jurvelin et al., 1990). Using the elastic model,
Hoch et al. (1983), for example, document that the equilibrium compressive modulus of
articular cartilage on the rabbit tibial plateau in central areas not covered by meniscus is
slightly higher on the medial than lateral facet. Similarly, using the same model,
Rasanen and Messner (1996) document a creep modulus after 15 sec. in the central
partially covered area of the rabbit plateau to be slightly higher on the medial than lateral
facet. In the same study a so called “ramp Young’s modulus” is also documented at the
200 nrillisecond time point to be higher on the medial than lateral facet. These
investigators hypothesize that the in vivo stresses on the medial facet are likely higher
than on the lateral facet, possibly due to higher loading pressures (Kernpson, 1979;
Swarm and Seedhom, 1986). This hypothesis, however, is in contrast to more recent
experimental data in which the investigators suggest that contact loads on the lateral
facet may often be greater than on the medial facet during hopping gait (Gushue et al.,
2005). Wei et al. (1998) also use the linear elastic model to determine the instantaneous
(at 200 ms) and 50-second creep modulus of cartilage centrally located on the rabbit
tibial plateau. A Poisson’s ratio of 0.3 is used for the calculation of 50 sec. creep
modulus, based on the literature, in this study. The investigators document a higher
instantaneous modulus on the lateral than medial facet, which is inconsistent with

previous ﬁndings by Rasanen and Messner (1996). On the other hand, the 50-sec

l8

modulus is slightly higher medial than lateral from the study, which is in support of
Rasanen and Messner’s ﬁndings. However, this notion is in contrast with the ideas
presented by Gushue et al., (2005).

The linear biphasic model of cartilage is able to determine the linearly elastic
equilibrium modulus and Poisson’s ratio of the matrix, as well as the tissue permeability
directly ﬁom experimental creep or relaxation data (Mow et al.1980). On the other hand,
the model does not include a parameter that directly relates to the instantaneous response
of the tissue. A recent study by Roemhildt et al. (2006) utilizes the linear biphasic model
to extract the material properties of rabbit articular cartilage across the tibial plateau
from creep indentation data. The investigators document that the aggregate modulus,
extracted from equilibrium data, is slightly lower medially than laterally in posterior
areas of the plateau covered by meniscus. This is in contrast with the ﬁndings that
suggest a higher creep modulus in areas partially covered by the meniscus documented
using the linear elastic model (Rasanen and Messner, 1996; Wei et al., 1998). No
signiﬁcant differences, however, were noted medial to lateral in the uncovered areas of
the plateau. The study also determines that Poisson’s ratio is smaller on the medial than
lateral facets, and the permeability of cartilage on the medial facet is signiﬁcantly greater
than on the lateral facet. The result is consistent with the observation that “the medial
compartments had slightly rougher surfaces and/or the presence of superﬁcial or deep
splits.” These morphological results, on the other hand, suggest that the collagen
network in cartilage from the medial compartment of the rabbit tibial plateau may be
more degraded than in the lateral compartment (Pelletier et al., 1983; Bank et al., 2000).

Yet a more degraded collagen network in the medial than lateral compartment suggests

l9

 

that Poisson’s ratio in the medial compartment may be greater than in the lateral
compartment, per a recent study by Kiviranta et al., (2006), using a ﬁbril-reinforced
biphasic model for cartilage. Recent studies in the literature suggest that this more
complex model of cartilage, which incorporates a collagen ﬁber network within the
matrix, may more accurately ﬁt the creep and relaxation responses of cartilage under
both unconﬁned and conﬁned compression, as well as indentation testing (Wilson et al.,
2005).

The objective of the current study was to document the mechanical properties of
articular cartilage in meniscal covered and uncovered areas of the rabbit tibial plateau
using the ﬁbril-reinforced poroelastic model for the tissue. Both the histological and
morphological characteristics were also documented across the tibial plateau in an
attempt to show that the model could provide material properties across the plateau that
were consistent with histological and morphological data, as well as to compare data
from earlier studies that have utilized less complex models of articular cartilage for

extraction of the material properties from indentation test data.

Methods

A total of 10 Flemish Giant rabbits (4.5 +/- 0.7 kg, 6-8 months of age) were used
in the study. This study was approved by an All-University Committee on Animal Use
and Care. The rabbits were exercised 10 min. per day on a treadmill at a speed of 0.3
mph, 5 days a week for this 12 month study (Oyen-Tiesma et al., 1998). A single

licensed veterinary technician (J .A.) conducted all exercise sessions.

20

After 12 months, the animals were sacriﬁced with Pentobarbital (85.9 mg/kg
body weight). The hind limbs were opened and examined for gross abnormalities, such
as advanced joint disease or inﬂammation in the joint. The meniscus was also examined,
and it was removed after marking its location on the tibial plateau. The surfaces were
stained with India ink and the length of surface ﬁssures was measured from a digital
photograph (Polaroid DMC2, Polaroid Corp, Waltharn, MA) taken under a dissecting
microscope (Wild TYP 374590, Heerbrugg, Switzerland), by a single observer (E.M.)
using image analysis software (SigrnaScan, SPSS Inc., Chicago, IL). The mechanical
properties of articular cartilage were documented at four sites across the medial and
lateral facets of the tibial plateau using indentation relaxation tests (Haut et al., 1995).
The sites were near the edge of the meniscus on the medial (Site 1) and lateral (Site 3)

facets in uncovered areas (Figure 2.1). Sites 2 and 4 were slightly posterior and under the

medial and lateral meniscus, respectively.

 

Figure 2.1. Indentation relaxation test sites located on the rabbit tibial plateau. Sites 1
and 3 correspond to the areas not covered by the meniscus on the medial and lateral
facets, respectively. Sites 2 (medial) and 4 (lateral) are the areas covered by the
meniscus.

21

The tibial plateau was cut from the tibia and was mounted in a clamp which was
bathed in room temperature phosphate buffered saline for mechanical tests. At each site
the surface of the plateau was adjusted to be perpendicular to the indenter probe with a
camera mount (model 3265, Bogen Manfrotto, Italy). After a preload of 0.03N was
applied, a 1 mm diameter, ﬂat non-porous probe was then pressed into the cartilage 0.1
mm in 30 ms and maintained for 150 s using a custom-built stepper motor device
(Physic Instruments, Waldbronn, Germany, Model M-168.3) (Figure 2.2). The resistive
loads were measured (Data Instruments, Acton MA, Model J P-25), ampliﬁed, and
sampled at 1000 Hz for the ﬁrst second and 20 Hz thereafter. After 5 minutes of rest, a
needle probe was pressed into the cartilage to measure thickness (Athanasiou et al.,

1991).

Indenter
Camera mount 4

X-Y mounting
plate

 

Figure 2.2. Photograph of the indentation test ﬁxture. The X-Y mounting plate allows
for left/right or forward/backward placement, and the Z plate allows for up.down
placement. The camera mount allowed for rotation of the sample to set the surface
perpendicular to the indenter.

22

The mechanical tests were simulated with a ﬁbril-reinforced biphasic cartilage
model (Li et al., 1999) and implemented in a ﬁnite element package (Abaqus v.6.3,

Hibbitt, Karlsson & Sorensen, Inc., Pawtucket, RI, USA). A compressible neo-Hookean

material with Young’s modulus Em and an assumed Poisson’s ratio (vm=0.3 ) were

assigned to the non-ﬁbrillar part of the matrix.
The ﬁber network was modeled as non-linear springs with Young’s modulus

given by

Ef=w(z,h)Egcf, 81.20, OSzSh, (1)

where E8 was an elastic modulus parameter for the ﬁbers, sf was the strain in the ﬁber

direction, w(z) was a non-dimensional weight function, and h was the tissue thickness.

To account for the mainly horizontal collagen ﬁber distribution in the superﬁcial zone, a
weight function was deﬁned (Li et al., 2000):
25 z 2 40 z z

[—j + Z G [0,0.8]

H? h Eh’
w(.z,h) =i (2)

 

1, is (O.8,0.1]

Permeability of the tissue (It) was assumed to be strain dependent (van der Voet, 1997):

M
1+e

k=ko 1+e0 , (3)

 

23

Where kg was the initial permeability, M was a stiffening constant, and e and co were

the current and initial void ratios, respectively. Initial water content was assumed to be
linearly decreasing from 85% superﬁcially to 70% at the cartilage-bone interface
(Lipshitz et al., 1975).

For simulation, the axisyrnmetric ﬁnite element mesh consisted of pore pressure
elements (model type CAX4P) for the matrix, and nonlinear spring elements (model type
SPRIN GA) for the ﬁbers. The specimen was modeled as a circular disc with radius of 4
mm. The bottom nodes where ﬁxed to simulate attachment to underlying bone. The
indenter was modeled as rigid, frictionless and non-permeable. To specify unrestricted
ﬂuid ﬂow out of the tissue, zero pore pressure was prescribed at the tOp nodes outside
the indenter and at the outer boundary. The mesh consisted of 64 elements in the radial
direction and 18-26 elements in the vertical direction, depending on the specimen
thickness. In preliminary studies by this lab (not published), mesh sensitivity analyses
were conducted on the ﬁnite-element model in order to optimize its rate of convergence
for typical model parameters.

To optimize agreement between experimental data and the numerical
simulations, a custom-written Gauss-Newton constrained nonlinear least-squares

minimization procedure was used in an iterative fashion (Lindstrom and Wedin, 1993).

Sensitivity analysis was performed for the four unknown material parameters (Em, E8,

kg, and M) to ensure that the constants were uniquely identiﬁable from a single

indentation experiment. The dimensionless sensitivity coefﬁcients were deﬁned as

24

 

.0 a
If ('61): 1261 6,2: (4)

where i is the i-th parameter, ﬂio is its nominal value, f (ﬂl, ,62,...,8,,) is a state

variable (reaction force), and max is its maximum value (Beck et al., 1977). It was

assumed that the experimental data were sensitive to a particular parameter

when I I f (,3,- )| > 0-1 . To ensure uniqueness of each parameter, the sensitivity plots

were checked for linear independence over time.

After mechanical tests, the specimen was placed in 10% buffered formalin and
decalciﬁed in 20% formic acid. Coronal-oriented tissue blocks were cut ﬁom the
plateau. The blocks were processed in parafﬁn and six sequential sections, 8 microns
thick, were stained with Saﬁ'anin O-Fast Green and examined under light microscopy.
The thicknesses Of articular cartilage, the zone of calciﬁed cartilage and subchondral
bone were determined by averaging across each facet with a calibrated eyepiece by 3
readers (EH, EM, J W). The readers also scored various histological parameters from the
articular cartilage and zone of calciﬁed cartilage using an established scoring system

(Figure 2.3) (Weaver and Haut, 2005; Columbo et al., 1983; Mazieres et al., 1987).

25

-—‘——n

 

 

Articular Articular Cartilage
Cartilage Fissures None 0 Disruptions None 0
1-3 Compression
Surface 1 Ridges 2
1-2 Horizontal
Midzone 2 Splits 4
3-4 Vertical
Midzone 3 Splits 4
4+
Midzone 4
1+ Deep
Zone 4
Proteoglycan Calcified Cartilage
Stain Normal 0 Stain Normal 0
Slight
Loss 1 Slight 1
Moderate
Loss 2 Moderate 2
Focally 3 Dark 3
Total
Loss 4

 

 

 

 

 

 

Figure 2.3. Histomorphometric scoring system used to quantify the characteristics for
cartilage across the tibial plateau.

The mechanical testing and thickness data from the right and left limbs of each
animal were compared using t-tests. The limb-averaged data was subjected to a two
factor (medial/lateral location; covered/uncovered) repeated (both factors) measures
AN OVA with post hoc Student-Newman-Keuls (S-N-K) tests. Non-parametric statistical
tests were used for the ordinal histological data analyses. Friedman, repeated measures
AN OVA on ranks was used for the analysis of differences between readers. Kruskal-
Wallis single factor ANOVA on Ranks with S-N-K post hoc testing was used to test for
differences between the right and left limbs. Wilcoxon Signed Rank tests and S-N-K
post hoc tests were used to test for differences between medial and lateral facets.
Spearman correlations were conducted on the mechanical and histomorphological data.

A signiﬁcant statistical effect was indicated for p < 0.05.

26

Results

Gross examination of the joints indicated no signs of joint disease. India ink
staining of the plateau indicated surface irregularities and ﬁssures on both facets (Figure
2.4). There was no difference in the length of the surface ﬁssures on right versus left
limbs. The length of the ﬁssures on the medial facet (72.6 :1: 21.6 mm) was signiﬁcantly

greater than on the lateral facet (54.8 :i: 10.0 mm).

Lateral

 

Figure 2.4. Gross photographs of the tibial surface, stained with India ink to highlight
ﬁssures. Surface irregularities were analyzed, comparing the medial and lateral facets.
Indentation relaxation testing on the tibial plateau indicated a high frequency
load response immediately after the probe was stopped. This led to a difﬁculty in
determining the time at “peak load”. Therefore, this time point was detennined by a
linear interpolation of the load data in the vicinity of this peak. The latter was used in the
ﬁnite element simulation of the indentation test in the optimization process. In all cases

the theoretical curves, with Optimized parameters, closely ﬁt the experimental data

(Figure 2.5).

27

 

 

 

 

 

 

 

 

 

 

 

 

 

Reaction Force (N)

 

 

 

 

0.0001 0.001 0.01

0.1 10 100
Time (see)

Figure 2.5. Theoretical curves were calculated to closely ﬁt the data collected during
experimental testing.

A sensitivity plot for typical values of the material parameters (Em, E8, kc, and
M) indicated that the response curves were highly sensitive to the matrix and ﬁber
modulus, as well as for k0 and M (Figure 2.6a). All parameters were also linearly

independent from one another, as indicated by the very distinct shape of each sensitivity
curve. This plot demonstrated that the peak load response was very sensitive to the value

of ﬁber modulus near the time of peak load, which is the time when the instantaneous

28

modulus would be determined using the linear elastic model. The experimental curves
also showed high sensitivity to the value of the matrix modulus throughout the entire
relaxation curve. Thus, the model parameter was determined to coincide closely with
values of the equilibrium modulus ﬁ'om the linear elastic theory. The response during the
transition from the peak load to equilibrium was sensitive to the value of permeability
and the stiffening parameter. Yet, during the curve ﬁtting process the values of the
stiffening parameter and permeability at site 4 were hard to determine using the same

nrunber of iterations as the other three test sites. Detailed sensitivity analyses were

therefore performed for the initial permeability value ko and stiffening parameter M

(Figure 2.6b). In sites 1-3, both k0 and M were linearly independent. However, analysis
at site 4 showed that the two parameters tended to be linearly dependent. Therefore, in
order to uniquely determine ko and M, the curve-ﬁtting procedure required ﬁve

additional iterations at site 4 than the required 5 for sites 1-3.

29

 

 

 

dogwoo— 3829 .3 Rubens
castes one? 2 use 3:53:59 02:. 3 63233 303 £33839 Bassoon. 05 com .39an bmﬁzmﬁcm Am .3 .3 RN ennui
a a
83 we: .83 as.»
9.: S F to 5o 82 Q: E a 3 5o .86
_—:—_-— — —=:——— — _=—-——— _ —-=—- n n mNAvi

A2 305
v cam

Emu“ if! A .1;
 \ . // W

.,/, \ ......

«I; ..\. -
, . gas
,, 7. 3 8%

w 36
again—2 .. A... I

 

‘

 

nuxmsueg

 

 

   

 

 

 

 

 

 

 

 

 

30

The analysis of indentation relaxation data from the left and right limbs at each
site indicated no signiﬁcant differences in any mechanical parameter or cartilage
thickness. Therefore, these data were averaged for this study (Figure 2.7). The
indentation tests did indicate, however, some differences in mechanical parameters

between the medial and lateral facets, as well as between covered and uncovered areas of
each facet. While in covered areas E1n was approximately 51% higher in the medial than
lateral compartments (p=0.05), no differences were noted between facets in the
uncovered areas (p=0.77). On the other hand, no signiﬁcant difference in Em was
determined overall between the medial and lateral facets on the mature rabbit’s tibial
plateau (p=0.33). Overall across the plateau, EIn was approximately 48% higher in the
uncovered areas than covered areas (p=0.005).

The ﬁber modulus, Ef, was signiﬁcantly different medial to lateral, as well as in
covered versus uncovered areas of the plateau. This modulus was, on average, nearly 4
times less in the medial than lateral compartments (p<0.001). While no difference in Ef
existed between the uncovered lateral and medial compartments (p=0.65), there was
nearly 7-fold decrease in Ef in the medial versus the lateral facets for meniscal covered

areas (p<0.001).
Differences in permeability were also noted between the medial and lateral

compartments, as well as between meniscal covered and uncovered areas. The initial

permeability kg and factor M were each increased by approximately 90% in the medial

31

versus the lateral compartments (p=0.005). Both parameters were also higher by 95%
(p=0.003) and 71% (p<0.001), respectively, in uncovered than covered areas.

The histological scores ﬁom the left and right limbs, as well as from the 3
readers were not signiﬁcantly different, so the data were averaged for each specimen. On
the other hand, some signiﬁcant differences were evident between the medial and lateral
facets (Figure 2.8).

More intense staining of the calciﬁed cartilage was noted for medial versus
lateral facets (Figure 2.9a,b). More disruptions of cartilage were also noted in the medial
than the lateral comparlrnents (Figure 2.9c,d). A trend (p = 0.088) was also indicated’for
there to be more histolog’cal-based ﬁssures in sections ﬁ'om the medial than lateral
compartments of the plateau (Figure 2.9e,f). On the other hand, no signiﬁcant
differences were recorded in the intensity of PG staining in the cartilage from the medial

versus lateral facet.

32

.mOmoﬁeoam
5 See me quﬁ 98 508 05 we eon/mm 0.8 338d .ﬂooﬂ 383 28 338 2: Bob 3:58 208 ﬁcEBSmE .w.~ 25w”...—

 

 

mz 6.08 No 85.8 mud macaw on.
Bod 5.3.8 No.0 :3 m: 858m
-88
need 5.3.8 3 8.4.08 SN 82523
:56 A8828 :3 8.3. 8 he saw 00
83> n. .923 .382

.038 m3..— 5 53m 08 $8058.89 .5 Bungee c8855 98 Gov—08

8.on BE. .38 commune—2 838585 05 89a 3:68 because Roam—#008 98 $0865 Owﬂgo 3.58— 23 382 5N 9.sz
cameo. 8 SEE...” sore 5.8 3.7m»: 8o+m88 Shame tonne. 5 Shane .2
838.8 imam.” 85: v c $.35 87%». : Smut... 858.8 3&3.» azsev 9.

8.38 was 8.88 38 388 9:. $.88 98 8%: .m
88.8 m: 68.8 as... :88 :4! 68.8 Ema ans: em
3885.: Enema moment 3&8." Comte: :38.» Emma: 8.8.3 EEC ;

3.98:: 3.900 8.98:: 3550 Eugene

 

.923 .885.

 

33

Medial

\Iedizll

 

e f

Figure 2.9 (a-f). Histological images of the articular cartilage and underlying bone show
signiﬁcant differences between the medial and lateral facets. The zone of calciﬁed
cartilage showed more intense staining for tissue proteoglycans medial (a) versus the
lateral (b) facet, using a Safranin O-Fast Green protocol. The cartilage on the medial
facet (c) showed more disruptions than the lateral facet (d). The tibial plateau showed
more surface disruptions, including ﬁssuring, on the medial (e) versus the lateral facet

(0-

Overall, there were no signiﬁcant differences between the thicknesses of
calciﬁed cartilage in medial (0.058 d: 0.014 mm) versus lateral (0.061 :b 0.012 mm)
compartments, as well as between the medial (0.78 i 0.07 mm) and lateral (0.80 at 0.11

mm) subchondral plates. In contrast, there was a difference in the histological-based

34

measures of articular cartilage thickness between the medial (1.36 :l: 0.19 mm) and
lateral (0.93 :l: 0.12 mm) compartments.
Statistically signiﬁcant correlations were also documented between some of the

histomorphological and mechanical parameters. For example, the length of surface

ﬁssures had a weak but signiﬁcant negative correlation with ﬁber modulus (R2 = -0.223,

p = 0.035), and a weak but signiﬁcant positive correlation with tissue permeability (R2 =
0.24, p = 0.028). The ﬁber modulus also had a weak but signiﬁcant negative correlation
with articular cartilage thickness (R2 = -0.264, p = 0.02). Finally, there was no indication
of a statistical correlation between the intensity of cartilage PG stain and matrix modulus

(R2 = 0.00032, p = 0.9) or tissue permeability (R2 = 0.043, p = 0.4) in the current study.

Discussion

The ﬁbril reinforced biphasic model has previously been used to extract the
mechanical properties of cartilage from experimental response data (Wilson et al., 2004).
The model accurately ﬁt the data from both the indentation and unconﬁned compression
tests utilized in this study. In the current study the ﬁbril reinforced biphasic model was
used to closely ﬁt the entire response curve from a single indentation relaxation
experiment. The properties extracted from this model showed variation across the tibial
plateau, as well as differences speciﬁcally in regions covered and uncovered by the
meniscus. Surface ﬁssuring and histologically identiﬁed matrix damage in articular

cartilage documented in the medial compartment paralleled with reductions in ﬁber

35

modulus, and increases in permeability and tissue thickness in the medial versus lateral
compartments of the plateau.

Numerous results from the current study could be compared with previous data
by others, using either linear elastic or linear biphasic models of cartilage. These studies
determined mechanical parameters from either the instantaneous or equilibrium
response, which paralleled with the ﬁber or matrix modulus, respectively, in the current
study using the ﬁbril-reinforced biphasic model. Hoch et al. (1983), using the linear
elastic cartilage model, as well as Roemhildt et al. (2006) using the linear biphasic
model, both document no signiﬁcant differences in the equilibrium modulus between
medial and lateral facets of the rabbit tibial plateau for the central, uncovered regions.
The current study also showed no signiﬁcant differences in matrix modulus between the
medial and lateral facets for the uncovered regions. However, the modulus was found to
be greater on the medial versus lateral facet for regions covered by meniscus. These data
compare to those of Rasanen and Messner (1996) using creep indentation tests in regions
partially covered by the meniscus which document slightly stiffer cartilage in the medial
than lateral facets. The current study also documented an over 40% higher matrix
modulus in meniscal uncovered than covered regions of the plateau. Previous studies
suggest that articular cartilage is stiffer, based on the equilibrium modulus, with a higher
content of matrix proteoglycans in areas subjected to high versus low levels of stress
(Kempson, 1979; Swarm and Seedhom, 1986). The results of the current study on
rabbits may then suggest larger joint pressures in uncovered than covered regions of the
plateau, and especially in the medial compartment as indicated by the larger Em in the

medial versus the lateral compartment. Studies on human specimens also show greater

36

 

pressures in the uncovered regions of the medial compartment, while high pressures are
carried on both the uncovered and covered areas in the lateral compartment
(Fukubayashi and Kurosawa, 1980; Walker and Erkrnan, 1975).

The equilibrium modulus correlates with the content of cartilage matrix PGs
(Armstrong and Mow, 1982). More recently studies with the ﬁbril reinforced biphasic
model correlate PG stainability of cartilage with matrix modulus (Julkunen et al., 2007)
However, in the current study, no statistical correlation was established between matrix
modulus and PG stain intensity in the cartilage. A previous study suggests that the
correlation between PG stain intensity and tissue stiffness may be related to the state of
cartilage health (Camplejohn and Allard, 1988). In normal cartilage a direct correlation
could be established between these tissue properties, but not in osteoarthritic cartilage.
This may be due to the reduction of anions in the matrix and therefore, a reduction of
possible binding sites for the safranin-O stain (Camplejohn and Allard, 1988). The lack
of a correlation in the current study may then suggest the tissue may be in a diseased
state of health.

The instantaneous response of the cartilage was also ﬁt with the ﬁbril-reinforced
biphasic model in this study. The sensitivity studies suggested that the peak load during
these experiments was most sensitive to ﬁber modulus, a mechanical parameter that
likely reﬂects the mechanical characteristics of the network of collagen (Korhonen et al.,
2003; Julkunen et al., 2007). In a previous study on central areas of the rabbit tibial
plateau a slightly lower instantaneous modulus using the linear elastic model was
documented in the medial than lateral facets (Wei et al., 1998). The result compares

well with the ﬁber modulus data of the current study. Additionally, the current study

37

established a statistically signiﬁcant negative correlation between ﬁber modulus and the
length of surface ﬁssures. Surface ﬁssures and tissue swelling are early characteristics of
osteoarthritis that have been related to the early degradation of the collagen network
(Bank et al., 2000; Pelletier et al., 1983; Verzijl et al., 2002). The correlation
documented between increases in tissue permeability with ﬁssure length helps explain
differences documented in the medial versus lateral compartments and meniscal
uncovered versus covered areas of the plateau, and the results compare with early studies
on the development of osteoarthritis using the knee ligament transection model (Setton
et al., 1994).

We also documented an increase in the PG stain intensity in the zone of calciﬁed
cartilage in the medial versus lateral facet. This histological feature may suggest that the
zone of calciﬁed cartilage in the medial facet was in more of an active state of
remodeling than the lateral facet (Oegema and Thompson, 1992). This histological
feature also paralleled with more matrix disruptions and surface ﬁssuring of cartilage on
the medial than lateral facets in this animal model. It is currently unclear whether these
changes in tissue quality result from more or less load being carried by the medial versus
lateral facets. In human (Kemp et al., 2008) and animal studies (Rasanen and
Messner,]996) the excessive levels of tissue degradation noted in medial versus lateral
facets are suggested to be the result of more loading on the medial facet. On the other
hand, studies suggest the degenerative state of this tissue in medial versus lateral facets
may be due to a lower level of load in the medial compartment (Chang et al., 1997). This
suggestion may well support the observation made recently in rabbit gait studies that

show a slightly larger joint load being carried by the lateral compartment during a hop

38

(Gushue et al., 2005). On the other hand, the current study indicates a larger matrix
modulus in the medial versus lateral compartments, suggestive of possibly greater levels
of loading medial than lateral. Thus, additional studies are needed to clarify these
apparent contradictory results.

In terms of tissue permeability and ﬁbril modulus in covered areas of the plateau,
the cm'rent study showed larger medial to lateral differences than most literature. This
could be due to relatively more intense loading on the medial than lateral facets, or vice
versa, in the exercised model used in the current study. None of the previous studies
have used a regularly exercised model, nor have results been gathered on the larger,
Flemish Giant rabbit. Previous studies on other animal models have shown that while
moderate intensities of regular exercise have a positive effect on joint cartilage (J urvelin
et al., 1990; Newton et al., 1997), more intense exercise or unloading can have a
negative effect on the joint cartilage (Helminen et al., 1992). While our laboratory has
not performed gait analysis on our model or quantiﬁed the intensity of the exercise
protocol, these rabbits did subjectively appear exhausted after each daily session of
treadmill activity, suggesting a rather intense level of exercise for the animal.

Future studies will be needed to better deﬁne potential relationships between
alterations in the joint loading and changes in tissue properties across the tibial plateau in
the human and various animal models. The current study helps support the notion that a
more complex ﬁbril-reinforced, biphasic model of the cartilage in such studies may lead
to more consistent mechanical results that correlate with histomorphological changes

noted in articular cartilage.

39

Acknowledgements

This study was supported by the Centers for Disease Control and Prevention, the
National Center for Injury Prevention and Control (CE000623). Its contents are the
responsibility of the authors and do not necessarily represent the official views of the
Centers for Disease Control and Prevention. The authors thank Jane Walsh (J W) and
Derek Handzo (DH) for their contributions on the histology aspects of the study, as well

as and Jean Atkinson (J A) for exercising and care of the animals and Eric Meyer for

P.-‘h.a+._.n’ I

mechanical testing and histological scoring.

40

References -

Armstrong, CG. and Mow, V.C. (1982) Variations in the intrinsic mechanical properties
of human articular cartilage with age, degeneration, and water content. Journal of Bone
and Joint Surgery Am. 64, 88-94.

Athanasio, K.A., Rosenwasser, M.P., Buckwalter, J .A., Malinin, TL, and Mow, V.C.
(1991) Interspecies comparison of in situ intrinsic mechanical properties of distal
femoral cartilage. Journal of Orthopaedic Research 9,.330—340.

Bank, R.A., Soudry, M., Maroudas, A., Mizrahi, J ., and TeKoppele, J .M. (2000) The
increased swelling and instantaneous deformation of osteoarthritic cartilage is highly
correlated with collagen degradation. Arthritis and Rheumatism 43, 2202-2210.

Beck, J. and Arnold, K. (1977). Parameter Estimation in Engineering and Science, New
York: John Wiley and sons.

Camplejohn, KL and Allard, SA. (1988) Limitations of safranin ’O’ staining in
proteoglycan-depleted cartilage demonstrated with monoclonal antibodies.
Histochemistry 89, 185-188.

Chang, D.G., Iverson, E.P., Schinagl, R.M., Sonoda, M., Amiel, D.,Coutts, R.D., and
Sah, KL, (1997) Quantitation and localization of cartilage degeneration following
induction of osteoarthritis in the rabbit knee. Osteoarthritis and Cartilage 5, 357-372.

Colombo, C., Butler, M., O’Byrne, E., Hickman, L., Swartzendruber, D., Selwyn, M.,
and Steinetz, B. (1983) A new model of osteoarthritis in rabbits. I. Development of knee
joint pathology following lateral meniscectomy and section of ﬁbular collateral and
sesarnoid ligaments. Arthritis Rheumatology 26, 87 5-886.

Ewers, B.J., Weaver, B.T., Sevensma, B.T., and Haut, R.C. (2002) Chronic changes in
rabbit retro-patellar cartilage and subchondral bone after blunt impact loading of the
patellofemoral joint. Journal of Orthopaedic Research 20, 545-550.

Fukubayashi, T. and Kurosawa, H. (1980) ‘The contact area and pressure distribution
pattern of the knee. Acta Orthopaedica Scandinavica 51, 871-879.

Gushue, D.L., Houck, J ., and Lerner, AL. (2005) Rabbit knee joint biomechanics:
motion analysis and modeling of forces during hopping. Journal of Orthopaedic
Research 23, 735-742.

Haut, R.C., Ide, T.M., and DeCamp, CE. (1995) Mechanical responses of the rabbit
petello-femoral joint to blunt impact. Journal of Biomechanical Engineering 117, 402-
408.

Hayes, W.C., Keer, I.M., Herrmann, G., and Mockros, LE. (1972) A mathematical
analysis for indentation tests of articular cartilage. Journal of Biomechanics 5, .541-551.

41

run“! 1

Helminen, H., Kiviranta, 1., Saamanen, A., Jurvelin, J ., Arokoski,J., Oettmeier, R. (1992)
Effect of motion and load on articular cartilage in animal models. Articular Cartilage
and Osteoarthritis 35, 501—510.

Hoch, D., Grodzinsky, A., Koob, T., Albert, M., and Eyre, D. (1983) Early changes in
material properties of rabbit articular cartilage after meniscectomy. Journal of
Orthopaedic Research 1, 4-12.

Jin, H. and Lewis, J .L. (2004) Determination of Poisson’s ratio of articular cartilage by
indentation using different-sized indenters. Journal of Biomechanical Engineering 126,
l 3 8-145.

Julkunen, P., Kiviranta, P., Wilson, W., Jurvelin, J .S., and Korhonen, R.K. (2007)
Characterization of articular cartilage by combining microscopic analysis with a ﬁbril-
reinforced ﬁnite-element model. Journal of Biomechanics 40, 1862-1870.

Jurvelin, J ., Kiviranta, I., Saamanen, A.M., Tammi, M., and Helminen, HI. (1990)
Indentation stiffness of young canine knee articular cartilage-inﬂuence of strenuous joint
loading. Journal of Biomechanics 23, 1239-1246.

Kemp, G., Crossley, K.M., Wrigley, T.V., and Metcalf, BR. (2008) Reducing joint
loading in medial knee osteoarthritis: shoes and canes. Arthritis and Rheumatism 59,
609-614.

Kernpson, GE. (1979) Mechanical properties of articular cartilage. In: Freeman, M.A.R.
(ed.), Adult Articular, 333—414 London: Pitrnan Medical.

Kiviranta, P., Rieppo, R.K., Julunen, P., Toyras, J ., and Jurvelin, J .S. (2006) Collagen
network primarily controls Poisson’s ratio of bovine articular cartilage in compression.
Journal of Orthopaedic Research 24, 690-699.

Korhonen, R.K., Laasanen, M.S., Toyras, J ., Lappalainen, R., Helminen, H.J., and
Jurvelin, J .S. (2003) Fibril reinforced poroelastic model predicts speciﬁcally mechanical
behavior of normal, proteoglycan depleted and collagen degraded articular cartilage.
Journal of Biomechanics 36, 1373-1379.

Li, L.P., Buschmann, M., and Shirazi-Adl, A. (2000) A ﬁbril reinforced
nonhomogeneous poroelastic model for articular cartilage: Inhomogeneous response in
unconﬁned compression. Journal of Biomechanics 33, 153 3-1 541 .

Li, L.P., Soulhat, J., Buschmann, M.D., and Shirazi-Adl, A. (1999) Nonlinear analysis of
cartilage in unconﬁned ramp compression using a ﬁbril reinforced poroelastic model.
Clinical Biomechanics 14, 673-682.

42

Lindstrom, P., and Wedin, RA. (1993) A method of gauss-newton type for nonlinear
least-squares problems with nonlinear constraints. Technical Report UMINF-l33.87,
Inst. Of Information Processing, University of Umea, Umea, Sweden.

Lipshitz, H., Etheridge, R., and Glirncher, M.J. (1975) In vitro wear of articular
cartilage. Journal of Bone and Joint Surgery 57, 527-537.

Mansour, J .M., Wentorf, P.A., and Degoede, KM. (1998) In vivo kinematics of the
rabbit knee in unstable models of osteoarthritis. Annals of Biomedical Engineering 26,
353-360.

Mazieres, B., Blanckaert, A., and Thiechart, M. (1987) Experimental post-contusive
osteoarthritis of the knee: quantitative microscope study of the patella and the femoral
condyles’, Journal of Rheumatology 14, 119-121.

Mow, V.C., Kuei, S.C., Lai, W.M., and Armstrong, CG. (1980) Biphasic creep and
stress relaxation of articular cartilage in compression: Theory and experiments. Journal
of Biomechanical Engineering 102, 73-84.

Newton, P.M., Mow, W.C., Gardner, T.R., Buckwalter, J .A., and Albright, J .P. (1997)
The effect of lifelong exercise on canine articular cartilage. American Journal of Sports
Medicine 25, 282—287.

Oegema, T.R., Carpenter, R.J., Hofrneister, F., and Thompson, R.C. Jr. (1997) The
interaction of the zone of calciﬁed cartilage and subchondral bone in osteoarthritis.
Microscopy Research and Technique 37, 324-332.

Oyen-Tiesma, M., Atkinson, J ., and Haut, R.C. (1998) A method for promoting regular
exercise in rabbits involved in orthopedics research. Contemporary Topics in Laboratory
Animal Science 37, 77-80.

Parsons, J .R. and Black, J. (1977) The viscoelastic shear behavior of normal rabbit
articular cartilage. Journal of Biomechanics 10, 21-29.

Pelletier, J .P., Martel-Palletier, J ., Altman, R.D., Ghandur/Mnaymneh, L., Hower, D.S.,
and Woessner, J.P., (1983) Collagenolytic activityand collagen matrix breakdown of the
articular cartilage in de Pond-Nuki dog model of osteoarthritis. Arthritis and
Rheumatism 26, 866-874.

Rasanen, T. and Messner, K. (1996) Regional variations of indentation stiffness and
thickness of normal rabbit knee articular cartilage. Journal of Biomedical Materials
Research 31, 519-524.

Roemhildt, M.L., Coughlin, K.M., Peura, G.D., Fleming, BC, and Beynnon, RD.

(2006) Material properties of articular cartilage in the rabbit tibial plateau. Journal of
Biomechanics 39, 2331-2337.

43

Sah, R.L., Yang, A.S., Chen, A.C., Hant, J .J ., Halili, R.B., Yoshioka, M., Amiel, D., and
Coutts, RD. (1997) Physical properties of rabbit articular cartilage after transition of the
anterior cruiate ligament. Journal of Orthopedic Research 15, 197-203.

Setton, L.A., Mow, V.C., Muller, P.J., Pita, J .C., and Howell, D.S. (1994) Mechanical
properties of canine articular cartilage are signiﬁcantly altered following transection of
the anterior cruciate ligament. Journal of Orthopaedic Research 12, 451 -463.

Swann, AC. and Seedhom, BB. (1986) The relationship between the stiffness of normal
articular cartilage and predominant acting stress levels. Publ. Univ. Kuopio Med. Orig.
Rep 6, C3.

van der Voet, A. (1997) A comparison of ﬁnite element codes for the solution of
biphasic poroelastic problems. Proceedings of the Institute of Mechanical Engineering
211, 209-211.

Verzijl, N., DeGroot, J ., Zaken, C.B., Bram-Benjamin, O., Maroudas, A., Bank, R.A.,
Mizrahi, J ., Schalkwaijk, C.G., Thorpe, S.R., Baynes, J .W., Bijlsma, J .W.J ., Lafeber,
F.P.J.G., and TeKoppele, J .M. (2002) Crosslinking by advanced glycation end products
increases the stiffness of the collagen network in human articular cartilage. Arthritis and
Rheumatism 46, 114-123.

Vignon, E., Bejui, J., Mathieu, P., Hartrnann, J.D., Ville, G., Evreux, J.C., and Descotes,
J. (1987) Histological cartilage changes in a rabbit model of osteoartluitis. Journal of
Rheumatology 14, 104-106.

Walker, PS. and Erkrnan, M.J. (1975) The role of the menisci in force transmission
across the knee. Clinical Orthopaedics 106, 184-192.

Weaver, B.T. and Haut, R.C. (2005) Enforced Exercise aﬁer blunt trauma signiﬁcantly
affects biomechanical and histological changes in rabbit retro-patellar cartilage. Journal
of Biomechanics 38, 1 177-1 183.

Wei, X., Rasanen, T., and Messner, K. (1998) Maturation-related compressive properties
of rabbit knee articular cartilage and volume fraction of subchondral tissue.
Osteoarthritis and Cartilage 6, 400-409.

Wilson, W., van Donkelaar, C.C., van Rietbergen, B., and Huiskes, R. (2005) A ﬁbril-
reinforced poroviscoelastic swelling model for articular cartilage. Journal of
Biomechanics 38, 1195-1204.

Wilson, W., van Donkelaar, C.C., van Rietbergen, B., Ito, K., and Huiskes, R. (2004)
Stress in the local collagen network of articular cartilage: a poroviscoelastic ﬁbril-
reinforced ﬁnite element study. Journal of Biomechanics 37, 357-366.

Zhang M., Zheng, Y.P., and Mak, AF. (1997) Estimating the effective young’s modulus
of soft tissues from indentation tests-nonlinear ﬁnite element analysis of effects of
ﬁiction and large deformation. Medical Engineering and Physics 19, 512-517.

45

Chapter 3
High levels of glucosamine-chondroitin sulfate can alter articular cartilage stiffness
and up—regulate proteoglycan content of bovine chondral explants following
unconﬁned compression injury
Abstract
Traumatic injury to articular cartilage results in chondrocyte loss, collagen

damage and decreased PG content and has been found to be load level and strain
dependent. Therapeutic agents, such as glcN-CS, have been shown to increase the PG
content of cartilage pre-trauma and; therefore, increase the compressive stifﬁress. A
hypothesis of the current study was that signiﬁcantly more damage will occur following a
higher level of trauma documented by lower mechanical properties and a decrease in PG
content of the tissue. A second hypothesis of the study was that treatment with glcN-CS
in the post-trauma period will up regulate the PG content of the injured explants and
therefore, help maintain its pre-trauma mechanical stiffness. Bovine chondral explants
were loaded with 10 MPa or 25 MPa of unconﬁned compression with and without glcN-
CS treatment. Mechanical properties were extracted at 7, 14, and 21 days and following
21 days PG content was analyzed. A 25 MPa unconﬁned compression signiﬁcantly
decreased the mechanical properties of the cartilage compared to tissues that were loaded
with 10 MPa of unconﬁned compression. Treatment with glcN-CS signiﬁcantly increased
the PG content and matrix modulus compared to untreated samples following both 10 and
25 MPa of unconﬁned compression. The results of this study conﬁrm the ability of glcN-
CS to increase the PG content in the tissue following traumatic injury, increasing the
matrix stiffness. Future studies should focus on the effects of glcN-CS treatment in the

long term as it may provide better outcomes for the ACL repaired patient.

46

Introduction

Traumatic injury to articular cartilage during participation in sports, recreation
and exercise (SRE) ﬁequently leads to the development of a chronic joint disease,
osteoarthritis (OA). Two speciﬁc types of injuries are associated with subsequent knee
OA: cruciate ligament damage and meniscal tears (F elson et al., 2004). Evidence in the
clinical literature suggests that 50—70% of patients with complete anterior cruciate
ligament (ACL) rupture and associated injuries develop radiological signs of OA within
15-20 years (Gillquist and Messner, 1999). And, surgical reconstruction of the torn ACL
is not effective in mitigating the incidence of joint CA, as a signiﬁcant portion of these
patients develop clinical symptoms of OA 5-10 years post-injury (Daniel et al., 1994).
Arthroscopic surgeries have documented softening and ﬁssuring of cartilage overlying so
call “geographic” bone bruises in the ACL injured patient, and biopsy specimens from
these patients reveal degeneration and death of chondrocytes in this area (Fang et al.,
2001; Johnson et al., 1998). This damage to articular cartilage is thought to be caused by
excessive compressive forces generated in the joint during the acute injury and has been
hypothesized to form the basis for the subsequent development of OA (Fang et al., 2001).

In vitro studies have documented signiﬁcant changes in the articular cartilage
following traumatic loading, such as an increase in tissue wet weight (Loening et al.,
2000, Huser and Davies, 2006), damage to the collagen network (Torzilli et al., 1999)
and decreased chondrocyte viability (Huser and Davies, 2006; Loening et al., 2000; Kurz
et al., 2001) and proteoglycan (PG) content (Huser and Davies, 2006; Patwari et al.,
2000). The level of damage is dependent on compressive loading levels and strain rates

(Torzilli et al., 1999). Speciﬁcally, traumatic loading to the cartilage explants increases

47

the percentage of dead cells with increasing loading levels (6-20 MPa), with signiﬁcantly
more apoptotic cells docmnented above 10 MPa (Loening et al., 2000). Additionally, a
study by Kurz et al. (2001) documents an increase in cell death with increasing strain rate
(0.01-ls") of compression and the remaining viable cells have a signiﬁcant decrease in
biosynthetic activity at 0.1 s'1 and 1 s". Chondrocyte damage leaves the cartilage with
few metabolically active cells to repair the degraded matrix (Loening et al., 2000). One
speciﬁc ﬁrnction of chondrocytes is the synthesis of proteoglycans (PGs) and loss of
chondrocytes has been correlated with a loss of tissue PGs (Huser and Davies, 2006;
Simon et al., 1976). Previous studies have documented the loss of PG in cartilage
explants following compressive loading (Huser and Davies, 2006; Patwari et al., 2000)
and this loss of PG has been correlated with a decrease in compressive stiffness (Patwari
et al. 2000). These degenerative changes may be due to an increase in degenerative
enzymes (Lin et al., 2004). Cartilage softening has been documented in patients
following traumatic injury and has been associated with the development of OA. Since
proteoglycans are responsible for the resistance to stress and strain and loss of PG is
correlated with cartilage softening (Patwari et al., 2000), prevention of PG loss may help
in retaining the cartilage stiffness and resistance to compressive loading after injury.
Therapeutic agents, such as glucosamine (glcN) and chondroitin sulfate (CS), can
increase the PG content of cartilage and therefore increase matrix properties like
compressive stiffness. Clinical trials have shown that daily oral supplements of glcN for 3
years can slow the progression of joint disease, by mitigating joint space narrowing
(Richy et al. 2003), although no evaluation of the mechanical or biological properties

were conducted in this particular study. Tiraloche et al. (2005), in another study,

48

 

document an increase in histological PG stain in ACL transected rabbits treated for 8
weeks with daily supplements of oral glcN compared to untreated animals. Additionally,
bathing chondral explants in a supplement of glcN and CS up-regulates the synthesis of
tissue PG’s (Lippiello, 2003). Treatment with glcN-CS has been documented to decrease
the effects of degradative enzymes (Dodge and Jimenez, 2003). The objective of the
current study was to document the mechanical properties and PG content of chondral
explants following unconﬁned compression, with and without glcN-CS treatment post-
trauma. Two levels of unconﬁned compression (10 MPa and 25 MPa) were used to
evaluate varying levels of trauma as previous studies have documented degeneration
effects to be load dependent. A hypothesis of the current study was that signiﬁcantly
more damage will occur following a higher level of trauma documented by lower
mechanical properties and a decrease in PG content of the tissue. A second hypothesis of
the study was that treatment with glcN-CS in the post-trauma period will up regulate the
PG content of the injured explants and therefore, help maintain its pre-trauma mechanical
stiffness. This may help maintain post-trauma tissue homeostasis and help preserve the

functionality of traumatized joint cartilage.

Method
Dissection and tissue culture

Skeletally mature bovine forelegs (18-24 months of age) were obtained from a
local abattoir within 2 hours of slaughter. The legs were rinsed with water and skinned
prior to exposing the metacarpal joint. A 6 mm diameter biopsy punch (Miltex Instrument

Company, Bethpage, NY) was used to make chondral explants ﬁom the lower metacarpal

49

sm'face of each limb under a laminar ﬂow hood. Explants were separated ﬁom the
underlying bone with a scalpel. All specimens were washed three times in Dulbecco’s
Modiﬁed Eagle Media: F12 (DMEM: F12) (Gibco, USA, #12500-039) supplemented
with additional amino acids and antibiotics (penicillin 100 U/ml, streptomycin 1 ug/rnl,
amphotericin B 0.25 rig/ml). The osrnolarity of the media was 300mosM (Osmete 2,
Precision Systems), and the pH was 7.4 as this has previously been shown to result in
physiological and metabolic stability of the explants (Phillips and Haut 2004; Baars et al.
2006; Wei et al. 2008; Wei and Haut 2009). The explants were randomly assigned to two
impact groups: 10 MPa and 25 MPa. Each of these groups was further divided into two
sub groups, with and without glcN-CS: ‘10 MPa with glcN-CS’ (n=8), ’10 MPa without
glcN-CS’ (n=12), ’25 MPa with glcN-CS’ (n=8), and ’25 MPa without glcN-CS’ (n=12).
The explants were then placed in a 24-well plate in media supplemented with 10% fetal
bovine serum (Gibco, USA, #16000). The treated samples were bathed in supplemented
media with glucosamine and chondroitin sulfate supplement (glcN-CS) (500 ug/mL glcN
(F CHG49®) and 250 ug/mL CS (TRH122®)). GlcN-CS concentrations were chosen
based on previous studies (Rundell et al., 2005; Wei et al., 2008) to maximize treatment

effects. The media was replaced every 2 days during the study.

Indentation testing of explants

All samples were allowed to equilibrate for 24 hours after harvesting inside of a
humidity-controlled incubator (3 7°C, 5% C02, 95% humidity). Prior to impact, each
explant was subjected to mechanical testing using an indentation stress relaxation test to

extract the mechanical properties of the explants prior to compressive loading. These data

50

served as baseline mechanical property data and were used to normalize the data
following compressive loading. Indentation tests on each explant were also performed 7,
14, and 21 days following impact. Prior to mechanical testing, the cartilage explant
thickness was measured twice at perpendicular orientations across the center of the
explant using a digital vernier caliper (Mitutoyo Corp.: Absolute Digimatic, Model No.

CD-6" CS) with a resolution of 0.01 mm (Steinrneyer et al., 1997 and 1999). The two

it‘l

thickness values were then averaged. The explants were placed on a ﬂat level surface so
that the face of the explant was perpendicular to the indenter tip (Figure 3.1B). A magnet l
with a 4.3 mm diameter hole was placed on top of the explant to secure the edges from
curling (Figure 3.1C). The explant and ﬁxture were then submerged into a room-
ternperature phosphate buffered saline solution (PBS with pH 7.2) (Figure 3.1A). A 2.39
mm diameter spherical, non-porous probe was lowered into the cartilage until a preload
of 0.05 N was attained and held for 60 s (Figure 3.1D). The indenter was then pressed
into the cartilage 25% of the thickness in 2 s and maintained for 600 3 while resistive
loads of relaxation were recorded (Data Instruments, Acton, MA: model J P-25, 25 lb
capacity), ampliﬁed and sampled at 1,000 Hz for the ﬁrst second, and 20 Hz thereafter.
The stress relaxation curves were ﬁtted to a ﬁbril-reinforced biphasic model (Soulhat et

al., 1999; Golenberg et al., 2009) with an assmned Poisson’s ratio of 0.25. The matrix

modulus (Em), ﬁber modulus (Bf) and tissue permeability (kg) were evaluated in the

computational model using a custom-written Gauss-Newton constrained nonlinear least

square minimization procedure.

51

‘*/. \

Al‘lagnet Explant

A

Magnets“

 

Figure 3.1. Explant indentation test system and ﬁxture (A). The explants were placed in
a hole in the bottom magnet on a ﬂat steel surface (B). A top magnet was lowered over
the top of the explant to hold down the edges (C). The spherical indentor tip was lowered
to a preload of 0.05 N (D).
Unconﬁned compression

Following a 5 N preload the explants were taken to either 282 N (~10 MPa) or
707 N (~25 MPa) in unconﬁned compression between two polished stainless steel plates
(Figure 3.2). A 0.5 Hz (1 s time to peak) haversine loading protocol was programmed in a
servo-controlled hydraulic testing machine (Instron, model 1331, retroﬁtted with 8500

plus electronics, Canton, MA). Immediately after compressive loading, all explants were

placed in the incubator for the duration of the study.

52

  
  

Stainless Steel
Plates

 
 

—

 

Figure 3.2. Cartilage explants were loaded (10 MPa or 25 MPa) in unconﬁned
compression between two polished stainless steel plates.
Determination of Proteoglycan (PG) Content

Aﬁer 21 days, the sample wet weights were recorded. Approximately, 4.5 mg
Chondroitin Sulfate A sodium salt ﬁ'orn bovine trachea (Sigma-ALDRICH GmbH
Steinheim, Germany) was measured out and used to generate standard curves
(Steinmeyer et al. 1999). The samples and the standards were digested overnight at 60 °C
in a papain solution: PBS, EDTA, cysteine and papain. Papain digested cartilage explants
and the standards were dimethyl-methylene blue (DMB) assayed for sulfated PGs by the
reaction with l, 9-DMB dye solution in polystyrene 96 well plates and quantitated with
spectrophotometry at wavelength 530 nm using a Bio Tek microplatc reader. PG content

was normalized to cartilage wet weights.

Statistical Analysis
Mechanical property data obtained ﬁ'om post-impact explants were normalized
using the pre-impact values to document change in each property. Statistical analysis was

used to evaluate differences in mechanical and biochemical properties. A two-factor (day,

53

load group) AN OVA with a posthoc Student-Newman-Keuls (SNK) test was used to
determine differences in mechanical properties between the treatment groups. A two-
factor (day, glcN-CS treatment) AN OVA with a posthoc SNK test was used to determine
differences in mechanical properties with the treatment of glcN-CS. A one factor (load
group) AN OVA was used to document differences in PG content between the two levels
of unconﬁned compression. A one factor (glcN-CS treatment) AN OVA was used to
analyze PG content of the unconﬁned compression groups. Statistical signiﬁcance was

indicated at p<0.05.

Results
Unconﬁned Compression

Both the 10 MPa and 25 MPa of unconﬁned compression resulted in surface
ﬁssures seen following staining of the articular surface with India ink. Gross assessment

revealed more surface lesions in the samples loaded to 25 MPa~ These samples also

appeared elliptical in shape.

 

» i A . A
igure 3.3. Gross analysis of the cartilage explants revealed surface lesions on both the
10 MPa (A) and 25 MPa (B) samples with more ﬁssures on the 25 MPa samples. These
samples also appeared elliptical in shape.

 

54

Indentation Testing

The Em decreased approximately 20% and 40% at 10 MPa and 25 MPa of
unconﬁned compression, respectively. No decrease in Em was noted over 21 days. A
signiﬁcant decrease in Em was found in the 25 MPa samples compared to the 10 MPa
samples at 7 (p<0.001), 14 (p=0.004) and 21 (p=0.004) days following unconﬁned

compression (Figure 3.4).

 

12° ‘ I 10 MPa

 

 

 

100‘

00
O
1

Matrix Modulus
(% Difference)
O\

o

 

 

4o —
20 —
0 _.
7 14 21
Time (day)

Figure 3.4. The matrix modulus was determined ﬁ'om the stress-relaxation curves
documented during each indentation test. The matrix modulus presented here is given as
a percentage of the original property prior to unconﬁned compression. A decrease,
compared to the initial property, in the matrix modulus was documented following both
10 MPa and 25 MPa of unconﬁned compression. Signiﬁcantly lower matrix modulus was
documented following 25 MPa of unconﬁned compression versus 10 MPa of unconﬁned
compression. ‘*’ represents statistical signiﬁcance compared to 10 MPa samples.

The permeability increased by approximately 140% and 250% following 10 MPa
and 25 MPa of unconﬁned compression, respectively. No changes in permeability were

documented between 7, l4, and 21 days. A statistical increase in the permeability was

55

noted in the 25 MPa versus 10 MPa samples at 7 (p=0.007), l4 (p=0.005), and 21

(p<0.001) days (Figure 3.5).

450 -
400 ~

' 350 —
300 J
250 —
200 —
150 —
100 —
so -
0 —4

IIOMPa ,

Permeability
(% Difference)

 

 

7 14 21
Time (day)

Figure 3.5. The permeability was determined from the stress-relaxation curves
documented during each indentation test. The permeability presented here is given as a
percentage of the original property prior to compression. A signiﬁcant increase in
permeability was documented following 25 MPa unconﬁned compression compared to
10 MPa unconﬁned compression. ‘*’ represents statistical signiﬁcance compared to 10
MPa samples.

A 150% increase in the Efwas documented in the 25 MPa samples. No such
increase was noted in samples compressed to 10 MPa. This increase in E; was
signiﬁcantly greater at 25 MPa compared to the 10 MPa at 7 (p=0.042) and 21 (p=0.016)
days with a statistical trend for an increase at 14 days (p=0.096) (Figure 3.6). No change

in Ef was documented between 7, 14, and 21 days.

56

250 ‘ I10 MPa *
I25 MPa *

   

 

200 ,
n A
2 8
§ § 150 —
E .a;'-:’
g n 1ool
a 8
50 —
0 _
7 14 21
Time (day)

Figure 3.6. The ﬁber modulus was determined from the stress-relaxation curves
documented during each indentation test. The ﬁber presented here is given as a
percentage of the original property prior to compression. An increase in the ﬁber modulus
was documented following 25 MPa unconﬁned compression compared to original
property and this increase was signiﬁcantly higher than samples following 10 MPa of
unconﬁned compression ‘*’ represents statistical signiﬁcance compared to the 10 MPa

samples.

Biochemical assays revealed a decrease in matrix PG content in the 25 MPa

samples compared to the 10 MPa, however, this decrease did not rise to a level of

statistical signiﬁcance (Figure 3.7).

57

8888’
411_J_l

154
104

 

Proteoglycan Content
(p9 P6! mg wet welght)

   

o O!
r r

 

10 MPa 25 MPa
Impact Level

Figure 3.7. Proteoglycan content in samples was determined using DMB assay. Results
are shown here as ug PG per mg wet weight. No statistical difference was documented in
the PG content between 10 MPa and 25 MPa of unconﬁned compression.

Treatment with glcN-CS
An increase in 13m was documented in the ‘10 MPa with glcN-CS’ versus ‘10MPa

without glcN-CS’ at day 7 (p=0.033) and 14 (p=0.048) days, with statistical trend for an
increase at day 21 (p=0.06) (Figure 3.8). Similarly, an increase in the matrix modulus was
documented between the samples treated with glcN-CS versus the untreated samples at 7,
14 (p=0.1), and 21 days following 25 MPa of unconﬁned compression, however, this
difference did not rise to a level of statistical signiﬁcance (Figure 3.9). No signiﬁcant
differences were documented between treated and untreated samples in the ﬁber modulus

or permeability in either unconﬁned compression group.

58

r—Iu—I
#0‘
00

no glcN-CS
I glcN-CS

Matrix Modulus
(% Difference)
o 8 3 8 8 8 '5’

 

7 14 21
Time (day)

Figure 3.8. The matrix modulus of the 10 MPa samples with and without glcN-CS was
determined from stress-relaxation curves. A signiﬁcant increase in the matrix modulus
was documented following treatment with glcN-CS compared to untreated samples. ‘*’
denotes statistical signiﬁcance compared to samples treated with glcN-CS.

 

I no glcN-CS
140-00 j I glcN-CS

120.00 a -.

 

 

100.00 i

 

80.00 i

60.00 4

Matrix Modulus
(% difference)

40.00 l

20.00 -

 

 

0.00 -

7 1 4 21
Time (days)

Figure 3.9. The matrix modulus of the 25 MPa samples with and without glcN-CS was
determined ﬁ'om stress-relaxation curves. An increase in the matrix modulus was
documented following treatment with glcN-CS compared to untreated samples; however,
statistical signiﬁcance was not reached.

A signiﬁcant increase in PG content was docmnented in the treated samples
compared to the untreated samples for both 10 MPa (p<0.001) and 25 MPa (p=0.023) of

unconﬁned compression (Figure 3.10).

701

a s

5 E

g ; INo glcN-CS
a E” I glcN-CS

e

.3 8’.

$2 °:i.

 

 

Impact Level (MPa)

Figure 3.10. Proteoglycan content in all samples was determined using DMB assay.
Results are shown here as ug PG per mg wet weight. A signiﬁcant increase in PG content
was documented in samples treated with glcN-CS compared to untreated samples. ‘*’
denotes statistical signiﬁcance compared to glcN—CS treated samples.
Discussion

The current study documented a decrease in the matrix modulus and PG content
with an increase in the tissue permeability and ﬁber modulus with an increasing level of
unconﬁned compression. These results correlate with previous studies that documented
decreased cartilage stiffness and increased ﬂuid parameters following traumatic injury
with a loss of PG in the tissue (Kurz etal., 2001; Loening et al., 2000). Interestingly, an

increase in the ﬁber modulus was documented following a 25 MPa of unconﬁned

compression, with no change in modulus following 10 MPa of unconﬁned compression.

60

Torzilli et al. (1999) documents that loading at or above a critical threshold of 15-20 MPa
causes permanent damage to the collagen network and an increase in the tissue water
content. In the current study, a change in the cartilage shape and an increase in surface
disruptions were documented following 25 MPa of unconﬁned compression. Previous
studies have documented that following an impact load, the cartilage becomes ﬂattened
ﬁssured and elliptical in shape and that the alignment of these distortions reﬂect the
orientation of the collagen ﬁbers (Jeffrey et al., 1995). In the current study, a change in
shape and increased ﬁssuring were found following 25 MPa of unconﬁned compression
and, therefore, reﬂect damage to the collagen network. Additionally, previous studies
have documented a correlation between damage to the collagen network and tissue
swelling (Bank et al., 2000; Khalsa and Eisenberg, 1997), as the collagen network is
known to resist the swelling of cartilage. Therefore I suggest that, following traumatic
injury to the collagen network, swelling pressures increase and therefore, the remairning
intact collagen ﬁbers bear more tension. The increase in ﬁber modulus from the current
study may reﬂect this increase in tension in the intact collagen ﬁbers and the increase in
swelling pressure following the 25 MPa of unconﬁned compression. However, cartilage
swelling and content of collagen were not measured in the current study.

The current study also showed an increase in tissue PG content with glcN-CS

treatment that was reﬂected by an increase in tissue Em. These results compare with

previous ﬁndings by our laboratory documenting an increase in PG content and matrix
stiffness following treatment of bovine chondral explants with glcN-CS (Wei and Haut,

2008). Interestingly, Tiraloche et al. (2005) also document an increase in the PG content

61

in the rabbit articular cartilage following ACL transection with an oral treatment of glcN-
CS.

Previous studies suggest a possible explanation for the decrease in mechanical
properties is the increase in degenerative enzymes. Injurious compression has been
shown to increase MMP-3 resulting in the degadation effects in the matrix parameters
(Lin et al., 2004). A previous study suggests that treatrrnent with glcN-CS may inhibit the
effects of MMP-3s (Dodge and Jimenez, 2003). The current study did not docurnente
MMP-3 concentrations. Future studies should focus on the effects of these inhibiting
these enzymes following traumatic injury with glcN-CS treatment post-trauma.

The current study suggests that post-trauma treatment with glcN-CS effectively
increased matrix stiffness and PG content of the injured tissue. Pre-trauma cyclic loading
increases the PG content and stiffness of the cartilage explants (Wei et al., 2008), and
increases glcN-CS incorporation in the tissue (Wei and Haut, 2009). However, the effects
of post-traumatic cyclic loading have yet to be elicited. The following chapter will focus
on the effects of post-traumatic cyclic loading as a potential means of increasing PG
synthesis and helping to maintain homeostasis of the damaged cartilage by enhancing the

effect of glcN-CS.

62

References

Baars, D.C., Rundell, S.A., Haut, R.C. (2006) Treatment with the non-ionic surfactant
poloxamer P188 reduces DNA ﬁ'agnentation in cells from bovine chondral explants

exposed to injurious unconﬁned compression. Biomechan Model Mechanobiol 5, 133-
139.

Bank, R.A., Soudry, M., Maroudas, A., Mizrahi, J ., TeKoppele, J .M. (2000) The
increased swelling and instantaneous deformation of osteoarthritic cartilage is highly
correlated with collagen degadation. Arthritis and Rheumatism 43, 2202-2210.

Daniel, D.M., Stone, M.L., Dobson, B.E., Fithian, D.C., Rossman, D.I., Kaufman, KR.
(1994) Fate of the ACL injured patient: a prospective outcome study. The American
Journal of Sports Medicine 22, 632-644.

Dodge, GR. and J irnenez, SA. (2003) Glucosamine sulfate modulates the levels of
aggecan and matrix meta110proteinase-3 synthesized by cultured human osteoarthritis
articular chondrocytes. Osteoarthritis and Cartilage 11, 424-432.

Fang, C., Johnson, D., Leslie, M.P., Carlson, C.S., Robbins, M., Di Cesare, PE. (2001)
Tissue distribution and measurement of cartilage oligomeric matrix protein in patients
with magnetic resonance imaging-detected bone bruises after acute anterior cruiate
ligament tears. Journal of Orthopaedic Research 19, 634-641.

Felson, D.T. (2004) An update on the pathogenesis and epiderrniology of osteoarthritis.
Radio] Clin North Am 42, 1-9.

Gillquist, J., Messner, K., (1999) Anterior cruciate ligament reconstruction and the long
term incidence of gonarthrosis. Sports Medicine 27, 143-156.

Golenberg, N., Kepich, E., Haut, R.C., (2009) Histomorphological and mechanical
property correlations in rabbit tibial plateau cartilage based on a ﬁbril-reinforced biphasic
model. International Journal of Experimental and Computational Biomechanics 1, 58-75.

Huser, C.A.M., Davies, ME. (2006) Validation of an in vitro single-impact load model
of the initiation of osteoarthritis-like changes in articular cartilage. Journal of
Orthopaedic Research 24, 725-732.

Jeffrey, J .E., Gregory, D.W., Aspden, RM. (1995) Matrix damage and chondrocyte
viability following a single impact load on articular cartilage. Archives of Biochemistry
and Biophysics 322, 87-96.

Johnson, D.L., Urban, W.P., Caborn, D.N.M., Vanarthos, W.J., Carlson, CS. (1998)
Articular cartilage changes seen with magnetic resonance imagine-detercted bone bruises

associated with acture anterior cruiate ligament rupture. The American Journal of Sports
Medicine 26, 409-414.

63

Khalsa, P.S., Eisenberg, SR. (1997) Compressive behavior of articular cartilage is not

completely explained by proteoglycan osmotic pressure. Journal of Biomechanics 30,
589-594.

Kurz, B., Jin, M., Patwari, P., Cheng, D.M., Lark, M.W., Grodzinsky, AJ. (2001)
Biosynthetic response and mechanical properties of articular cartilage after injurious
compression. Journal of Orthopaedic Research 19, 1140-1146.

Lin, P.M., Chen, C.-T.C., Torzilli, RA. (2004) Increased stromelysin-l (MMP-3),
proteoglycan degadation (333- and 7D4) and collagen damage in cyclically load-injured
articular cartilage. Osteoarthritis and Cartilage 12, 485—496.

Lippiello, L. (2003) Glucosamine and chondroitin sulfate: biological response modiﬁers
of chondrocytes under simulated conditions of j oint stress. OsteoArthritis and Cartilage
11, 335-342.

Loening, A.M., James, I.E., Levenston, M.E., Badger, A.M., Frank, E.H., Kurz, B.,
Nuttall, M.E., Hung, H.-H., Blake, S.M., Grodzinsky, A.J., Lark, M.W. (2000) Injurious
mechanical compression of bovine articular cartilage induces chondrocyte apoptosis.
Archives of Biochemistry and Biophysics 381, 205-212.

Patwari, P., Kurz, B., Sandy, J.D., Grodzinsky, A.J., (2000) Mannosarnine inhibits
aggrecanase-mediated changes in physical properties and biochemical composition of
articular cartilage. Archives of Biochemistry and Biophysics 374, 79-85.

Phillips, D.M. and Haut, R.C. (2004) The use of non-ionic surfactant (P188) to save
chondrocytes from necrosis following impact loading of chondral explants. Journal of
Orthopaedic Research 22, 1135-1 142.

Richy, F., Bruyere, O., Ethgen, O., Cucherat, M., Henrotin, Y., Reginster, J .-Y. (2003)
Structural and symptomatic efficacy of glucosamine and chondroitin in knee
osteoarthritis. Archives of Internal Medicine 163, 1514-1522.

Rundell, SA. (2005) Chapter Three: Glucosamine supplementation can help limit matrix
damage and adjacent cell death in traumatized explants. In: Investigation into the acute
injury response of articular cartilage in vitro and in vivo: analysis of various therapeutic
treatments. Theis for the degee of MS. Michigan State University, 70-95.

Simon, W.H., Richardson, S., Herman, W., Parsons, J.R., Lane, J. (1976) Long-term
effects of chondrocyte death on rabbit articular cartilage in vivo. Journal of Bone and
Joint Surgery, Am 58, 517-526.

Soulhat, J., Buschmann, M.D., Shirazi-Adl, A. (1999) A ﬁbril-network-reinforced

biphasic model of cartilage in unconﬁned compression. Journal of Biomechanical
Engineering 121, 340-347.

64

Steinmeyer, J ., Ackermann, B., Raiss, R.X. (1997) Intermittent cyclic loading of cartilage
explants modulates ﬁbronectin metabolism. Osteoarthritis and Cartilage 5, 331-341.

Steinmeyer, J ., Knue, S., Raiss, R.X., Pelzer, I. (1999) Effects of intermittently applied
cyclic loading on proteoglycans metabolism and swelling behavior of articular cartilage
explants. Osteoarthritis and Cartilage 7, 155-164.

Tiraloche, G., Girard, C., Chouinard, L., Sampalis, J ., Moquirn, L., Ionescu, M., Reiner,
A., Poole, A.R., Laverty, S. (2005) Effect of oral glucosamine on cartilage degadation in
a rabbit model of osteoarthritis. Arthritis and Rheumatism 52, 1 118-1 128.

Torzilli, P.A., Grigiene, R., Borrelli, J. Jr., Helfet, D.L. (1999) Effect of impact load on
articular cartilage: cell metabolism and viability, and matrix water content. Journal of
Biomechanical Engineering 121, 433-441.

Wei, P., Golenberg, N., Kepich, B.T., Haut, R.C. (2008) Effect of intermittent cyclic
preloads on the resoponse of articular cartilage explants to an excessive level of
unconﬁned compression. Journal of Orthopaedic Research 26, 1636-1642.

Wei, F. and Haut, R.C. (2009) High levels of glucosarnine—chondroitin sulfate can alter

the cyclic preload and acute overload responses of chondral explants. Journal of
Orthopaedic Research 27, 353-359.

65

Chapter 4

Investigation of low level cyclic loading following high levels of unconfined

compression with and without glucosamine chondroitin sulfate treatment
Abstract

A deﬁning feature of OA is softening of the articular cartilage and decreased

tissue proteoglycan (PG) content. Low level cyclic loading has been shown to increase
matrix stiffness and cell biosynthesis, speciﬁcally PG synthesis, and increase
incorporation of glcN-CS. The hypothesis of the current study was that the matrix
stiffness and tissue PG content would increase with cyclic loading post-trauma and that
the glcN-CS effect previously documented would be further enhanced with low level
cyclic loading post-trauma. Bovine chondral explants were subjected to 10 or 25 MPa of
unconﬁned compression followed by 0.5 MPa of cyclic loading. Treatment with glcN-CS
was also investigated. An increase in PG content was documented with glcN-CS
treatment and cyclic loading following both 10 MPa and 25 MPa of unconﬁned
compression compared to cyclic loading alone. No difference in the matrix modulus was
documented with glcN-CS treatment following either level of unconﬁned compression.
Previous studies have documented that the elastic properties of cartilage are inﬂuenced
by the aggegate structure of the proteoglycans, not only their tissue content.
Proteoglycans ﬁ'om degenerated tissue cartilage were found to be smaller and had lost
their ability to bind to hyaluronic acid and form aggegates. Therefore, future studies are
still needed to study the type and shape of the proteoglycans produced by injured cells
following unconﬁned compression with the additional stress associated with cyclic

loading.

66

 

Introduction

Injmies during participation in sports, recreation, and exercise (SRE) have been
associated with long-term development of osteoarthritis (OA) (Lane, 1996). A deﬁning
feature of OA is softening of the articular cartilage and decreased tissue proteoglycan
(PG) content. Matrix PGs are responsible for cartilage resistance to stress and strain and
the loss of mechanical stiffness in OA tissue is suggested to be caused by decreased PG
content (Patwari et al., 2000). The prevention of PG loss following traumatic injury has
also been suggested as a possible intervention aimed at the prevention of post-traumatic
OA. While treatments of CA are currently limited, mechanical stimulation and
nutraceutical treatments, such as glucosamine-chondroitin sulfate (glcN-CS), have been
investigated as possible means of repairing articular cartilage and increasing matrix PG
content following traumatic injury.

Low to moderate levels of exercise increase PG synthesis and increase the matrix
stiffness of articular cartilage. Clinically, following regular exercise, patients previously
diagnosed with high risk of developing knee OA saw an increase in PG content (R005
and Dahlberg, 2005). Similarly, in vivo animal models have been used to document
increases in matrix P63 and tissue stifﬁness with exercise. A previous study using a
canine model documents an increase in the tissue PG content, speciﬁcally in regions of
high loading following regular exercise (Kiviranta et al., 2005). A study by Jurvelin et al.
(1986) also documents that regular exercise increases compressive stiffness in carnine
articular cartilage. A post-tramnatic study using a Flemish Giant rabbit model documents

that exercise prevents the loss of matrix PGs (Weaver and Haut, 2005).

67

In vitro studies have also been used to investigate the effects of low level (0.5-5
MPa) mechanical loading on articular cartilage. Low level cyclic loading increases cell
biosynthesis, speciﬁcally PG synthesis (Torzilli et al., 1999; Sah et a1. 1989; Millward-
Sadler and Salter, 2004). Wei et al. (2008), using bovine chondral explants, document
that 0.5 MPa of cyclic loading over 14 days results in an increase in tissue matrix
modulus and decreases the tissue permeability. This increase in mechanical stiffrness is
correlated with an increase in matrix PGs and results in a decrease in the severity and
extent of surface ﬁssures and cell death in the tissue. Moreover, low level cyclic loading
increases the incorporation of glcN-CS throughout the tissue and results in a further
increase in PG content and mechanical stiffness of cartilage (Wei and Haut, 2009,
Sharma et al., 2008).

The previous chapter documented the ability of glcN-CS to increase matrix PG
content and matrix stifﬁness following 10 MPa and 25 MPa of unconﬁned compression
and pre-trauma cyclic loading ﬁnrther increases glcN-CS incorporation. However, post-
traumatic cyclic loading of cartilage explants has not been documented. Therefore, the
purpose of the current study was to determine whether low level cyclic loading would
increase the mechanical stiffness of cartilage explants post-trauma and investigate if
cyclic loading will enhance the incorporation and effects of glcN-CS documented in the
previous chapter. The hypothesis of the current study was that the matrix stiffness and
tissue PG content would increase with exercise and that the glcN-CS effect previously

documented would be further enhanced with low level cyclic loading post trauma.

68

Method
Dissection and tissue culture

Skeletally mature bovine forelegs were obtained from a local abattoir within 2
hours of slaughter. The legs were rinsed with water and skinned prior to exposing the
metacarpal joint under a laminar ﬂow hood. A 6 mm diameter biopsy punch (Miltex
Instrument Company, Bethpage, NY) was used to make chondral explants ﬁom the lower
metacarpal surface of the limbs. Each explant was separated from the underlying bone
with a scalpel. The explants were randomly assigned to two impact load goups (10 MPa
and 25 MPa). Each of these goups was furtlner divided into two sub goups, with and
witlnout glcN-CS treatment: ‘10 MPa with glcN-CS’ (n=8), ‘10 MPa without glcN-CS’
(n=12), ‘25 MPa with glcN-CS’ (n=8), and ‘25 MPa without glcN-CS’ (n=12). All
specimens were washed three times in Dulbecco’s Modiﬁed Eagle Media: F 12 (DMEM:
F12) (Gibco, USA, #12500-039) supplemented with additional amino acids and
antibiotics (penicillin 100 U/ml, streptomycin 1 rig/ml, amphotericin B 0.25 ug/ml). The
explants were then incubated in media supplemented with 10% fetal bovine serum in a
24—well plate. Glucosamine and chondroitin sulfate treated samples were placed in a 24-
well plate with media supplemented with glucosamine (500 ug/mL (F CHG49®)),
chondroitin sulfate (250 ug/mL CS (TRH122®)) (Rundell, 2005; Wei et al., 2008), and
10% fetal bovine serum. The concentration of glcN-CS was chosen to maximize effect of
glcN-CS treatment based on previous studies by this laboratory (Rundell, 2005; Wei et
al., 2008). The media was replaced every 2 days for the duration of the study. Following
initial indentation and unconﬁned compression at either 10 MPa or 25 MPa, the well

plates were placed in a mechanical loading device (the ‘cartilage exerciser’ (described

69

below)) inside of a humidity-controlled incubator (37°C, 5% C02, 95% humidity). The
osrnolarity of the media was 300mosM (Osmete 2, Precision Systems), and the pH was
7.4 as this results in physiological and metabolic stability of the explants (Phillips and

Haut 2004; Baars et al. 2006; Wei et al. 2008; Wei and Haut 2009).

Indentation testing of the explants

The samples were allowed to equilibrate for 24 hours after harvesting. Prior to
unconﬁned compression, each explant was subjected to an indentation stress relaxation
test. Mechanical tests were also performed 7, l4, and 21 days after impact. Before each
indentation test, the cartilage explant thickness was measured twice at perpendicular
orientations across the center of the explant using a digital verrnier caliper (Mitutoyo
Corp.: Absolute Digimatic, Model No. CD—6" CS) with a resolution of 0.01 mm
(Steinmeyer et al. 1997 and 1999). The two thickness values were then averaged. The
explants were then placed on a ﬂat level surface so that the face of the explant was
perpendicular to the indenter tip (Figure 4.1b). A magnet with a 4.3 mm diameter hole
was placed on top of the explant to secure the edges and help resist curling of the
explants (Figure 4.10). The explant and ﬁxture were then submerged into a room-
temperature phosphate buffered saline (PBS with pH 7.2) (Figure 4.1a). A 2.39 mm
diameter spherical, non-porous probe was lowered into the cartilage until a preload of
0.05 N was attained and held for 60 s (Figure 4.1d). The indenter was then pressed into
the cartilage 25% its total thickness in 2 s and maintained for 600 8 while resistive loads
of relaxation were measured (Data Instruments, Acton, MA: model J P-25, 25 lb

capacity), ampliﬁed and collected at 1,000 Hz for the ﬁrst second and 20 Hz, thereafter.

70

The stress relaxation curves were obtained and ﬁtted with a ﬁbril-reinforced biphasic

ﬁnite element model (Soulhat et al. 1999) with an assumed Poisson’s ratio of 0.25.
Cartilage matrix modulus (Em), ﬁber modulus (Ef) and tissue permeability (kg) were
evaluated with a custom-written Gauss-Newton constrained nonlinear least square

minimization procedure.

/'\

N-lagnct Explant

.2- ‘ ' ' Load
.- . c u
7 +1.; e
.,

5"
l

«*sz z

Bath Magnets“

  

;_ %

Indentor - .

 

Figure 4.1. Explant indentation test system and ﬁxture (A). The explants were placed in
a hold of the bottom magnet on a ﬂat steel surface (B). A top magnet was lowered over
the top of the explant to hold down the edges (C). The indentor tip was lowered to a
preload of 0.05 N (D).

71

Unconﬁned compression tests on the explants

Following a 5 N preload, the explants were taken to either 282 N (~10 MPa) or
707 N (~25 MPa) in unconﬁned compression between two polished stainless steel plates
(Figure 4.2). A 0.5 Hz (1 s time to peak) haversine loading protocol was progammed for
application onto the explants in a servo-controlled hydraulic testing machine (Instron,
model 1331, retroﬁtted with 8500 plus electronics, Canton, MA). Immediately after this
load protocol, the explants were placed in the ‘cartilage exerciser’ for the remainder of

the study.

   
 

Stairnless Steel
Plates

 

Figure 4.2. Cartilage explants were loaded in unconﬁned compression at either 10 MPa
or 25 MPa between two polished stainless steel plates.
Cyclic loading of the explants

All samples were cyclically loaded for the duration of the study. The ‘cartilage
exerciser’ consisted of 12 loading chambers simultaneously powered by air. Pneumatic
cylinders forced the pistons downward to apply a compressive load to the specimens
through 14.6 mm diameter non-porous Teﬂon® platens. The “cartilage exerciser” was
designed to hold a 24-well culture plate with 12 cartilage samples that could be

mecharnically loaded (Figure 4.3). Intermittent, uniaxial cyclic loads were applied using a

72

0.2 Hz sinusoidal waveform with a peak stress of 0.5 MPa. The cyclic loads were applied
for 10 cycles followed by a load-free period lasting 3600 3. During the period of

unloading the load platen was lifted from the cartilage surface.

 

pistons
cartilage Cyc. Load
explants . \
in media ‘

 

Figure 4.3. The “cartilage exerciser” mecharnical loading device applied compressive
loads to the cartilage explants in 12 separate loading chambers in a 24 well plate. The
samples were cyclically loaded 10 times with a peak stress of 0.5 MPa followed by 3600
seconds of rest. This protocol was then repeated for the duration of the test.
Determination of the tissue proteoglycan (PG) content

Following the 21 -day test period, samples were weighed and digested overnight at
60° C in a papain solution. Approximately, 4.5 mg chondroitin sulfate A sodium salt
from bovine trachea (Sigma-ALDRICH GmbH Steinheim, Germany) was digested using
the same protocol and was used as the standard during this assay. Papain digested

cartilage explants and the chondroitin sulfate standards were dimethyl-methylene blue

(DMB) assayed for sulfated PGs by the reaction with 1,9-DMB dye solution in

73

polystyrene 96 well plates and quantitated with spectrophotometry at wavelength 530 nm

using a Bio Tek micr0plate reader.

Statistical Analysis

Mechanical data was collected during indentation-relaxation testing. Post-
compression mechanical data were normalized by the pre-compression values. Statistical
analysis was used to evaluate differences in mechanical and biochemical properties. A
two-factor (day, glcN-CS treatrrnent) AN OVA with post hoc Student-Newman-Keuls
(SNK) test was used to determine differences in mecharnical properties due to glcN-CS
treatment. A one-factor (glcN-CS treatment) AN OVA with SNK post hoc test was used
to determine differences in PG content of supplemented and non-supplemented samples

at the various loading levels. Statistical signiﬁcance was indicated at p<0.05.

Results

A 20% decrease in the matrix modulus was documented following 10 MPa of
unconﬁned compression and exercise. This decrease in matrix modulus was not changed
with glcN-CS treatment (Figure 4.4a). Similarly, a decrease in the matrix modulus was
documented following 25 MP3 of unconﬁned compression and exercise. With glcN-CS
treatment, however, a statistical trend was noted for an increase in the matrix modulus at
7 (p=0.055) and 21 (p=0.1) days post trauma (Figure 4.4b). Treatment with glcN-CS
increased the PG content in the tissue following 10 (p=0.027) and 25 MPa (p=0.05) of

unconﬁned compression and post-trauma low level cyclic loading (Figure 4.5).

74

 

 

 

120 7 I_lcN-CS
a 100 -
'8 g 80 -
2 o
E
-E w: 60 -
’3 °\°
E 40 —
20 —
0 _
7 14 21
Time (day)
a)
160-
140-
120—

Matrix Modulus
% difference

 

 

 

7 14 21
Time @830

b)

Figure 4.4. The matrix modulus following 10 MPa (a) and 25 MPa (b) of unconﬁned
compression and low level cyclic loading with and without treatment with glcN-CS
supplement. No differences were documented with the treatment of glcN-CS following
10 MPa of unconﬁned compression, while an increase was documented following 25
MPa of unconﬁned compression and cyclic loading with glcN-CS treatment.

 

I no glcN-CS
I glcN-CS

 

 

 

 

 

 

10 MPa 25 MPa
Unconfined Compression Level

Figure 4.5. PG content of samples with and without glcN-CS supplement. A signiﬁcant
increase in the tissue PG content was documented with the treatment of glcN-CS. ‘*’
denotes a statistically signiﬁcant difference compared to samples treated with glcN-CS.
Discussion

Previous studies have documented the ability of glcN-CS to increase PG synthesis
in cartilage (Lippiello, 2003, Chapter 2). Treatment with cyclic loading and glcN-CS in
the current study increased the PG content following both 10 MPa and 25 MPa of
unconﬁned compression compared to cyclic loading alone. Treatment with glcN-CS and
cyclic loading after trauma also increased the matrix stiffrness of samples loaded with 25
MPa of unconﬁned compression, which supported the ﬁndings from Chapter 2 of this
thesis. Interestingly, while glcN-CS with cyclic loading after trauma was found to
increase the PG content in the tissue following 10 MPa of unconﬁned compression, there
was no tendency for an increase in the matrix stifﬁness of the tissue. Previous studies have

documented that the elastic properties of cartilage are inﬂuenced by the aggegate

76

structure of the proteoglycans, not only their tissue content (Inerot and Heinegard, 1978).
Proteoglycans ﬁorn degenerated tissue cartilage in the latter study were found to be
smaller and had lost their ability to bind to hyaluronic acid and form aggegates. While
the size and binding potential of the proteoglycans in the current study are unknown, it is
possible that the proteoglycans produced by cells following 10 MPa of unconﬁned
compression with cyclic post trauma loading may resemble those of osteoarthritic
cartilage previously shown to be smaller and lacking binding sites. This could help
explain the lack of a correlation between changes in the proteoglycans content of the
tissue exposed to 10 MPa with cyclic loading and its mechanical stiffrness. The lower
level loading may have damaged signiﬁcant number of cells causing a signiﬁcant level of
cellular dysfunction in the tissue. In contrast after 25 MPa of compression, the damaged
cells may have lost any degee of viability even to produce tlnese dysfunctional PGs.

Previous studies by Wei et al. (2008) document an increase in PG content and
matrix stifﬁness with low level cyclic loading. However, a comparison with the previous
chapter revealed no signiﬁcant changes in matrix stifﬁness with cyclic loading following
both 10 and 25 MPa of unconﬁned compression (Figure 4.6) along with no signiﬁcant
changes in PG content of the tissue with or witlnout the cyclic loading post-trauma
(Figure 4.7). A study by Kurz et al. (2001) documents that cells that are mechanically
injured may not be able to respond to dynamic mechanical stimulation either because the
cells have lost the ability to do so or because damage to the extracellular matrix has

disrupted the transduction of physical signals to the cells.

77

Iwithout CL
120 — Iwith CL
100 A

80*
60-

% difference

40—

Matrix Modulus

20-

   

 

7 14 21
Time (days)

a)

Iwithout CL

.0
80
70
60
50
40
30
20
1 0

Matrix Modulus
% difference

   

7 1 4 21
Time (days)
b)
Figure 4.6. The matrix modulus following 10 MPa (a) and 25 MPa (b) of unconﬁned

compression with and without exercise (from Chapter 2). No differences were
documented due to cyclic loading in either of the loading goups.

*‘Nssae
l l L 1

0010010
I 111

l

Proteoglycan Content
(pg PGI mg wet weight)

Without Cyclic With Cyclic Loading Without Cyclic Wiﬂn Cyclic Loading
Loading Loading

10 MPa 25 MPa
Unconﬁned compression level

 

 

Figure 4.7. The PG content following 10 MPa and 25 MPa of unconﬁned compression
with and without cyclic loading. No signiﬁcant differences were documented between the
samples with or without cyclic loading.

Multiple factors are associated with the cartilage properties. Cell viability depends
on loading levels and loading rates (Torzilli et al., 1999; Loening et al., 2000).
Additionally, the loading levels and loading rates alter the mechanical integity (Kurz et
al., 2001) of the cartilage as well as effect cellular biosynthesis (Wilkins et al., 2000).
Furthermore, these mechanical stimulations affect the level of pharmaceutical efﬁcacy
(Sah et al., 1989). Therefore, future studies should continue to focus on multiple loading
situations and their effects on changes in chondrocyte biosynthesis and cartilage integity.
The exact mecharnism that inhibited the increase in the matrix stiffrness is currently
unknown, therefore, future studies are still needed to study the type and shape of the
proteoglycans produced by injured cells following unconﬁned compression with the
additional stress associated with cyclic loading. Additional studies will also be needed to
investigate both cellular viability and cellular synthesis alterations with glcN-CS

treatments in combination with cyclic loading on mechanically injured cartilage.

79

References

Baars, D.C., Rundell, S.A., Haut, R.C. (2006) Treatment with the non-ionic surfactant
poloxamer P188 reduces DNA fragnentation in cells from bovine chondral explants
exposed to injurious unconﬁned compression. Biomechan Model Mechanobiol 5, 133-
1 39.

Inerot, S., and Heinegard, D. (1978) Articular-cartilage proteoglycans in aging and
osteoarthritis. Biochem. J. 169, 143-156.

Jurvelin, J ., Kiviranta, 1., Tammi, M., Hehninern, H.J., (1986) Effect of physical exercise
on indentation stiffrness of articular cartilage in the canine knee. Int. J. Sports Med. 7,
106-1 10.

Kiviranta, I., Tamrrni, M., Jurvelin, J ., Saamanen, A.-M., Helminen, H]. (2005) Moderate
running exercise augnents glycosarrninoglycans and thickrness of articular cartilage in the
krnee joint of young beagle dogs. Journal of Orthopaedic Research 6, 188-195.

Kurz, B., Jin, M., Patwari, P., Cheng, D.M., Lark, M.W., Grodzinsky, AJ. (2001)
Biosynthetic response and mecharnical properties of articular cartilage after injurious
compression. Journal of Orthopaedic Research 19, 1140-1146.

Lane, N. (1996) Physical activity at leisure and risk of osteoarthritis. Annals Rheumatic
Disease 55, 682-684.

Lippiello, L. (2003) Glucosarrnine and chondroitin sulfate: biological response modiﬁers
of chondrocytes under simulated conditions of j oint stress. OsteoArthritis and Cartilage
11, 335-342.

Loerning, A.M., James, 1.E., Levenston, M.E., Badger, A.M., Frank, E.H., Kurz, B.,
Nuttall, M.E., Hung, H.-H., Blake, S.M., Grodzirnsky, A.J., Lark, M.W. (2000) Injurious
mechanical compression of bovine articular cartilage induces chondrocyte apoptosis.
Archives of Biochemistry and Biophysics 381, 205-212.

Millward-Sadler, S.J., Salter, D.M. (2004) Integin-dependent signal cascades in
chondrocyte mechanotransduction. Annals of Biomedical Engineering 32, 43 5-446

Patwari, P., Kurz, B., Sandy, J .D., Grodzinsky, A.J., (2000) Mannosarnine inhibits
aggecanase-mediated changes in physical properties and biochemical composition of
articular cartilage. Archives of Biochemistry and Biophysics 374, 79-85.

Phillips, D.M. and Haut, R.C. (2004) The use of non-ionic surfactant (P188) to save

chondrocytes ﬁom necrosis following impact loading of chondral explants. Journal of
Orthopaedic Research 22, 1 135-1 142.

80

Roos, EM. and Dalnlberg, L. (2005) Positive effects of moderate exercise on
glycosarrninoglycan content in krnee cartilage: A four-month, randorrnized, controlled trial
in patients at risk of osteoartlnritis. Arthritis and Rheumatism 52, 3507-3514.

Rundell, SA. (2005) Chapter Three: Glucosamine supplementation can help limit matrix
damage and adjacent cell death in traumatized explants. In: Investigation into the acute
injury response of articular cartilage in vitro and in vivo: analysis of various therapeutic
treatments. Theis for the degee of MS. Michigan State University, 70—95.

Sah R.L.-Y., Kim, Y.-J., Doong, J .-Y.H., Grodzinsky, A.J., Plaas, A.H.K., Sandy, J .D.
(1989) Biosynthetic response of cartilage explants to dynamic compression. Journal of
Orthopaedic Research 7, 619-636.

Sharma, G., Sazena, R.K., Mishra, P. (2008) Synergistic effect of chondroitin sulfate and
cyclic pressure on biochennical and morphological properties of chondrocytes ﬁ'om
articular cartilage. Osteoarthritis and Cartilage 16, 1387-1394.

Soulhat, J ., Bsuhmann, M.D., Shirazi-Adl, A. (1999) A ﬁbril reinforced biphasic model
of cartilage in unconﬁned compression. Journal of Biomechanical Engineering 121, 340-
347.

Steinmeyer, J ., Ackermann, B., Raiss, R.X. (1997) Intermittent cyclic loading of cartilage
explants modulates ﬁbronectin metabolism. Osteoarthritis and Cartilage 5, 331-341 .

Steinmeyer, J ., Knue, S., Raiss, R.X., Pelzer, I. (1999) Effects of intermittently applied
cyclic loading on proteoglycans metabolism and swelling behavior of articular cartilage
explants. Osteoarthritis and Cartilage 7, 155-164.

Torzilli, P.A., Grigiene, R., Borrelli, J. Jr., Helfet, D.L. (1999) Effect of impact load on
articular cartilage: cell metabolism and viability, and matrix water content. Journal of
Biomechanical Engineering 121, 433-441.

Weaver, B.T., and Haut, R.C. (2005) Enforced exercise after blunt trauma signiﬁcantly
affects biomechanical and histological changes in rabbit retro-patellar cartilage. Journal
of Biomechanics 38, 1 177-1 183.

Wei, P., Golenberg, N., Kepich, E.T., Haut, R.C. (2008) Effect of intermittent cyclic
preloads on the response of articular cartilage explants to an excessive level of
unconﬁned compression. Journal of Orthopaedic Research 26, 1636-1642.

Wei, F. and Haut, R.C. (2009) High levels of glucosamine-chondroitin sulfate can alter

the cyclic preload and acute overload responses of chondral explants. Journal of
Orthopaedic Research 27, 353-359.

Wilkins, R.J., Browning, J.A., Urban, J PG. (2000) Chondrocyte regulation by
mechanical load. Bioheology 37, 67-74.

81

Chapter 5
Effects of acute repair of chondrocytes in the rabbit tibio-femoral joint 6 weeks
following blunt impact using P188 surfactant
Abstract
Recent studies have indicated that there may be a correlation between acute

chondrocyte damage and joint degeneration reminiscent of early stage OA. P188
surfactant has been shown to acutely restore the integity of damaged chondrocytes;
however, its long term efﬁcacy is unknown. The hypothesis of this study was that a
single injection of P1 88 into a traumatized joint would acutely repair damaged cell
membranes and maintain their viability in the longer term. Six rabbits were divided into
two goups, with and without P188 treatment and sacriﬁced after 6 weeks post-trauma.
P188 treatments were administered immediately post-trauma. A decrease in the density of
viable cells was documented in the untreated impacted limb versus its contralateral
control, while no difference in the density of viable cells was documented in the impacted
treated limb versus its contralateral control. The results of the current study conﬁrm the
acute efﬁcacy of P188 treatnnent in the longer term, but, additional studies are still needed

to investigate the chronic implications of the acute repair of cells on the traumatized joint.

82

Introduction

Participation in sports, recreation, and exercise (SRE) is becoming increasingly
popular and widespread in today’s culture. Participation in SRE increases the risk of
acute and chronic injuries such as ligament tears and osteoartlnritis (OA), respectively
(Lane, 1996). Knee joint injuries, such as cruciate ligament damage and meniscal tears
(F elson, 2004), have been associated with excessive compressive forces passing through
the joint causing damage to the articular cartilage. Speciﬁcally, damage to the anterior
cruciate ligament (ACL) leads to characteristic osteochondral lesions in the postero-
lateral aspect of the tibia and/or antero-lateral aspect of the lateral femoral chondyle
(Atkinson et al., 2008). Of concern in the current literature is evidence of damage to
articular cartilage and chondrocytes in regions overlying these bone bruises. Since 50%
of ACL tear patients develop radiological signs of OA within ten years following injury
with or without ACL reconstruction, acute injury to the cartilage and subchondral bone
may play an important role in the progession of chronic joint disease.

Deatln of chondrocytes following traumatic injury has been hypothesized to be
associated with the long-term development of OA as cell death has been correlated with
degeneration of the cartilage matrix (Hashimoto et al., 1998, Duda et al., 2001, Simon et
al., 1996). These degenerative changes result in a loss of tissue integity, represented by a
decrease in tissue stiffrness, and an increase in tissue permeability (Kurz et al., 2001;
Ewers and Haut, 2000; Ewers etal., 2001 ). A study using the porcine patella documents
considerable cellular dysfunction that may act to promote subsequent structural tissue
damage (Duda et al., 2001). This may be particularly important because the synthesis of

cartilage matrix proteins is directly dependant on cell viability and homeostasis (Duda et

83

al., 2001). Since chondrocytes are required for matrix repair and chondrocyte death leads
to matrix loss, chondrocyte death and repair has become a focus of OA research and more
recently cartilage trauma research.

A deﬁning feature of cellular necrosis is swelling of the cell due to a damaged
membrane. Damage to the plasma membrane allows an inﬂux of ﬂuid into the cell
resulting in the inability of the cell to maintain an ionic gadient. As a result the cell
swells and eventually ruptures (Duke et al., 1996). Surfactants, such as poloxamer P188
(PI 88), interact with this damaged cell membrane. P188 is an 8400-dalton triblock
copolymer containing both hydrophobic and hydrophilic regions. Marks et al. (2001)
shows that P188 surfactant speciﬁcally inserts into only the damaged areas of a cell
membrane. Studies by our laboratory, and others, have shown P188 to be effective in
reducing the loss of chondrocyte in articular cartilage following traumatic loading to the
PF joint (Rundell et al., 2005) and TF joint (Isaac et al., in review). Additionally, studies
have documented the ability of P1 88 to repair membrane damage and increase cell
viability in bovine chondral (Baars et al., 2006) and osteochondral explants (N atoli and
Atlnanasiou, 2008).

Because of the gowing interest in SRE, and the increasing number of injuries to
the TF joint, the current study focuses on the effects of a single traumatic load to the
rabbit TF joint. Previous studies have documented increases in cell viability acutely with
P188 treatment; however, to my krnowledge this is the ﬁrst study to investigate long term
efﬁcacy of P1 88. The hypothesis of the current study was that aisingle injection of P1 88
into the rabbit TF joint following traumatic injury would acutely repair the damaged cell

membrane and its efﬁcacy will be validated by a long term increase in the density of

84

viable cells.

Materials and Methods
Impact

Six Skeletally mature Flemish Giant rabbits aged 6-12 months (5.6 :I: 0.2 kg) were
used in this study after approval by an All-University Committee on Animal Use and
Care. All rabbits were housed in individual cages (152 x 152 x 36 cm) and allowed ﬁee
cage activity for this study. Using a previously described impact method, a 1.75 kg mass
with a pre-crushed, deformable impact head (Hexcel, 3.76 MPa crush strength) was
dropped onto the left tibio-femoral (TF) joint of the anestlnetized rabbit (2% isoﬂurane
and oxygen) (Isaac et al. 2008). The right limb served as an unimpacted control. The
impact interface was mounted in ﬁont of a 4.45 kN (1000 lb.) load transducer (Model
AL311CV, 1000 lb capacity, Sensotec, Columbus, Ohio) (Figure 5.1). The mass was
arrested electronically after the ﬁrst impact, avoiding multiple loadings on the joint.

The animals were randomly divided into two goups, ‘P188’ (n=3) and ‘no P188’
(n=3). The ‘P188’ animals received a 1.5 mL injection of P1 88 at 8 mg/mL concentration
(Rundell et al., 2005) in sterile phosphate buffered saline (PBS) into the left, impacted
joint. The right limb received a 1.5 mL sham sterile PBS injection. The ‘no P188’ rabbits
received a 1.5 mL sham injection into botln limbs. The combination of P1 88 in PBS and
PBS sham solutions were ﬁlter sterilized prior to injection using a 0.2 mm vacuum ﬁlter
(N a] gene, Nalge Nunc Int., Rochester, NY). To insure distribution of the injection into

the joint, the limb was manually ﬂexed a number of times following the treatrrnent.

85

 
 
   
  

Gravity accelerated
mass

 

Load transducer

 

Deformable interface

Tibial constraint ________,

 

 

Figure 5.1. Impact experiments were conducted by dropping a gavity-accelerated mass
onto the ﬂexed knee so that impacts were isolated on the TF joint.
Dissection and Harvesting

The animals were sacriﬁced 6 weeks following impact with 85.9 mg/kg BW
Pentobarbital IV. The joint was dissected immediately after sacriﬁce, and examined for
abnormalities. The media] (MTP) and lateral (LTP) tibial plateaus were then prepared for
cell viability analyses. A 6 mm trephine (TREPH-6, Salvin Dental Specialties, Charlotte,
NC) was used to core a region of the MTP and LTP in areas not covered by the menisci,
as these were determined to be regions of high contact pressure during impact (Isaac et
al., 2008). The cores were undercut using a diamond saw (Isomet 11-1180 Low Speed
Saw, Buehler, Lake Bluff, IL) leaving approximately 0.5 mm of bone underlying the
articular cartilage. Coronal slices were taken across the medial (MF C) and lateral (LFC)
femoral chondyles in a predetermined area of interest leaving approximately 0.5 mm of

bone underlying the articular cartilage. All explants were washed three times in

86

Dulbecco’s Modiﬁed Eagle Media: F12 (DMEM: F 12) (Gibco, USA, #12500-039)
supplemented with additional amino acids and antibiotics (penicillirn 100 U/rrnl,
streptomycin 1 ug/rrnl, amphotericin B 0.25 jig/ml), and placed in this supplemented
media with 10% fetal bovine serum in a 24-well plate. The samples were allowed to

incubate for 24 hours in a humidity-controlled incubator (37 C, 5% C02, 95% hurrnidity).

Cell viability

Following incubation, a specialized cutting device was used to obtain full
thickness sections of the explants for cell viability analyses (Ewers et al., 2001) (Figure
5.2). Prior to staining, the slices were rinsed three times with PBS. The slices were then
stained with calcein AM and etlnidiurn homodimer(EthD-1), according to the
manufacturer’s speciﬁcations (Live/Dead Cytotoxicity Kit, Molecular Probes, Eugene,
OR). Following the staining, each sample was rinsed three times with PBS to remove any
excess stain. The sections were viewed under a ﬂuorescence microscope (Leitz Dialux
20, Leitz Mikroskopie and System GmgH, Wetlzar, Germany). Viable cells were
distinguished by the presence of ﬂuorescent calcein AM (geen). The presence of a
damaged plasma membrane was identiﬁed by a bright red ﬂuorescence due to etlnidiurn
homodimer penetrating the damaged membrane. The number of viable and damaged cells
were manually counted by two blinded readers (DI, NG) using an image analysis
progam (Image J, National Institutes of Healtln, 2004). A representative area of each
explant was selected. Its thickrness and width were measured (Signa Scan, SPSS Inc.,

Chicago, IL) and used to calculate the density of cells for each sample.

87

 

Cartilage

Figure 5.2. A custom cartilage cutting device was used to prepare slices of tissue for cell
viability.
Statistical Analysis

A two factor (limb, facet) repeated measures ANOVA with post hoc Student-
Newman-Keuls (SNK) test was used to compare the density of viable cells between the
left impacted limb and the right unimpacted control of both the treated and untreated
animals. A one factor (limb) repeated measures ANOVA was used to document ,
differences between the unimpacted and impacted limb of both the treated and untreated

animals. Statistical signiﬁcance was indicated at p<0.05.

Results

Gross inspection of the joints at necropsy showed no signs of joint disease and no
damage to ligaments or menisci. No statistical differences were found in the times to
peak impact load or the magnitudes of the peak load between treatment goups. The
average peak, inertially compensated impact load and impact duration were 1102 :1: 92 N
and 23.0 :t 0.2 ms, respectively.

Compartrnental analyses revealed no statistically signiﬁcant differences in the

88

density of viable cells between the impacted and unimpacted limbs of either the ‘P188’
(Figure 5.3a) or the ‘no P188’ (Figure 5.3b) rabbits. A statistically signiﬁcant higher
density of viable cells was documented, however, in the riglnt limb (average across facets
excluding the medial tibial plateau) compared to the left limb of the ‘no P188’ rabbits
(p=0.026) (Figure 5.4). No signiﬁcant difference was documented in the impacted versus

unimpacted limb of the P188 treated rabbits.

89

I Impacted
I Unimpacted

dNN
N
§§°
0
l J

§§§§§§

Denslty of Viable Cells (cells/mmz)
8
o

 

 

N
o 8
l l

   

MFC

I Impacted
I Unimpacted

.5 .-
8 8
0 0
n 1

Density of Viable Cells (cells/mm“)
8 F3
0 O
O O

 

 

LFC MFC LTP MTP

b) untreated
Figure 5.3. No signiﬁcant differences were documented in the density of viable cells
between the impacted and unimpacted limbs in the lateral femoral chondyle (LFC),
medial femoral chondyle (MF C), lateral tibial plateau (LTP) or medial tibial plateau
(MTP) of the treated (a) and untreated (b) rabbits.

90

2200 -
2000 -
1800 — an:
1600 —
1400 -
1200 ~
1000 —
800 —
600 —
400 ~
200 -

I impacted
I unimpacted

A

Density of Viable Cells (cells/mm2

 

 

P1 88 Control

Figure 5.4. Analysis of the impacted and unimpacted limbs revealed a signiﬁcant
difference in the density of viable cells in the ‘no P188’ goup. No signiﬁcant difference,
however, was noted in the ‘P188’ goup. This suggests that P188 had a beneﬁt in
preventing the long term degeneration of acutely injured chondrocytes in the TF joint. ""’
indicates a statistically signiﬁcant difference between the impacted and the unimpacted
limbs.
Discussion

The current study was the ﬁrst to document the long term efﬁcacy of P188 on cell
membrane repair following a single, traumatic injury load on the rabbit TF joint. The
hypothesis of this study was that a single injection of P1 88 into the joint would acutely
repair the damaged cells, and that these cells would remain viable 6 weeks post-trauma.
Previous studies document that a 13 J impact to the rabbit TF joint results in a decrease in
cell viability 4 days post-trauma (Isaac, 2009). And, that an acute treatment with P188
results in an increase in the density of viable cells 4 days post-trauma. The current study
documented a statistical decrease in the viable cell density in the impacted limb versus

the contralateral, control limb in the untreated animals (p=0.026). On the other hand, no

signiﬁcant difference in the viable cell density was documented in the impacted limb

91

versus the contralateral, control limb of the P188 treated animals. This analysis supports
the previous ﬁndings, suggesting that P188 was effective in maintaining viability of
acutely repaired cells in the longer-term.

The mechanism that leads ﬁorn tramnatic injury to cell death is largely unknown,
although two pathways, necrosis and apoptosis, result in loss of chondrocytes. Cells
appear to enter an acute necrotic pathway, undergoing swelling and eventually cellular
lysis, following trauma induced damage to the cell membrane (Duke et al., 1996).
Excessive mechanical stress causes cell membrane damage and/or changes in membrane
transport pathways, such as Na+/K+ pump, which regulate cellular volume (Wilkins et
al., 2000). The current study documents cell viability by cell membrane damage, a
deﬁning feature of acute necrosis. Apoptosis may have also been initiated in the damaged
cells following trauma, but not documented by the current cell viability assays. Apoptotic
cells may be documented by TUNEL+ staining. Such cells occur as early as 48 hours
after excessive mecharnical loading (Chen et al., 2001). The percentage of these cells
increases up to 21 days post-trauma in other studies (Clements et al., 2004; Levin et al.,
2001). Apoptosis results in the fragnentation of the nucleus which condenses into
structures that may contain apoptotic enzymes (Majno and J oris, 1995), such as caspase
(D’Lirna et al., 2001). These apoptotic enzymes may be released into the matrix and
initiate cell apoptosis in viable cells (Levin et al., 2001). A previous study by Natoli and
Athanasiou (2008) suggests acute chondrocyte repair using P188 surfactant prevents both
apoptotic and necrotic cell death as preventing necrosis may also inhibit the release of
apoptotic initiators into the matrix. Apoptosis also is known to cause membrane damage

in its later stages (Columbano, 1995). In the current study, all cells with membrane

92

damage would have been identiﬁed as necrotic cells, but some of these may actually be in
the late stage of apoptotic deatln A previous study by our laboratory documents P188 is
effective in preventing apoptotic cell death documented by TUNEL staining in bovine
chondral explants 7 days post-trauma (Baars et al. 2006). Since the density of viable cells
remained at ‘control levels’ in the current study, P188 may be effective in preventing
botln cell death pathways in the longer-term.

A limitation of the current study was the relatively small sample size. Power
analyses revealed, on average, approximately 17 animals would be required to attain
statistical signiﬁcance in the density of viable cells between the impacted and
contralateral control in the LF C, MFC, and the LTP of untreated animals. On the other
hand, because the density of viable cells in the treated limb was actually, on average,
geater than the density of viable cells in the contralateral, control, a signiﬁcantly larger
sample size would be required to show fewer viable cells in P188 treated limbs than
controls. This suggests geater differences in the untreated rabbit, indicating P188 may
have actually restored the density of viable cells closer to ‘control levels’. The data ﬁ’om
the MTP revealed no differences between the impacted and contralateral, control limb in
eitlner the treated or untreated arnimals. Previous studies have documented that the medial
tibial plateau has signiﬁcantly more baseline damage (Golenberg et al., 2008) and,
therefore, excessive loading to the cartilage does not increase the percentage of damaged
cells (Isaac, 2009), limiting the efﬁcacy of P1 88. Therefore, in the above analysis the
MTP data has been excluded.

In an attempt to further examine the long term efficacy of P1 88 in our ﬁrst study

with a lirrnited number of specimens, the data from all compartments of the TF joint were

93

combined. A previous study by Isaac (2009) has previously documented that P188 was
effective in increasing viable cells in all of these compartments. Additionally, all
compartments of the joint were subjected to the same testing protocol; therefore, the
MFC, LFC, and LTP were combined in this analysis. The analysis revealed a signiﬁcant
decrease in the density of viable cells between the impacted limb and contralateral,
control limb of the untreated arnimals, but no difference was documented in the treated
animals. Since the impacted limbs of both the treated and untreated animals were handled
in the same manner both during testing and analysis, and a statistical signiﬁcance was
documented in the density of viable cells of the impacted versus contralateral, control
limb in the untreated animals but not in the treated animals, the efﬁcacy of P1 88 may
have been established in this current study with this limited number of specimens.

Early signs of OA, such as ﬁssuring, were not documented in either the treated or
untreated rabbits of the current study. But, a previous study by Armstrong and Mow
(1982) suggests that the visual or histological appearance of the cartilage is a poor
indication of its mechanical integity. The current study suggests that P188 was effective
in maintaining the viability of acutely damaged cells 6 weeks post-trauma by repairing
the damaged cell membranes, however, the effects of saving these cells on the
mecharnical integity of cartilage was not docnunented. Future studies should focus on the
effects that rescuing the cells with P188 has on maintaining the functional stiffrness of the
cartilage, since this is needed to maintain homeostasis of the joint tissue. Our irnitial
results indicate that pharmacologic approaches, such as treatment witln P188 surfactant,
directed speciﬁcally at the injured cartilage may provide a new approach for decreasing

cell damage and helping to ensure the survival of joint cartilage. Damage to the articular

94

cartilage overlying bone lesions in the ACL tear patient is thought to lead to the
development of OA. Treatment with P188 targeting these damaged cells may lead to

better outcomes in the snu'gically repaired patient and possible 0A prevention.

Acknowledgments

This study was supported by a gant from the Centers for Disease Prevention and
Control, Center for Injury Control & Prevention (CE000623). The contents of this
manuscript are solely the responsibility of the authors and do not necessarily represent
the official views of the Center for Injury Prevention and Control. The authors wish to
acknowledge Ms. Jean Atkinson (J .A.) for her assistance in animal care, Dan Isaac (D.I.)
for his assistance in cell viability analyses and impact procedures, and Eric Meyer for his

assistance in the impact procedures.

95

References

Armstrong, CG. and Mow, V.C. (1982) Variations in the intrinsic mechanical properties
of human articular cartilage with age, degeneration, and water content. The Journal of
Bone and Joint Surgery 64, 88-94.

Atkinson, P., Cooper, T., Anseth, 8., Walter, N., Kargus, R., Haut, R.C. (2008)
Association of knee bone bruise frequency with time post-injury and type of soft tissue
injury. Orthop 31, 440.

Bars, D.C., Rundell, S.A., Haut, R.C. (2006) Treatment with the non-ionic surfactant
poloxamer P188 reduces DNA fragnentation in cells ﬁom bovine chondral explants
exposed to injurious unconﬁned compression. Biomech Model Mechanobiol 5, 133-139.

Chen, C.T., Burton-Wurster, N., Borden, C., Hueffer, K., Bloom, S.E., Lust, G. (2001)
Chondrocyte necrosis and apoptosis in impact damaged articular cartilage. J Orthop Res
19, 703-711.

Clements, K.M., Bnnton-Wurster, N., Lust, G. (2004) The spread of cell death ﬁom
impact damaged cartilage: lack of evidence for the role of nitric oxide and caspases.
Osteoarthritis and Cartilage 12, 577-585.

Colurnbano, A. (1995) Cell death: current difﬁculties in discriminating apoptosis ﬁom
necrosis in the context of pathological processes in vivo. Journal of Cellular
Biochemistry 58, 181-190.

D’Lirna D.D., Hashimoto, S., Chen, P.C., Colwell, C.W. Jr., Lotz, MK. (2001) Impact of
mechanical trauma on matrix and cells. Clinical Orthopaedics and Related Research 1,
$90-$99.

Duda, G.N., Eilers, M., Loh, L., Hoffman, J .E., Kaab, M., Schaser, K. (2001)
Chondrocyte death precedes structural damage in blunt impact tramna. Clinical
Orthopaedics and Related Research 393, 302-309.

Duke, R.C., Ojcius, D.M., Young, J .D. (1996) Cell suicide in heath and disease. Scientiﬁc
American 275, 80-87.

Ewers, B.J., Haut, R.C. (2000) Polysulphated glycosarninoglycan treatments can mitigate
decreases in stifﬁness of articular cartilage in a traumatized arnimal joint. J Orthop Res 18,
756-761.

Ewers, B.J., Weaver, B.T., Sevensma, B.T., Haut, R.C., (2001) Chronic changes in rabbit

retro-patellar cartilage and subchondral bone after blunt impact loading of the
patellofemoral joint. Journal of Orthopaedic Research 20, 545-550.

96

Felson, D.T. (2004) An update on the pathogenesis and epidemiology of osteoarthritis.
Radio] Clin North Am 42, 1-9.

Golenberg, N., Kepich, E., Haut, R.C., (2009) Histomorphological and mechanical
property correlations in rabbit tibial plateau cartilage based on a ﬁbril-reinforced biphasic
model. International Journal of Experimental and Computational Biomechanics 1, 58-75.

Hashimoto, S., Ochs, R.L., Komiya, S., Lotz, M. (1998) Linkage of chondrocyte
apoptosis and cartilage degadation in human osteoarthritis. Arth Rheum 41, 1632-163 8.

Isaac, D.I., Golenberg, N., Haut, R.C. Acute repair of chondrocytes in the rabbit
tibiofemoral joint following blunt impact using P188 surfactant and a preliminary
investigation of its long-tenn efﬁcacy. Journal of Orthopaedic Research (In Review).

Isaac, D.I., Meyer, E.G., Haut, R.C. (2008) Chondrocyte damage and contact pressures
following impact on the rabbit tibiofemoral joint. J Biomech Eng 130, 0410181-5.

Isaac, DJ. (2009) Chapter 5: Acute repair of chondrocytes in the rabbit tibiofemoral joint
following blunt impact using P188 surfactant. In: Investigations on the response of knee
joint cartilage to blunt impact in a small animal model. Thesis for the degee of MS.
Michigan State University, 79-101.

Kurz, B., Jin, M., Patwari, P., Cheng, D.M., Lark, M.W., Grodzinsky, A.J. (2001)
Biosynthetic response and mechanical properties of articular cartilage after injurious
compression. J Orthop Res 19, 1 140-1146.

Lane, N. (1996) Physical activity at leisure and risk of osteoarthritis. Annals Rheumatic
Disease 55, 682-684.

Levin, A., Burton-Wurster, N., Chen, C.T., Lust, G. (2001) Intercellular signaling as a
cause of cell death in cyclically impacted cartilage explants. Osteoarthritis and Cartilage
9, 702-71 1.

Majno, G., Joris, I. (1995) Apoptosis, oncosis and necrosis: an overview of cell death.
American Journal of Pathology 146, 3-15.

Marks, J.D., Pan, C.Y., Bushell, T., Cromie, W., Lee, R.C. (2001) Amphiphilic tri-block
copolymers provide potent membrane-targeted neuroprotection. FASEB 15, 1 107-1109.

Natoli, R.M., Athanasiou, K.A. (2008) P188 reduces cell death and IGF-1 reduces GAG
release following single-impact loading of articular cartilage. J Biomech Eng 130,
041012-1-9.

97

 

Rundell, S.A., Baars, D.C., Phillips, D.M., Haut, R.C. (2005) The limitation of acute
necrosis in retro-patellar cartilage after a severe blunt impact to the in vivo rabbit patello-
femoral joint. J Orthop Res 23, 1363-1369.

Simon, W.H., Richardson, S., Herman, W., Parsons, J .R., Lane, J. (1976) Long-term
effects of chondrocyte death on rabbit articular cartilage in vivo. Journal of Bone and
Joint Surgery, Am 58, 517-526.

Wilkins, R.J., Browning, J .A., Urban, J PG. (2000) Chondrocyte regulation by
mechanical load. Bioheology 37, 67-74.

98

Chapter 6

Conclusions and Recommendations for Future Work

The previous chapters described investigations on the mecharnical and biological
properties of articular cartilage following various loading conditions using a small animal
model and chondral explants. Additionally, interventions using poloxamer 188 and
glucosannine chondroitin sulfate were investigated following injury to the cartilage.

In Chapter 1, differences in the mechanical properties of the rabbit tibial plateau
were documented between the medial and lateral facets, as well as in areas covered and
uncovered by the meniscus. Histological differences were also documented between the
medial and lateral facets. Additionally, surface ﬁssuring and histologically identiﬁed
matrix damage of articular cartilage were documented in the medial compartrrnent that
paralleled with reductions in ﬁber modulus and increases in permeability and tissue
thickness in the medial versus lateral compartments of the plateau. These results suggest
that in using the more complex model of cartilage, the mechanical properties in the model
better deﬁned the topogaphical mecharnical properties of the joint and correlated well
with the matrix components that had been previously suggested to regulate these
parameters. Using the more complex ﬁbril-reinforced biphasic model, future studies are
still necessary to deﬁne the changes in tissue properties due to normal and abnormal
loading of the joint.

Chapter 2 described investigations on the effects of 10 MPa and 25 MPa of
unconﬁned compression on bovine chondral explants, as well as the effects of post-
trauma treatment of the tissue with glucosamine-chondroitin sulfate. A signiﬁcant

decrease in the mecharnical properties of the tissue was documented following 25 MPa of

99

 

unconﬁned compression compared to 10 MPa of unconﬁned compression. Treatment
with glcN-CS resulted in an increase in the proteoglycan content and matrix stiffrness for
botln 10 and 25 MPa of unconﬁned compression. Previous studies have documented the
ability of cyclic loading and glcN-CS to increase PG content pre-trauma.

Cartilage explants were cyclically loaded following two levels of unconﬁned
compression in Chapter 3. Additionally, glcN-CS treatment with cyclic loading was
investigated. An increase in PG content was documented following cyclic loading and
glcN-CS treatment following 10 and 25 .MPa of unconﬁned compression compared to
cyclic loading alone. Similar to the previous chapter, an increase in the matrix stifﬁness
was also documented with glcN-CS treatment following 25 MPa of unconﬁned
compression. HoWever, following 10 MPa of unconﬁned compression and cyclic loading,
no increase in the matrix stifﬁness was docmnented with the treatment of glcN-CS. This
may be due to the size and binding sites of the proteoglycans in the matrix and its
inﬂuence on the mechanical properties of the tissue. A comparison with Chapter 2
revealed no differences between the cyclically loaded and non-loaded explants following
either level of unconﬁned compression. Future studies should investigate the cell viability
and synthesis following unconﬁned compression with and without cyclic loading as well
as the size and shape of the proteoglycans to ﬁrrther understand the effects of unconﬁned
compression. This may lead to possible intervention methods and better treatments for
the damaged cartilage.

Chapter 4 investigates the long-term efﬁcacy of P1 88 following a traumatic injury
to the rabbit tibio-femoral joint. A signiﬁcant decrease in the density of viable cells was

documented in the untreated limb versus the contralateral, control limb. This decrease

100

 

was not found in the treated animals. This would suggest a long term efﬁcacy of P188.
No visual signs of 0A were documented in either the treated or untreated rabbit. Previous
studies suggest that the visual appearance is not always a good indication of the integity
of the cartilage and that analysis of the mecharnical properties are still necessary.
Therefore, future studies should investigate the mechanical properties of the cartilage and
the effects of saving the cells with P188 treatment.

These chapters investigate the changes in homeostasis of the cartilage constituents

following traumatic injmy and possible therapeutic treatments. This may provide a better

 

understanding of the cause and pathways of OA. Future work may focus on these
degenerative changes and a combination of these therapeutic treatments to develop better

preventative and treatment methods of the OA patient in the long term.

101

Appendix A

Raw Data ﬁom Chapter 2

102

l

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

005.3 _. v.m000..5 0 _..mN00.0 0 wum05v.0 00.55500 00+mN w v... 01—. v 00.0 omEo><
005.05 2.00000 v 70000.0 0 70000 .0 00+w00 _. .5 000.000.? 00.0 0 F... '00<_N
000.00 _._.-m_.N_..0 3-00.0.0 20005.0 00+m 00.0 00+m_050.5 2.. E. 00.0 FMO—2MN
0NN.05 30000.0 30000.0 9.0.30.0 00...w00:ﬂ 00+mwvm0 00. E. 00.0 :0EmN
50.50 0 79080 3.030... 0 70000... 00+mv~00 00+MNNNN 5.3 00.0 w¢0><>
0v00 «Tm 3N0 33m 500 30500.0 00+mo00.0 00+00¢0._. 000—. N00 _.n_0><>
500.05 :0 _.00.0 3.00000 20000.0 00+w050.N 00+m~0v._. 00.0.. 00.0 wm00>
5 0.00 2.-va _.._. 3.000 Pd 0 _.-m0_. 5‘ 00+w5 P0._. 00+mv00€ 00.0.. 00.0 500>
.0060 :00 _.5.N 3.00 31‘ 20000.? 00+w500.v 0905000 500—. N00 wmon>
«00.05 3.005 v.0 v rum—000.0 0 70000.0 00.000006 00+M000 .0 V0.0? 0.0 _.._00>
000.00 _. _.-M000.0 S-MOV00 0 F-M55v0 00+w05v.v 00+mNmo._. 00. r _. V0.0 5.00.2!”—
50 _. .00 2.00000 0 9.005000 0 TMO 00 00.00000 00.00000 00. E. 00.0 .352sz
000.05 0.30030 070008? 00..m00 w... 00+M000.N 00.00000 00.0.. 00.0 Fm 905.3.
30. _.5 Sum F0_..N 33m v00... 0 Tw505é 00+M5000 00+w0050 000—. 00. _. 5 ..0 CE
000.00 300300 3.05054. 0 _.-m005.v 00+mw50d 00+M0550 0v. 2. 50.0 3.0me
500.00 N TM500.N v 7500;... 0 _.-m000.¢ 00+m0 r0... 00+m000 ._. 000—. 05.0 50 .5m
@0000 30000.5 $010.5 0 Tw00 _..5 00+.n._00 _..0 00+m000... Nv.0_. 00.0 vmozm
005.05 _. 70000 .0 V 90000.5q 0 Tmovvé 00+M000.0 00+MON F... 3.1:. 00.0 50:0
#0000 0 7005vd 0 F-M00v._. 000000... 00+M5 _.0.F 00.0500... 2.. E. 00.0 5.030
50...05 2300000 3-0900 0_.-w00w.0 00.00 0.0 00+M000._. 00. : 0.0 5030

v.0uo «0253‘s: 0c. «@3320 ox 3.: 0m Aonqﬁm ex; 0 AEEV : :08?on

 

 

 

 

 

 

Dmmm>ooza ASA—ms. . $4015.20”: —. mtw

 

 

._ 86 E 3508 3980 2.0 we 80.635 Bow—snug; H.< 033.

103

.N 86 5 .3508 35:8 20 00

35:82 3. 2%;

 

104

 

.m 0:... a

65:8 20 00

     

35:82 3. 2.3

105

 

.v 8% E 8508 3558 25 mo

35:82 .2 29¢

106

Appendix B

Raw Data from Chapter 3

107

Table B.1 Mechanical properties following 10 MPa of unconﬁned compression. Matrix

Modulus (a) Fiber Modulus (b) Permeability (c).

 

10 MPa (E...- % day 0)

 

 

 

 

 

 

 

10 MPa (ko- % day.0)

 

 

 

 

 

7 I 14 21

65.77 84.07 93.82

80.65 64.49 80.79
109.11 97.74 97.25
107.62

64.91 77.51 58.89

66.08 72.23 95.14

56.01 94.68 52.31

69.36 78.86 102.44
103.83

81.86

a
10 MPa (Ef- % day 0)

7 l 14 I 21
72.61 137.74 152.51
64.35 102.19

183.92 137.34

136.98 116.97 143.84
60.69 102.85 35.40
79.66 131.16 113.02
78.11 126.05 42.79
57.41 87.76 117.49

139.05

110.92

110.15

101.26

 

7 L 14 J 21
169.29 181.00 128.19
176.53 214.06 135.96

84.02 170.06 124.67
74.38 114.93 88.91
138.95 95.47 224.55
95.68 146.75 117.70
213.21 116.24
214.25 140.60 150.25
82.55
76.76
94.96
127.68

 

 

 

C

108

 

 

Table B.2 Mechanical properties following 25 MP8 of unconﬁned compression. Matrix
Modulus (a) Fiber Modulus (b) Permeability (c).

 

25 MPa (Em- % day 0)

 

 

 

 

 

 

 

25 MPa (ko- % day 0)

 

 

 

 

 

7 14 21
43.05 40.67 39.23
63.15 55.52 57.55
57.86 45.85 46.61
62.88 42.15 74.35
44.55 61 .24 54.57
82.1 1 58.82 90.83
80.95
53.10
96.22

25 MPa (Ef- % day 0)
7 14 j 21
222.13 221 .15
229.12 169.73 254.64
214.61
101.90 90.40 227.60
104.49 167.30 139.95
165.12 207.97 153.74
179.33 82.03 181.82
124.96 117.65 107.97
89.26
128.03
70.87

 

 

7 14 21
456.95 408.35
219.91 290.44 235.91
367.81 403.19 329.98
284.67 292.55
172.47 239.79 213.03
93.90 83.01 87.27
172.87 196.79 155.82
95.75 63.67 139.52
21 .78
137.82
167.77

 

 

C

109

 

 

Table B.3 Matrix Modulus following 10 MPa (8) and 25 MPa of unconﬁned compression
with glcN-CS treatment.

 

 

 

 

 

 

 

 

 

 

10 MPa (Em- % day 0) 25 MPa (Em- % day 0)
7 7 14 21 7 14 21
140.50 52.24
131.09 108.13
55.03 152.90
89.60 55.92
60.20 113.41
86.47 100.72
121.33 55.11
132.40 65.36
136.60 41.53
123.93 109.27 88.59 110.40
100.12 102.09 117.17 160.23
117.59 131.87 100.85 50.64 44.67 39.01
119.56 175.88 73.15 81.96 69.67 85.95
77.47 84.18 104.33
126.13 144.82 99.18

 

 

Table B.4 Proteoglycan content with and without glcN-CS treatment.

 

 

 

 

 

 

Proteoglycan Content

NoﬂN-CS glcN-CS
10 MPa [25 MPa 10 MP3 I 25 MPa
27.96 16.49 40.13 23.31
11.69 14.14 40.60 31.62
7.53 17.14 57.20 36.47
11.64 12.44 46.49 31.18
34.86 19.19 52.32 27.30
27.18 31.50 46.89 19.92
44.73 17.24 58.51 71.81
32.71 29.17 63.23 53.49

 

 

110

Appendix C

Raw Data ﬁ'om Chapter 4

111

Table C.1 Matrix modulus following unconﬁned compression and exercise.

 

 

 

10 MPa (Em- % day 0) 25 MPa (Em- % day 0)

7 I 14 21 7 L 14 21

70.18 53.15 81.66 58.07 56.94 57.82

94.94 83.26 53.84 57.98 38.26 46.90

92.44 63.76 100.10 65.58 36.08 49.20

42.17 39.29 38.81 41.33 24.93 41.77

86.92 90.80 83.49 85.54 62.78 66.41

47.70 35.28 57.79 50.31 62.35 62.85
149.91 100.00 67.91 43.38 55.48
104.36 49.21 59.83 62.87 55.90 37.90

80.10 40.70

84.04 47.66

69.31 60.09

50.36 37.58

 

 

 

Table C.2 Matrix modulus following unconﬁned compression and exercise with glcN-CS

 

 

 

 

treatment.
10 MPa (Em- % day 0) 25 MPa (Em- % day 0)
7 I 14 21 7 14 21
96.70 49.57
48.05 72.58
60.09 87.95
92.29 101.96
74.88 131.08
83.21 70.11
71.45 66.64
62.14 117.50
62.62 72.86 71.31 70.89
97.30 106.64 83.29 49.21 49.52 40.01
77.81 112.34 104.07
46.69 42.54 121.91
56.08 44.84 46.73

 

 

 

 

Table C.3 Proteoglycan content with and without glcN-CS treatment.

 

 

 

 

 

 

Proteoglycan Content
No glcN-CS glcN-CS
10 MPa 1 25 MPa 10 MPa 125 MPa
8.13 35.03 21.04 29.54
12.24 31.38 27.01 36.51
7.41 22.10 30.64 38.63
2.58 15.10 26.54 35.75
28.15 31.04 38.59 23.03
28.16 24.15 44.13 24.38
36.55 15.53 44.92 43.71
28.68 18.72 40.69 51.59

 

 

112

Appendix D

Raw Data from Chapter 5

113

Table 13.1 Live cell density analysis of the P188 treated rabbits (cells/m2).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

P188
LFL [ LFM I LTL [LTM [ RFL LRFM [ RTL [Rm
BF732 1749.29 1577.97 937.77 2200.14 1947.67 1334.51 1235.33
1614.04 1487.80 1091.05 1776.35 1758.31 937.77 74.84
1586.99 1289.43 1568.96 1217.29 973.83
Average 1650.11 1532.89 1106.08 1988.24 1758.31 1163.19 761.33
BF744 2064.89 1767.33 1496.82 820.55 2290.31 1054.99 1487.80 1145.16
1767.33 1704.21 1469.77 775.46 1803.40 1370.58 1027.94 757.43
1803.40 1577.97 1623.06 874.65 1397.63 1433.70
Average 1878.54 1683.17 1529.88 823.55 2046.86 1274.40 1316.48 951.29
BF738 1559.94 1181.22 1226.31 1127.12 1307.46 1325.50 1145.16 856.61
1641.09 1370.58 1740.28 757.43 1839.46 1154.17 1018.92 757.43
1568.96 1316.48 1361.56 1181.22 1731.26 1253.36 1054.99
Average 1589.99 1289.43 1442.72 1021.92 1626.06 1244.34 1082.04 889.68
Table D2 Live cell density analysis of the untreated rabbits (cells/mm2).
Control
LFL LLFMT LTL l LTM ERFLIRFM1RTL RTM
TF43 1424.68 1235.33 1163.19 1469.77 1605.02 1803.40 1334.51 775.46
1334.51 1136.14 1217.29 730.38 1695.19 1605.02 1100.07 640.21
1343.53 1190.24 955.80 1596.01
Average 1367.58 1187.24 1112.09 1100.07 1650.11 1668.14 1217.29 707.83
544 1713.23 1190.24 1199.26 766.44 1767.33 1379.60 1677.16 892.68
1722.24 1000.89 1100.07 685.29 1668.14 1397.63 1397.63 685.29
1298.45
Average 1717.74 1095.56 1149.67 725.87 1717.74 1358.56 1537.40 788.99
53BRF 1325.50 1289.43 1217.29 1064.00 1388.62 1550.92 1109.09 1109.09
1983.74 1442.72 1442.72 919.73 1641.09 1460.75 1009.90 892.68
1532.89 1605.02
Average 1614.04 1366.07 1330.00 991.87 1544.91 1505.84 105950 100.89

 

 

114