Endothelial Progenitor Cells for Rotator Cuff Healing

by

Hui Tung Tony Lin

A thesis submitted in conformity with the requirements for the degree of Master of Science

Institute of Medical Science University of Toronto

© Copyright by Hui Tung Tony Lin 2014

ii

Endothelial Progenitor Cells for Rotator Cuff Healing

Hui Tung Tony Lin

Master of Science

Institute of Medical Science

University of Toronto

2014

Abstract

The avascular environment at the tendon-bone interface is a major concern for rotator cuff

healing, contributing to the high re-tear rate. We hypothesized that better tendon-bone healing

would occur with transplantation of endothelial progenitor cells, a population of angiogenic cells,

during rotator cuff repair in rats. The surgical procedure involved a unilateral detachment and

repair of the supraspinatus tendon in a syngeneic population of Fischer 344 rats. The EPC group

received one 106 EPCs and the Fibrin (negative control) group received none. Biomechanical and

histological analyses were performed at two, four, and six weeks post mortem. At 6 weeks, there

was a statistically significant difference in the tendon attachment strength. While no significant

difference was found in histology, the EPC groups showed a trend of increased vascularity. The

EPCs provide improvements in tendon-bone attachment 6 weeks after the repair. However, such

biomechanical improvement was not detectable via conventional histological methods.

iii

Acknowledgments

I owe my deepest gratitude to my supervisors and mentors, Dr Emil Schemitsch and Dr Aaron

Nauth, for their guidance, encouragement, and support. The years in the Musculoskeletal (MSK)

Research Laboratory at St Michael’s Hospital (SMH) have been a wonderful learning

experience. My graduate study has been challenging at times, but Dr Schemitsch and Dr Nauth

were always positively optimistic. Their dedication to research and scholarship is nothing but

inspirational. Without their supervision and mentorship, a suffix “MSc” after my name would not

have been possible.

I would also like to express my gratitude to my program advisory committee members, Dr Jed

Davies and Dr Cari Whyne, for their time, expertise, constructive criticisms, and trips to my

meeting during the dreaded Toronto rush hour.

The companionship and camaraderie of members and former members of the MSK Research Lab

– Erica Giles, Michael Glick, Charles Godbout, and Sarah Ketcheson – shall always be

cherished. Our countless hours in the Vivarium, the lab, the culture hood, and the pub

contributed to the completion of my thesis project and a phenomenal research experience. In

addition, the contribution of our lab rats to scientific advancements will not be forgotten.

Dr Vladimir Iakovlev at Pathology, St Michael’s Hospital, Jian Wang at Dentistry, University of

Toronto, and Lily Morikawa at Toronto Centre for Phenogenomics, were key to my data

collection and analyses. Radovan Zdero of St Michael’s Hospital Biomechanics Lab was

extremely helpful providing guidance in biomechanics. The staff at the Core Research Facility at

SMH had been especially patient with me, taking me step by step through the methods.

Moreover, I am in debt to all the science nerds at SMH for lending their hands and equipment.

I would like to extend my thanks to my friends in fish bowl office 412, Lidice and Alaa, whom I

spent three years sharing my coffee maker with; and Trifecta, cough…Shariq and Yian…cough,

without them, I would have graduated a year earlier.

Finally, I would like to thank my family for their unconditional support and love throughout my

study and life.

iv

Table of Contents

Acknowledgments ____________________________________________________________ iii

Table of Contents _____________________________________________________________ iv

Contributions ________________________________________________________________ vii

List of Abbreviations __________________________________________________________ viii

List of Tables __________________________________________________________________ x

List of Figures ________________________________________________________________ xi

List of Appendices ____________________________________________________________ xiv

1 Introduction _____________________________________________________________ 1

1.1 The Rotator Cuff _____________________________________________________________ 1

1.1.1 Anatomy and Function ______________________________________________________________ 2

1.1.2 Healthy Tendon Insertion ____________________________________________________________ 4

1.1.3 Vascular Supply to the Rotator Cuff ____________________________________________________ 8

1.2 Rotator Cuff Pathology _______________________________________________________10

1.2.1 Clinical Spectrum of Disease _________________________________________________________ 10

1.2.2 Pathophysiology of Rotator Cuff Disease _______________________________________________ 12

1.2.3 Treatment Options ________________________________________________________________ 17

1.2.4 Biology of Rotator Cuff Healing _______________________________________________________ 19

1.2.5 Vascularity of the Rotator Cuff _______________________________________________________ 28

1.3 Rotator Cuff Repair Model ____________________________________________________31

1.3.1 Animal Models and the Rat Model ____________________________________________________ 31

1.3.2 Cell Therapy in the Rat Rotator Cuff Repair Model _______________________________________ 34

1.4 Angiogenesis and Endothelial Progenitor Cells ____________________________________39

1.4.1 Capillaries ________________________________________________________________________ 39

1.4.2 Angiogenesis _____________________________________________________________________ 41

1.4.3 Endothelial Progenitor Cells _________________________________________________________ 44

1.4.4 Applications of Endothelial Progenitor Cells ____________________________________________ 49

1.5 Rationale, Hypothesis, and Objectives __________________________________________53

1.5.1 Rationale ________________________________________________________________________ 53

v

1.5.2 Hypothesis _______________________________________________________________________ 54

1.5.3 Objectives _______________________________________________________________________ 54

2 Methods _______________________________________________________________ 56

2.1 Study Design _______________________________________________________________56

2.1.1 Statistical Significance of the Study ___________________________________________________ 58

2.1.2 Sample Size Calculation _____________________________________________________________ 59

2.2 Cell Isolation _______________________________________________________________60

2.2.1 Endothelial progenitor cell isolation ___________________________________________________ 60

2.2.2 Flow cytometry ___________________________________________________________________ 61

2.2.3 Functional assays __________________________________________________________________ 65

2.2.4 Cell survival in fibrin sealant _________________________________________________________ 67

2.3 Surgery ___________________________________________________________________68

2.4 End Points, Sacrifices, and Dissections __________________________________________71

2.5 Biomechanical Testing _______________________________________________________72

2.6 Histological Analysis _________________________________________________________76

2.6.1 Hematoxylin and Eosin Staining ______________________________________________________ 78

2.6.2 Safranin O/ Fast Green FCF Staining ___________________________________________________ 79

2.6.3 Collagen Birefringence _____________________________________________________________ 80

2.6.4 Vessel Staining ____________________________________________________________________ 82

2.7 Cell Tracking _______________________________________________________________83

3 Results _________________________________________________________________ 85

3.1 Endothelial Progenitor Cells ___________________________________________________85

3.1.1 Flow cytometry ___________________________________________________________________ 85

3.1.2 Functional Assays _________________________________________________________________ 87

3.1.3 Cell survival in fibrin sealant _________________________________________________________ 90

3.2 Biomechanical Testing _______________________________________________________91

3.2.1 Ultimate Load ____________________________________________________________________ 92

3.2.2 Interface Area and Ultimate Stress ____________________________________________________ 94

3.2.3 Stiffness _________________________________________________________________________ 98

3.3 Histological Analyses _______________________________________________________101

3.3.1 Hematoxylin and Eosin Staining _____________________________________________________ 101

vi

3.3.2 Safranin O staining _______________________________________________________________ 102

3.3.3 Picrosirius Red Staining ____________________________________________________________ 103

3.3.4 Vessel Staining ___________________________________________________________________ 106

3.4 Cell Tracking ______________________________________________________________108

4 Discussion _____________________________________________________________ 110

4.1 Endothelial Progenitor Cells __________________________________________________112

4.2 Biomechanical Testing ______________________________________________________113

4.3 Histology _________________________________________________________________115

4.4 Cell Tracking ______________________________________________________________118

4.5 Limitations _______________________________________________________________119

4.6 Clinical Relevance __________________________________________________________121

5 Conclusion and Future Directions __________________________________________ 122

5.1 Conclusion________________________________________________________________122

5.2 Future Directions __________________________________________________________122

5.2.1 Identifying the Angiogenic Mechanisms _______________________________________________ 123

5.2.2 The Mesenchymal Lineage _________________________________________________________ 124

5.2.3 Clinical Application _______________________________________________________________ 125

References ________________________________________________________________ 128

Appendix A ________________________________________________________________ 146

Appendix B ________________________________________________________________ 148

vii

Contributions

The figure of rotator cuff vascular anatomy was modified and adapted from Rothman and Parke.

Clin Orthop Relat Res. 1965 Jul-Aug;41:176-86.

The F344-Tg(UBC-EGFP)F455Rrrc (RRRC#: 307) rats were provided by the Rat Resource and

Research Center P40OD011062, Columbia, MO, USA.

viii

List of Abbreviations

AcLDL Acetylated-low density lipoprotein

APC Allophycocyanin

bFGF Basic fibroblastic growth factor

BMP Bone morphogenetic protein

BSA Bovine serum albumin

CCN Connective tissue growth factor/cysteine-rich 61/nephroblastoma

over-expressed protein

COMP Cartilage oligomeric matrix protein

CTGF Connective tissue growth factor

DAB 3, 3'-diaminobenzidine

DAPI 4', 6-diamidino-2-phenylindole

DiI-AcLDL 1,1' -dioctadecyl -3,3,3',3' -tetramethyl-indocarbocyanine

perchlorate

DNA Deoxyribonucleic acid

EGM2-MV Endothelial Growth Medium 2-MV

EPC Endothelial progenitor cell

FITC Fluorescein isothiocyanate

Flk-1 Fetal liver kinase-1

FMO Fluorescent minus one

GFP Green Fluorescent protein

GH joint Glenohumeral joint

ix

HRP Horseradish peroxidase

IGF Insulin-like growth factor

miRNA Micro ribonucleic acid

MRI Magnetic resonance imaging

MSC Mesenchymal stem cells

MMP Matrix metalloproteinase

MT1-MMP Membrane type 1 matrix metalloproteinase

NSAID Non-steroidal anti-inflammatory drug

PBS Phosphate buffered saline

PCR Polymerase chain reaction

PDGF Platelet derived growth factor

PE Phycoerythrin

PECAM-1 Platelet endothelial cell adhesion molecule 1

PRFM Platelet-rich fibrin matrix

PROM-1 Prominin 1

PRP Platelet-rich plasma

RCF Relative centrifugal force

Scx Scleraxis gene

TGF Transforming growth factor

TIMMP tissue inhibitors of matrix metalloproteinase

UEA-1 Ulex europaeus 1 agglutinin

VEGF Vascular endothelial growth factor

VEGFR-2 Vascular endothelial growth factor receptor-2

x

List of Tables

Table 1-1. Biomechanical results of rotator cuff studies using mesenchymal stem cells

Table 1-2. Histological results of rotator cuff studies using mesenchymal stem cells

Table 2-1. Staining Protocol for flow cytometry

Table 2-2. Tissue dehydration process by Leica TP1020

Table 2-3. Hematoxylin and eosin staining protocol using Lecia Autostainer XL

Table 2-4. Deparaffinization and rehydration of slides using Leica Autostainer XL

Table 2-5. Dehydration of slides using Leica Autostainer XL

Table 3-1. Ultimate Load (N)

Table 3-2. Interface Cross-Sectional Area (mm2)

Table 3-3. Ultimate Stress of Tendon-Bone Interface (N/mm2 or MPa)

Table 3-4. Stiffness of Tendons (N/mm)

Table 3-5. Summary of biomechanical values for 2, 4, and 6 Week Groups

Table 3-6. Collagen birefringence measured in mean optical density

Table 3-7. Vascular density (vessels/mm2)

xi

List of Figures

Figure 1-1. Cross-section (coronal plane) of the shoulder illustrating the relationships of the

glenohumeral joint, the joint capsule, the subacromial bursa, and the rotator cuff (supraspinatus

tendon). Figure adapted from http://what-when-how.com/rheumatology/periarticular-disorders-

of-the-extremities-disorders-of-the-joints-and-adjacent-tissues-rheumatology/

Figure 1-2. Pictorial representation of the tendon-bone transition (safranin O staining). At a

mature fibrocartilaginous enthesis, there are four distinctive, but continuous zones of specialized

tissues: dense fibrous connective tendinous tissue, unmineralized fibrocartilage, mineralized

fibrocartilage, and bone.

Figure 1-3. The vascular anatomy of the rotator cuff insertions (adapted from Rothman. Clin

Orthop Relat Res. 1965 Jul-Aug;41:176-86)

Figure 1-4. The ultimate load to failure at the insertion. The mesenchymal stem cells transduced

with MT1-MMP gene and Scx gene increase insertion strength at 4 weeks after the repair

compared to the control group/MSC group (MSC group was the baseline value for subsequent

genetic modifications). Mesenchymal stem cells alone were not sufficient to improve healing.

Figure 1-5. Surface antigens and possible angiogenic mechanisms of endothelial progenitor cells.

The conventional surface markers are CD 34, CD 133, and Flk-1 (VEGFR2). The mechanisms

for angiogenesis are endothelial differentiation and angiogenic signaling.

Figure 2-1. The study design.

Figure 2-2. Biomechanical apparatus setup.

Figure 3-1. Flow cytometry analysis: 98.3 % of the cultured cells expressed CD 34 on the surface

(left); of these CD 34 expressing cells, 96.8 % of the cells expressed CD 133 and Flk-1 (right). In

summary, 95 % of the cultured cells expressed all three surface antigens.

Figure 3-2. AcLDL and lectin stains: DiI-AcLDL (red fluorescence) staining and DAPI (blue

fluorescence) staining (top); GS-II lectin DAPI (green fluorescence) staining and DAPI DAPI

(blue fluorescence) staining (bottom). Bar = 50 μm.

Figure 3-3. Tube formation assay 24 hours after seeding: endothelial progenitor cells (after ten

days of culturing, left) and mesenchymal stem cells (right). The endothelial progenitor cells

formed tube-like structures while the mesenchymal stem cells were in isolated colonies.

xii

Figure 3-4. Cell survival: cells suspended in fibrin for 24 hours (left) and cell proliferation at 48

hours (right). After the cells were placed into a flask after being in the fibrin clot for 24 hours,

they were able to adhere, resume monolayer morphology, and proliferate.

Figure 3-5. The weight of rats. The weights of the rats in EPC groups and Fibrin groups were not

significantly different at any of the time points. Error bars represent standard deviations.

Figure 3-6. The ultimate load at the insertions. The ultimate load to failure is significantly higher

in the EPC treated repairs at 6 weeks (t-test, p = 0.0025). Error bars represent standard

deviations. The dotted line represents the ultimate load for healthy tendon, 26.39 N.

Figure 3-7. Area of tendon-bone attachments. No differences in the insertion area were found at

any time point. Error bars represent standard deviations. The dotted line represents the area of

tendon-bone attachment for healthy insertions, 2.71 mm2.

Figure 3-8. Ultimate stress at the insertion. The stress significantly higher in the EPC group 6

weeks after the repair (p < 0.0001). Error bars represent standard deviations. The stress for

healthy insertions is 9.84 MPa (not shown on the graph).

Figure 3-9. The stiffness of the tendons. No differences in stiffness were found at any time point.

Error bars represent standard deviations. The dotted line represents the stiffness for healthy

insertions, 23.17 N/mm.

Figure 3-10. Hematoxylin and eosin, Safranin O, picrosirius red, and CD 31 antigen staining

(from left to right) of healthy rotator cuff tendons. Bar = 2 mm.

Figure 3-11. Hematoxylin and eosin. The orientation and the depth of the shoulder sections were

appropriate. Bar = 2 mm.

Figure 3-12. Safranin O and fast green staining. No cartilage formation was detected at the

tendon-bone interface. Bar = 2 mm.

Figure 3-13. Picrosirius red staining. The organization of collagen fibers was assessed under

polarized light. There was no difference in collagen organization between the Fibrin and the EPC

groups. Bar = 2mm.

Figure 3-14. Polarization of tendinous insertion. No significant differences were found between

the groups at any time point. Error bars represent standard deviations.

Figure 3-15. CD 31 staining. The insertion region was selected for analysis. The EPC group

showed an increase in vascularity from Week 4 to Week 6. Bar = 2 mm.

xiii

Figure 3-16. Number of blood vessels per millimeter squared (vascular density) at the insertion

site as identified by CD 31 immunohistochemistry staining. Although the differences were not

statistically significant, the density of vessels was increased in the EPC group at both time points

and the EPC group showed a trend towards increasing vessel density from 4 to 6 weeks, whereas

the Fibrin group showed a decrease in vessel density from 4 to 6 weeks. Error bars represent

standard deviations.

Figure 3-17. The brown HRP-DAB stain indicated GFP-expressing endothelial progenitor cells

(transplanted cells) at Week 2 in the scar tissue at the injury site. Few of the GFP positive cells

incorporated into the endothelium as indicated by the red arrows.

xiv

List of Appendices

Appendix A: Flow cytometry analysis for CD 31, CD 34, CD 133, and Flk-1

Appendix B: Bonferroni’s multiple comparison test

1

1 Introduction

Rotator cuff pathology can cause serious disability in occupational and athletic settings as well as

a major functional impairment for those performing their daily activities. Rotator cuff pathology

is the most common soft tissue injury of the shoulder, and is ranked only second to back and

neck pain in workplace related injuries 86

. Over stressing of rotator cuff muscles by excessive

load during repetitive overhead motion or as a result of acute trauma can cause a rotator cuff tear.

Depending on the severity of the tear, surgical repair may be necessary to restore rotator cuff

function.

With more than 75,000 surgical repairs of the rotator cuff performed annually at an average cost

of $10,605 per operation 195

, the health care utility cost of rotator cuff treatment is estimated to

be over 800 million US dollars alone. The overall economic burden would be much higher as

rotator cuff disability limits physical movement and can have significant impact on work

productivity and sick days. It also compromises daily living activities and severely reduces the

quality of life.

1.1 The Rotator Cuff

The range of motion of the glenohumeral (GH) joint is second to none in the human body.

Ligaments and capsule at the joint provide passive resistance to instability. Active stabilization is

2

primarily provided by a group of four muscles and their tendons, collectively termed the rotator

cuff. From anterior to posterior, these muscles and tendons are the subscapularis, supraspinatus,

infraspinatus, and teres minor. Each of the four muscles and tendons are responsible for the fine

control and the dynamic stability of humeral movements. They centre the humeral head to the

glenoid and prevent excessive glenohumeral translation in addition to rotating the humeral head

about the GH joint. Weak or damaged rotator cuff muscles often lead to limited shoulder range

of motion, poor GH joint stability, and compromised function at the shoulder.

1.1.1 Anatomy and Function

The rotator cuff originates from various bony surfaces of the scapula and inserts into the

proximal region of the humerus. The subscapularis muscle originates from the medial two-thirds

of anterior scapula, which is commonly termed the subscapular fossa. It inserts onto the lesser

tuberosity of proximal humerus. As the sole anterior component of the rotator cuff, the

subscapularis muscle serves to internally rotate and flex the humerus at the GH joint as well as

limit anterior translation of the humeral head. On the posterior surface of the scapula, the

infraspinatus and the teres minor muscles are external rotators and extenders of the humerus. The

two muscles originate from the medial aspect of the infraspinatus fossa and insert into the middle

and lower facet of the greater tuberosity of proximal humerus.

The most superiorly situated muscle of the rotator cuff is the supraspinatus muscle. Originating

from the postero-superior surface of the scapular that is superior to the scapular spine, the

supraspinatus muscle inserts onto the superior facet of the greater tuberosity. The tendinous

3

insertion at the greater tuberosity is not completely separate from infraspinatus and teres minor.

It primarily participates in the abduction of the humerus as well as limits, to some extent, inferior

translation of the humeral head. In addition, the supraspinatus muscle compresses the humeral

head against the glenoid fossa when the arm is in an adducted position; the inferiorly directed

component force serves to depress the humeral head. Due to the complex nature of the bony

structures at the shoulder, the supraspinatus muscle/tendon passes through the subacromial

space. The distal portion of the supraspinatus tendon is wrapped around the contour of the

articular surface of the humeral head, where the articular surface of the tendon is subjected to

compressive forces in addition to tension during the initial phase of humeral abduction.

The subacromial space is defined by the acromion, the acromioclavicular joint, the clavicle, and

the coracoid. Situated between the acromion process and the rotator cuff tendons are the

subacromial and the subdeltoid bursae that help to reduce friction between the structures. In

addition to padding, the bursae also take part in the inflammatory response and provide a

potential vascular source during rotator cuff healing. However, adhesion of the subacromial

bursa to the rotator cuff tendons can effectively narrow the subacromial space and cause

impingement of the tendons.

4

Figure 1-1. Cross-section (coronal plane) of the shoulder illustrating the relationships of the

glenohumeral joint, the joint capsule, the subacromial bursa, and the rotator cuff (supraspinatus

tendon). Figure adapted from http://what-when-how.com/rheumatology/periarticular-disorders-

of-the-extremities-disorders-of-the-joints-and-adjacent-tissues-rheumatology/

1.1.2 Healthy Tendon Insertion

A typical healthy tendon consists roughly of 20 % cellular material and 80 % extracellular matrix

by volume. It is a regular dense connective tissue with relatively few resident cells known as

5

tenocytes. Tenocytes are responsible for producing the extracellular components of the tendon

matter.

A healthy tendon is 55 to 70 % water by weight 96, 162

and 95% of its dry mass is type I collagen.

Type III collagen accounts for less than 5 % of the tendon dry mass in a normal cadaver rotator

cuff tendon. Together, the collagen molecules comprise the majority of solid substance in the

extracellular matrix. The ratio of type III to type I collagen, however, increases with the severity

of degeneration, tendon injury, and age 102

.

In addition to the collagen molecules, the extracellular matrix contains a number of various

macromolecules, termed ground substances, which help stabilize the structural and material

properties. These major ground substances can be classified into two large molecular families,

proteoglycans and glycoproteins. Proteoglycans are shaped like a bottle brush: a core protein

with various glycosaminoglycan chains stemming from it. Glycosaminoglycans are linear

polysaccharides containing repeating units of disaccharides. The negatively charged

proteoglycan molecules are hydrophilic, and thus help retain water in the tissue. Of the

proteoglycan family, decorin and aggrecan are in the tendon and the fibrocartilage 144

.

The glycoproteins are branched molecules which are responsible in maintaining and repairing the

tendon structure. These structural glycoproteins facilitate interaction between cells and their

neighbours as well as cell adhesion to the specific tissues. For instance, fibronectin provides

binding sites for fibroblasts and collagen, and chondronectin is essential for adhesion of

chondrocytes to type II collagen.

Macroscopically, as the rotator cuff tendons approach their skeletal attachment, their

extracellular matrix composition changes. The skeletal attachment of the rotator cuff tendons are

fibrocartilaginous in nature. At a mature fibrocartilaginous enthesis, there are four distinctive,

6

but continuous zones of specialized tissues: dense fibrous connective tendinous tissue,

unmineralized fibrocartilage, mineralized fibrocartilage, and bone. This transition zone is

commonly referred to as the enthesis, footprint, osteotendinuous junction, or tendon-bone

interface. Microscopically, the biochemical composition in the extracellular matrix at the tendon

insertion reflects mechanical demands present at the soft and hard tissue interface. The interface

facilitates the force transmission from tendon to the bone while resisting bending forces when the

line of pull is not parallel to the insertion.

The tendon-bone insertion is where the two type I collagen-rich tissues, the bone and the tendon,

join. At this interface, the calcified and uncalcified fibrocartilage is rich in type II collagen, and

thus presenting a gap phenomenon with immunohistochemical staining of collagen molecules.

Other ground substances such as glycosaminoglycans and proteoglycans help to dissipate the

bending forces away from the insertion to the more flexible tendon body.

7

Tendon

Unmineralized fibrocartilage

Mineralized fibrocartilage

Bone

Figure 1-2. Pictorial representation of the tendon-bone transition (safranin O staining). At a

mature fibrocartilaginous enthesis, there are four distinctive, but continuous zones of specialized

tissues: dense fibrous connective tendinous tissue, unmineralized fibrocartilage, mineralized

fibrocartilage, and bone.

8

1.1.3 Vascular Supply to the Rotator Cuff

The head of humerus receives its blood supply mainly from the axillary artery. Anatomic studies

have demonstrated that the axillary artery branches into six vessels, with the suprascapular and

the anterior and posterior humeral circumflex arteries being the largest contributor of blood

supply to the head of humerus 148

. The ascending arcuate branch of anterior humeral artery

provides the major source of proximal humeral vascularity and the greater tuberosity and the

insertion of the cuff tendons. Immediately proximal to the supraspinatus tendon insertion, there

is an identified anastomotic area known as the critical zone where vascular deficiency is

observed. Most degenerative changes and tears occur in this area. It is believed that the

avascularity at this region compromises the tendon’s metabolic activity and its ability to heal

after micro-trauma. Studies have identified that this critical zone in the supraspinatus tendon

renders it prone to tear 43, 100, 142, 148

. A possible decrease in vascularity was observed in patients

up to six months after surgery 50

.

9

Figure 1-3. The vascular anatomy of the rotator cuff insertions (adapted from Rothman. Clin

Orthop Relat Res. 1965 Jul-Aug;41:176-86)

10

1.2 Rotator Cuff Pathology

While the etiology of pathological changes are still debated, the current proposed causes for

rotator cuff diseases fall under two categories, intrinsic and extrinsic factors. Intrinsic changes

suggest the alteration of tendon chemical environment which compromises the structural

integrity. Extrinsic insults, on the other hand, describe the mechanical influences that result in

structural wear and tear. These factors, in combination, contribute to the pathological progression

in rotator cuff degeneration.

1.2.1 Clinical Spectrum of Disease

The major clinical presentations of rotator cuff diseases include impingement, chronic tears, and

acute tears. Primary impingement implies a decrease in the subacromial space in which the cuff

tendons passes through. Increased friction with the surrounding tissues causes accelerated wear

and tear. The symptoms show characteristics of Neer’s three-stage classification for

impingement: (I) edema and hemorrhage, (II) fibrosis and tendinitis, and (III) degeneration, bony

changes, and tendon ruptures. Instability at the glenohumeral joint capsule or lagging of scapular

stabilizers can result in unfavourable humeral head translations, and consequently, secondary

impingement of the cuff tendons. Excessive contact with the bony structures such as the postero-

11

superior glenoid rim frequently causes partial tearing of the supraspinatus and infraspinatus

tendons.

Long term repetitive use in overhead motions combined with the inherent poor blood supply to

the rotator cuff tendons can lead to degenerative changes and ultimately chronic tears. The

disease here manifests in poor tendon property and compromised healing capacity, commonly

seen in overhead workers and the in elderly population. In contrast, acute tears of the rotator cuff

tendons are often the result of traumatic incidences such as glenohumeral joint dislocation.

Incidences of rotator cuff tears have been observed in cadaveric specimens over the age of 50 to

be between 39 and 60 % 40

. It was estimated that the prevalence of rotator cuff tears may be up to

40 % of asymptomatic individuals over the age of 50 202

. Recent magnetic resonance imaging

studies of rotator cuff tears in populations reported that 37 % of the patients had tears of the

supraspinatus and infraspinatus tendons 154

, and found that the majority of these tears (70%)

were at the rotator cuff insertion 194

.

Clinically rotator cuff tendon tears are categorized into either partial-thickness or full-thickness

(complete) tears. The Ellman classification 47

of partial-thickness tears indicates the location

(articular, bursal surfaces, or intrasubstance) of the tear. Then according to severity, the tears are

sub-classified into Grade I tear (tears less than 3 mm deep), Grade II lesion (tears 3 to 6 mm

deep), and Grade III lesion (tears greater 6 mm deep). Here the normal cuff tendon thickness is

considered to be between 10 to 12 mm. If, however, the tears are full-thickness, a separate but

similar classification system, the Cofield classification 41

, is used. A full-thickness tear suggests a

complete tear involving one or more cuff tendons. A small full-thickness tear is less than 1 cm in

size, a medium tear between 1 to 3 cm in size, a large tear between 3 to 5 cm, and a massive tear

greater than 5 cm.

12

1.2.2 Pathophysiology of Rotator Cuff Disease

Intrinsic Factors of Rotator Cuff Disease

From a histolopathological point of view, rotator cuff tendinopathy often manifests in

degenerative changes, and is widely believed to be the major factor leading to rotator cuff tendon

tears. Structural and chemical alterations can be detected using biochemical and histological

analyses. It is uncertain, however, if degeneration is the initiating pathology or secondary to

other conditions such as impingement or chronic overuse.

The proposed intrinsic factors include the cuff quality reflected by the collagen composition and

the presence of fatty infiltration, cytokines and inflammatory markers, tendon remodeling

molecules the matrix metalloproteinases and their inhibitors, and genetics, diet, and lifestyle

factors.

i. Rotator cuff quality

To determine the pathological progression of rotator cuff tendinopathy, markers of degeneration

are used to assess the tendon quality. Signs of degeneration can manifest in the absence of

inflammatory cells, lack of fibril structural organization, atrophy of fibrils, changes in cellularity,

cell apoptosis, glycosaminoglycan accumulation, and fatty infiltration. With no changes in the

expression of cytokines and receptors, however, inflammation is absent in the degenerative state.

It is not known whether inflammatory response has taken place in earlier stages. Evidence

13

suggests that the extracellular matrix turnover in chronic tendinopathy reflects the cell-mediated

remodeling process, but with prolonged impaired remodeling and gradual deterioration 144

.

Those tendons suffering from injury or degeneration have a higher ratio of type III to type I

collagen. High levels of type III collagen, along with increased hydroxylysine content and

excessive cross-links in collagen fibers, are characteristic features of mechanically weaker scar

tissue. In addition, changes in water content can be indicative of tendon pathology. Injured

tendon will synthesize more extracellular matrix proteins in an attempt to maintain structural

integrity. Consequently, these proteins draw water into the tendon, thereby altering the

mechanical property.

Torn rotator cuff tendons may lead to degenerative fat accumulation in the corresponding

muscles. Fatty infiltration reduces tendon quality and predisposes repaired tendon to re-tear. The

degree of fatty infiltration is associated with the severity of the tear, the type of tendon lesion, the

age of the patient, and the chronicity of the tear. While accumulation of fat can be arrested upon

successful repair and restoration of tendon function, the existing fatty content is not always

reversible 61, 205

.

ii. Cytokines and inflammation

In the early phase of inflammatory response, the key modulators are the signaling molecules –

cytokines. They act as molecular agents to promote cellular proliferation, metabolism, apoptosis,

and extracellular matrix protein turnover in tendon tissue. The degree of inflammation and

angiogenesis, however, is inversely proportional to the age of the patient, particularly on the

articular side. In degenerative tendinopathy, imbalances of cytokine levels and inflammatory

responses fails to trigger repair in the tendon, resulting in poor tendon quality 60, 102

. Chronic

14

overuse and repetitive stress significantly reduces the number of resident tendon cells, causing

impairment to the tendinous repair process. Specifically, oxidative stress, inflammatory

cytokines and regulators, and apoptotic mediators are potential factors for stress-induced

apoptosis of fibroblast-like cells 116, 197, 210

. The inflammatory response is a double edged sword

which may initially be detrimental to the healing of tendon and triggering tendon cell apoptosis,

but the process is essential for the repair of tendon tissue. The absence of acute inflammation

results in poor clinical outcome in pathologic supraspinatus tendons. The relations between

tendon degeneration, tendinopathy, and inflammatory responses have yet to be clearly

delineated.

Growth factors are biochemical molecules which stimulate cellular activities, namely in

proliferation and differentiation. Researchers have been trying to identify the roles of the various

growth factors that contribute to the process of tendon-to-bone healing. Well-defined spatial and

temporal expression of relevant growth factors will aid the progress in targeted treatment

modalities. The early phase temporal expression of growth factors and tendon-associated

proteins were examined at the tendon-bone interface in a rat supraspinatus repair model 203

.

These proteins include basic fibroblastic growth factor (bFGF), bone morphogenetic proteins

(BMP) -12, -13, and -14, cartilage oligomeric matrix protein, connective tissue growth factor,

platelet-derived growth factor-β (PDGF- β), and transforming growth factor-β (TGF- β). Their

temporal profiles are thought to correspond to the various stages and the associated responses of

acute tendon-bone healing. Increased levels of these proteins were found in the first week post-

operation, but the levels returned to the normal values by four months. These growth factors

have been independently investigated in many studies as well (see Biology of Rotator Cuff

Healing section).

15

iii. Matrix metalloproteinases and tissue inhibitors of matrix metalloproteinases

Tendons undergo continuous remodeling, and the tendons that experience high mechanical stress

such as the supraspinatus tendon will exhibit a greater extracellular matrix protein turnover rate

than those subjected to lesser stress, such as the distal biceps tendon. Extracellular matrix

metabolism is a balance between catabolic and anabolic activities. The major class of enzyme in

tendon degradation is the matrix metalloproteinases (MMPs). Their activity is rivaled by tissue

inhibitors of matrix metalloproteinases (TIMMPs). MMPs are found at low levels in normal

tendon tissue but high amounts are seen in injured tissues 39

. Elevated MMP-1 synthesis in

tendon cells is associated with up-regulation of cytokines and inflammatory modulators 197

. The

expression of TIMMPs acts in a reciprocal manner to MMPs in torn tendons 39, 57

. The imbalance

between these extracellular matrix enzymes and inhibitors reduces tendon quality and results in

the progression of degeneration.

iv. Genetics, lifestyle, and diet

Other possible factors which lead to degenerative changes in the rotator cuff tendons have been

investigated. Studies have proposed genetic predisposition to the pathogenesis of rotator cuff

disease. For instance, Harvie and co-workers found that familial influence doubled the risk of

developing rotator cuff tears and increased the chance to experience symptoms five folds

between patients and their siblings 69

. Not surprisingly, lifestyle choice such as smoking

increases tearing risk in a dose-dependent manner even after adjustment for age 13

. Longo et al

investigated rotator cuff tears in association to patient lipid and glucose profiles. In both the

16

prospective serum lipid level study and the case control plasma glucose study, there seemed to be

a trend towards a correlation between high values and the prevalence of rotator cuff tears 101

.

While no direct histological or pathological correlation was made, the studies alluded to macro-

circulatory and micro-circulatory impairment mediated tendon degeneration. However,

atherosclerosis, a macro-circulatory disease, did not correlate with rotator cuff tears 44

. Lastly,

hypertension is believed to be a potential factor for rotator cuff tear and its severity 67

. It was

postulated that circulatory impairment may initiate degenerative changes in the footprint, and the

hypoxic condition in the tendon was exacerbated by the administration of antihypertensive drugs.

Extrinsic factors of Rotator Cuff Disease

Intrinsic factors play a role in the degenerative changes of the tendons. However, the other major

contributor to rotator cuff disorders are the extrinsic factors that contribute to tendon damage.

Structural wear and tear often results from mechanical compression of the supraspinatus tendon

against the coracoacromial ligament and the anterior acromion. Neer observed bony

impingement in rotator cuff tears and this mechanical characteristic has been described as an

extrinsic factor potentially contributing to rotator cuff pathology. Subsequent classification of

acromial morphology showed correlation between hooked and curved acromions with rotator

cuff tendon tears 125

. The shape of the acromion can be congenital or acquired through age-

related changes. Excessive mechanical stress between the anatomical structures aided by

pathophysiological conditions at the acromial region may encourage the pathogenic progression

to a hooked acromion as well as the growth of acromial bony spurs.

17

Repetitive injuries and chronic overuse are recognized associations with rotator cuff pathology.

The micro-trauma theory proposed by Nho and colleagues suggested that small injuries

accumulate over time to result in significant structural deterioration, predisposing the tendon to

acute tear 126

. The overuse injury is frequently observed in the shoulder of the dominant arm in

patients. In one study, about 75 % of the patients presented full-thickness tear on their dominant

side, and 36 % to 50 % of the patients had bilateral tears 40

.

1.2.3 Treatment Options

Non-operative treatment is usually the first-line of treatment for most of the tears, with the

exception of large, acute tear in young and active patients. When non-operative management

fails, surgical intervention may be considered in hopes to reduce pain and restore function. The

standard surgical repair of rotator cuff tears involves approximating the remains of the ruptured

tendon to the humeral head and inserting them into a bony trough on the greater tuberosity.

Within recent decades, surgical methods have evolved from open, to mini-open, to arthroscopic

repairs. The attachment techniques have changed over the years as well, from transosseous

suture repairs to single rows and double rows of suture anchors, and recently the suture bridge

technique.

Generally, strenuous overhead activities are avoided for at least four to six weeks post-surgery

for most of the repairs. Passive range of motion exercise is prescribed for six weeks followed by

active motion in physiotherapy. As the shoulders approaches full range of motion, a more

aggressive rehabilitative program is prescribed. At least a six month recovery prior to return to

18

sports is recommended. Despite all the precautionary and rehabilitative strategies, functional

recovery to original strength is, however, is infrequent.

Re-tear

Despite the evolution of surgical techniques and improved repair methods, structural healing of

the rotator cuff tendon to the humeral head still remains a challenge. While studies have found

the tendons to heal after the rotator cuff repair, the incidence of re-tear is a major concern for

current repair methods. A recent systematic review of post-operative outcomes indicates

recurrence rates from 13 % in small tear less than 1 cm to 61 % in massive tears over 5 cm 45

.

The causes of re-tearing after repair have been proposed: mechanical failure of repair construct

and biological failure of healing at tendon-bone interface 117

. Re-tears that occur within three

months after the repair are attributed to mechanical failure; recurrences after the initiation of

physical therapy are more reflective of the failure of biological healing to provide adequate

mechanical strength.

The success of surgical repair relies heavily on the functional healing of the injury site, the

maintenance of contact between the separated structures, and the regeneration of the enthesis 92,

181, 182. The hypovascular and hypocellular nature of tendons may be the primary reason for poor

healing and lack of regeneration of a specialized tissue to connect tendon and bone 180

.

Insufficient blood supply hinders reparative activities at the injury site and the lack of resident

cells impedes healing and remodeling processes. Immature scar tissues are not replaced by the

strong, mature collagen types appropriate at the tendon-bone interface. This predisposes the

19

repair to failure in the future as the presence of fibrovascular scar tissue renders the interface

mechanically weaker than the native fibrocartilaginous attachment.

1.2.4 Biology of Rotator Cuff Healing

The process of tendon healing follows closely the general sequence for soft tissue healing:

inflammation, proliferation, and remodeling. Hematoma usually occurs immediately after the

initial trauma. While the platelets clot the damaged vessels to contain erythrocytes within the

circulatory system, the platelets release chemotactic factors to promote the infiltration of

inflammatory cells, mainly the monocytes and the macrophages. Angiogenesis and residential

tenocyte proliferation are initiated at this stage as well as fibroblast recruitment and collagen

synthesis. A few days after the initial trauma, the injured site enters the proliferative stage in

which fibroblasts and tenocytes are rapidly recruited with type III collagen synthesis reaching its

peak. During this stage, neovascularization reaches its peak as the metabolic demand is high. At

this point the wound becomes scar-like. The remodeling stage occurs weeks after the injury as

the overall cellularity and type III collagen decrease. Type III collagen is progressively replaced

by the structurally organized type I collagen to yield superior mechanical strength. The water

content and extracellular matrix components approach the normal levels of a tendon tissue. This

remodeling and maturation process may take up to four months 162

.

Although the healing process is slow, it is essential that the tendon remodeling achieves a

structure and content which is similar to the uninjured counterpart. Unfortunately, in many cases,

20

the healed tendon is unlikely to reach its original biomechanical capacity as scar tissue is inferior

to the native tendon tissue.

Growth Factors

Growth factors are biochemical molecules which stimulate cellular activities, namely in

proliferation and differentiation. Researchers have been trying to identify the roles of the various

growth factors that contribute to the process of tendon-to-bone healing. Well-defined spatial and

temporal expression of relevant growth factors will aid the progress in targeted treatment

modalities. The temporal expressions of eight early-phase growth factors in supraspinatus

tendon-to-bone healing were characterized by Würgler-Hauri et al 203

. In addition, Oliva et al has

recently compiled a list of known growth factors expressed in the three defined healing stages

127.

i. Basic fibroblastic growth factor

Consisting of a single chain polypeptide of 146 amino acids, basic fibroblastic growth factor

(bFGF) has been shown to promote wound healing via its angiogenic activity 10, 186

and induction

of a proliferative response 28, 183, 186

. The proliferative effect on tenocytes was confirmed by

Takahashi et al., but the administration of bFGF suppressed collagen synthesis in vitro 174, 183

.

The delivery of exogenous bFGF gene into tenocytes, however, significantly increased the

expression of type I and III collagen gene in vitro 198

. In vivo studies of rotator cuff healing

showed the peak of bFGF expression at week one 90, 203

, and another increase at week eight 203

.

21

Recent in vivo investigation of bFGF application in a rat supraspinatus model demonstrated

accelerated tendon-to-bone repair 79, 80

.

ii. Bone morphogenetic proteins

The bone morphogenetic proteins (BMPs) were first identified in bone extracts, and named

according to their ability to directly induce ectopic bone formation, although they have also been

found to participate in the growth and development of other tissues. BMPs constitute the largest

subfamily of the transforming growth factor beta (TGF-β) superfamily, with over twenty

identified members to date 23, 32

. With respect to their sequential similarity and functions, BMPs

are typically divided in four smaller groups: BMP-2/ -4, BMP-5/ -6/ -7/ -8a/ -8b, BMP-9/ -10,

and BMP-12/ -13/ -14. Within the smaller groups, BMP-12/ -13/ -14 are believed to be involved

in tendon regeneration 52, 122, 200

. Acute increase of these BMPs at the supraspinatus tendon-to-

bone attachment site within the first 2 weeks post-operation was reported by Würgler-Hauri et al,

with the changes of BMP-12 being the most significant 203

. BMP-2 and BMP-7 induced

moderate anabolic activities in rotator cuff tendon cells in vitro which were favourable for the

healing of the tendon 133

. In a sheep rotator cuff repair model, BMP-12 was found to improve

tissue healing and biomechanical properties 160

. Adenovirus-mediated BMP13 delivery to rotator

cuff tendon repair in rats resulted in up-regulation of the tendon matrix gene expression and

improved biomechanical strength 211

.

22

iii. Cartilage oligomeric matrix protein

Cartilage oligomeric matrix protein (COMP) is the fifth member of the thrombospondin protein

family, and is therefore also known as thrombospondin-5. It is a non-collagenous pentameric

glycoprotein present in the extra-cellular matrix of both tendon and cartilage 24

. It binds to both

type I and II collagen, the major components of fibrocartilagous tissues. Södersten et al observed

a positive correlation between the ultra-structural distribution of COMP in horse digital flexor

tendons with respect to age, maturation, and training 158

. A similar correlation was found in the

expression of COMP and the exercise load in rat Achilles tendons 159

. Although the function of

COMP has yet to be clearly defined, current evidences suggest its role in supporting structural

integrity of tissues. The expression of COMP is found in cultured monolayer chondrocyte,

tendon and ligament cells. In the study of temporal expression of growth factors, Würgler-Hauri

et al reported a significant increase of COMP level the first week immediate to the surgery 203

.

iv. Connective tissue growth factor

Connective tissue growth factor (CTGF) belongs to the connective tissue growth factor/cysteine-

rich 61/nephroblastoma over-expressed protein (CCN) family 24

. While CNN proteins have

similar structures, their biological functions vary greatly and are very dependent on the local

cellular context. Their participation in angiogenesis 25

, chrondrogenesis 121

, wound healing and

fibrosis 199

has been implicated in many studies. CTGF is a promoting factor for anabolic activity

of the extracellular matrix, including fibroblast and endothelial cell proliferation, migration,

adhesion, and survival 104, 118, 121, 163

. The expression of CTFG in mesenchymal stem cells

(MSCs) was able to direct the MSCs down the fibroblastic differentiating pathway 94

. Elevated

23

expression of CTFG was found in tendons after acute exercise 72

or injury 31

. The observations

were consistent with the report of an initial increase of CTGF in the supraspinatus tendon

insertion site at one week post-surgery 203

.

v. Platelet-derived growth factors

Platelet-derived growth factors (PDGFs) are a group of mitogens and chemo-attractants for cells

of mesenchymal origin. The PDGF family is composed of four structurally and functionally

similar polypeptides, designated PDGF-A/ -B/ -C/ -D, and the polypeptides can be joined by

disulphide-bonded to form dimers. The expression of PDGF-BB has been observed in tendon

healing models in close association with other growth factors such as bFGF, TGF-β, and IGF in

vitro209

and in vivo 90, 203

. Genetic modification in tenocytes to express PDGF-B 192

or the

addition of PDGF-BB to fibroblasts 185

enhanced collagen gene expression in vitro.

Thomopoulos et al investigated flexor tendon repair and found that PDGF-BB promoted cell

proliferation and collagen remodeling 188

as well as biomechanical properties 184

. The efficacy of

PDGF on rotator cuff tendon healing has been tested by two groups. Uggen et al evaluated both

in vitro and in vivo application of genetically transduced tendon fibroblasts, and reported a

significant increase in collagen synthesis in cell culture and accelerated rotator cuff repair

histomorphologically in rats 192

. The other study by Hee et al demonstrated superior

biomechanical and histological properties at the infraspinatus tendon insertion 70

.

24

vi. Transforming growth factors

As their names suggest, the transforming growth factor-βs (TGF-βs) gained their reputation from

their ability to induce and transform the growth of fibroblasts 151, 170

. With the discovery of more

structurally and functionally related proteins, the TGF-βs now represent one of the four major

subfamilies of the big TGF-β family of over 40 members 151

. Three isoforms of TGF-βs have

been identified in mammals: TGF-β1, TGF-β2, and TGF-β3. The TGF-βs play important roles in

the regulation of collagen synthesis during wound healing, tissue synthesis, fibrosis, and scarring

as well as many other cellular processes.

In vitro studies provided evidence for the fibrogenic potential of TGF-β1 in tendons 29, 88, 89

.

TGF-β1 has been widely associated with aggressive inflammation and wound healing responses

30, 151. The response of collagen formation, however, varies with cells isolated from different

parts of a tendon, as shown by Klass et al with rabbit flexor tendon cells 88, 89

. There is a general

consensus that over-expression of TGF-β1 may have adverse effect by increasing type III to type

I collagen ratio and tendon adhesion 30, 87-89

. Many studies of the supraspinatus tendon model

confirmed transient and localized increase of TGF-β1 one week after surgery or injury 54, 203

,

suggesting the role of TGF-β1 in the initial phase of tendon healing. Peak cell proliferation and

cellularity at the injury site coincided with the increased level of TGF-β1 90

.

Studies have noted the phenomenon that fetal wound healing is achieved in a regenerative

manner, often termed scarless healing, while post-natal healing is achieved in a scar-mediated

process 161, 168

. This difference in fetal and adult healing property can be attributed to expression

of growth factors unique to the different stages of development 20, 161

. Manning et al attempted to

mimic the fetal development and scarless healing stimulus via exogenous addition of TGF-β3.

25

They hypothesized the sustained delivery of TGF-β3 would stimulate regeneration of

fibrocartilage at the adult tendon-to-bone insertion rather than proliferation of scar tissue at the

region. The administration of TGF-β3, however, was insufficient to create a biochemical milieu

which promotes scarless healing in mature rat rotator cuff enthesis. In fact, Manning et al found

increased disorganized scar fibres at the fibrocartilaginous transition zone. The authors

speculated the importance of the role of tissue environment and cellular composition which

affected the healing property in the presence of TGF-β3. Nonetheless, early structural and

biomechanical advantages were observed. Compiling their observations, the data suggested an

accelerated response of all phases in the healing process with the addition of TGF-β3 111

. Kim et

al, however, reported that the application of exogenous TGF-β3 and neutralizing antibodies to

TGF-β1 and -β2 did not improve supraspinatus tendon-to-bone healing 87

. It is important to

better define the healing properties of the TGF-βs and the other key factors to modulate and

augment early tendon healing.

vii. Insulin-like growth factor

Insulin-like growth factors (IGFs) are liver-made proteins that participate in growth,

development, and metabolic activity of various tissues. The stimulatory effects of IGF-I have

been identified in musculoskeletal tissues. IGF-I has been implicated as a mediator in phases of

wound healing of various tissues. Analyses of temporal expression of growth factors revealed

that IGF-I expression occurred early on during the inflammatory phase 31, 90

, and then peaked

again at four weeks 37

. Other studies have found increased extra-cellular matrix synthesis and

protein turnover in vitro 1, 2

, and accelerated tendon healing in rat models in vivo 93

. To date,

26

studies on IGF-I applications have yet to examine their efficacy on the healing at the tendon-to-

bone interface.

viii. Matrix metalloproteinases

The matrix metalloproteinases (MMPs) are a family of zinc endopeptidases found in a range of

multicellular organisms. MMPs are responsible for timely degradation of extra-cellular matrix in

development, morphogenesis, tissue repair, and remodeling. A compilation of the MMPs and

their association to orthopaedic applications was provided by Pasternak and Aspenberg 132

, and

Garofalo et al 58

. It is commonly believed that general inhibition of MMPs will impair tendon

healing. Modulating the levels of MMPs present at the injury may facilitate tendon healing. Bedi

and coworkers reported a distinct histomorphologic difference at the supraspinatus tendon-to-

bone insertion with local delivery of a universal MMP inhibitor, α-2-macroglobulin protein.

They found greater fibrocartilage formation and collagen organization, and a significant

reduction in collagen degradation within four weeks while no biomechanical advantage was

detected at the insertion 17

. Gulotta et al took a different approach by introducing the

developmental gene MT1-MMP into mesenchymal stem cells before transplanting them to the

tendon-bone insertion in a rat supraspinatus tendon. At four weeks, they obtained a higher

presence of fibrocartilage and an improved biomechanical property at the tendon-bone insertion

site 64

. Future investigations are necessary to better characterize the effects of the different

MMPs to the structural integrity of the repair. Modulation of MMP activity in the perioperative

period may offer a novel biologic pathway to augment tendon-bone healing after rotator cuff

repair.

27

Platelet-Rich Plasma

Platelet-rich plasma (PRP) is a natural storage vehicle of growth factors. The current working

definition of PRP is a volume of autologous plasma with a platelet concentration of more than

106 platelets/μL

112. It is obtained from whole blood by separating red and white blood cells from

the plasma and platelets, and subsequently concentrating the plasma contents and platelets.

Taylor et al compiled a list of GFs and bioactive proteins found in PRP: TGF-β1, PDGF, bFGF,

VEGF, IGF-1, epithelial growth factor, and hepatocyte growth factor 177

. Many clinical studies

have been done with the application of PRP on musculoskeletal injuries, but only a handful have

focused on rotator cuff tendon repair 81, 110, 141

. The most recent study which investigated the

efficacy of PRP found that PRP application during arthroscopic repair of supraspinatus tendon

did not result in accelerated recovery 81

. These clinical studies and randomized controlled trials

81, 110, 141 have failed to demonstrate improved healing with PRP application at the time of rotator

cuff repair.

A variation of PRP, termed platelet-rich fibrin matrix (PRFM), is also being used in attempt to

augment rotator cuff repair. The safety and efficacy of the use of PRFM has been evaluated in

randomized controlled trials 27, 145

as well as in other clinical investigations 12, 21

. The randomized

trials by Castricini et al 27

and Rodeo et al 145

, and the clinical trial by Bergeson et al 12

found that

PRFM did not improve shoulder functions while Barber and colleagues showed lowered

incidents of re-tear rates 21

. However, no significant adverse effect was observed in any of the

studies. It should be noted that there are considerable difference exist amongst various

commercial PRP constructs available for use. Variability may present in the density of platelets

and growth factors, the presence of anticoagulants, and the leukocyte infusion.

28

1.2.5 Vascularity of the Rotator Cuff

Similar to other tissues of the body, vascular changes are indicative of tissue health and

condition. Normal tendon is relatively hypovascular when compared to its adjacent

musculoskeletal structures, the bone and the muscle. The vascular supply to the supraspinatus

tendon and the presence of its critical hypoxic zone has been the spotlight of debate. Rotator cuff

tears often involve the supraspinatus tendon, and some speculate the presence of a hypoxic

region which contributes to the progression of tendon degeneration in a similar fashion to

hypoxic degenerative tendinopathy in the Achilles tendon.

However, the theories of hypovascularity and predisposition to degeneration and injury of the

rotator cuff tendon in this region remain a source of controversy. The traditional description of

the critical zone corresponds to a region of anastomoses between skeletal and tendinous blood

supplies at the supraspinatus tendon. The area of hypovascularity at the supraspinatus tendon is

between 5 mm proximal to the humeral insertion and the musculotendinous junction. The critical

zone was more evident at the articular side of the tendon where almost no vessels were present

distally in comparison to the bursal side. It was suggested that this deficient vascularization

predisposes the supraspinatus to degenerative pathologies. Furthermore, full adduction of the

arm compresses the supraspinatus tendon against the humeral head and increased pressure on the

supraspinatus tendon effectively reduces perfusion to the articular side of the tendon 107, 142

.

The concept of enhancing blood flow to injured regions has been applied to therapeutic

treatments of chronic Achilles tendinopathy 130, 131

, patellar tendinopathy 171

, and lateral

29

epicondylitis 113

. Glyceryl trinitrate, a topical drug, was used in various studies to induce local

vasodilation and hence increase blood flow to the applied area. A double-blinded randomized

controlled trial examined the effect of glyceryl trinitrate for chronic rotator cuff tendinopathy,

and found positive effects including force and functional measures in patient outcomes 129

.

Recently, the use of contrast-enhanced ultrasonography allowed a better visualization of the

vascular dynamics of the rotator cuff tendon in human subjects, confirming a hypovascular zone

in the supraspinatus tendon. Studies have evaluated the spatial distribution and age-related

changes to circulation of the supraspinatus tendon in healthy adults as well as alternations after

tendon tear or repair 4, 55, 149

. There was a consistent region of decreased vascularity at the

articular medial margin of the asymptomatic rotator cuff tendons. While all regions experienced

an enhanced blood flow after exercise, the response at the articular medial region was

considerably lower in comparison to the articular, lateral and bursal aspects of the tendon.

Funakoshi et al did a longitudinal characterization of the vascular pattern at the supraspinatus

tendon and its associated anatomy and found unique temporal vascular responses in the

investigated structures 53

. The vascular response of the bursal aspect of the supraspinatus tendon

peaked one month after the surgery. The peak response for the articular aspect, however,

occurred at two months. The vascular activity decreased globally by three months after the

surgery in the supraspinatus tendon. In contrast, the vascularity at the anchor hole which was at

the region of tendon and bone contact showed elevated responses over the period of the study.

Proper blood supply is a pre-requisite for maintaining tendon quality. Hence, blood supply has

becoming an emphasized therapeutic target for preventing degeneration and promoting healing

of the rotator cuff tendon. Understanding the vascular dynamics of the rotator cuff will allow

researchers and physicians to gain deeper insights to the pathophysiology of degenerative

30

changes and potentially lead to novel therapeutic strategies. However, at the present time there

are no therapeutic options available for increasing vascularity in rotator cuff repair.

31

1.3 Rotator Cuff Repair Model

An appropriate rotator cuff model that is representative of the human rotator cuff consists of the

following criteria: the shoulder musculature of the rotator cuff and its surrounding muscles (the

biceps and deltoid muscles), the shoulder bony structures (the acromion, clavicle, coracoids, and

humerus), the articulations of the bony structures, and the motions of the joints.

1.3.1 Animal Models and the Rat Model

Various animal models have been explored in attempt to recapitulate the disease and healing

progression of the rotator cuff. Some of these animal models include the rabbit, the sheep, the

canine, the primate, and the most popularized – the rat.

The rabbit rotator cuff model was employed to examine muscular fatty infiltration and

degenerative changes, tendon-bone enthesis regeneration, and rotator cuff biologic augmentation

68. The sheep rotator cuff was large enough in size to introduce large tears and monitor its effect

and healing progression comparable to the timeline for humans 59

. Additionally, sheep rotator

cuff muscles experience degenerative changes similar to that of chronic tears in humans. This

model allows for temporal molecular analysis of pathological mechanisms after rotator cuff tear

51. The canine rotator cuff model provided a functional assessment of the scaffold augmentations

as the animals place load to the rotator cuff which reflects the stress experience by human

32

conditions 42

. While many of the quadrupedal models have similar shoulder architecture as the

human shoulder, their rotator cuff tendons insert onto the humeral head separately, and therefore

do not form a “true cuff” structure as that of the human rotator cuff 167

.

Non-human primates have the shoulder anatomy and physiology most similar to that of the

human shoulder 166, 204

. The cost, complexity of facility management, and ethical considerations

comprise the reasons for the lack of use of this model.

While the rats are also quadrupeds, their unique shoulder anatomy makes them a suitable animal

model for rotator cuff studies. The rat shoulder shares similar morphology with a human

shoulder, with the supraspinatus tendon passing through the subacromial space 169

. Expanding

from the rat supraspinatus model, an acute surgical repair model has been developed to

specifically examine the healing at the insertion 181

. The model has been extensively used to

examine the various factors in and strategies for tendon healing.

The use of biological approaches to enhance tendon-bone integration has been a popular theme

for the rat supraspinatus acute repair model. The surgery involves detachment of the

supraspinatus from its humeral insertion immediately followed by a bone tunnel suture repair.

Interventions are usually administered during the surgeries: growth factors and bioactive

molecules in healthy or diseased-condition, pharmacological drugs, and cell therapies. Outcome

measurements have included tendon biomechanical properties and histological examinations.

The characterization of early phase of supraspinatus tendon-bone healing in this model allowed

the characterization of the temporal expression profiles of growth factors 203

. Several studies

have investigated the application of fibroblastic growth factor 79, 80

, bone morphogenetic protein-

13 66

, insulin-like growth factor and platelet-derived growth factor-β 192

, and transforming

33

growth factor-β 111

in this model. All but the bone morphogenetic protein-13 study found that the

growth factors were moderately beneficial for tendon healing. However, platelet-rich plasma,

which contains a variety of growth factors that stimulates anabolic activities in tissues, was

found to have no effect on the tendon-bone strength after surgery 14

. Bioactive molecules such as

the recombinant human parathyroid hormone 73

(a hormone that increases osteogenesis and

chrondrogenesis) and alpha-2-macroglobulin 17

(a large generic protease inhibitor) yielded

superior macroscopic healing at the tendon-bone junction, but failed to increase strength,

suggesting the formation of bone and fibrocartilage at the enthesis was not sufficient to translate

into biomechanical improvements.

Diabetic and osteopenic conditions for the rat supraspinatus model have been developed.

Diabetes mellitus significantly compromises healing at the tendon-bone interface, resulting in

lower biomechanical load to failure and inferior tissue characteristics 16

. Although the study by

Cadet and co-workers did not perform surgical detachments of the rotator cuff tendons, estrogen

deficiency induced osteopenia at the greater tuberosity resulted in negative changes at the

insertion histomorphologically 26

. The failure stress, however, did not show statistical

significance.

Pharmacological agents such as bisphosphonate and non-steroidal anti-inflammatory drugs

(NSAIDs) were administered to examine their effects on tendon-bone healing. In the same study,

Cadet and co-workers also used zoledronic acid (bisphosphonate) to improve bone density at the

insertion of the supraspinatus tendon 26

. They found stronger mechanical strength at the tendon-

bone junction with increased bone density at the supraspinatus footprint. It is important,

however, to note that their study did not involve detachment and repair of the supraspinatus

tendon. Further investigation to address the efficacy of bisphosphonate drugs in tendon-bone

34

healing is necessary to delineate the relation. Cohen et al conducted an experiment to examine

the effect of NSAIDs on tendon-bone healing in the rat supraspinatus repair model 34

. Both of the

anti-inflammatory drugs indomethacin and celecoxib compromised soft tissue healing, including

some that failed to heal. The study confirmed the links between tendon-bone healing and the

production of cyclooxygenase-2.

As mentioned in the previous chapter, cell therapy is an attractive therapeutic for regenerative

medicine, especially when the cells themselves are both renewable and differentiable. The

following section will discuss the application of cell therapies to the rat supraspinatus repair

model.

1.3.2 Cell Therapy in the Rat Rotator Cuff Repair Model

Gulotta and colleagues investigated the use of mesenchymal stem cells (MSCs) during rotator

cuff repair to improve tendon-bone healing 63

. MSCs are speculated to differentiate into cells

associated with the tendon-bone interface and augment the tissue quality at the injury site.

However, they found that MSCs showed no benefit over the control groups. Gulotta and

colleagues then conducted a series of three subsequent studies with rotator cuff repair by

genetically modifying the MSCs: MSCs transduced with the membrane type 1 matrix

metalloproteinase gene (MSC-MT1-MMP) 64

, MSCs transduced with the scleraxis gene (MSC-

Scx) 65

, and MSCs transduced with the bone morphogenetic protein-13 gene (MSC-BMP13) 66

.

MSC-MT1-MMP and MSC-Scx demonstrated stronger interfaces, while MSC-BMP13 had no

improvement.

35

It was found that MSCs alone were insufficient to improve the attachment strength of the

supraspinatus tendon at 2 weeks and 4 weeks after repair. The following table provides a

summary of the biomechanical results of these studies.

Table 1-1. Biomechanical results of rotator cuff studies using mesenchymal stem cells

2 Weeks Control63

MSC63-66

MSC-

BMP1366

MSC-MT1-

MMP64

MSC-Scx

65

Load (N) 11.2 ± 2.3 10.5 ± 2.4 11.2 ± 2.5 12 13

Stress (MPa) 1.9 ± 0.6 1.7 ± 0.7 1.6 2.1 2.6 ± 0.9*

Stiffness

(N/mm) 5.7 ± 2.3 4.9 ± 1.8 5.1 6 8.4 ± 2.9*

4 Weeks Control63

MSC63-66

MSC-

BMP1366

MSC-MT1-

MMP64

MSC-Scx

65

Load (N) 22.1 ± 3.5 20.8 ± 4.4 20.8 ± 4.4 27.3 ± 4.2* 26.7 ± 4.6*

Stress (MPa) 3.5 ± 1.0 3.5 ± 1.0 3.9 5.0 ± 1.3* 4.7 ± 1.3*

Stiffness

(N/mm) 9.8 ± 4.7 9.3 ± 3.3 8.2 13.3 ± 5.1* 15.3 ± 3.4*

* significant compared to the control group/MSC group (MSC group was the baseline value for

subsequent genetic modifications)

36

0

5

10

15

20

25

30

35Fo

rce

Tendon Strength

2 Weeks

4 Weeks

*F

orc

e (

N)

Figure 1-4. The ultimate load to failure at the insertion. The mesenchymal stem cells transduced

with MT1-MMP gene and Scx gene increase insertion strength at 4 weeks after the repair

compared to the control group/MSC group (MSC group was the baseline value for subsequent

genetic modifications). Mesenchymal stem cells alone were not sufficient to improve healing.

Histologically, the MSCs were tracked using Ad-LacZ transduction prior to transplantation.

Their presence was evaluated by the activity of β-galactosidase post-mortem. Exposure to β-

galactosidase turned X-Gal solution from colourless to blue, indicating the presence of

metabolically active MSC-LacZ. The amount of blue was significantly more in the specimens

treated with MSC-LacZ than MSC-Null, and the β-galactosidase activity was lower at 4 weeks

comparing to that at 2 weeks.

37

At 4 weeks, the tendon-bone interface tissue was more robust than those at 2 weeks even under

gross observation. The continuity between the repaired tendon and the bone was observed and

the natural healing progression without intervention was noted.

The collagen fibre organization and the area of fibrocartilaginous tissue were quantified using

birefringence represented in gray scale and Safranin O/Fast Green FCF staining respectively.

The following table provides a summary of the histological results from these studies.

38

Table 1-2. Histological results of rotator cuff studies using mesenchymal stem cells

2 Weeks Control63

MSC63-66

MSC-

BMP1366

MSC-MT1-

MMP64

MSC-Scx

65

Organization

(gray scale) 16.5 ± 5.5 20.3 ± 5.9 20.2 ± 3.6 24.4 ± 4.4 22.1 ± 3.5

Fibrocartilage

(mm2)

0.342 ± 0.201

0.596 ± 0.192 0.603 ± 0.294 0.514 ± 0.305 0.528 ± 0.407

4 Weeks Control63

MSC63-66

MSC-

BMP1366

MSC-MT1-

MMP64

MSC-Scx

65

Organization

(gray scale) 24.8 ± 2.5 26.5 ± 5.5 26.5 ± 5.4 34.5 ± 3.5* 25.9 ± 6.3

Fibrocartilage

(mm2)

0.438 ± 0.115 0.343 ± 0.217 0.451 ± 0.215 1.047 ±

0.384*

0.729 ±

0.050*

* significant compared to the control group/MSC group (MSC group was the baseline value for

subsequent genetic modifications)

The mesenchymal stem cells alone did not statistically improve the biomechanical strength of the

repaired tendons. The hypoxic environment leading to compromised cell activities was likely a

factor for the result. Alternative cell types were considered and it was hypothesized that the issue

of poor vascularity should be addressed.

39

1.4 Angiogenesis and Endothelial Progenitor Cells

1.4.1 Capillaries

The microvasculature of the body is the site of interchange between the circulation and the body

tissues. These specialized vessels are called capillaries. The total diameter of the capillaries in a

human body is 800 times longer than that of the aorta, and their total surface area is roughly

6000 metres squared. Accordingly, the flow speed in the capillaries is 0.3 millimetres per second

whereas the average flow speed is 320 millimetres per second in the aorta. Their dimensions are

7 to 9 micrometres in diameter and 0.25 to 1 millimetres in length. Structural variations of

capillaries in different tissues reflect a variety of complex metabolic demand and function of the

associated tissues and organs.

The general structure of capillaries is composed of a layer of endothelium and a basement

membrane. The endothelium consists of a single layer of endothelial cells of mesenchymal

origin. Endothelial cells are polygonal in shape and elongated in the direction of blood flow.

They rest on basal lamina, are held together by zonula occludens (tight junctions), and are in

contact with neighbours through gap junctions. On the luminal membrane of endothelial cells is

surface proteins which are immunogenic and coagulating modulators. Present along some

locations surrounding the capillary endothelial cells are pericytes. These mesenchymal cells

possess the capacity to transform into cells associated with blood vessels.

40

The structure of the capillaries dictates the function of the capillaries: the macromolecule

permeability, the metabolic activities, and the anti-thrombogenic role. Accordingly, the various

types of capillaries include the continuous (somatic) capillary, the fenestrated (visceral)

capillary, renal fenestrated capillary, and the discontinuous sinusoidal capillary. Continuous

capillary is the major capillary type in muscles, connective tissues, and nervous tissues. The

prominent characteristic of the continuous capillary is the absence of fenestrae in the

endothelium. Here, exchanges between the blood and tissues rely heavily on pinocytotic vesicles

in the endothelial cells. Fenestrated, or visceral, capillaries are often seen in organs that allow

rapid interchange of substances such as in kidneys and intestinal organs. From 60 to 80

nanometres in diameter, fenestrations closed by a thin diaphragm in the endothelium enable

small molecules to cross freely. The renal fenestrated capillary is a specialized version of

fenestrated capillary, which is encapsulated by the glomerulus. Lastly, the discontinuous

sinusoidal capillary is marked by tortuous path, fenestration lacking diaphragms, enlarged lumen

from 30 to 40 micrometres in diameter, and presence of macrophages along the endothelium.

These capillaries have discontinued endothelium and basement membrane that allow

mobilization of cells and higher levels material exchange, especially mid-sized plasma proteins

such as albumin. Discontinued sinusoidal capillaries are present in the liver and hematopoietic

tissues and organs such as the bone marrow and the spleen.

The second component of the capillary, the basement membrane, is a connective tissue which is

by large composed of the extra-cellular matrix. The membrane provides structural integrity, acts

as a filtration barrier, and regulates cell attachment and chemotaxis. The membrane is rich in

collagen fibrils: type I and type III collagen.

41

1.4.2 Angiogenesis

The growth of vessels via pre-existing capillaries is termed angiogenesis. It is a complex process

that involves proliferation, migration, and remodeling of endothelial cells into functional

vasculature, and the tight balances between pro-angiogenic and angiostatic molecules.

Angiogenesis is regulated by cytokines and chemokines, the extracellular matrix as well as

physical forces of the vascular environment.

The most potent angiogenic cytokines are the vascular endothelial growth factors (VEGFs).

VEGFs can induce vasodilation, increased permeability of and fenestration in the endothelium,

activation of endothelial cells to a proliferative state 190

. On the surfaces of endothelial cells are

VEGF tyrosine kinase receptors, specifically, the VEGF receptor-2 which binds to the VEGF-A

isoform. Some of the major bioactive molecules involved in angiogenesis include fibroblastic

growth factors, transforming growth factor-β, insulin-like growth factor-1, platelet-derived

growth factors, angiopoeitin, erythropoietin, and tumor necrosis factor-α. Each drives the

angiogenic activity through various signaling pathways and downstream effects. The actions of

the molecules are not independent of each other, but work in concert to facilitate microvascular

expansion.

Successful migration and invasion of endothelial cells into the perivascular tissue is dependent

on the proteolysis and remodeling of the extracellular matrix. The extracellular matrix offers

molecular directional cues for angiogenesis as well as binding sites that provides spatial

information to the cells 190

. Present on the cell surface are transmembrane protein integrins which

modulate the motility and affinity of the cells through the extracellular matrices.

42

Lastly, hemodynamic forces in the vessels can affect the angiogenesis and subsequent

remodeling processes. Shear stress is essential for the maintenance of endothelial integrity and

responses; forms of shear stress can induce different angiogenic responses via distinct signaling

pathways 109

. Local inflammatory response alters the hemodynamics of the tissues, and the

capillary network remodels to accommodate the environment. It is observed that once the

inflammation subsides, the network remodels again to its original hemostatic architecture.

Mechanisms of Angiogenesis

Once the endothelial cells have been stimulated for angiogenesis, a cascade of events follows as

the vessels prepare for neovascularization. There are two identified physiological angiogenic

processes: sprouting and intussusceptive angiogenesis. Both types of microvascular growth have

been observed in vivo.

Sprouting angiogenesis, as the name suggests, involves paracrine induced migration of a tip cell

supported by lateral stalks cells into the perivascular tissue. Upon activation of the tip cell by

angiogenic stimulation, the adjacent basement membrane is degraded by proteases. The direction

of sprouting is guided toward higher gradient of growth factors until cellular contact is made

with other capillary sprouts. Trailing the tip cell, the stalk cells form the lumen of the vessel and

line the endothelium. Pericytes are recruited as the vessel architecture undergoes remodeling and

maturation. The tip and stalk cell phenotypes are transient and highly regulated by notch

signaling; the process of angiogenesis requires dynamic transitions between the endothelial, tip,

and stalk cell phenotypes 138

.

43

The other process of angiogenesis, intussusceptive angiogenesis, occurs via splitting the pre-

existing vascular network or capillary plexus. The characteristic hallmark of intussusceptive

angiogenesis is the formation of multiple intraluminal pillars from invaginations of opposing

vessel walls in its initial stages 38

. When the inter-endothelial junctions are re-organized, the core

of each pillar is perforated to allow mural cell invasion and growth factor trafficking. As the

pillar expands with increasing extracellular matrix deposition, it fuses with nearby pillars to

create bifurcation in a vessel.

Both sprouting and intussusceptive angiogenesis rely on pre-existing vasculature. While

intussusceptive angiogenesis is more limited in a sense that it can only occur in capillary

plexuses developed from vasculogenesis or sprouting angiogenesis, its process is much faster

and energy efficient. Capillary splitting does not require the disruption of basement membrane or

extensive proliferation of endothelial cells to line the endothelium. Microvascular growth studies

have identified the predominance of capillary sprouting in the early stages of vascular growth,

followed by extensive hierarchical splitting of vessel during remodeling 106

. The novel evidence

suggests the temporal dependence of the angiogenic mechanisms.

It is plausible that during tendon healing in which metabolic resource is scarce, there is a

tendency for the fibrovascular tissue to derive vessels via intussusceptive angiogenesis.

However, the tendon is hypovascular, leaving the bone and the bursa as the major sources for

pre-existing vasculature. The hypovascular nature of the tendon itself may impair its ability to

heal.

44

1.4.3 Endothelial Progenitor Cells

Cell therapy is an attractive therapeutic avenue for regenerative medicine, especially when the

cells themselves have the capacity to renew and differentiate. Stem cells and progenitor cells are

populations of cells which possess such capabilities. Much excitement was generated when

endothelial progenitor cells were identified. The first published detailed description for

endothelial progenitor cell isolation is by Asahara and colleagues in 1997 6. It sparked a new

paradigm in utilizing endothelial progenitor cells as a cellular strategy for therapeutic

neovascularization. It is believed that these cells can induce angiogenesis via the secretion of

angiogenic factors or by directly participating in the lining of endothelium. Endothelial

progenitor cells are a rare but normal component of the elements that make up the circulating

blood cells.

There are a several sources from which these cells can be isolated. These include the peripheral

blood 6, 77, 99

, the bone marrow 140

, and the umbilical cord blood 22, 123, 134

In addition, endothelial

progenitor cells have been found to reside in tissues and organs such as fat 134, 137

and vascular

tissues 150

, the fetal liver 134

, and the heart19

.

Currently, there is not a well-defined set of surface immune-phenotyping or functional assays to

identify putative populations of endothelial progenitor cells. Various combinations of cell surface

antigens or markers were used in attempt to characterize the cells. Most of these markers are not

endothelial lineage specific; the marker profile reflects the hematopoietic lineage 189

.

Nonetheless, the conventional positive markers for detection are CD 34, CD 133, and vascular

endothelial growth factor receptor-2 (VEGFR2) or fetal liver kinase-1 (Flk-1) 75, 189

. CD 34

45

positivity indicates that the putative population is of the hematopoietic stem/progenitor cell

population. The second marker, CD 133, also known as Prominin-1 or AC 133, is a strict

character for early hematopoietic stem/progenitor cell that reside in the bone marrow, fetal liver,

or in circulation 207

. Lastly, the presence of vascular endothelial growth factor receptor-2 on the

membrane warrants the ability for the cells to respond to vascular endothelial growth factor, a

potent signaling molecule for angiogenesis.

46

Endothelial Progenitor Cells

CD133

CD34

Endothelial progenitor cell

Endothelial cells

Paracrineeffect

Flk-1

Figure 1-5. Surface antigens and possible angiogenic mechanisms of endothelial progenitor cells.

The conventional surface markers are CD 34, CD 133, and Flk-1 (VEGFR2). The mechanisms

for angiogenesis are endothelial differentiation and angiogenic signaling.

In addition to surface immune-phenotyping, assays are often performed to demonstrate

functional characteristics of putative cells of the endothelial lineage. The uptake of acetylated-

low density lipoprotein (AcLDL) labeled with fluorescent probe 1,1' -dioctadecyl -3,3,3',3' -

tetramethyl-indocarbocyanine perchlorate (DiI-AcLDL), the binding of lectin Ulex europaeus 1

agglutinin (UEA-1), and forming tubes in 3 dimensional matrices are frequently used in vitro

behaviours 35, 74, 189

. In contrast to the surface markers which correlate more to that of the

hematopoietic stem cell phenotypes, the functional assays are indicative of endothelial

properties. AcLDL uptake occurs through the scavenger cell pathway of lipid metabolism. While

47

internalizing AcLDL is a property of endothelial cells, monocytes and macrophages are also

know to scavenge AcLDL macromolecules. Likewise, UEA-1 binds to glycoproteins that are

found predominately on the endothelial cell surface, but are present on epithelial cells and other

blood-derived cells. Accordingly, either of the functional assays alone is sufficient to distinguish

endothelial progenitor cells from other hematopoietic cells. In vitro tubulogenesis assay is

considered to be a very informative test; other cells do not typically form tubes in fibrous

matrices. Endothelial progenitor cells, when in contact with extracellular matrix components,

have the capacity to assemble and form a network of tubes that resembles the microvasculature

in vivo 74

.

Under the conventional isolation protocols, two distinct populations of endothelial progenitor

cells have been identified. Early and late endothelial progenitor cells, named according to their

time-dependent appearance in vitro, share common endothelial surface phenotypes but display

different functional potentials 77, 208

. Early endothelial progenitor cells have a shorter life span

(about 4 weeks) and higher levels of cytokine secretion. They are, however, poor in forming

tubes in vitro. Late endothelial progenitor cells, also known as late outgrowth cells, have better

endothelial incorporating and tubulogenic capacities; they are representative of mature

endothelial cells yet still maintain proliferative angiogenic characteristics. The contrasting

angiogenic profiles suggest the different roles during neovascularization 119

. The paracrine and

autocrine effects of the early cells augment the proliferation, migration, and incorporation of the

late cells. This synergistic interaction of the mixed populations has been demonstrated in vivo.

The transplantation of both types of progenitor cells together led to better perfusion than either of

the cells alone 208

.

48

The exact origin of endothelial progenitor cells is still uncertain. The cells are believed to be

derived from the bone marrow, as observed in gender-mismatched marrow transplant patients 99

.

There are two proposed sources, the mesenchymal and the hematopoietic differentiations. The

cells from the mesenchymal lineage lacked the early hematopoietic markers and were able to

differentiate in vitro into osteocytes and adipocytes 128, 143

. When they were cultured for

endothelial differentiation, the cells expressed endothelial antigens and formed vascular tubes

when plated on extracellular matrix 5, 128, 143

. In addition to the mesenchymal origin, the

hematopoietic origin has been indicated as an alternative source for endothelial progenitor cells.

A wealth of literature has described the hematopoietic stem cell-myeloid progenitor-monocyte

pathway for endothelial precursor differentiation 48, 139, 156

. Surface markers and angiogenic

assays confirmed the plasticity of monocytes to give rise to vasculogenic cells 135

. Myeloid

progenitors are also found to contribute to endothelial progenitor cell differentiation and

contribute to the vascular endothelium 11, 103

. The close association between the two cell lineages

in the bone marrow could explain the phenotypic overlap of their progenies. A recent study

highlighted the trans-differentiating potential from the mesenchymal lineage precursor

(mesenchymoangioblast) to the hematopoietic lineage precursor (hemangioblast) given the

appropriate stimulation 164

.

49

1.4.4 Applications of Endothelial Progenitor Cells

Bone marrow transplant experiment showed that endothelial progenitor cells have 0.2 to 0.4 %

engraftment rates in the vascular endothelium after four months 36

. Tissue specific vascular

progenitor cells may result in different cell responses and fates during neovascularization 108

.

Various animal models have been used for pre-clinical investigations of endothelial progenitor

cell therapy in ischemic conditions. Increased micro-vasculature in the myocardium, improved

myocardial remodeling and function, and reduced scarring were observed with the addition of

endothelial progenitor cells to ischemic heart condition 84, 85, 91, 157

. Similarly, anti-fibrogenesis,

remodeling, and hepatic regeneration were associated with the transplantation of endothelial

progenitor cells to the cirrhotic liver model 120, 175, 191

. The authors speculated that the secretion

of angiogenic growth factors and enhanced vascularization lead to the improvement in liver

function and animal survival. Additionally, extensive studies have examined the effects of

delivering endothelial progenitor cells in ischemic hind limb models. Capillary density and limb

blood flow increased, and the rate of limb loss was reduced 83, 155, 193

.

The pre-clinical animal studies done using endothelial progenitor cells have revolved around the

theme of re-vascularizing injured tissues to enhance the tissue’s innate reparative ability.

Increased vascular density and blood supply to target tissues along with ameliorated tissue

quality were the results. Indeed, the endothelial progenitor cell population was expected to

provide a source of angiogenic growth factors that stimulates endothelial proliferation. While

these studies have confirmed the angiogenic effects of introducing endothelial progenitor cells to

50

these tissues, more studies are needed to delineate the specific mechanisms by which the cells

promote neovascularization.

i. Clinical trials

Vascular progenitor cells have been investigated in clinical trials for critical limb ischemia and

heart diseases. Randomized controlled trials have been conducted to evaluate angiogenic cell

therapy in critical limb ischemia. Two sources of mononuclear cells were used in the studies:

bone marrow mononuclear cells 176

and peripheral blood mononuclear cells 76

. Both trials

demonstrated feasibility and safety of treatment.

Clinical trials for angiogenic cell therapy in heart diseases have been divided into two categories

165: acute myocardial infarction and chronic heart disease. Major trials for acute myocardial

infarction include TOPCARE-AMI 8, 152

, BOOST 114, 115, 201

, ASTAMI 18, 105

, and REPAIR-AMI

49, 153. On the other hand, for chronic ischemic heart diseases, there are TOPCARE-CHD

7 and

FOCUS-CCTRN 18, 105, 136

. All trials but ASTAMI and FOCUS-CCTRN showed positive

functional outcomes for the heart.

Two trials, TOPCARE-AMI and TOPCARE-CHD, employed both the bone marrow and the

peripheral blood endothelial progenitor cells, and both found favourable improvements in the left

ventricular function. Interestingly, the TOPCARE-CHD study compared circulating progenitor

cells with bone marrow progenitor cells and found better outcome in left ventricular ejection

fraction with the bone marrow progenitor cells 7. The long term results of TOPCARE-AMI

demonstrated safety and efficacy with progenitor cell restored cardiac function 95

.

51

ii. Orthopaedic applications

Vascular strategies have received much attention in the field of orthopaedics recently. The use of

vascular progenitor cells to promote neovascularization in tissues with poor or deficient

vascularity has been a major focus of recent investigations. This includes osteonecrosis of the

femoral head, segmental bone defects, and transected ligament.

While the etiology of osteonecrosis is multifaceted and not yet fully understood, circulatory

impediment is a widely accepted theory of pathogenesis 82

. The addition of endothelial

progenitor cells has shown beneficial effects on osteonecrosis of the femoral head in animal

models 173, 179

. Both osteogenic and angiogenic responses have been reported. Clinically, bone

marrow mononuclear cells containing endothelial progenitor cells have been continuously

investigated over the years 56, 196, 206

. While endothelial progenitor cell transplantations have

showed promises, more studies are required to ensure efficacy and safety.

Endothelial progenitor cell therapy has demonstrated great potential to heal segmental bone

defects in the rat femur. The addition of endothelial progenitor cells aided the healing of the bone

defect, while the control group without these cells did not heal 9, 98

. A subsequent study showed

that endothelial progenitor cell therapy enhanced expression of various VEGF isoforms in a bone

defect model 97

. The findings of these studies suggested that the endothelial progenitor cells

encouraged early osteogenesis via the facilitation of angiogenesis.

Ligaments are known for their hypovascularity under normal physiological conditions. Healing

of the medial collateral ligament was augmented by administering an endothelial progenitor cell-

enriched population 178

. The local transplantation of the CD 34-positive cells aided ligament

52

healing by increasing angiogenic activities and improving macroscopic, histological, and

biomechanical results. Real-time reverse transcription polymerase chain reaction also revealed

higher expressions of ligament-specific markers, collagen1A2 and tenomodulin, in the treatment

group. The addition of CD 34-positive cells could alter the local environment to one that is

favourable for ligament healing.

Gomes and colleagues investigated the supplement of bone marrow mononuclear cells during

conventional transosseous rotator cuff repair in a clinical trial 46

. The idea of using these

mononuclear cells was that they could revitalize tendons that have undergone avascular

degeneration after the repair. The patients demonstrated functional improvements and tendon

integrity after a twelve month follow up. The magnetic resonance imaging (MRI) results also

suggested the formation of new tissue at the footprint in some of the patients. Although the

preliminary results indicated positive clinical outcomes and MRI findings, the study did not

include a control group and were unable to provide definitive evidence for tissue quality.

To our knowledge, the use of endothelial progenitor cells to augment rotator cuff repair has not

been investigated. In vivo studies are necessary to assess the effectiveness of endothelial

progenitor cell transplantation during rotator cuff repair.

53

1.5 Rationale, Hypothesis, and Objectives

1.5.1 Rationale

Biological strategies for the augmentation of rotator cuff tendon repair have been a popular topic

of pre-clinical and clinical investigations, many of which employ growth factor supplementation

33, 172, extra-cellular matrices

146, 160, and/or anabolic progenitor cells

63, 64. Due to their absence of

immunogenic responses, autologous cell-based therapies have received much attention. With

recent investigations of mesenchymal stem cells (MSCs) strategies yielding disappointing

results63

, potentially due to impaired vascularization at the site of tendon-bone healing3, 78

,

alternative cell lines warrant consideration.

Endothelial progenitor cells are a population of heterogeneous cells that can be isolated from the

bone marrow and the peripheral blood. Endothelial progenitor cells are believed to promote

angiogenic activities through secretion of angiogenic factors (paracrine) and differentiation into

vascular endothelial cells 74

. This cell population has become a popular vehicle for

supplementation of angiogenic growth factors. Studies in animal models of ischemia have shown

their efficacy to induce vascular growth in tissues.

Recent orthopaedic application of endothelial progenitor cells showed superior bone healing

compared to using mesenchymal stem cells9, 124

. The endothelial progenitor cells increased

vascularity in the animals (on the basis of Laser Doppler) and demonstrated superior healing of

54

bone. Given the relevant literature, it stands to reason that the addition of endothelial progenitor

cells at the tendon-bone reattachment site in rotator cuff repair could enhance the vascular

response and improve healing. Endothelial progenitor cell therapy could have a therapeutic effect

on the hypovascular condition of the tendon and aid osteogenesis on the bone end to improve

tendon-bone integration. However, the use of endothelial progenitor cells for the biological

augmentation of rotator cuff repair has not previously been investigated. Our study aimed to

investigate the augmentation rotator cuff repair and vascularity with the therapeutic application

of endothelial progenitor cells.

1.5.2 Hypothesis

Our hypothesis was that the application of endothelial progenitor cells during rotator cuff repair

would increase the biomechanical strength at the tendon-bone interface and increase

angiogenesis during the early phase of rotator cuff healing.

1.5.3 Objectives

It was postulated that by promoting angiogenesis at the tendon-bone site, the endothelial

progenitor cell-treated subjects would yield better tendon repair in a rotator cuff model. The

study addressed the efficacy of endothelial progenitor cells to augment surgical rotator cuff

55

repair and also sought to investigate the fate of the transplanted endothelial progenitor cells. Our

first objective was to confirm the endothelial progenitor cell population from our isolation using

flow cytometry and functional assays. Our second objective was to assess the effect of

endothelial progenitor cell transplantation on the biomechanical strength and histology of the

healing rotator cuff tendon after surgical repair in a rat model. Our third objective was to

evaluate the effect of endothelial progenitor cell transplantation on the vascularity at the repair

site using immunohistochemistry. Our final objective was to examine the fate of the transplanted

endothelial progenitor cells using genetically labeled cells.

56

2 Methods

2.1 Study Design

The study aimed to complete three objectives: evaluate the effect of endothelial progenitor cell

transplantation on the strength and quality of healing tissue, quantify the angiogenic response,

and monitor cell survival at the injury site. Accordingly, the study was designed to address these

objectives.

All animals(80)

EPC(32)

Biomechanics (24)

Week 2(4)

Week 4(10)

Week 6(10)

Histology(8)

Week 4(4)

Week 6(4)

Fibrin(32)

Biomechanics (24)

Week 2(4)

Week 4(10)

Week 6(10)

Histology(8)

Week 4(4)

Week 6(4)

Tracking(16)

GFP-EPC(8)

Day 3 (2)

Week 1(2)

Week 2(2)

Week 4(2)

EPC(8)

Day 3 (2)

Week 1(2)

Week 2(2)

Week 4(2)

Figure 2-1. The study design.

57

A population of syngeneic Fischer 344 rats was used to allow the transplantation of endothelial

progenitor cells between specimens. They were mature healthy rats between 250 to 300 grams

when ordered which roughly translates to three months of age. The animals were housed in pairs

upon arrival from Charles River, and were allowed to acclimatize for at least two days prior to

their surgery. All rats were allowed normal weight bearing and cage activities, standard diet and

water ad libitum. The study was approved by the Animal Care Committee and the Research

Vivarium, St. Michael’s Hospital.

The rats were divided into three groups. The first group (experimental group) received

endothelial progenitor cells delivered in fibrin sealant during the surgical repair of the

supraspinatus tendon. The second group (control group) received only the fibrin sealant. These

two groups were allocated for biomechanical testing and histomorphometric analysis to answer

the first two study questions. The third group, which was used to track cells, received endothelial

progenitor cells which expressed green fluorescent protein delivered in fibrin sealant. This group

was dedicated only for histology examination for the presence of the fluorescent protein.

58

2.1.1 Statistical Significance of the Study

The primary outcome measure for the study was biomechanical strength of the rotator cuff

repair, specifically the ultimate load to failure at 4 weeks. Previous studies have set a 20 %

increase in biomechanical strength as the threshold for significance, and this was felt to be a

clinically significant difference. Accordingly, the statistical significance for this study was set at

20 % increase in biomechanical strength at 95 % confidence (α = 0.05) and 80 % power (β =

0.2).

59

2.1.2 Sample Size Calculation

The calculation for sample size with respect to estimated standard deviation and expected

differences was the following:

where α = the confidence interval; β = power; σ = the standard deviation; and δ = the expected

difference.

The pre-determined level of significance is 95 % with a power at 80 % gives

The same size variables were derived from previous data with regards to supraspinatus tendon

strength at 4 weeks 63

.

Therefore, the determined sample size for each group (the experimental and fibrin groups) was

ten.

60

2.2 Cell Isolation

2.2.1 Endothelial progenitor cell isolation

The endothelial progenitor cell isolation was performed following the developed protocol in the

Musculoskeletal Research Laboratory as previously described 9, 124

.

Syngeneic Fischer 344 rats (Charles River Laboratories, Wilmington, Massachusetts, USA) were

used for cell isolation to allow for transplantation without immunogenic complications.

Endothelial progenitor cells were isolated from rat long bone marrow. Rats were anaesthetized

with isoflurane and euthanized via cervical dislocation. Bilateral femurs and tibiae were carefully

disarticulated and soft tissues were sharply removed post-mortem under sterile conditions.

Dissected femurs and tibiae were immediately immersed in sterile phosphate buffered solution

(Dulbecco’s Phosphate-Buffered Saline, Gibco, Carlsbad, California, USA) and transferred to a

culture hood.

Bone marrow cells were obtained by flushing the intramedullary canals of the long bones with

PBS. The collected solution was centrifuged at 220 relative centrifugal force (RCF) for 5

minutes at room temperature. The supernatant was discarded and the pellet was suspended in

Endothelial Growth Medium 2-MV (EGM2-MV, Gibco, Carlsbad, California, USA), and

transferred into a tube containing Ficoll (Ficoll-Paque Plus, GE Healthcare, Little

Chalfont, United Kingdom) without disturbing the interface between the two solutions. The tube

was centrifuged at 400 RCF, 30 minutes, room temperature. The middle buffy layer containing

61

mononuclear cells was extracted, rinsed in EGM-2-MV, and centrifuged at 100 RCF for 10

minutes, room temperature. The pellet was re-suspended and rinsed once more, and centrifuged

at 100 RCF for 10 minutes, room temperature.

The mononuclear cells were re-suspended in EGM-2-MV and seeded on to human fibronectin-

coated (Discovery Labware, Billerica, Massachusetts, USA) culture flasks and incubated at 37°C

for four days to allow selection of endothelial progenitor cells via fibronectin adhesion and

endothelial cell-specific medium before washing non-adhering cells. Medium was changed every

two days after the initial four days. The cells were cultured for 9 to 10 days before

transplantation, by which time they reach 80 % confluence.

2.2.2 Flow cytometry

Once the endothelial progenitors cells were cultured in the flask for 9 to 10 days, they were

trypsinized (Trypsin-EDTA 1x, Mediatech, Manasass, Virginia, USA) for 5 minutes, harvested,

and rinsed. They were then transferred into microcentrifuge tubes (200 000 cells/tube). The cells

were pelleted and the supernatant was removed. A volume of 100 microlitres of 1 % bovine

serum albumin (Sigma-Aldrich, St Louis, Missouri, USA) in phosphate buffered saline

(BSA/PBS) was used to suspend the cells for each tube. The four antibodies were added

accordingly. The tubes were incubated at 4°C in the dark for 20 minutes to allow antibodies to

bind. The cell suspensions were transferred to flow tubes for analysis.

62

The cells were analyzed with MACSQuant flow cytometer. The event readings were selected for

single cells only, and their fluorescence was gated using the fluorescent minus one (FMO)

groups and the single stains. The surface antigens chosen for flow cytometry were CD 31, CD

34, CD 133, and Flk-1. The excitation formats for CD 31, CD 34, CD 133, and Flk-1 were

fluorescein isothiocyanate (FITC), AlexaFluor 647, phycoerythrin (PE), and VioBlue (v450)

respectively.

i. CD 31

Platelet endothelial cell adhesion molecule 1 (PECAM-1), designated CD 31, is a common

endothelial marker. It is a type I integral membrane glycoprotein (130 kDa) and a member of the

immunoglobulin family that is expressed at high concentrations on the surfaces and between cell

junctions. CD 31 mediates homophilic and heterophilic cell-cell adhesion. CD 31 mediated

endothelial cell-cell interactions are involved in angiogenesis in rats and mice. Accordingly, CD

31 antibodies are often used to measure angiogenesis in association with physiological processes

such as tumor growth. However, in addition to endothelial cells, CD 31 is also expressed on

platelets and subsets of leukocytes.

ii. CD 34

CD 34 is a type I transmembrane glycoprotein (110-120 kDa) of the CD 34/Podocalyxin family

of sialomucin proteins. It is recognized as a highly glycosylated hematopoietic stem/ progenitor

cell antigen. The surface glycoprotein functions as a cell-cell adhesion factor, which mediates the

63

attachment of stems cells to the bone marrow extracellular matrix or directly to stromal cells. CD

34 is also expressed on the endothelium of capillaries.

iii. CD 133

CD 133 (117 kDa), or prominin 1 (PROM-1), is a 5-transmembrane glycoprotein specifically

localized to cellular protrusions. Originally known as AC 133, CD 133 expression is restricted to

a subset of CD3 4 stem/ progenitor cells in the hematopoietic system in liver, bone marrow, cord

blood, and peripheral blood. Only a small portion of CD 34- CD 133

+ cells are found in these

tissues. CD 133 is present on circulating endothelial progenitor cells, fetal neural stem cells, and

other tissue specific stem cells.

iv. Fetal liver kinase-1/Vascular endothelial growth factor receptor-2

Fetal liver kinase-1 (Flk-1), also known as vascular endothelial growth factor receptor-2

(VEGFR-2), is a receptor protein tyrosine kinase of the immunoglobulin family. It is closely

related to CD 177 (c-kit) and CD 140a (PDGF receptor α chain). As the name VEGFR-2

suggests, Flk-1 is a receptor for vascular endothelial growth factor (VEGF). It is expressed on

endothelial cells, and in embryonic and adult tissues. In vivo and in vitro studies indicate that

Flk-1 is required for embryonic development of hematopoietic and vascular endothelial cells as

VEGF is an important signaling protein involved in vasculogenesis and angiogenesis.

64

Table 2-1. Staining Protocol for flow cytometry

Antibody CD 31 CD 34 CD 133 Flk-1

Manufacturer AbSeroTech BD Horizon Miltenyi Biotech BD Horizon

Stock concentration 0.1 mg/mL 0.5 mg/mL 0.022 mg/mL 0.2 mg/mL

Format FITC AlexaFluor 647 PE VioBlue

Aliquoted stock per tube (µL) 5 4 20 10

Aliquoted BSA/PBS per tube (µL) 15 16 0 10

Aliquoted volume (µL) 20 20 20 20

Total stock prepared (µL) 30 24 100 60

Added BSA/PBS prepared (µL) 90 96 0 60

Total diluted volume (µL) 120 120 100 120

Added Aliquoted Volume in µL

Stains Tube number Antibodies CD 31 CD 34 CD 133 Flk-1 BSA/PBS Total

Control 1 - 0 0 0 0 100 100

Single

2 CD 31 20 0 0 0 80 100

3 CD 34 0 20 0 0 80 100

4 CD 133 0 0 20 0 80 100

5 Flk-1 0 0 0 20 80 100

FMO

6 CD 31, 34, 133 20 20 20 0 40 100

7 CD 31, 34, Flk-1 20 20 0 20 40 100

8 CD 31, 133, Flk-1 20 0 20 20 40 100

9 CD 34, 133, Flk-1 0 20 20 20 40 100

Quadruple 10 CD 31, 34, 133, Flk-1 20 20 20 20 20 100

Total 100 100 100 100

65

2.2.3 Functional assays

Functional assays were performed to evaluate the putative endothelial progenitor cell

populations. Both AcLDL uptake and lectin binding assays were carried out on the same slides.

i. AcLDL uptake and lectin binding assays

The isolated endothelial progenitor cell population was evaluated for the uptake of acetylated-

low density lipoprotein labeled with fluorescent probe 1,1' -dioctadecyl -3,3,3',3' -tetramethyl-

indocarbocyanine perchlorate (DiI-AcLDL, Invitrogen, Carlsbad, California, USA) and the

binding of lectin GS II (Lectin GS II Alexa Fluor 488 Conjugate, Invitrogen, Carlsbad,

California, USA). Two-well chamber slides were coated with human fibronectin (10 µg/mL) and

cells were seeded at a density of 8000 cells per square centimetre with endothelial growth

medium at 37°C overnight.

The next morning, stock DiI-AcLDL was diluted to 10 µg/mL with culture medium, and 500 µL

of the solution was added to the appropriate chambers. The control chambers received 500 µL of

endothelial growth medium. They were all incubated at 37°C for four hours followed by three

phosphate buffered saline (PBS) rinses. The cells were then fixed with 2 % paraformaldehyde for

ten minutes, and washed three times with PBS.

The lectin molecule, GS II, was used for lectin binding. Stock GS II was diluted to 20 µg/mL

using PBS with calcium and magnesium (Dulbecco’s Phosphate-Buffered Saline with calcium

and magnesium, Gibco, Carlsbad, California, USA). The solution was added to the appropriate

66

chambers (500 µL/well) and the control chambers received only PBS, and were kept at 4°C

overnight on a shaker. The next morning, the chambers were washed three times with PBS. The

chambers were disassembled and the slide was allowed to air dry in the dark. 4', 6-diamidino-2-

phenylindole in permount (ProLong Gold with DAPI, Invitrogen, Carlsbad, California, USA)

was added to each slide (80 µL), and cover slips were placed on. When the slides were imaged

under fluorescent microscopy, the DiI-AcLDL appeared red, the GS II appeared green, and the

nucleus appeared blue.

ii. In vitro tube formation assay

The isolated population of endothelial progenitor cells was plated on basement membrane

Matrigel (BD Sciences, Bedford, Massachusetts, USA) for tube formation assay. Stock Matrigel

was diluted 1:1 with sterile cold PBS. Diluted Matrigel was aliquoted into a 24-well plate (300

µL/well) and was incubated for 30 minutes to allow polymerization. Endothelial progenitor cells

were seeded at 48 000 cells per well (cell density of approximately 25 000 cells/cm2) with 500

µL of endothelial growth medium. They were incubated for 24 hours and imaged with a

microscope. Mesenchymal stem cells were used as the negative control. The number of tubes

formed was counted and the total length was calculated.

67

2.2.4 Cell survival in fibrin sealant

Fibrin sealant has been widely used as cell delivery agent, namely with mesenchymal stem cells.

Endothelial progenitor cells, however, have yet to be delivered via this method. To ensure

endothelial progenitor cell survival in this delivery agent, the fibrin sealant, the following

procedure was performed.

Commercial fibrin sealant (Evicel, Johnson & Johnson, Somervile, New Jersey, USA) is

composed of two ingredients: (1) human clottable protein (often termed fibrinogen), and (2)

human thrombin. Human clottable protein is activated upon exposure to thrombin in the presence

of calcium ions, causing the mixture to clot.

The endothelial progenitor cells were trypsinized and counted. One million cells were re-

suspended with 12.5 microlitres of thrombin, as it is the less viscous of the 2 components of

fibrin sealant. The same volume of fibrinogen was added to the thrombin-cell suspension and

incubated at 37°C without endothelial growth medium for 24 hours. Five millilitres of

endothelial growth medium was added after 24 hours and the cells were incubated for another

day. The cell survival was assessed at 24 hours and 48 hours by examining cell morphology

(normal healthy cell appearance – adherence to flask, flat monolayer).

68

2.3 Surgery

Endothelial progenitor cells were prepared immediately prior to surgeries. After 9 to 10 days of

in vitro expansion, the cells were harvested by trypsinization, and washed in endothelial basal

medium without fetal bovine serum. For each rat, one million endothelial progenitor cells were

mixed in 12.5 microlitres of thrombin solution of fibrin sealant and kept on ice. The delivery of

the cells involved pipetting the 12.5 microlitres of thrombin and cell mixture with 12.5

microlitres of viscous fibrinogen solution to the central hole made on the greater tuberosity

during surgery.

The surgical procedure involves a unilateral detachment of the supraspinatus tendon from its

humeral insertion followed by a transosseous suture repair. The surgical model has been

developed by Thomopoulos and colleagues to examine tendon-bone healing at the insertion 181

.

During the surgery, 10 day-old endothelial progenitor cells expanded in vitro were delivered in

fibrin sealant. All surgical procedures were approved by the Animal Care Committee and the

Research Vivarium, St. Michael’s Hospital.

Prior to the surgeries, endothelial progenitors were harvested from the culture flasks and

suspended in thrombin, the less viscous solution of the fibrin sealant components. They were

then kept on ice until use.

All the surgeries were performed in the same manner. In each surgery, once the rat had been

anaesthetized with isoflurane (5 %) in combination with oxygen (2 L/min) in the induction

chamber, it was weighted and given a pre-operation dose of buprenorphine (0.05 mg/kg) via

69

intraperitoneal injection. The rat was shaved from the elbow to the neck to expose the skin,

disinfected with Betadine (povidone-iodine) and alcohol. The rat was transferred to the operating

table and was maintained anesthetized with isoflurane (2 %) in combination with oxygen (2

L/min) through the nose cone.

A three centimetre skin incision was made with a scalpel blade at the anterolateral aspect of the

rat’s right shoulder to expose the deltoid-acromion structure. The humerus of the rat was kept in

adduction and external rotation, and a longitudinal incision was made on the deltoid muscle,

splitting the muscle without detaching the deltoid from its acromion origin. Access to the

supraspinatus tendon was created. The supraspinatus tendon insertion on the greater tuberosity of

the proximal humerus could be effectively visualized by placing the humerus in adduction and

internal rotation at the glenohumeral joint.

Before detaching the supraspinatus tendon from it insertion, a Mason-Allen stitch to the distal

tendon was made using a 3-0 Ethicon suture (Johnson & Johnson, Somervile, New Jersey, USA).

Subsequently the tendon was sharply detached from the proximal humerus with a scalpel blade,

and any residual fibrocartilage at the footprint was removed by gently decorticating the greater

tuberosity to expose the bone marrow with a 1.0 mm crosscut burr (Stryker, Kalamazoo,

Michigan, USA). A bone tunnel was made using the burr at the anterior margins of the insertion

two millimetres from the articular surface. Both ends of the suture from the Mason-Allen stitch

were passed through this tunnel and tied over a bony bridge on the lateral aspect of the proximal

humerus.

Prior to securing the tendon repair, the rat was randomized to one of the groups to receive either

the fibrin sealant carrying endothelial progenitor cells or fibrin sealant only. The experimental

group received 106 endothelial progenitor cells suspended in 25 microlitres of the fibrin sealant

70

carrier and the fibrin group received only 25 microlitres of the fibrin sealant carrier. The cells

were first suspended in 12.5 microlitres of thrombin, and then injected at the central hole

simultaneously with 12.5 microlitres of fibrinogen. The clot formed at the bony trough made by

decorticating the fibrocartilaginous footprint. The supraspinatus tendon was approximated to the

exposed marrow by pulling the suture and tying it over the humeral metaphyseal cortex. The

quality of the repair was evaluated by applying slight tension to the tendon. The deltoid muscle

and the skin were closed separately using a standard suturing technique with 5-0 Ethicon suture

(Johnson & Johnson, Somervile, New Jersey, USA).

The rat was then transferred to the recovery chamber until it recovered from the anaesthesia, and

was then housed individually. Two doses of buprenorphine (0.05 mg/kg) were administrated to

each rat during the post-operative period for two days. All rats were allowed normal weight

bearing and cage activities, standard diet and water ad libitum.

71

2.4 End Points, Sacrifices, and Dissections

There were three end points for biomechanical testing (two, four, and six weeks after rotator cuff

repair), and two for histological analysis (four and six weeks after rotator cuff repair). End points

for cell tracking were three days, one week, two weeks, and four weeks after rotator cuff repair.

From the experimental and the fibrin control groups, ten rats were dedicated for biomechanics

and the other four for histology at each end point.

All dissections were performed post mortem. The rats were sacrificed via anaesthesia with

isoflurane (5 %) in combination with oxygen (2 L/min) in the induction chamber followed by

cardiac injection of a dose of diluted euthanizing agent, T61. Each rat received 0.5 millilitres of

T61 diluted with 0.5 millilitres of saline. The shoulder regions were then shaved to fully expose

the skin.

For biomechanics, the rat shoulders were dissected of irrelevant tissues; each specimen contained

only the humerus, the supraspinatus tendon, and the supraspinatus muscle. Each specimen was

immediately wrapped in gauze which was soaked in phosphate buffered saline, and kept in -80°C

freezer until use.

For histology, the rat shoulders were harvested en bloc; each specimen included the humerus, the

scapula, and the adjacent soft tissues. Each specimen was immediately immersed in 10 % neutral

buffered formalin and kept at room temperature for at least five days before any subsequent

tissue processing.

72

2.5 Biomechanical Testing

The rotator cuff specimens, consisting of the supraspinatus tendon attached to the humerus, were

stored in the -80°C freezer. The suture was removed and was not included in the testing. Before

the biomechanical testing, the specimens were thawed overnight in the 4°C refrigerator. They

were warmed to room temperature prior to testing. The area of the tendon-bone interface

(enthesis) was measured using a digital caliper. The area was estimated to be rectangular and

length and width were recorded for each specimen. The greater tuberosity of the humerus was

used as the main landmark for tendon/scar length measurement. Additionally, the central tunnel

for transosseous sutures was present in all repaired shoulders, and was used to approximate the

width of the healing tendon/scar.

i. Specimen mounting

The tendon end was stripped of muscle tissue, gripped with sandpaper and ethyl cyanoacrylate

adhesive (crazy glue), and clamped with custom screw grip. The tendon was aligned to the

humerus in a linear fashion 63

(see Figure 2-2). The tendon footprint area was measured using a

digital caliper.

73

ii. Tendon loading

The testing protocol included a preload to secure the tendon, a session of cyclic preconditioning

to obtain a consistent tendon loading history followed by a ramp to failure to determine the

ultimate load.

In detail, the tendons were preloaded to 0.10 N to eliminate the slack. Ten preconditioning cycles

from 0.10 N to 0.50 N at a rate of 1.0 mm/min were given to each of the tendon to reach a steady

state of mechanical equilibrium. The ramp to failure was performed at a rate of 1.0 mm/min.

Two sites of failure were anticipated: at the enthesis and at the tendon. An enthesis failure can be

a disruption at the tendon-bone interface with or without bony fragments attached. Disruption at

the tendon body can be considered tendon rupture, and the load to failure was not included in the

final analysis.

74

Grip

Jig

Clamps to secure the humerus

Humerus

Supraspinatus tendon

Figure 2-2. Biomechanical apparatus setup.

iii. Parameters

The measured parameters included the area of the footprint, tendon displacement, load, and

ultimate tensile load. The calculated parameters included the stiffness and stress.

(1) Tendon stress: tendon force relative to the tendon cross sectional area

75

(2) Tendon stiffness: change in tendon length in relation to the force applied to the tendon

These parameters were subjected to statistical analysis, comparing the experimental and the

fibrin control groups using the t-test for four and six weeks animals (n = 10 in each group), and

non-parametric test for two weeks animals (n = 4 in each group). Analysis of variance (ANOVA)

and multiple comparison test (Bonferroni’s test) were performed to compare the effect of healing

over time.

76

2.6 Histological Analysis

The specimens for histology consisted of the scapula, the humerus, the distal portion of the

clavicle, and all the attached musculature. The specimens were kept in 10 % neutral buffered

formalin and stored at room temperature for a minimum of five days. The shoulders were

processed in groups for consistency. After fixation, the specimens were trimmed of irrelevant

tissues and decalcified in Immunocal (Decal, Congers, New York, USA) for one week minimum.

The dehydration process was executed using Leica TP1020, an automated tissue processing

machine. The specimens were secured in a plastic cassette and went through 12 stations of

various solvents or solutions. See table below for details in each station.

Table 2-2. Tissue dehydration process by Leica TP1020

Station Reagent Duration (hr) Vacuum

1 Formalin 1 No

2 Formalin 1 No

3 70% EtOH 1.5 No

4 80% EtOH 1.5 No

5 95% EtOH 1.5 No

6 95% EtOH 1.5 No

7 100% EtOH 1.5 No

8 100% EtOH 2 No

9 Xylene 1.5 No

10 Xylene 2 No

11 Paraffin 60°C 1.5 Yes

12 Paraffin 60°C 2 Yes

77

After the specimens completed the dehydration process, they were embedded in paraffin in a

geometry which the supraspinatus was approximately perpendicular to the shaft of the humerus.

The specimens were positioned for sectioning in the coronal plane. The positioning of the tissue

sample was achieved with fast solidification of the paraffin wax on ice using Leica EG 1160.

The embedded samples were then stored at room temperature.

The sectioning was performed using a low profile blade on Leica RM 2235. Each section was

five microns thick in the coronal plane, cutting through the head of humerus, the tendon-bone

junction, and the supraspinatus tendon. The sections were mounted on positively charged

microscope slides.

78

2.6.1 Hematoxylin and Eosin Staining

The hematoxylin and eosin staining was performed using Leica Autostainer XL. Slides were then

examined for proper section orientation and depth as well as infection.

Table 2-3. Hematoxylin and eosin staining protocol using Lecia Autostainer XL

Station Reagent Time (min:sec) Exact Dips

Oven 65°C 10:00 No -

1 Xylene 3:00 No 3

2 Xylene 2:00 No 3

3 Xylene 3:00 No 3

4 100% EtOH 1:00 No 3

5 100% EtOH 1:00 No 3

6 95% EtOH 1:00 No 3

Wash 1 H2O 1:30 No 3

8 HAEMATOXYLIN 2:00 Yes 3

Wash 5 H2O 1:00 No 3

9 DEFINE 0:35 Yes 3

Wash 4 H2O 1:00 No 3

10 BLUE BUFFER 0:45 Yes 3

Wash 3 H2O 1:00 No 3

11 95% EtOH 1:00 No 3

12 EOSIN 1:00 Yes 3

13 100% EtOH 1:00 No 3

14 100% EtOH 1:00 No 3

16 100% EtOH 1:00 No 3

17 Xylene 1:00 No 3

18 Xylene 1:00 No 3

22 EXIT - - -

79

2.6.2 Safranin O/ Fast Green FCF Staining

The staining for cartilage involved three parts: (1) deparaffinization/rehydration, (2) cartilage

staining, and (3) dehydration. The deparaffinization/rehydration steps were a replication of

haematoxylin staining using Leica Autostainer XL.

Table 2-4. Deparaffinization and rehydration of slides using Leica Autostainer XL

Station Reagent Time (min:sec) Exact Dips

Oven 65°C 10:00 No -

1 Xylene 3:00 No 3

2 Xylene 2:00 No 3

3 Xylene 3:00 No 3

4 100% EtOH 1:00 No 3

5 100% EtOH 1:00 No 3

6 95% EtOH 1:00 No 3

Wash 1 H2O 1:30 No 3

22 Exit - - -

After the slides were retrieved, the cartilaginous tissues were stained with Safranin O and fast

green FCF. First the slides were submerged in 0.1 % aqueous fast green FCF stain (OmniPur,

EMD Chemicals, Darmstadt, Germany) for 3 minutes followed by an acetic acid (1 %) wash for

30 seconds. Then they were submerged in 0.1% aqueous Safranin O (Harleco, EMD Chemicals,

Darmstadt, Germany) for 5 minutes.

For the dehydration process, the slides were transferred to the Leica Autostainer XL.

Table 2-5. Dehydration of slides using Leica Autostainer XL

80

Station Reagent Time (min:sec) Exact Dips

Wash 3 H2O 1:00 No 3

11 95% EtOH 1:00 No 3

13 100% EtOH 1:00 No 3

14 100% EtOH 1:00 No 3

16 100% EtOH 1:00 No 3

17 Xylene 1:00 No 3

18 Xylene 1:00 No 3

22 EXIT - - -

The proteoglycans in the cartilage/fibrocartilage stained red, the nuclei stained black, and the

cytoplasm stained green.

Safranin O binds to proteoglycans in the fibrocartilage and presents a contrasting red colour

against the green background. The fibrocartilage was outlined on three sequential slides for each

rotator cuff with a software program and the area for each slide was calculated. The mean area

and the standard deviation for each group were determined and compared.

2.6.3 Collagen Birefringence

The collagen parallelism was examined with polarized light microscopy. When a beam of light

passes through a polarizing filter, the light will exist with vibration in one direction. When it

encounters another polarizing filter with a main axis perpendicular to that of the first filter, the

light will not pass through, resulting in the dark field effect. Collagen has the ability to rotate the

axis of light passing through called birefringence. Once the vibration direction of the light has

81

been rotated, the light will pass through the second filter, and collagen appears bright against the

dark background.

The parallelism of the collagen fibres close to the insertion can be evaluated with the parallelism

index. The index is calculated based on the amount of light signal fluctuation. Higher fluctuation

indicates more organized fibre directions as light intensity is greatest when collagen fibres are

parallel (anisotropy). Likewise, unorganized fibres will generate smaller fluctuation of light

intensity as multiple fibre orientations are represented.

The slides were first deparaffinized/rehydrated using Leica Autostainer XL, then stained in

picrosirius red (Sirius red, Sigma-Aldrich, St Louis, Missouri, USA) for one hour and washed

twice with an acetic acid (1 %) wash for 30 seconds each. The slides were dehydrated using the

Leica Autostainer XL.

Picrosirius red staining was used to enhance the birefringent property of collagen. The light

intensity, the rotation of the filter, and the focus of the camera were standardized for all the

measurements. The specimens were orientated so that the tendons were parallel to the edge of the

stage. The field of view was imaged and the mean light intensity was recorded for each specimen

using an image analysis program based on language IDL 6.3 ( ITT Visual Information Solutions,

Boulder, Colorado). The mean and the standard deviation for each group was determined and

compared.

82

2.6.4 Vessel Staining

The tissues were stained for CD 31 (PECAM-1). The slides were deparaffinized/rehydrated in

the same manner as before using the Leica AutoStainer XL. The antigen retrieval step involved

incubating the slides with 10 mM of sodium citrate at 98°C for 20 minutes. The slides were then

treated with 3 % hydrogen peroxide for 30 minutes to quench the endogenous peroxide. They

were washed thoroughly and followed by incubation with Dako protein block (Dako, Glostrup,

Denmark)) for 30 minutes. The serum was wiped off and the rabbit anti-CD 31 antibody

(Abcam, Cambridge, England, United Kingdom) was applied to the slides. After incubation

overnight at 4°C, the slides were washed and the secondary antibody (Goat anti-rabbit antibody:

Biotinylated, Dako, Glostrup, Denmark) was added at 1:200 dilution and incubated at room

temperature for 30 minutes. The VECTORSTAIN Elite ABC System (Vector Labs, Burlingame,

CA, USA) was used for detection. Solutions A and B were both diluted 1:50 for avidin-

biotin/horse raddish peroxidase binding. The DAB reagent was then applied for colour

development, incubated at room temperature for 10 minutes. The slides were rinsed, dehydrated,

and mounted for analysis.

Images of the new tendon insertion (the area of interest where the tendon boardered the bone and

the adjacent tendinous tissues) were captured and selected, and the vascular objects were counted

as vascular density (vessel/mm2) using an image analysis program, Visiopharm Integrator

System (Visiopharm, Hoersholm, Denmark). The mean and the standard deviation for each

group were determined and compared.

83

2.7 Cell Tracking

Green fluorescent protein (GFP) was used to track transplanted endothelial progenitor cells. The

endothelial progenitor cells were isolated from the bone marrow of transgenic GFP Fischer 344

rats [F344-Tg(UBC-EGFP)F455Rrrc, RRRC#: 307; Rat Resource and Research Center,

Columbia, MO, USA] using the same isolation techniques and procedure as previously

described. Additionally, GFP cell pellets were collected for positive control staining. The GFP

positive endothelial progenitor cells were cultured in the same manner, and delivered to the

rotator cuff tendon insertion with fibrin sealant. The rats were allowed to heal for three days, one

week, two weeks, and four weeks. Two rats were allocated per time point. Regular non-GFP

fluorescent endothelial progenitor cells were used for control.

Once the shoulders were dissected, they were processed in the same manner as previously

described for histology. Again, the specimens consisted of the scapula, the humerus, the distal

portion of the clavicle, and all the attached musculature. They were then fixed, decalcified, and

embedded. Five micron sections were cut in preparation to stain for GFP protein using

immunohistochemistry.

After the slides were cut, they were deparaffinized and rehydrated. The antigen retrieval step

involved incubating the slides with 10 mM of sodium citrate at 98°C for 20 minutes. Then the

endogenous hydrogen peroxidases were blocked with 3 % hydrogen peroxide solution for 30

minutes. Dako protein block (Dako, Glostrup, Denmark) was applied and incubated for 30

minutes at room temperature. The primary antibody (Chicken IgY anti-GFP antibody,

84

Invitrogen, Carlsbad, California, USA) was added at 1:500 dilution and incubated at 4°C

overnight. The secondary antibody (Goat anti-chicken antibody: Biotinylated, Vector Labs,

Burlingame, CA, USA) was added at 1:200 dilution and incubated at room temperature for 30

minutes. The VECTORSTAIN Elite ABC System (Vector Labs, Burlingame, CA, USA) was

used for detection. Solutions A and B were both diluted 1:50 for avidin-biotin/horse raddish

peroxidase binding. The DAB reagent was then applied for colour development, incubated at

room temperature for 10 minutes. The slides were rinsed, dehydrated, and mounted for analysis.

The transplanted endothelial progenitor cells (GFP-expressing) were identified by the presence

of the brown stain under a microscope.

85

3 Results

3.1 Endothelial Progenitor Cells

3.1.1 Flow cytometry

Cells were selected out of all detected events in the flow cytometry. Then single cells, CD 34

positive, and CD 133 and Flk-1 positives, and lastly, CD 31 positive cells were gated. The

analysis found 98.3 % of single cells positive for CD 34. Of the CD 34 positive cells, 96.8 %

were positive for both CD 133 and Flk-1. These antigens indicated progenitor and endothelial

characteristics. The “stemness” of the progenitor cells was illustrated by the high expression of

CD 34 and CD 133, reflective of the hematopoietic lineage. The presence of Flk-1, also known

as vascular endothelial growth factor receptor-2 (VEGFR-2), indicated the response to

angiogenic signals. With respect to the total single cell population, 98 % of the cells were

positive for all three surface antigens.

Lastly, CD 31 was gated to determine endothelial commitment. CD 31, platelet endothelial cell

adhesion molecule 1 (PECAM-1), is a mature endothelium marker. This set of surface antigen

profile in the EPC population indicated its origin from the hematopoietic lineage and

commitment to the endothelial lineage. The level of CD 31 expression, 60 % positivity, was not

as high as the other markers because the endothelial progenitor cells have yet to lose their

86

stem/progenitor cell phenotype. The cells possessed more of the characteristics of stem cells than

the terminally differentiated endothelial cells.

Please refer to Appendix A for the full cytometric analysis.

CD 34 CD 133

Flk-

1

Sid

e Sc

atte

r

Figure 3-1. Flow cytometry analysis: 98.3 % of the cultured cells expressed CD 34 on the surface

(left); of these CD 34 expressing cells, 96.8 % of the cells expressed CD 133 and Flk-1 (right). In

summary, 95 % of the cultured cells expressed all three surface antigens.

87

3.1.2 Functional Assays

i. AcLDL and lectin

The isolated cell population was identified by DAPI, staining the nucleus blue. The cells showed

high positivity for both DiI-AcLDL and Alexa Fluor 488 conjugated GS-II, suggesting the

uptake of AcLDL and the binding to lectin molecules. AcLDL uptake and lectin binding are

characteristics of the endothelial cell lineage. The results indicate that the isolation population

possesses endothelial phenotypes.

88

Figure 3-2. AcLDL and lectin stains: DiI-AcLDL (red fluorescence) staining and DAPI (blue

fluorescence) staining (top); GS-II lectin DAPI (green fluorescence) staining and DAPI DAPI

(blue fluorescence) staining (bottom). Bar = 50 μm.

89

ii. Tube formation

The angiogenic potential of the progenitor cells was assessed by their ability to form tube-like

structures on the basement membrane matrices, Matrigel. The mesenchymal stem cell control did

not have tube-like structure 24 hours after seeding.

Figure 3-3. Tube formation assay 24 hours after seeding: endothelial progenitor cells (after ten

days of culturing, left) and mesenchymal stem cells (right). The endothelial progenitor cells

formed tube-like structures while the mesenchymal stem cells were in isolated colonies.

90

3.1.3 Cell survival in fibrin sealant

The cells survived the clot formation from adding thrombin to fibrinogen. The general cell shape

remained intact at 24 hours after mixing the components to produce fibrin. After the growth

media was supplemented, the endothelial progenitor cells were able to proliferate at the edges of

the fibrin clot. They were able to find a surface to adhere to and assumed the cell adherent

morphology. The cells remained active after 48 hours in the fibrin clot.

When delivered in vitro, the fibrin clot is expected to be disintegrated by the host within a day.

The cells will then be exposed to the host tissue environment which will provide the essential

nutrients and oxygen for metabolism. From this result, the endothelial progenitor cells could be

expected to survive the delivery and remain active in the host.

Figure 3-4. Cell survival: cells suspended in fibrin for 24 hours (left) and cell proliferation at 48

hours (right). After the cells were placed into a flask after being in the fibrin clot for 24 hours,

they were able to adhere, resume monolayer morphology, and proliferate.

91

3.2 Biomechanical Testing

The average weight of the Fibrin group at the time of repair was 283 ± 21 g. The weight of the

Fibrin group at the three respective endpoints, two, four, and six weeks, was 297 ± 16 g, 319 ±

20 g, and 331 ± 14 g. The average weight of the EPC group at the time of repair was 281 ± 13 g.

The weight of the EPC group at the three respective endpoints, two, four, and six weeks, was 293

± 16 g, 315 ± 10 g, and 332 ± 18 g. There were no differences between the Fibrin and the EPC

groups at the time of the repair or at the endpoints.

W e ig h t o f R a ts

We

igh

t (g

)

S u r g e r y W e e k 2 W e e k 4 W e e k 6

0

1 0 0

2 0 0

3 0 0

4 0 0

F ib r in G ro u p

E P C G ro u p

W e ig h t o f R a ts

We

igh

t (g

)

S u r g e r y W e e k 2 W e e k 4 W e e k 6

0

1 0 0

2 0 0

3 0 0

4 0 0

F ib r in G ro u p

E P C G ro u p

Figure 3-5. The weight of rats. The weights of the rats in EPC groups and Fibrin groups were not

significantly different at any of the time points. Error bars represent standard deviations.

92

3.2.1 Ultimate Load

For the healthy contra-lateral shoulders, the failure sites were either at the physis or the tendinous

tissue close to the interface. Only the ones that failed at the tendon-bone interface were included.

However, the failure modes of the biomechanical tests for the operated tendons (with sutures

removed) were consistently at the tendon-bone interface. The fibrous scar at the attachment was

significantly weaker than the tendinous tissue.

The healthy shoulders tended to have a clean slope to failure with one distinctive peak and a

characteristic drop in force, indicating a complete rupture. The operated shoulders, however,

displayed discontinuous ramps, multiple peaks, and a gradual decrease in force.

The healthy insertions had an average ultimate force of 26.39 ± 5.27 N with a sample size of 10

specimens. For the Fibrin group, the ultimate force at two, four, and six weeks was determined to

be 9.33 ± 3.86 N, 10.67 ± 3.29 N, and 19.09 ± 3.48 N. The ultimate load for the EPC group at

two, four, and six weeks was 9.14 ± 3.00 N, 11.58 ± 2.74 N, and 24.64 ± 3.58 N*. However, at

six weeks, there was an appreciable difference between the Fibrin and the EPC groups

(p=0.0025). ANOVA and Bonferroni’s multiple comparison test indicated no differences in

ultimate force between at two and four weeks, but significant increase in ultimate force from four

to six weeks in both groups. Refer to Appendix B for details of Bonferroni’s multiple

comparison test.

93

Table 3-1. Ultimate Load (N)

Time Point Fibrin Group EPC Group

Week 2 9.33 ± 3.86 N 9.14 ± 3.00 N

Week 4 10.67 ± 3.29 N 11.58 ± 2.74 N

Week 6 19.09 ± 3.48 N 24.64 ± 3.58 N*

Significance: * p = 0.0025

Figure 3-6. The ultimate load at the insertions. The ultimate load to failure is significantly higher

in the EPC treated repairs at 6 weeks (t-test, p = 0.0025). Error bars represent standard

deviations. The dotted line represents the ultimate load for healthy tendon, 26.39 N.

94

3.2.2 Interface Area and Ultimate Stress

The area of the tendon-bone interface was significantly smaller in the healthy insertions

compared to the operated groups. These healthy insertions were better defined and their

boundaries were clear. On the other hand, the operated insertions in the Fibrin and the EPC

groups were filled with scar tissues. For the Fibrin group, the interface area at two, four, and six

weeks was determined to be 7.69 ± 0.49 mm2, 6.08 ± 0.76 mm

2, and 6.71 ± 0.49 mm

2. The

interface area for the EPC group at two, four, and six weeks was 7.58 ± 0.51 mm2, 6.24 ± 1.22

mm2, and 6.43 ± 0.54 mm

2. There was no significant difference in the area between the groups.

The interface area remained larger than that of the healthy tendon insertion at all the time points.

ANOVA and Bonferroni’s multiple comparison test indicated a decrease in the cross sectional

area in the Fibrin group from two to four weeks. The difference was not significant in the EPC

group due to large standard deviation. However, there were no significant differences between

the EPC and the Fibrin groups in either time point. Refer to Appendix B for details of

Bonferroni’s multiple comparison test.

95

Table 3-2. Interface Cross-Sectional Area (mm2)

Time Point Fibrin Group EPC Group

Week 2 7.69 ± 0.49 mm2

7.58 ± 0.51 mm2

Week 4 6.08 ± 0.76 mm2 6.24 ± 1.22 mm

2

Week 6 6.71 ± 0.49 mm2 6.43 ± 0.54 mm

2

Figure 3-7. Area of tendon-bone attachments. No differences in the insertion area were found at

any time point. Error bars represent standard deviations. The dotted line represents the area of

tendon-bone attachment for healthy insertions, 2.71 mm2.

96

The ultimate stress was determined by dividing the ultimate load with the corresponding tendon-

bone interface area. This normalized the attachment strength with respect to the size of the

insertion. For the Fibrin group, the ultimate stress at two, four, and six weeks was determined to

be 1.21 ± 0.52 MPa, 1.79 ± 0.66 MPa, and 2.83 ± 0.35 MPa. The ultimate stress for the EPC

group at two, four, and six weeks was 1.20 ± 0.33 MPa, 1.89 ± 0.44 MPa, and 3.83 ± 0.43 MPa*.

Significant difference was detected at six weeks, with the EPC group demonstrating significantly

higher ultimate stress (p<0.0001). ANOVA and Bonferroni’s multiple comparison test indicated

gradual improvements from two weeks to six weeks in the EPC group over time. The Fibrin

group did not show significant differences between two and four week due to large standard

deviation. However, differences between the Fibrin and EPC groups were not seen until six

weeks. Refer to Appendix B for details of Bonferroni’s multiple comparison test.

97

Table 3-3. Ultimate Stress of Tendon-Bone Interface (N/mm2 or MPa)

Time Point Fibrin Group EPC Group

Week 2 1.21 ± 0.52 MPa 1.20 ± 0.33 MPa

Week 4 1.79 ± 0.66 MPa 1.89 ± 0.44 MPa

Week 6 2.83 ± 0.35 MPa 3.83 ± 0.43 MPa*

Significance: * p < 0.0001

Stress

*

Figure 3-8. Ultimate stress at the insertion. The stress significantly higher in the EPC group 6

weeks after the repair (p < 0.0001). Error bars represent standard deviations. The stress for

healthy insertions is 9.84 MPa (not shown on the graph).

98

3.2.3 Stiffness

The stiffness of the tendon and scar tissues were determined by the largest value of slope in the

linear region of the ramp to failure. For the Fibrin group, the stiffness at two, four, and six weeks

was determined to be 6.77 ± 2.23 N/mm, 8.17 ± 3.11 N/mm, and 13.17 ± 3.93 N/mm. The

stiffness for the EPC group at two, four, and six weeks was 8.11 ± 1.75 N/mm, 8.56 ± 2.62

N/mm, and 12.78 ± 2.68 N/mm. There was no significant difference between the Fibrin and the

EPC group. ANOVA and Bonferroni’s multiple comparison test indicated no differences in

stiffness between at two and four weeks, but significant increase in stiffness from four to six

weeks in both groups. Refer to Appendix B for details of Bonferroni’s multiple comparison test.

99

Table 3-4. Stiffness of Tendons (N/mm)

Time Point Fibrin Group EPC Group

Week 2 6.77 ± 2.23 N/mm 8.11 ± 1.75 N/mm

Week 4 8.17 ± 3.11 N/mm 8.56 ± 2.62 N/mm

Week 6 13.17 ± 3.93 N/mm 12.78 ± 2.68 N/mm

Figure 3-9. The stiffness of the tendons. No differences in stiffness were found at any time point.

Error bars represent standard deviations. The dotted line represents the stiffness for healthy

insertions, 23.17 N/mm.

100

Table 3-5. Summary of biomechanical values for 2, 4, and 6 Week Groups

2 Week Groups Fibrin (n=4) EPC (n=4)

Ultimate Load (N) 9.33 ± 3.86 9.14 ± 3.00

Stiffness (N/mm) 6.77 ± 2.23 8.11 ± 1.75

Insertion Area (mm2) 7.69 ± 0.49 7.58 ± 0.51

Stress (N/mm2) 1.21 ± 0.52 1.20 ± 0.33

4 Week Groups Fibrin (n=10) EPC (n=10)

Ultimate Load (N) 10.67 ± 3.29 11.58 ± 2.74

Stiffness (N/mm) 8.17 ± 3.11 8.56 ± 2.62

Insertion Area (mm2) 6.08 ± 0.76 6.24 ± 1.22

Stress (N/mm2) 1.79 ± 0.66 1.89 ± 0.44

6 Week Groups Fibrin (n=10) EPC (n=10)

Ultimate Load (N) 19.09 ± 3.48 24.64 ± 3.58*

Stiffness (N/mm) 13.17 ± 3.93 12.78 ± 2.68

Insertion Area (mm2) 6.71 ± 0.49 6.43 ± 0.54

Stress (N/mm2) 2.83 ± 0.35 3.83 ± 0.43**

Significance: * p = 0.0025; ** p < 0.0001

101

3.3 Histological Analyses

All stains were performed on the operated shoulders as well as the unoperated, healthy

contralateral shoulders.

Figure 3-10. Hematoxylin and eosin, Safranin O, picrosirius red, and CD 31 antigen staining

(from left to right) of healthy rotator cuff tendons. Bar = 2 mm.

3.3.1 Hematoxylin and Eosin Staining

Qualitative analysis of the shoulder indicated that the operated tendons were damaged, and that

scar tissue was apparent. The central hole in the humeral head at the native tendon insertion was

obvious in the operated samples. The sections made were of the correct orientation and at similar

depth.

102

Fibrin Group EPC Group

Week 4

Week 6

Figure 3-11. Hematoxylin and eosin. The orientation and the depth of the shoulder sections were

appropriate. Bar = 2 mm.

3.3.2 Safranin O staining

No fibrocartilage was observed at the insertion site at four or six weeks after the repair in both

the Fibrin and the EPC groups.

103

Fibrin Group EPC Group

Week 4

Week 6

Figure 3-12. Safranin O and fast green staining. No cartilage formation was detected at the

tendon-bone interface. Bar = 2 mm.

3.3.3 Picrosirius Red Staining

The polarization birefringence property was enhanced using picrosirius red staining. The

measured brightness intensity demonstrated that there was no difference between the Fibrin and

the EPC groups, indicating that the degree of collagen immaturity and disorganization were

similar.

104

Fibrin Group EPC Group

Week 4

Week 6

Figure 3-13. Picrosirius red staining. The organization of collagen fibers was assessed under

polarized light. There was no difference in collagen organization between the Fibrin and the EPC

groups. Bar = 2mm.

105

Table 3-6. Collagen birefringence measured in mean optical density

Time Point Fibrin Group EPC Group

Week 4 89.82 ± 20.22 96.69 ± 12.30

Week 6 108.20 ± 9.52 91.23 ± 7.35

Figure 3-14. Polarization of tendinous insertion. No significant differences were found between

the groups at any time point. Error bars represent standard deviations.

106

3.3.4 Vessel Staining

There was no significant difference between the Fibrin and the EPC groups in the number of

vessels. Although the values did not reach statistical significance, the vascular density was

higher in the EPC group and both time points. In addition, an increase of vascularity was

observed in the EPC group between four and six weeks, while the Fibrin group exhibited the

opposite trend.

Fibrin Group EPC Group

Week 4

Week 6

Figure 3-15. CD 31 staining. The insertion region was selected for analysis. The EPC group

showed an increase in vascularity from Week 4 to Week 6. Bar = 2 mm.

107

Table 3-7. Vascular density (vessels/mm2)

Time Point Fibrin Group EPC Group

Week 4 279 ± 141 (n = 4) 376 ± 126 (n = 3)

Week 6 188 ± 78 (n = 4) 457 ± 146 (n = 3)

Figure 3-16. Number of blood vessels per millimeter squared (vascular density) at the insertion

site as identified by CD 31 immunohistochemistry staining. Although the differences were not

statistically significant, the density of vessels was increased in the EPC group at both time points

and the EPC group showed a trend towards increasing vessel density from 4 to 6 weeks, whereas

the Fibrin group showed a decrease in vessel density from 4 to 6 weeks. Error bars represent

standard deviations.

108

3.4 Cell Tracking

Green fluorescent protein (GFP) positive endothelial progenitor cells were observed at the rotator

cuff repair site at three days, one week, two weeks, and four weeks. The presence of the cells

indicates survival and retention at the site of injury.

Most GFP positive cells were surrounded by tendinous scar tissues. Only a few were

incorporated into the endothelium.

109

Figure 3-17. The brown HRP-DAB stain indicated GFP-expressing endothelial progenitor cells

(transplanted cells) at Week 2 in the scar tissue at the injury site. Few of the GFP positive cells

incorporated into the endothelium as indicated by the red arrows.

110

4 Discussion

Rotator cuff retear is a common complication after surgical repair. Causes for retear have been

proposed and strategies to augment tendon healing have been investigated. Scaffolds, biological

factors, surgical techniques, cells, and even rehabilitation have been indicated in variable studies

to improve integration of the tendinous and bony tissues at the insertion. However, there is

currently a lack of clinically effective means to augment rotator cuff repair biologically.

Recently, there has been significant interest in the investigation of adult pluripotent stem cells to

aid rotator cuff tendon healing. Cell-based therapies are of a particular interest among other

therapeutic strategies because of their ability to contribute to healing through self-renewal,

paracrine effects, and/or differentiation into tissue-specific cells that contribute to the healing

process.

Our study was built on previous findings of restricted blood supply to the supraspinatus tendon

insertion and the therapeutic angiogenic effect of endothelial progenitor cells. Vascular ingrowth

is critical to facilitate the delivery of nutrients and cells that respond to the requirements of the

stages of healing. Since the successful identification of endothelial progenitor cells by Asahara et

al 6, there has been a wealth of evidence that these cells induce angiogenic responses in various

in vivo models. With endothelial progenitor cells and their known ability for neovascularization,

we hoped to develop a novel cell therapy for rotator cuff repair to augment healing at the poorly

vascularized tendon-bone interface.

111

The rat rotator cuff repair model is a well-established surgical model for the investigation of

tendon-bone healing and therapeutic interventions 63, 169, 181

. The fibrin sealant delivery method

and the survival of transplanted cells were assessed in previous studies 187

. The use of autologous

progenitor cells minimized immunogenic responses in the host. No adverse immunogenic

response was observed in the animals.

The tracking of the transplanted green fluorescent protein-expressing endothelial progenitor cells

using immunohistochemistry at three days indicated the survival of the transplanted cells.

However, the survival percentage of transplanted cells is not quantified. Determining the survival

percentage should give light to the effectiveness of cell delivery through fibrin sealant.

112

4.1 Endothelial Progenitor Cells

The Ficoll density separation limited the cells collected from the bone marrow to be strictly

mononuclear, which contains the progenitor cells including endothelial progenitor cells as well

as mesenchymal stem cells. The fibronectin coating and the growth media supplementation

favoured the adherence and proliferation of endothelial progenitor cells in vitro culturing.

The high expression of CD 34 and CD 133 suggested that the isolated endothelial progenitor

cells possessed the characteristics of stem cell characteristics, especially that of the

hematopoietic lineages. As importantly, the presence of vascular endothelial growth factor

receptor-2 (VEGFR-2) indicated response to angiogenic signals. The same population of

endothelial progenitor cells also demonstrated endothelial capacity: expression of CD 31

(PECAM-1), uptake of AcLDL and binding of lectin, and in vitro tube formation. Collectively

these assays confirmed that the population of cells isolated from the bone marrow was indeed

functional endothelial progenitor cells.

The cells were only cultured for ten days prior to surgery, and thus were considered early

endothelial progenitor cells. Cells that were cultured for more than two weeks would be

considered late endothelial progenitor cells. Early endothelial cells are believed to predominantly

provide paracrine effects, and the presence of these cytokine secreting early endothelial

progenitor cells was important for in vivo angiogenic mechanisms 208

. The result in this study

supports the paracrine effects of transplanted early endothelial progenitor cells (see Chapter 4.4).

113

4.2 Biomechanical Testing

The ultimate force for this rat rotator cuff repair model is considered the most appropriate

biomechanical outcome as it is the most repeatable measurement with the least source of error.

The injury no doubt weakened the strength at the tendon-bone interface. All operated specimens

ruptured at the tendon-bone interface (repair site, sutures removed). Even at six weeks after the

repair, the attachment (Fibrin group) remained weaker than the healthy, uninjured insertions. In

this study, the percentage of strength recovered at each time point are 35 %, 43 %, and 72 % of

the normal insertion strength, versus the reported value at 22 % and 52 % at two and four weeks

15. Additionally, Gulotta et al found no benefits of applying fibrin sealant during rat rotator cuff

repair 63

.

We found no significant differences between the Fibrin and the EPC groups at two or four weeks

in the ultimate force, interface cross-sectional area, stiffness, and ultimate stress. Similar results

were reported by Gulotta and colleagues when they used mesenchymal stem cells alone in the

same rotator cuff repair model 63

. From gross observation, it is evident that fibrous tissues filled

the interface and attached over the greater tuberosity of the humeral head in a disorganized

fashion. The coverage area was much larger than the original insertion, likely a reparative

mechanism to compensate for the weaker scar fibers, providing stability during the initial stages

of healing. Results of rotator cuff repair using cell therapy by Gulotta et al also indicated large

amount of scar at the attachment 63

.

Significant improvements in the tendon ultimate load and stress were detected after six weeks of

healing with the addition of endothelial progenitor cells. The ultimate load of the rotator cuff

114

insertion in the EPC group (24.64 ± 3.58 N) was comparable to that in the healthy insertions

(26.34 ± 5.27 N). Manning and co-workers have reported that the attachment strength in repaired

tendon insertions (TGF-treated and control groups) approaches the normal/uninjured strength at

28.8 N eight weeks after repair 111

. The large cross-sectional area of these fibrous tissues at the

tendon-bone interface offered strength to the attachment more so than the quality of the tendon

tissue. Extensive scar formation was observed in studies using the same rotator cuff model 54, 63,

181. Consequently, the ultimate stress at six weeks (3.83 ± 0.43 N/mm

2) remained inferior to that

of the healthy insertions (9.84 ± 2.15 N/mm2).

The trend showed a slow healing progress in the first four weeks followed by a rapid increase in

strength. This is likely from the effect of the remodeling phase in which the weaker type III

collagen fibrils were replaced by the stronger, more organized type I collagen fibrils. Galatz and

colleagues reported an increased expression of mRNA and protein contents in both types I and

III collagen at ten days post-surgery 54

. Between two to four weeks, there was a drop in type I

collagen content and synthesis and that of the type III collagen plateaued. After four weeks,

types I and III collagen synthesis exhibited opposing trends, with increasing type I and

decreasing type III, suggesting structural remodeling. The biomechanical effect in our study was

consistent with the trend observed in the previous literature.

115

4.3 Histology

The histology results indicated that there was no significant difference between the Fibrin group

and the EPC group. At four weeks, the literature reported high type III collagen content and

synthesis 54

. Disorganized fibers were present in abundance even at six weeks in our study.

Collagen birefringence showed less dense and less mature collagen fibers at the insertion of

repaired shoulders compared to the healthy, non-injured tendons. Scarring has been associated

with transforming growth factor beta (TGF-β) signaling. TGF-β1 level peaked at day ten after

rotator cuff repair, corresponding to the peak cell proliferation and extracellular matrix synthesis

54. The administration of TGF-β3 during tendon-bone healing led to increased formation of

fibrovascular scar tissue 111

. It seems that scarring is an inevitable process in adult rotator cuff

healing, and provides stability and strength to the injured tissue prior to structural remodeling as

suggested in the previous TGF-β3 study. Whether or not having abundant scar tissue at this stage

of healing is therapeutic would require longer endpoints fully assess.

We did not observe fibrocartilage at the repair site in either treatment group. Ide et al did not find

safranin O-stained fibrocartilage in the FGF-untreated shoulders at two or six weeks 79

. Using

unmodified mesenchymal stem cells alone did not result in increased fibrocartilage 63

. In the cell

therapy studies where increased fibrocartilage was detected 64, 65

, molecular signaling

(transduction of MT1-MMP gene or Scleraxis gene into mesenchymal stem cells) was

incorporated to induce tissue regeneration. Collectively, this evidence suggests that early

regeneration of this fibrocartilaginous transition zone between the tendon and the bone requires

more than just the host’s intrinsic anabolic cells and the transplanted adult stem cells. The repair

116

site likely lacks the critical biological factors/molecules and the appropriate reparative cells to

recreate a fibrocartilaginous insertion. It is, however, uncertain if angiogenic therapy will

promote the recreation of the native fibrocartilaginous insertion due to the nature of the

hypovascular transition zone.

The attempt to increase the blood supply to this region did not improve the healing response or

encourage early remodeling at four weeks. While the vascularity data did not reach statistical

significance in either time point, it is apparent that the EPC group had increasing vascularity at

the injury site between four to six weeks after the repair. At the same time the Fibrin group

showed decreasing vascularity. The endothelial progenitor cells were capable of sustaining

vascular supply, and possibly enhancing it, weeks after the surgery when the tendon naturally

progresses into its “hypovascular state”. The presence of the avascular critical zone in healthy

insertions is believed to have negative impact on the subsequent rotator cuff tendon-bone

healing. The questionable blood supply to the repair site, especially after massive rotator cuff

repairs, has been implicated in repair failure in clinical studies 71

. Other clinical studies have

found increased vascular distribution to the repair site compared to the uninjured states; however,

the studies showed either delayed 55

or a lower magnitude of response than the surrounding

tissues at the repair site 53

. In our study, the observed increase in vascularity correlated with the

mechanical advantages of the attachment site seen in the biomechanical testing. It is plausible

that prolonged supply of nutrient and cellular access to the attachment was responsible for the

superior strength. The alteration of the vascular environment perhaps allowed reparative cells to

engage in a higher anabolic level, for a longer period of time, than those without. The finding

here confirms that the lack of sustained vascular supply during insertion healing could have

detrimental consequences. However, given the histological findings in the tissue quality at the

117

repair site, the specific mechanism by which vascular supply improves attachment strength

remains unclear.

The conventional histological methods were unable to detect differences between the Fibrin and

the EPC groups. This suggests that the increase in strength is independent of the presence of

fibrocartilage or better collagen fibre organization. It seems that more sensitive and targeted

histological techniques are required to detect changes in the healing rotator cuff tendons.

Immunohistochemical characterization of collagen types (eg. types I, II, III, and X) and

tendinous ground substances (eg. aggrecan, decorin, and biglycan) at the insertion may shed light

on small changes in repair tissue and the distribution of these extracellular matrix molecules.

Changes in tissue levels of matrix metalloproteinases, modulators of tendon remodeling, may

help uncover the remodeling processes that led to stronger attachments. Moreover, the transition

zone includes the tendon and the bone, and endothelial progenitor cells have been shown to

promote bone healing. While we did not specifically assess bone healing at the insertion

histologically, such healing may have contributed to the increase in attachment strength

observed.

118

4.4 Cell Tracking

The ubiquitous green fluorescent protein (GFP) expression allows the identification of the

transplanted endothelial progenitor cells in regular non-GFP expressing Fischer 344 rats. The

GFP positive endothelial progenitor cells were seen at all cell tracking time points, suggesting

survival in the host environment without adverse immunogenic responses. Endothelial

incorporation was not observed on day 3, but was seen at one, two, and four weeks. This is

consistent with reports in the literature that peak capillary proliferation was seen at seven days

post-surgery 54

. However, most of the GFP positive endothelial progenitor cells were stray cells

in the scar tissue rather than incorporating into endothelium, suggesting that their activity may be

limited to paracrine effects, such as secreting molecular factors beneficial to the repair process.

More studies are needed to delineate the fate of transplanted endothelial progenitor cells in

rotator cuff tendon healing.

Gulotta et al used the metabolism of β-galactosidase to determine the presence of transplanted

mesenchymal stem cells (MSCs) at the rotator cuff tendon repair site 187

. While the metabolism

of β-galactosidase indicated that the cells were alive and metabolically active at the site of injury,

neither the cells’ location nor morphology could be determined using such method. Our detection

of transplanted endothelial progenitor cells via immunohistochemistry allowed us to identify and

locate the cells in the shoulder sections. It also revealed cell fate (incorporation into the

endothelium) after transplantation.

119

4.5 Limitations

The study has several limitations that require consideration. As previously mentioned, several

potential sources of error are inherent to the biomechanical testing. While the testing was

performed in a consistent manner, variability between specimens was inevitable. This is

especially the case when mounting the specimens (gauge length) and measuring the attachment

area. The distribution glue between the sand paper and the tendons, the time needed for the glue

to dry, and the penetration of glue into the tissues are examples of potential variability. The

graphs of the biomechanical test results were not “clean” load to failure curves, making stiffness

calculations problematic. This likely contributed to the variability in stiffness and ultimate stress,

although the variability observed in this study was actually quite low. It is also important to note

that a single load-to-failure test construct used in this study (and others in the literature) does not

represent what the clinical population experiences on a daily basis. Continuous cyclic loading

may be a better testing regime for replicating tendon overuse.

While the shoulder sections were made by experienced technicians and the sample orientation

and depth were examined by the pathologist during the cut to the best of their abilities, these

variables could have affected our histological data. The histology serves as supplementary results

to support our biomechanical findings. Thus, the histology by itself cannot provide conclusive

results of rotator cuff healing. In spite of our biomechanical finding, our histology methods may

not have been sensitive enough to detect differences in the healing tissues.

In this study, a single dose of one million endothelial progenitor cells was used for

transplantation. This number was chosen in accordance to previous studies of cell therapy for

120

rotator cuff repairs 63, 64, 65, 66

. Varying numbers of cells were not tested. It is likely that more than

one million cells are required to result in detectable histological differences. However, it is

expected that at some point the aggregation of transplanted cells will become a burden to

healing. A dose-response curve will be crucial to determine the optimal amount of cells before

becoming detrimental to tendon healing.

Another potential limitation of this study is that the model used healthy young rats. This may not

be representative of the older population who often experiences rotator cuff tears. Compromised

healing rat models should be investigated in future studies. These may include diabetic models,

steroid or collagenase injections to simulate tendon degeneration, and induced overuse via

activity. Aged rats are expected to show slower healing due to their diminished cellular activities.

The augmentation of blood supply in these conditions will perhaps have a greater effect in

healing and result in further statistical significance.

121

4.6 Clinical Relevance

We found the transplantation of autologous endothelial progenitor cells during rotator cuff repair

to result in improved biomechanical strength without adverse effects or immunogenic responses.

The difference in the ultimate load to failure between the EPC group (24.64 N) and the Fibrin

group (19.09 N) is 5.55 N, a 29 % increase from the Fibrin group. This improvement in insertion

strength is considered clinically significant as we previously defined clinical significance to be

20 % or more, in agreement with the literature 15, 16, 34, 62, 63, 73

. Although it is unfortunate that we

cannot directly link the difference in insertion strength to retear rates in a clinical setting, we can

associate stronger attachment with better tendon-bone integration, and a higher threshold for re-

rupture. While the regeneration of the fibrocartilaginous transitional zone has been regarded as

the hallmark for the regeneration of native tendon-bone insertion 34, 62, 64, 79, 80

, we found

attachment strength increase to be independent of fibrocartilage formation at the insertion site in

this study.

122

5 Conclusion and Future Directions

5.1 Conclusion

The delivery of endothelial progenitor cells during rotator cuff repair improves the attachment

strength at the supraspinatus insertion six weeks after the surgery without detrimental effects to

the tendinous tissue. Further studies are required to gain a better understanding of the multiple

facets of tendon-bone healing and to ultimately lead to effective biological therapies for rotator

cuff repair.

5.2 Future Directions

Attributing a single factor or cause to rotator cuff injury is difficult due to the large number of

pathogenic factors that may come into play. Targeting each pathogenic condition is necessary for

individualized medical treatment. Similarly, while rotator cuff tears can be treated surgically or

managed non-operatively, the post-operation treatments will need to address the various facets of

the condition for the rotator cuff healing to be optimized.

123

In this experiment, a cellular strategy for vascular enhancement was investigated. Endothelial

progenitor cells have been shown this study, and many others, to have the ability to induce and

enhance localized vascular growth in injured tissues. The cells are appropriate candidates for

therapeutic angiogenesis.

5.2.1 Identifying the Angiogenic Mechanisms

The current study sought to address the functional outcome via biomechanical testing. The load

to failure represents the strength that the healing tendinous structure can sustain before it

detaches completely. Biomechanical testing is the most clinically representative outcome, as

functional strength is a well-defined clinical assessment for rotator cuff injuries. In this study, the

histomorphometric analyses provide only supplementary confirmation of tissue healing.

Histology generates pictures for the healing phases, and histological grading can be related to

biomechanical functions 147

.

The study draws connections between the transplanted cells and the outcomes. The cellular

mechanism by which vessel growth is promoted by the endothelial progenitor cell population is

still yet to be fully delineated. Possible mechanisms include secretion of angiogenic growth

factors such as vascular endothelial growth factors, angiopoeitin-1, and platelet derived growth

factor, and the recently proposed mechanism of secreting vesicles containing miRNA, and

differentiation into endothelial cells to aid the expansion of the local vascular network.

124

The factors responsible for the angiogenic mechanisms could be confirmed by genetically

modifying the endothelial progenitor cells to enhance the synthesis of angiogenic molecules,

such as vascular endothelial growth factors, angiopoeitin-1, and platelet derived growth factor.

Pharmaceutical agents, alternatively, could be employed to block the activity or production of

angiogenic molecules to more precisely define the downstream effect of the factors at play.

5.2.2 The Mesenchymal Lineage

The transition zone helps transfer mechanical forces from the tendon to the bone by minimizing

stress concentration at the interface. Accordingly, for optimal healing at the tendon-bone

interface, the regeneration of this specialized tendon-bone transition zone must occur. In addition

to meeting the metabolic and vascular demands, this healing process requires the proliferation

and hypertrophy of chondrocytes at the interface, proper mineral deposition in the extra-cellular

matrices, and remodeling of collagenous structures.

The current study focused on the augmentation of vascular supply in hopes that this would set

stage for the following healing and remodeling processes. Indeed, meeting the local metabolic

demands of the tissue is crucial for optimizing tendon-bone healing. However, fibrocartilage

chrondrogenesis and structural remodeling are other facets of the healing that could be targeted.

In the future, a mixture of endothelial progenitor cells and mesenchymal stem cells could be

investigated. Both cell types are found in the bone marrow mononuclear cell population. The

functions and effects of the endothelial progenitor cells are described previously. The

125

mesenchymal stem cells can differentiate into the various stromal cell lineages that aid the repair

of musculoskeletal tissues; osteoblasts, chondrocytes, and tenocytes/fibroblasts that are essential

for the production of extracellular matrix of the tendon-bone junction. This is where blood

supply plays a crucial role in meeting the metabolic demands of these active cells. As previously

mentioned, the fact that autologous transplantation of mesenchymal stem cells alone did not

yield superior healing 63

is possibly due to the discrepancy between nutrient supply and

metabolic demand at the repair site. With the presence of endothelial progenitor cells, the

vascularized tendon-bone interface would possibly support the high metabolic requirement of

active cells at the repair site.

While the mononuclear cell population has already been employed in clinical studies, endothelial

progenitor cells constitute only a small fraction of the population 46

. Given the results of this

study, it would reasonable to investigate an endothelial progenitor cell-enriched as well as

mesenchymal stem cell-enriched mononuclear population in rotator cuff tendon healing after

surgery.

5.2.3 Clinical Application

Moving forward, a dose-response relationship between the quantity of endothelial progenitor

cells and the healing/angiogenic response needs to be established. The study was done in an

animal model in which the rats received one million endothelial progenitor cells for their

treatment. Whether the positive effect can be seen in a much larger scale in that of a human

rotator cuff is still unknown. A recent study of mesenchymal stem cell therapy in human rotator

126

cuff repair has shown successful healing with an average of 51 000 cells given to each patient 212

compared to the one million mesenchymal stem cells given in previous studies. Various amounts

of endothelial progenitor cells need to be transplanted to determine a curve that will indicate the

appropriate dose of endothelial progenitor cells to administer to the patients.

The cells are rare and their phenotype alters during in vitro expansion, and thus finding the

optimal number cells for transplantation requires ensuring the quality of the cells as well as the

quantity. Accordingly, it is crucial to standardize and optimize methods to rapidly and efficiently

acquire sufficient endothelial progenitor cells from the bone marrow with or without in vitro

culturing for use during the surgery. Maintaining early endothelial progenitor cell phenotypes

may be key to higher levels of angiogenic secretions. In addition, investigations of multiple

deliveries over the span of the recovery period could further determine the best treatment regime.

While the first dose of endothelial progenitor cells could be given during the surgery, subsequent

doses would have to be administered via local injection or intravenous delivery. Accordingly, the

method of cell delivery is also a topic that requires further investigation.

Most importantly, not all rotator cuff injuries should be treated the same. Identifying the patients

who would be most likely to benefit from endothelial progenitor cell transplantation during

rotator cuff repair is paramount. Older populations suffering from tendon degeneration are more

likely to benefit from the addition of endothelial progenitor cells than the young, athletic

population with traumatic rotator cuff injury. Comparisons of healing potential between aged rat,

diabetic rats, and young mature rats will confirm the possible different responses to endothelial

progenitor cell therapy.

The use of endothelial progenitor cells in surgical settings to aid the healing of rotator cuff

tendon is feasible as the cell population can be extracted from autologous sources: the bone

127

marrow and the blood. Employing adult stem cell-like progenitor cells for rotator cuff surgeries

bypasses the many ethical issues and technical issues that surround the use of embryonic stem

cells and induced pluripotent stem cells. Nonetheless, the cost-efficiency of such procedure

would need to be addressed.

128

References

1. Abrahamsson SO: Similar effects of recombinant human insulin-like growth factor-I and II on

cellular activities in flexor tendons of young rabbits: experimental studies in vitro. Journal of

Orthopaedic Research. 1997;15(2):256-262.

2. Abrahamsson SO, Lohmander S: Differential effects of insulin-like growth factor-I on matrix

and DNA synthesis in various regions and types of rabbit tendons. Journal of Orthopaedic

Research. 1996;14(3):370-376.

3. Adams J, Zobitz M, Reach J, et al: Rotator cuff repair using an acellular dermal matrix graft:

an in vivo study in a canine model. Arthroscopy. 2006;22(7):700-709.

4. Adler RS, Fealy S, Rudzki JR, et al: Rotator cuff in asymptomatic volunteers: contrast-

enhanced US depiction of intratendinous and peritendinous vascularity. Radiology.

2008;248(3):954-961.

5. Al-Khaldi A, Eliopoulos N, Martineau D, et al: Postnatal bone marrow stromal cells elicit a

potent VEGF-dependent neoangiogenic response in vivo. Gene Ther. 2003;10(8):621-629.

6. Asahara T, Murohara T, Sullivan A, et al: Isolation of putative progenitor endothelial cells for

angiogenesis. Science. 1997;275(5302):964-967.

7. Assmus B, Honold J, Schachinger V, et al: Transcoronary transplantation of progenitor cells

after myocardial infarction. N Engl J Med. 2006;355(12):1222-1232.

8. Assmus B, Schachinger V, Teupe C, et al: Transplantation of Progenitor Cells and

Regeneration Enhancement in Acute Myocardial Infarction (TOPCARE-AMI). Circulation.

2002;106(24):3009-3017.

9. Atesok K, Li R, Stewart D, et al: Endothelial progenitor cells promote fracture healing in a

segmental bone defect model. Journal of Orthopaedic Research. 2010;28(8):1007-1014.

10. Aviezer D, Hecht D, Safran M, et al: Perlecan, basal lamina proteoglycan, promotes basic

fibroblast growth factor-receptor binding, mitogenesis, and angiogenesis. Cell. 1994;79(6):1005-

1013.

11. Bailey AS, Willenbring H, Jiang S, et al: Myeloid lineage progenitors give rise to vascular

endothelium. Proc Natl Acad Sci U S A. 2006;103(35):13156-13161.

12. Barber FA, Hrnack S, Snyder S, et al: Rotator cuff repair healing influenced by platelet-rich

plasma construct augmentation. Arthroscopy. 2011;27(8):1029-1035.

129

13. Baumgarten KM, Gerlach D, Galatz LM, et al: Cigarette smoking increases the risk for

rotator cuff tears. Clin Orthop Relat Res. 2010;468(6):1534-1541.

14. Beck J, Evans D, Tonino P, et al: The Biomechanical and Histologic Effects of Platelet-Rich

Plasma on Rat Rotator Cuff Repairs. Am J Sports Med. 2012;40(9):2037-2044.

15. Bedi A, Fox AJ, Kovacevic D, et al: Doxycycline-mediated inhibition of matrix

metalloproteinases improves healing after rotator cuff repair. Am J Sports Med. 2010;38(2):308-

317.

16. Bedi A, Fox AJS, Harris P, et al: Diabetes mellitus impairs tendon-bone healing after rotator

cuff repair. Journal of shoulder and elbow surgery. 2010;19(7):978-988.

17. Bedi A, Kovacevic D, Hettrich C, et al: The effect of matrix metalloproteinase inhibition on

tendon-to-bone healing in a rotator cuff repair model. Journal of shoulder and elbow surgery.

2010;19(3):384-391.

18. Beitnes JO, Hopp E, Lunde K, et al: Long-term results after intracoronary injection of

autologous mononuclear bone marrow cells in acute myocardial infarction: the ASTAMI

randomised, controlled study. Heart. 2009;95(24):1983-1989.

19. Beltrami AP, Barlucchi L, Torella D, et al: Adult cardiac stem cells are multipotent and

support myocardial regeneration. Cell. 2003;114(6):763-776.

20. Beredjiklian P, Favata M, Cartmell J, et al: Regenerative versus reparative healing in tendon:

a study of biomechanical and histological properties in fetal sheep. Ann Biomed Eng.

2003;31(10):1143-1152.

21. Bergeson A, Tashjian R, Greis P, et al: Effects of Platelet-Rich Fibrin Matrix on Repair

Integrity of At-Risk Rotator Cuff Tears. Am J Sports Med. 2011.

22. Bompais H, Chagraoui J, Canron X, et al: Human endothelial cells derived from circulating

progenitors display specific functional properties compared with mature vessel wall endothelial

cells. Blood. 2004;103(7):2577-2584.

23. Bragdon B, Moseychuk O, Saldanha S, et al: Bone morphogenetic proteins: a critical review.

Cell Signal. 2011;23(4):609-620.

24. Brigstock DR: The connective tissue growth factor/cysteine-rich 61/nephroblastoma

overexpressed (CCN) family. Endocr Rev. 1999;20(2):189-206.

25. Brigstock D: Regulation of angiogenesis and endothelial cell function by connective tissue

growth factor (CTGF) and cysteine-rich 61 (CYR61). Angiogenesis. 2002;5(3):153-165.

130

26. Cadet E, Vorys G, Rahman R, et al: Improving bone density at the rotator cuff footprint

increases supraspinatus tendon failure stress in a rat model. Journal of Orthopaedic Research.

2010;28(3):308-314.

27. Castricini R, Longo U, De Benedetto M, et al: Platelet-rich plasma augmentation for

arthroscopic rotator cuff repair: a randomized controlled trial. Am J Sports Med. 2011;39(2):258-

265.

28. Chan BP, Chan KM, Maffulli N, et al: Effect of basic fibroblast growth factor. An in vitro

study of tendon healing. Clin Orthop. 1997(342):239-247.

29. Chan K, Fu S, Wong Y, et al: Expression of transforming growth factor beta isoforms and

their roles in tendon healing. Wound repair and regeneration. 2008;16(3):399-407.

30. Chang J, Thunder R, Most D, et al: Studies in flexor tendon wound healing: neutralizing

antibody to TGF-beta1 increases postoperative range of motion. Plast Reconstr Surg.

2000;105(1):148-155.

31. Chen C, Cao Y, Wu Y, et al: Tendon healing in vivo: gene expression and production of

multiple growth factors in early tendon healing period. J Hand Surg. 2008;33(10):1834-1842.

32. Chen D, Zhao M, Mundy G: Bone morphogenetic proteins. Growth Factors. 2004;22(4):233-

241.

33. Cheung E, Silverio L, Sperling J: Strategies in biologic augmentation of rotator cuff repair: a

review. Clin Orthop. 2010;468(6):1476-1484.

34. Cohen DB, Kawamura S, Ehteshami JR, et al: Indomethacin and celecoxib impair rotator

cuff tendon-to-bone healing. Am J Sports Med. 2006;34(3):362-369.

35. Crabtree B, Subramanian V, Sato JD: Behavior of endothelial cells on Matrigel and

development of a method for a rapid and reproducible in vitro angiogenesis assay. In Vitro

Cellular & Developmental Biology - Animal. 2007;43(2):87-94.

36. Crosby JR, Kaminski WE, Schatteman G, et al: Endothelial cells of hematopoietic origin

make a significant contribution to adult blood vessel formation. Circ Res. 2000;87(9):728-730.

37. Dahlgren L, Mohammed H, Nixon A: Temporal expression of growth factors and matrix

molecules in healing tendon lesions. Journal of Orthopaedic Research. 2005;23(1):84-92.

38. De Spiegelaere W, Casteleyn C, Van den Broeck W, et al: Intussusceptive angiogenesis: a

biologically relevant form of angiogenesis. J Vasc Res. 2012;49(5):390-404.

39. Del Buono A, Oliva F, Longo UG, et al: Metalloproteases and rotator cuff disease. J

Shoulder Elbow Surg. 2012;21(2):200-208.

131

40. DeLee JD, Drez D, Miller MD: in DeLee & Drez's Orthopaedic Sports MNedicine:

Priniciples and Practice 2nd Edition. Philadelphia, Saunders Elsevier, 2002, pp 2624.

41. DeOrio JK, Cofield RH: Results of a second attempt at surgical repair of a failed initial

rotator-cuff repair. J Bone Joint Surg Am. 1984;66(4):563-567.

42. Derwin KA, Baker AR, Codsi MJ, et al: Assessment of the canine model of rotator cuff

injury and repair. J Shoulder Elbow Surg. 2007;16(5 Suppl):S140-8.

43. Determe D, Rongieres M, Kany J, et al: Anatomic study of the tendinous rotator cuff of the

shoulder. Surg Radiol Anat. 1996;18(3):195-200.

44. Donovan A, Schweitzer M, Bencardino J, et al: Correlation between Rotator Cuff Tears and

Systemic Atherosclerotic Disease. Radiol Res Pract. 2011;2011:128353.

45. Duquin T, Buyea C, Bisson L: Which Method of Rotator Cuff Repair Leads to the Highest

Rate of Structural Healing? Am J Sports Med. 2010;38(4):835-841.

46. Ellera Gomes JL, da Silva RC, Silla LM, et al: Conventional rotator cuff repair

complemented by the aid of mononuclear autologous stem cells. Knee Surg Sports Traumatol

Arthrosc. 2012;20(2):373-377.

47. Ellman HMD: Diagnosis and Treatment of Incomplete Rotator Cuff Tears. Clinical

Orthopaedics & Related Research. 1990;254:64-74.

48. Elsheikh E, Uzunel M, He Z, et al: Only a specific subset of human peripheral-blood

monocytes has endothelial-like functional capacity. Blood. 2005;106(7):2347-2355.

49. Erbs S, Linke A, Schachinger V, et al: Restoration of microvascular function in the infarct-

related artery by intracoronary transplantation of bone marrow progenitor cells in patients with

acute myocardial infarction: the Doppler Substudy of the Reinfusion of Enriched Progenitor

Cells and Infarct Remodeling in Acute Myocardial Infarction (REPAIR-AMI) trial. Circulation.

2007;116(4):366-374.

50. Fealy S, Adler R, Drakos M, et al: Patterns of vascular and anatomical response after rotator

cuff repair. Am J Sports Med. 2006;34(1):120-127.

51. Frey E, Regenfelder F, Sussmann P, et al: Adipogenic and myogenic gene expression in

rotator cuff muscle of the sheep after tendon tear. J Orthop Res. 2009;27(4):504-509.

52. Fu S, Wong Y, Chan B, et al: The roles of bone morphogenetic protein (BMP) 12 in

stimulating the proliferation and matrix production of human patellar tendon fibroblasts. Life

Sci. 2003;72(26):2965-2974.

132

53. Funakoshi T, Iwasaki N, Kamishima T, et al: In vivo vascularity alterations in repaired

rotator cuffs determined by contrast-enhanced ultrasound. Am J Sports Med. 2011;39(12):2640-

2646.

54. Galatz L, Sandell L, Rothermich S, et al: Characteristics of the rat supraspinatus tendon

during tendon-to-bone healing after acute injury. Journal of Orthopaedic Research.

2006;24(3):541-550.

55. Gamradt SC, Gallo RA, Adler RS, et al: Vascularity of the supraspinatus tendon three

months after repair: characterization using contrast-enhanced ultrasound. J Shoulder Elbow Surg.

2010;19(1):73-80.

56. Gangji V, De Maertelaer V, Hauzeur J: Autologous bone marrow cell implantation in the

treatment of non-traumatic osteonecrosis of the femoral head: Five year follow-up of a

prospective controlled study. Bone. 2011;49(5):1005-1009.

57. Garofalo R, Cesari E, Vinci E, et al: Role of metalloproteinases in rotator cuff tear. Sports

Med Arthrosc. 2011;19(3):207-212.

58. Garofalo R, Cesari E, Vinci E, et al: Role of metalloproteinases in rotator cuff tear. Sports

medicine and arthroscopy review. 2011;19(3):207-212.

59. Gerber C, Meyer D, Von Rechenberg B, et al: Rotator Cuff Muscles Lose Responsiveness to

Anabolic Steroids After Tendon Tear and Musculotendinous Retraction: An Experimental Study

in Sheep. Am J Sports Med. 2012;40(11):2454-2461.

60. Giuseppe Longo U, Franceschi F, Ruzzini L, et al: Histopathology of the Supraspinatus

Tendon in Rotator Cuff Tears. Am J Sports Med. 2008;36(3):533-538.

61. Gladstone JN, Bishop JY, Lo IK, et al: Fatty infiltration and atrophy of the rotator cuff do not

improve after rotator cuff repair and correlate with poor functional outcome. Am J Sports Med.

2007;35(5):719-728.

62. Gulotta LV, Kovacevic D, Cordasco F, et al: Evaluation of Tumor Necrosis Factor α

Blockade on Early Tendon-to-Bone Healing in a Rat Rotator Cuff Repair Model. Arthroscopy:

The Journal of Arthroscopic & Related Surgery. 2011;27(10):1351-1357.

63. Gulotta L, Kovacevic D, Ehteshami J, et al: Application of bone marrow-derived

mesenchymal stem cells in a rotator cuff repair model. Am J Sports Med. 2009;37(11):2126-

2133.

64. Gulotta L, Kovacevic D, Montgomery S, et al: Stem cells genetically modified with the

developmental gene MT1-MMP improve regeneration of the supraspinatus tendon-to-bone

insertion site. Am J Sports Med. 2010;38(7):1429-1437.

133

65. Gulotta L, Kovacevic D, Packer J, et al: Bone marrow-derived mesenchymal stem cells

transduced with scleraxis improve rotator cuff healing in a rat model. Am J Sports Med.

2011;39(6):1282-1289.

66. Gulotta L, Kovacevic D, Packer J, et al: Adenoviral-mediated gene transfer of human bone

morphogenetic protein-13 does not improve rotator cuff healing in a rat model. Am J Sports

Med. 2011;39(1):180-187.

67. Gumina S, Arceri V, Carbone S, et al: The association between arterial hypertension and

rotator cuff tear: the influence on rotator cuff tear sizes. J Shoulder Elbow Surg. 2012.

68. Gupta R, Lee TQ: Contributions of the different rabbit models to our understanding of rotator

cuff pathology. J Shoulder Elbow Surg. 2007;16(5 Suppl):S149-57.

69. Harvie P, Ostlere SJ, Teh J, et al: Genetic influences in the aetiology of tears of the rotator

cuff. Sibling risk of a full-thickness tear. J Bone Joint Surg Br. 2004;86(5):696-700.

70. Hee C, Dines J, Dines D, et al: Augmentation of a rotator cuff suture repair using rhPDGF-

BB and a type I bovine collagen matrix in an ovine model. Am J Sports Med. 2011;39(8):1630-

1639.

71. Hegedus EJ, Cook C, Brennan M, et al: Vascularity and tendon pathology in the rotator cuff:

a review of literature and implications for rehabilitation and surgery. Br J Sports Med.

2010;44(12):838-847.

72. Heinemeier KM, Bjerrum SS, Schjerling P, et al: Expression of extracellular matrix

components and related growth factors in human tendon and muscle after acute exercise.

Scandinavian journal of medicine science in sports. 2011.

73. Hettrich CM, Beamer BS, Bedi A, et al: The effect of rhPTH on the healing of tendon to

bone in a rat model. J Orthop Res. 2012;30(5):769-774.

74. Hirschi KK, Ingram DA, Yoder MC: Assessing identity, phenotype, and fate of endothelial

progenitor cells. Arterioscler Thromb Vasc Biol. 2008;28(9):1584-1595.

75. Hristov M, Erl W, Weber PC: Endothelial progenitor cells: mobilization, differentiation, and

homing. Arterioscler Thromb Vasc Biol. 2003;23(7):1185-1189.

76. Huang P, Li S, Han M, et al: Autologous transplantation of granulocyte colony-stimulating

factor-mobilized peripheral blood mononuclear cells improves critical limb ischemia in diabetes.

Diabetes Care. 2005;28(9):2155-2160.

134

77. Hur J, Yoon CH, Kim HS, et al: Characterization of two types of endothelial progenitor cells

and their different contributions to neovasculogenesis. Arterioscler Thromb Vasc Biol.

2004;24(2):288-293.

78. Iannotti J, Codsi M, Kwon Y, et al: Porcine small intestine submucosa augmentation of

surgical repair of chronic two-tendon rotator cuff tears. A randomized, controlled trial. Journal of

Bone and Joint Surgery American volume. 2006;88(6):1238-1244.

79. Ide J, Kikukawa K, Hirose J, et al: The effect of a local application of fibroblast growth

factor-2 on tendon-to-bone remodeling in rats with acute injury and repair of the supraspinatus

tendon. Journal of shoulder and elbow surgery. 2009;18(3):391-398.

80. Ide J, Kikukawa K, Hirose J, et al: The effects of fibroblast growth factor-2 on rotator cuff

reconstruction with acellular dermal matrix grafts. Arthroscopy. 2009;25(6):608-616.

81. Jo C, Kim J, Yoon K, et al: Does platelet-rich plasma accelerate recovery after rotator cuff

repair? A prospective cohort study. Am J Sports Med. 2011;39(10):2082-2090.

82. Jones KB, Seshadri T, Krantz R, et al: Cell-based therapies for osteonecrosis of the femoral

head. Biol Blood Marrow Transplant. 2008;14(10):1081-1087.

83. Kalka C, Masuda H, Takahashi T, et al: Transplantation of ex vivo expanded endothelial

progenitor cells for therapeutic neovascularization. Proc Natl Acad Sci U S A. 2000;97(7):3422-

3427.

84. Kawamoto A, Gwon HC, Iwaguro H, et al: Therapeutic potential of ex vivo expanded

endothelial progenitor cells for myocardial ischemia. Circulation. 2001;103(5):634-637.

85. Kawamoto A, Tkebuchava T, Yamaguchi J, et al: Intramyocardial transplantation of

autologous endothelial progenitor cells for therapeutic neovascularization of myocardial

ischemia. Circulation. 2003;107(3):461-468.

86. Kemp KA, Sheps DM, Luciak-Corea C, et al: Systematic review of rotator cuff tears in

workers' compensation patients. Occup Med (Lond). 2011;61(8):556-562.

87. Kim HM, Galatz L, Das R, et al: The role of transforming growth factor beta isoforms in

tendon-to-bone healing. Connect Tissue Res. 2011;52(2):87-98.

88. Klass BR, Rolfe KJ, Grobbelaar AO: In vitro flexor tendon cell response to TGF-beta1: a

gene expression study. J Hand Surg. 2009;34(3):495-503.

89. Klein M, Yalamanchi N, Pham H, et al: Flexor tendon healing in vitro: effects of TGF-beta

on tendon cell collagen production. J Hand Surg. 2002;27(4):615-620.

135

90. Kobayashi M, Itoi E, Minagawa H, et al: Expression of growth factors in the early phase of

supraspinatus tendon healing in rabbits. Journal of shoulder and elbow surgery. 2006;15(3):371-

377.

91. Kocher AA, Schuster MD, Szabolcs MJ, et al: Neovascularization of ischemic myocardium

by human bone-marrow-derived angioblasts prevents cardiomyocyte apoptosis, reduces

remodeling and improves cardiac function. Nat Med. 2001;7(4):430-436.

92. Koike Y, Trudel G, Uhthoff H: Formation of a new enthesis after attachment of the

supraspinatus tendon: A quantitative histologic study in rabbits. Journal of Orthopaedic

Research. 2005;23(6):1433-1440.

93. Kurtz CA, Loebig TG, Anderson DD, et al: Insulin-like growth factor I accelerates functional

recovery from Achilles tendon injury in a rat model. Am J Sports Med. 1999;27(3):363-369.

94. Lee C, Moioli E, Mao J: Fibroblastic differentiation of human mesenchymal stem cells using

connective tissue growth factor. IEEE Engineering in medicine and biology society conference

proceedings. 2006;1:775-778.

95. Leistner DM, Fischer-Rasokat U, Honold J, et al: Transplantation of progenitor cells and

regeneration enhancement in acute myocardial infarction (TOPCARE-AMI): final 5-year results

suggest long-term safety and efficacy. Clin Res Cardiol. 2011;100(10):925-934.

96. Lewis JS: Rotator cuff tendinopathy. Br J Sports Med. 2009;43(4):236-241.

97. Li R, Nauth A, Li C, et al: Expression of VEGF Gene Isoforms in a Rat Segmental Bone

Defect Model Treated With EPCs. J Orthop Trauma. 2012;26(12):689-692.

98. Li R, Stewart D, von Schroeder H, et al: Effect of cell-based VEGF gene therapy on healing

of a segmental bone defect. Journal of Orthopaedic Research. 2009;27(1):8-14.

99. Lin Y, Weisdorf DJ, Solovey A, et al: Origins of circulating endothelial cells and endothelial

outgrowth from blood. J Clin Invest. 2000;105(1):71-77.

100. Lohr JF, Uhthoff HK: The microvascular pattern of the supraspinatus tendon. Clin Orthop

Relat Res. 1990;(254)(254):35-38.

101. Longo UG, Franceschi F, Spiezia F, et al: Triglycerides and total serum cholesterol in

rotator cuff tears: do they matter? Br J Sports Med. 2010;44(13):948-951.

102. Longo UG, Berton A, Khan WS, et al: Histopathology of Rotator Cuff Tears. Sports

Medicine & Arthroscopy Review. 2011;19(3):227-236.

136

103. Loomans CJ, Wan H, de Crom R, et al: Angiogenic murine endothelial progenitor cells are

derived from a myeloid bone marrow fraction and can be identified by endothelial NO synthase

expression. Arterioscler Thromb Vasc Biol. 2006;26(8):1760-1767.

104. Luft F: CCN2, the connective tissue growth factor. Journal of molecular medicine.

2008;86(1):1-3.

105. Lunde K, Solheim S, Aakhus S, et al: Autologous stem cell transplantation in acute

myocardial infarction: The ASTAMI randomized controlled trial. Intracoronary transplantation

of autologous mononuclear bone marrow cells, study design and safety aspects. Scandinavian

Cardiovascular Journal. 2005;39(3):150-158.

106. Macchiarelli G, Jiang JY, Nottola SA, et al: Morphological patterns of angiogenesis in

ovarian follicle capillary networks. A scanning electron microscopy study of corrosion cast.

Microsc Res Tech. 2006;69(6):459-468.

107. Maffulli N, Longo UG, Berton A, et al: Biological factors in the pathogenesis of rotator cuff

tears. Sports Med Arthrosc. 2011;19(3):194-201.

108. Majka SM, Jackson KA, Kienstra KA, et al: Distinct progenitor populations in skeletal

muscle are bone marrow derived and exhibit different cell fates during vascular regeneration. J

Clin Invest. 2003;111(1):71-79.

109. Makanya A, Hlushchuk R, Djonov V: Intussusceptive angiogenesis and its role in vascular

morphogenesis, patterning, and remodeling. Angiogenesis. 2009;12(2):113-123.

110. Maniscalco P, Gambera D, Lunati A, et al: The "Cascade" membrane: a new PRP device for

tendon ruptures. Description and case report on rotator cuff tendon. Acta bio-medica de L'Ateneo

parmense. 2008;79(3):223-226.

111. Manning C, Kim HM, Sakiyama Elbert S, et al: Sustained delivery of transforming growth

factor beta three enhances tendon-to-bone healing in a rat model. Journal of Orthopaedic

Research. 2011;29(7):1099-1105.

112. Marx RE: Platelet-rich plasma (PRP): what is PRP and what is not PRP? Implant Dent.

2001;10(4):225-228.

113. McCallum SD, Paoloni JA, Murrell GA: Five-year prospective comparison study of topical

glyceryl trinitrate treatment of chronic lateral epicondylosis at the elbow. Br J Sports Med.

2011;45(5):416-420.

114. Meyer GP, Wollert KC, Lotz J, et al: Intracoronary bone marrow cell transfer after

myocardial infarction: 5-year follow-up from the randomized-controlled BOOST trial. Eur Heart

J. 2009;30(24):2978-2984.

137

115. Meyer GP, Wollert KC, Lotz J, et al: Intracoronary bone marrow cell transfer after

myocardial infarction: eighteen months' follow-up data from the randomized, controlled BOOST

(BOne marrOw transfer to enhance ST-elevation infarct regeneration) trial. Circulation.

2006;113(10):1287-1294.

116. Millar NL, Wei AQ, Molloy TJ, et al: Cytokines and apoptosis in supraspinatus

tendinopathy. J Bone Joint Surg Br. 2009;91(3):417-424.

117. Miller B, Downie B, Kohen R, et al: When Do Rotator Cuff Repairs Fail? Serial Ultrasound

Examination After Arthroscopic Repair of Large and Massive Rotator Cuff Tears. Am J Sports

Med. 2011;39(10):2064-2070.

118. Moussad EE, Brigstock DR: Connective tissue growth factor: what's in a name? Mol Genet

Metab. 2000;71(1-2):276-292.

119. Mukai N, Akahori T, Komaki M, et al: A comparison of the tube forming potentials of early

and late endothelial progenitor cells. Exp Cell Res. 2008;314(3):430-440.

120. Nakamura T, Torimura T, Sakamoto M, et al: Significance and Therapeutic Potential of

Endothelial Progenitor Cell Transplantation in a Cirrhotic Liver Rat Model. Gastroenterology.

2007;133(1):91-e1.

121. Nakanishi T, Nishida T, Shimo T, et al: Effects of CTGF/Hcs24, a product of a

hypertrophic chondrocyte-specific gene, on the proliferation and differentiation of chondrocytes

in culture. Endocrinology. 2000;141(1):264-273.

122. Nakase T, Sugamoto K, Miyamoto T, et al: Activation of cartilage-derived morphogenetic

protein-1 in torn rotator cuff. Clin Orthop. 2002(399):140-145.

123. Naruse K, Hamada Y, Nakashima E, et al: Therapeutic neovascularization using cord

blood-derived endothelial progenitor cells for diabetic neuropathy. Diabetes. 2005;54(6):1823-

1828.

124. Nauth A, Li R, Schemitsch EH: Endothelial progenitor cells for fracture healing and

angiogenesis: A comparison with Mesenchymal Stem Cells. Annual Meeting of the American

Academy of Orthopaedic Surgeons, Las Vegas, NV. 2010.

125. Neer CS,2nd: Anterior acromioplasty for the chronic impingement syndrome in the

shoulder: a preliminary report. J Bone Joint Surg Am. 1972;54(1):41-50.

126. Nho SJ, Yadav HBS, Shindle MK, et al: Rotator Cuff Degeneration: Etiology and

Pathogenesis. Am J Sports Med. 2008;36(5):987-993.

138

127. Oliva F, Via A, Maffulli N: Role of growth factors in rotator cuff healing. Sports medicine

and arthroscopy review. 2011;19(3):218-226.

128. Oswald J, Boxberger S, Jorgensen B, et al: Mesenchymal stem cells can be differentiated

into endothelial cells in vitro. Stem Cells. 2004;22(3):377-384.

129. Paoloni JA, Appleyard RC, Nelson J, et al: Topical glyceryl trinitrate application in the

treatment of chronic supraspinatus tendinopathy: a randomized, double-blinded, placebo-

controlled clinical trial. Am J Sports Med. 2005;33(6):806-813.

130. Paoloni JA, Appleyard RC, Nelson J, et al: Topical glyceryl trinitrate treatment of chronic

noninsertional achilles tendinopathy. A randomized, double-blind, placebo-controlled trial. J

Bone Joint Surg Am. 2004;86-A(5):916-922.

131. Paoloni JA, Murrell GA: Three-year followup study of topical glyceryl trinitrate treatment

of chronic noninsertional Achilles tendinopathy. Foot Ankle Int. 2007;28(10):1064-1068.

132. Pasternak B, Aspenberg P: Metalloproteinases and their inhibitors-diagnostic and

therapeutic opportunities in orthopedics. Acta orthopaedica. 2009;80(6):693-703.

133. Pauly S, Klatte F, Strobel C, et al: BMP-2 and BMP-7 affect human rotator cuff tendon

cells in vitro. J Shoulder Elbow Surg. 2012;21(4):464-473.

134. Peichev M, Naiyer AJ, Pereira D, et al: Expression of VEGFR-2 and AC133 by circulating

human CD34(+) cells identifies a population of functional endothelial precursors. Blood.

2000;95(3):952-958.

135. Pelosi E, Castelli G, Martin-Padura I, et al: Human Haemato-Endothelial Precursors: Cord

Blood CD34+ Cells Produce Haemogenic Endothelium. PLoS One. 2012;7(12):e51109.

136. Perin EC, Willerson JT, Pepine CJ, et al: Effect of transendocardial delivery of autologous

bone marrow mononuclear cells on functional capacity, left ventricular function, and perfusion in

chronic heart failure: the FOCUS-CCTRN trial. JAMA. 2012;307(16):1717-1726.

137. Planat-Benard V, Silvestre JS, Cousin B, et al: Plasticity of human adipose lineage cells

toward endothelial cells: physiological and therapeutic perspectives. Circulation.

2004;109(5):656-663.

138. Potente M, Gerhardt H, Carmeliet P: Basic and Therapeutic Aspects of Angiogenesis. Cell.

2011;146(6):873-887.

139. Pujol BF, Lucibello FC, Gehling UM, et al: Endothelial-like cells derived from human

CD14 positive monocytes. Differentiation. 2000;65(5):287-300.

139

140. Quirici N, Soligo D, Caneva L, et al: Differentiation and expansion of endothelial cells from

human bone marrow CD133+ cells. Br J Haematol. 2001;115(1):186-194.

141. Randelli P, Arrigoni P, Cabitza P, et al: Autologous platelet rich plasma for arthroscopic

rotator cuff repair. A pilot study. Disabil Rehabil. 2008;30(20-22):1584-1589.

142. Rathbun JB, Macnab I: The microvascular pattern of the rotator cuff. J Bone Joint Surg Br.

1970;52(3):540-553.

143. Reyes M, Lund T, Lenvik T, et al: Purification and ex vivo expansion of postnatal human

marrow mesodermal progenitor cells. Blood. 2001;98(9):2615-2625.

144. Riley G: Chronic tendon pathology: molecular basis and therapeutic implications. Expert

Reviews in Molecular Medicine. 2005;7(05):1.

145. Rodeo SA, Delos D, Williams RJ, et al: The effect of platelet-rich fibrin matrix on rotator

cuff tendon healing: a prospective, randomized clinical study. Am J Sports Med.

2012;40(6):1234-1241.

146. Rodeo S, Potter H, Kawamura S, et al: Biologic augmentation of rotator cuff tendon-healing

with use of a mixture of osteoinductive growth factors. Journal of Bone and Joint Surgery

American volume. 2007;89(11):2485-2497.

147. Rosenbaum AJ, Wicker JF, Dines JS, et al: Histologic stages of healing correlate with

restoration of tensile strength in a model of experimental tendon repair. HSS J. 2010;6(2):164-

170.

148. Rothman RH, Parke WW: The vascular anatomy of the rotator cuff. Clin Orthop Relat Res.

1965;41:176-186.

149. Rudzki JR, Adler RS, Warren RF, et al: Contrast-enhanced ultrasound characterization of

the vascularity of the rotator cuff tendon: age- and activity-related changes in the intact

asymptomatic rotator cuff. J Shoulder Elbow Surg. 2008;17(1 Suppl):96S-100S.

150. Sainz J, Al Haj Zen A, Caligiuri G, et al: Isolation of "side population" progenitor cells

from healthy arteries of adult mice. Arterioscler Thromb Vasc Biol. 2006;26(2):281-286.

151. Santibaez J, Quintanilla M, Bernabeu C: TGF-β/TGF-β receptor system and its role in

physiological and pathological conditions. Clin Sci. 2011;121(6):233-251.

152. Schachinger V, Assmus B, Britten MB, et al: Transplantation of progenitor cells and

regeneration enhancement in acute myocardial infarction: final one-year results of the

TOPCARE-AMI Trial. J Am Coll Cardiol. 2004;44(8):1690-1699.

140

153. Schachinger V, Erbs S, Elsasser A, et al: Improved clinical outcome after intracoronary

administration of bone-marrow-derived progenitor cells in acute myocardial infarction: final 1-

year results of the REPAIR-AMI trial. Eur Heart J. 2006;27(23):2775-2783.

154. Schaeffeler C, Mueller D, Kirchhoff C, et al: Tears at the rotator cuff footprint: Prevalence

and imaging characteristics in 305 MR arthrograms of the shoulder. Eur Radiol.

2011;21(7):1477-1484.

155. Schatteman GC, Hanlon HD, Jiao C, et al: Blood-derived angioblasts accelerate blood-flow

restoration in diabetic mice. J Clin Invest. 2000;106(4):571-578.

156. Schmeisser A, GarlichsThese authors have contributed equally to this work.,Christoph D.,

Zhang H, et al: Monocytes coexpress endothelial and macrophagocytic lineage markers and form

cord-like structures in Matrigel® under angiogenic conditions. Cardiovasc Res. 2001;49(3):671-

680.

157. Schuh A, Liehn EA, Sasse A, et al: Transplantation of endothelial progenitor cells improves

neovascularization and left ventricular function after myocardial infarction in a rat model. Basic

Res Cardiol. 2008;103(1):69-77.

158. Sdersten F, Ekman S, Eloranta M, et al: Ultrastructural immunolocalization of cartilage

oligomeric matrix protein (COMP) in relation to collagen fibrils in the equine tendon. Matrix

biology. 2005;24(5):376-385.

159. Sdersten F, Ekman S, Niehoff A, et al: Ultrastructural immunolocalization of cartilage

oligomeric matrix protein, thrombospondin-4, and collagen fibril size in rodent achilles tendon in

relation to exercise. Connect Tissue Res. 2007;48(5):254-262.

160. Seeherman H, Archambault J, Rodeo S, et al: rhBMP-12 accelerates healing of rotator cuff

repairs in a sheep model. Journal of Bone and Joint Surgery American volume.

2008;90(10):2206-2219.

161. Shah M, Foreman DM, Ferguson MW: Neutralisation of TGF-beta 1 and TGF-beta 2 or

exogenous addition of TGF-beta 3 to cutaneous rat wounds reduces scarring. J Cell Sci.

1995;108(3):985-1002.

162. Sharma P, Maffulli N: Biology of tendon injury: healing, modeling and remodeling. Journal

of musculoskeletal neuronal interactions. 2006;6(2):181-190.

163. Shimo T, Nakanishi T, Nishida T, et al: Connective tissue growth factor induces the

proliferation, migration, and tube formation of vascular endothelial cells in vitro, and

angiogenesis in vivo. J Biochem. 1999;126(1):137-145.

141

164. Slukvin II, Vodyanik M: Endothelial origin of mesenchymal stem cells. Cell Cycle.

2011;10(9):1370-1373.

165. Smadja DM, Cornet A, Emmerich J, et al: Endothelial progenitor cells: characterization, in

vitro expansion, and prospects for autologous cell therapy. Cell Biol Toxicol. 2007;23(4):223-

239.

166. Sonnabend DH, Howlett CR, Young AA: Histological evaluation of repair of the rotator

cuff in a primate model. J Bone Joint Surg Br. 2010;92(4):586-594.

167. Sonnabend DH, Young AA: Comparative anatomy of the rotator cuff. J Bone Joint Surg Br.

2009;91(12):1632-1637.

168. Soo C, Beanes S, Hu F, et al: Ontogenetic transition in fetal wound transforming growth

factor-beta regulation correlates with collagen organization. The American journal of pathology.

2003;163(6):2459-2476.

169. Soslowsky LJ, Carpenter JE, DeBano CM, et al: Development and use of an animal model

for investigations on rotator cuff disease. Journal of Shoulder and Elbow Surgery.

1996;5(5):383-392.

170. Sporn M: The early history of TGF-beta, and a brief glimpse of its future. Cytokine growth

factor reviews. 2006;17(1-2):3-7.

171. Steunebrink M, Zwerver J, Brandsema R, et al: Topical glyceryl trinitrate treatment of

chronic patellar tendinopathy: a randomised, double-blind, placebo-controlled clinical trial. Br J

Sports Med. 2013;47(1):34-39.

172. Sugaya H, Maeda K, Matsuki K, et al: Repair integrity and functional outcome after

arthroscopic double-row rotator cuff repair. A prospective outcome study. Journal of Bone and

Joint Surgery American volume. 2007;89(5):953-960.

173. Sun Y, Feng Y, Zhang C, et al: Beneficial effect of autologous transplantation of

endothelial progenitor cells on steroid-induced femoral head osteonecrosis in rabbits. Cell

Transplant. 2011;20(2):233-243.

174. Takahasih S, Nakajima M, Kobayashi M, et al: Effect of recombinant basic fibroblast

growth factor (bFGF) on fibroblast-like cells from human rotator cuff tendon. Tohoku J Exp

Med. 2002;198(4):207-214.

175. Taniguchi E, Kin M, Torimura T, et al: Endothelial Progenitor Cell Transplantation

Improves the Survival Following Liver Injury in Mice. Gastroenterology. 2006;130(2):521-531.

142

176. Tateishi-Yuyama E, Matsubara H, Murohara T, et al: Therapeutic angiogenesis for patients

with limb ischaemia by autologous transplantation of bone-marrow cells: a pilot study and a

randomised controlled trial. Lancet. 2002;360(9331):427-435.

177. Taylor D, Petrera M, Hendry M, et al: A systematic review of the use of platelet-rich

plasma in sports medicine as a new treatment for tendon and ligament injuries. Clinical journal

of sport medicine. 2011;21(4):344-352.

178. Tei K, Matsumoto T, Mifune Y, et al: Administrations of peripheral blood CD34-positive

cells contribute to medial collateral ligament healing via vasculogenesis. Stem Cells.

2008;26(3):819-830.

179. Terayama H, Ishikawa M, Yasunaga Y, et al: Prevention of osteonecrosis by intravenous

administration of human peripheral blood-derived CD34-positive cells in a rat osteonecrosis

model. J Tissue Eng Regen Med. 2011;5(1):32-40.

180. Thomopoulos S, Genin GM, Galatz LM: The development and morphogenesis of the

tendon-to-bone insertion - what development can teach us about healing -. Journal of

musculoskeletal neuronal interactions. 2010;10(1):35-45.

181. Thomopoulos S, Hattersley G, Rosen V, et al: The localized expression of extracellular

matrix components in healing tendon insertion sites: an in situ hybridization study. Journal of

Orthopaedic Research. 2002;20(3):454-463.

182. Thomopoulos S, Williams GR, Soslowsky LJ: Tendon to bone healing: differences in

biomechanical, structural, and compositional properties due to a range of activity levels. J

Biomech Eng. 2003;125(1):106-113.

183. Thomopoulos S, Das R, Sakiyama Elbert S, et al: bFGF and PDGF-BB for tendon repair:

controlled release and biologic activity by tendon fibroblasts in vitro. Ann Biomed Eng.

2010;38(2):225-234.

184. Thomopoulos S, Das R, Silva M, et al: Enhanced flexor tendon healing through controlled

delivery of PDGF-BB. Journal of Orthopaedic Research. 2009;27(9):1209-1215.

185. Thomopoulos S, Harwood F, Silva M, et al: Effect of several growth factors on canine

flexor tendon fibroblast proliferation and collagen synthesis in vitro. J Hand Surg.

2005;30(3):441-447.

186. Thomopoulos S, Kim HM, Das R, et al: The effects of exogenous basic fibroblast growth

factor on intrasynovial flexor tendon healing in a canine model. Journal of Bone and Joint

Surgery American volume. 2010;92(13):2285-2293.

143

187. Thomopoulos S, Soslowsky LJ, Flanagan CL, et al: The effect of fibrin clot on healing rat

supraspinatus tendon defects. Journal of Shoulder and Elbow Surgery. 2002;11(3):239-247.

188. Thomopoulos S, Zaegel M, Das R, et al: PDGF-BB released in tendon repair using a novel

delivery system promotes cell proliferation and collagen remodeling. Journal of Orthopaedic

Research. 2007;25(10):1358-1368.

189. Timmermans F, Plum J, Yoder MC, et al: Endothelial progenitor cells: identity defined? J

Cell Mol Med. 2009;13(1):87-102.

190. Ucuzian AA, Gassman AA, East AT, et al: Molecular mediators of angiogenesis. J Burn

Care Res. 2010;31(1):158-175.

191. Ueno T, Nakamura T, Torimura T, et al: Angiogenic cell therapy for hepatic fibrosis. Med

Mol Morphol. 2006;39(1):16-21.

192. Uggen J, Dines J, Uggen C, et al: Tendon gene therapy modulates the local repair

environment in the shoulder. J Am Osteopath Assoc. 2005;105(1):20-21.

193. Urbich C, Heeschen C, Aicher A, et al: Relevance of monocytic features for

neovascularization capacity of circulating endothelial progenitor cells. Circulation.

2003;108(20):2511-2516.

194. Vinson EN, Helms CA, Higgins LD: Rim-rent tear of the rotator cuff: a common and easily

overlooked partial tear. AJR Am J Roentgenol. 2007;189(4):943-946.

195. Vitale MA, Vitale MG, Zivin JG, et al: Rotator cuff repair: An analysis of utility scores and

cost-effectiveness. Journal of Shoulder and Elbow Surgery. 2007;16(2):181-187.

196. Wang B, Sun W, Shi Z, et al: Treatment of nontraumatic osteonecrosis of the femoral head

with the implantation of core decompression and concentrated autologous bone marrow

containing mononuclear cells. Arch Orthop Trauma Surg. 2010;130(7):859-865.

197. Wang F, Murrell GA, Wang MX: Oxidative stress-induced c-Jun N-terminal kinase (JNK)

activation in tendon cells upregulates MMP1 mRNA and protein expression. J Orthop Res.

2007;25(3):378-389.

198. Wang X, Liu P, Xin K, et al: Tendon healing in vitro: bFGF gene transfer to tenocytes by

adeno-associated viral vectors promotes expression of collagen genes. J Hand Surg.

2005;30(6):1255-1261.

199. Widgerow A: Cellular/extracellular matrix cross-talk in scar evolution and control. Wound

repair and regeneration. 2011;19(2):117-133.

144

200. Wolfman NM, Hattersley G, Cox K, et al: Ectopic induction of tendon and ligament in rats

by growth and differentiation factors 5, 6, and 7, members of the TGF-beta gene family. J Clin

Invest. 1997;100(2):321-330.

201. Wollert KC, Meyer GP, Lotz J, et al: Intracoronary autologous bone-marrow cell transfer

after myocardial infarction: the BOOST randomised controlled clinical trial. Lancet.

2004;364(9429):141-148.

202. Worland RL, Lee D, Orozco CG, et al: Correlation of age, acromial morphology, and

rotator cuff tear pathology diagnosed by ultrasound in asymptomatic patients. J South Orthop

Assoc. 2003;12(1):23-26.

203. Wrgler-Hauri C, Dourte L, Baradet T, et al: Temporal expression of 8 growth factors in

tendon-to-bone healing in a rat supraspinatus model. Journal of shoulder and elbow surgery.

2007;16(5 Suppl):S198-S203.

204. Xu H, Sandor M, Qi S, et al: Implantation of a porcine acellular dermal graft in a primate

model of rotator cuff repair. Journal of Shoulder and Elbow Surgery. 2012;21(5):580-588.

205. Yamaguchi H, Suenaga N, Oizumi N, et al: Will preoperative atrophy and Fatty

degeneration of the shoulder muscles improve after rotator cuff repair in patients with massive

rotator cuff tears? Adv Orthop. 2012;2012:195876.

206. Yamasaki T, Yasunaga Y, Terayama H, et al: Transplantation of bone marrow mononuclear

cells enables simultaneous treatment with osteotomy for osteonecrosis of the bilateral femoral

head. Med Sci Monit. 2008;14(4):CS23-30.

207. Yin AH, Miraglia S, Zanjani ED, et al: AC133, a novel marker for human hematopoietic

stem and progenitor cells. Blood. 1997;90(12):5002-5012.

208. Yoon CH, Hur J, Park KW, et al: Synergistic neovascularization by mixed transplantation

of early endothelial progenitor cells and late outgrowth endothelial cells: the role of angiogenic

cytokines and matrix metalloproteinases. Circulation. 2005;112(11):1618-1627.

209. Yoshikawa Y, Abrahamsson SO: Dose-related cellular effects of platelet-derived growth

factor-BB differ in various types of rabbit tendons in vitro. Acta Orthop Scand. 2001;72(3):287-

292.

210. Yuan J, Murrell GAC, Wei A, et al: Apoptosis in rotator cuff tendonopathy. Journal of

Orthopaedic Research. 2002;20(6):1372-1379.

211. Lamplot JD, Angeline M, et al: Distinct effects of platelet-rich plasma and BMP13 on

rotator cuff tendon injury. Am J Sports Med. 2014 [Epub ahead print]

145

212. Hernigou P, Lachaniette CH, et al: Biologic augmentation of rotator cuff repair with

mesenchymal stem cells during arthroscopy improves healing and prevents further tears: a case-

controlled study. International Orthopaedics. 2014;38(18):1811-1818

146

Appendix A

Flow cytometry analysis for CD 31, CD 34, CD 133, and Flk-1

Forward Scatter – Area

Forw

ard

Sca

tter

–H

eigh

t

Forward Scatter – Area

CD 133CD 34

Flk-

1

Sid

e Sc

atte

r –

Are

aSi

de

Scat

ter –

Are

a

The first gate selected cells from the total events registered by the flow cytometer (top left). The

second gate limited the analysis to single cells only (top right). Using unstained cells as well as

fluorescence minus one (FMO) samples, we determined the threshold for CD 34 positivity to be

at 103 (bottom left). Cells with a fluorescence of 10

3 or higher were considered CD 34 positive.

The analysis indicated 98.3 % of the single cells express CD 34. Then we gated for both CD 133

and Flk-1 at the same time (bottom right). Again, using unstained cells as well as FMO samples,

we determined the threshold for CD 34 positivity to about 103. The analysis indicated 96.8 % of

the CD 34 expressing cells express both CD 133 and Flk-1. In other words, approximately 95 %

of the single cells express all three markers.

147

CD 31 is a mature endothelial marker. In this study, CD 31 was

used to evaluate endothelial commitment of the endothelial

progenitor cells.

The unstained endothelial progenitor cell population exhibited

some level of auto-fluorescence (top picture). The gating

showed that approximately 60 % (after accounting for auto-

fluorescence) of the endothelial progenitor cells were positive

for CD 31 (middle picture). While the shift in fluorescence was

small, the detection of CD 31 with the antibody was sensitive as

indicated by overlapping the unstained cell data with the stained

(bottom picture).

Flow cytometry showed that most cells possessed the

endothelial progenitor cell markers CD 34, CD 133, and Flk-1,

but only some expressed the endothelial marker C 31. This

indicates that the cell population exhibited stronger

stem/progenitor cell characteristics than the terminally

differentiated endothelial cells.

148

Appendix B

Bonferroni’s Multiple Comparison Test

Ultimate Force

Fibrin Group Mean Diff. Significant? Adjusted P Value

Week 2 vs. Week 4 -1.341 No > 0.9999

Week 2 vs. Week 6 -9.761 Yes 0.0003

Week 4 vs. Week 6 -8.42 Yes < 0.0001

EPC Group Mean Diff. Significant? Adjusted P Value

Week 2 vs. Week 4 -2.44 No 0.6187

Week 2 vs. Week 6 -15.5 Yes < 0.0001

Week 4 vs. Week 6 -13.06 Yes < 0.0001

Stiffness

Fibrin Group Mean Diff. Significant? Adjusted P Value

Week 2 vs. Week 4 -1.397 No > 0.9999

Week 2 vs. Week 6 -6.395 Yes 0.0133

Week 4 vs. Week 6 -4.998 Yes 0.0103

EPC Group Mean Diff. Significant? Adjusted P Value

Week 2 vs. Week 4 -0.453 No > 0.9999

Week 2 vs. Week 6 -4.67 Yes 0.0161

Week 4 vs. Week 6 -4.217 Yes 0.0039

149

Cross Sectional Area

Fibrin Group Mean Diff. Significant? Adjusted P Value

Week 2 vs. Week 4 1.612 Yes 0.0008

Week 2 vs. Week 6 0.9835 Yes 0.0429

Week 4 vs. Week 6 -0.628 No 0.1046

EPC Group Mean Diff. Significant? Adjusted P Value

Week 2 vs. Week 4 1.346 No 0.0569

Week 2 vs. Week 6 1.158 No 0.1206

Week 4 vs. Week 6 -0.188 No > 0.9999

Ultimate Stress

Fibrin Group Mean Diff. Significant? Adjusted P Value

Week 2 vs. Week 4 -0.5802 No 0.233

Week 2 vs. Week 6 -1.62 Yes 0.0001

Week 4 vs. Week 6 -1.04 Yes 0.0008

EPC Group Mean Diff. Significant? Adjusted P Value

Week 2 vs. Week 4 -0.6939 Yes 0.0322

Week 2 vs. Week 6 -2.638 Yes < 0.0001

Week 4 vs. Week 6 -1.944 Yes < 0.0001