Tissue engineering approaches to tendon repair studies on the use of bone marrow stromal cells and knitted poly (d, l lactide co glycolide scaffold

TISSUE ENGINEERING APPROACHES TO TENDON REPAIR: STUDIES ON THE USE OF BONE MARROW STROMAL CELLS AND KNITTED POLY D, L-LACTIDE-CO-GLYCOLIDE SCAFFOLD OUYANG HONGWEI Bachelor of Medicine/Ba

TISSUE ENGINEERING APPROACHES TO TENDON REPAIR: STUDIES ON THE USE OF BONE MARROW STROMAL CELLS AND KNITTED POLY (D, L-LACTIDE-CO-GLYCOLIDE) SCAFFOLD

OUYANG HONGWEI (Bachelor of Medicine/Bachelor of Surgery)

A THESIS SUBMITED FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

DEPARTMENT OF ORTHOPAEDIC SURGERY

NATIONAL UNIVERSITY OF SINGAPORE

2003

Acknowledgements

I wish to express my deepest gratitude and heartfelt thanks to my supervisors:

Professor Lee Eng Hin, Dean, Faculty of medicine, National University of

Singapore, and Associate Professor James Goh Director of research, Department of

Orthopedic Surgery, National University of Singapore, for their constant

encouragement, invaluable guidance and infinite patience throughout the course of

this study

I would like to express my sincere thanks to Professor K Satkunanantham Head,

Department of Orthopedic Surgery, National University of Singapore, for his support

Without the excellent facilities, this work would not have been possible

I owe my thanks to Professor Teoh Swee Hin, Assistant Professor Dietmar

Hutmacher, Dr Mo Xui-Mei, Division of Bioengineering, NUS for their assistance

in the provision of biomaterials and reagents, as well as technique assistance in

manufacturing of scaffolds

I would also like to express my appreciation to the following staff members Ms

Chong Sue-Wee, Ms Julee Chan, Mr Ashvin Thambyah, Ms Grace Lee, Mr Barry P Pereira, Ms Jessie Tan, Mr Yong Soon Chiong, Mr Dominic Tey, Dr Li

Li, Dr Ge Zi-Gang, and Dr Shao Xin-Xin for their kind help

I will always remember my friends in Singapore for their constant encouragement and

kind help This work was support by a grant from NMRC, Singapore

Last but not least, I am grateful to my families, my parents, my wife Zou Xiao-Hui

and my baby OuYang Xin-yi for their understanding and love during the years of my

Ph.D pursuit

Table of Content

I Acknowledgement………i

II Table of content……… iii

III Summary……….ix

IV Publications………xii

Chapter I: Introduction and Literature Review………1

Chapter II Literature Review……….……… 5

2.0 Introduction to Literature Review……….… … 6

2.1 Tendon Anatomy, Physiology And Injury……… …6

2.1.1 Tendon Anatomy……… 6

2.1.2 Tendon Composite……… ……….… ………8

2.1.3 Tendon to Bone Insertion…… ……… ….….……13

2.1.4 Tendon Biomechanics…… ……… ……14

2.1.4.1 Tendon Mechanical Properties………… ……… …… 14

2.1.4.2 Effect of Biomechanical Load on Tendon……….16

2.2 Tendon Injury Healing and Current Therapy……… 18

2.2.1 Tendon Injury Healing ……… 18

2.2.2 Several Concerns about Tendon Healing……… 20

2.2.2.1 The Amount of Reparative Cells………20

2.2.2.2 The Mobilization of Reparative Tendon……….…22

2.2.2.3 The Evaluation of Tendon Healing………22

2.2.3 Current Therapy……… 23

2.3 Tissue Engineering Approaches To Improve Tendon Healing……… 24

2.3.1 Tissue Engineering Principles……….24

2.3.2 Cell Source……… 25

2.3.3 Biomaterials and Scaffold……… 29

2.3.4 Biomolecules……… 33

2.3.5 Animal Model……….36

2.3.6 Tissue Engineering Techniques for Tendon Insertion Healing……… 39

2.4 Hypotheses and Objective of This study……… …….…… 40

Chapter III: Materials and Methods……….… ……42

3.0 Introduction to Material and Methodology……….………43

3.1 Stage1 bMSCs Differentiation Study……….…………43

3.1.1 Isolation And Culture Of Bone Marrow Stromal Cells……… 43

3.1.2 Osteogenesis Induction and Detection……….……… 44

3.1.2.1 In vitro Osteogenesis Induction ……… ……….……… 44

3.1.2.2 Vonkossa Staining……… 45

3.1.3 Chondrogensis Induction And Detection………46

3.1.3.1 In Vitro Chondrogenesis Induction ……… ……….…….…46

3.1.3.2 Collagen Type II Immunoassaying……….…47

3.1.4 Adipogenesis Induction And Detection……… 48

3.1.4.2 Oil Red Staining……….50

3.2 Stage II Trace Study On The Fate of bMSCs After Implantation……….51

3.2.1 Animal Model ……… 51

3.2.2 Cells Labeling And Detection……….52

3.2.2.1 CFDA Labeling And Detection……… 52

3.2.2.2 GFP Gene Transfection And Detection……….53

3.2.3 Tissue Preparation For Cryostat Section……….54

3.3 Stage 3 Study On The behavior of bMSCs On Various Polymer Films …… 55

3.3.1 Materials……….55

3.3.2 Polymer Film Manufacture……….56

3.3.3 Polymer film Sterilization and Prewetting……….……….56

3.3.4 Water Contact Angle Test……… 57

3.3.5 bMSCs Adhesion Assay ………57

3.3.6 bMSCs Proliferation Assay……….57

3.3.7 MTS Assay For Cell Proliferation……… 58

3.3.8 Statistic Analysis Method……… 58

3.4 Stage 4 Study On The Effect Of bMSCs Seeded knitted PLGA For Achilles Tendon Repair……….……59

3.4.1 Animal Surgery……… 59

3.4.2 Knitting PLGA Fiber Scaffold………61

3.4.3 Tissue Preparation For Paraffin Section……….61

3.4.4 H&E Staining……… …….62

3.4.5 Immunohistochemical Staining……… 63

3.4.6 Transmission Electronic Microscopy……… ….64

3.4.7 Biomechanical Test……….…64

3.5 Stage 5 Study on the Effect Of bMSCs On the Tendon Insertion Healing… … 66

3.5.1 Animal Surgery……… ……….……… 66

3.5.2 Hard Tissue preparation For Paraffin Section……… ……….………67

3.5.3 H&E & Immunohistochemical Staining ……… 67

Chapter IV: Results………68

4.1 Stage 1.The Differentiation Of bMSCs………… ……….….69

4.1.1 bMSCs Isolation……… …69

4.1.2 Osteo-lineage Differentiation……… 70

4.1.3 Chondro-lineage Differentiation ……… … 71

4.1.4 Adipo-lineage Differentiation……… … ……72

4.2 Stage 2 The Fate of bMSCs After Implantation…… ……… ……73

4.3 Stage3 The Adhesion and Proliferation of bMSCs on Various Polymer Films……… 76

4.3.1 Polymer Films………76

4.3.2 Cell Adhesion……….77

4.3.3 Cell Proliferation ……… ……….………79

4.3.4 Cell Morphology……… ……….……….81

4.4 Stage4 The Efficacy Of Allogeneic bMSCs Seeded Knitted PLGA Scaffold For Achilles Tendon Repair……… ….82

4.4.2 The Immunohistology of Tendon Repair……… ………90

4.4.3 The Biomechanics of Tendon Repair……… …… … … 91

4.5 Stage5 The Effect of bMSCs on Tendon Insertion Healing………… …….….94

5.5.1 The Natural Healing of Tendon Insertion……… ….… 94

5.5.2 The Effect of bMSCs on Tendon Insertion Healing ……… … …96

Chapter V: Discussion………99

5.1 The Isolation and Differentiation of bMSCs……….……… 100

5.1.1 The isolation of bMSCs……….…… 100

5.1.2 The Multipotential of bMSCs……….…….101

5.1.3 Bone Marrow Stromal Cells as Tendon progenitor Cells……….… 102

5.2 The Fate of bMSCs After Implantation……… … 103

5.2.1 The Methods of Cell Trace Study……… … 103

5.2.2 The Fate of bMSCs after Implantation into Tendon Wound Site….……… …103

5.2.3 The Possibility of Allogeneic bMSCs for Tissue Engineering……….… 105

5.2.4 The Potential of bMSCs For Gene Delivery……… …….…….107

5.3 The Behavior of bMSCs on Various Polymer Films……… ……… …108

5.3.1 Method for Characterizing Cell-Polymer Interaction……… ….108

5.3.2 The Effect of Cell Source on The Cell-Polymer Interaction……… 109

5.3.3 The Effect of Substrate on The bMSCs Adhesion And Proliferation……… 110

5.4 The Efficacy of bMSCs and Knitted PLGA scaffold for Achilles Tendon Repair……… … 113

5.4.1 The Effect of Knitted PLGA on Tendon Repair……… ….113

5.4.2 The Effect of bMSCs on Tendon Repair……… 116

5.4.3 Tendon Repair versus Tendon Regeneration……… …118

5.5 The Effect of bMSCs on Tendon Insertion Healing ……… …120

5.5.1 The Natural Healing of Tendon Insertion……… ….120

5.5.2 The Effect of bMSCs on Tendon Insertion Healing……… ….121

Chapter VI Conclusion and Recommended Future Study……… ….…124

ChapterVII References……… …… 128

Summary

Background: Unlike bone that is able to heal by regenerating normal bone in most

cases, tendons often heal by forming scar tissue The limited capacity for injured

tendon to regenerate poses a great challenge and creates an opportunity for

engineering new tendons To date, less work has been done on tendon tissue

engineering compared to the extensive work on the bone and cartilage tissue

engineering Several studies have investigated the use of gel and braided scaffold

with or without cells for tendon repair; however, the inferior mechanical strength of

gel carrier and the poor tissue ingrowths associated with the braided fiber scaffold has

limited the success Many other problems have yet to be addressed: the fate of

implanted bone marrow stromal cells (bMSCs) at the tendon site has not yet been

studied; no general principles have been established to select material for bMSCs

delivery; the role of bMSCs in tendon repair has been arguable due to the lack of

appropriate controls in previous studies, and limited attention has been placed on the

tendon-to-bone healing when engineering tendon graft for tendon repair

Hypotheses: The main hypothesis is that tissue engineered graft composed of bMSCs

and knitted PLGA can improve tendon repair To support this hypothesis: (a) bMSCs

should have multipotential and be good candidates of cell source for tendon repair

(b) The knitted PLGA should be promising for bMSCs delivery and tendon repair

Methods: With the aim to verify the above hypotheses, five sequential stages of

experiments were designed They are as follows:

• Stage 1: The bMSCs differentiation test to determine the potential of the

bMSCs used in this study to differentiate into different tissues

• Stage 2: Cell trace test to determine whether the allogeneic MSCs can

survive and differentiate into tenocytes after implantation

• Stage 3: Cell- matrix interaction test to determine the appropriate material for MSCs’ delivery

• Stage 4: Tendon repair with bMSCs/knitted PLGA composites to evaluate the efficacy of this treatment modality

• Stage 5: The effect of bMSCs on the healing of tendon to bone

Results: The stage 1 experiment showed that the bMSCs isolated by short-term

plastic adhesion were able to differentiate into multi-mesenchymal lineages such as

osteo-, chondro- and adipo-lineages The stage 2 experiment illustrated that the

implanted allogeneic bMSCs could survive as long as 8 weeks and was able to

differentiate into spindle-shape cells 5 weeks after implantation at rabbit patella

tendon window wound site The stage 3 experiment selected optimal material for

bMSCs delivery by verifying that PLGA was more likely to allow bMSCs to adhere

and grow as compare to other five synthetic biodegradable polymers The stage 4

experiment exhibited that the composite of bMSCs and knitted PLGA scaffold

could improve the structure and biomechanics of tendon repair in rabbit Achilles

stage 5, the bMSCs exhibited the potential to restore the native structure at the

tendon to bone interface healing in the stage 5 experiment

Conclusion: In all, these sequential experiments proved that the bMSCs were able

to be the seed cells for tendon repair; the knitted PLGA scaffolds possess optimal

material and structural properties for bMSCs delivery and tendon tissue formation;

and the composite of bMSCs and knitted PLGA could be an ideal substitute for

tendon repair This work could be logically extended to ligament repair

polymeric films Mat Sci Eng C-BIO S 20 (1-2): 63-69 Sp Iss SI MAY 31 2002

2 Ouyang HW, Goh JCH, Mo XM, Teoh SH, Lee EH The efficacy of bone marrow

stromal cell-seeded knitted PLGA fiber scaffold for Achilles tendon repair Ann N

Y Acad Sci Jun; 961:126-9 2002

3 Ouyang HW, Goh JCH, Thambyah A, Teoh SH, Lee EH The use of knitted

PLGA and MSCs for Achilles tendon repair in rabbit model Tissue Engineering Vol 9, No.3, 431-439, 2003

4 Goh JCH, Ouyang HW, Chan CK, Teoh SH, Lee EH Tissue engineering

approaches to tendon and ligament repair and regeneration Tissue Engineering

Vol.9 Sup1, 31-44, 2003

5 Ouyang HW, Goh JCH, Lee EH The effect of MSCs on the tendon to bone

healing Am J Sport Med (In Press) 2003

6 Ouyang HW, Goh JCH, Lee EH Viability of allogeneic bone marrow stromal cells following local delivery into patella tendon in rabbit model J Orthop Res

(submitted) 2003

International Conference:

marrow stromal cells following implantation in tendon regeneration in rabbit model 49th annual meeting of orthopedic research society New Orleans, LA, USA Feb 2-5, 2003

2 Ouyang HW, Goh JCH, Lee EH Application of mesenchymal stem cells in the

repair of tendon and ligament 11th international conference of biological and

3 Ouyang HW, Bini TB, Mo XM, Yang F, Wang S, Ramakrishna S, Goh JCH,

Teoh SH, Lee EH Fibrous scaffolds in regeneration of tendon and nerve tissues BECON 2001 Repair Medicine: Growth Tissue and Organs, Jun 25-26,

2001, NIH, Maryland, USA

marrow stromal cell-seeded knitted PLGA fiber scaffold for Achilles tendon repair BECON 2001 Repair Medicine: Growth Tissue and Organs, Jun 25-26,

2001, NIH, Maryland, USA

5 Ouyang HW, Goh JCH, Mo XM, Teoh SH, Chong SW, Wang Z, Lee EH Cell-

material systems for ACL regeneration: The behavior of biodegradable

polymer films ICMAT 2001, Singapore

stromal cells on the interface healing of tendon to bone ICMAT 2001, Singapore

marrow stromal cell-seeded knitted PLGA fiber scaffold for Achilles tendon repair ICMAT 2001, Singapore

8 Goh JCH, Ouyang HW, Lee EH Use of tissue engineering techniques in the

repair of tendon and ligament Conference on biomedical engineering technology Taiwan 2001

Chapter I Introduction

Musculoskeletal conditions are increasingly becoming one of the major health

concerns because of an aging population and increased occurrence of sport-related

injuries To date, orthopedic surgeons still rely on conventional methods of repair, such

as the use of autograft, allograft and prostheses Although these procedures have been

fairly successful, the various shortcomings of these procedures have prompted surgeons

and scientists to look for viable alternatives Recently developments in tissue

engineering have shown great potential in solving the problems associated with tissue or

organ transplantation As compared to the extensive work on bone and cartilage tissue

engineering, tendon tissue engineering has not received much attention

Tendon is arguably the least complex of the connective tissues with respect to its

composition and architecture This might reasonably lead to the expectation that it would

be more amenable to tissue engineering approaches than other tissues However, decades

of experience have showed how difficult it is for tendon to regenerate after treatment

This limited capability of tendon injuries to regenerate poses a challenge to tendon tissue

engineering, and emphasizes the importance of developing a procedure to do so

So far, few works have been done on tendon tissue engineering compared to the

extensive work on the bone and cartilage tissue engineering Several studies have

investigated the use of gel and braided scaffold with or without cells for tendon repair;

however, the inferior mechanical strength of the gel carrier and the poor tissue ingrowths

associated with the braided fiber scaffold has limited its success Many other issues have

not yet to be addressed, for example: 1) the fate of implanted bone marrow stromal cells

(bMSCs) at the tendon site has not yet been studied; 2) no general principles have been

established to select material for bMSCs delivery; 3) the role of bMSCs in tendon repair

has been arguable due to the lack of appropriate controls in previous studies, and 4)

limited attention has been placed on the tendon-to-bone healing when engineering tendon

graft for tendon repair

It is clear that the continued development of tendon tissue engineering will depend

upon identification and characterization of appropriate sources of cells as well as the

development of new scaffolds The identification of an optimal cell source for a particular

tissue engineering application will depend on rigorous characterization with regards to

plasticity, propagation, and control of differentiation To guide the organization, growth,

and differentiation of cells in tissue engineered constructs, appropriate scaffolds are

needed to provide mechanical support as well as physical, chemical, and mechanical cues

in forming functional tissues

The main hypothesis of this study is that tissue engineered graft composed of bMSCs

and knitted PLGA can improve tendon repair Specific mechanistic hypotheses on why

bMSCs can be the seed cells and knitted PLGA scaffolds can be the vehicle and template

for tendon repair are as follows:

1) bMSCs can differentiate into multimesenchymal tissues including teno-lineage

2) bMSCs can survive and differentiate into tenocytes after local delivery into tendon

site

3) bMSCs attach and grow faster on PLGA substrate than other five polymers

4) Knitted PLGA scaffold can allow a large number of cells infiltration and connective

fibrous tissue ingrowths; and bMSCs/knitted PLGA graft can accelerate tissue

5) bMSCs can restore physical structure of tendon-to-bone interface by inducing

fibrocartilage zone formation at the insertion

This thesis includes seven chapters In chapter I, I briefly introduced the background of

tendon repair, tissue engineering, the hypotheses of this study, and the scope/ structure of

this thesis

The literature review (Chapter II) of this thesis covers the tendon anatomy, physiology,

biomechanics, injury healing, clinical therapies and current tendon tissue engineering

researches Chapters III, IV and V of this thesis respectively describe the material &

methodology, results and discussion of the five sequential experiments which were

designed to verify the hypotheses of this study In Chapter VI, a conclusion is drawn and

suggestions for future studies are proposed The references are in Chapter VII

Chapter II Literature Review and hypotheses

2.0 Introduction to Literature Review

Understanding of the cellular and extracellular components of tendon is essential for

determining the methodology required to successfully engineer tissue, and provide useful

information in the development of analogs of the extracellular matrix as implants to

facilitate tendon regeneration Insights into tendon biomechanical properties and the effect

of load on tendon structure can help design the tissue engineering graft with appropriate

biomechanical properties Moreover, knowledge of the process of spontaneous healing of

tendon injuries and the results of selected treatment modalities can provide a guide to

strategies for tissue engineering Finally, a review of the principles of tissue engineering

and current tissue engineering studies in developing grafts to facilitate tendon regeneration

can serve as the foundation for this tendon tissue engineering work

2.1 Tendon Anatomy, Physiology And Injury

2.1.1 Tendon Anatomy

Tendon is a specialized dense connective tissue that links muscle to bone and allows for

transmission of muscle contraction forces to the bone for skeletal locomotion; for example,

the Achilles tendon, one of the largest in the body, links the triceps surae muscle, the

grouping of gastrocnemius, the soleus, and the plantaris muscle to the calcaneus bone As

organs, tendons consist of three parts: the muscle attachment region, the substance of the

tendon itself, and tendon to bone insertion region These three parts vary in their cellular

composite, histology, and function This study will focus on the midsubstance of the tendon

and give some insight into the healing of tendon insertion

Healthy tendons are brilliant white in color and fibroblastic in texture, showing great

resistance to mechanical loads Besides a few tendons in the hands and feet that have the

two-layer tendon sheaths, most of the tendons are surrounded by loose areolar connective

tissue called paratenon The main components of the paratenon are collagen type I and III

fibrils( Kvist, 1985), elastic fibrils and synovial cells that form the inner lining ( Williams,

1986) Paratenon functions like an elastic sleeve and permits free movement of the tendon

against the surrounding tissues (Hess et al 1989) Under the paratenon, the entire tendon is

surrounded by a fine connective tissue sheath called the epitenon or peritenon On its outer

surface, the epitenon is continuous with the paratenon and on its inner surface with the

endotenon The epitenon is a relatively dense fibrillar network of collagen with strands of 8

nm to 10 nm in thickness (Jozsa, 1991) The endotenon is a thin reticular network of

connective tissue inside the tendon that has a well-developed crisscross pattern of collagen

fibrils (Jozsa, 1982, Kastelic, 1978, Rowe, 1985) The endotenon fibrils connect tendon

fibers and bind fibers together A bunch of collagen fibrils forms a collagen fiber, which is

the basic unit of a tendon and is aligned from end to end in a tendon, and a bunch of tendon

fibers forms a primary fiber bundle (subfascicle), and groups of these bundles form

secondary bundles (fascicles) A group of secondary bundles forms tertiary bundles, which

make up the tendon

Fig2 The hierarchical organization of the tendon structure from collagen fibrils (20nm-

150 nm) to collagen fiber (1 µm -50 µm), fascicle and the entire tendon.( Adapted from

Josza 1997)

2.1.2 Tendon Composite

Tendons have relatively few cells but a large amount of extracellular matrixes that give

them extremely high tensile strength The extracellular tendon matrix is composed of

collagen and elastin fibers, the ground substance, and the inorganic components Collagen

constitutes approximately 90% of the total protein of the tendons or 65-75% of the dry

mass of tendons Elastin accounts for only about 2 % of the dry mass of tendons (Hess,

1989, Jozsa 1989b) The ground substance, which surrounds the collagen, consists of

proteoglycan, glycosaminoglycans(GAGs), structural glycoproteins and a wide variety of

other small molecules These elements are produced by tenoblasts and tenocytes that are

the elongated fibroblasts and fibrocytes lying between the collagen fibers

All collagen molecules are long stiff rods consisting of a triple helix of three polypeptide

chain To date, 13 different types of collagen molecules have been found in the mammalian

body ( Karpakka 1991) The 13 types are named by roman numbers The relative proportion

of each type of collagen in each tissue is characteristic and specific In tendon and

ligament, type I collagen predominates, but small amounts of other types of collagen are

also found (see table 1)

Table 1 The chemical composition, amount, and location of the different collagen types in

tendon

Collagen

type

Molecular formula

Molecular Weight(Dalton)

Amount in tendon ( %)

,vascular walls, Muscle tendon junction

membrane, Muscle tendon junction

Muscle tendon junction

Type I collagen has predilection to form parallel fibers One collagen type I molecule

consists of three helical polypeptide alpha-chains that wind around each other to form a

right-hand triple helix that extends collinear through out the length of the tropocollagen

molecule or microfibril (Ghadially 1988) A number of tropocollagen molecules or

between 20 nm and 150 nm (Demel et al 1982) Within a fibril, the microfibrils are

surrounded by proteoglycans and GAGs A bunch of collagen fibrils then form a collagen

fiber whose diameter may range between 1 and 50 um Within a fiber, proteoglycan and

GAGs surround the fibrils So each unit in the collagen fiber structure is surrounded by

proteoglycan and GAGs Type I collagen has high tensile strength with limited elasticity

and is therefore suitable for force transmission Other collagen types constitute only a small

part of tendinous collagen Their roles are not entirely clear With the advancement in basic

science of collagen biochemistry, the present concepts of collagen types and their functions

may well be changed

Fig 3 Schematic figure of a mature type I collagen fibril (Adapted from Ghadially, 1998)

Elastic fibers are barely present in human tendons (Carlstedt, 1987) The function of

elastic fibers is not entirely clear, but they may contribute to the recovery of the wavy

configuration of the collagen fibers after tendinous stretch (Buter, 1978) Proteoglycans are

composites of a protein core in which one or more glycosaminoglycan (GAG) are

covalently attached They are large (WM 106 daltons), negatively charged hydrophilic

molecules that can retain water 50 times their weight By virtue of their high fixed charge

density and charge-to- charge repulsion force, proteoglycans are stiffly extended to provide

the collagen fibrils with high capacity to resist high compressive and tensile forces

(Karpakka, 1991) In addition, the proteoglycans enable rapid diffusion of water soluble

molecules and migration of cells In tendon, a wide variety of inorganic components have

been detected(Lappalainen 1982) Calcium is found in the highest concentrations, with

concentration of 0.001% to 0.01% of tendon dry weight in the tensional area of a normal

tendon( Josza 1989b) Other detected components have been magnesium, manganese,

cadmium, cobalt, copper, zinc, nickel, lithium, lead, fluoride, phosphor, and silicon(

Spadaro 1970, Strehlow 1969)

The organization of the extracellular matrix molecules of the tendon is the principal

determinant of the physical function of this tissue The degree to which tissue engineering

approaches are successful will depend on the degree to which the normal composition and

architecture of the extracellular matrix is restored

Apart from the extensive work on the biology of the extracellullar components of

tendons, there are also many studies that focused on understanding the biology of tendon

cells Tendons have a variety of cell types, including fibroblasts, fibrocartilage cells and

occasional fat cells Fibroblasts are the most common and can be found in all regions of

tendons and ligaments They are typically arranged in elongated rows within the parallel

bundles of collagen fibers In longitudinal section, the cells are elongated and have spindle

shape nuclei In addition to the responsibility of producing extracellular matrix, fibroblasts

As the parenchymal cell of tendon, the tenocyte has the role of maintaining matrix

structure through the degenerative and formative processes comprising remodeling, and to

some extent healing But tendon has a relatively low density of cells that also display a low

mitotic activity So how much these cells can promote intrinsic healing is the question

Moreover, this fact may suggest the need for the application of exogenous cells in tissue

engineering approaches for tendon repair and regeneration

Besides the extracellular matrix and cells, tendons have an organized peri- and

intratendinous network of blood vessels (Elliott 1965) For example, the human Achilles

tendon is supplied by small branches of the posterior tibia, anterior tibia and peroneal

arteries (Schmidt-Rohfing 1992) Small branches from the above-mentioned arteries run

transversely through the paratendon, branch repeatedly to become longitudinally and

transversely oriented along the fiber fascicles to form a uniform mesh-like vascular system

along the length of the tendon (Reynolds 1991) The vessels of this network then penetrate

the tendon (Schmid-Rohlfing 1992) It has been demonstrated that there is a zone of

decreased vascularity in the Achilles tendon at 2 to 6 cm proximal to the tendon insertion

(Carr 1989) The number and diameter of intratendious vessels diminish in the distal end

towards the middle portion of the Achilles tendon Several authors have speculated that this

may be an important etiological factor in the development of Achilles tendon degeneration

and rupture (Archambault 1995, Clement 1984, Kvist 1992, Reynolds 1991) The majority

of ruptures occur in this hypovascularizated zone Also the poor vascularization limits the

capability of healing in tendon injuries This fact may point to the need of tissue

engineering approaches to initiate revascularization, to improve oxygenation and to initiate

tissue healing

2.1.3 Tendon-to-Bone Insertion

One of the keys to the success in surgical reconstruction of damaged tendons or ligaments

is the tendon-to-bone healing Acceleration of the appropriate healing process at the

insertion site would allow patients to go for early rehabilitation Tendons attach to bone

through fibrous or fibrocartilaginous junctions In limbs, fibrous insertions are

characteristic of tendons that attach to the diaphyses, whereas fibrocartilaginous insertions

are typical of attachment to epiphyses (Benjamin, 1986, 2002) The fibrocartilaginous

insertion is characterized by four distinct zones: bone, mineralized cartilage, fibrocartilage

and tendon (Cooper and Misol 1970)

Fig4 Direct insertion of rat quadriceps tendon to patella: T= Tendon, F= Fibrocartilage, M

= Mineralized fibrocartilage, B=Bone (Adapted from Cooper 1970)

The tendinous part of tendon-to-bone junction consists of dense collagen bundles with

Gradually the cells lose their elongated appearance and become round, and begin to look

like the chondrocytes They lie in pairs or rows surrounded by lacunar spaces in the

extracellullar matrix Around the chondrocytes, 10nm to 30 nm tendinous collagen fibrils

can be seen (Ippolito 1986) Immunohistochemical studies have shown that they are type II

collagen, which is the main collagen of articular cartilage The fibrocartilage is separated

from the mineralized fibrocartilage by a distinct border named as the “cementing” or

“basophilic tidemark” or “blue’ line due to the intensive affinity of the line when stained

with hematoxylin and eosin (Ippolito 1986) Similar to tendon and fibrocartilage, the

boundary between mineralized fibrocartilage and bone is seamless and the collagen fibers

of these two zones are indistinguishable The bone tissue shows no special features that

would separate it from normal bone This four-zone specific insertion site structure is

designed to distribute longitudinal and shear forces from the tendon proper into the

subchondral bone plate, thus minimizing stresses on individual collagen bundles (Benjamin

1986) However, the fibrocartilaginous transition would not form during normal healing of

tendon to bone in a bone tunnel (Rodeo 1993, Liu 1997)

2.1.4 Tendon Biomechanics

2.1.4.1 Tendon Mechanical Properties

Tendons transmit load with minimal energy loss and deformation But it is not an

entirely inextensible cable that directly transfers the length change or force of a contracting

muscle to the bone It is elastic as well and capable of deforming and returning to its

original length (Best& Garrett 1994) Thus, tendons are viscoelastic materials A tissue is

said to be elastic if it returns to its original shape after removing the stress If it does not

return, the tissue is viscous (Viidik, 1966) A tendon is perfectly elastic as long as the strain

does not exceed 4 %, after which the viscous range commences This can be illustrated by

the stress-strain curve of tendon

Stress

Fig 4 A stress-strain curve for tendon

The initial concave portion of the stress-strain curve( region I ) has been termed the “toe”

region.( Butler 1978, Viidik 1973, Best 1994) This toe portion of the curve results when the

waviness( crimped ) tendon fiber bundle straightens out The strain of the tissue at the end

of this region has been reported to be between 1.5% and 4 %( Bulter 1978, Viidik 1973,

Whittaker 1991) Following the toe region, the tendon shows a relatively linear response to

stress (region II) The fibers of the tendon become more parallel and have lost their wavy

region is often referred to as the elastic stiffness, or Young’s modulus of elasticity of the

tendon( Best 1994) Microfailure of fibers occurs at the end of this linear loading region

Beyond the linear load (region III), more fiber failure occur in an unpredictable fashion

(Bulter 1978) This region corresponds to strains of 3% to 8% In the fourth part of the

curve (Rregion IV), macroscopic failure occurs because of the tensile failure of the fibers

and shear failure between fibers (O’Brien 1992) Besides the stress-strain response, tendon

also undergoes stress-relaxation, creep, and hysteresis loop Stress-relaxation means that

with the same degree of stretch, or sufficient deformation, the load required to maintain the

stretch decreases over time (Best 1994, Fyfe 1992) Creep means that with a constant load,

deformation increases over time Hysteresis loop is a measure of the energy that is lost

during the loading-unloading test of the tendon; it therefore is an indication of the viscous

properties of the tissue (Butler 1978)

The mechanical properties of tendon have been modeled as functions of the morphology

and hierarchical structure of the collagen fibers (Comninou M 1976, Frank C 1992, 1996)

Also several studies have investigated the relationship between morphology and the

restoration of mechanical properties in healing of tendon (Frank C 1992, 1996, Dervin,

1996) These considerations emphasized the need for functional tissue engineering

approaches to regenerate the molecule composition and hierarchal structure of tendon in

order to be assured that the material will be able to function as normal

2.1.4.2 Effect of Biomechanical Load on Tendon

Animal experiments provide evidence that tensile strength, elastic stiffness, and total

weight of tendon increased by gradually increasing physical exercise (Woo 1980, 1981,

1982, Archambault 1995, Curwin 1988) The changes in tendons can be explained by

exercise-induced acceleration in the synthesis of collagen and proteoglycan matrix due to

increased tenocyte activity (Heikkinen 1972, 1973, 1975) Microscopically, the collagen

fibrils and fibers are thickened and their tropocollagen cross-links increase in number The

orientations of tendon fibers become more parallel with the stress lines of the tendon

(Michna 1984, 1989, Maffulli 1992) As far as we know, the Achilles tendons of athletes

were thicker than those of non athletes It suggests that localized physiological adaptation

to mechanical stress occurs in human tendons (Kainberger 1990, Maffulli 1992)

In in-vitro studies, tendon cells have been shown to respond to mechanical loading The

responses include release of intracellular calcium, alteration of their cytoplasmic filament

organization and content, polymerization of actins and alteration of protein expression

(Banes 1999a, Ralphs 2002) However, little is known about the mechanism of cellular

response to mechanical load It seems that stress-generated potentials, mechanosensitive

ion channel, and membrane-bound stretch receptors may play important roles in triggering

these responses (Sutker 1990, Davidson 1990) A recent study showed that cyclic loading

of whole avian flexor tendons stimulated DNA and collagen synthesis, and this response

could be blocked with octanol, a reversible gap junction blocker (Banes 1999b) Confocal

study by McNeilly et al (1996) on adult rat tendons showed that cells are in direct physical

contact and they form a communicating network throughout the tendon It is significant

that the adjacent cells communicate via the gap between them and within the rows This

indicates that there is a network of mechanically responsive cells throughout the tendons,

and the network ensures that the overall tissue response is well coordinated

As mechanical loading has an effect on the tendon extracellular matrix organization and

the tenocyte activity, one concern of tissue engineering approaches is to enable the

mobilization of reparative tendon immediately after surgery, which then physically

regulates cell behavior and restores the structure and function of the reparative tendon

early

2.2 Tendon Injury Healing and Current Therapy

2.2.1 Tendon Injury Healing

Tendons and ligaments repairs are of major concerns Injury to tendons tends to be site

dependent and may occur by direct or indirect mechanism For example, the majority of

tendon injuries of the hand are caused by direct injury such as lacerations, while Achilles

tendon injuries are caused by either direct injury or indirect injury resulting in spontaneous

ruptures The spontaneous healing of tendon has been studied extensively in the Achilles

tendon and the flexor tendons of the hand Following a full transection, there is a

spontaneous retraction of the cut tendon ends In the Achilles tendon of both a rat and

rabbit model, the gap formed by the retraction of the ends, with the joints held in neutral

position, was observed to be approximately 9-12 mm(Buck 1953) Additional retraction of

the ends can occur with movement of the calcaneous or the knee joint

The healing of the Achilles tendon, reviewed in several articles (Buckwalter, 1987,

Amadio, 1992; Lui, 1995) is similar to the healing of other connective tissues such as skin

and ligament According to several authors (Buckwalter, 1987; Woo, 1982; Buck, 1953;

Amadio, 1992.), the sequence of tendon healing is generally broken down into four

overlapping phases: Phase I (Hemorrhagic) is within the first few hours of injury After the

disruption of tendon, the gap is filled with blood clots (Buck, 1953; Enwemeka, 1989;

Flynn, 1965; Postacchini, 1978) Triggered by the cytokines released within the clot,

leukocytes and lymphocytes appear within several hours These cells expand the

inflammatory response and recruit other types of cells to the wound Phase II (

Inflammatory) starts within several hours of injury and takes 3-10 days to complete This

stage is associated with a “clean-up” of the lesion site Macrophages arrived by 24-48

hours and are the predominant cell type in phase II Macrophages are responsible for

phagocytosis of necrotic tissue Macrophages also secrete growth factors to induce

neovascularization and the formation of granulation tissue Phase III (Fibroblasts migration

and proliferation) starts as early as 10 days after injury and takes 2-5 weeks to complete

Fibroblast is the last cell type to arrive within the wound Although debate continues, it is

currently thought that fibroblasts are recruited from neighboring tissues and the systemic

circulation (Woo 1998) These fibroblasts have abundant rough endoplastic reticulum and

begin to proliferate in the wound site within the fibrin mesh of the clot Simultaneously,

endothelial cells of surrounding vessels enlarge and proliferate forming capillary buds that

follow the migrating fibroblasts Together, the fibroblasts, macrophages and capillaries

form the granulation tissue in the wound site (Enwemeka, 1991) This phase is

characterized by increased cellularity The fibrin clot is gradually replaced by collagen

matrix comprising predominantly collagen type III The collagen type III fibers do not

aggregate in a preferential direction The matrix is loose and disorganized Phase IV(

decrease in the cellularity of the healed tissue, a reduction in the production of type III and

reorganization of the type I collagen fibers(Flynn, 1965) During this stage, the matrix

becomes denser and longitudinally oriented The tensile strength of the tendon increases

through the period of remodeling even though the total volume is decreased (Flynn, 1965;

Gonzalez, 1949; McGaw, 1986)

Tendon healing may occur in three ways: by tissue regeneration, by scar repair, or by a

combination of both Regeneration is a form of repair that produces new tissue that is

structurally and functionally identical to normal tissue (Ledbetter 1992) Regeneration thus

represents the ideal soft-tissue injury healing which, however, is not always the case in

tendon injuries Instead, tendon injury is often repaired by scar tissue which has poorly

organized extracellular matrix and inferior biomechanical strength With time, the scar

tissue may assume some of the characteristics of the tendon, but complete regeneration

does not appear to occur The goal of tissue engineering approaches is to improve the

quality of tendon repair

2.2.2 Issues in Tendon Healing:

2.2.2.1 The Amount of Reparative Cells

Many cell-mediated processes related to the generation of skeletal tissue depend on the

number of cells involved, both in the rate and magnitude of the effect For example, in the

in-vitro model of production of connective tissue, the rate of collagen gel contraction by

fibroblasts embedded in the gel is dependent on the number of cells(Bell, 1979).The extent

of fibroblast orientation in cultures grown on collagen gel is directly related to the initial

cell density(Klebe, 1989) This cell orientation effect has been correlated with the

observation of an “organization center” in the culture, the number of which has been

suggested to be a direct indicator of morphogenetic capacity at the molecular and cellular

level(Bab, 1984).In addition, cell density-dependent differentiation was demonstrated in

the culture of limb bud cells(Caplan, 1970) The cells exhibited osteogenesis and

chondrogenesis at higher density cultures So the numbers of cells initially present will

strongly influence the nature of cell mediated processes involve in tissue formation and the

rate at which these developmental and physiological processes occur If these observations

are applied to the reparative processes of skeletal tissue, it seems that some minimum

threshold of cell number may be required at the repair site before formation of normal

neotissue can occur Muschleret, et al (2002) estimated that for one cubic centimeter of

new bone formation would require 70 million osteoblasts This concept can probably be

extended to tendon regeneration There is controversy on the identity and location of the

cells responsible for collagen synthesis during tendon repair Some believe that tendon has

the necessary cells (Mass, 1990; Manske, 1984; Garner, 1989), while others believe that

the cells are from outside of the tendon (Potenza, 1962, 1975), It is currently thought that

both intrinsic (tenocyte) and extrinsic (neighboring tissue and the systemic circulation)

cells contribute to tendon healing (Russell IE 1990, Mason ML 1941) Questions remain

about whether there is a sufficient pool of tenocytes or connective tissue progenitor cells to

be recruited to the wound site If the cell number is insufficient, the extent to which the

reparative process can occur maybe limited This uncertainty warrants the application of

2.2.2.2 The Mobilization of Reparative Tendon

As the mechanical load physically regulates cell behavior and extracellular matrix

organization, the effect of post-operative immobilization and mobilization has been

investigated in many studies (Frank C 1991, Buck RC 1953, Meislin RI 1990, Howard CB

1985, Takai S 1991, Gelberman RH 1983, Murrel GAC 1994, Enwemeka CS 1988) Most

of the studies have shown the positive effect of post-operative mobilization Wada A, et al

(2001) reported that actively mobilized tendons healed without the extrinsic adhesions

while large tendon calluses were found in immobilized tendons For the Achilles tendon,

Pneumaticos SG (2001) reported that early mobilization of the repaired tendon seemed to

restore the functional properties of the tendon more rapidly than continuous immobilization

in following an identical surgical repair So the need for early post-operative mobilization

of the reparative tendon requires the initial mechanical strength of the tissue engineering

graft to be strong enough for early physical movement of the tendon

2.2.2.3 The Evaluation of Tendon Healing

Structural and functional techniques are used by many studies for the evaluation of

tendon healing (Kato YP 1991 Buck RC 1953, Postacchini F 1978, Gelberman RH 1983,

Stein SR 1976, Ketchum LD 1977, Gelberman LD 1977) Structural techniques have

included morphological assessment and biochemical analysis Morphology was evaluated

qualitatively by comparing repair tendon to normal tendon at macrostructural,

microstructural(microscopy) and ultrastructural (electromicroscopy) levels Some structural

parameters such as collagen fiber orientation, the crimping pattern, the diameter of collagen

fibrils and the interface of repair tendon to normal tendon were considered very important

to evaluate the structure of reparative tendon The components and the ratio of the

components of extracellular matrix were evaluated by biochemical techniques Functional

techniques have included biomechanical testing (Best TM 1993, Goldin B 1980) and gait

analysis, with mechanical testing being the most widely used method to determine the

degree to which restoration of function has been achieved Stiffness, maximum load and

elastic modulus (Young’s modulus) were also used to evaluate the function of reparative

tendon

2.2.3 Current Therapy

Currently the therapeutic options to treat tendon and ligament injuries are direct suture

and autograft, allograft and permanent tendon prostheses in case of tendon defects

(Jorgensen, 2001; Steebrugge, 2001; Kato, 1991) The autograft has many advantages, such

as absence of immunological and infectious problems, quick incorporation, and good

remodeling However, the autograft is also accompanied by several disadvantages, such as

1) the increased duration of the operation due to the time used for graft harvest and

preparation, 2) the limited donor tendon source, and 3) the sacrifice of the function of the

original tendon Animal experiments have confirmed that the fresh autograft undergoes a

gradual but complete reorganization The tenocytes of the graft degenerate and original

collagen fibers swell and finally disappear Salamon, 1970; Kao,1991 in Achilles tendon

neotenon was not identical to the normal Achilles tendon The mechanical property of the

neotendon was 60% of the normal at 1 year and the crimp length was only 20% of the

normal at 1 year Allografts are accounted with problems of immunological rejection,

disease transmission and limited donor tendons The disadvantages of biological grafts

encourage further research for alternative repair options, including the use of tendon

prostheses Various non-biodegradable materials, such as polyester (Levine, 1968),

polyethylene (Grau, 1958), dexon, teflon (Gonzales, 1958; Grau 1958), carbon (Benson

1971, Janeke 1974, Jenkins 1977, 1978, Goodship 1980), and silicon (Hunter 1987), have

been used in tendon prostheses However, artificial tendons are not able to duplicate the

fiber organization, attachment site anatomy, or function of the original tendon, and the

wear products of the prostheses stimulated foreign body granulation, scar formation and

tendon tissue necrosis

In summary, the disadvantages of biological grafts and questions about long term

performance of permanent prostheses have fueled the search for tissue engineering

substitutes

2.4 Tissue Engineering Approaches To Improve Tendon Healing

2.3.1 Tissue Engineering Principles

The term tissue engineering was initially defined in 1988 as “application of the principles and methods of engineering and life sciences toward fundamental understanding

of structure-function relationship in normal and pathological mammalian tissues and the

development of biological substitutes for the repair and regeneration of tissue or organ

function.”(Skalak 1988) In 1993, Langer and Vancanti summarized the early

developments in this field and defined tissue engineering as “an interdisciplinary field that

applies the principles of engineering and life sciences toward the development of biological

substitutes that restore, maintain or improve tissue or organ function.”(Langer 1993) Tissue

engineering has now emerged as a potential alternative to tissue or organ transplantation

With this technology, tissue loss or organ failure can be treated by implantation of a tissue

engineered graft Tissue engineered grafts are composed of either two or all of the three

major components-cells, biomaterials/scaffold and bio-molecules Currently there are two

approaches to tissue engineering: one is to implant the cells-scaffold composite directly

into the damaged site, as such the body acts like a “bioreactor”; the other is to culture the

cells-scaffold composite in a bioreactor ex vivo for a period of time before transplantation

The first approach is widely applied for connective tissue repair and regeneration So far,

tissue engineered products such as bioartificial skin (Apligraf from organogenesis) and

autologous cultured chondrocytes(Carticel from Genzyme Tissue repairs) have reached the

market, which gives encouragement for the investigation on tendon tissue engineering As

compared to other tissues, such as bone and cartilage, tendon has its specific

characteristics So the choice of cell source, biomaterial/scaffold and bio-molecules need to

be different and specific for tendon engineering

2.3.2 Cell Source

A key factor in the tissue engineering approach to tissue repair and regeneration is the

proliferation potential, cell-to-cell signaling, bio-molecule production and formation of

extra-cellular matrix The number of cells initially seeded will strongly influence the nature

of cell-mediated processes involved in skeletal tissue formation and the rate at which these

developmental and physiological processes occur It seems clear that some minimum

threshold quantity of cells may be required at the repair site for normal neotissue formation

(Caplan 1993) As such, exogenous cells are essential in facilitating tissue regeneration

There are a number of cell sources They can be classified into parenchymal cells (e.g

fibroblasts from tendon), progenitor cells (e.g bone marrow stromal cells) according to the

cell potential, or into autologous cells (e.g cell from one’s own body) and allogeneic cells

(e.g cells from another human donor) For the parenchymal cell source, several groups

(Huang, 1993; Dun, 1995; Goulet, 1997; Bellincampi, 1998) have used fibroblast-seeded

collagen scaffold for tendon/ligament regeneration The viability and proliferation of the

fibroblasts were investigated on the scaffolds Lin et al (1999) cultured fibroblasts on PGA

fiber scaffolds in-vitro and achieved tissue formation after 5 weeks of culture The in vitro

studies exhibited the possibility of engineering tendon or ligament analogs However, the

function of the tendon/ligament analogs needed to be evaluated in vivo In-vivo studies by

Bellincampi et al (1998) showed that fibroblast-seeded collagen scaffolds were viable for at

least 8 weeks after re-implantation in the knee joint or subcutaneous tissues of the donor

rabbits But the collagen scaffolds were completely degraded at 8 weeks There was no

tendon or ligament like tissue formation Cao et al(1995) isolated fibroblasts from newborn

calves and seeded the cells onto non-woven meshes of PGA fibers After one week of

culture, the cell/PGA composite were implanted subcutaneously in nude mice for up to 10

weeks Histology at 10 weeks showed parallel linear organization of collagen bundles