autologous tissue engineering for bone repair

TOWARDS AN AUTOLOGOUS TISSUE ENGINEERING CONSTRUCT FOR CRANIOFACIAL BONE REPAIR

BY

AARON JEFFREY MAKI

DISSERTATION

Submitted in partial fulfillment of the requirements

for the degree of Doctor of Philosophy in Bioengineering in the Graduate College of the

University of Illinois at Urbana-Champaign, 2013

Urbana, Illinois

Doctoral Committee:

Professor Matthew B. Wheeler, Chair, Director of Research Professor Brian Cunningham Professor Walter Hurley Assistant Professor Brendan Harley

ii

ABSTRACT

Patients with critical-size bone defects, as a result of trauma, congenital malformations or

tumor resections, generally have limited healing without clinical intervention. The autograft is

the current standard of care for repair of these defects due to capacity for osteointegration and

immunological compatibility. However, potential limitations, such as donor site morbidity, have

motivated the development of alternative autologous approaches for the treatment of these

defects. Materials used in tissue engineering, such as scaffolds, growth factors and adult stem

cells, can be derived from patient blood and adipose tissue and are potential autologous

therapeutic options. This dissertation investigates a prospective procedure to improve

craniofacial bone healing using fibrin scaffolds and platelet rich plasma from patient blood, and

adipose-derived stem cells from liposuction. The objectives of these studies are to evaluate the

effects of fibrin scaffolds and platelet-rich plasma on adipose-derived stem cells and their ability

to heal critical-size bone defects in a porcine animal model.

During coagulation of whole blood, fibrin scaffolds were modified using treatments to

reduce red blood cell density and porosity or increase concentrations of calcium and phosphate

ions. Platelet-rich plasma was collected using an anticoagulant with subsequent centrifugations

to acquire the fraction of plasma with high concentrations of growth factor-releasing platelets.

Both fibrin scaffolds and platelet rich plasma were cultured with adipose-derived stem cells to

determine proliferation, migration, and osteogenic differentiation potential. Autologous adipose-

derived stem cells, platelet-rich plasma, and fibrin scaffolds were injected into critical-size

defects in the porcine mandible. Analysis of bone healing after 8 weeks indicated higher bone

mineral density and bone volume fraction compared to untreated controls for all three treatments

using ASCs. Addition of both platelet rich plasma and fibrin scaffolds to autologous ASCs from

liposuction improved bone volume fraction of critical-size defects. Based on these results,

addition of either platelet-rich plasma or calcium phosphate-fibrin composite scaffolds to

autologous adipose-derived stem cells are recommended to for further improvement in healing of

critical-sized bone defects.

iii

ACKNOWLEDGMENTS

I would like to thank my advisor, Dr. Matthew Wheeler, for everything he has done for

me. Even during times when things got tough, he never gave up on me, which took endless

amounts of patience sometimes. I am truly lucky to have him as an advisor. I would also like to

thank my lab group members including Elisa Monaco, Massimo Bionaz, Ijeoma Omelogu,

Jamey Cooper, Chanaka Rabel, Ekta Khetan, Siddhant Jain, Anna Ercolin, and Kelly Roballo.

This work would not have been possible without their willingness to help and their ability to be

excellent friends. Jonathon Mosley and the rest of his staff at the Imported Swine Research

Laboratory did an outstanding job of caring for the pigs and helping me with the surgeries. Ted

Limpoco at the Materials Research Laboratory was instrumental in helping me figure out how to

use nanoindentation. Peter Fitschen was very helpful in showing me the ropes for dual-energy x-

ray absorptiometry. Lucas Osterbur’s assistance with scanning electron microscopy was very

helpful, and at the Beckman Institute of Science and Technology, Scott Robinson’s help with

histological preparation was much appreciated. And the suggestions of my doctoral committee,

Drs. Brian Cunningham, Walter Hurley and Brendan Harley, greatly improved this dissertation.

I want to thank my mother, father, and brother for their support and love. Though the

years fly by, they are constants in this world and have always been there for me. I would also

like to thank my wonderful wife, Agatha, for just being herself as an amazing and strong woman.

I can only hope our new daughter, Katherine Abigail, will be just like her. I can’t thank you all

enough for all that you do for me.

iv

TABLE OF CONTENTS

CHAPTER 1: INTRODUCTION………………………………………………………… ……..1

CHAPTER 2: IN VITRO OSTEOGENESIS ON FIBRIN SCAFFOLDS………………….…..24

CHAPTER 3: IN VITRO MIGRATION IN PLATELET-RICH PLASMA……………….…...44

CHAPTER 4: IN VIVO AUTOLOGOUS BONE TISSUE ENGINEERING…………….……60

CHAPTER 5: SUMMARY AND CONCLUSIONS………………………………….…….….82

APPENDIX A: PROTOCOLS…………………………………………………………………..85

1

CHAPTER 1

INTRODUCTION

1.1 Clinical Rationale

1.1.1 Current Practices

Delayed or non-union bone defects do not heal spontaneously without clinical

intervention and are the result of numerous clinical conditions, including tumor resections,

congenital abnormalities and traumatic injuries [1]. About 500,000 patients in the United States

are affected annually, an estimated 5-10% of all bone fractures [2]. While the mechanism of

delayed or non-union is under investigation, one primary factor is defect size. A non-union,

defined as less than 10% bone regeneration, occurs at a critical size, generally when the length of

the defect is approximately 1.5-2 times the thickness of the damaged bone [3]. Several factors

likely play a role in the determination of the critical size, including surrounding blood supply,

intactness of periosteum, biomechanical fixation, age, and presence of co-morbidities such as

diabetes [4].

Currently, the gold standard for treatment is the autograft, where bone is harvested from a

donor site in the same patient (usually iliac crest, rib, or fibula) and is shaped by the surgeon to

fill the defect. This approach confers many advantages in terms of patient outcomes, including

superior immunological compatibility and low risk of graft failure. However, complications

occur in up to 30% of all cases primarily due to the second surgical site, including pain in the

harvested area, as well as increased blood loss [5]. The drawbacks of donor site morbidity and

limited tissue supply have prompted development of alternative methods which can overcome

these disadvantages, especially for cases in which a large amount of grafted bone is required.

One currently used alternative is the allograft, which utilizes bone from a human donor.

While removing the problems of the second surgery site, allografts generally have limited

promotion of bone growth and have the potential for immune rejection leading to graft failure

[6]. In addition, mechanical weakening can occur due to the sterilization procedure using gamma

irradiation or during packaging or storage by freeze drying. And while small, a theoretical risk of

2

disease transmission remains [7,8]. Therefore, allografts are typically utilized when a patient is

not a candidate for autografting.

Other alternatives include the use of xenografts, typically decellularized bovine bone.

While generally available and inexpensive, xenografts are immunologically incompatible and

may require the use of immunosuppressant drugs, which lower resistance to infection [9]. For

patients who would have limited success with bone grafts, metallic implants can be used to

restore form and function and improve quality of life. Currently, alloys of cobalt chromium or

titanium are most widely used. However, these metals are more than ten times stiffer than bone

causing tissue at the implant interface to be shielded from normal stresses, which can result in

resorption over time, leading to implant failure [10]. Though immunologically less problematic,

metallic implants cannot repair or remodel and have limited integration with surrounding bone,

which serves to limit their useful lifetime before revision surgery is required.

The current state of the art for treating delayed and non-union defects involves tradeoffs.

To obtain the generally superior results of the autograft, a second surgical site is required. In

order to eliminate this site, allografts or xenografts have issues with immunological

compatibility. To improve these issues, metallic implants have limited integration with

surrounding bone [10]. The ideal bone graft would be biocompatible without requiring a second

surgical site. Potential bone grafting substitutes generally require low risk-benefit ratios for

widespread clinical use. Therefore, complications of the bone grafting procedures and metallic

implants have prompted investigation into alternative methods, such as the tissue engineering

approach.

1.1.2 Bone Tissue Engineering

The fields of tissue engineering and regenerative medicine aim to interact with host

tissues to stimulate regeneration and reduce formation of scar tissue [11]. Newly regenerated

bone would retain its ability to repair and remodel itself over the patient’s lifetime, potentially

achieving similar levels of integration as an autograft without requiring a second surgical site.

The tissue engineering triad involves the use of scaffolds, growth factors, and cells [12].

Scaffolds provide an intermediate surface for cells to attach and guide the process of tissue

formation. Over time, these scaffolds are resorbed to make space for newly forming tissue.

3

Growth factors provide instructions for migration, proliferation, and differentiation of cells to

form bone and other important support structures such as vascular networks. The sources of

these essential elements are important in determining the outcome of any proposed therapy.

Currently, most tissue engineering strategies utilize synthetic scaffolds, recombinant growth

factors, and/or culture-expanded stem cells.

Tissue engineering strategies vary depending on the amount of cell culture work needed

[13]. The first approach, in situ (in place), utilizes scaffolds and growth factors and relies on

interaction with the host tissue microenvironment. The second approach, ex vivo (outside the

living), adds cells isolated from a patient or donor and then re-inserted into the patient. The third

approach, in vitro (on glass), produces most tissue in a laboratory bioreactor before implantation.

This dissertation focuses on the second approach, ex vivo bone tissue engineering.

As a tissue, bone is a composite material of collagen I reinforced with hydroxyapatite

crystals. Following a bone fracture, an initial fibrin blood clot forms and the inflammatory

process is initiated to degrade necrotic tissue. Mesenchymal stem cells (MSCs) from the marrow

stroma, periosteum and surrounding tissues migrate to the wound site and proliferate to form a

soft, collagenous callus, which is invaded by new blood vessels. On the order of weeks, new

bone matrix will be secreted for calcification into a hard callus which is slowly remodeled back

to the original organization [14]. Compared to other tissues, bone has evolved a high capacity for

regeneration, but its effectiveness is still limited by defect size. Migrating MSC number and

activity, blood vessel invasion, and callus calcification are all potential limiting factors, which

are current targets for tissue engineering strategies [15,16]

An array of scaffolding materials has been developed for the purpose of bone tissue

engineering. These include natural polymers such as collagen, hyaluronic acid and fibrin, as well

as synthetic polymers such as poly-glycolic acid (PGA), poly-lactic acid (PLA) and its

copolymer (PLGA), and polycaprolactone (PCL) among others [17]. Ceramics composed of

calcium phosphates and calcium sulfates have also been used. Each of these materials has their

respective strengths and weaknesses in terms of strength, degradation speed and products,

biocompatibility, and ease of processing. Depending on the application, anatomical bone

location, and loading requirements, each material has potential to be used for the repair of bone

4

defects. The natural polymer, fibrin, a primary constituent of blood clots, is the focus of this

dissertation.

The principal growth factors involved in bone regeneration include fibroblast growth

factor (FGF), platelet-derived growth factor (PDGF), vascular endothelial growth factor (VEGF),

transforming growth factor (TGF), and the bone morphogenetic proteins (BMP), among others

[18]. The sources of these growth factors include platelets, white blood cells, surrounding bone,

and/or recombinant proteins added therapeutically. Each growth factor plays overlapping roles in

the process of bone formation, and one prevailing notion in tissue engineering is that increasing

the concentration by the correct amount, location and time course of one or more of these growth

factors may enhance the healing process. While each growth factor family varies in its amino

acid sequence and intracellular pathways activated, many activate serine threonine kinase

receptors and are biologically active as dimers (PDGF, TGF, BMP).

In particular, PDGF and FGF play a key initiatory role, increasing cellular migration

towards the wound and proliferation at its site [19]. FGF and VEGF are potent stimulators of

proliferation and angiogenesis to build vascular support networks. TGFs and BMPs have

overlapping roles in proliferation and are essential to begin the cascade toward the eventual

upregulation of the transcription factors Runx2, and Osterix. This results in the upregulation of

genes which are markers of osteoblasts, including collagen I, alkaline phosphatase, osteocalcin

and osteopontin, resulting in the synthesis of collagen and the enzymes which cause its eventual

mineralization into bone [20].

Cells used in bone tissue engineering tend to focus on adult stem cells for their ease of

isolation and ability to expand in culture [21]. Adult stem cells from different tissue sources,

such as bone marrow, fat, periosteum, and teeth have been effectively used in bone tissue

engineering. Embryonic/induced pluripotent stem cells have not been as widely used due to

possibility of teratoma formation, immune rejection, and/or ethical concerns. Differentiated

osteoblasts have potential but are also less widely used due to more difficult isolation and

reduced proliferation.

While both autografts and tissue engineering are the future of bone defect repair, it is

possible that a hybrid, autologous tissue engineering approach, where cells, scaffolds and growth

factors are acquired from the patient, may capitalize on the advantages of each technique. An

5

autologous tissue engineering method, as detailed in this dissertation, has the potential to

maintain similar performance and compatibility to an autograft without requiring harvesting of

patient bone, avoiding its associated complications.

1.2 Fibrin Scaffolds

1.2.1 Background and Biology

One factor less-studied in the bone fracture healing process is the natural polymer, fibrin,

which is a primary component of blood clots. Blood contains constituents essential for the

wound healing process, including coagulation cascade enzymes and fibrinogen. In fibrin’s native

form, it exists as a soluble monomer, fibrinogen, which circulates through the cardiovascular

system. Upon activation through damage-induced exposure to collagen or other mechanisms,

fibrinogen converts to insoluble fibrin under the control of a series of enzymatic steps of the

coagulation cascade [22]. These polymerizing fibers cross-link together to form a stiff clot able

to limit blood loss and act as a scaffold in the wound repair process.

Fibrinogen is a glycoprotein produced by the liver, 340 kDa in size and circulates at a

concentration of 1.5-4.0 g/L (7 µM) in blood. Fibrinogen is composed of three peptide chains

linked by disulfide bonds (α, β, γ) for which coiled α-helices predominate in their secondary

structures [23]. These coiled-coils are a major source of elasticity for the polymer allowing for

stretching 2-3 times the original length [24]. Two N-terminals associate (E domain), forming a

hexamer with the C-terminals (D domain) left externally exposed as recognition sites. Bonding

between N-terminals is on the α-chain, but two non-reacting peptides (fibrinopeptide B) are

maintained on the β-chain, which aids in maintaining solubility.

Following activation through the intrinsic and/or extrinsic pathways, a cascading series of

enzymes, in which the preceding enzyme acts as a catalyst for activation of the next, result in the

eventual activation of Factor X [25]. This enzyme cleaves and exposes cryptic sites on the

zymogen prothrombin, converting to thrombin, a serine protease capable of interacting with

fibrin. Thrombin’s activated protease sites cleave fibrinopeptide B from both β-chains, allowing

for the formation of insoluble protofibrils. Lateral bonding of protofibrils between similar

domains result in the formation of fibrin fibers. These insoluble fibers intertwine and form a

6

three-dimensional fiber network , trapping red blood cells and platelets to rapidly form a soft

blood clot. On the order of minutes, another enzyme in the coagulation cascade, the

transglutaminase Factor XIII, links N-terminal domains with lateral C-terminal domains, serving

to cross-link and stiffen the fibrin gel [26].

The completed fibrin gel is viscoelastic and approximately 30 kPa in stiffness [27].

Several factors during coagulation play a role in the properties of the gel, including ion and

fibrinogen concentrations, fiber thickness and density, and branch point density [28,29]. Charged

inter-helical interactions between chains allow for positively-charged hydrogen ions, among

others, to interact, serving to increase resistance to stretch by electrostatic repulsion, which

increases stiffness of the gel. There is a general inverse relationship between fiber thickness and

density, depending primarily upon the ratio of fibrinogen to thrombin during polymerization.

Because individual fibers are highly elastic, fibrin gels with thick fibers and low density will

have low stiffness. Fibers at high density are highly entangled and provide more points of

resistance to movement and a stiffening effect. The number of potential branch points also

increases, further stiffening the gel.

However, while fibrin provides the initial wound microenvironment for bone fracture

repair, some aspects of fibrin clots may present challenges for the success of potential ASC

therapies. Two possible limitations of fibrin include its comparatively low stiffness and ion

concentrations [30]. For example, trabecular bone is approximately four orders of magnitude

stiffer than fibrin [31]. Scaffold stiffness on the order of trabecular bone and presence of calcium

and phosphate ions have both been shown to increase osteoblast differentiation [32]. Therefore,

modifications which can stiffen fibrin and reasonably increase its local ion concentrations may

result in improved outcomes for ASC therapy for bone defects.

Potential methods to increase blood clot stiffness include reduction of red blood cells and

porosity, reducing pH, and addition of stiff granules to form a composite. Reducing pH increases

protonation of residues which results in more charged inter-helical interactions. Forming

composites with stiffer, insoluble materials, such as calcium phosphates, could result in more

balanced properties, similar to the collagen-hydroxyapatite composite of bone [33]. Highly-

deformable red blood cells, which are designed to flexibly move through capillaries, decrease the

stiffness of blood clots due to their low stiffness (order of 10 Pa) [34]. Reducing the number and

7

intactness of red blood cells in a blood clot may increase stiffness. In addition, cross-linking

density of fibrin is below its maximum [26]. Finally, reducing porous empty space between

fibers significantly increases stiffness [35]. In summary, fibrin is a flexible hydrogel whose

polymerization conditions determine its structure and properties which may be modified for

improved bone defect healing.

1.2.2 Therapeutic Use

Blood may present an alternative source of autologous materials for use in craniofacial

bone defect repair. Compared to bone, blood is more accessible, more plentiful in supply, and

results in less structural deformity following its collection by blood draws. Therapeutically,

clotted blood and its derivatives, such as fibrin glue and fibrin sealant, have been recognized for

their useful properties of hemostasis and tissue adhesion since the 1980s and have been widely

used to reduce blood loss during and bleeding after surgical procedures [36]. Fibrin sealants also

can be utilized as carriers for the release of drugs, such as growth factors and antibiotics due to

their biocompatibility and resorption on the order of weeks.

In the United States, sources have historically been autologous from the patient due to the

risk of viral contamination, though virally-inactivated products from donors have now been

approved for clinical use [37]. Fibrin sealants/glues vary in formulation but consist of a

concentrated solution of fibrinogen mixed with thrombin (human or bovine) and calcium

chloride during the operation. Administration can be through syringe injection or sprayed at the

wound site. Fibrin sealants are generally tolerated well and have limited side effects, making

them candidate biomaterials for bone defect repair.

Despite its demonstrated effectiveness in the repair of soft connective tissue defects, the

use of fibrin to promote healing of hard bone tissue defects is generally controversial. Some

studies report reductions in bone formation [38,39] while others report no effect when treated

with fibrin sealant alone. Clinical usage of fibrin sealants for bone applications has been limited

and has focused primarily upon its hemostatic properties. One successful orthopedic application

of fibrin sealants has been in total knee arthroplasty and have been shown to reduce blood loss

during surgery [40].

8

General surgical techniques such as suturing aim to limit empty space in the wound, as

the subsequently large internal blood clots tend to result in scar tissue formation. Fibrin’s density

may impede cell infiltration and slow the wound healing response. In contrast, fibrin is generally

thought to support wound healing through the promotion of angiogenesis, cell proliferation and

attachment [41]. Therefore, when adult stem cells are combined with coagulating blood,

bypassing the requirement of cell invasion, the results may be more positive in terms of bone

growth.

Combining fibrin scaffolds with calcium phosphate granules to form a composite is a

more recent development and has been demonstrated to improve bone formation in defects and

ectopic sites [33,42]. Calcium phosphate ceramics such as hydroxyapatite and tri-calcium

phosphate have been used as bone graft and filling substitutes since the 1980s. Similar to the

ceramic portion of native bone, calcium phosphate granules may provide calcium and phosphate

ions and local stiffness on the scale of native bone, which has been demonstrated to promote

osteoblast differentiation [43].

However, calcium phosphate granules alone, while osteoconductive, are difficult to

handle and are mechanically unstable in large defects, which can be amended by mixing the

granules in polymerizing fibrin to immobilize granules to the implantation site. Clinical studies

of fibrin-calcium phosphate composites have been limited to large animal models and individual

surgical case reports, but generally have demonstrated improved defect filling in calvarial defects

[44]. In summary, fibrin sealants and glues represent an excellent surgical tool for the promotion

of hemostasis, tissue adhesion, and soft tissue defect healing. Forming a composite with calcium

phosphate granules may result in improved hard tissue defect healing as well.

1.3 Platelet-Rich Plasma


Bone regeneration is dependent upon various signals, including growth factor

concentration gradients and mechanical stresses and strains [45]. One cell type in particular, the

platelet, plays a key initiatory role in the wound healing process as an important source of

growth factors directing the initial healing response. These small, circulating, anucleate cells

9

store high densities of growth factors within their cytoplasmic granules [46]. Platelets originate

from within the bone marrow as portions of exocytosed cytoplasm from megakaryocytes and

circulate at a concentration of 200,000 per µL blood. Within minutes of activation at the site of a

wound, platelets begin to aggregate to form a sticky platelet plug for initial cessation of bleeding,

as well as release factors for the promotion of vasoconstriction, inflammation, and initiation of

the coagulation cascade [47]. One primary pathway of platelet activation is damage-induced

exposure to collagen and von Willebrand factor, which is not normally present in intact blood

vessels.

Following activation from the wound microenvironment, platelets degranulate, releasing

significant quantities of factors for the initiation of inflammation and the healing response. Dense

granules, containing mostly inflammatory factors such as adenosine diphosphate (ADP), act to

regulate platelet activity by positive feedback through recruitment of additional platelets to the

site of the wound and release of their contents. In addition, arachidonic acid is converted to

thromboxane A2, a potent initiator of inflammation and white blood cell migration. Previously

smooth and spheroid in shape, degranulation induces surface and morphological changes causing

platelets to be sticky and irregular, allowing them to clump together as a plug. Many enzymatic

conversions of the coagulation cascade occur on surfaces of platelets [48].

Upon activation, platelets also release the contents of their α-granules, containing mostly

growth factors [46]. As relatively small signaling proteins, growth factors play an essential role

for bone fracture healing and include platelet-derived growth factor (PDGF) which is a potent

mitogen, fibroblast growth factor (FGF) which is important for vasculogenesis, and transforming

growth factor-β (TGF-β) which induces cell differentiation into collagen-producing cell types,

among others. The only growth factor family not generally present in platelets is the bone

morphogenetic proteins (BMPs), which are found in native bone and released following damage

and resorption.

PDGF concentration increases over 4-fold following platelet activation and is a potent

stimulator of chemotaxis and proliferation. It consists of two isoforms (A or B) and is active as a

heterodimer (AB) or homodimer (AA or BB) linked by disulfide bonds. Tyrosine kinase

membrane receptors are activated in target cells, resulting in intracellular signaling through the

PI3K pathway, of which calcium ions are essential. Its primary intracellular effect on gene

10

transcription in the nucleus is to allow cells to skip the G1 regulatory checkpoint in mitosis to

promote rapid cell division [49].

FGFs are a family of growth factors which play important roles in embryonic

development, wound healing and angiogenesis. In addition, they have been shown to increase the

rate of adipose-derived stem cell proliferation [50]. FGFs act through FGF receptors, a type of

tyrosine kinase receptor. In particular, FGF-2 (basic FGF) is potent mitogen of mesodermal

tissues such as bone and is involved in patterning of limb development. FGF-2 is a key

stimulator of vasculogenesis during bone healing.

The TGF-β superfamily, which includes the BMPs and TGF-β among others, originates

from higher molecular weight precursors and are activated by proteolytic enzymes. On target

cells such as mesenchymal stem cells, they act on serine/threonine kinase membrane receptors.

Intracellular signaling is through the Smad pathway for many family members, leading to

eventual alteration of gene expression in the cell nucleus [20]. In particular, TGF-β1 is the most

abundant growth factor in human bone and plays a key role in the initial embryonic development

and patterning of the skeleton. Whether the source is native bone or degranulation, platelets

possess key proteolytic enzymes for the activation of TGF-β1, which is a potent growth factor

for the chemotaxis, proliferation, and differentiation of mesenchymal stem cells [51]. This

growth factor concentration increases 3.5-fold following platelet activation and degranulation.

Together, these growth factors make platelet activation and growth factor release important for

the initiation of the wound healing.

Following a fracture, a cascade of signaling molecules directs the bone’s eventual

regeneration. Due to its highly vascularized nature, significant hematoma formation and

inflammation develops, which is initiated by platelet release of inflammatory mediators.

Neutrophils and macrophages are then recruited for debridement of ischemic tissue and release

of inflammatory cytokines to enlist more white blood cells. By the first 24 hours, mesenchymal

stem cells are recruited primarily through chemotaxis toward gradients of PDGF and TGF-β

released by platelets [52]. The process of angiogenesis is initiated by the third day post-fracture

with FGF and vascular endothelial growth factor as the primary growth factors coordinating the

response [53].

11


Therapeutically, platelets have been utilized as a readily accessible source of growth

factors through the use of platelet-rich plasma (PRP), in which autologous blood is collected

with anticoagulant (sodium citrate) and spun in a centrifuge [54]. PRP contains a high density of

platelets and is acquired from the lowest fraction of plasma and upper fraction of the white blood

cell-containing buffy coat. The PRP fraction is isolated and stimulators of platelet activation

(thrombin and/or calcium chloride) are added to inactivate residual anticoagulant and release

growth factors through platelet degranulation.

Autologous PRP was first used in 1987 by Ferrari to avoid excessive blood product

transfusions [55]. PRP has since been used for a range of therapies, including orthopedic, dental,

and neurosurgical applications, among others [56]. Administration is generally considered safe,

with a low risk of infection occurring when sterility precautions are not sufficiently followed.

Typically, 30-60 mL of venous blood is withdrawn, then an initial centrifugation step is

performed to remove red blood cells and obtain plasma. A second centrifugation at a higher

speed separates platelet-poor and platelet-rich plasma fractions to obtain a total of 3-6 mL of

PRP.

Improvement of the rate of bone fracture healing has been demonstrated through the use

of PRP in preclinical animal models and case reports, while clinical trials have yet to be

performed [57,58]. The mechanism of healing is most likely due to increased growth factor

concentrations at the wound site. While PRP appears to be efficacious in promoting wound

healing, clinical outcomes of total effectiveness have been mixed, due primarily to the strong

pro-inflammatory component of platelet-released factors [59]. Therefore, clinical use of platelet-

rich plasma may require balancing with anti-inflammatory drugs. Or, anti-inflammatory cytokine

release, such as from adipose-derived stem cells [60], may provide inflammatory suppression to

remedy this potential limitation.

Other current limitations of PRP therapy are variability between methods for its

collection and between patients. Currently, two systems are approved for PRP collection in the

United States, but substantial variability remains in the clinical protocols used, including amount

of blood extracted, centrifugation speed and time, fractions isolated, as well as stimulants of

platelet activation. These differences have resulted in an assortment of platelet concentrations

12

used clinically (3-8 fold increase above baseline), which may explain some of the mixed results.

In addition, the thrombin commonly used to activate platelets uses bovine sources, which may

further exacerbate inflammation. Therefore, protocols for large-scale clinical use will require

standardization to improve treatment consistency.

Compared to growth factors obtained from recombinant (bacterial or mammalian cell)

sources, growth factors from PRP are lower in cost and allow for the administration of multiple

growth factors simultaneously, a technique which is more difficult using recombinant sources.

No studies have documented PRP to promote tumor growth, while using recombinant growth

factors at high concentrations may present an increased risk [61]. The key tradeoffs are reduced

control of dosage and side effects which stem from inflammation. Despite current limitations,

PRP therapy represents an autologous alternative to recombinant growth factors. The general

consensus for PRP indicates a promising potential therapy to promote enhanced bone healing,

but further study in clinical trials is needed before widespread adoption.

1.4 Adipose-Derived Stem Cells


Mesenchymal stem cells (MSCs) descend from embryological mesoderm and are defined

to retain into adulthood the capacity to differentiate into cells which produce and maintain the

connective tissues, such as bone, adipose, muscle, cartilage, and marrow stroma, among others

[62]. MSCs are capable of self-renewal primarily through asymmetric cell division, in which one

daughter cell differentiates into the desired cell type while the other remains an undifferentiated

clone and capable of self-renewal. These adult stem cells are isolated for primary culture through

tissue biopsy and digestion using an enzyme such as collagenase or trypsin. Following

centrifugation, the subsequent stromal-vascular fraction is plated onto tissue culture plastic

(irradiated polystyrene), and the fraction of cells which attach and form colonies originating from

a single cell are defined to be mesenchymal stem cells. The gold standard to prove stemness

would be to isolate a single colony and then differentiate into multiple lineages such as

osteogenic, adipogenic, and chondrogenic, among others using lineage-specific induction factors.

However, plastic adherence is typically sufficient in practice, resulting in a cell population which

13

is heterogeneous and fibroblast-like in nature. MSCs are capable of expansion of over one

billion-fold in culture without loss of multipotency [63]. Over multiple passages, the population

increases in homogeneity as more rapidly proliferating cell types predominate, but retain their

differentiation capacity.

In vivo, MSCs are less well-defined, but perictyes are one population of cells, residing

outside of capillaries, which share many characteristics with MSCs [64,65]. Both cell types are

angiogenic, aiding in the stabilization of blood vessels, as well as anti-apoptotic, reducing the

field of injury and reducing scar tissue formation. Pericytes and MSCs are also generally anti-

inflammatory through the release of cytokines such as interferon-gamma (IFN-γ), tumor necrosis

factor-alpha (TNF-α), interleukin-6 (IL-6), and interleukin-10 (IL-10) [66]. These cytokines

primarily act through suppression of helper T-cells, reducing length of inflammation and severity

post-injury. During a bone fracture, MSC proliferation and cytokine production becomes a major

factor in the substantial decrease in inflammation by the third day.

Historically, MSCs were first isolated from bone marrow. Bone marrow-derived stem

cells (BMSCs) remain the gold standard clinically for their well-developed protocols for marrow

biopsy, expansion, and osteogenic differentiation. However, limitations using bone marrow as a

source for MSCs include low tissue volumes per donor and low density of cells per volume of

bone marrow. In addition, this density appears to decline with age as red marrow slowly converts

to yellow, more fatty marrow [67]. Biopsy procedures also result in significant pain for an

already clinically valuable material used in bone marrow transplants for the treatment of

hematopoietic disorders. Therefore, most strategies involving BMSCs involve expansion in

culture to compensate for low cell numbers, a significant obstacle for widespread clinical use.

In comparison, adipose-derived stem cells (ASCs) have been more recently isolated, but

retain several similarities to BMSCs, including plastic adherence, cell morphology, and

capability to undergo osteogenic differentiation [68]. In addition, they share a 97% correlation in

transcriptome during osteogenic differentiation and a 90% similarity in expression of cell surface

markers [69]. Though some differences are present, such as response to some growth factors and

morphology of osteogenesis in vitro, ASCs may be a suitable substitute cell population for

BMSCs for bone tissue engineering applications. However, suitable preclinical studies in large

animal models would aid in improving potential clinical protocols.

14

The isolation of ASCs presents key advantages. Adipose tissue is typically abundant in

the body, and its anatomically superficial location improves accessibility by liposuction or

lipectomy. An estimated 1 out of 10,000 nucleated cells in adipose tissue is estimated to be an

ASC, which does not appear to change significantly with age, a significantly higher density (10-

fold) than BMSCs. In addition, 2-3 L of lipoaspirate may be collected under local anesthesia per

liposuction procedure, which would otherwise be discarded [70]. Compared to less than 300 mL

for bone marrow biopsy, a higher density of cells and capability of collecting more tissue

provides an opportunity for collecting clinically relevant numbers of ASCs without the need of

expansion in culture. This creates an opportunity to utilize ASCs as an autologous therapy

without requiring specialized culturing facilities as is generally required for BMSCs.


Past advances in the isolation, characterization, and differentiation of these relatively rare

adult stem cell populations have opened up new opportunities for the treatment of a wide range

of clinical conditions. Due to their properties of promoting vascularization, modulating

inflammation, and promoting tissue formation [71], ASCs have been investigated for treating

muscular dystrophy, diabetes, autoimmune disorders, and graft versus host disease, among others

[72,73]. However, this review will focus on the use of ASCs to promote healing of bone defects.

While precise mechanisms of action for ASC healing of bone defects are unknown,

vascular infusion or direct injection of undifferentiated ASCs have the potential to remedy

limiting factors of critical-size bone defects by migration, proliferation, and differentiation into

the proper cell types required by the regenerating tissue. Other effects include their generally

anti-inflammatory cytokine release profile and recruitment of surrounding cells by secretion of

growth factors and cytokines [74]. At the site of a bone defect, ASCs may modulate the

inflammatory response, recruit surrounding cells, and/or differentiate into osteoblasts to directly

build new bone.

Several options are available regarding the source and method of administration of ASCs.

Cross-tissue autologous cell transplants are one method in which the patient donates tissue, such

as fat by liposuction, for isolation of ASCs to be received by the patient’s bone. ASCs can be

administered intra-operatively, expanded in culture prior to treatment, or allowed to form bone at

15

an ectopic site before engraftment at the wound site. Currently, ASCs from liposuction are in use

clinically for adipose tissue grafts and other procedures involving fat transfer, showing

improvement in graft vascularization and integration [75]. Intra-operative use of ASCs for

calvarial bone defects has shown improved bone formation in case studies, and has potential for

widespread use if safety and efficacy can be established [76]. In one clinical case, maxillary

reconstruction was successfully accomplished by culturing ASCs using good manufacturing

practice (GMP) methods along with tri-calcium phosphate and BMP-2 and insertion in an

ectopic site before transplantation [77].

Another option in ASC therapy lies in the use of allogenic cells. There are an estimated 1

million liposuction procedures worldwide, and most of the resulting lipoaspirate is currently

discarded [70]. MSCs do not express B7 or CD40, which are key cell surface markers necessary

to activate T-cells to stimulate the host-transplant rejection response [78]. The immune-

privileged nature of ASCs provides donor flexibility for an abundant and accessible tissue

source. In a porcine mandible animal model, infusion of allogenic ASCs either intravascularly or

directly injected resulted in an increased rate of bone formation [79]. Addition of BMP-2 and

ASCs in a fibrin matrix in the rat femur reduced callus volume with no loss of bone formation

compared to administration of BMP-2 alone [80]. Human ASCs genetically modified to

overexpress BMP-2 were injected into critical-size rat femoral defects and found to improve

healing [81].

A current therapeutic question for ASC therapy for bone defects is whether or not an

inductive factor is required before administration or whether the wound microenvironment

provides sufficient induction. Most studies favor using growth factor administration which

directly stimulates osteogenic differentiation, such as BMP-2, along with ASC infusion. One

assumption of this approach is that administered cells produce the majority of the newly formed

bone tissue. The results of this combination have been mixed, with some evidence suggesting

ASCs do not directly respond to some growth factors [82]. Growth factor administration which

favors proliferation of undifferentiated ASCs in vivo is a currently less utilized approach which

may capitalize on other therapeutic benefits, such as anti-inflammatory and angiogenic properties

while maintaining capacity for osteogenic differentiation.

16

In summary, autologous ASCs present advantages of biocompatibility and reduced

complications compared to harvesting of patient bone for autografting. An autologous tissue

engineering approach utilizing ASCs from lipoaspirate may improve the healing outcomes of

critical-size bone defects. Combining blood-derived growth factors and fibrin scaffolds with

adult stem cells derived from fat comprises a construct which is consistent with the autologous

tissue engineering paradigm and may result in further improvements in bone healing.

1.5 References

1. Panagiotis, M. Classification of non-union. Injury 36, S30–S37 (2005).

2. Cattermole, H. R., Hardy, J. R. W. & Gregg, P. J. The footballer’s fracture. Br J Sports Med 30, 171–175 (1996).

3. Sd, C., Mw, W., Sl, S. & Dc, R. Effect of recombinant human osteogenic protein-1 on healing of segmental defects in non-human primates. J Bone Joint Surg Am 77, 734–750 (1995).

4. Iain H. Kalfas Principles of bone healing. Neurosurg Focus 10:1–4 (2007).

5. Younger, E. & Chapman, M. Morbidity at Bone Graft Donor Sites. Journal of Orthopaedic Trauma September 1989 3, 192–195 (1989).

6. Bonfiglio, M. & Jeter, W. S. Immunological responses to bone. Clinical Orthopaedics and Related Research 87, 19–27 (1972).

7. Conrad, E. U. Conrad EU, Gretch DR, Obermeyer KR, Moogk MS, Sayers M, Wilson JJ, Strong DM. Transmission of the hepatitis-C virus by tissue transplantation. Journal of Bone and Joint Surgery - Series A 77, 214–224 (1995).

8. Boyce, T., Edwards, J. & Scarborough, N. Allograft bone: The influence of processing on safety and performance. Orthopedic Clinics of North America 30, 571–581 (1999).

9. Horowitz, I. & Bodner, L. Use of xenograft bone with aspirated bone marrow for treatment of cystic defect of the jaws. Head & Neck 11, 516–523 (1989).

10. Kennady, M. C., Tucker, M. R., Lester, G. E. & Buckley, M. J. Stress shielding effect of rigid internal fixation plates on mandibular bone grafts. A photon absorption densitometry and quantitative computerized tomographic evaluation. International Journal of Oral and Maxillofacial Surgery 18, 307–310 (1989).

http://europepmc.org/search/;jsessionid=Nr5UmdSeUeT1nDrMZ2Bp.0?page=1&query=AUTH:%22Conrad+EU%22

http://europepmc.org/search/;jsessionid=Nr5UmdSeUeT1nDrMZ2Bp.0?page=1&query=AUTH:%22Gretch+DR%22

http://europepmc.org/search/;jsessionid=Nr5UmdSeUeT1nDrMZ2Bp.0?page=1&query=AUTH:%22Obermeyer+KR%22

http://europepmc.org/search/;jsessionid=Nr5UmdSeUeT1nDrMZ2Bp.0?page=1&query=AUTH:%22Moogk+MS%22

http://europepmc.org/search/;jsessionid=Nr5UmdSeUeT1nDrMZ2Bp.0?page=1&query=AUTH:%22Sayers+M%22

http://europepmc.org/search/;jsessionid=Nr5UmdSeUeT1nDrMZ2Bp.0?page=1&query=AUTH:%22Wilson+JJ%22

http://europepmc.org/search/;jsessionid=Nr5UmdSeUeT1nDrMZ2Bp.0?page=1&query=AUTH:%22Strong+DM%22

17

11. Kneser, U., Schaefer, D. J., Polykandriotis, E. & Horch, R. E. Tissue engineering of bone: the reconstructive surgeon’s point of view. Journal of Cellular and Molecular Medicine 10, 7–19 (2006).

12. Langer, R. & Vacanti, J. P. Tissue engineering. Science 260, 920–926 (1993).

13. Bianco, P. & Robey, P. G. Stem cells in tissue engineering. NATURE-LONDON- 118–121 (2001).

14. Einhorn, T. A. The cell and molecular biology of fracture healing. Clinical Orthopaedics and Related Research S7–S21 (1998).

15. Kadiyala, S., Jaiswal, N. & Bruder, S. P. Culture-Expanded, Bone Marrow-Derived Mesenchymal Stem Cells Can Regenerate a Critical-Sized Segmental Bone Defect. Tissue Engineering 3, 173–185 (1997).

16. Street, J., Bao, M., DeGuzman, L., Bunting, S., Peale, F. V., Ferrara, N., & Filvaroff, E. H. Vascular endothelial growth factor stimulates bone repair by promoting angiogenesis and bone turnover. PNAS 99, 9656–9661 (2002).

17. Hutmacher, D. W. Scaffolds in tissue engineering bone and cartilage. Biomaterials 21, 2529–2543 (2000).

18. Schliephake, H. Bone growth factors in maxillofacial skeletal reconstruction. International Journal of Oral and Maxillofacial Surgery 31, 469–484 (2002).

19. Bolander, M. E. Regulation of Fracture Repair by Growth Factors. Proc Soc Exp Biol Med 200, 165–170 (1992).

20. Lee, K. S., Kim, H. J., Li, Q. L., Chi, X. Z., Ueta, C., Komori, T., Wozney, J., & Bae, S. C. Runx2 Is a Common Target of Transforming Growth Factor β1 and Bone Morphogenetic Protein 2, and Cooperation between Runx2 and Smad5 Induces Osteoblast-Specific Gene Expression in the Pluripotent Mesenchymal Precursor Cell Line C2C12. Mol. Cell. Biol. 20, 8783–8792 (2000).

21. Mauney, J. R., Volloch, V. & Kaplan, D. L. Role of Adult Mesenchymal Stem Cells in Bone Tissue Engineering Applications: Current Status and Future Prospects. Tissue Engineering 11, 787–802 (2005).

22. Davie, E. W., Fujikawa, K. & Kisiel, W. The coagulation cascade: initiation, maintenance, and regulation. Biochemistry 30, 10363–10370 (1991).

18

23. Mosesson, M. W. Fibrinogen and fibrin structure and functions. Journal of Thrombosis and Haemostasis 3, 1894–1904 (2005).

24. Piechocka, I. K., Bacabac, R. G., Potters, M., MacKintosh, F. C. & Koenderink, G. H. Structural Hierarchy Governs Fibrin Gel Mechanics. Biophysical Journal 98, 2281–2289 (2010).

25. Hoffman, M. Remodeling the blood coagulation cascade. Journal of thrombosis and thrombolysis 16, 17–20 (2003).

26. Lewis, K. B., Teller, D. C., Fry, J., Lasser, G. W. & Bishop, P. D. Crosslinking Kinetics of the Human Transglutaminase, Factor XIII[A2], Acting on Fibrin Gels and γ-Chain Peptides†. Biochemistry 36, 995–1002 (1997).

27. Bryant, S. J. & Anseth, K. S. Hydrogel properties influence ECM production by chondrocytes photoencapsulated in poly(ethylene glycol) hydrogels. Journal of Biomedical Materials Research 59, 63–72 (2002).

28. Shen, L. L., McDonagh, R. P., McDonagh, J. & Hermans Jr., J. Fibrin gel structure: Influence of calcium and covalent cross-linking on the elasticity. Biochemical and Biophysical Research Communications 56, 793–798 (1974).

29. Blombäck, B., Carlsson, K., Hessel, B., Liljeborg, A., Procyk, R., & Åslund, N. Native fibrin gel networks observed by 3D microscopy, permeation and turbidity. Biochimica et Biophysica Acta (BBA) - Protein Structure and Molecular Enzymology 997, 96–110 (1989).

30. Shen, L. L., Hermans, J., McDonagh, J., McDonagh, R. P. & Carr, M. Effects of calcium ion and covalent crosslinking on formation and elasticity of fibrin gels. Thrombosis Research 6, 255–265 (1975).

31.Ashman, R. B. & Jae Young Rho Elastic modulus of trabecular bone material. Journal of Biomechanics 21, 177–181 (1988).

32. Maeno, S., Niki, Y., Matsumoto, H., Morioka, H., Yatabe, T., Funayama, A., & Tanaka, J. The effect of calcium ion concentration on osteoblast viability, proliferation and differentiation in monolayer and 3D culture. Biomaterials 26, 4847–4855 (2005).

33. Le Nihouannen, D., Saffarzadeh, A., Gauthier, O., Moreau, F., Pilet, P., Spaethe, R., & Daculsi, G. Bone tissue formation in sheep muscles induced by a biphasic calcium phosphate ceramic and fibrin glue composite. J Mater Sci: Mater Med 19, 667–675 (2008).

34. Dao, M., Lim, C. T. & Suresh, S. Mechanics of the human red blood cell deformed by optical tweezers. Journal of the Mechanics and Physics of Solids 51, 2259–2280 (2003).

19

35. Duong, H., Wu, B. & Tawil, B. Modulation of 3D Fibrin Matrix Stiffness by Intrinsic Fibrinogen–Thrombin Compositions and by Extrinsic Cellular Activity. Tissue Engineering Part A 15, 1865–1876 (2009).

36. Gestring, G. F. & Lerner, R. Autologous Fibrinogen for Tissue-Adhesion, Hemostasis and Embolization. Vasc Endovascular Surg 17, 294–304 (1983).

37. Jackson, M. R. Fibrin sealants in surgical practice: An overview. The American Journal of Surgery 182, S1–S7 (2001).

38. G, S. & H, R. Fibrin sealant in orthopedic surgery. Clin Orthop Relat Res 227, 269–285 (1988).

39. Karp, J. M., Sarraf, F., Shoichet, M. S. & Davies, J. E. Fibrin-filled scaffolds for bone-tissue engineering: An in vivo study. Journal of Biomedical Materials Research Part A 71A, 162–171 (2004).

40. Levy, O., Martinowitz, U., Oran, A., Tauber, C. & Horoszowski, H. The Use of Fibrin Tissue Adhesive to Reduce Blood Loss and the Need for Blood Transfusion After Total Knee Arthroplasty. A Prospective, Randomized, Multicenter Study. The Journal of Bone and Joint Surgery (American) 81, 1580–8 (1999).

41. Dvorak, H. F., Harvey, V. S., Estrella, P., Brown, L. F., McDonagh, J., & Dvorak, A. M. Fibrin containing gels induce angiogenesis. Implications for tumor stroma generation and wound healing. Lab Invest 57, 673–686 (1987).

42. Le Nihouannen, D., Guehennec, L. L., Rouillon, T., Pilet, P., Bilban, M., Layrolle, P., & Daculsi, G. Micro-architecture of calcium phosphate granules and fibrin glue composites for bone tissue engineering. Biomaterials 27, 2716–2722 (2006).

43. Engler, A. J., Sen, S., Sweeney, H. L. & Discher, D. E. Matrix Elasticity Directs Stem Cell Lineage Specification. Cell 126, 677–689 (2006).

44. Osathanon, T., Linnes, M. L., Rajachar, R. M., Ratner, B. D., Somerman, M. J., & Giachelli, C. M. Microporous nanofibrous fibrin-based scaffolds for bone tissue engineering. Biomaterials 29, 4091–4099 (2008).

45. Buser, D., Brägger, U., Lang, N. P. & Nyman, S. Regeneration and enlargement of jaw bone using guided tissue regeneration. Clinical Oral Implants Research 1, 22–32 (1990).

46. Kaplan, D. R., Chao, F. C., Stiles, C. D., Antoniades, H. N. & Scher, C. D. Platelet alpha granules contain a growth factor for fibroblasts. Blood 53, 1043–1052 (1979).

20

47. Meyers, K. M., Holmsen, H. & Seachord, C. L. Comparative study of platelet dense granule constituents. Am J Physiol Regul Integr Comp Physiol 243, R454–R461 (1982).

48. McNicol, A. & Israels, S. J. Platelet Dense Granules. Thrombosis Research 95, 1–18 (1999).

49. Haase, H. R., Clarkson, R. W., Waters, M. J. & Bartold, P. M. Growth factor modulation of mitogenic responses and proteoglycan synthesis by human periodontal fibroblasts. Journal of Cellular Physiology 174, 353–361 (1998).

50. Suga, H., Shigeura, T., Matsumoto, D., Inoue, K., Kato, H., Aoi, N., Murase , S., Sato, K., Gonda, K., Koshima, I., & Yoshimura, K. Rapid expansion of human adipose-derived stromal cells preserving multipotency. Cytotherapy 9, 738–745 (2007).

51. Ng, F., Boucher, S., Koh, S., Sastry, K. S., Chase, L., Lakshmipathy, U., Choong, C., Yang, Z., Vemuri, M., Rao, M., & Tanavde, V. PDGF, TGF-β, and FGF signaling is important for differentiation and growth of mesenchymal stem cells (MSCs): transcriptional profiling can identify markers and signaling pathways important in differentiation of MSCs into adipogenic, chondrogenic, and osteogenic lineages. Blood 112, 295–307 (2008).

52. Ponte, A. L., Marais, E., Gallay, N., Langonne, A., Delorme, B., Herault, O., Charbord, P., & Domenech, J. The In Vitro Migration Capacity of Human Bone Marrow Mesenchymal Stem Cells: Comparison of Chemokine and Growth Factor Chemotactic Activities. STEM CELLS 25, 1737–1745 (2007).

53. Drake, C. J., LaRue, A., Ferrara, N. & Little, C. D. Vascular Endothelial Growth Factor Regulates Cell Behavior during Vasculogenesis. Developmental Biology 224, 178–188 (2000).

54. Marx, R. E., Carlson, E. R., Eichstaedt, R. M., Schimmele, S. R., Strauss, J. E., & Georgeff, K. R. Platelet-rich plasma: Growth factor enhancement for bone grafts. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology, and Endodontology 85, 638–646 (1998).

55. Ferrari, M., Zia, S., Valbonesi, M., Henriquet, F., Venere, G., Spagnolo, S., Grasso, M., & Panzani, I. A new technique for hemodilution, preparation of autologous platelet-rich plasma and intraoperative blood salvage in cardiac surgery. Int J Artif Organs 10, 47–50 (1987).

56. Foster, T. E., Puskas, B. L., Mandelbaum, B. R., Gerhardt, M. B. & Rodeo, S. A. Platelet-Rich Plasma From Basic Science to Clinical Applications. Am J Sports Med 37, 2259–2272 (2009).

57. Intini, G. The use of platelet-rich plasma in bone reconstruction therapy. Biomaterials 30, 4956–4966 (2009).

21

58. Messora, M. R., Nagata, M. J. H., Dornelles, R. C. M., Bomfim, S. R. M., Furlaneto, F. A. C., De Melo, L. G. N., & Fucini, S. E. Bone healing in critical-size defects treated with platelet-rich plasma activated by two different methods. A histologic and histometric study in rat calvaria. Journal of Periodontal Research 43, 723–729 (2008).

59. Jensen, T. B., Rahbek, O., Overgaard, S. & Søballe, K. No effect of platelet-rich plasma with frozen or processed bone allograft around noncemented implants. International Orthopaedics (SICOT) 29, 67–72 (2005).

60. González, M. A., Gonzalez–Rey, E., Rico, L., Büscher, D. & Delgado, M. Adipose-Derived Mesenchymal Stem Cells Alleviate Experimental Colitis by Inhibiting Inflammatory and Autoimmune Responses. Gastroenterology 136, 978–989 (2009).

61. Raida, M., Clement, J. H., Leek, R. D., Ameri, K., Bicknell, R., Niederwieser, D., & Harris, A. L. Bone morphogenetic protein 2 (BMP-2) and induction of tumor angiogenesis. J Cancer Res Clin Oncol 131, 741–750 (2005).

62. Caplan, A. I. Mesenchymal stem cells. Journal of Orthopaedic Research 9, 641–650 (1991).

63. Muraglia, A., Cancedda, R. & Quarto, R. Clonal mesenchymal progenitors from human bone marrow differentiate in vitro according to a hierarchical model. J Cell Sci 113, 1161–1166 (2000).

64. Bexell, D., Gunnarsson, S., Tormin, A., Darabi, A., Gisselsson, D., Roybon, L., Scheding, S., & Bengzon, J. Bone Marrow Multipotent Mesenchymal Stroma Cells Act as Pericyte-like Migratory Vehicles in Experimental Gliomas. Molecular Therapy 17, 183–190 (2008).

65. Covas, D. T., Panepucci, R. A., Fontes, A. M., Silva, W. A., Orellana, M. D., Freitas, M. C., Neder, L., Santos, A., Peres, L., Jamur, M. & Zago, M. A. Multipotent mesenchymal stromal cells obtained from diverse human tissues share functional properties and gene-expression profile with CD146+ perivascular cells and fibroblasts. Experimental Hematology 36, 642–654 (2008).

66. Allt, G. & Lawrenson, J. G. Pericytes: Cell Biology and Pathology. Cells Tissues Organs 169, 1–11 (2001).

67. Stolzing, A., Jones, E., McGonagle, D. & Scutt, A. Age-related changes in human bone marrow-derived mesenchymal stem cells: Consequences for cell therapies. Mechanisms of Ageing and Development 129, 163–173 (2008).

22

68. Zuk, P. A., Zhu, M., Mizuno, H., Huang, J., Futrell, J. W., Katz, A. J., Benhaim, P., Lorenz, H., & Hedrick, M. H. Multilineage Cells from Human Adipose Tissue: Implications for Cell-Based Therapies. Tissue Engineering 7, 211–228 (2001).

69. Monaco, E., Bionaz, M., Hollister, S. J. & Wheeler, M. B. Strategies for regeneration of the bone using porcine adult adipose-derived mesenchymal stem cells. Theriogenology 75, 1381–1399 (2011).

70. Gimble, J. M., Katz, A. J. & Bunnell, B. A. Adipose-Derived Stem Cells for Regenerative Medicine. Circulation Research 100, 1249–1260 (2007).

71. Gnecchi, M., Zhang, Z., Ni, A. & Dzau, V. J. Paracrine Mechanisms in Adult Stem Cell Signaling and Therapy. Circulation Research 103, 1204–1219 (2008).

72. Liu, Y., Yan, X., Sun, Z., Chen, B., Han, Q., Li, J., & Zhao, R. C. Flk-1 Adipose-Derived Mesenchymal Stem Cells Differentiate into Skeletal Muscle Satellite Cells and Ameliorate Muscular Dystrophy in MDX Mice. Stem Cells and Development 16, 695–706 (2007).

73. Yanez, R., Lamana, M. L., García‐Castro, J., Colmenero, I., Ramirez, M., & Bueren, J. A. Adipose Tissue-Derived Mesenchymal Stem Cells Have In Vivo Immunosuppressive Properties Applicable for the Control of the Graft-Versus-Host Disease. Stem Cells 24, 2582–2591 (2006).

74. Kilroy, G. E., Foster, S. J., Wu, X., Ruiz, J., Sherwood, S., Heifetz, A., & Gimble, J. M. Cytokine profile of human adipose-derived stem cells: Expression of angiogenic, hematopoietic, and pro-inflammatory factors. Journal of Cellular Physiology 212, 702–709 (2007).

75. Yoshimura, K., Sato, K., Aoi, N., Kurita, M., Hirohi, T., & Harii, K. Cell-Assisted Lipotransfer for Facial Lipoatrophy: Efficacy of Clinical Use of Adipose-Derived Stem Cells. Dermatologic Surgery 34, 1178–1185 (2008).

76. Lendeckel, S., Jödicke, A., Christophis, P., Heidinger, K., Wolff, J., Fraser, J. K., & Howaldt, H. P. Autologous stem cells (adipose) and fibrin glue used to treat widespread traumatic calvarial defects: case report. Journal of Cranio-Maxillofacial Surgery 32, 370–373 (2004).

77. Mesimäki, K., Lindroos, B., Törnwall, J., Mauno, J., Lindqvist, C., Kontio, R., & Suuronen, R. Novel maxillary reconstruction with ectopic bone formation by GMP adipose stem cells. International Journal of Oral and Maxillofacial Surgery 38, 201–209 (2009).

78.Tse, W. T. 1, Pendleton, J. D., Beyer, W. M., Egalka, M. C. & Guinan, E. C. Suppression of allogeneic T-cell proliferation by human marrow stromal cells: implications in transplantation. [Miscellaneous Article]. Transplantation February 15, 2003 75, 389–397 (2003).

23

79. Wilson, S. M., Goldwasser, M. S., Clark, S. G., Monaco, E., Bionaz, M., Hurley, W. L., Rodriguez-Zas, S., Feng, L., Dymon, Z., & Wheeler, M. B. Adipose-Derived Mesenchymal Stem Cells Enhance Healing of Mandibular Defects in the Ramus of Swine. Journal of Oral and Maxillofacial Surgery 70, e193–e203 (2012).

80. Keibl, C., Fügl, A., Zanoni, G., Tangl, S., Wolbank, S., Redl, H., & Van Griensven, M. Human adipose derived stem cells reduce callus volume upon BMP-2 administration in bone regeneration. Injury 42, 814–820 (2011).

81. Peterson, B., Zhang, J., Iglesias, R., Kabo, M., Hedrick, M., Benhaim, P., & Lieberman, J. R. Healing of Critically Sized Femoral Defects, Using Genetically Modified Mesenchymal Stem Cells from Human Adipose Tissue. Tissue Engineering 11, 120–129 (2005).

82. Chou, Y.-F., Zuk, P. A., Chang, T.-L., Benhaim, P. & Wu, B. M. Adipose-derived stem cells and BMP2: Part 1. BMP2-treated adipose-derived stem cells do not improve repair of segmental femoral defects. Connective Tissue Research 52, 109–118 (2011).

24

CHAPTER 2

IN VITRO OSTEOGENESIS ON FIBRIN SCAFFOLDS

2.1 Overview

This chapter covers the process of collecting modified fibrin scaffolds, analysis of their

mechanical properties, and culture with ASCs for osteogenic differentiation. The primary aim

was to develop a fibrin scaffold for optimum in vitro osteogenic differentiation of porcine ASCs.

Fibrin scaffolds physically and chemically modified to be stiffer or have a higher concentration

of calcium and phosphate ions were hypothesized to enhance osteogenic differentiation

compared to control fibrin scaffolds. Treatments during coagulation resulted in 6 scaffold types

for comparison: whole blood controls, red blood cell lysis buffer, calcium chloride, calcium

hydrogen phosphate, vacuum, and mechanical compression. Fibrin scaffolds treated with red

blood cell lysis buffer and calcium hydrogen phosphate were determined to be stiffer compared

to untreated controls. In addition, treatment with calcium phosphate was found to accelerate

coagulation. Osteogenic differentiation was enhanced on scaffolds treated with calcium chloride

and calcium hydrogen phosphate. It is likely these results are explained in part by ASC

attachment and fibrin polymerization during coagulation. Based on these results, calcium

phosphate was selected as the method to modify fibrin for the in vivo study (Chapter 4).

2.2 Methods

2.2.1 Blood Collection and Fibrin Modification

Whole blood was collected in sterile 1-liter containers from 6 pigs and allowed to clot

with different treatments, which included solutions (2.4 and 7.2 mM) of calcium chloride (CaCl2,

Sigma C7902, St. Louis, MO), or calcium hydrogen phosphate (CaHPO4, Mallinckrodt #4272,

Hazelwood, MO), a precursor to completed bone mineral (hydroxyapatite, Ca5(PO4)3OH).

Highly flexible red blood cells (RBCs) were removed osmotically with red blood cell lysis buffer

(25% or 50% to blood, 8.3 g/L ammonium chloride, Sigma R7757). To reduce porosity, samples

25

were mechanically compressed under pressure (200 MPa) in a syringe using free weights or

placed under vacuum (130 kPa) during coagulation. Samples were stored at -20°C [1].

2.2.2 Coagulation Time

During whole blood collection, samples from 4 pigs were timed to determine the rate of

coagulation under various treatments. Blood was collected in 50 mL centrifuge tubes and

agitated manually. Timing began upon collection in an individual tube and ended upon cessation

of motion during tube rocking [2]. Coagulation time was normalized to untreated whole blood

for each individual pig.

2.2.3 In Vitro Degradation

Samples were cut into approximately 2 gram sections and pressed flat onto 10 cm2 tissue

culture plates with 5 mL of Dulbecco’s Modified Eagle’s Medium (DMEM, Sigma D5648) and

varying concentrations (0, 1, 5, 10%) of fetal bovine serum (FBS, BenchMark, Gemini Bio-

Products, Sacramento, CA) [3]. At weekly intervals, samples were digitally scanned to observe

degradation. Surface area of each scaffold was calculated using an image analysis program

(ImageJ, National Institutes of Health, Bethesda, MA). Area was normalized by its original value

(day 0) to calculate percent change in area. After 28 days, samples were fixed in 10% formalin

(Fisher Scientific, Fair Lawn, NJ).

Microstructural changes were compared using scanning electron microscopy (SEM).

Samples were dehydrated daily in an increasing ethanol series (70%, 80%, 90%, 100%) and

stored in a desiccation chamber overnight. Samples were then critical-point dried using liquid

carbon dioxide (Tousimis, Cleveland, OH). After applying a 50 mbar vacuum and inducing

eccentric rotation, samples were sputter coated with gold-palladium (Au-Pd) for 45 seconds [4].

Samples were imaged using a Phillips XL30 (FEI Company, Hillsboro, OR) SEM. Beam voltage

was 10 kV with a spot size of 5. Images were recorded at 500x magnification.

2.2.4 Mechanical Testing

Test specimens were prepared as right circular cylinders 25 mm in diameter and 30 mm

in height for mechanical testing (Instron Model 5567, Grove City, PA) [5]. A compression test at

26

constant velocity (1 mm/min) was performed on hydrated samples (n = 5). Original dimensions

were used to calculate stress and strain, and resulting curves were used to determine elastic

modulus. Analysis was limited to the initial linear region of 10% strain [6].

Cylindrical samples of 1 cm diameter and 1 cm height were prepared in centrifuge tube

caps for nanoindentation (n = 4). A TI950 Hysitron Tribometer (Minneapolis, MN), using a 90°

conical PMMA tip, indented samples 10 µm in depth to determine relative sample stiffness

(Figure 2.4). Stiffness was calculated as slope of tip removal curve between 40-90%

displacement [7].

2.2.5 Adipose-Derived Stem Cell Co-Culture and Osteogenic Differentiation

Fibrin scaffolds were thawed and sectioned (1 mm thickness). Sections were sterilized

using ethylene oxide gas (Anprolene, Andersen Products, Haw River, NC) exposure overnight

before washing with phosphate buffered saline (PBS, Sigma D5773). Scaffolds were placed in

tissue culture plates and cultured overnight under standard conditions before addition of passage

2-3 ASCs (1x105 cells per cm2), which were previously isolated from a pig transgenic for green

fluorescent protein (GFP) [8]. ASCs were cultured in DMEM with 10% FBS incubated with 5%

carbon dioxide at 39°C. Cell attachment to fibrin was observed with a fluorescent microscope.

At 80% confluence, media was supplemented with 10 mM β-glycerophosphate (Sigma

G9891), 50 µM ascorbic acid (Sigma A4403) and 1 µM dexamethasone (Sigma D4902) to

induce osteogenic differentiation [9]. Cells were cultured for 0, 14, and 28 days before fixation in

10% formalin. Circular nodules form during osteogenic differentiation, which were imaged using

a fluorescent microscope (Nikon Diaphot-TMD, Melville, NY) and digital camera (Nikon

DXM1200). After fluorescent imaging, average area per nodule (n=30) was quantified using

ImageJ (National Institutes of Health, Bethesda, MA). Average nodule radius was calculated

assuming a circular nodule, using the equation [9]:

Radius = sqrt(Area / π)

27

2.2.6 Statistical Analysis

Quantitative data are presented as mean ± standard error. One-way analysis of

variance (ANOVA) was performed using SAS 9.2 statistical analysis software (SAS Institute,

Cary, NC). Fisher’s exact test was used for pairwise mean comparisons. A total of 3 scaffolds

per treatment per timepoint were analyzed for a total of 18 scaffolds and 30 nodules per

treatment per timepoint (n = 30). Significance was evaluated using an alpha level of 0.05.

2.3 Results

2.3.1 Coagulation Time

Average coagulation time of untreated whole blood was 121 seconds (Figure 2.1a).

Treatment with calcium phosphate reduced coagulation time by 25% (p = 0.25) while treatment

with red blood cell lysis buffer increased clotting time by 35% (p = 0.081, Figure 2.1b).

Following normalization, differences between treatments were significant (calcium phosphate p

= 0.022, RBC lysis p = 0.0071). Calcium chloride treatments could not be assessed due to low

stiffness, as samples remained in motion during manual agitation even after completing

coagulation.

2.3.2 In Vitro Degradation

Evidence of bulk degradation was present in most treatments (Figure 2.2). In addition,

increasing fetal bovine serum concentration generally resulted in a higher amount which was

degraded. Control blood clots had 98% remaining after 28 days in culture with serum-free

DMEM (0% FBS) and 80% remaining after culture in DMEM with 10% FBS (Figure 2.3). Clots

treated with calcium chloride were more resistant to degradation (92% remaining, p = 0.097)

while clots treated with red blood cell lysis buffer were less resistant (74% remaining, p = 0.14).

Analysis of scaffolds using scanning electron microscopy (SEM, Figure 2.4) displayed

limited structural differences for whole blood, calcium chloride, and red blood cell lysis buffer

scaffolds. Calcium phosphate treatment resulted in a more granular surface with notable

roughness. Vacuum and compression-treated scaffolds appeared smoother.

28

2.3.3 Mechanical Testing

Measured by compression testing, the average elastic modulus of fibin scaffolds (Figure

2.5) derived from whole blood was 1.9 kPa. Scaffolds treated with RBC lysis buffer were higher

in stiffness by 25% (p = 0.092). In addition, scaffolds treated with CaHPO4 had a similar higher

stiffness (p = 0.089). Scaffolds coagulated under vaccum or mechanical compression showed no

change in stiffness. Treatment with CaCl2 significantly reduced stiffness by 37% (p = 0.025,

Figure 2.6).

Nanoindentation (Figure 2.7) stiffness measurements were significantly higher for

scaffolds treated with calcium phosphate (average 94 kPa, p = 0.00024) compared to whole

blood (47 kPa). Scaffolds treated with red blood cell lysis buffer also were higher compared to

whole blood (p = 0.095). Normalization by pig (Figure 2.8) resulted in stiffness nearly two times

higher for calcium phosphate (p = 0.0053) and a trend higher for red blood cell lysis buffer (p =

0.064). Scaffolds from other treatments could not be accurately measured due to low stiffness.

2.3.4 Osteogenic Differentiation

GFP-ASCs attached onto fibrin within 2 days and differentiated into an osteogenic

phenotype when cultured in media supplemented with dexamethasone, β-glycerophosphate and

ascorbic acid (Figure 2.9). Over time, bone-like nodules formed which were observed using

fluorescent microscopy. Osteogenic nodules were observed by 7 days and increased in density

until 14 days. Nodules increased in size through 21 days, then appeared to grow more slowly

until 28 days.

Average area per nodule (Figure 2.10) and average radius (Figure 2.11) generally were

lower on fibrin derived from whole blood compared to control polystyrene (tissue culture plastic,

p = 0.029). However, nodules on fibrin scaffolds treated with calcium phosphate (p = 0.042)

were greater in size compared to control whole blood.

29

Figure 2.1: Coagulation time of blood in a centrifuge tube (a). Times normalized relative to untreated whole blood for each individual pig (b). Treatments with different superscripts significantly differ (p < 0.05).

a

b

b

c

a

b

a, b

a

30

Figure 2.2: Representative 2-D cross-sections of fibrin scaffolds after 28 days in culture. Presence of FBS increased degradation rate, which varied depending on scaffold treatment.

Figure 2.3: Relative change in cross-sectional area during culture in 10% fetal bovine serum. Scaffold treatment affected rate of degradation, but treatments remained stable over the timeframe of osteogenic differentiation. Treatments with different superscripts significantly differ (p < 0.05).

a

a

a

b

a, b

a, b a, b

31

Figure 2.4: Representative scanning electron microscope (SEM) images in fibrin scaffolds. While whole blood, calcium chloride, and red blood cell lysis buffer scaffolds displayed limited differences in structure, calcium phosphate treatment resulted in a more granular surface with notable roughness. Vacuum and compression-treated scaffolds appeared smoother.

32

Figure 2.5: Experimental set ups for compression testing (left) included preparing right circular cylinders 25 mm in diameter and 30 mm in height and compressing samples using an Instron Model 5567. Calculated stresses and strains were used to determine elastic modulus. Nanoindentation (right) used samples 1 cm in diameter and 1 cm in height. Samples were indented using a 90° conical PMMA tip, indenting 10 µm in depth. Stiffness was calculated as slope of tip removal curve between 40-90% displacement.

Figure 2.6: Compression testing results displayed higher stiffness in scaffolds treated with RBC lysis buffer and calcium hydrogen phosphate. Scaffolds coagulated under vacuum or mechanical compression showed little difference and scaffolds treated with calcium chloride were lower in stiffness. Treatments with different superscripts significantly differ (p < 0.05).

a

b

a

a

a, b a

33

Figure 2.7: Compared to whole blood, nanoindentation results were significantly higher in stiffness for scaffolds treated with calcium phosphate. Scaffolds treated with red blood cell lysis buffer trended higher compared to whole blood but not significantly. Treatments with different superscripts significantly differ (p < 0.05).

Figure 2.8: Stiffness normalized by individual pig were significantly higher in stiffness for scaffolds treated with calcium phosphate. Scaffolds treated with red blood cell lysis buffer trended higher compared to whole blood but not significantly. Treatments with different superscripts significantly differ (p < 0.05).

b

b

a

a

a

a

34

Figure 2.9: Paired phase contrast and flourescent images of ASCs (white arrows) expressing green fluorescent protein attaching to (a,b and c,d) and differentiating (e,f) on fibrin scaffolds. Scale a and b – 25 µm. Scale c and d – 50 µm. Scale e and f – 100 µm.

35

Figure 2.10: Average nodule area during osteogenic differentiation on scaffolds of various treatments. Treatments with different superscripts significantly differ (p < 0.05).

a, b

a

a

b b

b

a, b

36

Figure 2.11: Average nodule radius during osteogenic differentiation on scaffolds of various treatments. Treatments with different superscripts significantly differ (p < 0.05).

0

20

40

60

80

100

120

140

160

Polystyrene Whole Lysis CaCl2 CaHPO4 vacuum compression

Aver

age

Nod

ule

Rad

ius (

µm)

a

b

a, b a, b

a

b

c

37

2.4 Discussion

The primary objective of these experiments was to develop an optimal preparation of

whole blood-derived fibrin which was practical, biocompatible, and osteoconductive. Based on

the results, fibrin supplemented with granules of calcium hydrogen phosphate (CaHPO4) was

determined to be the most suitable formulation of those examined and was appropriated for use

in the in vivo experiment (Chapter 4). Using calcium phosphate resulted in a fibrin scaffold

which coagulated faster (p = 0.022), had a rougher surface, and higher stiffness, which are

desirable properties for practical usage during surgical operations and for scaffolds used in bone

tissue engineering.

Calcium phosphates and calcium chlorides act as natural pro-coagulants due to increased

concentrations of soluble calcium ions, an essential element for activation of both factor X and

thrombin in the coagulation cascade [10]. However, these ionic compounds are on separate

extremes of the solubility scale, with CaCl2 being almost completely soluble in water while

CaHPO4 has limited solubility, though comparatively high among calcium phosphates.

Differences in solubility affect bioavailability of calcium ions and thus modify their effect during

fibrin polymerization. From calcium chloride, the increased soluble calcium ion concentration

likely allows for more simultaneous initiation sites of fibrin polymerization [11], resulting in a

scaffold that gels more rapidly with lower average chain length and reduced stiffness.

For calcium phosphate, this effect is mitigated by fewer available calcium ions, while

providing relatively insoluble nucleation sites for the initiation and stabilization of fibrin

polymerization. Initial contact activation is the longest phase of the coagulation process [2].

Addition of calcium phosphate granules significantly increases available surface area for contact

activation and is likely to account for the observed reduction in coagulation time with calcium

phosphate. In addition, concentration of fibrinogen also has an important effect on coagulation

with lower concentrations resulting in fibrin gels which polymerize slower. Dilution of

fibrinogen and other coagulation factors by red blood cell lysis buffer is likely to explain its

slower coagulation time.

For both in vitro and in vivo environments, fibrinolysis proceeds through two related

mechanisms, bulk and enzymatic degradation [12]. Bulk degradation is the result of water

infiltration and hydrolysis, reducing chain lengths and network connectivity [13]. Enzymatic

38

degradation proceeds primarily through the enzyme plasmin, which circulates in an inactive form

in the blood [12] and is present in fetal bovine serum. Addition of FBS to culture media

increased fibrinolysis regardless of treatment, indicating the significant role of enzymatic

degradation, while bulk degradation was slower in comparison.

Fibrin scaffolds treated with calcium chloride or coagulated under vacuum displayed

notable reductions in degradation (higher area remaining) compared to fibrin scaffolds derived

from whole blood. Though multiple mechanisms may determine this result, the chemical

property of shorter average chain length during is likely to be a dominant factor for calcium

chloride treatment [14]. However, for vacuum coagulation, the physical property of reduced

porosity aids in reducing water infiltration and surface exposure to plasmin, slowing both bulk

and enzymatic degradation [15]. In contrast, scaffolds treated with red blood cell lysis buffer

likely had increased water infiltration and plasmin exposure, due to the osmotic dissolution of

red blood cells, resulting in was higher (p = 0.097) degradation (lower area remaining) compared

to controls.

For attachment-dependent cell types such as adipose-derived stem cells, a rough surface

is generally more preferable than a smooth surface [16]. Surface roughness provides more

opportunities for multidirectional cell attachments more similar to that experienced by cells in

vivo, as well as increased surface area for attachment and material transport. For fibrin scaffolds,

roughness is primarily due to red blood cells (7 µm average diameter). Red blood cells likely

account for the surface roughness viewable by scanning electron microscopy for whole blood

and calcium chloride treatments, while osmotic destruction of red blood cells by red blood cell

lysis buffer may explain the slight increase in surface roughness. However, ASCs generally

cannot directly attach to red blood cells. In contrast, attachment-dependent cell types are capable

of attachment to calcium phosphate granules [17], which have granule sizes on a similar scale.

Addition of granules appears to provide increased surface roughness to fibrin scaffolds. Use of

vacuum or mechanical compression appears to flatten red blood cells, causing fibrin to appear

smoother.

In addition, adipose-derived stem cells, among other fibroblast-like cell types, are not

known to maintain receptors for direct attachment to fibrin. Therefore, cell attachment is likely to

proceed through intermediates such as fibronectin [18], which circulates in blood along with

39

fibrinogen and becomes entrapped within the developing blood clot during coagulation [19].

Reduced direct attachment to fibrin may help explain in part the limited success in the use of

fibrin-only glues for tissue regeneration. ASCs appeared to viably attach and differentiate on

whole blood derived fibrin scaffolds.

Fibrin scaffolds derived from whole blood maintain several advantages over those

prepared from only fibrinogen and thrombin. These include simple and inexpensive collection in

addition to retaining essential blood constituents in the wound healing process such as

fibronectin and platelet aggregates [20]. However, the use of whole blood-derived fibrin comes

at the cost of reduced operative control of fibrin polymerization as well as reduced stiffness of

the clot due to the presence of red blood cells.

Bulk stiffness of fibrin scaffolds, as measured by compression testing, did not

appreciably increase from a practical standpoint regardless of treatment, and remains several

orders of magnitude lower than trabecular bone [21]. However, microscale stiffness, as measured

by nanoindentation, significantly increased with calcium phosphate treatment compared to

controls. The effect of calcium phosphate may be related to its inherently high stiffness as a

ceramic as well as the potential of it granules to act as anchors for the fibrin network. The

mechanical properties of fibrin-calcium phosphate composites can be approximated as a

weighted average of the two substances [22]. Increased stiffness at the microscale has been

demonstrated to favor osteogenic differentiation of adipose-derived stem cells, whereas softer

surfaces favor other cell types, such as neurons and myocytes [23]. The mechanisms for this

phenomenon are under investigation, but are related to the interdependent relationship between

mechanical forces on the extracellular matrix and gene expression.

A difference in measured stiffness values of approximately one order of magnitude was

observed between the methods of compression testing and nanoindentation. These results are

likely explained by the hydrated, physiological testing conditions and the role of water in fibrin

viscoelasticity [24]. Surface tension is a much more dominant force in nanoindentation

measurements. Water is allowed to flow more freely under a bulk compression testing regimen,

which may reduce observed elastic (storage) modulus. Application of a sinusoidal stress, used in

methods such as dynamic mechanical analysis [25], could potentially rectify this difference in

40

observed values and better account for the viscoelastic behavior of fibrin scaffolds, which are

similar to hydrogel polymers.

In terms of osteoconductivity, one notable result is the tendency of ASCs to form larger

nodules on tissue culture plastic (irradiated polystyrene) compared to whole blood-derived fibrin.

This result is likely explained in part by methods used in ASC isolation [9], which includes

expansion of cells which adhere onto tissue culture plastic. While starting as a heterogeneous cell

population, the cells which attach and divide most rapidly dominate in number [26]. Therefore,

ASC isolation methods introduce a strong selection for attachment to polystyrene.

The ability of ASCs to attach and migrate towards each other is essential for the

formation of osteogenic nodules over a timescale of weeks. The potential factors of improved

cell attachment, presence of calcium and phosphate ions, surface roughness, and increased

stiffness on the microscale may each play a role in the increase in osteoconductivity of ASCs for

fibrin treatments of red blood cell lysis buffer, calcium chloride and calcium phosphate, as

measured by histomorphometry. Treatment of fibrin scaffolds with calcium phosphate appears to

improve each of these factors, resulting in the decision to utilize calcium phosphate as a

supplement to fibrin for the in vivo experiment.

2.5 References

1. Pieters, M., Jerling, J. C. & Weisel, J. W. Effect of freeze-drying, freezing and frozen storage

of blood plasma on fibrin network characteristics. Thrombosis Research 107, 263–269 (2002).

2. Hattersley PG Activated coagulation time of whole blood. JAMA 196, 436–440 (1966).

3. Marder, V. J. & Francis, C. W. Plasmin Degradation of Cross-Linked Fibrin. Annals of the

New York Academy of Sciences 408, 397–406 (1983).

4. Le Nihouannen, D., Guehennec, L. L., Rouillon, T., Pilet, P., Bilban, M., Layrolle, P., &

Daculsi, G. Micro-architecture of calcium phosphate granules and fibrin glue composites for

bone tissue engineering. Biomaterials 27, 2716–2722 (2006).

41

5. Johnson, B., Bauer, J. M., Niedermaier, D. J., Crone, W. C. & Beebe, D. J. Experimental

techniques for mechanical characterization of hydrogels at the microscale. Experimental

Mechanics 44, 21–28 (2004).

6. Kong, H.-J., Lee, K. Y. & Mooney, D. J. Decoupling the dependence of

rheological/mechanical properties of hydrogels from solids concentration. Polymer 43, 6239–

6246 (2002).

7. Oyen, M. l. Nanoindentation of Biological and Biomimetic Materials. Experimental

Techniques 37, 73–87 (2013).

8. Liu, Z., Song, J., Wang, Z., Tian, J., Kong, Q., Zheng, Z., & Prather, R. S. Green fluorescent

protein (GFP) transgenic pig produced by somatic cell nuclear transfer. Chin. Sci. Bull. 53,

1035–1039 (2008).

9. Monaco, E., Lima, A., Bionaz, M., Maki, A., Wilson, S., Hurley, W. L., & Wheeler, M.

B.Morphological and Transcriptomic Comparison of Adipose and Bone Marrow Derived

Porcine Stem Cells. Open Tissue Engineering & Regenerative Medicine Journal 2, 20–33

(2009).

10. Davie, E. W., Fujikawa, K. & Kisiel, W. The coagulation cascade: initiation, maintenance,

and regulation. Biochemistry 30, 10363–10370 (1991).

11. Boyer, M. H., Shainoff, J. R. & Ratnoff, O. D. Acceleration of Fibrin Polymerization by

Calcium Ions. Blood 39, 382–387 (1972).

12. Wiman, B. & Collen, D. Molecular mechanism of physiological fibrinolysis. 6 April 1978;

272, 549–550 (1978).

13. Göpferich, A. Polymer Bulk Erosion. Macromolecules 30, 2598–2604 (1997).

14. Blomback, B., Hogg, D. H., Gårdlund, B., Hessel, B. & Kudryk, B. Fibrinogen and fibrin

formation. Thrombosis Research 8, Supplement 2, 329–346 (1976).

42

15. Kolev, K., Tenekedjiev, K., Komorowicz, E. & Machovich, R. Functional Evaluation of the

Structural Features of Proteases and Their Substrate in Fibrin Surface Degradation. J. Biol.

Chem. 272, 13666–13675 (1997).

16. Lampin, M., Warocquier-Clérout, R., Legris, C., Degrange, M. & Sigot-Luizard, M. F.

Correlation between substratum roughness and wettability, cell adhesion, and cell migration.

Journal of Biomedical Materials Research 36, 99–108 (1997).

17. Deligianni, D. D., Katsala, N. D., Koutsoukos, P. G. & Missirlis, Y. F. Effect of surface

roughness of hydroxyapatite on human bone marrow cell adhesion, proliferation, differentiation

and detachment strength. Biomaterials 22, 87–96 (2000).

18. Clark, R. A., Lanigan, J. M., DellaPelle, P., Manseau, E., Dvorak, H. F., & Colvin, R. B.

Fibronectin and Fibrin Provide a Provisional Matrix for Epidermal Cell Migration During

Wound Reepithelialization..Investigative Dermatology 79, 264–269 (1982).

19. Proctor, R. A. Fibronectin: A Brief Overview of Its Structure, Function, and Physiology.

Clinical Infectious Diseases 9, S317–S321 (1987).

20. Rand, M. D., Lock, J. B., Veer, C. van’t, Gaffney, D. P. & Mann, K. G. Blood clotting in

minimally altered whole blood. Blood 88, 3432–3445 (1996).

21. Ashman, R. B. & Jae Young Rho Elastic modulus of trabecular bone material. Journal of

Biomechanics 21, 177–181 (1988).

22. Mooney, R. G., Costales, C. A., Freeman, E. G., Curtin, J. M., Corrin, A. A., Lee, J. T.,

Reynolds, S., Tawil, B., & Shaw, M. C. Indentation micromechanics of three-dimensional

fibrin/collagen biomaterial scaffolds. Journal of Materials Research 21, 2023–2034 (2006).

23. Engler, A. J., Sen, S., Sweeney, H. L. & Discher, D. E. Matrix Elasticity Directs Stem Cell

Lineage Specification. Cell 126, 677–689 (2006).

24. Fukada, E. & Kaibara, M. Rheological measurements of fibrin gels during clotting.

Thrombosis Research 8, Supplement 2, 49–58 (1976).

43

25. Weisel, J. W. The mechanical properties of fibrin for basic scientists and clinicians.

Biophysical Chemistry 112, 267–276 (2004).

26. Oedayrajsingh-Varma, M. J., Van Ham, S. M., Knippenberg, M., Helder, M. N., Klein-

Nulend, J., Schouten, T. E., & Van Milligen, F. J Adipose tissue-derived mesenchymal stem cell

yield and growth characteristics are affected by the tissue-harvesting procedure. Cytotherapy 8,

166–177 (2006).

44

CHAPTER 3

IN VITRO MIGRATION IN PLATELET-RICH PLASMA

3.1 Overview

This chapter details the protocol for collecting platelet-rich plasma (PRP), its

characterization, and its effects on the proliferation and migration of porcine adipose-derived

stem cells (ASCs). The primary aim of these studies was to analyze in vitro ASC migration in

varying concentrations of growth factors derived from the platelet-rich fraction of centrifuged

plasma. ASCs were hypothesized to migrate at a faster average velocity with increasing PRP

concentration. Whole blood was collected with sodium citrate anticoagulant and underwent two

centrifugations. The first slower spin (905 xg) was to remove red blood cells to collect plasma

and the second faster spin (2510 xg) was to collect the fraction with a high platelet concentration.

Platelet concentration increased 3.5-fold compared to plasma, within the prescribed therapeutic

range. For long-term culture of ASCs, dilution with DMEM to concentrations less than 30% was

found to be necessary for cell viability. ASC rate of proliferation was comparable to that in fetal

bovine serum (FBS), a standard cell culture media supplement. In addition, velocity of ASC

migration increased in cultures supplemented with 20% or 30% PRP. Generally, PRP was

determined to be a media supplement with similar effects as FBS, potentially making it a suitable

substitute for in vitro expansion of ASC populations. It is likely these results are explained in

part by similarities in growth factor concentrations and their effects. Based on these results, 20%

PRP was selected as the concentration for the in vivo study (Chapter 4).

3.2 Methods

3.2.1 Blood Collection and Processing

PRP was collected in sterile 1-liter containers from 4 pigs using sodium citrate (3.3%,

Sigma W302600, St. Louis, MO) as the anticoagulant at a ratio of 8 mL blood to 1 mL

anticoagulant. An initial centrifugation spin was performed at 1800 rpm (905 xg) for 15 minutes

for removal of RBCs and collection of plasma. A subsequent spin of collected plasma at 3000

45

rpm (2510 xg) for 10 minutes separated platelets [1]. The lower half was considered PRP while

the upper fraction was considered platelet-poor plasma (P3). Samples were stored at -20°C.

3.2.2 Platelet Concentrations

Plasma, PRP and P3 were assessed for platelet concentration using a hemocytometer and

phase contrast microscope (Nikon Diaphot-TMD, Melville, NY). Platelets were counted at 400x

magnification in triplicate.

3.2.3 Adipose-Derived Stem Cell Proliferation

Serum was thawed to room temperature, and then heated to 56°C for 30 minutes to

inactivate complement. Platelets were degranulated using 2.4 mM calcium chloride (Sigma

C7902) before filtration (0.2 µm syringe filters, Nalgene, Rochester, NY). ASCs were thawed

and cultured in DMEM (Sigma D5648) supplemented with FBS (BenchMark, Gemini Bio-

Products, Sacramento, CA), PRP, or P3 at concentrations of 10% or 20%. At passage 3-4, cells

were seeded at 1.0 x 104 cells/mL in 6-well plates. Cells were trypsinized and counted in

duplicate on a hemocytometer after 0, 1, 2, 4, 6 and 10 days. Average doubling time (Td) was

calculated using the equation [2,3]:

Td = (t2 – t1) * log (2) / log (q2/q1)

using counts (q1,2) from day 2 (t1) and day 6 (t2) time points during the logarithmic growth phase.

3.2.4 Transwell Migration Assay

A polyethylene terephthalate (PET) 8-µm pore insert for 6-well plates (BD Biosciences,

San Jose, CA) was used to assess ASC migration. ASCs were starved in serum-free DMEM

overnight before seeding (1 x 105 cells/mL) onto the top migration well [4]. The lower chamber

contained DMEM (control), FBS, PRP, or P3 (10, 20%). After incubation at 39°C in 5% CO2 for

3, 8 and 24 hours, ASCs on the bottom well were counted, performed at 100x magnification in

duplicate.

46

3.2.5 PDMS Stamp Production

Poly-dimethylsiloxane (PDMS, Dow Corning, Midland, MI) stamps were produced using

standard soft lithographic methods, which consist of a silicon master pattern to produce PDMS

molds [5]. In a clean room (Beckman Institute), the master was formed using a photomask

printed with the 2 mm circular patterns which selectively blocked the passage of UV light onto

an epoxy photoresist (SU-8, MicroChem, St. Newton, MA). Photoresist was coated onto silicon

wafers at 5 µm thickness using a 1000 rpm spin speed and 1 minute pre-bake at 65°C. Sections

of photoresist exposed to UV light polymerized and attached to the silicon wafer while

unexposed portions were subsequently washed away using 70% ethanol, resulting in a negative

of the original photomask. Wafers with 2 mm photoresist patterns were baked at 95°C for 1

minute to complete polymerization.

After liquid monomer was mixed in a 10:1 mass ratio with catalyst (Sylgard, Dow

Corning), PDMS was poured onto the master. Following removal of bubbles by vacuum, PDMS

was allowed to set under 50 kg of weight overnight. Heating at 60°C for 2 hours completed

polymerization. PDMS formed a mold which matched the pattern of the original photomask.

3.2.6 Stamp Migration Assay

PDMS stamps were sterilized with 70% ethanol and allowed to air dry before placing in

12-well plates. Serum-starved ASCs (1.0 x 107 cells/mL) were loaded in each well and cultured

in serum-free DMEM for 4 hours to allow cell attachment. Treatments included medium

consisting of DMEM with varying PRP (0, 10, 20, 50, 80, 90, 100%) or FBS (10, 20%)

concentrations. Plates were placed into a specialized incubator equipped with a phase contrast

microscope (Olympus WeatherStation, Precision Control, Tacoma, WA) and a camera capable of

time-lapse imaging to track cell migration. Images were captured every 15 minutes for 48 hours.

Displacement and distance traveled of randomly selected ASCs over time (n = 30) were

measured using ImageJ [6]. Average displacement was used to calculate the average ASC

velocity.

47



variance (ANOVA) was performed using SAS statistical analysis software (SAS Institute, Cary,

NC). Fisher’s exact test was used for pairwise mean comparisons. A total of 30 cells per

treatment (n = 30) were analyzed. Significance was evaluated using an alpha value of 0.05.

3.3 Results

3.3.1 Platelet Concentration

For every 5 mL of whole blood, 1 mL of platelet rich plasma was acquired (Figure 3.1).

Through centrifugation, platelet concentration of PRP was significantly higher (p < 0.001) by a

factor of 3.5 over plasma from the same animal (Figure 3.2).

3.3.2 Adipose-Derived Stem Cell Proliferation

Addition of serum or plasma to DMEM resulted in a trend of shorter average doubling

time of ASCs (Figure 3.3). While limited for platelet-poor plasma (P3, p = 0.17), the effect was

more pronounced for media supplemented with PRP (p = 0.073) or FBS (p = 0.091). Increasing

PRP concentration appeared to have a limited effect on doubling time (p = 0.069, Figure 3.4).

3.3.3 Transwell Migration Assay

After 24 hours, ASC migration towards 20% PRP was 2.8 times higher than DMEM

controls (p = 0.023). This effect was similarly observed in 20% FBS (p = 0.031), while migration

was more limited towards 20% P3 (p = 0.035).

3.3.4 Stamp Migration Assay

After tracking for 48 hours (Figure 3.6), average cell velocity (Figure 3.7) was higher in

media supplemented with FBS (p= 0.0014) or PRP (p = 0.0013). Velocity was significantly

higher at 20% PRP compared to 10% PRP (p = 0.0003) with no significant difference between

30% PRP and 20% PRP (p = 0.42). ASCs were not viable long-term at concentrations above

30% PRP.

48

Figure 3.1: Platelet-rich plasma centrifugation process, (a) whole blood with 3.3% sodium citrate anticoagulant. (b) Separation into plasma (black arrow), buffy coat (grey arrow) and red blood cell (white arrow) fractions after first centrifugation. (c) Platelet-rich plasma consisting of upper portion of buffy coat and lower fraction of plasma after second centrifugation.

Figure 3.2: Platelet concentration increased 3.5 fold in platelet-rich plasma over whole plasma while platelet concentration decreased in platelet-poor plasma. Treatments with different superscripts significantly differ (p < 0.05).

b

c

a b c

a

49

Figure 3.3: Representative pictures of ASC proliferation in DMEM, 10% FBS, and 10% PRP after 2, 4 and 6 days. Scale bar is 100 µm.

50

Figure 3.4: Average doubling time of ASCs trended towards decrease (0.1 > p > 0.05) in platelet-rich plasma and fetal bovine serum compared to control DMEM and platelet poor plasma though no results were significantly different. Treatments with different superscripts trended toward difference but were not significantly different (0.01 > p > 0.05).

b b

b

a

a, b

51

Figure 3.5: Migration across polyethylteraphthalate wells significantly increased using platelet-rich plasma and fetal bovine serum. Treatments with different superscripts significantly differ (p < 0.05).

b

a

b, c

c

52

Figure 3.6: Representative images of cell migration in DMEM control and 20% PRP after 0, 24, and 48 hours. Scale bar is 200 µm.

53

Figure 3.7 Average cell velocity significantly increased in fetal bovine serum and platelet-rich plasma compared to control DMEM. Effect of PRP dose was significant up to 20% PRP. ASCs were not viable at PRP concentrations above 30%. Treatments with different superscripts significantly differ (p < 0.05).

a

b b

c

c

54

3.4 Discussion

The goal of these experiments was to determine an optimal concentration of platelet-rich

plasma which increased adipose-derived stem cell proliferation and migration while maintaining

long-term cell viability. Based on the results, a mixture of 20% platelet-rich plasma and 80%

Dulbecco’s modified Eagle medium was determined to be the optimum mixture and was adopted

for use in the in vivo experiment (Chapter 4). Using PRP at this concentration resulted in ASCs

which traversed through narrow channels in greater number, trended towards proliferating at a

faster rate, and migrated at a faster velocity while maintaining viability in culture long-term.

Faster proliferation and migration ultimately results in greater effective numbers of ASCs at the

site of the defect and helps to relieve the requirement of isolating and collecting high numbers of

these adult stem cells during surgery.

Injection of PRP along with adult stem cells at the site of the wound may be preferable to

expansion in cell culture, which carries risks of contamination, oncogenic mutation, as well as

significant regulatory and clinical infrastructure burdens [7]. However, as cell-based therapies

continue to be developed, optimal growth mediums which are safe and effective will be

important. Currently, fetal bovine serum and custom serum-free formulations are widely

employed, with FBS carrying advantages of effectiveness and cost but disadvantages of potential

disease transference and immune reactions, such as spongiform encephalopathy or serum

sickness, respectively [8,9]. Platelet-rich plasma poses an alternative for in vitro ASC expansion

[10]. According to the results, effectiveness in promoting proliferation and migration were

roughly equivalent to FBS. In addition, PRP could be individualized for each patient by blood

draws and processing before a grafting procedure. While the chance for contamination remains,

risks of disease transference are generally eliminated, making PRP an attractive potential

substitute for FBS or serum-free for growth factor mediums.

Protocols for concentrating platelets vary by amount of blood collected, centrifugation

speed and time, as well as fractions considered to be platelet-rich plasma [11]. The two-step

approach used in this study, which has been used clinically, resulted in a 3.5-fold increase (p <

0.001) in platelet concentration compared to plasma. Ranges of platelet concentration

approximately 3 to 6-fold are thought to be therapeutic and generally recommended for clinical

use of platelet-rich plasma [12]. Concentrations above this level may result in unacceptably

55

severe symptoms of inflammation, such as pain and swelling, or potential paradoxical effects in

wound healing. Translating this platelet concentration into growth factor yield, an estimated 0.06

ng of PDGF is produced per one million platelets, or about 1,200 molecules of PDGF per platelet

[13,14]. Therefore, while PRP is a potent autologous therapy, dosing remains an important but

less-studied factor in therapeutic regimens, in order to optimize both growth factors as well as

inflammatory factors which are simultaneously released during platelet degranulation.

Another important result of this study is the low viability of ASCs at concentrations

above 30% PRP, which is important because PRP appeared to form a stable gel at concentrations

above 50% and remained in solution at lower concentrations. While initially thought to be

toxicity from residual sodium citrate anticoagulant or calcium chloride for platelet degranulation

in the culture media, the observed low viability at high concentrations may be more likely due to

osmotic imbalance because of the high protein concentrations as well as lower concentrations of

DMEM, which has essential elements for cell growth such as amino acids, salts, vitamins and

glucose. Studies which utilize PRP at high concentration along with cellular therapy may have

reduced performance due to cell losses. Standard concentrations of FBS used with DMEM are

generally 10% to 20% depending on application [15]. Culturing ASCs with PRP at higher

concentrations may require modification of standard DMEM to preserve ASC viability.

The effect of platelets on ASC proliferation is the result of two primary growth factors,

PDGF and TGF-β [16]. Average doubling time decreased (faster proliferation) for PRP and FBS

treatments while doubling time was longer in platelet-poor plasma and control DMEM. For this

study, doubling time was independent of concentration for PRP. This result may be due to limits

in the rate of mitosis set by steps such as DNA synthesis and cell division [17], which might not

be further stimulated by growth factor administration.

For the transwell migration assay, the ability of ASCs to traverse channels on the order of

the size of the nucleus towards a growth factor gradient is simulated. During a bone injury, tissue

becomes necrotic and marked for debridement by macrophages and osteoclasts resulting in the

rapid formation of channels for migration throughout the damaged bone. A baseline level of

ASCs were found to migrate with no growth factor gradient present (DMEM controls) and is

likely due following other gradients, such as glucose , oxygen and pH, found in the fresh media

of the lower chamber [18]. Addition of PRP, FBS, or platelet-poor plasma all appeared to

56

enhance the process of migration. Growth factor gradients stimulate migration by differential

membrane receptor stimulation on the forward migrating end with higher concentration

compared to the end with lower concentration and are important for increasing the number of

MSCs to the site of the defect.

In the stamp migration assay, ASCs did not migrate towards any gradient, but instead

generally moved towards newly opened space in the cell culture plate, at a greater velocity in

PRP (p = 0.0014) or FBS (p = 0.0013) compared to control DMEM. The stamp migration assay

represents a rough two-dimensional approximation of a bone defect in that live cell density

decreases as capillary networks are damaged. Cell-cell signaling results in proliferation and

migration at low local confluencies and is reduced as confluency increases, known as contact

inhibition, a property of normal, non-cancerous cells [19]. In addition to the removal of contact

inhibition, average cell velocity appeared to be concentration-dependent up to 20% PRP (p =

0.0003) before becoming independent and less viable at 30% PRP (p = 0.42). These results may

suggest a therapeutic limit for the dosage of PRP which would be effective for the promotion of

proliferation and migration of ASCs, as membrane receptors may become saturated and gene

expression maximally upregulated.

Because the growth factors released from platelets primarily affect cell proliferation and

migration, a possible improvement to better simulate the bone regeneration process could be

bone morphogenetic protein (BMP-2 or BMP-7) or differentiation media (dexamethasone,

ascorbic acid, β-glycerophosphate) supplementation. Regeneration is initiated by platelets,

primarily through the release of PDGF, FGF, and TGF-β, which act to induce mitosis and

migration of scarce (1 in 250,000 cells for an adult [20]) mesenchymal stem cells, greatly

increasing their number during an injury. While TGF-β plays an important role in collagen

production and organization, it is the BMPs, not found in PRP, which direct the maturation of

osteoblasts as well as the eventual remodeling of the tissue. Supplementing with these factors

may provide a better understanding of the entire bone regeneration process. In summary,

platelet-rich plasma at the appropriate concentration increases proliferation, migration, and

velocity of cell spreading and presents an alternative culture media supplement.

57

3.5 References

1. Kon, E., Filardo, G., Delcogliano, M., Presti, M. L., Russo, A., Bondi, A., & Marcacci, M.

Platelet-rich plasma: new clinical application: a pilot study for treatment of jumper’s knee. Injury

40, 598–603 (2009).

2. Lee, R. H., Kim, B., Choi, I., Kim, H., Choi, H., Suh, K., & Jung, J. S Characterization and

Expression Analysis of Mesenchymal Stem Cells from Human Bone Marrow and Adipose

Tissue. Cellular Physiology and Biochemistry 14, 311–324 (2004).

3. Steel, G. G. Cell loss as a factor in the growth rate of human tumours. European Journal of

Cancer (1965) 3, 381–387 (1967).

4. Potapova, I. A., Gaudette, G. R., Brink, P. R., Robinson, R. B., Rosen, M. R., Cohen, I. S., &

Doronin, S. V. Mesenchymal Stem Cells Support Migration, Extracellular Matrix Invasion,

Proliferation, and Survival of Endothelial Cells. Stem Cells 25, 1761–1768 (2007).

5. Zhao, X.-M., Xia, Y. & Whitesides, G. M. Soft lithographic methods for nano-fabrication.

Journal of Materials Chemistry 7, 1069–1074 (1997).

6. Deasy, B. M., Chirieleison, S. M., Witt, A. M., Peyton, M. J. & Bissell, T. A. Tracking Stem

Cell Function with Computers Via Live Cell Imaging: Identifying Donor Variability in Human

Stem Cells. Operative Techniques in Orthopaedics 20, 127–135 (2010).

7. Prockop, D. J., Brenner, M., Fibbe, W. E., Horwitz, E., Le Blanc, K., Phinney, D. G., &

Keating, A. Defining the risks of mesenchymal stromal cell therapy. Cytotherapy 12, 576–578

(2010).

8. Erickson, G. A., Bolin, S. R. & Landgraf, J. G. Viral contamination of fetal bovine serum used

for tissue culture: risks and concerns. Dev. Biol. Stand. 75, 173–175 (1991).

58

9. Mackensen, A., Dräger, R., Schlesier, M., Mertelsmann, R. & Lindemann, A. Presence of IgE

antibodies to bovine serum albumin in a patient developing anaphylaxis after vaccination with

human peptide-pulsed dendritic cells. Cancer Immunol Immunother 49, 152–156 (2000).

10. Schallmoser, K., Bartmann, C., Rohde, E., Reinisch, A., Kashofer, K., Stadelmeyer, E., &

Strunk, D. Human platelet lysate can replace fetal bovine serum for clinical-scale expansion of

functional mesenchymal stromal cells. Transfusion 47, 1436–1446 (2007).

11. Dohan Ehrenfest, D. M., Rasmusson, L. & Albrektsson, T. Classification of platelet

concentrates: from pure platelet-rich plasma (P-PRP) to leucocyte- and platelet-rich fibrin (L-

PRF). Trends in Biotechnology 27, 158–167 (2009).

12. Dolder, J. V. D., Mooren, R., Vloon, A. P. G., Stoelinga, P. J. W. & Jansen, J. A. Platelet-

Rich Plasma: Quantification of Growth Factor Levels and the Effect on Growth and

Differentiation of Rat Bone Marrow Cells. Tissue Engineering 12, 3067–3073 (2006).

13. Singh, J. P., Chaikin, M. A. & Stiles, C. D. Phylogenetic analysis of platelet-derived growth

factor by radio-receptor assay. J Cell Biol 95, 667–671 (1982).

14. Bowen-Pope, D. F., Vogel, A. & Ross, R. Production of platelet-derived growth factor-like

molecules and reduced expression of platelet-derived growth factor receptors accompany

transformation by a wide spectrum of agents. PNAS 81, 2396–2400 (1984).

15. Lennon, D. P., Haynesworth, S. E., Bruder, S. P., Jaiswal, N. & Caplan, A. I. Human and

animal mesenchymal progenitor cells from bone marrow: Identification of serum for optimal

selection and proliferation. In Vitro Cell.Dev.Biol.-Animal 32, 602–611 (1996).

16. Kakudo, N., Minakata, T., Mitsui, T., Kushida, S., Notodihardjo, F. Z., & Kusumoto, K.

Proliferation-Promoting Effect of Platelet-Rich Plasma on Human Adipose–Derived Stem Cells

and Human Dermal Fibroblasts. Plastic and Reconstructive Surgery 122, 1352–1360 (2008).

17. Polymenis, M. & Schmidt, E. V. Coordination of cell growth with cell division. Current

Opinion in Genetics & Development 9, 76–80 (1999).

59

18. van Noort, D., Ong, S. M., Zhang, C., Zhang, S., Arooz, T., & Yu, H. Stem cells in

microfluidics. Biotechnology Progress 25, 52–60 (2009).

19. Kim, S., Chin, K., Gray, J. W. & Bishop, J. M. A screen for genes that suppress loss of

contact inhibition: Identification of ING4 as a candidate tumor suppressor gene in human cancer.

PNAS 101, 16251–16256 (2004).

20. Lazarus, H. M., Haynesworth, S. E., Gerson, S. L., Rosenthal, N. S., & Caplan, A. I. Ex vivo

expansion and subsequent infusion of human bone marrow-derived stromal progenitor cells

(mesenchymal progenitor cells): implications for therapeutic use. Bone Marrow Transplant 16,

557–564 (1995).

60

CHAPTER 4

IN VIVO AUTOLOGOUS BONE TISSUE ENGINEERING

4.1 Overview

This chapter covers the process of obtaining adipose-derived stem cells from liposuction,

platelet-rich plasma and fibrin scaffolds from blood draws, creation of critical-size bone defects,

and analysis of bone healing after 8 weeks. The primary aim was to assess therapeutic effects of

autologous PRP and fibrin scaffolds combined with autologous ASCs for critical-size defects in

the pig mandible. Fibrin scaffolds of the appropriate composition (calcium hydrogen phosphate

composite, Chapter 2) and platelet-derived growth factors of the appropriate concentration (20%,

Chapter 3) were hypothesized to accelerate healing in porcine mandible bone compared to ASC-

only injections and untreated controls. Three treatments included the use of autologous ASCs

from liposuction with the addition of platelet-rich plasma, fibrin scaffolds, or as cell-only

controls. All three treatments using ASCs were determined to increase bone mineral density and

bone volume fraction compared to untreated controls. In general, addition of both platelet rich

plasma and fibrin scaffolds to autologous ASCs from liposuction improved bone healing of

critical-size defects. Based on these results, adipose-derived stem cell therapies for bone healing

are recommended to add together with either platelet-rich plasma or use a calcium phosphate-

fibrin composite to encapsulate the cells for further improvement in healing of critical-sized

defects.

4.2 Methods

4.2.1 Liposuction

The University of Illinois Institutional Animal Care and Use Committee approved all

following procedures. Twelve female Yorkshire pigs (Sus scrofa domesticus) aged 1-3 years and

weighing 170-275 kg were included in the study. For ASC extraction, liposuction was performed

using standard protocols (Protocol A.1) [1]. Following general anesthesia (5% Isoflurane,

Baxter, Deerfield, IL) and sufficient skin sterilization with Betadine (Butler Animal Health,

Dublin, OH), 70% ethanol, and zephiran chloride (Winthrop Laboratories, New York, NY),

61

pinpoint incisions were made at 4-6 locations on either side of the dorsal midline using a trocar.

Sterile 0.9% saline (Vedco, St. Joseph, MO) with epinephrine (1 mg/L, IMS Limited, El Monte,

CA) was injected into subcutaneous adipose deposition sites using a 60 mL syringe connected to

a Cobra cannula (3 mm diameter, 25 cm length, Shippert Medical, Centennial, CO).

Approximately 10-20 mL of saline was infused per injection and allowed to remain several

minutes before collection to promote vasoconstriction and reduce blood loss. A liposuction

cannula (Triport, 4 mm diameter, 15 cm length, Shippert Medical) connected to a suction pump

(75 kPa, Schuco Inc., Indianapolis, IN) was inserted to collect the lipoaspirate (300-400 mL

total). Liposuction time was 60-75 minutes.

4.2.2 Isolation of Adipose-Derived Stem Cells

Collected lipoaspirate was added to 250 mL centrifuge tubes (Corning Inc., Corning, NY)

and mixed on a shaker plate with an equal volume of type I collagenase (0.075%, Sigma C2674)

for 60 minutes at 37°C to digest extracellular matrix [2]. The mixture was centrifuged for 10

minutes at 1400 rpm (547 xg) to separate low-density adipocytes from the higher-density stromal

vascular cell fraction, which contains ASCs along with other additional cell types, such as

fibroblasts, endothelial cells and smooth muscle cells. The remaining cell pellet was removed

and placed in 50 mL conical tubes to be centrifuged for 5 minutes at 1400 rpm (547 xg). Two

mL of red blood cell lysis buffer (Sigma R7757, St. Louis, MO) was added to the pellet for 2

minutes, then diluted with PBS (Sigma D5773) and centrifuged (repeated 1-2 times for removal

of red blood cells). The final pellet was mixed with serum-free DMEM (Sigma D5648) and

filtered with a 100 µm cell strainer (Fisher Scientific, Fair Lawn, NJ). Cells collected were split

into two syringes, each having a volume of 5 mL. Processing times for isolation of ASCs was

120-150 minutes. Following 10% formalin (Fisher Scientific) fixation after 5 and 60 minutes, the

number of nucleated cells was estimated using DAPI (Sigma D9542) staining and

hemocytometer counting. A tracer amount (1 million) of GFP-ASCs was added to the cell

injections of 3 pigs to analyze cell engraftment.

62

4.2.3 Surgical Procedures

Following completion of liposuction, pigs were placed in a supine position with sufficient

padding. Lower jaws were shaved and sterilized before placing iodine (Ioban, 3M, St. Paul, MN)

and thyroid drapes (Kimberly Clark, Roswell, GA). Initial approach through skin was made by

scalpel and followed up by electrocautery pencil (ConMed, Utica, NY) through fat and muscle.

Periosteum was moved by periosteal elevator. Transcortical osteotomies, 25 mm in diameter,

were performed on the posterior region of the mandible (ramus) using a trephine (Irwin Tools,

Huntersville, NC) and power drill (Milwaukee Tools, Brookfield, WI) with adequate irrigation

[3,4]. One surgical defect was created on each side of the ramus of the mandible (two total per

pig). Removed bone cylinders were stored at -20ºC for later analysis.

Following removal of bone, the periosteum was sutured to form a pouch with 2-0 or 3-0

polyglactin resorbable suture (Vicryl, Ethicon, Somerville, NJ). Treatments (5 mL volume) were

injected into the defect and the periosteum was closed. Muscle, fat, and skin layers were then

sutured continuously and the wound was sealed with cyanoacrylate skin glue (Vetbond, 3M, St.

Paul, MN). Pigs were administered an analgesic (Banamine, Purdue Products, Somerville, CT)

and antibiotic (Excede, Pfizer, New York, NY) and allowed to recover. Pigs were maintained on

a soft diet for 1-3 days post-surgery and then resumed standard feed.

4.2.4 Blood Extraction and Processing

Blood draws were used to acquire the fibrin scaffolds and PRP. For PRP, blood was

collected from the ear vein using a 21 gauge butterfly needle (Jorgenson, Loveland, CO) and a

10 mL syringe filled with 1.5 mL of 3.3% sodium citrate. Blood was drawn to fill syringe up to

10 mL total (8.5 mL blood). Anticoagulated blood was transferred to 15 mL centrifuge tubes. An

initial centrifugation was performed at 1800 rpm (905 xg) for 15 minutes for removal of RBCs

and collection of plasma. A subsequent spin of collected plasma at 3000 rpm (2510 xg) for 10

minutes separated platelets [5]. A total volume of 2-3 mL of PRP was collected for injection (1-

1.5 mL per defect). Before injection, PRP was combined with cell/DMEM mixture (20% PRP

concentration) and 50 µL of 10% calcium chloride for platelet activation. Cell-PRP mixture was

then injected and the periosteum closed.

63

For fibrin scaffolds using whole blood, pigs were placed in a laterally recumbent position

for each defect. Blood was collected from the ear vein in a 10 mL syringe already filled with 1

mL of cell/DMEM mixture. Blood was drawn to fill syringe up to 4-5 mL total (3-4 mL blood),

and then quickly injected into the defect while still in the liquid phase. Calcium hydrogen

phosphate (0.1 mL of 0.3 M) was added and stirred in the defect. Following coagulation,

periosteum was then sutured.

4.2.5 Computed Tomography

Eight weeks post-surgery, mandibles were collected for analysis of bone healing. Four

freshly harvested whole mandibles were scanned using computed tomography (GE Medical

Systems, Waukesha, WI) at the University of Illinois Veterinary Medicine Teaching Hospital.

Scanning parameters were 120 kV and 64 mA. Three-dimensional images were analyzed for

gross anatomical information (Carestream, Rochester, NY).

4.2.6 Dual Energy X-ray Absorptiometry

Following removal of soft tissue, samples were scanned using DXA (Hologic QDR

4500A, Bedford, MA) to measure bone mineral density (BMD). Two X-ray beams with different

energy levels are passed through the bone and intensities are measured for each beam. Intensity

varies depending on the thickness of the bone and amount of calcification. Based on the

difference between the two beams, BMD was estimated. The original bone removed during the

surgery served as a control for comparison and was scanned alongside the defect. Relative

change in BMD was calculated as BMD of the defect divided by BMD of the original bone

removed during surgery.

4.2.7 Micro-Computed Tomography

Samples were trimmed with a band saw and were scanned using microCT (Skyscan 1172,

Kontich, Belgium) to determine the degree and location of bone defect healing [6]. MicroCT

utilizes the variable X-ray attenuations of different materials to image samples. X-rays pass

through the sample form a 2-D projection image. The sample was then slowly rotated, allowing

the capture and output of a series of images. The scanner was operated at 75 kV and 100 mA

64

with a 1mm aluminum filter to reduce low energy noise. The medium resolution setting at 95%

camera gain was utilized so that projection images were taken at 0.4 degree increments over a

range of 180 degrees. A frame average of five projection images eliminated artifacts due to noise

from the detector. To help eliminate ring artifacts caused by the detector elements, a random

vertical movement of 5 was included. The projection radiographs were 16 bit TIFF files with

1048 x 2000 pixels.

A reconstruction algorithm (NRecon 1.1.4, Micro Photonics, Allentown, PA) digitally

stacked and aligned these 2-D slices, resulting in a 3-D model of the sample which was used to

analyze the two and three-dimensional morphological parameters of the defect. Quantitative

analysis was based on the different attenuations of bone and soft tissue due to their different

densities and compositions. Each volume element, or voxel (25 µm resolution), was assigned a

gray scale value which was threshholded (Amira 5.0, Visualization Sciences, Burlington, MA).

Bone volume fraction was calculated as bone pixels divided by total defect pixels.

4.2.8 Histology

Freshly isolated samples were exposed under a UV fluorescent flashlight and

photographed to analyze cell engraftment (Figure 4.7). Following imaging, samples were fixed

in 10% neutral buffered formalin (Fisher Scientific, Fair Lawn, NJ) for two weeks, dehydrated in

an ethanol series (70, 80, 90, 100%) and infiltrated with methylmethacrylate (MMA, Acros

Organics, Geel, Belgium) liquid monomer. Poly-methylmethacrylate (PMMA, Polysciences,

Warrington, PA) solid beads and benzoyl peroxide (Sigma 33581) were added to induce

polymerization and embed the samples. Samples were then sectioned perpendicular to the long

axis of the cylindrical defect using a Buehler Isomet 100 diamond saw (Lake Bluff, IL) to yield

500 µm thick sections. Sections were polished to 300 μm thickness with increasing grades of

zirconium sandpaper. Three sections, one in the center and two at opposing edges, were used in

the analysis.

Sections were stained with Sanderson’s Rapid Bone Stain (Dorn & Hart Microedge, Villa

Park, IL) and counterstained with acid fuchsin (Sigma F8129) to differentiate calcification (red)

from soft callus (blue) and scar tissue (unstained). A digital scanner (HP Deskjet F4400, Miami,

FL) captured images at full scale for quantification using image analysis software (ImageJ,

65

National Institutes of Health, Bethesda, MA). Mineralized bone area was calculated as area of

tissue staining positively for bone (purple) per total defect area.



variance (ANOVA) was performed using SAS 9.2 statistical analysis software (SAS Institute,

Cary, NC). Fisher’s exact test was used for pairwise mean comparisons. A total of 6 defects per

treatment (n = 6) were analyzed for a total of 24 defects. Statistical significance was evaluated

using an alpha value of 0.05.

4.3 Results

4.3.1 Nucleated Cells Administered

While the number of nucleated cells harvested varied by pig due to variables such as

amount and consistency of fat, the number of nucleated cells administered did not significantly

differ according to treatment (p > 0.34, Figure 4.1).

4.3.2 Computed Tomography Scans

Computed tomography scans (Figure 4.2) detail anatomical defect location in the porcine

mandible. Defects were located in the ramus at varying levels of healing.

4.3.3 Bone Mineral Density

Bone mineral density (Figure 4.3) of the defect and relative change in BMD compared to

the original drilled bone (Figure 4.4) were significantly higher for ASCs (p = 0.032), platelet-rich

plasma (p = 0.028), and fibrin (p = 0.036). Relative change in BMD was significantly higher for

fibrin (p = 0.041) compared to ASCs-only.

4.3.4 Bone Volume Fraction

Bone volume fraction (Figure 4.5) in the defect was significantly higher for ASC (p =

0.0039), PRP (p = 0.0011), and fibrin (p = 0.0035) treatments compared to controls (Figure 4.6).

66

Bone volume fraction of both PRP (p = 0.044) and fibrin (p = 0.016) was significantly higher

compared to ASC-only injections.

4.3.5 Histology

Representative fluorescent images of bone defects injected with GFP tracer ASCs (1

million cells per defect) displayed areas of green color representing potential cell engraftment

within healing bone tissue (Figure 4.7). However, individual cells could not be located using

confocal microscopy.

Portions of mineralized bone in unstained sections appeared white while soft tissue and

scar tissue appeared yellow in color. Organized collagen in bone stains blue with Sanderson’s

Rapid Bone Stain while scar tissue is unstained (Figure 4.8). Mineralized tissue stains red with

acid fuchsin counterstaining. Mineralized bone which stains blue with Rapid Bone Stain and red

with acid fuchsin appears purple in color (Figure 4.9).

Mineralized tissue area in the defect was significantly higher for ASC (p = 0.0037), PRP

(p = 0.0015), and fibrin (p = 0.010) treatments compared to controls (Figure 4.10). Mineralized

tissue area of fibrin was significantly higher compared to ASC-only injections (p = 0.048).

67

Figure 4.1: The mean number of nucleated cells administered per pig did not vary significantly according to treatment. Treatments with different superscripts significantly differ (p < 0.05).

a

a a

68

Figure 4.2: Representative computed tomography (CT) scans of whole mandibles with critical size defects. White pixels indicate high bone mineral density while black pixels indicate low density.

2382 2567

2420 2649

69

Figure 4.3: Representative dual energy X-ray absorptiometry (DXA) scans of critical size defects of various treatments. White pixels indicate high bone mineral density, grey is intermediate, while black pixels indicate low density. Mature cortical bone is indicated by black arrows. Immature woven bone from regeneration in the defect is indicated by blue arrows. Soft tissue is indicated by green arrows.

70

Figure 4.4: Bone mineral density of the defect (top) and relative change in BMD compared original drilled bone (bottom) was significantly higher for all three treatments using ASCs. Relative change in BMD was significantly higher for fibrin compared to ASC-only. Treatments with different superscripts significantly differ (p < 0.05).

a

a

b

b

b

b b, c

c

71

Figure 4.5: Representative micro-computed tomography reconstructions of bone defects of various treatments. Yellow/orange pixels represent mineralized tissue while grey pixels indicate soft tissue. Mature cortical bone is indicated by black arrows. Immature woven bone from regeneration in the defect is indicated by blue arrows. Soft tissue is indicated by green arrows.

Figure 4.6: Bone volume fraction in the defect was significantly higher for all three cell treatments compared to controls. Bone volume fraction of both PRP and fibrin was significantly higher compared to ASC-only injections. Treatments with different superscripts significantly differ (p < 0.05).

a

c, d

b

d

72

Figure 4.7: Representative fluorescent images of bone defects injected with GFP tracer ASCs. Areas of green color (black arrows) represent potential cell engraftment within healing bone tissue. However, individual cells could not be located using confocal microscopy.

73

Figure 4.8: Representative histological sections with different stains. Portions of mineralized bone in unstained sections appear white while soft tissue and scar tissue is yellow in color. Organized collagen in bone stains blue with Sanderson’s Rapid Bone Stain while scar tissue is unstained. Mineralized tissue stains red with acid fuchsin counterstaining. Mineralized bone which stains blue with Rapid Bone Stain and red with acid fuchsin appears purple in color. Mature cortical bone is indicated by black arrows. Immature woven bone from regeneration in the defect is indicated by blue arrows. Soft tissue is indicated by green arrows.

74

Figure 4.9: Representative histological sections. Organized collagen in bone stains blue with Sanderson’s Rapid Bone Stain while scar tissue is unstained. Mineralized tissue stains red with acid fuchsin counterstaining. Mineralized bone which stains blue with Rapid Bone Stain and red with acid fuchsin appears purple in color. Mature cortical bone is indicated by black arrows. Immature woven bone from regeneration in the defect is indicated by blue arrows. Soft tissue is indicated by green arrows.

75

Figure 4.10: Mineralized tissue area in the defect was significantly higher for all three cell treatments compared to controls. Mineralized tissue area of fibrin was significantly higher compared to ASC-only injections. Treatments with different superscripts significantly differ (p < 0.05).

b

a

b, c

c

76

4.4 Discussion

The objective of this porcine animal model experiment was to compare autologous

therapeutic options using adipose-derived stem cells alone, ASCs supplemented with platelet-

rich plasma, and ASCs encapsulated in fibrin scaffolds derived from whole blood in terms of

new bone formation after 8 weeks. Based on the results, autologous ASCs encapsulated in fibrin

supplemented with calcium hydrogen phosphate were determined to be the treatment which

resulted in the most bone formation of those tested, while platelet-rich plasma treatment resulted

in only a slight reduction in bone volume fraction (p = 0.25). These supplements to ASCs

significantly increased bone volume fraction compared to ASCs alone (p < 0.044), indicating

potential for improvement in adult stem cell therapies for bone defects.

The positive results of this study and others [7] may in part be explained by the critical-

size defect model employed. Our mandibular ramus model (Figure 4.2) maintains several

advantages, including non-weight bearing defects, quantitative analysis, reproducibility, and low

requirements for additional animal care. However, this model also has conditions favorable

toward bone growth independent of treatment, including mechanical stabilization [8], defined

bony edges with complete debridement, intactness of periosteum [9], and cyclical biomechanical

loading due to mastication [10]. In addition, while the animals analyzed were sexually and

skeletally mature [11], the pigs were relatively young and in good health. Despite these model

advantages, vehicle controls treated with only DMEM resulted in bone volume fraction recovery

of less than 10% suggesting that the 25 mm defects being utilized are critical-size defects.

Administration of autologous ASCs aided in healing of this critical-size defect resulting

in a defect which would eventually heal and remodel spontaneously, resulting in more than 3-

fold improvement in bone formation compared to controls (Figures 4.4, 4.6). The mechanism of

healing is likely to be closely related to the limitations of regeneration and production of scar

tissue of a critical-size defect. Simultaneous factors such as insufficient signaling molecules,

neovascularization, or MSC recruitment may all play a role in the limited regenerative response

during a critical-size defect [12]. As a heterogeneous mixture of fibroblasts, endothelial cells,

smooth muscle cells, and other cell types, ASCs have potential to alleviate each of these

limitations through the release of growth factors, formation of new blood vessels, and further

recruitment of regenerative cells to the site of the defect. Though not definitive, the apparent

77

location of GFP from tracer ASCs injected at the defect site (Figure 4.7) would suggest that these

cells are capable of differentiation and engraftment into newly forming tissue.

However, one potential problem limiting therapeutic success of ASC administration may

be lack of cell encapsulation at the site of the defect, if only a liquid vehicle is used. Cell

encapsulation constricts cells and prevents their potential migration to sites other than the defect,

reducing dilution of cell concentrations as well as the theoretical risk of tumor formation at

ectopic sites [13]. Furthermore, bone has the potential to be better able to form in the center of

the defect in addition to the edges, allowing healing to proceed more rapidly in multiple

directions, eventually capable of forming bridges. Geometrically, bone formation at multiple

sites serves to functionally reduce defect spaces and increase the size by which healing becomes

critically limited. Therefore, due to differences in ability to encapsulate cells at the

concentrations tested, the comparison between PRP and fibrin is imperfect, as ASCs were not

viable at concentrations where PRP formed a stable gel (above 50%). Better cell encapsulation

using fibrin likely explains the moderate improvement in bone formation of fibrin compared to

PRP treatments.

Differences in radiographic methods used in the study were noted. Dual energy X-ray

absorptiometry (Figure 4.3) is two-dimensional method dependent in part on thickness of the

bone sample, which varied in the pig mandibles analyzed, resulting in variation in bone mineral

density values. Accounting for this thickness variation by normalization by the original,

surgically removed bone reduced this discrepancy and likely improved precision of the

measurement. In comparison, the three-dimensional method of micro-computed tomography

(Figure 4.5) takes thickness into account and appears to be a more precise method for the circular

ramus defect model. Based on measurements of bone mineral density and volume fraction,

average bone mineral mass produced in the defect was estimated to range from 0.235 g for

controls, 1.76 g for ASCs, 3.09 g for PRP, and 4.08 g for fibrin. Only 9.3 mg of calcium

phosphate was added to fibrin scaffolds, so the contribution of exogenous CaHPO4 towards the

bone mineral density measurement was likely to be less than 1%. While DXA and micro-CT

have different precisions using this model, both provide useful information regarding the degree

of bone healing in the cylindrical defect, and the results are in general agreement.

78

The approach of using fibrin scaffolds or platelet-rich plasma has several advantages,

including rapid clinical translation, a low risk-benefit ratio, and effective improvement in bone

formation. Nearly all materials in this prospective therapy already have Food and Drug

Administration approval for clinical use. Use of autologous materials decreases risk of immune

rejection, microbial contamination, and disease transmission, as well as practical, inexpensive

collection and processing The rate of complications is expected to be less than current autograft

methods, so this proposed treatment may eventually lead to reduced costs as well as improved

patient outcomes.

Both fibrin and PRP treatments have minor drawbacks. The drawing of non-clotted blood

requires rapid processing and is vulnerable to syringe blockage. Excess fibrin has also been

associated with the formation of scar tissue [14,15], some of which was observed during sample

collection. In the craniofacial region, this scar tissue may result in reduced patient satisfaction. In

the case of PRP, the pro-inflammatory component of the released platelet factors results in more

pain and swelling than a standard surgery for several days post-surgery. This additional swelling

may require thicker, stronger sutures to stay within breaking strength which may result in larger

scars at the incision site. Generally, the benefits of improved bone formation in these treatments

derived from autologous sources may be worth the risks and drawbacks posed.

Future work to improve these autologous therapies may include the substitution of

autologous bone dust or chips in place of calcium phosphate [16]. Besides being a more

autologous approach, bone chips harvested during the surgical removal of bone maintain lacunae

which house osteocytes that remain viable if collected and re-implanted within the timeframe of

a standard bone graft surgery. Furthermore, bone chips contain remnants of trabecular

architecture capable of infiltration by newly forming blood vessels, compared to calcium

phosphate granules which are generally internally solid and impenetrable. Improvements in PRP

processing to improve ASC viability at higher concentrations may improve bone formation, as

the biological adhesive properties of gelled PRP may improve bone grafting. Osmotic balancing

along with fine-tuning of anticoagulant:calcium ratios may result in cell viability suitable for

ASC administration. Further studies using more clinically-relevant bone defects, such as

calvarial, mandibular segmental, or femoral traction defects [17,18], among others, or using pigs

in states of compromised bone healing, such as diabetes [19] or older in age, are warranted.

79

Because the autologous materials studied in this dissertation cannot bear significant loads, a

much stiffer scaffold and/or other methods of stabilization would be required for body weight-

bearing applications. We conclude that addition of autologous blood products, either 20%

platelet-rich plasma or fibrin scaffolds derived from whole blood supplemented with calcium

phosphate, to adipose-derived stem cells derived from lipoaspirate, may be advantageous due to

improved bone formation without requiring harvesting of patient bone. In summary, the present

study demonstrates the potential for improvement in cell therapies towards an autologous bone

tissue construct for craniofacial bone repair.

4.5 References

1. Runyan, C. M., Jones, D. C., Bove, K. E., Maercks, R. A., Simpson, D. S., & Taylor, J. A . Porcine Allograft Mandible Revitalization Using Autologous Adipose-Derived Stem Cells, Bone Morphogenetic Protein-2, and Periosteum. Plastic and Reconstructive Surgery 125, 1372–1382 (2010).

2. Zuk, P. A., Zhu, M., Mizuno, H., Huang, J., Futrell, J. W., Katz, A. J., Benhaim, P., Lorenz, H., & Hedrick, M. H. Multilineage Cells from Human Adipose Tissue: Implications for Cell-Based Therapies. Tissue Engineering 7, 211–228 (2001).

3. Wilson, S. M., Goldwasser, M. S., Clark, S. G., Monaco, E., Bionaz, M., Hurley, W. L., Rodriguez-Zas, S., Feng, L., Dymon, Z., & Wheeler, M. B. Adipose-Derived Mesenchymal Stem Cells Enhance Healing of Mandibular Defects in the Ramus of Swine. Journal of Oral and Maxillofacial Surgery 70, e193–e203 (2012).

4. Maki, A. J., Clark, S. G., Woodard, J. R., Goldwasser, M. & Wheeler, M. B. A Critical-Size Craniofacial Bone Defect Model in the Yorkshire Pig. Reproduction, Fertility and Development 23, 159–159 (2010).

5. Kon, E., Filardo, G., Delcogliano, M., Presti, M. L., Russo, A., Bondi, A., Di Martino, A., Cenachhi, A., Fornasari, P., & Marcacci, M. Platelet-rich plasma: new clinical application: a pilot study for treatment of jumper’s knee. Injury 40, 598–603 (2009).

6. Lan Levengood, S. K., Polak, S. J., Poellmann, M. J., Hoelzle, D. J., Maki, A. J., Clark, S. G., & Wagoner Johnson, A. J. The effect of BMP-2 on micro- and macroscale osteointegration of biphasic calcium phosphate scaffolds with multiscale porosity. Acta Biomaterialia 6, 3283–3291 (2010).

80

7. Hicok, K. C., Du Laney, T. V., Zhou, Y. S., Halvorsen, Y. D. C., Hitt, D. C., Cooper, L. F., & Gimble, J. M Human Adipose-Derived Adult Stem Cells Produce Osteoid in Vivo. Tissue Engineering 10, 371–380 (2004).

8. Hiltunen, A., Vuorio, E. & Aro, H. T. A standardized experimental fracture in the mouse tibia. Journal of Orthopaedic Research 11, 305–312 (1993).

9. Ozaki, A., Tsunoda, M., Kinoshita, S. & Saura, R. Role of fracture hematoma and periosteum during fracture healing in rats: interaction of fracture hematoma and the periosteum in the initial step of the healing process. J Orthop Sci 5, 64–70 (2000).

10. Aro, H. T. & Chao, E. Y. S. Bone-healing patterns affected by loading, fracture fragment stability, fracture type, and fracture site compression. Clinical orthopaedics and related research 293, 8–17 (1993).

11. White, B. R., Lan, Y. H., McKeith, F. K., Novakofski, J., Wheeler, M. B., & McLaren, D. G. Growth and body composition of Meishan and Yorkshire barrows and gilts. J ANIM SCI 73, 738–749 (1995).

12. Deschaseaux, F., Sensébé, L. & Heymann, D. Mechanisms of bone repair and regeneration. Trends in Molecular Medicine 15, 417–429 (2009).

13. Yu, J. M., Jun, E. S., Bae, Y. C. & Jung, J. S. Mesenchymal Stem Cells Derived from Human Adipose Tissues Favor Tumor Cell Growth in vivo. Stem Cells and Development 17, 463–474 (2008).

14. Overgaard, K., Sereghy, T., Boysen, G., Pedersen, H., Høyer, S., & Diemer, N. H. A Rat Model of Reproducible Cerebral Infarction Using Thrombotic Blood Clot Emboli. Journal of Cerebral Blood Flow & Metabolism 12, 484–490 (1992).

15. Staindl, O. The healing of wounds and scar formation under the influence of a tissue adhesion system with fibrinogen, thrombin, and coagulation factor XIII. Arch Otorhinolaryngol 222, 241–245 (1979).

16. Tayapongsak, P., O’Brien, D. A., Monteiro, C. B. & Arceo-Diaz, L. Y. Autologous fibrin adhesive in mandibular reconstruction with particulate cancellous bone and marrow. J. Oral Maxillofac. Surg. 52, 161–165; discussion 166 (1994).

17. Bosch, C., Melsen, B. & Vargervik, K. Importance of the critical-size bone defect in testing bone-regenerating materials. The Journal of craniofacial surgery 9, 310 (1998).

18. Schmitz, J. P. & Hollinger, J. O. The critical size defect as an experimental model for craniomandibulofacial nonunions. Clin Orthop Relat Res 205, 299–308 (1986).

81

19. Brem, H. & Tomic-Canic, M. Cellular and molecular basis of wound healing in diabetes. Journal of Clinical Investigation 117, 1219–1222 (2007).

82

CHAPTER 5

SUMMARY AND CONCLUSIONS

This dissertation evaluated the effectiveness of autologous therapies for bone

reconstruction in the mandible using the Yorkshire pig as a preclinical animal model. The tissue

engineering approach of scaffolds, growth factors and stem cells were sourced from autologous

blood and fat from liposuction. Modifications of fibrin scaffolds were evaluated in terms of

coagulation speed, degradation, surface roughness, stiffness, and osteogenesis. Platelet-rich

plasma concentrations were assessed for proliferation, migration, and velocity of cell spreading.

In a 25-mm critical-size defect, treatments of fibrin scaffolds supplemented with calcium

hydrogen phosphate or 20% platelet-rich plasma combined with autologous adipose-derived

stem cells were compared with adipose-derived stem cells alone and untreated controls over an 8

week time point. Bone formation was assessed using dual-energy X-ray absorptiometry, micro-

computed tomography, and histology.

Results included improved stiffness, coagulation speed, and surface roughness in fibrin

scaffolds supplemented with calcium phosphate. A platelet-rich plasma concentration of 20%

promoted increased adipose-derived stem cell proliferation, increased migration, as well as

increased cell velocity. These results prompted the selection of calcium phosphate-supplemented

fibrin and 20% platelet-rich plasma for the subsequent in vivo study. These treatments resulted in

improved bone formation in a 25-mm porcine mandibular defect model compared to

administration of ASCs alone. All treatments with ASCs resulted in significantly more bone

formation than untreated controls. For untreated controls, ASCs only, or ASCs with platelet-rich

plasma, bone formation was observed to proceed from the outside towards the center similar to

the natural healing process. For ASCs with fibrin, bone formation was observed throughout the

defect, including the center, due to likely cell encapsulation and viable cells. Tracer GFP cells

were observed, indicating that these ASCs graft successfully in forming tissue and participate in

healing in part by differentiation into cell types which either support or form new bone tissue.

For the assessment of bone formation, micro-computed tomography appeared to be a more

precise measurement than dual energy X-ray tomography. In summary, addition of autologous

blood treatments further improved bone healing in a critical-size defect model (Figure 5.1).

83

Future studies should examine the use of these autologous bone tissue engineering

materials in more clinically-relevant defects, such as calvarial, mandibular segmental, or femoral

traction. The mechanism of adipose-derived stem cell attachment to fibrin scaffolds, including

molecules and physical characteristics required, would aid in improving scaffold design.

Addition of autologous bone chips acquired during the surgical procedure may improve results

over calcium phosphate granules. Efforts to increase cell viability at higher platelet-rich plasma

concentrations and its subsequent gelation may provide added benefits of cell encapsulation.

Stiffer scaffolds such as autologous bone, hydroxyapatite or poly-caprolactone, which are able to

bear body weight, will be required for successful treatment of the majority of bone defects. In

addition, combining fibrin scaffolds with platelet-rich plasma may result in synergistic

improvement above each component separately and would comprise a complete tissue

engineering construct. While more study is needed, this autologous approach has potential to

present a viable alternative to current bone grafting methods.

Analytical methods used in this study focused primarily on digital image analysis, which

has limitations of image resolution and subjectivity of the observer. Despite these limitations,

image analysis remains a highly flexible, adaptable method able to measure high volumes of

information for a wide range of applications, as demonstrated in this dissertation. Efforts in

automation would result in higher-throughput measurements with reduced subjectivity. Because

metrics are essential for evaluation of the suitability of bone tissue engineering scaffolds, image

analysis will remain one of key methods for quantitative, reproducible assessment. Some

measurements developed in this dissertation may have potential to be utilized for evaluation of

other prospective bone tissue engineering treatments or perhaps for clinical assessment of bone

grafting.

The approach detailed in this dissertation has several advantages, including rapid clinical

translation, a low risk-benefit ratio, and effective improvement in bone formation. Nearly all

materials in this prospective therapy already have Food and Drug Administration approval for

clinical use. Use of autologous materials decreases risk of immune rejection, microbial

contamination, and disease transmission, as well as practical, inexpensive collection and

processing. One clinical drawback to this procedure could be the multi-disciplinary cooperation

required due to simultaneous liposuction and bone grafting in one procedure, which may

84

contribute to increased cost in the short-term. However, the rate of complications is expected to

be less than current autograft methods, so this proposed autologous treatment may eventually

lead to reduced costs as well as improved patient outcomes. We conclude that addition of

autologous blood products, either 20% platelet-rich plasma or fibrin scaffolds derived from

whole blood supplemented with calcium phosphate, to adipose-derived stem cells derived from

lipoaspirate, may be advantageous due to improved bone formation without requiring harvesting

of patient bone. The studies in this dissertation represent a small, but significant step towards an

autologous bone tissue engineering construct for craniofacial bone repair.

Figure 5.1 Proposed model for beneficial effects of platelet-rich plasma and fibrin for adipose-derived stem cell therapy for critical-size bone defects.

1. Fibrin clot 2. Low cell infiltration 3. Scar tissue - Limited growth from edges

1. Increased proliferation 2. Stimulus keeps ASCs near wound 3. Reduced cell wash out - Further increased growth from edges, some bridging

1. Vascularization, differentiation, reduced inflammation 2. Some cells wash out - Increased growth from edges 1. Cell encapsulation 2. Surface for cell attachment in center 3. No cell wash out - Bone growth in center, more dense

85

APPENDIX A

PROTOCOLS

A.1 Liposuction

The following protocol assumes you are using a 15 cm cannula and are attempting to retrieve

300-450 mL of aspirate.

1. Shave approximately 0.75 meters in length and 0.6 meters in diameter from the back side of the pig.

2. Clean area from the center outward in a spiraling motion first using 70% Ethanol, then betadine solution. (repeat 3 times)

3. Apply a layer of Zephiran, (cover with gauze pads soaked in zephiran until ready to begin liposuction)

4. Choose 6 spots approximately 6 inches from each other within the clean area to penetrate the pig.

5. In the first spot using a trocar (chisel with sharp point) approximately 1cm in diameter (diameter should be slightly wider than the width of cannula) place chisel at 90 degree angle to “catch” skin. Then maintain pressure on the skin while turning the chisel as close to 180 degrees along the length side of the back away from a previously agitated area. Hammer the chisel in far enough to stretch the hole to be the bulk of the width of the trocar (chisel).

6. Remove trocar (chisel). 7. Using a 60mL syringe with a cannula attached fill the syringe with 20mL of saline with

epinephrine (1 mg per liter, 1:1,000,000). 8. Insert the cannula into the hole at an angle as close to 180 degrees along the pigs back

side as possible. 9. Use a back and forth motion to agitate the fat cells while slowly and gradually releasing

the saline with epinephrine (to reduce bleeding). 10. Remove the syringe with cannula. 11. Using suction with a large insert and collect all tissue agitated by the canula during the

previous step. 12. When no tissue (cells) is visibly being collected by suction, remove suction. 13. Repeat steps 7-12 for same location about 4-5 times or until limited fat is being collected.

Then start a new location. 14. Using your 5 other chosen points for penetration repeat steps 5-12.

86

A.2 Adipose Derived Stem Cell Isolation Using Lipoaspirate

The protocol assumes that the volume of the aspirate obtained from liposuction is 300ml.

1. Aspirate obtained from liposuction is typically ~300ml divide and transfer 100ml of aspirate into 250ml tubes (3 tubes if total volume was 300ml)

2. Add 300ml PBS to the original container (so that aspirate:DPBS is 1:1) 3. Add 225mg of collagenase to 300ml PBS (so that you have 75mg collagenase/100ml

PBS) and mix well 4. Divide and transfer 100ml of the PBS with collagenase mixture to the three 250ml tubes

(so that they have a total of 200ml now) 5. Place the three 250ml flasks on the orbital shaker that is inside the 370C incubator 30

mins 6. Centrifuge @ 1400rpm (547 xg, the centrifuge on the floor of the lab) for 10 mins 7. Remove tubes from the centrifuge there will be a floating fat pellet at the top and a

cell pellet at the bottom. Pass a 10ml pipette along the wall, through the floating fat, and aspirate out the cells at the bottom and transfer them to 50ml falcon tubes. Repeat until all cells are removed.

8. Centrifuge 50ml falcon tubes @1400 rpm for 5 mins 9. Remove supernatant using 10ml pipette remove as much as possible, but take care not

to remove the cells on the surface of the pellet 10. Add 2ml EL buffer (erythrocyte lysis buffer) to each tube mix gently by hand for ~

2mins 11. Add PBS to make the final volume ~45-50ml (still working with the same 50ml falcon

tubes) 12. Centrifuge @1400rpm for 5 mins 13. Remove supernatant and resuspend the pellet in one of the 50ml falcon tubes in 10ml

DMEM mix well and transfer it to the 2nd tube containing the cell pellet mix well and transfer everything to the 3rd tube containing the cell pellet

14. Filter DMEM/cell suspension using strainer 15. Centrifuge filtrate @1400rpm for 5 mins remove supernatant so that you have the

right volume for transplanting

autologous tissue engineering for bone repair

Documents