ON THE PERMANENT LIFE OF TISSUE OUTSIDE THE ORGANISM

The Pennsylvania State University

The Graduate School

Department of Materials Science and Engineering

ON THE PERMANENT LIFE OF TISSUE OUTSIDE THE ORGANISM

A Thesis in

Materials Science and Engineering

by

Dhurjati Ravi

© 2007 Dhurjati Ravi

Submitted in Partial Fulfillment of the Requirements

for the Degree of

Doctor of Philosophy

December 2007

The thesis of Dhurjati Ravi was reviewed and approved* by the following:

Erwin A. Vogler Professor of Materials Science and Engineering and Bioengineering Thesis Advisor Chair of Committee

Andrea M. Mastro Professor of Microbiology and Cell Biology Paul W. Brown Professor of Ceramics Ronald Hedden Assistant Professor of Materials Science and Engineering Sandeep K. Prabhu Assistant Professor of Veterinary and Biomedical Science

Gary L. Messing Distinguished Professor of Ceramic Science and Engineering Head of the Department of Materials Science and Engineering

*Signatures are on file in the Graduate School

iii

ABSTRACT

Growth and maintenance of cells/tissue outside the body is central to the practice of

biomedical sciences and technology. In vitro culture techniques, developed almost a century

ago, are fundamental enabling tools for the study of life processes and are routinely

employed in many applications that impact human health care. Emergence of the field of

tissue engineering, focused on developing tissue surrogates for transplantation, has renewed

the emphasis on critically examining the conditions under which we culture cells/tissue

outside the body. It has been well recognized that isolation of cells from the complexity of

their native physiological environment results in adaptive responses that often limit cell

viability and function in vitro. Bridging the gap between physiological complexity and in

vitro culture environments is critical to the successful implementation of the tissue

engineering strategy. We show herein that a compartmentalized bioreactor based on the

principle of continuous-growth–and-dialysis provides stable culture conditions that better

simulate the physiological environment. The bioreactor was used to grow mineralizing,

collagenous bone tissue up to 150 mμ thick from an inoculum of isolated murine (mouse

calvaria MC3T3-E1, ATCC CRL-2593) or human (hFOB 1.19 ATCC CRL-11372)

osteoblasts over uninterrupted culture periods up to a year. Proliferation and phenotypic

progression of an osteogenic-cell monolayer into a tissue comprised of cell layers of

mature osteoblasts in the bioreactor was compared to cell performance in conventional

tissue-culture polystyrene (TCPS) controls. Cells in the bioreactor basically matched results

obtained in TCPS over a 15 d culture interval, but loss of insoluble ECM (iECM) and ~2X

increase in apoptosis rates in TCPS after 30 d indicated progressive instability of cultures

maintained in TCPS with periodic refeeding but without subculture. By contrast, stable

cultures were maintained in the bioreactor for up to a year, suggesting that extended-term

6≥

iv

tissue maintenance is feasible with little-or-no special technique. Month’s long culture

interval in the bioreactor lead to progression of pre-osteoblasts to osteocyte-like cells

embedded in mineralized matrix observed in normal bone and production of visually-

apparent (macroscopic) bone. Challenging bioreactor-derived bone tissue at different stages

of development with metastatic breast cancer cells (MDA-MB-231) known to invade the

skeleton created a system-in-crisis that captured early stages breast cancer colonization of

bone. In situ confocal microscopy revealed sequential steps of breast cancer cell adhesion,

penetration, and degradation of the 3D bone-like osteoblast tissue over co-culture intervals

ranging from 3 to 10 days. Bioreactor enabled direct observation of interactions of breast

cancer cells with engineered 3D bone like tissue and simulated an important step in the

progression of the disease that may be a target for therapeutic intervention. Using bone as a

model tissue, this study demonstrates that complex physiological and pathological processes

of great importance to human health can be simulated using engineered cellular

environments for potential applications in drug discovery and toxicology.

v

TABLE OF CONTENTS

LIST OF TABLES ........................................................................................................vii

LIST OF FIGURES..................................................................................................... viii

Acknowledgements ........................................................................................................xii

Chapter 1 Introduction 1

In vivo and in vitro 1

The gap between in vivo and in vitro 3

Bridging the culture gap 4

References 7

Chapter 2 Literature Survey 13

Osteogenesis In Vivo 13

Bone Development 13

Cellular Origins Of Bone Cells 15

Bone Formation 15

The Osteoblast Milieu 18

Osteogenesis In Vitro 19

Organ Cultures 19

vi

Culture Of Bone Slices And Explants 20

Culture Of Isolated Bone Cells 22

In Vitro Models Of Bone Formation: An Assessment 23

Homogeneity Versus Heterogeneity 23

Length Of Culture Interval 25

Effect Of The Microenvironment 25

3D Models of Bone Formation 29

References 33

Chapter 3 Extended-Term Culture Of Bone Cells in a

Compartmentalized Bioreactor

61

Introduction 63

Methods And Materials 66

Results And Discussion 75

Conclusions 80

References 83

Chapter 4 Metastatic Colonization Of Bone Tissue By Breast

Cancer In Vitro

101

Introduction 102

Materials and Methods 105

Results and Discussion 108

Conclusions and Future Work 111

References 113

vii

LIST OF TABLES


Table 1 Comparison of in vivo and in vitro techniques 10


Table 1 Principal extracellular matrix proteins secreted by the

osteoblasts and their sequence of expression

56

Table 2 Local factors in the bone microenvironment that

condition the growth and differentiation of the

osteoblasts

57

Table 3 Historical time line of the development of in vitro

models of bone formation

58

Table 4 Representative bioreactors for bone tissue engineering

applications

59

Table 5 Unique design features of the compartmentalized bioreactor 60

Chapter 3 Extended-Term Culture of Bone Cells in a


61

Table 1 hFOB 1.19 cells in conventional culture compared to

compartmentalized bioreactor

88

Table 2 MC3T3-E1 Cells in conventional culture compared

to compartmentalized bioreactor

89

viii

LIST OF FIGURES


Figure 1 Cells and/or tissues are cultured in an artificial environment within

a defined control volume. Physical, biological and biochemical

factors that condition the cellular response within the control

volume are listed

12


Chapter 3 Extended-Term Culture of Bone Cells in a


61

Figure 1 Compartmentalized bioreactor design. Panel (a) is a cross-sectional

diagram through the device showing separation of the cell-growth

space (A) from the basal-medium reservoir (B) by a dialysis

membrane (C). Cells are grown on gas-permeable but liquid-

impermeable film (E). The device is ventilated through film (D),

which is the same material as (E) as described herein but can be

different. The whole device is brought together in a liquid-tight

fashion using screws shown in the laboratory implementation Panel

(b) and Panel (c) which is an exploded-view identifying separate

components. Liquid-access is through leur-taper ports (J, K) which

mate to standard pipettes. See Methods and Materials for theory of

operation.

93

Figure 2 Growth dynamics of hFOB 1.19 on tissue-culture-grade polystyrene

controls (TCPS, filled triangles) compared to the

compartmentalized bioreactor (open squares; note logarithmic

ordinate). Inset expands short-term attachment rates using a linear

ordinate. Note that the initial two-fold cell-attachment preference

for TCPS persists through the exponential-growth (3 7t 2< < hr)

and post-confluence expansion ( 72 720t< < hr) phases

94

ix

Figure 3 Comparison of hFOB 1.19 on TCPS (Panels A, B) and in the

compartmentalized bioreactor (Panels C, D) by cross-sectional

TEM. Notice that cell layers are thicker in the bioreactor than on

TCPS (compare Panels A, B to C, D). and formation of apoptotic

bodies (arrows) after 30 d culture in TCPS

95

Figure 4 Comparison of insoluble extracellular matrix (iECM) production by

hFOB 1.19 on TCPS (dark bar) and in the bioreactor (light bar).

Notice that iECM significantly decreased after 30 d on TCPS but

remained constant in the bioreactor. Bar values are mean of

duplicate samples measured in triplicate with standard deviation

represented by error bars

96

Figure 5 Representative confocal image of apoptotic bodies (green stain) in

TCPS and in the bioreactor selected from 8-10 similar in-depth

image planes probing hFOB 1.19-derived tissue. All cells were

stained red (Sytox Orange, scale bar = 50 μm). Percent apoptotic

cells quoted in each panel are the range of values observed within

the image planes

97

Figure 6 Ultramorphology of hFOB 1.19–derived tissue (15 d) recovered

from the bioreactor by cross-sectional TEM. Arrows point to cell

protrusions that occasionally connect two cells, as shown in Panel

C. Annotations: jNC = gap junction; MV = matrix vesicles; N =

nucleus; rER = rough endoplasmic reticulum

98

Figure 7 Analysis of mineral nodules taken from MC3T3-E1 cultured in a

bioreactor. Panel A is an SEM of a large nodule taken from a 70 d

bioreactor. Panel B is a high-magnification, cross-sectional TEM of

a similar nodule. Panel C is an x-ray spectrum obtained by

SEM/EDAX of a nodule taken from a 30 d bioreactor confirming

Ca and P as major constituents

99

Figure 8 Microscopic examination of matrix derived from MC3T3-E1

cultured in a bioreactor for 70 d. Panel A is an SEM of the outer

layer showing a highly fibrous network with bone nodules. Panel B

is an optical micrograph of a histologic workup showing layers of

100

x

collagen with imbedded osteoblasts. Panel C is a cross-sectional

TEM showing mineralized fibers running in (IP annotation) and

out (OP annotation) of the plane

Chapter 4 Metastatic Colonization of Bone Tissue by Breast

Cancer In Vitro

101

Figure 1 Schematic showing the sequential steps of cancer cell attachment,

penetration, replication and possible host tissue destruction

observed in the bioreactor as a result of interactions between GFP-

labeled breast cancer cells and multiple layer 3D osteoblast tissue.

118

Figure 2 (A) Light micrographs of a Hemotoxylin and Eosin stained 70-day

tissue (40X) reveal osteoblasts (pale red with dark-red nuclei) lining

a collagenous matrix (pale pink). (B)Transmission electron

micrographs (TEM) of 22-day tissue cross sections show up to 5 cell

layers (1500X). (C) Scanning electron micrographs of the surface of

a 70 day culture are studded with mineral nodules (1000X) that

prove positive for calcium and phosphorous by energy-dispersive

analysis of x-rays (not shown). (D) Bone chip recovered from a 5

month MC3T3-E1 bioreactor. X-ray diffraction (inset) indicates

similarity with a bovine bone reference spectrum (lines).

119

Figure 3 Transmission electron micrograph (TEM) of a cross-section of an

isolated MDA-MB-231 breast cancer cell crawling on the base film

of the bioreactor used in this work (cultured for 3 days in

osteoblast-conditioned medium) reveals an amoeboid morphology

with protrusive structures at the leading edge, consistent with a

migratory phenotype (Panel A, 1500X). TEM of breast cancer cells

(BC in Panel B, 5000X) co-cultured for 10-days with a 60-day

MC3T3-E1 tissue shows that cancer cells displaced osteoblasts

originally adhering to the base film (see Figure 2) and exhibit

protrusive structures that may be involved in penetration. (arrows;

compare to Panel A). Fluorescence microscopy oft breast cancer

cells (green) in co-culture with osteoid tissue (red) organize in

120

xi

single-file order (Panel C, 40X), captured in cross-section by TEM

(inset, 2000X).

Figure 4 Serial optical sections through an MC3T3-E1 derived osteoid tissue

(5-month culture interval stained with Cell Tracker Orange ™) co-

cultured with GFP-labeled MDA-MB-231 breast cancer cells

(green) for 3 days. Optical sections are from bottom [0μm] to top

[16μm]) of the tissue. (40X, scale bar = 50 μm)

121

Figure 5 3D reconstruction of serial optical sections taken through the

thickness of the 5 month osteoblast tissue cultured in the bioreactor

and subject to MDA-MB-231 breast cancer challenge over 3 day co-

culture interval.

122

xii

ACKNOWLEDGEMENTS

I would like to acknowledge my gratitude to Dr. Vogler for sharing his tremendous

enthusiasm and passion for science. It has been a pleasure to learn from a gifted teacher,

generous with his time and energies for all his students. I appreciate the continuous,

unwavering support through the years of graduate school and will always treasure the

conversations I have had with him about science and the philosophy of science.

I would also like to extend my gratitude to my thesis committee members- Dr. Mastro,

Dr. Brown, Dr. Prabhu and Dr. Hedden for helpful suggestions, critiques and constant

encouragement. I have benefited from my interactions with every one of my committee

members representing a diverse group of faculty from engineering and life sciences. I am

particularly indebted to Dr. Mastro for giving me the opportunity to work in the area of

cancer biology and for generously sharing her expertise and her valuable insight from the

many years of research on breast cancer.

I have had the opportunity to work with a wonderful group of people as part of this project.

My colleagues deserve special thanks for sharing the many joys and the minor tribulations of

the life in research.

Chapter 1

Introduction

1.1 In Vivo and In Vitro

Fundamental biological processes can be studied within the living

organism (in vivo) or in a controlled artificial environment outside the body (in

vitro). Investigations of biological phenomena in vivo using animal models has

the distinct advantage that the responses to particular dose are integrated across

the length scales of cells, tissues and organs, and are, therefore, representative of

physiological complexity. However, from an experimental design standpoint, it

is not very easy to separate cause and effect because the response to a dose

cannot be limited to a particular physiological compartment. In vivo studies offer

very limited control over the experimental conditions and it is difficult to tease

out the relationships between the numerous variables involved. Therefore, the

very complexity that confers physiological relevance to in vivo studies often

makes them difficult to interpret (Table 1).

Isolation of cells, tissues or even organs into a controlled artificial

environment has several advantages. In vitro techniques provide greater control

over the experimental conditions and the ability to limit the number of variables.

In other words they allow us to isolate a control volume defined by the biological

1

structure under investigation and its environment. Techniques that make it

possible to isolate, grow and sustain cells and tissues in controlled environments

outside the body were developed in the early part of the 20th century by pioneers

such as Harrison and Carrell.1-3 The incredible power of this approach to

provide fundamental insights into biological phenomenon was illustrated by the

very first experiment utilizing the technique. In 1907, Harrison2 isolated

fragments of the nerve tube of larval frog into a medium of clotted frog lymph in

a hollowed out glass slide and observed the outgrowth of nerve fibers from the

explants. Harrison’s seminal experiment2,4 provided, for the first time, the means

to observe living processes outside the body under controlled conditions. The

experimental setup allowed Harrison to separate the growth of the embryonic

nerve tissue from the overall complexity of embryonic development. Harrison’s

method was perfectly suited for testing his hypothesis that the nerve fibers were

outgrowths from the neurons and it helped him furnish experimental evidence to

resolve definitively one the major scientific controversies of his time on the

development of the nervous system.2,4

Early in vitro culture techniques including those used by Harrison and

Carrell were primarily culture of explants or fragments of tissue. The shift from

culture of tissues to the culture of isolated cells became possible only with the

development of enzymatic methods of cell separation in the 1950s.5 Rapid

2

adoption of these methods has resulted in culture of isolated cells being the norm

today. These developments have been consolidated into the standard practice of

monolayer cell culture where cells are cultured on a 2D substrate surrounded by

a medium that sustains the metabolic requirements of the cell.5 In almost a

century since their development, in vitro culture techniques have become integral

to the practice of biomedical sciences and technology. These enabling techniques

remain to this day, fundamental tools for the study of life processes and are

routinely employed in many applications that impact the delivery of human

health care.5

1.2 The Gap between In Vivo and In Vitro

In spite of the widespread success in the use of in vitro techniques, it is

important to recognize that there are profound differences between in vivo and in

vitro cellular environments. In vivo, cells held together by cell-cell contacts and

the extracellular matrix in a 3D structure, interact with neighboring cells of the

same type and/or different type with their responses coordinated by a variety of

biochemical and mechanical factors.6 In other words, cells function within the

context of a highly specialized microenvironment that is specific to the cell type

and the anatomical location. Growth, differentiation and ultimate fate of cells is

determined by complex interactions between the cell and its

microenvironment.7-9

3

Culture of cells outside the body using standard 2D monolayer techniques

deprives cells of their physiological context and brings into sharp focus the

impact of culture conditions on cellular function. Physical, biological and

biochemical factors affect cellular function within this controlled environment

(Fig. 1-1). To illustrate the nature of these environmental effects, consider the

affect of changes in osmolarity of the medium on cell volume. Decreasing the

osmolarity of the medium surrounding a cell leads to a compensatory response

of cell swelling; increase in osmolarity leads to cell shrinkage.10 In addition to

these physical factors, biological factors like cell-cell11 and cell-matrix

interactions7,8 condition cellular responses. Absence of proper environmental

cues often results in a less differentiated phenotype in vitro.7,8,12 Thus, there is

strong experimental evidence that cells undergo several adaptive responses

when transferred from their physiological environment to a radically different

exogenous environment.13 In aggregate, these adaptive responses have been

characterized as “culture shock” and they are often thought to limit cell viability

and function in vitro.13

1.3 Bridging the culture gap

The basic principles underlying in vitro culture techniques have not

changed significantly over the many years of their development. It is only with

4

the emergence of fields like tissue engineering14 that the environment in which

we routinely grow cells is being critically appraised. The gap between

conventional cell culture and the physiological environment is driving efforts to

develop engineered 3D cellular environments that can better simulate

physiological complexity.6,15 These approaches have the potential to integrate

physiological responses across the length scales of cells and tissues under

controlled conditions enabling us to model complex biological processes in vitro.

Working on the overall theme of the influence of the microenvironment

on cellular responses, this study showed that improved simulation of

physiological complexity can be achieved through suitable engineering of the

cellular environment. Particular focus was on the stability of culture

environment- an issue that has not received much attention in the literature.

Stable culture environments were engineered by separating cell growth and cell

feeding functions and through retention of cell-secreted growth factors and

cytokines in the cellular microenvironment.16 Using bone forming cells as a

model system, it was shown that stable culture environments promote assembly

of isolated cells into complex biological structures that can develop/mature over

extended periods.16 These in vitro grown structures permit a high level of

control over important experimental variables yet provide sufficient biological

complexity (e.g. three-dimensional, tissue-like cell density) that outcomes are a

5

reasonable predictor of the whole-organism physiological response to the

experimental variables. These tissue surrogates will be used as model systems to

study biological phenomenon of importance to human health care such as bone

formation and breast cancer colonization of bone.

6

References

1. Carrel A, Burrows MT. Cultivation of Adult Tissues and Organs Outside

the Body. J. Amer. Med. Assn. 1910;55:1379-1381.

2. Harrison RG. Experiments in transplanting limbs and their bearing upon

the problems of the development of nerves. Volume 4; 1907. p 239-281.

3. Oppenheimer JM. Taking Things Apart and Putting them Together Again.

Bull. Hist. Med. 1978;52(2):149-161.

4. Harrison RG. The outgrowth of the nerve fiber as a mode of protoplasmic

movement. Volume 9; 1910. p 787-846.

5. Freshney RI. Culture of Animal Cells: A Manual of Basic Technique. New

York: Wiley-Liss Inc; 2000.

6. Griffith LG, Swartz MA. Capturing complex 3D tissue physiology in vitro.

Nat Rev Mol Cell Biol 2006;7(3):211-224.

7. Roskelley CD, Desprez PY, Bissell MJ. Extracellular matrix-dependent

tissue-specific gene expression in mammary epithelial cells requires both

physical and biochemical signal transduction. Proc. Natl Acad. Sci. USA

1994;91:12378-12382.

8. Bissell MJ, Rizki A, Mian IS. Tissue architecture: the ultimate regulator of

breast epithelial function. Curr. Opin. Cell Biol. 2003;15:753-762.

7

9. Gurdon J, Lemaire P, Kato K. Community effects and related phenomena

in development. Cell 1993;75(5):831-4.

10. Sorkin AM, Dee KC, Knothe Tate ML. "Culture shock" from the bone cell's

perspective: emulating physiological conditions for mechanobiological

investigations. Volume 287; 2004. p C1527-1536.

11. Lauffenburger DA, Griffith LG. Who's Got Pull around Here? Cell

Organization in Development and Tissue Engineering. Proceedings of the

National Academy of Sciences of the United States of America

2001;98(8):4282-4284.

12. Abbot J, Holtzer H. The loss of phenotypic traits by differentiated cells. J.

Cell Biol. 1966;28:473-487.

13. Sherr C, DePinho R. Cellular senescence: mitotic clock or culture shock?

Cell 2000;102(4):407-10.

14. Langer R, Vacanti JP. Tissue Engineering. Science 1993;260(5110):920-926.

15. Griffith LG, Naughton G. Tissue engineering current challenges and

expanding opportunities. Science 2002;295:1009-1014.

16. Dhurjati R, Liu X, Gay CV, Mastro AM, Vogler EA. Extended-term culture

of bone cells in a compartmentalized bioreactor. Tissue Engineering

2006;12(11):3045-3054.

8

List of Table Legends

Table 1: Comparision of in vivo versus in vitro techniques.

9

Table 1: Comparison of in vivo and in vitro techniques

In Vivo In Vitro 1. Difficult to separate cause and effect because of the complexity of physiological environment and the number of independent variables involved.

1. Easier to separate cause and effect under controlled conditions and isolate the effect of individual variables.

2. Responses to dose are integrated across the length scales of cells, tissue and organs and are therefore representative of physiological reality.

2. Results obtained at the different length scales have to be integrated to create informed hypotheses about function at the organism level.

3. Tissue structure and organization and intercellular relationships are largely preserved.

3. Tissue structure and organization is often disrupted. Depending on the purpose of the investigation cellular physiology is sometimes studied in the absence of cell-cell and cell-ECM relationships found in vivo.

4. Does not easily permit reductionist approach of studying complex problems by breaking them down to their individual components.

4. Permits reductionist approach.

5. Difficult to study molecular level interactions including signaling and regulatory mechanisms.

5. Easier to study molecular level interactions.

6. Expensive, time-consuming and requires the maintenance of animals under the required test conditions for extended periods.

6. Simpler and cheaper to execute and permit greater control over experimental conditions

7. Studies using animals have to consider the inherent heterogeneity of subjects and require careful statistical analysis.

7. Permits greater control over test conditions and therefore repeatability.

10

List of Figure Legends

Figure 1-1: Cells and/or tissues are cultured in an artificial environment within a

defined control volume. Physical, biological and biochemical factors that

condition the cellular response within the control volume are listed.

11

Cont

rol V

olum

e

Tem

pera

ture

(typ

ical

ly 3

7oC)

Gas

com

posi

tion

(5 %

CO

2)M

ediu

m c

ompo

sitio

n, p

H, a

nd O

smol

arity

Oxy

gen

tens

ion

Mec

hani

cal s

tres

s

Pres

ence

of c

ytok

ines

, gro

wth

fact

ors,

ser

um-d

eriv

ed

prot

eins

, or

othe

r m

acro

mol

ecul

es in

the

med

ium

eith

er

by d

esig

n or

thro

ugh

secr

etio

n by

the

cells

as

part

of t

heir

norm

al fu

nctio

n.

Cell-

Cell

, Cel

l-EC

M a

nd C

ell-

Subs

trat

e In

tera

ctio

ns.

Phys

ical

Fac

tors

Biol

ogic

al F

acto

rs

Cellu

lar

Envi

ronm

ent a

nd C

ellu

lar

Func

tion

Fig

ure

1

12

Chapter 2

Literature Survey

This chapter starts with a short tutorial on bone growth and development and

introduces critical aspects of the physiology of bone-forming cells. Historical review of in

vitro models used to study these cells revealed specific design considerations for meeting the

need for improved bone cell culture models.

2.1 Osteogenesis In Vivo

2.1.1 Bone Development

Bone is a highly specialized organ that performs multiple functions within the body.1

In addition to providing structural support, bone acts as an storehouse of essential minerals

and is the site for hematopoiesis.2 Bone consists of a network of interconnected cells,

mineralized extracellular matrix and spaces that include bone marrow cavity, vascular

supply, canaliculi and lacunae.2,3 Growth, development and maintenance of the structural

integrity of bone require the coordinated actions of a complex cellular network.2,4,5

The primary modes of bone formation are endochondral and intramembranous.1,2

The process of endochondral bone formation that gives rise to the long bones and the

vertebrae requires the presence of a cartilaginous template that is replaced by or remodeled

into bone.2 Structurally, endochondral bone is further organized as either a dense outer

13

shell called the cortical bone or as a relatively thin inner network of connecting rods and

plates called trabecular bone.2,3,5 Intramembranous bones like the flat bones of the skull,

scapula and the ilieum form directly through differentiation of the mesenchymal cells into

bone forming osteoblasts without the need for an intermediate cartilage model.6

The principal functional activities involved in bone development and maintenance

are the formation of bone by osteoblasts7 and resorption of bone by osteoclasts.8 Removal of

bone from one site and resorption of bone at a different site occurs through a process called

“modeling” and is responsible for the shape and structure of bone during development.9 In

adults, resorption and new bone formation occur at the same site resulting in regeneration of

bone. This process called “remodeling” continues throughout life and becomes the

dominant process by the time bone reaches its peak mass in the 20s.3 Remodeling results in

complete regeneration of the adult skeleton every 10 years.3

Bone remodeling occurs through a tightly regulated sequence of osteoclastic bone

resorption followed by osteoblastic bone formation.10 The specific sequence of events

involved are osteoclast resorption and apoptosis followed by migration of osteoblast

precursors to the resorption site, proliferation and differentiation of osteoblasts, formation of

the extracellular matrix and its mineralization and finally cessation of osteoblast activity.11

Remodeling occurs through focal and discrete packets through the agency of a temporary

cellular unit called the “basic multicellular unit” (BMU) comprising of osteoblasts and

osteoclasts, central vascular capillary, nerve supply and related connective tissue.11 It is

estimated that in healthy human adults, 1 million BMUs operating at any given time and

over a course of a year 3-4 million BMUs are initiated to support the highly dynamic

function of bone tissue.11 The resorption stage of the bone remodeling sequence takes place

14

over the time frame of 10-14 days whereas formation of new bone at the site of resorption

takes up to 6 months.11 Delicate balance between resorption and bone formation that is the

hallmark of the remodeling process becomes disrupted in several pathological conditions12

including osteoporosis and breast cancer metastasis13 to bone resulting in loss of the

structural integrity of bone.

2.1.2 Cellular Origins of Bone Cells.

Osteoblasts and osteoclasts are derived from precursor cells within the bone

marrow.14 The bone marrow consists of cellular elements belonging to either the stromal

lineage or the hematopoietic lineage. The hematopoietic group is responsible for giving rise

to blood cells such as lymphocytes, erythrocytes, granulocytes, megakaryocytes and

monocytes.15 Osteoclasts are derived from hematopoietic cells of the

monocyte/macrophage lineage.15 Osteoblasts belong to the stromal group and are derived

from multipotent mesenchymal stem cells, which also give rise to bone marrow stromal cells,

chondrocytes, muscle cells and adipocytes.16 The complex sequence of steps involved in the

transformation of these early progenitor cells into functional osteoblasts and the various

factors that regulate the lineage development are only beginning to be understood.16 The

focus of this study was the functional expression of the osteoblast phenotype through the

sequence of proliferation, extracellular matrix formation, mineralization and finally

osteocytic transformation.17

2.1.3 Bone Formation

Bone consists of a mineral phase (70-90%) dispersed within an organic matrix

(osteoid) consisting primarily of Type I collagen. The rest of the organic matrix is composed

15

of non-collagenous proteins such as osteocalcin, osteonectin, bone sialoprotein, fibronectin,

vitronectin , thrombospondin as well as several proteoglycans, glycosoaminoglycans and

lipids (Table 1).18,19 The organic matrix of bone also sequesters several cytokines and growth

factors derived from bone cells and other cells that play important role in modulating growth

and differentiation of bone cells.20 These include growth factors such as Insulin-like growth

factor (IGF-I), Epidermal Growth Factor (EGF), Fibroblast Growth Factor (FGF),

Transforming Growth Factor (TGF-β), Bone Morphogenetic Proteins (BMP) and cytokines

such as IL-1 (Table 2).

Synthesis, deposition and mineralization of the organic matrix of bone is carried out

by osteoblasts.7 Bone formation involves proliferation of preosteoblasts and their

differentiation into functional osteoblasts capable of deposition, organization and progressive

mineralization of the organic matrix of bone.21 Stages of pre-osteoblast proliferation,

differentiation, extracellular matrix formation, mineralization and osteocytic transformation

advance through a tightly regulated spatial and temporal sequence.22

In the final stage of development osteoblasts can either undergo programmed cell

death (apoptosis) or transform into bone lining cells or become trapped within the bone

matrix as osteocytes.6 The proportion of osteoblasts that follow a particular fate is highly

variable and depends on several factors including age, species and anatomical location.6 In

the human cancellous bone it has been estimated that about 65% of the osteoblasts undergo

apoptosis23 with 30% transforming into osteocytes.24 In addition to pre-osteoblasts, active

osteoblasts, bone lining cells and osteocytes, a number of transitional stages have been

identified in the developmental program of the osteoblasts.25 These transitional stages

include preosteoblastic osteoblast, osteoblastic osteocyte (Type I pre-osteocyte), osteoid-

16

osteocyte (Type II pre-osteocyte), Type III pre-osteocyte, young osteocyte and old osteocyte.

Characteristic feature of this developmental program is that osteoblasts at different stages of

differentiation are part of a functional network of cells that communicate with each other

through gap junctions.26,27,28 As a result, bone tissue has been described as a functional

syncytium with cells ranging from osteoprogenitors to mature osteocytes connected in a vast

network,28 and coordinating responses to various mechanical and biochemical stimuli

through dynamic cell-cell contacts.

Various stages in the osteoblast differentiation are identified on the basis of

morphology, function, responsiveness to hormones such as parathyroid hormone and the

expression of a suite of molecular markers.29 Pre-osteoblasts are actively dividing cells that

express bone specific markers such as alkaline phosphatase and osteonectin.4 Osteoblasts are

characterized as non-dividing, cuboidal shaped cells, with large eccentric nucleus and one to

three nucleoli. Osteoblasts have extensive rough endoplasmic reticulum and golgi areas

consistent with their primarily secretory function.4 Osteoblasts can be distinguished from

pre-osteoblasts by the upregulation of number of bone markers including osteocalcin, bone

sialoprotein and type I collagen, alkaline phosphatase, vitamin D3 receptor and others.29 As

osteoblasts lay down the matrix, they become trapped within matrix and undergo

transformation to osteocytes.30,31 Nascent osteocytes radiate several cell processes32,33

towards the mineralizing matrix and have organelles very similar to those of osteoblasts. As

the osteocytes become embedded deeper in the matrix, changes in morphology and cell

organelle distribution occur.34 Mature osteocytes have a characteristic stellate shape, are

embedded within spaces called lacunae and extend long thin cellular processes that contact

other osteocytes and osteoblasts.32 Osteocytic transformation results in up to 70% decrease

in cell volume compared with original osteoblasts, and simplication of organelles with

17

decrease in golgi apparatus, decrease in the size of mitochondria and decrease in

endoplasmic reticulum.25,34 Osteocytes with an estimated half life of 25 years in humans

have a very long life span compared to the 3-4 month life span of the osteoblasts.33

Osteocytes die as a result of senescence, degeneration, and/or entrapment by osteoclasts.30

Several aspects of osteocyte function and lifecycle are yet to be defined because it is very

difficult to study osteocytes outside the body.30 Osteocytes are non-dividing, and they do not

maintain their differentiated state and function when isolated from the matrix that surrounds

them in their native state. As a result of this, several fundamental questions regarding

osteocytic transformation and the function of osteocytes are not yet fully understood.3,29,30 In

summary, the complex developmental program of osteoblasts associated functionally with

the process of bone formation ranges from pre-osteoblasts to osteocytes with a number of

transitional stages between these two stages and is modulated by a number of cellular and

molecular level interactions.33,35 In spite of significant progress in our understanding of the

developmental program of osteoblast and the various transitional stages there are still

considerable gaps in our knowledge. Filling these gaps requires experimental models that

can define the various stages not only in terms of morphology but also by the levels of

expression of specific molecular markers like alkaline phosphatase, osteocalcin, osteonectin,

parathyroid hormone related protein, as well as markers considered to be to specific to

osteocytes such as E11 and DMP1.

2.1.4 The Osteoblast Mileu

Growth, development, and function of osteoblasts is modulated by number of local

factors through both autocrine and paracrine mechanisms.20 The vast majority of these

factors are secreted by the osteoblasts themselves (Table 2).5 Mediators of osteoblast

response are also derived from other cells in the marrow environment and from the bone

18

matrix which has a high affinity for systemic and local factors.5 The complexity of the

osteoblast milieu is defined by cell-cell, cell-matrix interactions as well as the presence of a

number of regulators in the local environment of the osteoblasts that include growth factors

such as insulin-like growth factor (IGF-I), epidermal growth factor (EGF), fibroblast growth

factor (FGF), transforming growth factor (TGF-β), bone morphogenetic proteins (BMP)

and cytokines such as IL-1. For extensive discussion of these regulatory factors see

References 17 and 20.

2.2 Osteogenesis In Vitro

Tissue culture techniques have contributed tremendously to our understanding of the

mechanisms of bone formation. Bone has been studied in vitro as an organ, tissue or as

isolated cells. This section traces the historical development of these techniques outlining

the advantages and disadvantages associated with each of these approaches.

2. 2. 1 Organ Cultures

Strangeways and Fell pioneered the use of organ cultures36 to study the embryonic

development of skeletal tissue. Early organ cultures involved isolation of explants from the

chick embryo at different stages of development into a semi-solid medium of clotted plasma

and embryo extract. Cultures were incubated in test tubes36, watch glasses37 or in hollowed

out glass slides.38 Watch glass cultures were further enclosed in a petridish with a layer of

moist cotton at the bottom to provide a humidified culture environment.37 In the first

successful adoption of the technique, Strangeways and Fell36 demonstrated the growth and

development of cartilage from the leg buds of a 3-day chick embryo over a 14 day culture

interval. These early experiments revealed the potential of the technique to study the

developmental processes of bone under controlled conditions. Organ culture technique was

19

further refined and developed by Fell to study the development of avian embryonic long

bones. In 1929, Fell published the first paper37 dealing with calcification of bone in vitro

using organ cultures of isolated femora and tibiae from 5 1/2 to 6 day chick embryo. In the

1930s Fell used this technique to establish the importance of alkaline phosphatase in

mineralization,37,39,40 to demonstrate the osteogenic capacity of the periosteum and

endosteum in vitro38 and to investigate the in vitro development of the avian knee joint.41

These early studies and Fell’s contributions in particular are highlighted here because they

represent a significant progress from static histological studies to the dynamic study of the

development of bone. Fell’s pioneering studies form the basis for our understanding of

different stages of cartilage and bone cell differentiation during the process of embryonic

bone development. Organ cultures have been used since then to describe the embryonic and

fetal development of the chick long bone42 and to study bone development in mammalian

models like fetal and rat long bones.43 Organ cultures have also been found to be particularly

suitable to evaluate the action of vitamins and hormones on bone44 and to carry out

resorption studies.43,45

2.2.2 Culture of Bone Slices and Explants

Culture of slices of bone derived from the long bone, calvaria or metatarsal tissue

provide another in vitro alternative to study bone.46-48 The most successful study employing

slices of bone tissue49 in vitro was carried out by Rose in 1960 using tissue culture chambers

made of cellophane.50-52 Rose was among the first to recognize the importance of retaining

cell-secreted factors in the cellular environment for maintaining the differentiated state of the

cells. Culture chambers employed by Rose separated cell growth and cell feeding functions

through use of a dialysis membrane and represented a significant innovation in technique

20

(see chapter 3 for extended discussion). Rose reported outgrowth of osteoblasts and

fibroblasts from embryonic chick bone slices in culture.49 Although calcification was not

demonstrated, explants were maintained for extended periods up to weeks.49,52 Utility of

these early studies in understanding the cellular basis of bone formation was limited by the

heterogeneity of the cell populations involved. In the absence of modern genomic and

proteomic tools it was difficult to identify and separate the responses of bone cells from other

cell types involved. Therefore, the primary contribution of these early studies using slices of

bone tissue in vitro is limited to short-term studies on the metabolism of bone and in

determining the effect of hormones on bone tissue.46-48

Explant cultures of the periosteum and endosteum have also been used to study bone

formation.38 Culture of periosteum was initiated by Fell who showed that explants of

periosteum from 6-10 day chick embryos could form osteoblast-like cells that produced an

osteoid. Culture of folded periostea, developed by Nijweide, is a particularily useful form of

the explant technique. 53 Nijweide and Vanderplas showed that in cultures of folded

periostea from 16-18 day old embryonic chick calvaria, osteoprogenitor cells at the center of

the fold differentiate into matrix-secreting osteoblasts.53 Folded periostea preserve critical

cell-cell contacts and the overall 3D structural organization of the tissue and serve as a

functional model for the study of osteogenesis in vitro. Culture of folded periostea from

embryonic chick calvariae has been used to study hormonal regulation,54,55 metabolic

effects55, the regulation of calcium transport53,55, to demonstrate osteoid formation53,56 and

mineralization in the presence of β-glycerophosphate57, and to study the effect of

dexamethasone on bone formation in vitro.58

21

2.2.3 Culture of Isolated Bone Cells

The real breakthrough in our understanding of the physiology of bone forming cells

came through the development of enzymatic methods of isolation of bone cells by Peck and

his colleagues59-61 in the 1960s. Peck61 employed buffered collagenase to isolate rat calvarial

osteoblasts and cultured them using monolayer cell culture techniques. Isolated cells were

shown to be viable in culture for days although matrix deposition or bone formation was not

noted.61 In subsequent studies, it was shown that osteoblastic cells synthesize collagen,

respond to ascorbic acid with enhanced collagen synthesis59 and react to hormones such as

parathyroid hormone.62 Isolation of osteoblasts in culture under these controlled conditions

made it possible for the first time to undertake a wide range of biochemical and metabolic

studies. It was possible, for example, to study calcium transport,63,64 oxygen consumption

and lactic acid production65 over short-term incubation periods.

Methods of enzymatic isolation developed by Peck and his co-workers and the

refinements of these methods66 provided the basis for much of subsequent work on

understanding the function of bone forming cells. Sequential extraction of bone cells using

proteases such as trypsin and collagenase67-71 resulted in isolation of subpopulations of

osteogenic mouse calvarial cells. A further advance in developing homogenous cell

populations was the dissociation and cloning of fetal rat and mouse calvarial cells72-79 in the

early 1980’s. Cloning allowed isolation of cell populations based on specific characteristics

such as growth patterns, responsiveness to parathyroid hormone or alkaline phosphatase

activity. Formation of discrete mineralized bone nodules in vitro was reported in cultures of

fetal rat calvarial cells in the presence of ascorbic acid and β-glycerophosphate.80 Ability of

cultured osteoblasts to form mineral deposits was found to be dependent on several factors

22

including initial cell density, length of culture interval and the presence of ascorbic acid and

organic phosphates like β-glycerophosphate.72-74,80-89 Isolated osteoblasts in culture were

shown to synthesize several proteins involved in bone formation including osteocalcin, type I

collagen, osteonectin and osteopontin.22,90 Identification of these proteins and enzymes and

their temporal expression during the bone development process was actively investigated.22,91

Studies using cultures fetal rat and chick calvarial osteoblasts identified the developmental

sequence of osteoblasts consisting of proliferation, extracellular matrix formation and

mineralization and co-related this sequence with the gene expression of specific proteins

involved in bone formation.22

2.3 In Vitro Models of Bone Formation: An Assessment

In summary, bone cells/tissue in culture isolated from different

sites (long bones, calvaria, mandibles or the iliac crest) and from different species (rat,

mouse, avian, human) using a variety of isolation techniques (enzymatic dissociation,

outgrowths from explants) have contributed to our understanding of bone formation (Table

3). These isolation and culture techniques continue to be in use today.92 Careful analysis of

these disparate studies reveal important issues that are critical to understanding past efforts at

culturing bone cells and devising better culture models for the study of bone formation.

Some of these issues are discussed in the next few sections.

2.3.1 Homogeneity versus heterogeneity

A convincing argument can be made that the overwhelming trend in the

historical development of bone culture models is a reductionist shift from the heterogeneity

and complexity of tissue/organ/explant cultures to the simplicity and homogeneity of

cultures of isolated bone cells. Shift to the study of homogenous cell populations was driven

23

by a need to isolate the cellular level responses of specific population of bone cells from

various other influences (other cell types, biochemical factors) in the bone

microenvironment. However a purely reductionist approach brings into question the

physiological relevance of culturing isolated bone cells deprived of critical cell-cell and cell-

ECM interactions. Choice between organ cultures and cultures of isolated bone cells is

ultimately an experimental design issue and has to be resolved by the needs of the problem

being investigated. Traditionally organ cultures have been used for studying the systemic

effects of various hormones, drugs or other agents on bone and while cultures of isolated

bone cells have been used investigate cellular and molecular level interactions involved in the

formation of bone. Culture of tissue explants have been the bridge between the length scales

of cell and organ.

Another way of resolving the issue of length scale is to define a functional unit of structure

with the necessary complexity to simulate the physiological process of bone formation.

Osteoblasts have been known to lose their differentiated state when cultured as a monolayer

and reassume their differentiated state when cultured as multiple layers. There is also

extensive evidence in the literature that multiple-layered osteoblast tissue is required to

demonstrate mineralization in vitro.93-95 Tissue surrogates grown from isolated cells can

integrate the requirements of 3D structure,93,94 increased cell-cell contact and cell-ECM

interactions in the developmental processes of the bone cell and in the formation of bone in

vitro. Developing 3D tissue from isolated cells in a controlled manner has the potential to

realize functional biological structures that can bridge the gap between culture of isolated

bone cells and organ cultures. Further, tissue engineering methods differ fundamentally

from cultures of explant tissues because they allow us to build complexity in a controlled

manner using relatively homogenous cells as the starting materials. Aspects of cell

24

proliferation and stages of differentiation can be controlled by providing suitable

environmental cues through use of scaffolds, specially designed culture vessels called

bioreactors96 and by controlling the presence of various biochemical factors in the cellular

environment. Numerous studies in recent years have adopted this approach95,97-104 to develop

3D bone tissue from isolated bone cells seeded on to scaffolds and cultured in vitro in dishes,

spinner flasks, perfused catridges or rotating vessels.95,100,105-108

2.3.2 Length of Culture Interval

The length of culture interval is a critical variable in studies of bone

formation in vitro.109,110 Demonstration of osteogenesis in vitro requires continuous culture

intervals of weeks to months because the underlying physiological processes of bone cell

proliferation, extracellular matrix formation and mineralization takes place over extended

periods up to months. Quick survey of the historical development of in vitro bone cultures

shows that the most successful studies of osteogenesis involve extended culture intervals.

The classic example is the study by Binderman and his colleagues80 where they

demonstrated for the first time mineralization in cultures of isolated bone cells after 8 weeks

of continuous culture. In vitro cultures of bone cells must not only be viable over these

extended culture intervals109 but they must simulate the different stages in the development of

bone forming cells in a physiologically relevant manner. Recent studies have also shown

that the formation of bone in vitro is dependent on the length of culture interval.109

2.3.3 Effect of the Microenvironment.

Biochemical Factors: Growth and function of bone cells is regulated by the presence of a

number of local factors in their microenvironment.20,111-113 Osteoblasts secrete a number of

regulatory factors that in turn control the growth and differentiation of these cells through

25

autocrine and paracrine mechanisms.111,114 Importance of retaining these biochemical factors

in the cellular microenvironment was first articulated by Rose in the 1960s49,52 and has been

elegantly demonstrated by Tenenbaum and Heersche using cultures of folded chick

periosteaum.53,57 Success of periosteal osteogenesis model has to do largely with the fact

that folding the tissue not only increases cell-cell contact but also provides a

microenvironment in which the cell-secreted products are sequestered. To demonstrate this

effect, periostea were cultured without folding under two different configurations. In the first

configuration cell-secreted products were allowed to accumulate in the immediate

environment of the tissue. In the alternate configuration cell-secreted factors could diffuse

away from the tissue leading to reduced local concentrations. Under these experimental

conditions it was found that osteo-differentiation took place only in those situations where

the cell-secreted products were trapped within the local cellular environment.53,57

Refinements of the experimental setup employing Diaflo membranes with different

molecular weight cut-offs showed that allowing large molecules (100-300kD) to accumulate

in the local cellular environment was critical for inducing differentiation of the

osteoprogenitor cells. Recognition of these ideas has to the led to the discovery of several

factors in the bone microenvironment that play a critical role in the development of the bone

cell.113 Soluble factors that influence bone formation include bone morphogenetic proteins

(BMP), fibroblast growth factor (FGF), insulin-like growth factor (IGF), platelet-derived

growth factor (PDGF) and Interleukin-1 (IL-1). Complete description of the complex role

these factors play in the cellular physiology and development of bone is beyond the scope of

this work. However, it is critical to incorporate the effect of local regulators of bone

formation in the development of in vitro models of osteogenesis. Recent studies using 3D cell

culture models have confirmed the importance of the mass transfer of biochemical factors in

the development of bone tissue.115

26

Influence of Medium/Substrate: Early in vitro studies of bone tissue were carried out in the

semi-solid medium of plasma clot/embryo extracts and with air as the gas phase.

Refinement of explant culture techniques have resulted in culture of explants on a floating

raft/grid/gel at the interface between liquid medium and air to ensure sufficient gas

exchange.116,117 Development of techniques for the culture of isolated bone cells resulted in

widespread adoption of fluid medium consisting primarily of buffered salt solution with

addition of various nutrients and vitamins.117-119 The conventional practice is to culture

monolayer of cells in polystyrene dishes/flasks completely surrounded by a fluid medium118

containing nutrients and to incubate these dishes in 5% CO2 in air.

Higher CO2 tensions used in conventional culture have been shown to stimulate

bone cell proliferation and bone formation and calcification in organ cultures.120 In studies

examining the role of oxygen tension, lower O2 tension has been shown to stimulate bone

matrix formation and calcification in some culture systems.117,120-122 Studies of oxygen

consumption are diverse in terms of cells/tissues used and therefore it is difficult to

extrapolate the results of these studies to arrive at a specific oxygen requirement for bone

cells. Ensuring optimal delivery of this critical nutrient is critical to the development of

in vitro bone cell culture models. Oxygen tensions have to be tailored to the thickness of the

cell/tissue layer and the specific metabolic requirements of the cells being cultured.

Conventional polystyrene flasks are relatively impermeable to CO2 and O2 and therefore

oxygen supply is mostly through the dissolved oxygen supplied through the medium.

Oxygen diffusivity in aqueous solutions is rather low.123,124 Depending on the rate of oxygen

consumption, cells in conventional culture may be placed in anoxic conditions or hyperoxic

conditions. Therefore, the level of medium in the culture dishes (between 2-5mm) has to be

27

carefully controlled to provide sufficient oxygen tensions for the metabolic requirements of

the cells. Local oxygen concentrations regulate cell behavior in many different ways and

therefore it is important to consider not just the bulk oxygen concentration but also the

effects of oxygen gradients on cellular function.125 In summary, supply of oxygen in optimal

pericellular concentrations is a critical design consideration for developing better in vitro

models of bone formation.

Mechanical Stress: Growth and development of bone is conditioned by mechanical loads

imposed on the skeleton. Number of model systems have been developed to study the effect

of stretching and pressure forces on bone and organ cultures.126-132 More recently fluid shear

stresses133 have been shown to have an important role in regulating the development of bone

tissue in 3D cell culture models. Although, mechanical conditioning of in vitro bone

cells/tissue is an important area of research, it is not being addressed directly through this

study.

Review of the literature on bone cell physiology and historical development of the

in vitro methods used to study the function and development of bone cells, revealed

fundamental requirements for the design of culture models that can bridge the gap between

multiple length scales and demonstrate osteogenesis in vitro. These requirements include 3D

structural organization, optimal mass transport of nutrients and oxygen tension, stability of

the culture environment achieved through retention of cell-secreted biochemical factors and

the ability to grow/maintain tissue over extended time periods. Design and implementation

of a compartmentalized bioreactor that meets these requirements is described in the next

chapter.

28

2.4 3D models of Bone Formation

Identification of improved methods of hard tissue repair, augmentation or

replacements represents a major clinical and socioeconomic need.12,134-136 The field of bone

tissue engineering has emerged in recent years to meet this need through the development of

engineered bone tissue from a combination of cells, scaffolds and growth factors. Rapid

growth in the field has led to the development of several 3D models of bone formation

employing the basic strategy of culturing cell-scaffold constructs under controlled conditions

in devices called bioreactors.95,100,137 Bone tissue engineering approach has received extensive

attention in recent literature. Examples of primary studies,95,100,138,139 reviews,97,98,101,104 and

even books devoted to the subject 140,141 indicate the extent of interest in the field.

Central to the development of engineered bone tissue are the design and

implementation of bioreactors that provide a controlled physical environment and suitable

microenvironmental cues to direct the assembly of isolated bone cells into 3D tissue.142

Numerous studies in recent years have investigated the use of bioreactors to engineer bone

tissue to repair or replace bone lost due to disease or injury.142 Many different bioreactor

designs such as spinner flasks105,143, rotating wall vessels144, perfusion reactors106,145 and

rotating vessels139,146 have evolved for this purpose in recent times (Refer to Table 4 for a few

examples drawn from many refs. 147-150) , each with unique attributes and drawbacks.

Common design considerations underlying these different approaches to bioreactor

development are discussed below -

2.4.1 Cell-scaffold Interactions: The surface of the scaffold is designed to promote cell-

attachment through careful choice of scaffold material and biomaterial surface engineering.

The internal structure of the 3D scaffold is designed to promote efficient cell seeding, to

29

provide a suitable microenvironment for the cells, to allow for transfer of fluids and to

provide a template for the development of 3D tissue. Scaffolds made of biodegradable

polymers are designed to degrade and to be ultimately replaced by the developing tissue.100

2.4.2 Mass Transport of Nutrients and Oxygen: Critical limiting factors in the growth and

maintenance of 3D tissue are the delivery of soluble nutrients and oxygen to the cells while

simultaneously removing metabolic waste products. Design of bioreactors has to take into

account concentration gradients in nutrients and oxygen concentrations that may develop

through the thickness of the tissue and the tissue/scaffold constructs. Without adequate

nutrient supply, the scaffolds may either develop a necrotic core or diminished cell survival

and function at the center of the scaffold compared to cell/tissue layer at the surface.

Bioreactor designs take different approaches to address this fundamental issue. Spinner

flasks provide increased mass transport through continuous mixing of the medium.

However, the turbulence generated in spinner flasks can be detrimental to tissue

development. Mass transport through the tissue construct could be increased by continuous

rotation of the culture vessel itself or by continuous direct perfusion of medium through the

cell/scaffold constructs. Rotating wall vessels employ dynamic laminar flow under

conditions of low shear stress for efficient mass transport.

2.4.3 Mechanical conditioning: Bioreactors have also been designed with specific

emphasis on mechanical conditioning to the 3D bone tissue constructs.151 Mechanical

conditioning can be provided through direct application of stress or through application of

fluid shear stress.133,152-154

30

2.4.4 Mass transport of biochemical factors: A design feature that has received far less

attention in the development of bioreactors is the mass transport of biochemical factors.

Source of biochemical factors could be either endogenous (cells) or exogenous (added to the

bioreactor environment). It is well recognized that cell-secreted growth factors and cytokines

play important roles in modulating cellular function through both autocrine and paracrine

mechanisms. Retaining these factors within the cellular microenvironment could be critical

for the growth/differentiation and maturation of the 3D bone tissue. (For complete

discussion see section 2.3.3 in this chapter). Continuous or scheduled perfusion of medium

typically employed in bioreactors can result in removal of cell-secreted regulatory factors,

perturbation of the cellular environment and instability in cultures. There is experimental

evidence to suggest that increased flow rates through 3D bone constructs lead to diminished

cell survival and function that could be a result of the loss of cell-secreted regulatory

molecules.155

As a consequence of the above factors, we have focused our attention on the

design and implementation of a compartmentalized bioreactor based on the principle of

continuous-growth and dialysis. The compartmentalized bioreactor was a significant

departure from existing designs because of its emphasis on stability of the pericellular

environment (Table 5). Extraordinarily stable culture environments were achieved in the

bioreactor through the separation of cell-growth and cell- feeding functions by a dialysis

membrane. Carefully selected molecular weight cut-off resulted in continuous supply of

nutrients and removal of low molecular weight waste products while retaining cell-secreted

biochemical factors. Whole device was enclosed by gas permeable films that provide

sufficient oxygen tension to meet the metabolic requirements of the cells. Continuous

dialysis resulted in controlled delivery of nutrients and removal of waste products. As a

31

result of the compartmentalized design, it was not necessary to disturb the medium in the

cell chamber through the course of the month’s long experiment. At the same time, nutrients

from the medium reservoir gradually dialyzed into the cell chamber while low-molecular

weight nutrient were continuously removed from the cell chamber. Hence a decided

advantage to the simultaneous-growth-and-dialysis critical to this study is that the

microenvironment remained substantially constant. Design features that further distinguish

the bioreactor include scaffold-free design that permitted direct observation of the growth of

3D bone- like tissue and its interaction with metastatic cancer cells through in situ confocal

microscopy. Scaffold free design also permitted routine examination of tissue through

phase-contrast microscopy and relatively easy isolation of tissue (avoiding the complexity of

separating tissue from a 3D scaffold) at different stages of development for various analyses.

The surface area of the films was designed to be equivalent to standard T-25 tissue culture

dishes (25 cm2) to enable direct comparison culture conditions in the bioreactor and standard

2D cell culture. The bioreactor was designed, in principle, to accommodate application of

mechanical stress on the cells. This design feature, however, was not evaluated in this

particular study.

The implementation of the compartmentalized bioreactor for extended term culture of bone

cells is discussed in the next chapter.

32

References

1. Cormack DH. Ham's Histology. Philadelphia: J. B. Lippincott Company; 1987.

2. Ross MH, Kaye GI, W. P. Bone. Histology: A Text and Atlas. Philadelphia:

Lippincott Williams &Wilkins 2003. p 180–213.

3. Manolagas SSC. Birth and death of bone cells: basic regulatory mechanisms

and implications for the pathogenesis and treatment of osteoporosis. Endocrine

reviews 2000;21(2):115-37.

4. Holtrop ME. Light and electron microscopical study of bone forming cells. In:

Hall BK, editor. Bone: The Osteoblast and the Osteocyte. Caldwell, NJ: Telford

Press; 1990. p 1-40.

5. Simmons DJ, Grynpas MD. Mechanisms of bone formation in vivo. In: Hall

BK, editor. Bone: The Osteoblast and Osteocyte. Caldwell: Telford Press; 1990.

p 193-302.

6. Franz-Odendaal TA, Hall BK, Witten PE. Buried alive: How osteoblasts

become osteocytes. Developmental Dynamics 2006;235(1):176-190.

7. Ducy P, Schinke T, Karsenty G. The Osteoblast: A Sophisticated Fibroblast

under Central Surveillance. Volume 289; 2000. p 1501-1504.

33

8. Teitelbaum SL. Bone Resorption by Osteoclasts. Science 2000;289(5484):1504-

1508.

9. Frost HM. Bone Remodeling and Its Relationship to Metabolic Bone Disease.

Springfields MA: Charles C. Thomas; 1973.

10. Eriksen EF. Normal and pathological remodeling of human trabecular bone:

three dimensional reconstruction of the remodeling sequence in normals and

in metabolic bone disease. Endocr Rev 1986;7(4):379-408.

11. Parfitt AM. Osteonal and hemi-osteonal remodeling: the spatial and temporal

framework for signal traffic in adult human bone. Journal of cellular

biochemistry 1994;55(3):273-86.

12. Rodan GA, Martin TJ. Therapeutic Approaches to Bone DIseases. Science

2000;289(5484):1508-1514.

13. Rubens RD, Mundy GR. Cancer and the skeleton. London: Martin Dunitz;

2000.

14. Owen M. Lineage of osteogenic cells and their relationship to the stromal

system. In: Peck WA, editor. Bone and Mineral Research. Amsterdam:

Elsevier; 1985. p 1-25.

15. Tenenbaum HC. Cellular origins and theories of bone forming cells. In: Hall

BK, editor. Bone: The Osteoblast and Osteocyte. Caldwell, NJ: Telford Press;

1990. p 41-69.

34

16. Aubin J. Advances in the osteoblast lineage. Biochem Cell Biol 1998;76(6):899-

910.

17. Marks S, Popoff S. Bone cell biology: the regulation of development, structure,

and function in the skeleton. Am J Anat 1988;183(1):1-44.

18. Boskey AL. Matrix proteins and mineralization: an overview. Connect Tissue

Res 1996;35(1-4):357-63.

19. Roach HHI. Why does bone matrix contain non-collagenous proteins? The

possible roles of osteocalcin, osteonectin, osteopontin and bone sialoprotein in

bone mineralisation and resorption. Cell biology international 1994;18(6):617-

28.

20. Canalis E, McCarthy T, Centrella M. Growth factors and the regulation of

bone remodeling. J Clin Invest 1988;81(2):277-81.

21. Bronner F, Farrach-Carson MC. Bone Formation. New York: Springer; 2004.

22. Lian JB, Stein GS. Concepts of osteoblast growth and differentiation: basis for

modulation of bone cell development and tissue formation. Crit Rev Oral Biol

Med 1992;3(3):269-305.

23. Hock JM, Krishnan V, Onyia JE, Bidwell JP, Milas J, Stanislaus D. Osteoblast

Apoptosis and Bone Turnover. Volume 16; 2001. p 975-984.

24. Parfitt AM. Bone forming cells in clinical conditions. In: Hall BK, editor. Bone:

The Osteoblast and Osteocyte. Caldwell: Telford Press; 1990. p 351-429.

35

25. Palumbo CC. A three-dimensional ultrastructural study of osteoid-osteocytes

in the tibia of chick embryos. Cell and tissue research 1986;246(1):125-31.

26. Yellowley CE, Li Z, Zhou Z, Jacobs CR, Donahue HJ. Functional gap junctions

between osteocytic and osteoblastic cells. J Bone Miner Res 2000;15(2):209-17.

27. Donahue HJ, Li Z, Zhou Z, Yellowley CE. Differentiation of human fetal

osteoblastic cells and gap junctional intercellular communication. Am J

Physiol Cell Physiol 2000;278(2):C315-22.

28. Marotti GG. The osteocyte as a wiring transmission system. Journal of

musculoskeletal & neuronal interactions 2000;1(2):133-6.

29. Franz-Odendaal TA, Hall BK, Witten PE. Buried alive: How osteoblasts

become osteocytes. Developmental Dynamics 2006;235(1):176-190.

30. Knothe Tate ML, Adamson JR, Tami AE, Bauer TW. The osteocyte. The

International Journal of Biochemistry & Cell Biology 2004;36(1):1-8.

31. Menton DN, Simmons DJ, Orr BY, Plurad SB. A cellular investment of bone

marrow. Anat Rec 1982;203(1):157-64.

32. Palumbo C, Palazzini S, Marotti G. Morphological study of intercellular

junctions during osteocyte differentiation. Bone 1990;11(6):401-6.

33. Palumbo C, Palazzini S, Zaffe D, Marotti G. Osteocyte differentiation in the

tibia of newborn rabbit: an ultrastructural study of the formation of

cytoplasmic processes. Acta Anat (Basel) 1990;137(4):350-8.

36

34. Dudley HR, Spiro D. The fine structure of bone cells. Journal of Cell Biology

1961;11(3):627-649.

35. Bukka P, McKee MD, Karaplis AC. Molecular Recognition of Osteoblast

Differentiation. In: Bronner F, Farach-Carson MC, editors. Bone Formation.

London: Springer-Verlag; 2004. p 1-17.

36. Strangeways TSP, Fell HB. Experimental studies on the differentiation of

embryonic tissues growing in vivo and in vitro.--i. The development of the

undifferentiated limb-bud (a) when subcutaneously grafted into the post-

embryonic chick and (b) when cultivated in vitro. Proceedings of the Royal

Society of London 1926;99(698):340.

37. Fell HB, Robison R. The growth, development and phosphatase activity of

embryonic avian femora and limb-buds cultivated in vitro. The Biochemical

journal 1929;23(4):767-784.5.

38. Fell HB. The osteogenic capacity in vitro of periosteum and endosteum

isolated from the limb skeleton of fowl embryos and young chicks. Journal of

anatomy 1932;66(Pt 2):157-180.11.

39. Fell HB, Robison R. The development and phosphatase activity in vivo and in

vitro of the mandibular skeletal tissue of the embryonic fowl. The Biochemical

journal 1930;24(6):1905-21.

37

40. Fell HB, Robison R. The development of the calcifying mechanism in avian

cartilage and osteoid tissue. The Biochemical journal 1934;28(6):2243-53.

41. Fell HB. Experiments on the development in vitro of the avian knee-joint.

Proceedings of the Royal Society of London 1934;116(799):316.

42. Caplan AI, Pechak DG, . Cellular and molecular embyology of bone formation.

In: Peck WA, editor. Bone and Mineral Research Amsterdam: Elsevier Science

Publishers BV; 1987. p 117-183.

43. Stern PH, Raisz LG. Organ culture of bone. In: Simmons DJ, Kunin AS, editors.

Skeletal Research. New York: Academic Publisher; 1979. p 21-59.

44. Holtrop ME, Raisz LG, Simmons HA. The effects of parathyroid hormone,

colchicine, and calcitonin on the ultrastructure and the activity of osteoclasts

in organ culture. The Journal of cell biology 1974;60(2):346-55.

45. Stern PH, Krieger NS. Comparison of fetal rat limb bones and neonatal mouse

calvaria: effects of parathyroid hormone and 1,25-dihydroxyvitamin D3. Calcif

Tissue Int 1983;35(2):172-6.

46. Deiss WP, Jr., Holmes LB, Johnston CC, Jr. Bone Matrix Biosynthesis in Vitro.

I. Labeling of hexosamine and collagen of normal bone. 1962;237(11):3555-

3559.

47. Park HZ, Talmage RV. Relation of endogenous parathyroid secretion to 3H-

cytidine incorporation into bone cells. Endocrinology 1967;80(4):552-60.

38

48. Borle AB, Nichols N, Nichols G, Jr. Metabolic Studies of Bone in Vitro. I.

Normal bone. Volume 235; 1960. p 1206-1210.

49. Rose GG, Shindler TO. The Cytodifferentiation of Osteoblasts in Tissue

Culture: A Description of Cellular Emigrations from Embryo Chick-Leg Bones.

Volume 42; 1960. p 485-563.

50. Rose GG. A separable and multipurpose tissue culture chamber. Texas reports

on biology and medicine 1954;12(4):1074-83.

51. Rose GG. Special uses of the multipurpose tissue culture chamber. Texas

reports on biology and medicine 1957;15(2):310-2.

52. Rose GG. Cytopathophysiology of Tissue Cultures Growing Under Cellophane

Membranes. Int. Rev. Exp. Pathology. New York: Academic Press; 1966. p

111-178.

53. Nijweide PJ. Embryonic chicken periosteum in tissue culture, osteoid

formation and calcium uptake. Proc. Kon. Ned. Akad. Wed. Ser. C. Biol. Med.

Sci. 1975;78:410-417.

54. Nijweide PJ, van der Plas A, Scherft JP. Biochemical and histological studies

on various bone cell preparations. Calcified Tissue International

1981;V33(1):529-540.

39

55. Nijweide PJ, van der Plas A. Regulation of calcium transport in isolated

periosteal cells, effects of hormones and metabolic inhibitors. Calcified tissue

international 1979;29(2):155-61.

56. Tenenbaum HC, Heersche JN. Differentiation of osteoblasts and formation of

mineralized bone in vitro. Calcified tissue international 1982;34(1):76-9.

57. Tenenbaum HC, Heersche JN. Differentiation of osteoid-producing cells in

vitro: possible evidence for the requirement of a microenvironment. Calcified

tissue international 1986;38(5):262-7.

58. Tenenbaum HC, Heersche JN. Dexamethasone stimulates osteogenesis in chick

periosteum in vitro. Endocrinology 1985;117(5):2211-7.

59. Peck WA, Birge SJ, Brandt JJ. Collagen synthesis by isolated bone cells:

stimulation by ascorbic acid in vitro. Biochimica et biophysica acta

1967;142(2):512-25.

60. Peck WA, Dirksen TR. The metabolism of bone tissue in vitro. Clin Orthop

Relat Res 1966;48:243-65.

61. Peck WA, Birge SJ, Fedak SA. Bone Cells: Biochemical and Biological Studies

after Enzymatic Isolation. Science 1964;146(3650):1476-1477.

62. Peck WA, Burks JK, Wilkins J, Rodan SB, Rodan GA. Evidence for preferential

effects of parathyroid hormone, calcitonin and adenosine on bone and

periosteum. Endocrinology 1977;100(5):1357-64.

40

63. Dziak R, Brand JS. Calcium transport in isolated bone cells I. Bone cell

isolation procedures. Volume 84; 1974. p 75-83.

64. Dziak R, Brand JS. Calcium transport in isolated bone cells II. Calcium

transport studies. Volume 84; 1974. p 85-96.

65. Smith DM, Johnston CC, Jr., Severson AR. Studies of the metabolism of

separated bone cells. I. Techniques of separation and identification. Calcif

Tissue Res 1973;11(1):56-69.

66. Yagiela JA, Woodbury DM. Enzymatic isolation of osteoblasts from fetal rat

calvaria. Anat Rec 1977;188(3):287-306.

67. Wong G, Cohn DV. Separation of parathyroid hormone and calcitonin-

sensitive cells from non-responsive bone cells. Nature 1974;252(5485):713-5.

68. Wong GL, Cohn DV. Target cells in bone for parathormone and calcitonin are

different: enrichment for each cell type by sequential digestion of mouse

calvaria and selective adhesion to polymeric surfaces. Proc Natl Acad Sci U S A

1975;72(8):3167-71.

69. Rao LG, Ng B, Brunette DM, Heersche JN. Parathyroid hormone- and

prostaglandin E1-response in a selected population of bone cells after repeated

subculture and storage at -80C. Endocrinology 1977;100(5):1233-41.

70. Heersche JN, Moe HK, Brunette DM, Rao AV, Riemers S, Rao LG. Long-term

culture of parathyroid hormone and prostaglandin E1 responsive bone cells in

41

monolayers and on artificial capillaries. Calcif Tissue Res 1977;22 Suppl:275-

82.

71. Luben RA, Wong GL, Cohn DV. Parathormone-stimulated resorption of

devitalised bone by cultured osteoclast-type bone cells. Nature

1977;265(5595):629-30.

72. Bellows CG, Aubin JE, Heersche JNM, Antosz ME. Mineralized Bone Nodules

from In Vitro from Enzymatically Released Rat Calvaria Cell Populations.

Calcified Tissue International 1986;38:143-154.

73. Sudo H, Kodama HA, Amagai Y, Yamamoto S, Kasai S. In Vitro Differentiation

and Calcification in a New Clonal Osteogenic Cell Line Derived from

Newborn Mouse Calvaria. The Journal of Cell Biology 1983;96:191-198.

74. Ecarot-Charrier B, Glorieux FH, Rest Mvd, Pereira G. Osteoblasts Isolated

from Mouse Calvaria Initiate Matrix Mineralization in Culture. The Journal of

Cell Biology 1983;96:639-643.

75. Koshihara Y, Kawamura M, Oda H, Higaki S. In vitro calcification in human

osteoblastic cell line derived from periosteum. Biochem Biophys Res Commun

1987;145(2):651-7.

76. Kadis B, Goodson JM, Offenbacher S, Bruns JW, Seibert S. Characterization of

osteoblast-like cells from fetal rat calvaria. J Dent Res 1980;59(11):2006-13.

42

77. Binderman I, Berger E, Fine N, Shimshoni Z, Harell A, Somjen D. Calvaria

derived osteogenic cells: phenotypic expression in culture. Connect Tissue Res

1989;20(1-4):41-7.

78. Benayahu D, Kletter Y, Zipori D, Wientroub S. Bone marrow-derived stromal

cell line expressing osteoblastic phenotype in vitro and osteogenic capacity in

vivo. J Cell Physiol 1989;140(1):1-7.

79. Aubin JE, Heersche JN, Merrilees MJ, Sodek J. Isolation of bone cell clones

with differences in growth, hormone responses, and extracellular matrix

production. J Cell Biol 1982;92(2):452-61.

80. Binderman I, Duksin D, Harell A, (Katchalski) EK, Sachs L. Formation of Bone

Tissue in Culture from Isolated Bone Cells. The Journal of Cell Biology

1974;61:427-439.

81. Bellows CG, Aubin JE, Heersche JNM. Physiological concentrations of

glucocorticoids stimulate formation of bone nodules from isolated rat calvaria

cells in vitro. Endocrinology 1987;121(6):1985-1992.

82. Ecarot-Charrier B, Shepard N, Charette G, Grynpas M, Glorieux F.

Mineralization in osteoblast cultures: a light and electron microscopic study.

Bone 1988;9(3):147-54.

43

83. Gerstenfeld LC, Chipman SD, Glowacki J, Lian JB. Expression of Differentiated

Function by Mineralizing Cultures of Chicken Osteoblasts. Developmental

Biology 1987;122:49-60.

84. Maniatopoulos C, Sodek J, Melcher AH. Bone formation in vitro by stromal

cells obtained from bone marrow of young adult rats. Cell and Tissue Research

1988;254:317-330.

85. Nefussi JR, Boy-Lefevre ML, Boulekbache H, Forest N. Mineralization in vitro

of matrix formed by osteoblasts isolated in collagenase digestion.

Differentiation 1985;29:160-168.

86. Robey PG, Termine JD. Human Bone Cells In Vitro. Calcified Tissue

International 1985;37:453-460.

87. Satomura K, Hiraiwa K, Nagayama M. Mineralized nodule formation in rat

bone marrow stromal cell culture without Beta-glycerophosphate. Bone and

Mineral 1991;14:41-54.

88. Satomura K, Nagayama M. Ultrastructure of Mineralized Nodules Formed in

Rat Bone Marrow Stromal Cell Culture in vitro. Acta Anatomica 1991;142:97-

104.

89. Tenenbaum HC, Heersche JNM. Differentiation of Osteoblasts and Formation

of Mineralized Bone in Vitro. Calcified Tissue International 1982;34:76-79.

44

90. Malaval L, Modrowski D, Gupta A, Aubin J. Cellular expression of bone-

related proteins during in vitro osteogenesis in rat bone marrow stromal cell

cultures. J Cell Physiol 1994;158(3):555-72.

91. Rodan GA, Rodan SB. Expression of the osteoblastic phenotype. Bone and

Mineral Research (edited by William A. Peck) 1984;2:244-285.

92. Declercq H, Van den Vreken N, De Maeyer E, Verbeeck R, Schacht E, De

Ridder L, Cornelissen M. Isolation, proliferation and differentiation of

osteoblastic cells to study cell/biomaterial interactions: comparison of different

isolation techniques and source. Biomaterials 2004;25(5):757-768.

93. Osdoby P, Caplan AI. The possible differentiation of osteogenic elements in

vitro from chick limb mesodermal cells. I. Morphological evidence. Dev Biol

1976;52(2):283-99.

94. Williams DC, Boder GB, Toomey RE, Paul DC, Hillman CC, Jr., King KL, Van

Frank RM, Johnston CC, Jr. Mineralization and metabolic response in serially

passaged adult rat bone cells. Calcif Tissue Int 1980;30(3):233-46.

95. Kale S, Biermann S, Edwards C, Tarnowski C, Morris M, Long M. Three-

dimensional cellular development is essential for ex vivo formation of human

bone. Nat Biotechnol 2000;18(9):954-8.

96. Martin I, Wendt D, Heberer M. The role of bioreactors in tissue engineering.

Trends in Biotechnology 2004;22(2):80-86.

45

97. Braddock M, Houston P, Campbell C, Ashcroft P. Born Again Bone: Tissue

Engineering for Bone Repair. News Physiol Sci 2001;16(5):208-213.

98. Bruder S, Fox B. Tissue engineering of bone. Cell based strategies. Clin Orthop

1999(367 Suppl):S68-83.

99. Bruder S, Kurth A, Shea M, Hayes W, Jaiswal N, Kadiyala S. Bone regeneration

by implantation of purified, culture-expanded human mesenchymal stem cells.

J Orthop Res 1998;16(2):155-62.

100. Holy C, Shoichet M, Davies J. Engineering three-dimensional bone tissue in

vitro using biodegradable scaffolds: investigating initial cell-seeding density

and culture period. J Biomed Mater Res 2000;51(3):376-82.

101. Laurencin CT, Ambrosio AMA, Borden MD, Cooper JA. Tissue Engineering:

Orthopedic Applications. Annual Review of Biomedical Engineering

1999;1(1):19-46.

102. Petite H, Viateau V, Bensaid W, Meunier A, de Pollak C, Bourguignon M,

Oudina K, Sedel L, Guillemin G. Tissue-engineered bone regeneration. Nat

Biotechnol 2000;18(9):959-63.

103. Shin M, Yoshimoto H, Vacanti J. In vivo bone tissue engineering using

mesenchymal stem cells on a novel electrospun nanofibrous scaffold. Tissue

Eng 2004;10(1-2):33-41.

46

104. Vacanti C, Bonassar L. An overview of tissue engineered bone. Clin Orthop

1999(367 Suppl):S375-81.

105. Sikavitsas V, Bancroft G, Mikos A. Formation of three-dimensional

cell/polymer constructs for bone tissue engineering in a spinner flask and a

rotating wall vessel bioreactor. J Biomed Mater Res 2002;62(1):136-48.

106. Wang Y, Uemura T, Dong J, Kojima H, Tanaka J, Tateishi T. Application of

perfusion culture system improves in vitro and in vivo osteogenesis of bone

marrow-derived osteoblastic cells in porous ceramic materials. Tissue Eng

2003;9(6):1205-14.

107. Freed LE, Vunjak-Novakovic G. Culture of organized cell communities.

Advanced Drug Delivery Reviews 1998;33:15-30.

108. Ferrera D, Poggi S, Biassoni C, Dickson GR, Astigiano S, Barbieri O, Favre A,

Franzi AT, Strangio A, Federici A and others. Three-dimensional cultures of

normal human osteoblasts: proliferation and differentiation potential in vitro

and upon ectopic implantation in nude mice. Bone 2002;30(5):718-725.

109. van den Dolder J, Vehof J, Spauwen P, Jansen J. Bone formation by rat bone

marrow cells cultured on titanium fiber mesh: effect of in vitro culture time. J

Biomed Mater Res 2002;62(3):350-8.

110. Sikavitsas V, van den Dolder J, Bancroft G, Jansen J, Mikos A. Influence of the

in vitro culture period on the in vivo performance of cell/titanium bone tissue-

47

engineered constructs using a rat cranial critical size defect model. J Biomed

Mater Res A 2003;67(3):944-51.

111. Centrella MM, Canalis EE. Local regulators of skeletal growth: a perspective.

Endocrine reviews 1985;6(4):544-51.

112. Canalis E. Effect of growth factors on bone cell replication and differentiation.

Clin Orthop Relat Res 1985(193):246-63.

113. Glimcher MJ. On the form and function of bone: from molecules to organs.

Wolff's law revisited. In: Veis A, editor. The Chemistry and Biology of

Mineralized Connective Tissues. New York: Elsevier North Holland; 1981. p

617-73.

114. Raisz LG, Kream BE. Regulation of bone formation. The New England journal

of medicine 1983;309(1):29-35.

115. Freed LE, Vunjak-Novakovic G. Culture of organized cell communities.

Advanced Drug Delivery Reviews 1998;33(1-2):15-30.

116. Freshney RI. Culture of Animal Cells: A Manual of Basic Technique. New

York: Wiley-Liss Inc; 2000.

117. Reynolds JJ. Skeletal Tissue In Culture. In: Bourne GH, editor. The

Biochemistry and Physiology of Bone. New York: Academic Press; 1972.

48

118. Biggers JD, Gwatkin RB, Heyner S. Growth of embryonic avian and

mammalian tibiae on a relatively simple chemically defined medium. Exp Cell

Res 1961;25:41-58.

119. Burks JK, Peck WA. Bone cells:a serum-free medium supports proliferation in

primary culture. Science 1978;199:542-544.

120. Nijweide PJ, Burger EH, Hekkelman JW, Herrmann-Erlee MPM, Gaillard PJ.

Regulatory mechanisms in the development of bone and cartilage: the use of

tissue culture techniques in the study of the development of embryonic bone

and cartilage: a perspective. Factors and Mechanisms influencing bone growth.

New York: Alan R. Liss Inc; 1982.

121. Nijweide PJ, van Iperen-van Gent AS, Kawilarang-de Haas EW, van der Plas

A, Wassenaar AM. Bone formation and calcification by isolated osteoblastlike

cells. J Cell Biol 1982;93(2):318-23.

122. Hall BK. Cellular differentiation in skeletal tissues. Biol Rev Camb Philos Soc

1970;45(4):455-84.

123. McLimans WF. Kinetics of gas diffusion in mammalian cell culture systems. I.

Experimental. Biotechnology and bioengineering 1968;10(6):725.

124. McLimans WF. Kinetics of gas diffusion in mammalian cell culture systems. II.

Theory. Biotechnology and bioengineering 1968;10(6):741.

49

125. Hill AV. The Diffusion of Oxygen and Lactic Acid through Tissues.

Proceedings of the Royal Society of London 1928;104(728):39.

126. Hasegawa S, Sato S, Saito S, Suzuki Y, Brunette DM. Mechanical stretching

increases the number of cultured bone cells synthesizing DNA and alters their

pattern of protein synthesis. Calcif Tissue Int 1985;37(4):431-6.

127. Meikle MC, Reynolds JJ, Sellers A, Dingle JT. Rabbit cranial sutures in vitro: a

new experimental model for studying the response of fibrous joints to

mechanical stress. Calcif Tissue Int 1979;28(2):137-44.

128. Hall BK. In vitro studies on the mechanical evocation of advenitious cartilage

in the chick. J Exp Zool 1968;168(3):283-305.

129. Copray JC, Jansen HW, Duterloo HS. Effect of compressive forces on

phosphatase activity in mandibular condylar cartilage of the rat in vitro. J Anat

1985;140 ( Pt 3):479-89.

130. Copray JC, Jansen HW, Duterloo HS. An in-vitro system for studying the

effect of variable compressive forces on the mandibular condylar cartilage of

the rat. Arch Oral Biol 1985;30(4):305-11.

131. Copray JC, Jansen HW, Duterloo HS. Effects of compressive forces on

proliferation and matrix synthesis in mandibular condylar cartilage of the rat

in vitro. Arch Oral Biol 1985;30(4):299-304.

50

132. Klein-Nulend J, Veldhuijzen JP, de Jong M, Burger EH. Increased bone

formation and decreased bone resorption in fetal mouse calvaria as a result of

intermittent compressive force in vitro. Bone Miner 1987;2(6):441-8.

133. Bancroft GN, Sikavitsas VI, van den Dolder J, Sheffield TL, Ambrose CG,

Jansen JA, Mikos AG. Fluid flow increases mineralized matrix deposition in 3D

perfusion culture of marrow stromal osteoblasts in a dose-dependent manner.

PNAS 2002;99(20):12600-12605.

134. Delmas PD, Anderson M. Launch of the Bone and Joint Decade 2000-2010.

Osteoporos Int. 2000;11:95-97.

135. Kiberstis P, Smith O, Norman C. Bone Health in the Balance. Science

2000;289(5484):1497.

136. Praemer A, Furner S, Rice DF. Musculoskeletal Conditions in the United

States. Rosemont, IL: American Academy of Orthopeadic Surgeons; 1999.

137. Bancroft G, Sikavitsas V, Mikos A. Design of a flow perfusion bioreactor

system for bone tissue-engineering applications. Tissue Eng 2003;9(3):549-54.

138. Granet C, Laroche N, Vico L, Alexandre C, Lafage-Proust M. Rotating-wall

vessels, promising bioreactors for osteoblastic cell culture: comparison with

other 3D conditions. Med Biol Eng Comput 1998;36(4):513-9.

51

139. Botchwey E, Pollack S, Levine E, Laurencin C. Bone tissue engineering in a

rotating bioreactor using a microcarrier matrix system. J Biomed Mater Res

2001;55(2):242-53.

140. J.O. Hollinger, T.O. Einhorn, B. A. Doll, C. Sfeir, editors. Bone Tissue

Engineering. Boca Raton, FL CRC Press; 2005.

141. Bronner F, Farach-Carson MC, Mikos AG, editors. Engineering of Functional

Skeletal Tissues. London: Springer-Verlag; 2007.

142. Abousleiman RI, Sikavitsas VI. Bioreactors for tissues of the musculoskeletal

system. Adv Exp Med Biol 2006;585:243-59.

143. Overstreet M, Sohrabi A, Polotsky A, Hungerford DS, Frondoza CG. Collagen

microcarrier spinner culture promotes osteoblast proliferation and synthesis of

matrix proteins. In vitro cellular & developmental biology. Animal 2003;39(5-

6):228-34.

144. Granet C, Laroche N, Vico L, Alexandre C, Lafage-Proust MH. Rotating-wall

vessels, promising bioreactors for osteoblastic cell culture: comparison with

other 3D conditions. Medical & biological engineering & computing

1998;36(4):513-9.

145. Sikavitsas VI, Bancroft GN, Holtorf HL, Jansen JA, Mikos AG. Mineralized

matrix deposition by marrow stromal osteoblasts in 3D perfusion culture

increases with increasing fluid shear forces. PNAS 2003;100(25):14683-14688.

52

146. Terai H. In vitro engineering of bone using a rotational oxygen-permeable

bioreactor system. Materials science and engineering 2002;20(1-2):3.

147. Botchwey EA, Pollack SR, Lavine EM, Laurencin CT. Bone Tissue Engineering

in a Rotating Bioreactor Using a Microcarrier Matrix System. J. Biomed. Mater.

Res. 2001;55:242-253.

148. Vassilios I. Sikavitsas GNBAGM. Formation of three-dimensional cell/polymer

constructs for bone tissue engineering in a spinner flask and a rotating wall

vessel bioreactor. Journal of Biomedical Materials Research 2002;62(1):136-

148.

149. Dhurjati R, Liu X, Gay CV, Mastro AM, Vogler EA. Extended-Term Culture of

Bone Cells in a Compartmentalized Bioreactor. Tissue Engineering

2006;12:3045-3054.

150. Jansen FW, Oostra J, Oorshot Av, Blitterswijk CAv. A perfusion bioreactor

system capable of producing clinically relevant volumes of tissue-engineered

bone: In vivo bone formation showing proof of concept. Biomaterials

2006;27(3):315-323.

151. Matziolis G, Tuischer J, Kasper G, Thompson M, Bartmeyer B, Krocker D,

Perka C, Duda G. Simulation of cell differentiation in fracture healing:

mechanically loaded composite scaffolds in a novel bioreactor system. Tissue

engineering 2006;12(1):201-8.

53

152. Sikavitsas VI, Temenoff JS, Mikos AG. Biomaterials and bone

mechanotransduction. Biomaterials 2001;22(19):2581-93.

153. Owan I, Burr DB, Turner CH, Qiu J, Tu Y, Onyia JE, Duncan RL.

Mechanotransduction in bone: osteoblasts are more responsive to fluid forces

than mechanical strain. American journal of physiology 1997;273(3 Pt 1):5.

154. Ignatius A, Blessing H, Liedert A, Schmidt C, Neidlinger-Wilke C, Kaspar D,

Friemert B, Claes L. Tissue engineering of bone: effects of mechanical strain on

osteoblastic cells in type I collagen matrices. Biomaterials 2005;26(3):311-318.

155. Cartmell SH, Porter BD, Garcia AJ, Guldberg RE. Effects of medium perfusion

rate on cell-seeded three-dimensional bone constructs in vitro. Tissue Eng.

2003;9:1197-1203.

54

List of Table Legends

Table 1: Principal extracellular matrix proteins secreted by the osteoblasts and their

sequence of expression.

Table 2: Factors in the bone microenvironment secreted that condition the growth and

differentiation of the osteoblasts.

Table 3: Historical time line of the development of in vitro models of bone formation.

Table 4: Representative bioreactors for bone tissue engineering applications.

Table 5: Unique design features of the compartmentalized bioreactor

55

Principal Extracellular Matrix Proteins

MW(kDa)

Characteristics Sequence of Expression

Collagen type I ~320 Principal structural component of bone extracellular matrix.

Expressed early, during the proliferation period. Gene expression gradually downregulated with mRNA maintained at low ostoecalcin basal level at subsequent stages of differentiation.

Osteopontin 60 Acidic glycoprotein.Arg-Gly-Asp containing sequence mediates cell attachment. Binds to Hydroxyapatite.

Expressed early and late in the osteoblastdevelopment sequence.

Osteocalcin 5.7 Binds to Hydroxyapatite.Vitamin K dependentCharacterized by three γ-carboxyglutamic acid residues(Gla). Marker for mature osteoblasts. Possible role in coupling of bone formation to resorption.

Expressed post-proliferatively with onset of nodule formation.

Maximal expression with mineralizaton.

Expression highly correlated to calcification of the extracellular matrix.

Phosphorylatedglycoprotein, binds calcium, cell attachment.

Osteonectin 33 Associated with mineralization.

Table1: Principal extracellular matrix proteins secreted by the osteoblasts and their sequence of expression. Adapted from Reference (22)

56

Table 2: Factors in the bone microenvironment secreted that condition the growth and differentiation of the osteoblasts.

Protein MW(kDa) IGF-I 7.5 IGF-II 7.5 aFGF ~16 bFGF ~16

IL-1

Growth factors and cytokines secreted by osteoblasts

TGF-β 25 BMP 18.5 OIF 22-28

PDGF 30 EGF ~6.4

BDGF 11

Factors derived from the Bone and other sources

Prostaglandins <1 IGF- Insulin Derived Growth Factor aFGF- Acidic Fibroblast Derived Growth Factor bFGF-Basic Fibroblast Derived Growth Factor BMP- Bone Morphogenic Protein OIF-Osteoinductive Factor PDGF- Platelet Derived Growth Factor IL-Interleukin EGF-Epidermal Growth Factor BDGF-Blood Derived Growth Factor * Cytokines are polypeptides (proteins) that regulate many cell functions. Cytokines that act on the same cell that produced them are called autocrine factors; those that act on other cells are called paracrine factors; those that act systemically (through the vascular system) are referred to as endocrine factors. Molecules that regulate cell division are referred to as growth factors.

57

Table 3: Historical time line of the development of in vitro models of bone formation.

Organ Cultures Fell and Robison (1929)

Cultivation of isolated femora and tibiae from 51/2 to 6 day chick embryos using watch glass technique.

First successful demonstration of calcification of bone in vitro

[39]

Explant Cultures Fell (1932)

Culture of periosteum and endosteum from embryonic chick calvaria using hanging drop method.

Demonstration of the osteogenic capacity of the periosteum and endosteum in vitro.

[40]

Rose (1960)

Culture of bone explants from the long bone of embryonic chick in culture chambers made of cellophane

Weeks long culture of explants in specially designed culture chambers

[51]

Nijweide (1975) Culture of folded periostea from embryonic chick calvaria

Study of osteoid formation , calcium transport and effects of hormones

[55]

Tenenbaum and Heersche (1982)

Culture of folded periostea from embryonic chick calvaria

Osteoid deposition and mineralization in the presence of β-glycerophosphate

[58]

Culture of Isolated Osteoblasts Peck et al. (1964) Isolation of osteoblasts in

culture from rat calvariae using buffered collagenase.

Landmark study that pioneered isolation of osteoblasts using enzymatic techniques.

[63]

Smith et al. (1973) Isolation of four different types of cells by mechanical separation from rat calvaria

Studies on the short term (hours) metabolism of isolated bone cells.

[67]

Binderman et al (1974)

Isolation of osteoblasts in culture from rat calvariae using trypsin and EDTA

Weeks long culture of osteoblasts showing deposition and mineralization of extracellular matrix.

[82]

Wong and Cohn (1974) Rao et al. (1977) Leuben et al . (1977)

Isolation of osteoblasts using sequential enzymatic treatment

Isolation of sub-populations of osteogenic cells

[69] [71] [73]

Kadis et al. (1980) Aubin et al. (1982) Ecarot-Charier et al. (1983) Sudo et al (1983) Bellows et al (1986)

Development of clonal populations from fetal rat and mouse calvarial cells

Increased homogeneity in cell populations

[78] [81] [84] [75] [74]

Robey and Termine (1985)

Establishment of human bone cells in vitro

[88]

Nefussi et al (1985) Bellows et al(1986)

Formation of discrete mineralized nodules in the presence of ascorbic acid and organic phosphate.

Mineralized nodule formation in culture

[87] [74]

Lian and Stein (1990)

Sequential progression of osteoblast differentiation using fetal rat and chick calvarial oteoblasts

Osteoblast development sequence

[22]

58

Table 4: Representative bioreactors for bone tissue engineering applications.

Types of Bioreactors

Principle of Operation Scaffolds Cells Ref.

Rotating-wall vessels (RWV)

RWV consists of two concentric cylinders- a stationary gas-permeable inner cylinder and a rotating outer cylinder. Scaffolds are placed in the annual space immersed in culture medium. Carefully selected rotational rates balance gravitational and centrifugal forces acting on the scaffolds resulting in microgravity-like conditions.

Microcarrier beads made of dextran based polymers.

Rat osteosarcoma cell line ROS 17/2.8 cells

[144]

Perfusion Culture Reactors

Continuous perfusion of medium through the scaffolds using a peristaltic pump.

Porous ceramic materials composed of β-tricalcium phosphate

Osteoblasts derived from the bone marrow of 7 week-old Fischer 344 male rats

[106]

Rotating Vessels

50 ml polypropylene centrifuge tube with a 30x40mm silicone window for gas exchange placed on roller culture apparatus rotating at 6 RPM

3D poly(DL-lactide-co-glycolide) PLGA foams

Osteoblastic cells from ther marrow of 3-7 days old neonatal Lewis rats

[146]

Spinner Flasks

Scaffolds seeded with cells are attached to needles hanging from the cover of the flask. Magnetic stir bar at the bottom of the flask is used to effect mixing of the medium

3D poly(DL-lactide-co-glycolide)PLGA foams

Rat marrow stromal cells

[143]

59

Table 5: Unique Design Features of the Compartmentalized Bioreactor Requirement Design Implementations Mass Transport of Nutrients Continuous- growth and dialysis Mass Transport of Biochemical Factors Limited to the cell growth compartment by the

dialysis membrane Flow of Medium Static Delivery of Oxygen Gas-permeable films made of a co-polymer of

ethylene methacrylate designed to provide optimal oxygen tensions.

Cell-Scaffold Interactions Scaffold free design. Cell-substrate function is performed by the dialysis membrane.

Microscopy Optically clear gas-permeable films enable observation of tissue through phase contrast or confocal microscopy.

60

Chapter 3

Extended Term Culture of Bone Cells in a Compartmentalized Bioreactor

Abstract

A specialized bioreactor is used to grow mineralizing, collagenous tissue up to

150 mμ thick from an inoculum of isolated murine (mouse calvaria MC3T3-E1,

ATCC CRL-2593) or human (hFOB 1.19 ATCC CRL-11372) osteoblasts over

uninterrupted culture periods longer than 120 d (4 months ). Proliferation and

phenotypic progression of an osteogenic-cell monolayer into a tissue comprised

of 6≥ cell layers of mature osteoblasts in the bioreactor was compared to cell

performance in conventional tissue-culture polystyrene (TCPS) controls. Cells in

the bioreactor basically matched results obtained in TCPS over a 15 d culture

interval, but loss of insoluble ECM (iECM) and ~2X increase in apoptosis rates in

TCPS after 30 d indicated progressive instability of cultures maintained in TCPS

with periodic refeeding but without subculture. By contrast, stable cultures were

maintained in the bioreactor for more that 120 d, suggesting that extended-term

tissue maintenance is feasible with little-or-no special technique. TEM

ultramorphology of tissue derived from hFOB 1.19 cells recovered from the

bioreactor after only 15 d of culture showed evidence of osteocytic-like processes

61

and gap junctions between cells like that observed in vivo, over-and-above

elaboration of the usual osteoblastic markers such as alkaline phosphatase

activity and mineralization (alizarin-red). Thus, the bioreactor design based on

the principle of simultaneous-growth-and-dialysis was shown to create an

extraordinarily-stable pericellular environment that better simulates the in vivo

condition than conventional tissue-culture. The bioreactor shows promise as a

tool for the in vitro study of osteogenesis and osteopathology.

62

1. Introduction

Cells extracted from the native, in vivo physiological state and placed into

a culture system undergo adaptive responses often referred to as “culture shock”

(1). The impact and duration of culture shock strongly depends on the stability

of the pericellular (micro) environment, the extent to which this

microenvironment simulates the in vivo condition, and the ability of cells to

actively interact/transform the pericellular milieu by secreting a variety of

macromolecules found in extracellular matrix (ECM) (2, 3). Indeed, culture

shock and the accumulated damage cells sustain in culture has long been

thought to limit cell viability in vitro (4, 5). Thus, one of the bioengineering

objectives for any in vitro culture device is to create a stable pericellular

environment simulating the in vivo condition in a manner that mitigates, rather

than amplifies, the impact of culture shock. The conventional tissue-culture

approach is to surround cells held in dishes or flasks with a buffered medium

containing various nutrients (amino acids, glucose, serum proteins, vitamins,

etc.). As cells grow, nutrients are depleted, waste products accumulate

(especially lactic acid), and pericellular pH decreases to unacceptable levels. The

typical solution to this problem is to exchange spent growth medium with fresh,

either on a continuous basis (perfusion) or as a matter of scheduled maintenance.

A decided disadvantage with either of these approaches is that the

63

aforementioned secreted macromolecules are removed along with waste

products. In the case when periodic media exchanges are employed, the

resulting oscillation in both pH and nutrient concentrations imparts an instability

in cultures that can lead to excessive cell vacuolization, ruffled cell margins, and

increasing rates of cell-surface detachment. Thus, conventional tissue-culture

strategies, especially the flask-and-dish type serviced with periodic media

exchanges, fail to maintain a stable pericellular environment critical to recovery

from culture shock. As applied specifically to the culture of bone cells in vitro,

continuous removal or periodic exchange of growth medium also means that

cell/protein-mediated dissolution-precipitation reactions involved in

mineralization-resorption are likewise perturbed. Thus it is difficult to simulate

the natural developmental sequence of bone cells that includes proliferation,

post-mitotic expression of a differentiated phenotype, extracellular matrix

formation, and mineralization. As an example relevant to this work,

transformation of mature osteoblasts into an osteocytic phenotype in vitro has not

been reported to our knowledge.

More than a decade before widespread use of conventional tissue culture

in biotechnology, G. G. Rose pioneered the “simultaneous-growth-and-dialysis”

method (6) by culturing cells beneath a cellulosic membrane (commercial

cellulose wrapping film in Rose’s implementation). The core idea was to

64

continuously feed cells with low-molecular-weight metabolites by dialysis

through a cellulose membrane that also retained secreted high-molecular-weight

macromolecules within the growth space created by the bounding membrane.

Metabolic waste products such as lactic acid dialyzed out of the pericellular

space and into the basal medium where waste could be removed without

perturbing cells by wholesale growth-medium removal. Rose’s idea languished

in the literature until its rediscovery more than twenty years later when

simultaneous-growth-and-dialysis was bioengineered into a routinely-usable, so-

called compartmentalized culture device (7, 8), one version of which is shown

Fig. 1. This bioreactor was shown to facilitate a number of advantages in the

culture/maintenance of soft-tissue and hybridoma cells, perhaps most notably of

which was the ability to sustain cells for more than 30 d without user

intervention and the accumulation of biosynthesized macromolecules (such as

IgG) within the growth space for periodic harvest (8). Unpublished work

(Vogler) suggests advantages in the culture of recalcitrant primary cells and in

vitro immunization as well.

Our interest in the development of improved orthopedic biomaterials and

scaffolds for orthopedic tissue engineering (9-11) has led the need for in vitro test

vehicles that permit extended-term (>30 d) contact of osteoblasts, osteoclasts, and

co-cultures thereof with candidate biomaterials. Finding many limitations with

65

conventional culture, not the least of which was loss of culture integrity over

long culture intervals, we were motivated to implement the compartmentalized

bioreactor for bone-cell culture. Herein we report phenotypic development of

two continuous osteoblast cell lines from an osteogenic-cell monolayer to a

mineralizing, collagenous tissue; mouse calvaria MC3T3-E1 (ATCC CRL-2593)

and human fetal osteoblasts (hFOB 1.19, ATCC CRL-11372). TEM

ultramorphology of hFOB 1.19-derived tissue recovered from the bioreactor after

15 d shows evidence of osteocytic-like processes and regions of intercellular

contacts between cells, suggesting that osteoblasts undergo complex phenotypic

development within the bioreactor like that observed in vivo when osteoblasts

are engulfed in a mineralizing tissue. Thus, the compartmentalized bioreactor

shows promise as a tool for the in vitro study of osteogenesis and osteopathology.

2. Methods and Materials

Bioreactor Design and Implementation: The compartmentalized

bioreactor (Fig.1, Panel a) separates a cell-growth space (A, 5 mL) from a larger-

volume medium reservoir (B, 30 mL) with a dialysis membrane (C). Cells were

cultured on a transparent, liquid-impermeable film (E) selected for

cytocompatibility (12-14) and gas (O2 and CO2) permeability (see further below).

During culture, cells were continuously bathed in pH-equilibrated and

oxygenated medium dialyzing from the reservoir (B). At the same time,

66

metabolic waste products such as lactic acid dialyzed out of the growth

compartment (A), maintaining low-pericellular concentrations. Serum

constituents or macromolecules synthesized by cells with molecular weights in

excess of the dialysis-membrane cutoff (6-8 kDa in our implementation) were

retained and concentrated within the growth compartment. The entire vessel was

ventilated through transparent gas-permeable films bounding cell growth and

reservoir compartments (D, E). The assembled bioreactor had a 25 cm2 cell–

growth area. The medium reservoir volume was designed for medium-

replenishment intervals ranging from 15-45 d, depending on the metabolic

activity of the cells. Access to the cell-growth space and medium reservoir was

through luer taper ports (J, K; Panel c Fig. 1) using standard pipettes. The

bioreactor was designed to work with most standard inverted phase-contrast

microscopes with minor stage modification and allowed adequate optical

microscopy throughout the culture interval; although resolution at the cellular

level naturally was compromised by the development of a thick, collagenous

multi-cellular tissue. The bioreactor was machined from 316L stainless-steel

stock (8). The body of the compartmentalized device consisted of four main rings

(F, G H, I; Panel c). Two chambers (A, B) were created with three films (C, D, E)

sandwiched between inner rings (F, G). The whole device was held together in a

liquid tight fashion using 6 stainless steel screws, as can be seen in the laboratory

implementation of Panel b. Access to, and venting of, growth and reservoir

67

chambers was through Luer-taper ports (J, K). The gas permeable films (G, H)

forming the outer barriers for chambers (A, B) were ~3 mil thick and made by

hot pressing Surlyn 1702 resin (DuPont, Wilmington, DE) by simultaneous

application of heat (220 oC) and pressure (245 Pa) in a laboratory hot press

(Model 2699, Fred S. Carver Inc.). The internal film barrier was cellulosic-

dialysis membrane (Spectrapor-13266, Spectrum Medical Industries) and was

hydrated in deionized water for at least 1 hr before assembly of the bioreactor.

Fabricated bioreactors were filled with 0.1% sodium azide prepared in phosphate

buffer saline, packaged in plastic bags, and sterilized using 10 Mrad γ-ray

irradiation at the Breazeale Nuclear Reactor on the campus of the Pennsylvania

State University. Sterile-packed bioreactors were opened within the confines of a

laminar-flow biosafety cabinet, drained of azide storage solution, and 3X rinsed

with basal medium using conventional aseptic technique just before cell

inoculation as described below.

Cells and Cell Culture: Murine calvaria pre-osteoblasts (MC3T3-E1,

ATCC CRL-2593) and human fetal osteoblastic cells (hFOB 1.19, ATCC CRL-

11372) were obtained form American Type Culture Collection (ATCC). MC3T3-

E1 were cultured in alpha minimum-essential medium (α-MEM) supplemented

with 10% charcoal-stripped fetal bovine serum, and 1% penicillin-streptomycin.

Post-confluent MC3T3-E1 were cultured in a differentiation medium additionally

68

containing 50 µg/mL ascorbic acid and 10 mM β-glycerophosphate. hFOB 1.19

were cultured using Dulbecco’s modified Eagle medium-Ham’s F-12 (1:1) basal

media supplemented with 10% charcoal-stripped fetal bovine serum, and 1%

penicillin-streptomycin. Post-confluent hFOB 1.19 were cultured in a more

complex differentiation medium additionally containing 50 µg/mL ascorbic

acid, 10-8M 1,25-dihydroxy vitamin D3 and 10-8M menadione. The hFOB 1.19 cell

line is conditionally immortalized by transfection with a gene encoding for the

temperature-sensitive mutant (tsA58) of the SV40 large T antigen. Transfection

confers a 33.5 oC continuous proliferation “permissive” temperature and 39 oC

quiescent temperature at which cells stop proliferating without undergoing

apoptosis (15). We cultured hFOB 1.19 at 37 oC and obtained a doubling time

similar to that observed at the permissive temperature (16).

Culture experiments were initiated in rinsed bioreactors (see above) by

filling the growth compartment with approximately 5 mL of serum-containing

medium containing a suspension of approximately 2X104 cells/mL (sub-

confluent cell density). The reservoir was filled with serum-free basal medium

containing no proteins. Cells from the same inoculum were plated in 25 cm2

standard tissue-culture-grade polystyrene (TCPS) dishes (Corning Life Science)

which served as the controls. Cultures were maintained in a water-jacketed 5%

CO2 incubator (Model 3110, ThermoForma) held at 37 oC. After cells had

69

reached confluence (3-5 d), growth medium was replaced with differentiation

medium and cells were permitted to grow in control plates and in the bioreactor

without subculture. Medium was replaced in TCPS controls every 3-5 d as

dictated by color of the methyl red indicator, with a hint of yellow (acid

indicator) leading to full exchange. Basal media within the bioreactor reservoir

(but not growth space) was refreshed every 15-30 d, again depending on the

perceived color of the pH indicator, with a hint of yellow leading to full

exchange of only the reservoir contents. For example, 120 d MC3T3-E1 cultures

in the bioreactor were sustained with four replacements of basal media within

the reservoir every 30 d.

Cell Attachment and Proliferation Rates: Short-term ( 0 3t< < hr) cell

attachment assays were performed as described previously (16, 17). Briefly, cells

were plated onto 15 identically-prepared plates at 2X104 cells/cm2 in whole

medium. TCPS culture dishes were the comparison controls. Dishes with the

substratum used in the bioreactor were prepared by adhering Surlyn film into

the bottom of TCPS culture dishes using double-sided adhesive tape. Cells were

allowed to adhere from the sessile medium while incubated at 37 oC in a CO2

incubator. At various time intervals, a single plate was selected for destructive

analysis (every 5 min from 0 30t< < min, every 10 min from 30 60t< < min,

every 15 min for 60 120t< < min, and every 30 min for 120 180t< < min) by

70

discarding suspension medium and rinsing substrata 3X with PBS. Attached

cells were released with trypsin and counted by hemacytometry. Surfaces were

analyzed in triplicate. Proliferation assays were performed in the same basic

way, except that after 3h attachment time, substrata were rinsed with PBS to

remove non-adherent cells and refreshed with growth medium. Remaining cells

were allowed to proliferate for 6, 12, 24, 48, and 72 h before destructive analysis.

Long term ( 15≥ d) proliferation was monitored by SYBR green nuclear-staining.

Duplicate TCPS and substrata removed from the bioreactor were fixed with 4%

paraformaldehyde in PBS, permeabilized with 0.1% Triton X-100 in PBS, stained

with 0.02% SYBR Green in PBS (Molecular Probes), and visualized using an

Olympus BX-60 epi-fluorescent microscope. Average nuclei number was

determined by counting 15 representative spots on each substrate using image

analysis (ImagePro Software, MediaCybernetics Inc).

Alkaline phosphatase activity: Alkaline phosphatase (ALP) activity was

quantified at the end of each culture period using a chromogenic assay involving

conversion of p-nitrophenyl phosphate to p-nitrophenol as described in (18).

Briefly, cells washed in PBS were lysed by rinsing with PBS and 0.1% (w/v)

Triton X-100 (Sigma) in PBS. Lysates were subjected to two freeze-thaw cycles,

after which ALP reaction buffer (1:1 mixture of 0.75 M 2-amino-2-methyl-1-

propanol and 2 mg/mL p-nitrophenol phosphate) was added and incubated at

71

37°C for 15 min. This reaction mixture was stopped with 0.05 N NaOH, and the

absorption was measured at 410 nm with a Beckman Spectrophotometer (DU-70

series). A calibration curve was prepared by serial dilution of p-nitrophenol

standard solution. A portion of the reaction mixture was used to measure total

protein concentration using Bio-Rad protein assay kit. ALP activity was

normalized to total cell protein. Experiments were conducted in triplicate for

duplicate cultures. Results are reported as mean±standard deviation (1σ ).

Apoptosis: Cell apoptosis was analyzed using the enzyme terminal

deoxynucleotidyl transferase assay (TUNEL assay, Promega) to catalytically

incorporate fluorescein-12-dUTP at 3’-OH DNA ends, so that apoptotic cells

fluoresce green. All cells were stained red with SYTOX® Orange (Molecular

Probes) so that apoptotic cells could differentiated from normal by fluorescent

microscopy. At the end of each growth period, duplicate cultures were rinsed

twice with PBS, fixed with 4% paraformaldehyde, and permeablized with 0.2%

Triton X-100. Cells were then equilibrated in buffer for 5-10 min at room

temperature, followed by incubation with TdT buffer (that included equilibration

buffer, nucleotide mix, and TdT enzyme) in a humidified chamber for 60 min at

37oC. The reaction was terminated using 2X “SSC washes” for 15 min at room

temperature according to the TUNEL assay protocol. A positive and negative

control was prepared in parallel. After removing unincorporated fluorescein-12-

72

dUTP, cells were stained with SYTOX® Orange for 10 min at room temperature.

After staining, substrata were mounted with ProLong® Gold and photographed

with a Laser Scanning Confocal microscope (Olympus Fluoview300). Images

were analyzed with ImagePro software (MediaCybernetics Inc.). The percentage

of apoptotic cells in the whole cell population was calculated from 10-20 random

spots on each substrate.

Insoluble Extracellular Matrix (iECM) Protein assay: Cells were

incubated in a bioreactor or in TCPS for 15 and 30 d. After medium was

removed, cell layers were washed twice with PBS and extracted in 4 M guanidine

hydrochloride, 50mM Tris-HCl, 100mM 6-aminocaproic acid, 5 mM benzamidine

hydrochloride, and 1mM phenylmethylsulphonyl fluoride, pH 7.4, at 4oC with

constant rocking. Cells were separated from the extract by centrifugation (2000g

X 5 min; the extract was routinely checked for cell debris by microscopy with

trypan-blue staining to enhance visualization of cell debris). Protein

concentration in the supernatant was quantified using Bio-Rad protein assay kit,

and normalized with cell number for each growth condition.

Hematoxylin and Eosin Staining: Tissue excised from the bioreactor was

fixed in 2.5% glutaraldehyde. Dehydration and paraffin embedding were carried

using an automated tissue processor (Citadel 2000, Thermo Electron). 5μm

73

sections were cut using a microtome (Model 2040, Leica Micro System) and

mounted on glass coverslips. Hematoxylin and Eosin staining was carried

according to a standard protocol using automated staining equipment (Shandon

Gemini, Thermo Electron) and the stained sections were observed under an

optical light microscope (Olympus BX50, HiTech).

Scanning and Transmission Electron Microscopy (SEM and TEM):

Cultures were washed in buffer (0.1M, pH 7.2 sodium cacodylate) and fixed

overnight with 2.5% glutaraldehyde in cacodylate buffer at 4oC followed by

staining with 1% osmium tetroxide in cacodylate buffer for an hour at room

temperature. Fixed cells were dehydrated using graded series of ethanol

concentrations, dried in a critical point dryer (BALTEC SCD030, Techno Trade

Inc.) using dry CO2 , mounted on to an aluminum stub, and sputter coated with

10 nm of gold/palladium in an automated sputter coater (BALTEC SCD030,

Techno Trade Inc.). Cells were examined under an SEM (JSM 5400, JEOL,

Peabody, MA) at an accelerating voltage of 20 kV. Energy dispersive X-Ray

analysis was carried out using image analysis software (IMIX-PC v.10, Princeton

Gamma Tech Inc.). For TEM, primary fixation with 2.5% glutaraldehyde and

secondary fixation with 1% osmium tetroxide were followed by en bloc staining

with 2% aqueous uranyl acetate for an hour. Dehydration was carried out using a

graded series of ethanol concentrations followed by impregnation and

74

embedding in Spurr’s resin. Ultra-thin sections were cut with a diamond knife

(Diatome Ultra 45) on a microtome (Ultracut UCT, Leica), placed on uncoated

copper grids and stained with both 0. 2% aqueous uranyl acetate and 0.2% lead

citrate. The cross-sections were examined using TEM (JEM 1200 EXII, JEOL) and

images were collected using an attached high-resolution camera (Tietz F224,

Gauting).

3. Results and Discussion

Tables 1 and 2 compile semi-quantitative cell-growth data over 30 d for

hFOB 1.19 and 120 d for MC3T3-E1 cells, respectively, in the bioreactor

compared to that obtained in conventional tissue-culture grade polystyrene

(TCPS). Neither cells grown on TCPS nor those grown in the bioreactor were

subcultured over the entire culture interval (see Methods and Materials). For

both hFOB 1.19 and MC3T3-E1, between 1-2 cell layers formed in TCPS

compared to 4-6 layers in the bioreactor over 30 d culture interval (the exact

number of layers varied with position in the bioreactor and how a cell layer was

defined). Evidence of a mature osteoblastic phenotype was indicated by alkaline

phosphatase activity which was positive for both cell lines in both culture

devices. Mineralization was especially evident by von Kossa and formation of

75

nodules in the MC3T3-E1 case, but hFOB 1.19 did not mineralize as well in either

culture device. Based on the data summarized in Tables 1 and 2, it is evident that

performance of hFOB 1.19 and MC3T3-E1 in the bioreactor was similar that of

cells grown on TCPS in the first 15 d of culture. However, cell performance in

the bioreactor clearly exceeded that for TCPS over a 15-30 d culture interval, with

double the number of cell layers and increased alkaline phosphatase activity and

alizarin-red staining. TCPS controls were not carried for more than 30 d because

it was both visually and quantitatively evident that these cultures were failing

(see further below).

Fig. 2 compares attachment and growth dynamics of hFOB 1.19 in the

bioreactor and TCPS over 0 720t< < hr culture period. In the short-term

( 0 3t< < hr), there was ~ 2X attachment preference for TCPS over the polymer-

film substrata used in the bioreactor (see inset expanding time axis and

expressing % inoculum on a linear axis; see also ref. (17) for more discussion of

polymer-film cytocompatibility). However, proliferation into confluent

monolayer (on TCPS) from this sub-confluent cell density (3 72t< < hr) and post-

monolayer growth into multilayers (72 720t< < hr) was nearly the same in TCPS

and the bioreactor. In fact, actual growth rates k within 3 72t< < hr were

statistically identical at 95% confidence (TCPS = 0.23 ± 0.01 10-2 hr-1; bioreactor =

0.29 ± 0.03 10-2 hr-1). Data within the 72 720t< < hr interval were too sparse for

76

statistical comparison, but it is interesting that the total-cell number (at constant

surface area = 25 cm2) continued to significantly increase within both the

bioreactor and TCPS at approximately the same rate, achieving ~5X increase in

total cell number over this time span. The data show that the initial two-fold

cell-attachment preference for TCPS persisted through the exponential-growth

(3 72t< < hr) and post-confluence expansion ( 72 720t< < hr) phases. Further

interpretation in terms of the data in Table 1, this persistent-attachment

preference implies that more cells are packed into fewer cell layers on TCPS than

within the bioreactor, consistent with the general observation that cells on TCPS

appear thinner than in the bioreactor (compare, for example, Panel A, C Fig. 3);

although a focused study would be required to be definitive in this regard.

Despite of the initial advantage on TCPS, hFOB 1.19 began to appear less

robust than in the bioreactor within 720 hr (30 d), as evidenced by TEM images

showing numerous apoptotic bodies, cytoplasmic vesiculation, and chromatin

margination (Fig. 3) and a distinct loss of insoluble iECM (Fig. 4). TUNEL assay

results (summarized in Fig. 5) confirmed that at 30 d, apoptosis was indeed ~2X

greater in TCPS than in the bioreactor (37-47% in TCPS compared to 16-23% in

the bioreactor), although 30 d apoptosis levels were roughly 3X 15 d levels in

both devices. Examination of hFOB 1.19 tissue recovered from the bioreactor

after 15 d culture by cross-sectional TEM (Fig. 6) revealed 3-4 layers of cells

77

(Panels A and B) with many cytoplasmic protrusions (arrowheads) and healthy-

appearing nuclei showing no evidence of apoptosis. The general impression

derived from an examination of many such TEM cross-sections is that cell layers

closest to the substratum (presumably older cells) were flatter and more dish-

shaped compared to cells within upper layers (presumably younger cells) that

had a more cuboidal shape. No effort was made to quantify this visual

interpretation. Arrow annotations on Fig. 6C point to what appear to be

osteocytic-like processes between two cells and Fig. 6D shows a close contact

between cells. Exocytosis of vesicles is evident in Fig. 6D.

Results obtained with MC3T3-E1 osteoblasts in TCPS and the bioreactor

were similar to those seen for hFOB 1.19 with the notable exception that

mineralization was much more striking with MC3T3-E1 cells. One of many large

nodules (Panel A, SEM) that were found on and within MC3T3-E1 tissue after 70

d culture (see also Fig. 8 Panel A) was analyzed (Fig. 7). Panel B is a high-

resolution cross-sectional view through a nodule similar to that shown in Panel

A. Nodules such as these proved positive for calcium and phosphorous by

energy-dispersive x-ray analysis (Panel C, SEM/EDAX of a nodule taken from a

30 d bioreactor). These nodules were enmeshed in a fibrous, apparently

collagenous, network that was very evident in SEM preparations such as that

shown in Fig. 8A. Histological preparations of MC3T3-E1 tissue recovered from

78

70 d bioreactors (Panel B) confirm substantial matrix development with

osteoblasts oriented along sheets of matrix. TEM of von Kossa-stained 70 d

tissue shown in Panel C is consistent with collagen fibers running in (IP

annotation) and along (OP annotation) the plane with interspersed mineral

nodules (arrows). We estimated that tissue taken from a 120 d bioreactor

was 150 mμ thick. After 120 d in culture, there was no indication that MC3T3-E1

tissue grown in the bioreactor was unstable and the absolute term-of-culture

within the bioreactor is regarded as indefinite until experimentation finds an

endpoint at which significant instability can be detected.

79

4. Conclusions

Evidence summarized in Fig. 2 and supporting micrographs (Figs. 3-8)

show that long-term maintenance of hFOB 1.19 (ATCC CRL-11372) and MC3T3-

E1 (ATCC CRL-2593) cells without subculturing leads to formation of cell

multilayers within a thick, mineralized matrix that cannot be realized when cells

are regularly passaged at confluence. Achievement and maintenance of such a

3D ‘tissue’ that promotes cell-cell contact has been shown to be critical to the

development of osteogenic cells into a fully-differentiated osteoblast phenotype

(19-22). We demonstrated that such tissue can be grown and maintained long-

term (> 120 d) in a bioreactor based on the principle of simultaneous-growth-and

dialysis with improved efficiency over conventional tissue-culture labware

(TCPS). In fact, tissue maintained 30 d in TCPS with medium replacement every

3 d became unstable, with 2X apoptosis rates and loss of insoluble extracellular

matrix (iECM) compared to cells grown in the bioreactor, even though results

obtained in TCPS and the bioreactor were similar in the first 15 d of culture.

Furthermore, morphology of cells grown in the bioreactor suggested

development of an osteoid structure with older cells taking on a flattened

osteocytic-like phenotype with inter-cellular connections. The matrix

surrounding up to 6 layers of cells was collagenous and mineralized. This

comparison of the bioreactor to TCPS suggests that the eventual instability

observed in TCPS was primarily due to the accumulated damage arising from

80

periodic medium replacement that removes not only waste products but also

‘luxury macromolecules’ secreted by the cells. Hence, the standard cell-culture

method tends to propagate, rather than mitigate, the effects of culture shock. By

contrast, the simultaneous-growth-and-dialysis culture method maintained an

extraordinarily-stable pericellular environment. With neither subculturing steps

to push cells closer-and-closer to the Hayflick limit (4) nor the environmental

stress of an ever-changing pericellular milieu, tissue grown with the bioreactor is

apparently stable for very long periods of time measured at least in months.

Thus the compartmentalized bioreactor shows promise as a tool for the in vitro

study of osteogenesis and osteopathology.

81

Acknowledgements

This work was supported, in part, by a grant from the Pennsylvania Department of

Health. Pennsylvania Department of Health specifically disclaims responsibility for any

analyses, interpretations or conclusions. This work was also supported by The Pennsylvania

State Tobacco Settlement Formula Fund and NIH AG13087-10. This work was also

supported, in part, by grants from the US Army Medical and Materiel Command Breast

Cancer Program (W81XWH-06-1-0432), and the Susan G. Komen Breast Cancer

Foundation (BCTR 0601044. Authors appreciate additional support from the Huck

Institutes of Life Sciences, Materials Research Institute, and Departments of Bioengineering

and Materials Science and Engineering of the Pennsylvania State University.

82

References

1. Sherr, C., and DePinho, R. Cellular senescence: mitotic clock or

culture shock? Cell 102, 407, 2000.

2. Anselme, K. Osteoblast Adhesion on Biomaterials. Biomaterials 21,

667, 2000.

3. Schnaper, H.W., and Kleinman, H.K. Regulation of cell function by

extracellular matrix. Pediatr Nephrol 7, 96, 1993.

4. Shay, J.W., and Wright, W.E. Hayflick, his limit, and Cellular

Ageing. Nature 1, 72, 2000.

5. Rubin, H. Telomerase and Cellular Lifespan: Ending the Debate?

Nature Biotechnology 16, 396, 1998.

6. Rose, G.G. Cytopathophysiology of Tissue Cultures Growing

Under Cellophane Membranes. In: Richter, G.W., and Epstein,

M.A., eds. Int. Rev. Exp. Pathology. New York: Academic Press,

1966, pp. 111-178.

7. Vogler, E.A. Compartmentalized Cell-Culture Device and Method.

USA: DuPont de Nemours Inc., 1988.

83

8. Vogler, E.A. A Compartmentalized Device for the Culture of

Animal Cells. J. Biomaterials, Artificial Cells, and Artificial Organs

17, 597, 1989.

9. Donahue, H.J., Siedlecki, C.A., and Vogler, E. Osteoblastic and

Osteocytic Biology and Bone Tissue Engineering. In: Hollinger, J.O.,

Einhorn, T.A., Doll, B., and Sfeir, C., eds. Bone Tissue Engineering.

Boca Raton: CRC Press, 2004, pp. 44-54.

10. Lim, J.Y., Liu, X., Vogler, E.A., and Donahue, H.J. Systematic

Variation in Osteoblast Adhesion and Phenotype with Substratum

Surface Characteristics. J. Biomed. Mater. Res. 68A, 504, 2004.

11. Lim, J.Y., Taylor, A.F., Li, Z., Vogler, E.A., and Donahue, H.J.

Integrin Expression and Osteopontin Regulation in Human Fetal

Osteoblastic Cells Mediated by Substratum Surface Characteristics.

Tissue Engineering 11, 19, 2005.

12. Vogler, E.A., and Bussian, R.W. Short-Term Cell-Attachment Rates:

A Surface Sensitive Test of Cell-Substrate Compatibility. J. Biomed.

Mat. Res. 21, 1197, 1987.

13. Vogler, E.A. Thermodynamics of Short-Term Cell Adhesion in vitro.

Biophysical J. 53, 759, 1988.

14. Vogler, E.A. A Thermodynamic Model of Short-Term Cell

Adhesion In vitro. Colloids and Surfaces 42, 233, 1989.

84

15. Harris, S.A., Enger, R.J., Riggs, B.L., and spelsberg, T.C.

Development and Characterization of a Conditionally

Immortalized Human Fetal Osteoblastic Cell Line. J. Bone and

Mineral Res. 10, 178, 1995.

16. Lim, J.Y., Liu, X., Vogler, E.A., and Donahue, H.J. Systematic

variation in osteoblast adhesion and phenotype with substratum

surface characteristics. J. Biomed. Mat. Res. 68A, 504, 2004.

17. Vogler, E.A., and Bussian, R.W. Short-term cell-attachment rates: a

surface-sensitive test of cell-substrate compatibility. J. Biomed. Mat.

Res. 21, 1197, 1987.

18. Lim, J.Y., Taylor, A.F., Li, Z., Vogler, E.A., and Donahue, H.J.

Integrin expression and osteopontin regulation in human fetal

osteoblastic cells mediated by substratum surface characteristics.

Tissue Engineering 11, 19, 2005.

19. Freed, L.E., and Vunjak-Novakovic, G. Culture of organized cell

communities. Advanced Drug Delivery Reviews 33, 15, 1998.

20. Gurdon, J., Lemaire, P., and Kato, K. Community effects and

related phenomena in development. Cell 75, 831, 1993.

21. Ferrera, D., Poggi, S., Biassoni, C., Dickson, G.R., Astigiano, S.,

Barbieri, O., Favre, A., Franzi, A.T., Strangio, A., Federici, A., and

Manduca, P. Three-dimensional cultures of normal human

85

osteoblasts: proliferation and differentiation potential in vitro and

upon ectopic implantation in nude mice. Bone 30, 718, 2002.

22. Gerber, I., and Gwynn, I. Differentiation of rat osteoblast-like cells

in monolayer and micromass cultures. Eur Cell Mater 3, 19, 2002.

86

List of Table Legends:

Table 1. hFOB 1.19 Cells in Conventional Culture Compared to


Table 2. MC3T3-E1 Cells in Conventional Culture Compared to


87

Table 1. hFOB 1.19 Cells in Conventional Culture Compared to Compartmentalized Bioreactor

Conventional Culture Compartmentalized

Bioreactor

Days in Culture 15 30 15 30

Number of cell layers 1-2 1-2 3-4 4-6

Alkaline Phosphatasea, b 4.39 ± 0.26 4.47 ± 0.38 28.89 ± 1.99 5.79 ± 0.40

Alizarin redc (μmol) 2.89 ± 0.04 3.03 ± 0.02 2.97 ± 0.18 5.17 ± 0.14

von Kossac - - - -

Mineralized nodules - - - -

+ = observed; - = not observed; ameasures of osteoblast maturity; b(nmol/mg

protein/min); cextent of mineralization.

88

Table 2. MC3T3-E1 Cells in Conventional Culture Compared to Compartmentalized Bioreactor

Conventional Culture Compartmentalized Bioreactor

Days in Culture 15 30 15 30 60 90 120

Number of cell layers 1-2 1-2 1-2 4-6 4-6 4-6 4-6

Alkaline Phosphatasea, b 1.14±0.01 1.27±0.17 2.94±0.48 3.36±0.46 ND ND ND

Alizarin redc (μmol) 4.59±0.12 5.07±0.09 5.28±0.05 6.88±0.55 ND ND ND

von Kossac + + + ++ ++ ++ ++

Mineralized nodules - +/- + ++ ++ ++ ++

+ = observed; - = not observed; ameasures of osteoblast maturity; b(nmol/mg protein/min); cextent of

mineralization; ND = not determined

89

List of Figure Legends:

Figure 1: Compartmentalized bioreactor design. Panel (a) is a cross-sectional

diagram through the device showing separation of the cell-growth space (A)

from the basal-medium reservoir (B) by a dialysis membrane (C). Cells are

grown on gas-permeable but liquid-impermeable film (E). The device is

ventilated through film (D), which is the same material as (E) as described herein

but can be different. The whole device is brought together in a liquid-tight

fashion using screws shown in the laboratory implementation Panel (b) and

Panel (c) which is an exploded-view identifying separate components. Liquid-

access is through leur-taper ports (J, K) which mate to standard pipettes. See

Methods and Materials for theory of operation.

Figure 2: Growth dynamics of hFOB 1.19 on tissue-culture-grade polystyrene

controls (TCPS, filled triangles) compared to the compartmentalized bioreactor

(open squares; note logarithmic ordinate). Inset expands short-term attachment

rates using a linear ordinate. Note that the initial two-fold cell-attachment

preference for TCPS persists through the exponential-growth (3 72t< < hr) and

post-confluence expansion ( 72 720t< < hr) phases.

90

Figure 3: Comparison of hFOB 1.19 on TCPS (Panels A, B) and in the

compartmentalized bioreactor (Panels C, D) by cross-sectional TEM. Notice that

cell layers are thicker in the bioreactor than on TCPS (compare Panels A, B to C,

D). and formation of apoptotic bodies (arrows) after 30 d culture in TCPS.

Figure 4: Comparison of insoluble extracellular matrix (iECM) production by

hFOB 1.19 on TCPS (dark bar) and in the bioreactor (light bar). Notice that iECM

significantly decreased after 30 d on TCPS but remained constant in the

bioreactor. Bar values are mean of duplicate samples measured in triplicate with

standard deviation represented by error bars.

Figure 5: Representative confocoal image of apoptotic bodies (green stain) in

TCPS and in the bioreactor selected from 8-10 similar in-depth image planes

probing hFOB 1.19-derived tissue. All cells were stained red (Sytox Orange,

scale bar = 50 μm). Percent apoptotic cells quoted in each panel are the range of

values observed within the image planes.

Figure 6: Ultramorphology of hFOB 1.19–derived tissue (15 d) recovered from

the bioreactor by cross-sectional TEM. Arrows point to cell protrusions that

91

occasionally connect two cells, as shown in Panel C. Annotations: jNC = gap

junction; MV = matrix vesicles; N = nucleus; rER = rough endoplasmic

reticulum.

Figure 7: Analysis of mineral nodules taken from MC3T3-E1 cultured in a

bioreactor. Panel A is an SEM of a large nodule taken from a 70 d bioreactor.

Panel B is a high-magnification, cross-sectional TEM of a similar nodule. Panel C

is an x-ray spectrum obtained by SEM/EDAX of a nodule taken from a 30 d

bioreactor confirming Ca and P as major constituents.

Figure 8: Microscopic examination of matrix derived from MC3T3-E1 cultured

in a bioreactor for 70 d. Panel A is an SEM of the outer layer showing a highly

fibrous network with bone nodules. Panel B is an optical micrograph of a

histologic workup showing layers of collagen with imbedded osteoblasts. Panel

C is a cross-sectional TEM showing mineralized fibers running in (IP annotation)

and out (OP annotation) of the plane.

92

50mm

(E)(E)

(C)(C)

(D)(D)

(I)

(H)(H)

(F)(F)

(G)(J)(J)

(K)(K)

a c

b

Figure 1

93

Time (hr)

Log

(% In

ocul

um)

0

1

2

3

4

0 1 3 360 720

Confluent monolayer

72

Time (min)0 50 100 150 200

% In

ocul

um0

20

40

60

80

Figure 2

94

A B

DC

substrate

substrate

substrate

substrate

Figure 3

95

iEC

M p

rodu

ctio

n (n

g/ce

ll)

0.0

0.1

0.2

0.3

0.4

0.5

0.6TCPSBioR

15 days 30 days

Growth period

* * * *

Figure 4

96

Figure 5

97

Substrate A

n

n

NN

ASubstrateSubstrate A

rER

BSubstrate

N

N

MV

N

C

rER

rER

MV

DMV

jNC

N

NN

ECM

ECM

ECM

Figure 6

98

A 5μm

500nm

B

C

Figure 7

99

5μm

cfl

cfccfl

50μm

Collagen

OB

C

Bone Nodule

Bone Nodule

A B

C

Figure 8

100

Chapter 4

System in Crisis- Colonization of Bone Tissue by Metastatic Breast Cancer In Vitro

Abstract

Metastasis is the ultimate cause of mortality in over 90% of cancer patients. Metastasis is a

multi-step process that starts with the shedding of neoplastic cells from the primary tumor,

entry into circulation, exit into a distant organ and ends with colonization of the target

organ. The final step in the metastatic cascade, colonization of distant tissue, is a critical

step that determines the eventual disease outcome and is therefore a valuable target for

therapeutics. Dynamic interaction between tumor cells and the host microenvironment

determines the survival, growth and the eventual development of clinically relevant tumors

in the target organ. Lack of model systems with sufficient biological complexity to serve as

relevant surrogates for the host microenvironment is a significant impediment to

understanding the cellular and molecular basis for colonization and discovery of therapeutic

interventions. We show herein that a three-dimensional, mineralizing, bone-like tissue

grown from isolated mouse calvarial pre-osteoblasts (MC3T3-E1) over months-long culture

in a specialized bioreactor is an effective surrogate for studying bone-tissue colonization by

metastatic breast cancer. Bioreactor derived bone-like tissue challenged with metastatic

breast cancer cells (MDA-MB-231) known to invade the skeleton captured important

hallmarks of metastatic colonization. In situ confocal microscopy revealed sequential steps

of breast cancer cell adhesion, penetration, and apparent degradation of the 3D bone-like

tissue over co-culture intervals ranging from 3 to 10 days. Bioreactor enabled direct

observation of interactions of breast cancer cells with engineered bone-like tissue simulating

101

an important step in the progression of the disease that may be a target for therapeutic

intervention.

Introduction Tissue engineering has focused considerable interest in the growth of three-dimensional

(3D), multiple-cell-layer tissues from isolated cells of purposely-selected type(s). It has been

well recognized that these engineered tissues can be used to great advantage in drug

discovery, development of therapeutic strategies, and toxicology1 by bridging the gap

between conventional monolayer (2D) cell culture and whole animals. These in vitro tissue

models confer considerable control over important experimental variables yet provide

sufficient biological complexity that outcomes are a reasonable predictor of the whole-

organism physiological response to experimental variables.2 A further advance in the use of

engineered tissue for drug discovery is to create an in vitro “system in crisis” that simulates an

in vivo disease state. Various candidate drugs and/or therapeutic strategies could then be

screened to find those that most effectively resolve the purposely-imposed crisis. Tissue

surrogates are useful in this pursuit because systemic aspects of disease can be broken down

into physiological subsystems to be studied separately in vitro. Efficiency of drug discovery

can be greatly improved with a clear understanding of the cellular and molecular basis of

disease because opportune targets of therapeutic intervention can be clearly identified.

We have focused on metastatic breast cancer colonization of bone as a proof-of-principle

problem with both stringent analytical requirements and great importance in human

healthcare. Colonization is the final step in the multi-step process of metastasis that starts

with the shedding of neoplastic cells from the primary tumor, entry of these cells into

circulation, arrest in a distant organ and finally growth and tumor formation in the target

organ.3 Cancers that colonize bone are particularly pernicious because, once bone

102

colonization occurs, the cure rate is almost zero.4-6 Cancers in bone progress with significant

morbidity related to massive bone loss, bone pain, hypercalcemia, pathological fractures,

and spinal cord compression.7 Of the major cancers that colonize bone (breast, prostate, and

multiple myeloma of hematopoietic origin), breast cancer has perhaps attracted the most

attention because of its devastating effects on women - approximately 25% of breast cancers

metastasize and the target organ is bone in 46% of the primary cases. Cancer can reoccur

after successful treatment of primary tumor and years of remission – in patients with first

relapse the frequency of metastatic bone disease is much higher at 70%.8 Critical rate

limiting steps in cancer colonization of bone include survival of tumor cells in the bone

microenvironment, proliferation and organization into overt tumors and angiogenesis or

recruitment of blood vessels necessary to sustain the tumor mass.9 Cancer cells can also

remain dormant and persist as isolated cells for years or even decades ready to be activated

into fully developed tumors under the right set of conditions.9 Cancer dormancy is a poorly

understood stage in disease progression characterized by the presence of cancer cells in bone

that have not progressed to form clinically detectable tumors. Difficulty in accessing the

bone marrow cavity probably accounts for the fact that bone tumors are usually detected

after cancer colonization has reached a very advanced stage, further complicating effective

treatment strategies.

Bone degradation in metastatic breast cancer is thought to be primarily mediated by bone-

resorbing cells (osteoclasts)10-13 that are activated by cancer cells. Excessive resorption results

in the formation of osteolytic lesions that reduce the strength and integrity of bone.

Consequently, drugs of the bisphosphonate family are currently used to inhibit osteoclasts.

Bisphosphonate treatment slows lesion formation but does not result in the restoration of

bone. Furthermore, it is not a cure in that it does not eliminate the cancer cells.14

Osteoblasts, responsible for bone formation, also play a critical role in cancer colonization.

103

Osteoblasts not only regulate osteoclast activity but also interact with cancer cells.11,15,16 As a

consequence, we have focused our attention on understanding the cellular and molecular

basis of cancer-cell colonization of bone and the related interactions with osteoblasts.17-22

These interactions include suppression of osteoblastic differentiation, increased osteoblast

apoptosis, an immune-like osteoblastic stress response, and the up-regulated secretion of

cytokines observed in conventional 2D cell culture. These findings imply a significant

osteoblastic pathology associated with cancers in bone, in addition to (and perhaps coupled

with) the known activation of osteoclasts. Understanding the effect of breast cancer on

osteoblast function can lead to therapeutic strategies that not only inhibit development of

osteolytic lesions but also restore or renew the normal functional activity of bone formation.

In spite of its clinical importance, the biology of cancer colonization is not completely

understood. Lack of experimental models that simulate cellular interactions involved in

breast cancer colonization of bone is a significant impediment to progress in the field.

Whole animal models often obscure details of the colonization process, especially when the

target is a difficult-to-access tissue such as bone. Models based on excised bone are not only

technically problematic but also difficult to interface with modern microscopic methods of

investigation. Further, most experimental models access only the end stage of the disease

defined by presence of clinically relevant tumors – critical early stages that determine the

eventual disease outcome remain largely “hidden” and inaccessible. In principle, a sub-set

of the metastatic process can be studied in vitro if the model system under consideration

retains sufficient biological complexity to be a reasonable surrogate for host tissue. Three-

dimensional (3D) tissue models have become a focus of recent investigation for this reason.

We have shown recently that a compartmentalized bioreactor based on the principle of

simultaneous–growth-and dialysis can promote the development of isolated osteoblasts into

mature 3D tissue over extended culture intervals up to a year.23 Resulting bone like

104

osteoblast tissue recapitulated progression of pre-osteoblasts to osteocyte-like cells observed

in normal bone, including production of visually- apparent (macroscopic) bone. Challenging

the bioreactor derived osteoblast tissue at various stages of development with metastatic

breast cancer cells known to invade the skeleton revealed the sequential steps of cancer cell-

tissue adhesion, penetration, and ultimate degradation of the host tissue, modeling a critical

stage in the progression of the disease that may be a target for therapeutic intervention.

Materials and Methods

Cells and Cell Culture: Murine calvaria pre-osteoblasts (MC3T3-E1, ATCC CRL-2593)

were a gift from Dr. Norman Karen at the University of Delaware. Human metastatic breast

cancer cell line (MDA-MB-231, ATCC-HTB 26) labeled with green fluorescent protein

(GFP) were a gift from Dr. Danny Welch from the University of Alabama at Birmingham.

MC3T3-E1 were inoculated into the bioreactors at a sub-confluent density (104 cells/cm2)

and cultured in alpha minimum-essential medium (α-MEM) (10-022-CV, Mediatech Inc,

Manassas VA) supplemented with 10% fetal bovine serum (FBS) and 100U/ml penicillin

and 100μg/ml streptomycin. Once the cells reached confluence, the medium was replaced

with differentiation medium containing the additional ingredients 50μg/mL ascorbic acid

and 10mM β-glycerophosphate. MC3T3-E1 cultures were maintained in the bioreactor

without sub-culture for extended intervals up to10 months using culture conditions and

protocols described in chapter 3. MDA-MB-231 cells were cultured in standard tissue

culture dishes in Dulbecco’s modified Eagle medium (DMEM) containing 5% fetal bovine

serum and 100U/ml penicillin and 100μg/ml streptomycin.

105

Bioreactor Co-Cultures: MC3T3-E1 osteoblast tissue in the bioreactor was stained with

Cell Tracker Orange ™ (Invitrogen Corp., Carlsbad, CA; prepared according to vendor

instructions) by replacing original culture medium with fresh medium containing 2.5μM cell

tracker orange for 45 minutes at 37oC, followed by 30 minutes incubation in dye-free culture

medium. Co-cultures were initiated by inoculating a suspension of GFP-labeled MDA-MB-

231 in osteoblast differentiation medium into the bioreactor on top on the MC3T3-E1 tissue

at an estimated cancer cell/ osteoblast ratio of 1/10. All operations were performed under

sterile conditions in a level II biosafety cabinet. Bioreactor co-cultures were maintained for

periods ranging from 3-10 days under standard culture conditions (5% CO2 and 95% air) in a

humidified incubator at 37oC. In situ confocal microscopy was used to follow the interaction

of breast cancer cells with 3D osteoblast tissue. At the end of the culture interval, the

bioreactor was dismantled and substratum film with adherent tissue was cut into pieces for

various assays. Bioreactors cultured with MC3T3-E1 for equivalent periods of time under

exactly the same conditions with similar medium changes but without the addition of cancer

cells were used as controls.

Confocal Microscopy: In situ laser-scanning confocal microscopy of co-cultures in the

bioreactor was performed using Olympus FV-300 laser scanning microscope (Olympus

America Inc, Melville, NY). Sections were observed with a 40X Olympus UPlanF1

objective with a numerical aperture of 0.85. Cell Tracker Orange ™ was excited using a 543

nm line from a helium-neon laser and collected through a 565 nm long-pass filter. GFP was

excited using a 488 nm argon laser and collected through 510 nm long-pass and 530nm

short-pass filters. A 570nm dichroic long-pass filter was used to split the emission. Serial

optical sections were taken at 1μm intervals throughout the tissue. Confocal images were

processed using Fluoview 300, Version 4.3b, Olympus America Inc. 3D optical

106

reconstructions of 2-D serial sections were obtained using AutoQuant AutoDeblur and

AutoVisualize software (Version 9.3, Media Cybernetics, Inc., Silver Spring, MD).

Transmission Electron Microscopy: Tissue excised from the bioreactor was fixed overnight

with 2.5% glutaraldehyde in 0.1M sodium cacodylate buffer at 4oC. Secondary fixation with

1% osmium tetroxide in 0.1M sodium cacodylate was carried out for an hour was followed

by enbloc staining with 2% aqueus uranyl acetate. Buffer washes in 0.1M sodium cacodylate

(3 washes, each 5 minute duration) were carried out after each fixation step and after the

enbloc staining step. Tissue was then dehydrated using a graded series of ethanol

concentrations ( 50 %, 70%, 85%, 95%, 100%) and finally immersed in 100% acetone.

Infiltration with Spurrs resin was carried out gradually over 3 infiltration steps with each step

carried out overnight at room temperature. Samples were embedded in Spurr’s resin an

cured overnight at 55oC. The cured block of embedded tissue was thin-sectioned with a

diamond knife (Ultra 45, Diatome, Hatfield PA) on a microtome (Ultracut UCT, Leica,

Bannockburn, IL). Thin sections of tissue were placed on uncoated copper grids and stained

with both 0.2% aqueous uranyl acetate and 0.2% lead citrate. Cross-sections of tissue were

examined using TEM (JEM 1200 EXII, JEOL, Peabody MA) and images were collected

using an attached high-resolution camera (Tietz F224, Tietz Video and Image Processing

Systems GmbH, Gauting Germany).

Scanning Electron Microscopy: Cultures were washed in buffer (0.1M, pH 7.2 sodium

cacodylate) and fixed overnight with 2.5% glutaraldehyde in cacodylate buffer at 4oC

followed by staining with 1% osmium tetroxide in cacodylate buffer for an hour at room

temperature. Fixed samples were dehydrated using graded series of ethanol concentrations,

107

dried in a critical point dryer (BALTEC SCD030, Techno Trade Inc.) using dry CO2,

mounted on to an aluminum stub, and sputter coated with 10 nm of gold/palladium in an

automated sputter coater (BALTEC SCD030, Techno Trade Inc.). Samples were examined

under an SEM (JSM 5400, JEOL, Peabody, MA) at an accelerating voltage of 20 kV.

Hematoxylin and Eosin Staining: Tissue excised from the bioreactor was fixed in 2.5%

glutaraldehyde. Dehydration and paraffin embedding were carried using an automated

tissue processor (Citadel 2000, Thermo Electron, Waltham MA). 5μm sections were cut

using a microtome (Model 2040, Leica Micro System, Peabody, MA) and mounted on glass

coverslips. Hematoxylin and Eosin staining was carried according to a standard protocol

using automated staining equipment (Shandon Gemini, Thermo Electron, Waltham MA)

and the stained sections were observed under an optical light microscope (BX50, Olympus

USA, Center Valley, PA).

Results and Discussion:

Figure 1 represents schematically results of interaction between GFP-labeled human breast

cancer cells (MDA-MB-231) and the multiple layer, mineralizing osteoblast tissue in the

bioreactor. We observed (Figure 2) that murine MC3T3E-1 osteoblasts maintained in the

bioreactor for periods up to 10 months develop a tissue-like network (Panel A and B) of 6-8

layers of differentiated cells within a highly collagenous matrix that stain for alkaline

phosphatase activity and mineralization by Von Kossa (Panel B), as well as form

mineralized nodules (Panel C) that prove positive for calcium and phosphorous by scanning

electron microscopy/energy-dispersive analysis of x-rays. 5 months of MC3T3-E1 culture

resulted in formation of visually apparent macroscopic bone especially in the form of

108

deposits on the dialysis membrane (Fig. 2D). Injection of MDA-MB-231 human breast

cancer cells (Figure 3, Panel A) directly onto the osteoblast tissue leads to cancer cell,

adhesion and tissue penetration within 24 hours (Panel C) probably accomplished through

the extension of long cellular processes (Panel B). In figures 4 and 5 are results of in situ

confocal microscopy study of breast cancer cell penetration of, and replication within, an

MC3T3E-1 osteoblast tissue developed in the bioreactor over 5 months of continuous

culture. Figure 4, a collection of optical sections through the thickness of the tissue

corresponding to 1-3 days of cancer cell/bone tissue co-culture indicates that solitary cancer

cells (day 1), replicate (day 2 ) and co-localize with the osteoblasts and penetrate the entire

thickness of the tissue (day 3). Figure 5 shows 3D reconstructions of optical sections like

those shown in Figure 4 corresponding to days 1-3 of co-culture, respectively. We interpret

these images to be multi-layers of MC3T3E-1 incorporated in a thick collagenous matrix

(black, compare to Figure 2) into which columns of cancer cells penetrate in a few locations

during the first day of co-culture (Panel B). Within 2 days of co-culture, breast cancer cells

replicate (Panel B) and begin to organize into linear files especially evident in Panel C of

Figure 4. Close inspection of both 2D optical sections and 3D reconstructions (Figure 5)

suggests concomitant remodeling of the osteoid tissue. Before cancer-cell challenge,

osteoblasts exhibit a rounded, cuboidal morphology. Over 3 days of cancer cell co-culture,

osteoblasts take on a definitively elongated appearance and align with cancer cells. Similar

morphological changes have been observed in osteoblasts exposed to breast-cancer-cell-

conditioned medium in conventional culture. It was also observed that tissue from co-

culture experiments was much more fragile than originating tissue not exposed to cancer

cells, requiring very careful processing to prevent wholesale cell sloughing during the wash

steps involved in preparation of specimens for histology and electron microscopy. It was

plainly evident that the tissue matrix was eroded and easily separated from the base film of

the bioreactors used in this work. We speculate that osteoblast tissue destruction occurs

109

through increased apoptosis and suppression of osteoblast differentiation that has been

observed in conventional 2D tissue culture24, as well as wholesale degradation of the thick

extracellular matrix (ECM) in which osteoblasts are embedded (Figure 1) through the action

of matrix metalloproteinases and cathepsin K synthesized by cancer cells.25-27 Further studies

using genomic and proteomic tools can delineate the specific mechanisms responsible for

degradation of osteoblast tissue by breast cancer cells observed in the bioreactor.

Cancer-related bone loss appears to occur through multiple pathways, including the

osteoclast-mediated resorption.10,11,28 In particular, destruction of devitalized bone directly

by cancer cells has been reported10,11,28 and it has been found that, late in metastasis when

bone-degradation rate is highest, osteoclast cell numbers are actually in precipitous decline.29

These lines of evidence support the idea that osteoclasts are not solely responsible for

excessive bone degradation and that cancer cells directly contribute to bone loss.

Degradation of the osteoblast-derived osteoid tissue by co-culture with breast cancer cells

observed in the model system presented here (that purposely excludes osteoclasts) strongly

suggests that yet another mechanism of bone loss is related to disruption of the bone-

accretion process by destruction of osteoblastic tissue. There is clinical and experimental

literature to support this concept. For example, quantitative histomorphometric analyses of

bone biopsies from patients with hypercalcemia due to bone metastasis indicated a dramatic

decrease in osteoblast activity.30 Histomorphometric analysis of rodents inoculated with lytic

human breast cancer cells (MDA-MB-231) indicated that, even though administration of

risedronate (a bisphosphonate) reduced the number of osteoclasts, slowed bone lysis, and

significantly reduced tumor burden, there was no evidence of new bone deposition or repair.

Similarly, administration of bisphosphonates to humans with osteolytic metastasis slowed

lesion progression but did not bring about healing.31 All taken together, these observations

strongly suggest that normal osteoblast function (i.e. deposition of matrix) is not only

110

impaired in the presence of breast cancer cells but, in fact, osteoblastic tissue is destroyed by

an orchestrated attack by cancer cells, possibly by enlisting a cooperative response by

osteoblasts themselves.

Conclusions and Future Work:

We have shown that metastatic breast cancer challenge of 3D osteoblast tissue grown in a

specialized bioreactor over 5 months of continuous culture permitted direct observation of

interaction of breast cancer cells with the osteoblast tissue. Pathological outcomes such as

tissue penetration and cancer cell filing which are not well modeled in conventional culture

were clearly observed in the bioreactor. The bioreactor allowed us to control and select

stages in osteoblast development for challenge with cancer cells. By varying the breast

cancer cells/ osteoblast ratio the model allows us to probe the conditions under which cancer

cells remain as solitary cells or associate to form tumors simulating early stages of cancer

colonization of bone. A decided advantage of the bioreactor cultures is that medium in the

cell culture chamber is not disturbed throughout the course of the experiment and various

cell-secreted factors are allowed to accumulate within the cell chamber. Supernatant from

the culture medium can be collected at periodic co-culture intervals to assay for specific

factors that may be secreted by cancer cells over the course of their interaction with 3D

tissue. Isolating the RNA from the osteoblast cultures at different co-culture intervals can

elucidate the mechanisms that lead to changes in osteoblast physiology in the presence of

breast cancer. The bioreactor model also permits study of the effect of specific drugs that

may inhibit cancer cell penetration, tumor formation and osteoblast tissue destruction under

controlled conditions over extended culture intervals. Further understanding of the cellular

and molecular basis for breast-cancer interactions with osteoblasts and discovery of

111

therapeutic interventions will be greatly expedited by the use of three-dimensional bone

tissue models such as the one reported herein.

112

References

1. Griffith LG, Naughton G. Tissue Engineering--Current Challenges and Expanding

Opportunities. Science 2002;295(5557):1009-1014.

2. Kim JB, Stein R, O'Hare MJ. Three-Dimensional in vitro Tissue Culture Models of

Breast Cancer - A Review. Breast Cancer Research and Treatment 2004;85:281-291.

3. Steeg PS. Cancer biology: Emissaries set up new sites. 2005;438(7069):750-751.

4. Woodhouse EC, Chuaqui RF, Liotta LA. General mechanisms of metastasis.

Cancer 1997;80(8 Suppl):1529-37.

5. Rubens RD. Bone metastases--the cllinical problem. Eur J Cancer 1998;34:210-213.

6. Rubens RD, Mundy GR. Cancer and the skeleton. London: Martin Dunitz; 2000.

7. Nielsen OS, Munro AJ, Tannock IF. Bone metastases: pathophysiology and

management policy. J Clin Oncol 1991;9(3):509-24.

8. Landis SH, Murray T, Bolden S, Wingo PA. Cancer statistics. CA Cancer J Clin

1999;49:8-31.

9. Chambers AF, Groom AC, MacDonald IC. Dissemination and growth of cancer

cells in metastatic sites. Nature reviews. Cancer 2002;2(8):563-72.

10. Mundy GR. Metastasis to bone: causes, consequences and therapeutic opportunities.

Nat Rev Cancer 2002;2:584-593.

11. Kozlow W, Guise TA. Breast cancer metastasis to bone: mechanisms of osteolysis

and implications for therapy. J Mammary Gland Biol Neoplasia 2005;10(2):169-80.

12. Sanchez-Sweatman OH, Lee J, Orr FW, Singh G. Direct osteolysis induced by

metastatic murine melanoma cells: role of matrix metalloproteinases. European

journal of cancer 1997;33(6):918-25.

13. Clohisy DR, Perkins SL, Ramnaraine ML. Review of cellular mechanisms of tumor

osteolysis. Clin Orthop Relat Res 2000. p 104-14.

113

14. Clemons M, Dranitsaris G, Cole D, Gainford MC. Too much, too little, too late to

start again? Assessing the efficacy of bisphosphonates in patients with bone

metastases from breast cancer. Oncologist 2006;11(3):227-233.

15. Guise TA, Mundy GR. Cancer and bone. Endocr Rev 1998;19:18-54.

16. Mastro AM, Gay CV, Welch DR. The skeleton as a unique environment for breast

cancer cells. Clinical and Experimental Metastasis 2003;20(3):275-284.

17. Mastro AM, Gay CV, Welch DR. The skeleton as a unique environment for breast

cancer cells. Clin & Exp Metas 2003;20:275-284.

18. Mercer R, Miyasaka C, Mastro AM. Metastatic breast cancer cells suppress

osteoblast adhesion and differentiation. Clin & Exp Metas 2004;21(5):427-435.

19. Mastro A, Gay C, Welch D, Donahue H, Jewell J, Mercer R, DiGirolamo D,

Chislock E, Guttridge K. Breast cancer cells induce osteoblast apoptosis: a possible

contributor to bone degradation. J Cell Biochem 2004;91(2):265-276.

20. Mercer R, Miyasaka C, Mastro A. Metastatic breast cancer cells suppress osteoblast

adhesion and differentiation. Clin Exp Metastasis 2004;21(5):427-35.

21. Phadke PA, Mercer RR, Harms JF, Jia Y, Frost AR, Jewell JL, Bussard KM,

Nelson S, Moore C, Kappes JC and others. Kinetics of metastatic breast cancer cell

trafficking in bone. Clinical cancer research 2006;12(5):1431-40.

22. Welch DR, Harms JJF, Mastro AM, Gay CV, Donahue HJ. Breast cancer

metastasis to bone: evolving models and research challenges. Journal of

musculoskeletal & neuronal interactions 2003;3(1):30-8.

23. Dhurjati R, Liu X, Gay CV, Mastro AM, Vogler EA. Extended-Term Culture of

Bone Cells in a Compartmentalized Bioreactor. Tissue Engineering 2006;12:3045-

3054.

114

24. Mastro AM, Gay CV, Welch DR, Donahue HJ, Jewell J, Mercer R, DiGirolamo D,

Chislock EM, Guttridge K. Breast cancer cells induce osteoblast apoptosis: a possible

contributor to bone degradation. J Cell Biochem 2004;91(2):265-76.

25. Chen WT. Proteolytic activity of specialized surface protrusions formed at rosette

contact sites of transformed cells. J Exp Zool 1989;251(2):167-85.

26. Weaver VM, Fischer AH, Peterson OW, Bissell MJ. The importance of the

microenvironment in breast cancer progression: recapitulation of mammary

tumorigenesis using a unique human mammary epithelial cell model and a three-

dimensional culture assay. Biochem Cell Biol 1996;74(6):833-51.

27. Kelly T, Mueller SC, Yeh Y, Chen WT. Invadopodia promote proteolysis of a wide

variety of extracellular matrix proteins. J Cell Physiol 1994;158(2):299-308.

28. Sanchez-Sweatman OH, Lee J, Orr FW, Singh G. Direct osteolysis induced by

metastatic murine melanoma cells: role of matrix metalloproteinases. Eur J Cancer

1997;33(6):918-25.

29. Phadke PA, Mercer RR, Harms JF, Jia Y, Frost AR, Jewell JL, Bussard KM,

Nelson S, Moore C, Kappes JC and others. Kinetics of metastatic breast cancer cell

trafficking in bone. Clin Cancer Res 2006;12(5):1431-40.

30. Stewart AF, Vignery A, Silverglate A, Ravin ND, Livolsi V, Broadus AE, Baron R.

Quantitative bone histomorphology in humoral hypercalcemia of malignancy:

uncoupling of bone cell activity. J Clin Endocrinol Metab 1982;55:219-227.

31. Lipton A, Theriault RL, Hortobagyi GN, Simeone J, Knight RD, Mellars K,

Reitsma DJ, Heffernan M, Seaman JJ. Pamidronate prevents skeletal complications

and is effective palliative treatment in women with breast carcinoma and osteolytic

bone metastases. Cancer 2000;88(5):1082-1090.

115

List of Figure Legends

Figure 1: Schematic showing the sequential steps of cancer cell attachment, penetration,

replication and possible host tissue destruction observed in the bioreactor as a result of

interactions between green fluorescent protein (GFP) labeled breast cancer cells and multiple

layer 3D osteoblast tissue.

Figure 2: (A) Light micrographs of a Hemotoxylin and Eosin stained 70-day tissue (40X)

reveal osteoblasts (pale red with dark-red nuclei) lining a collagenous matrix (pale pink).

(B)Transmission electron micrographs (TEM) of 22-day tissue cross sections show up to 5

cell layers (1500X). (C) Scanning electron micrographs of the surface of a 70 day culture are

studded with mineral nodules (1000X) that prove positive for calcium and phosphorous by

energy-dispersive analysis of x-rays (not shown). (D) Bone chip recovered from a 5 month

MC3T3-E1 bioreactor. X-ray diffraction (inset) indicates similarity with a bovine bone

reference spectrum (lines).

Figure 3: Transmission electron micrograph (TEM) of a cross-section of an isolated MDA-

MB-231 breast cancer cell crawling on the base film of the bioreactor used in this work

(cultured for 3 days in osteoblast-conditioned medium) reveals an amoeboid morphology

with protrusive structures at the leading edge, consistent with a migratory phenotype (Panel

A, 1500X). TEM of breast cancer cells (BC in Panel B, 5000X) co-cultured for 10-days with

a 60-day MC3T3-E1 tissue shows that cancer cells displaced osteoblasts originally adhering

to the base film (see Figure 2) and exhibit protrusive structures that may be involved in

penetration. (arrows; compare to Panel A). Fluorescence microscopy oft breast cancer cells

116

(green) in co-culture with osteoid tissue (red) organize in single-file order (Panel C, 40X),

captured in cross-section by TEM (inset, 2000X).

Figure 4: Serial optical sections through an MC3T3-E1 derived osteoid tissue (5-month

culture interval stained with Cell Tracker Orange ™) co-cultured with GFP-labeled MDA-

MB-231 breast cancer cells (green) for 3 days. Optical sections are from bottom [0μm] to top

[16μm]) of the tissue. (40X, scale bar = 50 μm)

Figure 5: 3D reconstruction of serial optical sections taken through the thickness of the 5

month osteoblast tissue cultured in the bioreactor and subject to MDA-MB-231 breast cancer

challenge over 3 day co-culture interval.

117

Figure 1

118

BC

A B

Figure 2

D

119

BC

BC

B

5µm

A

C

Figure 3

120

Day 2

Day 3

Day 1

0µm 4µm 8µm 12µm 16µm

Figure 4

121

A

B

C

Figure 5

122

VITA

Dhurjati Ravi

Dhurjati Ravi was born in Kakinada in India on December 08, 1976. Dhurjati graduated from

Kerala University with a degree in Mechanical Engineering. Dhurjati came to the United States

to pursue higher education and received his Masters Degree in Materials from Penn State

University in 2003. Dhurjati has been pursuing doctoral degree in Materials Science and

Engineering since Jan 2004. Dhurjati’s research interests are in the broad area of tissue

engineering and with specific interest in simulating complex physiological and pathological

processes using engineered tissue.

Research Accomplishments

Osteogenesis and Osteocytic Transformation In Vitro

Continuous culture of bone forming cells for extended periods up to a

year resulted in a specialized bioreactor resulted in growth/maturation

of 3D multilayered tissue, osteocytic transformation and macroscopic

bone formation.

System in Crisis: Breast Cancer Colonization of Bone In Vitro

In vitro bone tissue surrogate challenged with metastatic breast cancer

cells, effectively captured fundamental aspects of breast cancer

colonization of bone.

Publications Ravi D, Liu X, Mastro AM, Gay CV and Vogler EA Extended-Term Culture of Bone Cells in a Compartmentalized Bioreactor Tissue Engineering, [12], 3045-3054, Nov 2006.

Liu X, Lim JY, Donahue JH, Ravi D, Mastro AM and Vogler EA Influence of Substratum Surface Chemistry/Energy and Topography on the Human Fetal Osteoblast Cell Line HFOB 1.19: Phenotypic and Genotypic Responses Observed In Vitro Biomaterials, [28], 4535-4550 Nov 2007

Ravi D, Green DJ. Sintering Stresses and Distortion Produced By Density Differences inBi-layer Structures. J. Eur. Ceram. Soc, [26], 17-25, Jan 2006

Documents

ON THE PERMANENT LIFE OF TISSUE OUTSIDE THE ORGANISM