<%BANNER%>

Differentiation of Putative Cancer Stem Cells and Its Effect on Tumorigenicity in Osteosarcoma

Permanent Link: http://ufdc.ufl.edu/UFE0022577/00001

Material Information

Title: Differentiation of Putative Cancer Stem Cells and Its Effect on Tumorigenicity in Osteosarcoma
Physical Description: 1 online resource (68 p.)
Language: english
Creator: Currie, Thomas
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: Medicine -- Dissertations, Academic -- UF
Genre: Medical Sciences thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: In previous work, our group had shown that a subpopulation of cells in osteosarcoma, a mesenchymal malignancy, have certain stem cell like properties. They expressed embryonic stem cell transcription genes essential in maintaining pluripotency (Oct-4 and Nanog), they were capable of growing as spherical colonies, and they had the capacity for self renewal in vitro. Additionally, some of our cell lines formed tumors when xenografted subcutaneously into non-obese diabetic/severe combined immune deficient (NOD/SCID) mice. To investigate the relationship between ?stemness? and tumorigenesis, a cell line (OS521Oct-4/GFP) was transfected with a reporter construct containing the human Oct-4 promoter linked to the gene for green fluorescent protein, and an independently regulated neomycin resistance gene. These transgenic cultures were observed to express GFP heterogeneously, as were the tumors that were formed from them. Using fluorescent activated cell sorting, the GFP+ cells were found to be 100 fold more tumorigenic than GFP- cells in terms of penetrance and kinetics, thus indicating that cells capable of expressing the Oct-4 gene had enhanced tumorigenic capacity. To further investigate the putative osteosarcoma stem cell, we clonally expanded the Oct-4/GFP+ cells and characterized their capacity to form tumors following xenotransplantation in NOD/SCID mice. Delivery of the cells yielded tumors that were heterogeneous for Oct-4 expression, showing that Oct-4/GFP+ cells can give rise to GFP- cells in vivo. We also demonstrated that subsequent passage of these cells gives rise to a population heterogeneous for Oct-4/GFP expression. This suggested that the loss of tumorigenicity was associated with cellular differentiation as indicated by the inability to activate the Oct-4 promoter. Secondly, because differentiation of tumor cells has proven to be therapeutic in other forms of cancer, we wanted to examine the effect of induced differentiation of the putative osteosarcoma stem cell. OS521Oct-4/GFP clones in monolayer were incubated in commercially available osteogenic differentiation media for 21 days. We observed a 32% reduction in the proportion of cells expressing Oct-4, suggesting that certain tumor cells differentiated and no longer possessed a stem like phenotype. The clonally derived tumor cells were then incubated in the presence of BMP4 or BMP 7 at 200 > ng/ml. Flow cytometry showed a 58% and 36% reduction in GFP expression in the BMP 4 and BMP7 treated cells, respectively. These cells were then xenografted subcutaneously into NOD/SCID mice, and despite apparent differentiation, tumors still formed in both conditions. Analysis of the corresponding BMP4 tumors showed GFP expression at 10%, and the BMP 7 condition showed GFP expression at 70%. These data suggest that a clonally derived population, homogeneous for Oct-4 expression, can undergo differentiation events leading to a loss of expression of the gene when passaged in vivo. The resulting tumors recapitulate the heterogeneity of the parental line and confirm the notion that GFP- cells arise from GFP+ cells. Further induced differentiation via BMP4/7 exposure can significantly decrease the proportion of Oct-4 expression, but not to the extent where tumor formation is completely inhibited.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Thomas Currie.
Thesis: Thesis (M.S.)--University of Florida, 2008.
Local: Adviser: Ghivizzani, Steven.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022577:00001

Permanent Link: http://ufdc.ufl.edu/UFE0022577/00001

Material Information

Title: Differentiation of Putative Cancer Stem Cells and Its Effect on Tumorigenicity in Osteosarcoma
Physical Description: 1 online resource (68 p.)
Language: english
Creator: Currie, Thomas
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: Medicine -- Dissertations, Academic -- UF
Genre: Medical Sciences thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: In previous work, our group had shown that a subpopulation of cells in osteosarcoma, a mesenchymal malignancy, have certain stem cell like properties. They expressed embryonic stem cell transcription genes essential in maintaining pluripotency (Oct-4 and Nanog), they were capable of growing as spherical colonies, and they had the capacity for self renewal in vitro. Additionally, some of our cell lines formed tumors when xenografted subcutaneously into non-obese diabetic/severe combined immune deficient (NOD/SCID) mice. To investigate the relationship between ?stemness? and tumorigenesis, a cell line (OS521Oct-4/GFP) was transfected with a reporter construct containing the human Oct-4 promoter linked to the gene for green fluorescent protein, and an independently regulated neomycin resistance gene. These transgenic cultures were observed to express GFP heterogeneously, as were the tumors that were formed from them. Using fluorescent activated cell sorting, the GFP+ cells were found to be 100 fold more tumorigenic than GFP- cells in terms of penetrance and kinetics, thus indicating that cells capable of expressing the Oct-4 gene had enhanced tumorigenic capacity. To further investigate the putative osteosarcoma stem cell, we clonally expanded the Oct-4/GFP+ cells and characterized their capacity to form tumors following xenotransplantation in NOD/SCID mice. Delivery of the cells yielded tumors that were heterogeneous for Oct-4 expression, showing that Oct-4/GFP+ cells can give rise to GFP- cells in vivo. We also demonstrated that subsequent passage of these cells gives rise to a population heterogeneous for Oct-4/GFP expression. This suggested that the loss of tumorigenicity was associated with cellular differentiation as indicated by the inability to activate the Oct-4 promoter. Secondly, because differentiation of tumor cells has proven to be therapeutic in other forms of cancer, we wanted to examine the effect of induced differentiation of the putative osteosarcoma stem cell. OS521Oct-4/GFP clones in monolayer were incubated in commercially available osteogenic differentiation media for 21 days. We observed a 32% reduction in the proportion of cells expressing Oct-4, suggesting that certain tumor cells differentiated and no longer possessed a stem like phenotype. The clonally derived tumor cells were then incubated in the presence of BMP4 or BMP 7 at 200 > ng/ml. Flow cytometry showed a 58% and 36% reduction in GFP expression in the BMP 4 and BMP7 treated cells, respectively. These cells were then xenografted subcutaneously into NOD/SCID mice, and despite apparent differentiation, tumors still formed in both conditions. Analysis of the corresponding BMP4 tumors showed GFP expression at 10%, and the BMP 7 condition showed GFP expression at 70%. These data suggest that a clonally derived population, homogeneous for Oct-4 expression, can undergo differentiation events leading to a loss of expression of the gene when passaged in vivo. The resulting tumors recapitulate the heterogeneity of the parental line and confirm the notion that GFP- cells arise from GFP+ cells. Further induced differentiation via BMP4/7 exposure can significantly decrease the proportion of Oct-4 expression, but not to the extent where tumor formation is completely inhibited.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Thomas Currie.
Thesis: Thesis (M.S.)--University of Florida, 2008.
Local: Adviser: Ghivizzani, Steven.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022577:00001


This item has the following downloads:


Full Text





DIFFERENTIATION OF PUTATIVE CANCER STEM CELLS AND ITS EFFECT ON
TUMORIGENICITY IN OSTEOSARCOMA



















By

THOMAS PATRICK CURRIE


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA

2008


































2008 Thomas Patrick Currie




































To my Mother









ACKNOWLEDGMENTS

First and foremost, I would like to thank my parents for supporting me through all of my

educational endeavors. Without them, the road would have been impossible. They always pushed

me to be the best that I could be, and to pursue my dreams- promising that they would support

my decisions along the way. To this day they have kept their word, and because of this I have

been able to further my education with uncompromised focus. As I begin dental school in the fall

of 2008, I am still comforted by their promise that they are behind me 100%.

Most importantly I would like to thank Steven Ghivizzani PhD, Parker Gibbs MD, and

Padraic Levings PhD for guiding me through my classes, research, and requirements for the

masters program. They graciously included me in their project, allowed me to focus my thesis on

one of their specific aims, and provided direction along the way. They truly mentored me in the

laboratory, ensuring my success through their advice and wisdom. They understood my mishaps

and mistakes, offering constructive criticism and providing learning experiences, rather than

acting in a belittling manner. All of the long hours and late nights have not gone unnoticed and I

appreciate them for their dedication in helping me pursue my degree.

I must recognize the importance of my committee, which included Steven Ghivizzani

PhD, David Bloom PhD, and Ammon Peck PhD. They amicably critiqued my work and steered

me down the path to success, while ensuring my data would be scientifically sound. They raised

the bar by challenging me to investigate questions that I was otherwise naive to, allowing me to

see things from new perspectives. I know they have dedicated numerous hours to reviewing and

helping me and with my work, and I cannot forget that. I truly appreciate their assistance and

commitment to my education.

I am also grateful to Steve McClellan, who taught me how to use the flow cytometer. He

spent a great deal of time educating me on its usage and answering my questions when things









went awry. He was very patient with me, and besides being an excellent teacher, kept the flow

cytometry lab interesting as well with his wit and humor.

I would also like to thank all of my professors that have been gracious enough to lecture

with enthusiasm and expertise, maintaining a stimulating learning environment along the way. I

can truly say that I have learned a great deal from them, vastly increasing my understanding of

the biological sciences and improving my view on the mechanisms of life.

I must include special thanks to Mrs. Conners for her help in all aspects of the program-

from the time I applied, right up until graduation. Without her help, even registering for classes

would have been a confusing process. Her advice was always in my best interest, and she always

urged me to seek her help before going things alone. I know that she truly cares for all of the

students in the program and that their well being is at the forefront of her priorities.

I would like to acknowledge the rest of the Ghivizzani lab and their contribution to my

success. Carrie, Anthony, Rachael, Marsha, Jesse, Kristen, Olga and Celine were all very

helpful, but more importantly I think that they made working in the lab a fun and exciting place

to be. Spending long hours performing bench top procedures would have been unnerving had it

not been for their company and companionship throughout the past year.

Of all of my mentors, I must express my greatest gratitude to Padraic Levings, PhD.

"Paddy." Through his selflessness and benevolence, he has provided me with more assistance

than anyone else in attaining my goals. He patiently walked me step by step through the paces of

laboratory work, making sure that I was competent enough before I ventured on my own. His

upbeat demeanor kept me motivated and eager to learn; making this program a very exciting and

fun experience for me, while also allowing us to form a great friendship. Overshadowing all of

this was his unrelenting efforts to ensure my success in the program. I know that he has spent









uncountable hours helping me with my experiments, presentations, and manuscripts; none of

which he was required to do. I must say that I owe many of my accomplishments to him and his

desire to see me succeed.

Lastly I would like to thank some miscellaneous people. First I would like to thank the

knuckle-headed patrons who frequent the Swamp Restaurant during late night hours. They have

kept my desire to further my education salient, while also reminding me that I do not want to be

a bouncer for the rest of my life. Secondly I would like to thank my two dogs, Moose and

Legend, for keeping me sane and tranquil during the stressful parts of typing my thesis. Lastly, I

would like to thank the Discovery channel, specifically Dirty Jobs and Myth Busters, for

entertaining my brain during the time when I could not concentrate on my school work.









TABLE OF CONTENTS

page

A C K N O W L E D G M E N T S ..............................................................................................................4

LIST OF FIGURES .................................. .. ..... ..... ................. .9

ABSTRAC T ................................................... ............... 10

CHAPTER

1 INTRODUCTION ............... ................. ........... .............................. 12

Aim 1: To Identify the Origin of the Heterogeneous Expression Of Oct-4/GFP in the
O S521O ct-4GFP Cell Line .............. .... .......... ..... ................................................14
Aim 2: To Address the Role of Differentiation on Tumorigenic Capacity of OS5210ct-
4/GFP Clones ................ .................. .......................................14

2 M E T H O D S ........................ ......... .......... .............................................. 15

Derivation of Osteosarcoma Cell Lines from Patient Biopsies and Description of phOct-
4 /G F P ............................................................................................................................. . 1 5
Xenotransplantation and Tumorigenicity Assays..... .......... ........................................ 16
Fluorescence Activated Cell Sorting and Flow Cytometry .......................................... 17
Clonal Expansion of OS521Oct-4/GFP Cells.. ............................................................. 18
In vitro O steogenic D differentiation ............................................ ....................................... 18
Treatm ent with Bone M orphogenic Proteins................................... .................................... 18

3 L ITE R A TU R E R E V IE W ........................................................................ .. .......................20

The Biology of Cancer ..................................... .............. ................. ........... 20
T he B biology of O steosarcom a ....................................................................... ...................2 1
Stem C ells................ ...........................................................22
C cancer Stem C ell Theory................. .... .......... .. .. ...................................... ......... ......25
Use of Agents That Induce Differentiation for the Treatment of Cancer............................26
P relim inary R results ............... ................... ................ .... ...........................28
Cells Isolated from Bone Sarcoma Cultures Exhibit Stem-Like Attributes ....................28
Development of an In vivo Model to Examine the Role of Stem-Like cells in the
Pathogenesis of Osteosarcoma......................................... ...............32
Expression of an Oct-4 Promoter/GFP Reporter Construct That Selectively
Identifies Cancer Stem Cells in Osteosarcoma................................. ...... ............ ...33

4 R E SU L T S ...........................................................................................3 7

R rationale for A im 1 ...........................................................................37
R rationale for A im 2 ........................................................................ 38
Clonal Expansion of OS5210 ct-4/GFP cells ................................................................... 39









Flow Cytometric Analysis of Clones and Clonally Derived Tumors............... ........... 40
Culture of OS5210ct-4/GFP Clones in Osteogenic Media..............................................41
Culture of OS5210ct-4/GFP Clonally Derived Tumor Cells in BMP Conditioned Media...42

5 D ISCU SSIO N ............. .. ....... ..................................... ........................ 44

L IST O F R E F E R E N C E S ......... .. ............... ................. ............................................................64

B IO G R A PH IC A L SK E T C H .........................................................................................................68









LIST OF FIGURES


Figure p e

5-1 Osteosarcoma: H&E stain of a femoral osteosarcoma. .................................................... 51

5-2 Cancer Stem Cells ............. .................... ........ .. ......... .. .............. 52

5-3 ESC-specific Genes in Sarcospheres ........................................ ........................... 53

5-4 Immunohistochemical staining for Oct-4 and Nanog in sections from tumor biopsies
of chondro- and osteosarcom a ........................................ ............................................54

5-5 Analyses of bone sarcoma cultures for expression of genes of endo- and ectodermal
lin e ag e s ........................................................ ...................................5 5

5-6 M ultipotent cells in bone sarcom a ....................................................................... 56

5-7 Schem atic of phO ct-4/G FP ....................................... ............ .............. .........................57

5-8 Expression of phOct-4/GFP and surface antigens in OS521 in vitro and in vivo .............58

5-9 The Oct-4/GFP enriched tumor fraction of OS 521 is highly tumorigenic following
delivery into N O D /SCID m ice................................................. .............................. 59

5-10 Serial transplant of Oct-4/GFP clones in vivo ........................................ 60

5-11 The Oct-4/GFP-enriched fraction of clonally derived tumors is highly tumorigenic in
v iv o ............................................................................6 1

5-12 Culture in osteogenic media induces silencing of the Oct-4/GFP transgene.................62

5-13 Treatment of OS tumor cells with BMPs induced silencing of Oct-4/GFP expression
in vitro ......... ............... ....... ......... .......................................... ...... 63









Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

DIFFERENTIATION OF PUTATIVE CANCER STEM CELLS AND ITS EFFECT ON
TUMORIGENICITY IN OSTEOSARCOMA

By

Thomas Patrick Currie

August 2008

Chair: Steven Ghivizzani
Major: Medical Sciences

In previous work, our group had shown that a subpopulation of cells in osteosarcoma, a

mesenchymal malignancy, have certain stem cell like properties. They expressed embryonic

stem cell transcription genes essential in maintaining pluripotency (Oct-4 and Nanog), they were

capable of growing as spherical colonies, and they had the capacity for self renewal in vitro.

Additionally, some of our cell lines formed tumors when xenografted subcutaneously into non-

obese diabetic/severe combined immune deficient (NOD/SCID) mice.

To investigate the relationship between "stemness" and tumorigenesis, a cell line

(OS5210ct-4/GFP) was transfected with a reporter construct containing the human Oct-4

promoter linked to the gene for green fluorescent protein, and an independently regulated

neomycin resistance gene. These transgenic cultures were observed to express GFP

heterogeneously, as were the tumors that were formed from them. Using fluorescent activated

cell sorting, the GFP+ cells were found to be 100 fold more tumorigenic than GFP- cells in terms

of penetrance and kinetics, thus indicating that cells capable of expressing the Oct-4 gene had

enhanced tumorigenic capacity.

To further investigate the putative osteosarcoma stem cell, we clonally expanded the Oct-

4/GFP+ cells and characterized their capacity to form tumors following xenotransplantation in









NOD/SCID mice. Delivery of the cells yielded tumors that were heterogeneous for Oct-4

expression, showing that Oct-4/GFP+ cells can give rise to GFP- cells in vivo. We also

demonstrated that subsequent passage of these cells gives rise to a population heterogeneous for

Oct-4/GFP expression. This suggested that the loss of tumorigenicity was associated with

cellular differentiation as indicated by the inability to activate the Oct-4 promoter.

Secondly, because differentiation of tumor cells has proven to be therapeutic in other

forms of cancer, we wanted to examine the effect of induced differentiation of the putative

osteosarcoma stem cell. OS521Oct-4/GFP clones in monolayer were incubated in commercially

available osteogenic differentiation media for 21 days. We observed a 32% reduction in the

proportion of cells expressing Oct-4, suggesting that certain tumor cells differentiated and no

longer possessed a stem like phenotype.

The clonally derived tumor cells were then incubated in the presence of BMP4 or BMP 7

at 200>ng/ml. Flow cytometry showed a 58% and 36% reduction in GFP expression in the BMP

4 and BMP7 treated cells, respectively. These cells were then xenografted subcutaneously into

NOD/SCID mice, and despite apparent differentiation, tumors still formed in both conditions.

Analysis of the corresponding BMP4 tumors showed GFP expression at 10%, and the BMP 7

condition showed GFP expression at 70%.

These data suggest that a clonally derived population, homogeneous for Oct-4

expression, can undergo differentiation events leading to a loss of expression of the gene when

passage in vivo. The resulting tumors recapitulate the heterogeneity of the parental line and

confirm the notion that GFP- cells arise from GFP+ cells. Further induced differentiation via

BMP4/7 exposure can significantly decrease the proportion of Oct-4 expression, but not to the

extent where tumor formation is completely inhibited.









CHAPTER 1
INTRODUCTION

Osteosarcoma (OS) is the most common primary bone malignancy of childhood and

adolescence. Despite advances in surgery and chemotherapy, long-term survival rates have

stagnated over the last 30 years, and still 40% of patients diagnosed with osteosarcoma die of

their disease. The cancer stem cell theory, though, may account for the lack of effective

chemotherapeutics in this and other types of cancer. Recent studies of leukemia, brain and

breast cancer, suggest that within a tumor a subpopulation of cells with properties of stem cells,

is responsible for tumorigenesis and malignancy while the remaining tumor mass is comprised

of differentiated progeny of the tumor stem cell that are relatively benign. Thus, agents designed

to eradicate the bulk of rapidly dividing cells within a tumor mass may not eliminate the actual

tumorigenic cell. Following these reports, we initiated studies to explore the participation of

stem-like cells in bone sarcomas. Identification of such a cancer stem cell could provide

valuable insight into the process of tumorigenesis in osteosarcoma and perhaps a more effective

target for therapeutic intervention.

In previous and ongoing work we have demonstrated that a subpopulation of cells in

bone sarcomas share attributes with certain stem-, or stem-like cells: 1) They demonstrated the

capacity for self-renewal in vitro in clonal expansion assays (they formed sarcospheres in

anchorage-independent serum starved conditions), and in vivo following serial

xenotransplantation in immunocompromised mice; 2) cells from tumors were capable of being

induced to differentiate along mesenchymal lineages, and 3) they differentially expressed the

embryonic stem (ES) cell-specific genes Oct-4 and Nanog. We saw a direct correlation between

the expression of Oct-4 and the functional behavior of cells isolated from bone sarcomas. We

found that in vitro, as our sphere cultures enlarged and differentiated, the percentage of cells









expressing these genes decreased. In at least one patient derived osteosarcoma cell line

(OS521Oct-4/GFP), tumorigenesis, as indicated by the capacity to form tumors following

xenotransplantation in immunocompromised mice, appeared to be functionally linked to the

ability of the cells to express an exogenous reporter construct. This construct contained the full

length Oct-4 promoter linked to the coding region for green fluorescent protein GFP) and an

independent SV40-promoter driven neomycin resistance cassette. This allowed positive

selection of cells that acquired the plasmid, irrespective of their capacity to transcribe the Oct-4

promoter. Thus, in medium supplemented with G418, all surviving cells would contain the

plasmid and express the neomycin resistance gene; however, only the subset of cells that were

capable of activating the Oct-4 promoter sequence would fluoresce green. We found that

following stable transfection of the osteosarcoma line, expression of Oct4 was heterogeneous.

Only about 24% of the G418 resistant cells were GFP+. This subpopulation of GFP+ cells had

greater than 100 fold enhanced tumorigenicity relative to populations that did not transcribe the

Oct-4 promoter. Those cells that in vivo spontaneously lost the capacity to express the Oct-4

reporter construct also lost their ability to form tumors following xenotransplantation.

These data suggest that the tumorigenic cells in osteosarcoma, a somatic malignancy,

have been reprogrammed to an "ES cell-like" phenotype and that this phenotype drives

tumorigenicity.

The goal of my research was to investigate the relationship between cellular

differentiation and tumorigenic capacity in osteosarcoma. We hypothesized that a subset

osteosarcoma cells, which expressed the embryonic transcription factor Oct-4, retained a stem

cell like phenotype; and because of this, these cells were responsible for tumor initiation.

Cellular differentiation of these cells should silence the expression of this gene, resulting in a









loss of their stem cell like qualities, thereby negatively impacting their tumorigenic capacity.

We addressed this hypothesis through two specific aims.

Aim 1: To Identify the Origin of the Heterogeneous Expression Of Oct-4/GFP in the
OS521Oct-4GFP Cell Line

We isolated and expanded individual GFP+clones from xenograft tumors to determine if

the heterogeneity in Oct-4/GFP expression observed in vivo (e.g. 67% Oct-4/GFP+ vs. 33%

GFP-) arose from expansion of pre-existing GFP+ and GFP- cell populations in the cell

inoculums, or represented changes in expression of the Oct-4/GFP reporter arising from

proliferation/differentiation of Oct-4/GFP+ tumor initiating cells. We have previously shown that

activity of the Oct-4/GFP reporter identifies a highly tumorigenic sub-population of cells in

osteosarcoma tumors. We reasoned that any changes in expression of the exogenous Oct-4

promoter could be attributed to changes in the biology of the cells and not to pre-existing

heterogeneity in the injected cell population.

Aim 2: To Address the Role of Differentiation on Tumorigenic Capacity of OS5210ct-
4/GFP Clones

We examined the ability of culture conditions and agents that induce cellular

differentiation of stem cells to alter expression of the Oct-4/GFP reporter in vitro, and inhibit

tumorigenesis in vivo. As indicated in Aim 1, we hypothesized that change in expression of the

Oct-4/GFP reporter and loss of tumorigenic potential in vivo are the results of

proliferation/differentiation of Oct-4/GFP+ tumor initiating cells.









CHAPTER 2
METHODS

Derivation of Osteosarcoma Cell Lines from Patient Biopsies and Description of phOct-
4/GFP

For each tumor sample, cells were isolated from -1 cubic cm of tumor from an open

biopsy taken from patients that have not previously received any form of chemotherapy.

Histologic analysis of a frozen section taken from an adjacent area of the biopsy specimen was

used to confirm the diagnosis of osteosarcoma prior to cell culture preparation. From the biopsy,

the tissue matrix was enzymatically digested with 0.06g each of collagenase type I and II, and

0.3[tg of neutral protease for thirty minutes under agitation. The isolated cells were then

expanded in monolayer culture. To assess the tumorigenicity of the respective osteosarcoma cell

lines, the cells were first trypsinized, washed and counted. The cells were resuspended in a

minimal volume of saline solution, and 3x104 cells were injected subcutaneously into the backs

of separate groups of 6 NOD/SCID mice. (From previous experience with this model system we

have found that the incidence of tumor formation does not increase significantly at cell doses

beyond 3 x 104/mouse.) The animals were then be monitored for tumor growth over a period of 8

weeks. For this initial screen we worked to identify cell lines that formed tumors in at least 3

animals/group at 6 weeks post-injection. From cell lines that were capable of tumorigenesis at a

sufficient frequency, the resulting tumors were harvested, enzymatically digested and the

isolated cells pooled.

The tumor cultures were then expanded and split into thirds. One portion of the cells were

frozen in liquid nitrogen and archived. The second was reseeded into culture and remained

unmodified. The third portion was transfected with the phOct-4/GFP construct, and 24 hours

later seeded into media containing G418 to positively select for cells containing the plasmid.

Following the initial selection (typically 70-80% survive), the transfected cells were then split on









alternate days for 14-21 days at a ratio of 1:3 to ensure stable transfection. The cells in culture

were then expanded to obtain sufficient numbers for evaluation of the relative tumorigenicity of

GFP-enriched and GFP-depleted cell populations in NOD/SCID mice. These cells were also

used to establish clonal populations.

Xenotransplantation and Tumorigenicity Assays

The animals used in this project were maintained according to University of Florida

IACUC guidelines for tumor studies in experimental animals. All animal studies were performed

according to IUCUC protocol number D780.

Cells to be xenografted in to NOD/SCID mice from monolayer were first trypsinized,

washed and counted, then resuspended in Optimem. The desired concentration of cells was

calculated so that injection volumes equaled 100L/mouse.

Fresh tumors, from which cells were to be xenografted into NOD/SCID mice, were first

enzymatically digested as described above. The isolated cells were then expanded for a period of

24 hours in monolayer with standard culture media (DMEM/F-12, 500tg/ml G418, 10% fetal

bovine serum, 3% penstrep). Preparation of these cells for injection then followed monolayer

preparation protocol.

Injection protocols were as follows. Mice were lightly anesthetized via isoflourane

inhalation and, using lml insulin syringes, were subcutaneously inoculated on the dorsal side

between the scapulas. Mice recovered from anesthesia on their own, and were then monitored

daily for tumor formation. The skin of the back was palpated for tumor growth, and the numbers

of animals with tumors, days to tumor onset, and tumor size was scored for tumorigenicity

assays. When tumors reached > 1.0 cm in diameter or 12 weeks post injection, they were

harvested and prepared as necessary for desired assays. The lungs of all animals with tumors

were also harvested and examined for metastases.









Fluorescence Activated Cell Sorting and Flow Cytometry

Cells to be analyzed via FACS and/or flow cytometry were first plated at a density of

lx106 cells in Coming 25cm2 flasks, 18-24 hours prior to the assay. The cells were allowed to

expand in 7.5ml of Gibco DMEM F-12 media, with the addition of 500pg/ml G4-18. Directly

before analysis, the cells were trypsinized and washed two times in a freshly prepared 0.05%

PBS/BSA solution. Cells were recovered by centrifugation at 1,000RPM for 4 minutes and

resuspended at a density of lx105 cells per 5001l of the PBS/BSA solution. Analysis was then

performed on the BD FACS ARIA for sorting, or the BD LSR II Flow Cytometer to determine

the number of cells expressing GFP, as well as the intensity of the fluorescence.

If antibodies were used to assay for the presence of specific cell surface antigens,

following the two initial washes, additional steps are as follows. (First, a solution of IgG was

used to block nonspecific binding using 401l / 1x106 cells for 15 minutes.) Immediately

following this, 101 of the desired specific phycoerythrin conjugated antibody was incubated

with an aliquot of the cells. Following two more 0.05% PBS/BSA solution washes, a 20 minute

incubation was performed.

The flow cytometer was set to record 10,000 events, at the same settings for all tests; FSC

Voltage: E-l, FSC Amp Gain: 868, SSC Voltage: 304, SSC Amp Gain: 1.0, FL1 Voltage: 412,

FL1 Amp Gain: 1.0, FL2 Voltage: 427, FL2 Amp Gain 1.0. Compensation was adjusted

accordingly for each sample, yet FL2 compensation consistently ranged from 4.0% FL1 to 12%

FL1 for optimal data acquirement.

Analysis of the data included setting the isotype or negative control at no more than a

3.0% error, in order to compensate for any autofluorescence or particulate matter that may have

been contaminating the sample. No gates were used to isolate any specific populations in the

analysis of the data, thus all recorded events have been included in the histograms.









Clonal Expansion of OS5210ct-4/GFP Cells

Cells from an OS521Oct-4/GFP tumor that were heterogeneous for GFP expression were

fractionated into enriched and depleted populations using FACS. The resulting populations were

then counted, serially diluted, and plated in separate 96 well dishes at a density of one cell per

well. Two hours post plating, allowing the cells enough time to adhere to the dish, the wells were

visually inspected to confirm the presence of only a single cell per well. Wells with multiple

cells or those without cells were discarded at that time. The cells were then visually checked

every 24 hours to monitor proliferation. Additionally, if cells began to form colonies in two

different areas in the well, they would be discarded, ensuring that clones used for

experimentation proliferated from a single cell. The cells were incubated in standard culture

media and allowed to expand for two weeks. As the populations reached confluence, they were

expanded to larger wells, until they could be maintained in 25cm2 flasks.

In vitro Osteogenic Differentiation

OS5210ct-4/GFP clones from monolayer were seeded at a density of 4x104 cells/cm2 in

24 well plates- in either standard culture media or in StemXVivo complete osteogenic base

media (R&D Systems). The active ingredients in this media includes P-glycerol-phosphate, L-

ascorbic acid, and dexomethasone. Samples were then cultured for 21-28 days, during which

time the cells were passage and analyzed for GFP expression by flow cytometry every 7 days.

Treatment with Bone Morphogenic Proteins

Two different methods were used to examine the effect of BMPs on Oct-4/GFP

expression and tumorigenicity of osteosarcoma tumor cells: adenovirus (Ad.) mediated

transduction of BMP 2, 4, or 7, or culture in conditioned media derived from transduced cells.

Adenovirus-mediated transduction of BMP 2, 4, or 7 into osteosarcoma tumor cells: For

adenovirus infection, cells were isolated by enzymatic dissociation of freshly harvested tumors









and cultured overnight. The following day cells were trypsinized and reseeded at a density of 104

cells/cm2 in 24 well dishes in standard culture media and cultured for an additional 12-16 hours.

The media were then aspirated, and the cells were washed 3 times with PBS. Cultures were then

transduced in Optimem for 4 hours with Ad.BMP 2, 4, or 7 at an estimated multiplicity of

infection of 10 infectious particles per cell. In control experiments using Ad.GFP this dose was

found to result in transduction of 90-100% of the cells as determined by flow cytometry.

Samples were then cultured for an additional five days. Cell media was collected from each

sample on days 2 and 5 to determine approximate BMP concentrations using commercially

available ELISAs according to the manufacturer's protocol (Quantikine, R&D Systems). The

respective BMP conditioned media were then stored at -200C for later use (see below). On day 7

post-transduction samples were trypsinized and the cells were counted. 1x105 cells were

transplanted into NOD/SCID mice and monitored as previously indicated for tumorigenicity

experiments. Aliquots of the samples were analyzed by flow cytometry for Oct-4/GFP

expression.

Treatment of osteosarcoma tumor cells in ith BMP 2, 4, or 7 conditioned media: Additional

aliquots of cells isolated from xenotransplanted tumors were initially seeded as described above,

but instead of transduction with adenovirus, they were incubated in the BMP conditioned media

from adenovirus infected tumor cell cultures, in a 1:1 ratio with standard culture media, for 7-10

days. BMP treated and untreated samples were then analyzed by flow cytometry for Oct-4/GFP

expression and transplanted into NOD/SCID mice (1X105 cells/mouse) and monitored for tumor

formation as before.









CHAPTER 3
LITERATURE REVIEW

The Biology of Cancer

Cancer is typified by a culmination of genotypic changes in otherwise normal functioning

cells that lead to malignant behavior. These genetic mutations can be characterized as either loss

or gain of function which then lead to observed cancer phenotypes. Loss of function occurs in

tumor suppressor genes, which normally function to repress proliferation, repair DNA damage,

or promote apoptosis. Gene silencing of tumor suppressors such as p53 is seen in a majority of

human cancers (Meyers et al. 2005) and often arises from epigenetic changes such as

hypermethylation of the CpG islands in the proximal promoter regions of the gene (Jones and

Baylin 2002). Proto-oncogenes are genes whose aberrant activity causes normal cells to become

cancerous either because they are mutated or expressed at the wrong time in development. The

gene products can cause increased cellular proliferation independent of extrinsic growth signals

as well as resistance to apoptotic mechanisms; these in turn drive cancer progression (Hanahan

and Weinberg 2000). Tumorigenesis generally does not occur after just a single mutation. It has

been suggested that several mutations, typically between four and seven rate-limiting stochastic

events, must occur in normal human cells for them to acquire tumorigenic properties (Gibbs et

al. 2002). According to Hananan and Weinberg (Hanahan and Weinberg 2000), the evolution

from a normal cell to a cancer cell is the culmination of at least six essential alterations in cell

physiology that collectively result in the capacity for malignant growth. These alterations

include: self sufficiency in growth signals, insensitivity to growth-inhibitory signals, evasion of

apoptosis, limitless replicative potential, sustained angiogenesis, and tissue invasion and

metastasis.









These are generalizations about changes that have been observed in all types of cancer.

Most cancer types have hallmark genes that have been mutated, and each of these changes results

in alterations in essential biological mechanisms. Understanding these mutations can change

cancer treatments into a rational science with practitioners aiming treatments a specific

molecular targets, thus improving patient comfort during treatment but more importantly

increasing patient survival rates (Hanahan and Weinberg 2000).

The Biology of Osteosarcoma

Osteosarcoma is a highly malignant mesenchymal tumor of bone in which the malignant

cells produce osteoid. It can arise in any bone, but occurs primarily in the juxta-epiphyseal

regions of rapidly growing long bones (Figure 5-1). The histopathologic appearance of high-

grade intramedullary osteosarcoma is one of malignant spindle cells producing osteoid and

immature bone. The bone structure is disorganized and can appear as a fine lacey trabecular

pattern or as irregular clumps of osteoid, distinctly unlike normal bone formation. Classic

osteosarcoma may also appear to be predominantly fibrous or chondroid with only small areas of

osteoid formation (Gibbs et al. 2002). Grossly, osteosarcoma begins as a process destructive of

medullary bone which progresses to destroy cortical bone, often with a large associated soft

tissue component. The natural history of osteosarcoma is one of relentless local progression with

loss of the function of the affected extremity and distant metastasis, most often to the lung(Chi et

al. 2004; Meyers et al. 2005).A small percentage of patients develop bone metastases which are

almost always fatal(Wuisman and Enneking 1990). Despite advances in the understanding of

cancer biology, the characterization of the events that lead to the development of conventional

osteosarcoma have yet to be defined. Difficulties impeding the molecular study of osteosarcoma

include obtaining living samples that can be decalcified, as well as samples that are viable after









aggressive chemotherapies. Thus far, unlike many other cancers, there have not been specific

translocations or chromosomal rearrangement events that can typify this malignancy.

Two genetic lesions that are often seen in osteosarcoma inactivate the p53 and

retinoblastoma (Rb) tumor suppressor genes; however, because of the variety of genetic

mutations involved in osteosarcoma, the relative contributions of these genes have yet to be

determined (Sandberg and Bridge 2003). Alterations of the RB1 gene have been shown in up to

70% of reported cases, and loss of heterozygosity for RB 1 has been shown to be a marker of

poor prognosis (Li et al. 2005). Overexpression of the oncogene MDM2 has also been implicated

as an important genetic alteration (Miller et al. 1996). Although this information has provided

insight into aspects of the molecular dysregulation of osteosarcoma and its heterogeneous nature,

to date these types of studies have been of limited value in establishing the molecular

determinants of tumorigenesis or in the development of effective therapies(Ragland et al. 2002).

Additionally, drug resistance, primarily to methotrexate, by osteosarcoma cells, only

increases the difficulty of treating patients. Two proposed reasons for drug resistance point to a

defect in the reduced folate carrier, or alterations and/ or amplifications of the dihydrofolate

reductase target. Likewise methotrexate resistance can stem from the functional loss of the RB

protein. Resistance to other chemo therapeutic drugs can also be attributed to overexpression of

p-glycoprotein, an ATP dependent transmembrane protein (Sandberg and Bridge 2003)

Stem Cells

Stem cells are considered to be a rare group of cells with the capacity to self-renew and

generate a developmental hierarchy of differentiating progeny. Pluripotent stem cells derived

from the inner cell mass (ICM) of the blastocyst expand and differentiate to ultimately generate

all the cells that comprise the adult individual. ES cells are derived from the ICM and retain the

widest developmental capacity. They are pluripotent, indicative of few epigenetic changes due to









methylation, allowing them to differentiate into ectodermal, mesodermal, and endodermal

lineages. Additionally, they maintain a stable karyotype and grow indefinitely in culture (Varga

and Wrana 2005).

At the time of embryonic differentiation, alleged permanent silencing of embryonic genes

is thought to occur via epigenetic modifications of their respective promoter regions. Histone

modifications and DNA methylation serve to block the reactivation of these genes and thus

prevent dedifferentiation of somatic cells (Hattori et al. 2004; Feldman et al. 2006).

Interestingly, several recent studies have challenged the apparent permanency of

embryonic gene silencing by demonstrating the ability of differentiated mouse and human skin

fibroblasts to be reprogrammed back to a pluripotent undifferentiated state (Meissner et al. 2007;

Okita et al. 2007; Wernig et al. 2007; Yu et al. 2007). In these studies, murine fibroblasts were

retrovirally transduced to express four transcription factors associated with pluripotency. Ectopic

expression of these proteins was sufficient to induce epigenetic reprogramming of the somatic

genome into an embryonic pluripotent state. The induced pluripotent stem (iPS) cells were

shown to form viable chimeras and generate live late-term embryos when injected into tetraploid

blastocysts (Meissner et al. 2007; Wernig et al. 2007). More recently, similar results have been

obtained using human dermal fibroblasts lentivirally transduced to express transcription factors

associated with pluripotency (Yu et al. 2007). Each of the generated iPS cell clones displayed

human ES cell morphology, normal karyotype, and expressed cell surface markers and genes

characteristic of human ES cells. All of the clones analyzed exhibited a demethylation pattern

similar to that of human ES cells. These cells also demonstrated the capacity to differentiate into

advanced cell types of each of the three primary germ layers (Okita et al. 2007; Yu et al. 2007).









These experiments indicate that the silencing of embryonic genes in adult cells may not be

permanent.

Adult stem cells generate mature, differentiated cells that form specific tissues (Eckfeldt

et al. 2005). Adult stem cells are characterized by their ability to self renew as well as the ability

to differentiate into mature cells of a particular lineage (Reya et al. 2001). These characteristics

allow these cells to maintain their population, while also allowing a more specialized response to

enable repair of injured tissues.

In vivo, stem cells reside in a "niche," which is a specific location in a tissue where the

cells can reside for an indefinite period of time and produce progeny cells while self-renewing

(Ohlstein et al. 2004). Cellular division within the niche can take two different directions and

must be carefully regulated. Too little proliferation may lead to a depletion of the necessary stem

cell population; yet, it is possible for unchecked proliferation to result in tumorigenesis.

Normally, self renewal or symmetric division occurs when a stem cell divides into two identical

daughter cells. These cells remain in the niche and serve as functioning stem cells, maintaining

pluripotency and sustaining the population. Asymmetric division occurs where two daughter

cells are formed, but one cell stays within the niche to act as a stem cell, and the other cell leaves

the niche in order to differentiate and proliferate into additional progeny. Due to different

physiologic conditions, both symmetric and asymmetric division can occur within two different

cells residing in the same niche (Yin and Li 2006).

Stem cells are often identified through in vitro studies, where cell surface/cytoplasmic

proteins, transcription factors, and proliferative behaviors are often the defining features

observed. In these studies, the natural in vivo environments are mimicked as closely as possible

through exposure to important growth factors and morphogenetic molecules. Yet, these attempts









at reproducing the in vivo environment fail to include many important events and interactions.

Cellular communication through cytokines, nonphysiological amounts of growth factors, and

morphogens can have potent effects on stem cell behavior not typically observed in vivo

(Steindler 2007).

Cancer Stem Cell Theory

The cancer stem cell theory holds that there is a sub-population of cells within a tumor

which, like normal stem cells, has the ability to self-renew. These cells can divide

asymmetrically, producing an identical daughter stem-like cell and a more differentiated cell

which upon subsequent divisions generates the vast majority of the tumor bulk, which is

essentially benign. This stem-like cell is responsible for initiating and maintaining the growth of

the tumor and if not completely eradicated by surgical extirpation or chemotherapy is responsible

for local and distant recurrence (Figure 5-2) (Pardal et al. 2003). Along these lines, Weismann,

drawing parallels between cancer stem cells and normal stem cells, has suggested that

tumorigenesis can be viewed as aberrant organogenesis (Reya et al. 2001)

The first definitive work describing a cancer stem cell was performed by John Dick and

colleagues in studies of acute myeloid leukemia (AML) (Bonnet and Dick 1997). They identified

a rare population of human SCID leukemia initiating cells that were able to propagate AML in a

xenograft transplant system. The leukemic grafts generated were representative of the patients'

original disease phenotype. They demonstrated that the human AML stem cells purified from

patient samples were CD34+ CD38-' resembling the normal hematopoietic stem cell phenotype.

Cells from the CD34+CD38+ fraction could not transfer the disease despite having a leukemic

blast phenotype. This suggested that the normal hematopoietic stem cell was the target of

leukemic transformation. Others have subsequently implicated stem-like cells in the









pathogenesis of brain and breast malignancies suggesting a broader involvement of stem cells in

carcinogenesis (Ignatova et al. 2002; Al-Hajj et al. 2003; Hemmati et al. 2003; Galli et al. 2004).

It has been suggested that cancer is a disease of unregulated self-renewal in which

abnormal stem cells utilize the machinery of self-renewal to drive neoplastic proliferation (Pardal

et al. 2003). That cancer could arise from a primitive stem-like cell or other precursor seems

reasonable as it would require far fewer genetic or epigenetic alterations to effect a malignant

change in a cell already equipped with the capacity for self-renewal. Several of the genes shown

to play a role in the regulation of normal stem cell self renewal (WNT, Sonic Hedgehog, Notch)

have been found to be active in cancer (Jhappan et al. 1992; van de Wetering et al. 2002; Lessard

and Sauvageau 2003; Pasca di Magliano and Hebrok 2003; Qiang et al. 2003).

Use of Agents That Induce Differentiation for the Treatment of Cancer

Several lines of evidence suggest that aberrant expression of the key regulatory proteins

of ES cell pluripotency can directly contribute to tumorigenesis in several cell types. Therefore,

agents that serve to inhibit the activity, or alternatively block the expression of these proteins

could be beneficial in the treatment of cancer. Along these lines alterations in the differentiation

programs of cancer cells often result in changes to their phenotype related to survival, rate of

growth, loss of differentiation and the ability to proliferate and invade surrounding tissue

(Hanahan and Weinberg 2000)Anaplasia, the loss of differentiation, is associated with aggressive

clinical behavior, suggesting differentiation confers a restraint on tumorigenesis (Thomas and

Kansara 2006). In fact, apart from the presence of metastatic disease, the degree of

differentiation of a sarcoma is the most powerful negative prognostic indicator. Interestingly,

differentiation therapy is already in clinical use. The most striking example may be the use of all-

trans-retinoic acid (ATRA) in acute promyelocytic leukemia, felt my most to be a stem cell









malignancy. 90% of patients can expect remission of their disease with combination therapy

consisting of both conventional chemotherapy and ATRA (Ohnishi 2007).

The role of BMP signaling in cellular biology is far reaching and diverse. One of these

roles lies in determining cell fate choices during differentiation. Data from work with BMP's in

mouse ES cells shows that their role sharply contrasts with that of analogous human cells, where

neural differentiation is blocked by BMPs in mice, yet induced in humans (Varga and Wrana

2005).

BMPs are secreted proteins and can direct mesenchymal stem cells (MSCs) to

chondrogenic and osteogenic cell lineages, and in the presence of fibroblast growth factors, they

can direct ES cells to differentiate into the trophoblast lineage in humans. BMPs are part of the

TGF-P superfamily, and transmit signals through a defined pathway. The BMP binds tightly to

BMPI/II receptor heterodimer upon which the type II receptor phosphorylates the type I receptor.

In the cytoplasm, the downstream effect of BMP binding causes phosphorylation of BMP R-

Smads, specifically 1, 5, and 8, which complex with Smad4. The heteromeric complexes then

translocate to the nucleus where they can regulate transcription either directly or in concert with

other transcription factors (Varga and Wrana 2005).

BMPs may have an important role in controlling the biology of stem cell cancers as they

play a crucial role in early stem cell development, as well as self-renewal; losing the ability to

regulate these functions may lead to tumorigenesis (Varga and Wrana 2005). In support of this

are studies showing that mutation of the BMP I receptor is a huge risk factor in developing

gastrointestinal cancers. Expanding the potential of differentiation therapy beyond leukemias, a

recent report by Piccirillo et al. (Piccirillo and Vescovi 2006)showed that stem-like, tumor-

initiating cells isolated from human glioblastomas when incubated with certain BMPs increased









the expression of markers of neural differentiation, and showed decreased proliferation and

tumor formation when transplanted into experimental animals. Their work implies that certain

populations of tumor stem cells retain an ability to respond to normal signals of maturation

induction, and efforts to devise therapies to differentiate cancer cells might be fruitful.

Histone deacetylases (HDAC) play a major role in the epigenetic changes which regulate

gene expression within a cell. HDACs catalyze the removal of acetyl groups, and thus stimulate

chromatin condensation and promote transcriptional repression. Transcriptional repression can

translate to the loss expression of tumor suppressor genes, contributing to the formation of

cancer. HDAC over-expression has been observed in colon, breast, prostate, and other cancers.

Due to their wide-spread involvement in cancer, HDACs have become a novel target for therapy.

HDAC inhibitors, which include a variety of therapeutic agents, can lead to the reversal of

epigenetic silencing. HDAC inhibitors come in the form of short chain fatty acids, hydroxamic

acids, benzamides, and cyclic tetrapeptides. The actions of these drugs affect cellular processes

such as inducing cell cycle arrest, stimulating tumor cell death, and promoting differentiation.

HDAC inhibitors induce cellular differentiation and promote cell cycle arrest at the G1/S

checkpoint. Clinical trials have shown that HDAC inhibitors have anti cancer activity and are

being tested as either monotherapies or in combination with chemotherapy (Carew et al. 2008).

Preliminary Results

Cells Isolated from Bone Sarcoma Cultures Exhibit Stem-Like Attributes

We initiated a preliminary series of experiments to explore the existence of stem-like

cells in bone sarcomas. The culture system originally used by Reynolds and Weiss (Reynolds

and Weiss 1992) to generate neurosphere clones from adult mammalian brain, has since been

found to isolate cells possessing attributes of stem and progenitor cells. The stressful growth

conditions of this system were found to positively select for primitive cells by eliminating the









differentiated cells, which are unable to survive. Similarly, suspending dissociated cancerous

tissue in semi-solid media without serum selects primitive clonogenic cells that can be expanded

and give rise to different classes of cells. This system enables isolation of stem-like cells from

malignant human brain tumors by exploiting anchorage independence, serum starvation and

necessary pleiotropic growth factors(Ignatova et al. 2002). Similar sphere culture systems have

been used to identify tumor stem cells from both brain and breast malignancies that are capable

of self-renewal in mouse models (Al-Hajj et al. 2003; Hemmati et al. 2003; Singh et al. 2003;

Singh et al. 2004).

We found that all bone sarcoma cultures formed spherical colonies ("sarcospheres") at a

frequency of 102 10-3 similar to that reported by others for brain and breast

malignancies(Ignatova et al. 2002; Al-Hajj et al. 2003; Hemmati et al. 2003). Sarcospheres were

also generated at a similar frequency from fresh tumor dissociates produced at the time of

biopsy. Furthermore, all cultures tested demonstrated the capacity for self-renewal by the

formation of secondary spheres at a similar or increased frequency of approximately 102.

We also examined these cultures for expression of Oct-4 and Nanog transcription factors

found to be indispensable for the maintenance of pluripotency and self-renewal in ES cells. Using

semi-quantitative RT-PCR we found detectable transcripts in each culture type for both

transcription factors (Figure 5-3).

Immunohistochemical staining of sarcospheres from paraffin sections enabled detection

of both Nanog and Oct-4 in similar patterns. Interestingly, the staining of the smaller spheres

showed a high proportion of cells that were positive for both transcription factors. As the spheres

increased in size and cell number, the cells showed increasingly greater heterogeneity of

immunostaining with a lesser percentage expressing these ES cell markers. These observations









suggested a situation whereby during early sphere formation relatively primitive cells undergo a

transition from symmetric to asymmetric division and thereupon produce daughter cells of

various states of differentiation.

Following these observations, we addressed whether Oct-4 and Nanog are expressed in

actual tumor tissue. For this, paraffin sections from eight bone sarcoma patients were evaluated

using immunohistochemistry (Figure 5-4). Nanog and Oct-4 nuclear staining was observed in

seven of the eight tumors studied. In each case, as determined by histologic criteria, the stained

nuclei were of malignant cells, and not from infiltrating normal cells. Among different tumor

specimens, the number of Oct-4 and Nanog positive cells varied considerably. Positive Oct-4

staining ranged from a few percent of the cells in some tumors to up to 25% in others. Nanog

staining was also quite variable ranging from -1% to nearly 50% in certain samples. These

results showed that subpopulations of cells in bone sarcomas express regulatory proteins

typically restricted specifically to embryonal cells.

We reasoned that if bone sarcomas express some of the molecular machinery of ES cells,

they might, in addition to mesodermal genes, express genes from endodermal and ectodermal

lineages. RT-PCR analyses of mRNA from adherent and sarcosphere cultures revealed

expression of Gata-4, Gata-6 and alpha fetoprotein (AFP), indicative of endodermal

differentiation (Figure 5-5a). Expression of P-III tubulin RNA was seen as well, which is

believed to be a marker of neural-ectoderm, but has also been demonstrated in some poorly

differentiated malignancies (Katsetos et al. 2003). Using Western blot analyses, expression of

AFP and P-III tubulin was demonstrated at the protein level in each of seven cultures tested

(Figure 5-5b). P-III tubulin was also detected in paraffin sections of tumors using

immunohistochemistry and in cell culture samples by immunocytochemistry (Figure 5-5c-f). At









this point, we do not know the position that a putative bone-sarcoma stem cell might occupy in

the stem cell hierarchy. It is possible that the expression of genes such as Gata-4 and Gata-6 is

simply a consequence of a larger global pattern of dysregulated gene expression in these tumor

cells. An alternative explanation is that these genes are indicative of aberrant pluripotent

differentiation of cancer stem cells. Regardless, the detection of expression of ectodermal and

endodermal genes implies that the tumor cells are not effectively lineage restricted and suggest

epigenetic reorganization of the genome in bone sarcomas.

Although the bone sarcomas we have studied appear to express transcription factors

associated with pluripotent ES cells as well as genes of endodermal and ectodermal lineages, the

histologic phenotype of these tumors is by definition one of arrested mesenchymal

differentiation. Thus, it seems reasonable to expect that a cancer stem cell in bone sarcoma

would arise from a mutagenic lesion in a mesenchymal progenitor. Therefore, we examined the

respective cultures for the presence of cells with characteristics of mesenchymal stem cells.

Although the precise cell-surface phenotype of an MSC has not been determined, MSCs have

been found to variously express the cell surface proteins Stro-1, CD90, CD29, CD73 CD44 and

CD105(Simmons et al. 1994; Stewart et al. 1999; Li et al. 2005). Using immunocytochemistry

and flow cytometry, we screened our tumor cell cultures and found that all the cultures expressed

these antigens, but to different degrees. Although differences in the intensity of

immunoreactivity of specific surface antigens were seen between cell cultures, within any

particular culture the cells appeared largely homogenous for expression of these antigens.

Unfortunately the apparent uniformity of the cultures with regard to expression of surface

antigens precluded fractionation of the cells based on surface antigen, and thus could not be used

strategically as a potential method to explore markers of tumorigenic cells. To determine if cells









within the cultures were indeed multipotent, we attempted to induce adherent cultures of bone

sarcoma cells to differentiate along two distinct mesenchymal lineages by culture in osteogenic

and adipogenic media. Within each adherent culture tested, we observed discrete foci of

mineralization in cells grown in osteogenic medium, and fields of lipid laden cells in those

grown in adipogenic medium (Figure 5-6).

Cumulatively, the data from the previous sections suggested that at least subpopulations

of cells in bone sarcomas were capable of self-renewal, expressed transcription factors of ES

cells as well as expression of ecto- and endodermal genes, and showed the capacity for

differentiation along multiple mesenchymal lineages. Altogether these data support the

involvement of stem like cells in bone sarcomas.

Development of an In vivo Model to Examine the Role of Stem-Like cells in the
Pathogenesis of Osteosarcoma

To explore in more detail the functional relationship between "stemness" and

tumorigenicity of osteosarcoma, we focused our studies on the OS521 line, derived from a high

grade, poorly differentiated human osteosarcoma. This cell line was found to cause robust tumor

formation following subcutaneous xenograft into the backs of NOD/SCID mice. In initial

experiments we found that delivery of as few as 3 x 104 cells in saline suspension reproducibly

produced tumors of >1 cm diameter in 4-6 weeks following injection. Similar to that observed

from the biopsy of the original patient, tumors arising from the xenograft showed clear evidence

of osteoid, recapitulating the characteristic phenotype of an osteosarcoma. Characterization of

the OS521 culture for several cell surface markers showed that the cells in monolayer were

comprised of a largely homogenous population without striking differences with regard to cell

surface antigens. The cells were MHC class I+, CD90+, and NCAM+. Interestingly they were

uniformly strongly positive for expression of CD44, a marker of breast cancer stem cells, and









negative for the presence of CD133 (Figure 5-8b) a marker of stem cells in colon, brain and

prostate cancer.

Expression of an Oct-4 Promoter/GFP Reporter Construct That Selectively Identifies
Cancer Stem Cells in Osteosarcoma

In an effort to selectively visualize and track living cells in culture that express ES cell-

specific genes and determine their relative participation in tumorigenesis, we transfected the

OS521 cells in monolayer with the plasmid construct phOct-4/GFP, (a generous gift from Dr.

Wei Cui of the Rosalin Institute, UK) containing the human Oct-4 promoter sequence (spanning

-3917 to +55, relative to the transcription start site) linked to the coding sequence for enhanced

green fluorescent protein (Figure 5-7). This plasmid also contains an independent SV40-

promoter driven neomycin resistance cassette, which allows positive selection of cells that have

acquired the plasmid construct, irrespective of their capacity to transcribe the Oct-4 promoter.

Thus in medium supplemented with G418 all surviving cells will contain the plasmid and

express the neomycin resistance gene; however, only the subset of cells that are capable of

activating the Oct-4 promoter sequence will fluoresce green. Gerrard et al.(Gerrard et al.

2005)showed that in human ES cells stably transfected with this construct, GFP expression

driven by the Oct-4 promoter faithfully represented expression of Oct-4 in undifferentiated ES

cells and during their differentiation. In these studies, GFP expression co-localized with

endogenous Oct-4 protein as well as surface antigens SSEA-4 and Tra-1-60. Neural

differentiation of the cells, as well as targeted knockdown of endogenous Oct-4 expression by

RNAi, down-regulated GFP and correlated closely with the reduction in endogenous Oct-4

protein.

Following transfection of the OS521 cells with phOct-4/GFP and positive selection of

transfectants in media containing G418 we characterized Oct-4 driven-GFP expression of cells in









monolayer using fluorescence microscopy and flow cytometry. Somewhat surprisingly, despite

the apparent homogeneity of the cells in culture with regard to cell surface proteins, we found

that following stable transfection of the OS521 line only about 24% of the G418 resistant cells

were GFP+ (Figure 5-8a). To begin to determine the relative participation of the GFP+ cells

(those that transcribe the Oct-4 promoter) in tumor initiation, we injected 3 x 104 cells from the

total neo resistant cell population (all cells both GFP positive and negative) subcutaneously into

the backs of 6 NOD/SCID mice; the same dose of untransfected OS521 cells were injected into a

separate group of 6 animals. At 3-5 weeks post-injection, tumors >0.5 cm diameter had formed

in 5/6 and 4/6 animals of the two respective groups, which indicated that transfection of the

phOct-4/GFP plasmid did not adversely influence the tumor-initiating potential of the OS521

cells. Tumors that were > 1.0 cm diameter were harvested. Tumors from animals receiving the

phOct-4/GFP transfected cells were brightly fluorescent under UV light, while tumors from

animals receiving untransfected OS521 cells showed no evidence of GFP expression (not

shown). Histologic section showed large clusters or foci of GFP+ cells distributed throughout the

tumor mass (Figure 5-8a). Following harvest, the tumors were dissociated, and the cells

recovered were characterized for GFP expression by flow cytometry. The proportion of GFP+

cells isolated from the tumors had increased to -67%, nearly 3-fold over that observed in

monolayer culture. There was no apparent change, however, in the expression of the various

surface antigens with respect to the GFP positive and negative cell populations (Figure 5-8b).

Altogether, these results suggested a selective amplification of cells that express the Oct-4

promoter during tumorigenesis.

To determine the relative tumor initiating capacity of the cells that expressed the Oct-4

promoter construct versus those that did not (i.e. GFP+ cells vs. GFP- cells, respectively), the









cells recovered from the harvested tumors were pooled and fractionated by FACS into GFP-

enriched and GFP-depleted populations, as shown in Figure 5-9a-c. Subsequent flow cytometry

analysis of a portion of the respective fractions showed that in the enriched fraction -92% of the

cells were GFP+; in the GFP-depleted fraction the number of GFP+ cells was reduced to about

3%. From a starting dose of 3 x 104 cells, we injected ten-fold dilutions of the respective

fractions, as well as equivalent numbers of unfractionated, phOct-4/GFP transfected cells, into

individual groups of NOD/SCID mice at 8 animals/group and examined the rate of tumor

formation.

The GFP-enriched fraction proved to be significantly more tumorigenic (>100-fold) than the

GFP-depleted fraction (Figure 5-9d). At the 3 x104 cell dose it produced tumors in all mice with

a mean time to onset of 22 days, while the GFP-depleted fraction only produced tumors in -60%

of the mice with a mean time to onset of over 6 weeks. At 3 x 103 cells, again, all 8 of the

animals receiving cells from the GFP-enriched fraction developed tumors, with a mean time to

onset of 34 days. For the GFP-depleted group, only 1 of 8 mice developed a tumor over the 90-

day time course. At the 3 x 102 cell dose none of the mice from the GFP-depleted group

developed tumors, while all of the animals receiving the GFP-enriched cells developed tumors,

with a mean time to onset of 42 days. At this dose only 3 of 8 animals receiving unsorted/phOct-

4/GFP transfected cells formed tumors, with a mean time to onset of -60 days. Visualization of

the freshly excised tumors using inverted fluorescence microscopy showed that all tumors

formed in all groups were highly GFP+. Following dissociation, flow cytometric analysis of the

recovered cells showed that tumors formed from the GFP-enriched fractions were comprised of

-70-80% GFP+ cells, while those from the GFP-depleted fractions were -50-55% GFP+ (not

shown).









Interestingly, we passage the GFP+ cells through at least 3 rounds in mice, whereby the

cells were injected, harvested from tumors, enriched and reinjected. At the 300 cell dose (-the

lowest dose that we can reasonably expect to deliver) we found that the tumors appeared to

increase in virulence with passage, producing tumors with shorter time to onset and more rapid

growth rate. We also noted the formation of multiple local tumor nodules following a single

injection. Analysis of the lungs of these mice using inverted fluorescence microscopy showed

clear evidence of metastases, with clusters of GFP+ cells readily identified throughout. These

data demonstrated that the cells that expressed the Oct-4 promoter construct displayed self-

renewal in vivo and further supported the participation of a cancer stem cell in tumorigenesis of

osteosarcoma.









CHAPTER 4
RESULTS

Rationale for Aim 1

Based on previous data, we found that our exogenous Oct-4/GFP reporter could identify a

subpopulation of cells with enhanced tumor forming capacity. These cells were capable of

activating the Oct-4 promoter whose gene product is responsible for maintaining a pluri-potency

in ES cells, hinting to the existence of a stem like tumor cell in human osteosarcoma. These cells

were 100 fold more tumorigenic than those not expressing the Oct-4 promoter, and the resulting

tumors were heterogeneous in nature for GFP expression. We questioned whether this

heterogeneity was due to existing differences in the proliferation rates of the GFP+/-

subpopulations, since fractions from FACS contained up to 8.0% cross contamination (Figures 5-

9b, 5-9c), or if the differences represented changes in the ability of certain cells to express the

Oct4 promoter. We aimed to investigate this by creating clonal populations of the respective

phenotypes and repeating the tumorigenicity assays. Clonal expansion would create a population

of cells that were homogeneous in nature, allowing any changes in Oct4/GFP expression to be

attributed to changes in the biology of the progeny cells and not to selective amplification of

GFP+ and GFP- subpopulations.

We reasoned that in vivo, tumor initiation occurs from GFP+ cells, which proliferate and

differentiate to form cells that have lost Oct-4 expression as well as their tumorigenic potential.

These GFP+ cells might be thought of as moving away from the stem like phenotype and

becoming more specialized, as the transform from GFP+ to GFP-. Along these lines, we would

not expect to see cells lacking the expression of the Oct-4 promoter to spontaneously gain the

ability to express the gene either in vivo or in vitro, because this would signify an act of

dedifferentiation towards a more stem like state. Due to epigenetic silencing, this is usually an









irreversible phenomenon. Using cells of a pure clonal population would allow us to observe any

shifts in Oct4/GFP expression and to determine if the GFP- cell populations arise from GFP+

populations and not vice versa. Additionally, it would show that the appearance of a GFP-

population was a direct result of biological changes in the tumorigenic cell and not a shift in the

dynamics of the population due to proliferative differences.

Rationale for Aim 2

If in fact, cancer stem cells in osteosarcoma rely on the molecular machinery of ES cells

that function to maintain pluripotency, then it would be rational to expect that the forced

induction of cellular differentiation in these cells should either reduce the expression of these

proteins, or alternatively modulate their effects, and in so doing reduce tumorigenicity. Along

these lines, Feinberg and others have suggested that cancer stem cells from certain types of

tumors retain the ability to respond to extrinsic differentiation signals (Jones and Baylin 2002).

In exploring the effects of differentiation agents on osteosarcoma stem cells, we

examined if exposure of OS521Oct-4/GFP clones in monolayer to osteogenic differentiation

media would stimulate cellular differentiation and change the expression of osteogenic-

associated genes in these cells. If we could associate an increase in osteogenic associated gene

transcription with a lack of expression of Oct-4, then we could assume that loss the stem like

phenotype has occurred along with the differentiation event. This would be important in linking

the lack of Oct-4 expression with a cell that has experienced differentiation.

We also chose to use BMPs 2, 4 and 7, which are well known inducers of osteogenic

differentiation in human mesenchymal stem cells, to perform tumorigenicity assays with. If

treatment with these agents induces a loss of Oct-4 expression via differentiation, resulting in a

loss of tumorigenic capacity, we could assume that the tumor initiating potential of osteosarcoma

depends on the retention of a stem like phenotype.









If we find that any of these differentiating agents decrease tumorigenicity, these data

could form the platform for the development of novel therapies for this devastating disease.

Clonal Expansion of OS5210ct-4/GFP cells

In order to investigate our hypothesis that heterogeneity can be derived from a clonal

population, we had to first create clones from our patient derived OS521 cell line. After

completing this task, we observed the following characteristics of the different clones.

For the GFP+ fractions, we observed that 90% of the 96 wells held viable cells following

serial dilutions and allowing cellular adherence. Any wells that were observed to contain more or

less than one cell were immediately discarded. Over the course of two weeks, all of the GFP+

cells began to divide, yet three of the clones displayed more aggressive proliferation and had

reached confluency. These clones, named Al, S1, and T1, were allowed to expand for

experimental purposes. At six weeks the S1 and T1 clones had divided enough to populate a

25cm2 flask. The Al clone inexplicably stopped dividing and expired four weeks into the

expansion process. Fluorescence microscopy and flow cytometry were used to confirm the

homogeneous expression of GFP in both clones. Figure 5-10a shows that the S1 clonal

population in monolayer was 98% positive for GFP expression, which was also comparable to

that seen in the Tlpopulation in culture.

Likewise, 90% of the GFP- population survived initial plating, yet after two weeks the

cells had failed to achieve proficient division rates. Additional time was allotted for the

expansion of the GFP- clones, but none of the cells succeeded and eventually died by 8 weeks.

Fluorescent microscopy confirmed the absence of GFP in any of the cells, but there were not

enough cells in any of the wells to perform a flow cytometry assay.









Flow Cytometric Analysis of Clones and Clonally Derived Tumors

We wanted to determine if the tumors formed from our largely homogeneous clonal

populations would recapitulate the heterogeneity observed from the parental OS521 lines. This

would allude to the ability of the cells to lose the expression of Oct-4 and proceed towards a

more differentiated phenotype.

Following delivery of 3 x 104 cells of the respective clones into NOD/SCID mice, tumors

readily formed in animals within 2-3 weeks. Analysis by flow cytometry showed that the cells

recovered from the tumors were markedly heterogeneous with respect to GFP expression, as

shown in Figure 5-10b, with a clear reduction in the mean intensity of fluorescence. On first

passage (Figure 5-10b), tumors exhibited fluorescence over a range of three logarithmic orders.

The histogram shows a clear shift to the left in fluorescence intensity compared to that of the

parental in vitro population, though there is no clear distinction between the two populations in

the histogram. Interestingly, serial passage of unsorted cells from freshly dissociated tumors

derived from these clones resulted in the production of discrete GFP+ and GFP- populations.

They were composed of approximately 60% GFP+ and 40%GFP- by the third passage (Figure 5-

10c). The GFP+ positive fraction more closely resembled that of the parent clonal population,

with the majority of the cells having a fluorescence intensity greater than 104.

We questioned whether these more discrete GFP+ and GFP- populations might also be

distinct with regards to tumorigenic capacity. To this end we harvested passage three tumors,

which are composed of the most segregated populations, from one of the clones (S ) and

fractionated the cells by FACS into GFP enriched and depleted populations. We then

transplanted them as previously described for tumorigenicity experiments. We observed results

similar to those shown for our parental cultures/tumors using these clonally derived populations

regarding the frequency and kinetics of tumor formation by the two fractions at the indicated









doses (compare Figure 5-9d and 5-1 Ib). Analysis of the resulting tumor cell populations for Oct-

4/GFP expression by flow cytometry showed that GFP enriched fractions formed tumors

composed of 85% GFP+ cells, and GFP depleted fractions formed tumors composed 55% GFP+

cells (Figure 5-lic).

The data to this point show that Oct-4/GFP+ and Oct-4/GFP- cells generate

heterogeneous tumors upon transplant, and are capable of extensive self-renewal in vivo: all of

which are traits associated with cancer stem cells. However, in this instance the cancer stem cell

is not a rare, slowly dividing stem-like cell but a highly proliferative cell whose active cell

division directly supports the growth of the tumor.

Culture of OS5210ct-4/GFP Clones in Osteogenic Media

To attempt to link differentiation to the loss of the expression of Oct-4, we took

OS521Oct-4/GFP clones from monolayer and attempted to stimulate differentiation with

commercially available osteogenic media for a period of 21 days. Culturing of human ESCs and

MSCs in these conditions has been shown to result in the activation of genes associated with

osteogenic differentiation and matrix deposition (Gronthos and Simmons 1995). We believe that

a loss of Oct-4 expression during the incubation would show that Oct-4/GFP- cells are indeed

more differentiated than their fluorescent counterparts.

In order to prevent "rolling up" of the confluent monolayers, the cultures were

trypsinized and reseeded every 7 days to remove dead cells and for analysis by flow cytometry.

We were unable to detect significant matrix deposition or changes in the expression of genes

associated with osteogenesis in samples cultured in osteogenic media by staining of the confluent

monolayers with Alizarin Red S and RT-PCR, respectively (not shown). As shown in Figure 5-

12, OS521Oct-4/GFP clones cultured in osteogenic media for 14 and 21 days were composed of

both GFP+ and GFP- cells. On day 14 there were 57% GFP+ and 43% GFP- cells, and by day 21









the population was composed of 64% GFP+ and 46% GFP- cells. Controls grown under standard

culture conditions remained unchanged for GFP expression (95%). These cells were not used in

tumorigenicity assays.

Culture of OS5210ct-4/GFP Clonally Derived Tumor Cells in BMP Conditioned Media

We hypothesized that inducing differentiation through various agents, namely BMPs,

would adversely affect the tumorigenic capacity of the OS521 clonally derived tumors. We

attempted to stimulate differentiation using either adenovirus transduction or exposure of the

cells to BMP treated media. Subsequent injection of the treated cells yielded the following.

BMP conditioned media was generated by infection of cells from tumors derived from

clone S1, with adenovirus encoding the cDNAs for either BMP 4, or 7 as described. BMP 2

appeared to be toxic to the cells and they did not survive the transduction; as a result this portion

was omitted from the experiment. Analysis by ELISA for BMP concentrations of cell free

supernatants from both infections showed high concentrations of BMP protein (200>ng/ml).

Approximately 3 days following the start of induction we first noted by visual observation that

BMP-treated cultures adopted a flatter cellular morphology and took longer to grow to

confluence than untreated controls. Unsorted tumor cell populations cultured in BMP 4 or -7

conditioned media for 7-10 days also resulted in alterations in the proportion of the GFP+ and

GFP- fractions, with a clear decrease in mean fluorescence intensity (compare 13A and B).

Interestingly this effect was more pronounced in the BMP 4 treated samples.

Tumors formed by transplantation of these cells (1.0X105 cells/mouse) also showed

differences in the relative proportions of GFP+ and GFP- cell fractions (Figure 5-13c). We

observed that for the BMP 4 treated sample, 10% of the cells were GFP+ following

transplantation. This result is in stark contrast to the BMP 7 treated sample, where 70% of the

cells were GFP+. This result contradicted what we had consistently observed for untreated









samples following transplantation; the GFP+ cell fraction usually expands or remains constant

(compare Figures 5-13c and 5-1 Ic). Time to tumor onset, penetrance and kinetics were all

comparable to tumors formed by untreated controls.









CHAPTER 5
DISCUSSION

Our group has identified a putative CSC population in osteosarcoma using an Oct-4

promoter driven GFP reporter. We observed that the GFP enriched fraction of xenotransplanted

tumor cells was roughly 100-fold more tumorigenic than the GFP depleted fraction when

transplanted subcutaneously into NOD/SCID mice. We believe that silencing of the Oct-4/GFP

transgene and concomitant loss of tumorigenic ability is the result of differentiation/proliferation

of these cells during tumor initiation and growth.

In the present study we have focused on two distinct aims. We first showed that

individual Oct-4/GFP+ clones were capable of generating heterogeneous tumors composed of

both GFP+ and GFP- cells. This confirms that this cell retains a stem cell like phenotype and

possesses key stem cell attributes, such as the ability to self renew and recapitulate heterogeneity

of the initial population. Additionally it supports our hypothesis that changes observed in our

experiments were due to biological effects and not the heterogeneous nature of the collective

population. Secondly, we cultured cells from these tumors in osteogenic media or in the presence

of BMP 4 or 7 and found that we were able to induce silencing of the Oct-4 promoter. We then

took these treated cells and performed tumorigenicity assays. At a dose of 105 unsorted cells

tumors formed within 4 weeks however there were distinct differences between tumors formed

by BMP 4 treated samples compared to samples exposed to BMP 7 regarding days to tumor

onset and the proportions of GFP+ and GFP- cells fractions.

In order to obtain a truly homogeneous population to which variation could be attributed

to biological changes, we clonally expanded the cells from OS521 Oct-4/GFP cell line.

Successful expansion of the population expressing the Oct-4 gene product can be attributed to

the fact that these cells maintain their pluripotency in vitro and thus their stem like qualities. The









ability of stem cell like cancer cells to divide both symmetrically and asymmetrically can explain

the expansion and restoration of the cell population. Symmetric division, which is what we saw

in our clones in vitro, contributed to the homogeneous nature of the clonal population. This

makes sense since the cells still expressing Oct-4 maintain an ES cell like phenotype. We did not

observe any asymmetric division in vitro, leading to more differentiated cells, until the cells were

exposed to the in vivo environment in our tumorigenicity assays. The in vitro environment

apparently lacks the necessary cues to induce change or gene silencing in the population.

The lack of Oct-4 expression in the GFP- cells allude to the reason why they cells were

not able to clonally expand. Loss of Oct-4 expression points directly to a cell which no longer

transcribes the Oct-4 gene, which is crucial in maintaining pluripotency in ES cells. Along these

lines, a cell that is no longer behaving as a stem cell would be unable to divide in the manner

required to reinstate a stable self maintaining population.

Phenotypic characteristics of the clones that were established in vitro showed that they

were 98% positive for GFP expression. Upon multiple passages of these cells in vitro, the

relative homogeneity remained at the highly uniform levels. Loss of Oct-4 promoter activity was

not observed in our cultures, which we attribute to the absence of the proper cues that a cell

would normally be exposed to in an in vivo environment. Basic media and cell/cell contacts do

not provide stimulation for the cells to differentiate or initiate any of the associated epigenetic

changes, and they symmetrically divide to maintain their population and existence.

Changes in GFP expression can be induced through serial passage in vivo in the

NOD/SCID mouse model. Tumor progression is often viewed as driven by epigenetic plasticity

and genomic instability. Epigenetic heterogeneity within stem and progenitor populations could

in part account for tumor cell heterogeneity. From these observations we postulate that one









possible explanation for our results is that Oct-4/GFP activity identifies tumor cells with a stem-

like epigenetic signature with the capacity for infinite self-renewal. Loss of Oct-4/GFP

expression may be the result of differentiation of the cancer stem cell population during tumor

growth and then an adoption of a more differentiated epigenetic signature. This leads to a loss of

stem-associated activities and the ability to initiate and sustain tumorigenesis in vivo.

As the number of serial passages increased, we observed that the gap between GFP+ and

GFP- populations increased as well (Our results clearly show that from initial injection and a

nearly pure GFP+ cell population, we can form tumors that are comprised of a substantial

number of GFP- cells.) We believe that this phenomenon occurs when the tumor initiating GFP+

cells are exposed to the in vivo micro-environment providing contact with the extracellular

matrix, cellular cues of proliferation, and growth and proliferation factors. These cells are now

stimulated to divide asymmetrically, where changes induce a loss of Oct-4/GFP expression, and

thus increase the number of GFP- cells. These differentiated cells accumulate and contribute to

the bulk of the tumor mass, while the less differentiated GFP+ cells maintain their primitive stem

like qualities and their tumorigenic potential. Additionally, these changes cause the GFP- cells to

lose their tumor initiating potential. With passage in vivo, the clonally expanded Oct-4/GFP+

cells diverge, eventually adopting a profile of GFP expression similar to the parental line (Figure

5-10). They display a population composed of both GFP+ and GFP- cells. This observation

supports two concepts consistent with a cancer stem cell. First, this shows that tumors formed

from the clonally expanded line are capable of recapitulating the heterogeneity of the Oct-4

expression of the original population. More importantly though, this suggests that GFP- cells are

derived from GFP+ cells, suggesting that any GFP- cells arise as a result of cellular

differentiation.









FACS of clonally derived tumor cells into GFP enriched and GFP depleted fractions,

followed by xenotransplantation over a range of cell doses, showed that cells that continued to

express the Oct-4 promoter sequence were significantly more tumorigenic than cells that had lost

that capacity (Figure 5-11). Consistent with results from the non-clonally derived cell population,

the GFP+ cell fraction produced tumors in all animals at all doses. Administration of as few as

300 cells was sufficient to cause tumor formation in all animals tested. The GFP- fraction in

contrast formed tumors in only 50% of the mice at 30,000 cells and 25% at 3,000 cells. Analysis

of the resultant tumors from each cell fraction showed that first passage tumors from the GFP+

fraction were comprised of 85% fluorescent cells. Tumors from the GFP- fraction were also

highly GFP+ with about 55% of the cells showing fluorescence above background.

As before, we believe the tumors arising from the GFP- population in all probability are

the result of low level contamination of GFP+ cells present in the depleted fraction. Foremost,

the majority of the cells in tumors generated from the GFP depleted fraction are GFP+. In our

experience, we have never observed cells reacquire the capacity to express the Oct-4 reporter

once lost. Furthermore, as seen in Figure 5-1 lb a dose of only 300 GFP+ cells is sufficient to

form tumors in all mice. Since typically between 3-8% of the cells in the respective fractions are

contaminants of the opposing fraction, we would expect between 900-2400 GFP+ cells in the

30,000 cell dose and 90-240 in the 3,000 dose, numbers at both dilutions that approach or exceed

the minimum established tumorigenic dose.

Cells from the GFP- fraction only formed tumors in doses of 3x103 or greater, and in 50%

or less of the mice. Tumor formation in this condition is most likely due to cross contamination

of GFP+ cells in the GFP depleted fraction. Figure 5-9a-c shows that the level of cross

contamination can reach 8.0% in the fractionated samples. Along these lines, a dose of 3x103









GFP- cells contains enough GFP+ cells to initiate tumor formation, and doses of 300 cells or less

do not contain enough cells with tumor forming capacity. Additionally, we see that the tumor is

composed of 55% GFP+ and 45% GFP- cells (Figure 5-1 Ic). This heterogeneity is also most

likely due to the cross contamination of GFP+, putative tumor initiating cells that were injected.

This small subpopulation would be required to divide more times to form a tumor, when

compared to tumors formed from a GFP enriched fraction. As this small number of GFP+ cells

divide, they will lose the ability to express the Oct-4 promoter, resulting in a larger population of

GFP- cells in the tumor.

As the population of the serial passed clones regressed back to resemble the Oct-4/GFP

expression of the parental line, we found that we could diverge the two populations further by

exposure of the cells to osteogenic differentiation media as well as BMP 4 and BMP 7.

Exposure to osteogenic media resulted in an overall reduction in GFP expression from

the Oct-4 promoter producing a heterogeneous population of approximately 60% GFP+ and 40%

GFP- cells. Despite these changes in transcription of the Oct-4 promoter were unable to detect

significant matrix deposition by staining of the confluent monolayers with Alizarin Red S, or

changes in the expression of genes associated with osteogenesis using RT-PCR. This may have

occurred for several reasons. First, OS521 is classified histologically as a highly dedifferentiated

osteosarcoma and thus may be relatively resistant to certain differentiation cues. Along these

lines, it is possible that this cell line may require a more extended incubation to stimulate the

cells to begin secreting matrix. It is also possible that deposited extracellular matrix components

were lost during passage of the confluent cultures during the extended incubation. Additionally,

changes in the expression of genes associated with osteogenesis may not have been detectable









because of the heterogeneity of the treated cell population. Fractionating the cells into their

respective GFP+ and GFP- populations, before RT-PCR analysis, may have resolved this issue.

.Differentiation from BMP stimulation occurs when the protein causes the

transmembrane BMPI receptor to phosphorylate the BMPII receptor and form a heteromeric

complex. Intracellularly, the BMPII receptor phosphorylates the SMAD1/5 proteins, which are

cytoplasmic. These, in turn, induce the upregulation of the transcription factors RUNX2 and

Osterix, which are critical to osteoinduction. BMPs are also known to activate MAPKs which

upregulates RUNX2. We postulate that by activating these pathways, we have induced

osteogenic differentiation which causes a loss of the expression of our transgenic reporter gene.

This in turn would allow a population of more differentiated but less tumorigenic cells to

increase in number, consistent with the increased GFP- population.

Upon xenografting the cells at a dose of 100k cells/mouse, we observed tumorigenesis in

2/3 of the mice inoculated in both BMP conditions. We attribute this to the large number of cells

injected. It should be noted that in previous work we established that in GFP enriched

populations, tumors readily formed at doses of three hundred cells. Along these lines it is highly

probable that at least three hundred Oct-4/GFP+ cells were delivered that maintained their

tumorigenic potential. Even if a smaller dose was administered (< 30k cells/mouse), we believe

we would most likely observe tumor formation because we would still be injecting a sufficient

dose of tumorigenic cells. Though a large portion of the cells from the sample were indeed

induced to differentiate (loss of GFP expression), and thus had lost their tumorigenic potential,

the underlying fact that a small dose of cells retain their tumorigenic capacity does not change.

The remaining GFP+ population, still retaining their tumorigenic potential, had perhaps not been

given enough time to respond to the BMP signaling. It may also be the case that under the









conditions used; they did not receive a sufficient amount of stimulation from the BMPs. As a

treatment option, differentiation therapy using agents such as BMP4 or BMP7 exposure might do

little to halt tumor formation due to the very small number of cells required to initiate a tumor.

Even though these molecules do induce differentiation, the relatively small subpopulation of

cells that can form a tumor persists. If this were to be used as an effective therapy, stimulatory

effects must induce virtually every cell in the tumor to differentiate.

What is the nature of the relationship between Oct-4/GFP expression, tumorigenesis, and

differentiation at the molecular level? This is a difficult question to address since tumorigenic

ability is a complex phenotype involving multiple factors and pathways and BMPs have

pleiotropic effects. It is likely that Oct-4/GFP activity is a product of the activity of a pathway or

group of factors critical to the tumorigenic phenotype.

Future work on this project will involve further characterization of the isolated cancer

stem cells to identify differences at the molecular level between the GFP+ and GFP- populations.

These studies should provide insight into the pathways and mechanisms responsible for

conferring tumorigenicity and provide clues to the development of therapeutic strategies that

target them.








































Figure 5-1. Osteosarcoma: H&E stain of a femoral osteosarcoma.














Recurrent
growth


Chemotherapy


Primary Tumor


Metastases


Figure 5-2. Cancer Stem Cells. Treatment of the primary tumor with chemotherapeutic agents
kills the rapidly dividing cells of the tumor bulk (blue to black) but fails to eradicate
the tumor stem cells (green) which can then cause re-growth of the primary tumor or
disseminate to form distal metastases.












A a a g a ia d


OOt4

Nonog-- - -
STATS


r-actln k -- -- - -- -


------~


122
1.0
0-8
0-0

0-2


C
111t-4 5W
STAT3 IMM ON MW OW MIO -87 kDO
pSTAT3 4M m Mn- 90 kDa
-ctsin F- 42 k1a

D Nanog *

''
3

i3*, dP %
OCt.4
e


0 0a 04 0,6 o. 10 IF l -* W6
Oct-4 t


Figure 5-3. ESC-specific Genes in Sarcospheres. Genes specific to ESCs show increased
expression in sarcosphere cultures derived from bone sarcomas. (A) Monolayer and
sarco-sphere (SP) cultures from five osteosarcoma (OS) and three chondrosarcomas
(CS) were analyzed for Oct-4, Nanog and STAT3 mRNA by RT-PCR; P-actin
expression was used as a positive control. Sphere cultures demonstrate increased
transcription of both Oct-4 and Nanog over adherent cultures; STAT3 expression was
uniform between both culture types. (B) Relative band intensities for Oct-4 and
Nanog for each culture from (A) were quantitated by densitometry, normalized
relative to p-actin and plotted on the graph shown (Oct-4, x-axis; Nanog, y-axis). As
indicated by the grouping, the sphere cultures of each sarcoma showed significantly
greater expression of both Oct-4 and Nanog than adherent monolayer cultures
(p<0.05, Pearson's correlation). (C) Western blot analysis of lysates from
representative bone sarcoma cell cultures for protein expression of Oct-4, STAT3 and
activated (phosphorylated, p) STAT3. P-actin was used as a positive control for
loading, membrane transfer and immunoblotting. All cultures showed positive
staining of protein bands of the appropriate sizes as indicated. (D) Small and large
sarcospheres were embedded in fibrin and then paraffin section and stained using
immunohistochemistry for Nanog and Oct-4 as indicated. Small spheres show intense
staining in cells in the periphery. Large spheres show similar numbers of darkly
staining cells with dramatically increased numbers of poorly staining cells in the
interior of the sphere.


B


a
2


4









Oct-4 Nanog



S" OS 154 ". :- "
.CS 187 s; -sg n p. f
,. i .. .. -- *. ,





Fetal t-,
Testis -' ."l r




Figure 5-4. Immunohistochemical staining for Oct-4 and Nanog in sections from tumor biopsies
of chondro- and osteosarcoma. One representative osteosarcoma (OS-154) and
chondrosarcoma (CS-187) and a positive control, human fetal testis, are shown as
indicated. CS-187 shows a single nuclei positive (brown) for Oct-4 and multiple
nuclei positive (brown) for Nanog in a lung metastasis from a chondrosarcoma. OS
154 sections demonstrate scattered Oct-4 nuclear staining and near complete nuclear
Nanog staining in a primary fibular osteosarcoma. Twenty-six week fetal testis with
scattered Oct-4 and Nanog nuclear staining are shown as positive controls.










A

Gata4 -- --
Gata6
AFP .--. .- -
p-fI tubulin E-- ---
0-actin





AFP om 70 kDa

p-actin --- --- p 42 kDa
(C and D) Expression of b-III tubulin in tissue specimens from bone sarcoma as
"- : "

-_ .- 7
'l ,"" .'.'. .-*', ".
mi i ,i ,..;;.: '. i








Figure 5-5. Analyses of bone sarcoma cultures for expression of genes of endo- and ectodermal
lineages. (A) RT-PCR analyses of adherent and sarcosphere cultures for transcription
of endoderm-associated genes (Gata-4, Gata-6, and alpha fetoprotein [AFP]) and the
neuro-ectoderm marker, P3-III tubulin. Primers for b-actin were used as positive
reaction controls as indicated. Control lane represents parallel RT-PCR reactions
performed without reverse transcriptase. (B) Western blot analyses for expression of
b-III tubulin and AFP from lysates of adherent cultures shown in panel A,
demonstrating protein expression of endoderm and neuro-ectoderm-associated genes.
(C and D) Expression ofb-III tubulin in tissue specimens from bone sarcomas as
detected by immunohistochemistry, and in adherent cultures (E and F) demonstrated
by immunocytochemistry. In panels C andD, arrows indicate regions of positive
staining. In panels E and F areas of b-III tubulin staining are seen in red. Nuclei were
counterstained blue using Hoechst's stain.









OS 99-1 CS 828


Osteogenic '' .J


,,. ..-, :, B .- :


Adipogenic




Figure 5-6. Multipotent cells in bone sarcoma. Following incubation in osteogenic or adipogenic
media to induce differentiation along mesenchymal lineages, the respective cultures
were analyzed for mineralization by Von Kossa staining or for lipid vacuoles by
staining with Oil-red -0. As shown, both cultures showed focal staining for
osteogenic and adipogenic differentiation.






















\ S. K V0. -Human Oct-4




GFP

Figure 5-7. Schematic of phOct-4/GFP. The human Oct-4 promoter (-4kb) was cloned upstream
of the gene for Green Fluorescent Protein (GFP). The plasmid also contains an
SV40promoter-driven Neomycin resistance gene allowing for selection using G418
(NeoR).














B. Monolayer




FITC4A




PE-A

PEA







PEA


Tumor




ITCA
MHC II



PE-A









PE A
NCAM








PE-A


Figure 5-8. Expression of phOct-4/GFP and surface antigens in OS521 in vitro and in vivo. (A)

Visualization Oct-4/GFP transfected monolayer cultures using inverted fluorescence

microscopy (top panel) and in tumors by immunofluorescence (bottom panel). (B)

Analysis of OCT-4/GFP and surface antigen expression by flow cytometry in

monolayer culture and xenotranplanted tumors of OS521.













SCell Dose CFP-enriched Unsorted GFP-depleted

30,000 8/8 22 days 6/8 26 days 5/8 47 days

FITC-A 3.000 3', 3-1 dly 6'3 1 4-1 dyvs I'J f51 daji

1 P 300 8/8 45 days 3/8 60 days 0/8 90 days





c GFP-depted






Figure 5-9. The Oct-4/GFP enriched tumor fraction of OS 521 is highly tumorigenic following
delivery into NOD/SCID mice. (A) Xenotranplanted tumors were harvested and
fractionated by FACS into GFP-enriched and depleted populations for tumorigenicity
experiments. The resulting populations were 92% and 97% pure as shown (B and C,
respectively). (D) The frequency and rate of tumor formation of the two fractions.






























Stumors-Simon P1

P3 P4

l l l l lll I I I l IIll I I l I I 1 I I II III I
102 10 10' 10
FITC-A

tumors-simon P3

P- P4


S010 102 103 1 10
FITC-A


Figure 5-10. Serial transplant of Oct-4/GFP clones in vivo. Flow cytometric analysis for Oct-
4/GFP expression in clones isolated from xenotransplants showing the emergence of
a GFP- population following serial passage in vivo (A) Clone Si in vitro prior to
xenotransplantation (B) following 1 passage in vivo (C) following 3 passages in vivo
tumor (D and E) Clone T1 in vitro (red) and passage-1 tumor (blue) (E) third
passage tumor.














Cell Dose GFP + GFP -


30,000 4/4 23 days 2/4 47 days

3,000 4/4 36 days 1/4 50 days

300 4/4 44 days 0/4 90 days


55%GFP


Figure 5-11. The Oct-4/GFP-enriched fraction of clonally derived tumors is highly tumorigenic
in vivo. (A) Transplanted tumrs were fractionated by FACS and the Oct-4/GFP
enriched and depleted fractions were transplanted into NOD/SCID mice. (B) The
clonal Oct-4/GFP-enriched fraction is more tumorigenic than the depleted fraction.
(C) Analysis by flow cytometry of tumor transplants shows the tumors arising from
the GFP+ and GFP- fractions are composed primarily of Oct-4/GFP+ cells.


tumors-simon P3

R3P P4

I 1 1111111 1' 11111 1 111111 1'
1 010 102 10 10 10
FITC-A


C
85%GFP+


3X104 GFP+






SFP FITC 0 -A
OFP FITC-A


3X104 GFP-






1 1' i 2 io P 1(1 I-3
GFPFITC-A











A Control


B Day 14


C Day 21


GFP


Figure 5-12. Culture in osteogenic media induces silencing of the Oct-4/GFP transgene. FACS
analysis of cells isolated from transplanted tumors cultured in standard culture media
for 21 days (A) or in osteogenic media for 14 or 21 days (B and C respectively).













tumors-simon P3


S P3 P4 60%EGFP+

I l lI '1 1 I I 1111113 II I I II I
18 010 102 10 10 10
FITC-A

B C
simon-BMP4 simon-BMP4 p1


25% EGFP+ P3 EGFP

'0 '2 .' ', 1 2 010 10 10
FITC-A 6_3 ____-A
simon-BMP7
simon-BMP7 p1

C-e
38% EGFP Pa 70% EGFP+


i22 010 1 1 1'0 10
FITC-A -2 -10 010 10 t 10
GFP FITC-A

Figure 5-13. Treatment of OS tumor cells with BMPs induced silencing of Oct-4/GFP expression
in vitro. (A) FACS analysis of tumor samples cultured in standard culture media. (B)
Tumor samples cultured in BMP-4 or -7 conditioned media for 10 days. (C) Tumors
generated by transplantation of 1X10 5 treated tumor cells.










LIST OF REFERENCES


Al-Hajj, M., Wicha, M.S., Benito-Hemandez, A., Morrison, S.J., and Clarke, M.F. (2003).
Prospective identification of tumorigenic breast cancer cells. Proc Natl Acad Sci U S A.
7, 3983-3988.

Bonnet, D. and Dick, J.E. (1997). Human acute myeloid leukemia is organized as a hierarchy
that originates from a primitive hematopoietic cell. Nat Med. 7, 730-737.

Carew, J.S., Giles, F.J., and Nawrocki, S.T. (2008). Histone deacetylase inhibitors: Mechanisms
of cell death and promise in combination cancer therapy. Cancer Lett.

Chi, S.N., Conklin, L.S., Qin, J., Meyers, P.A., Huvos, A.G., Healey, J.H., and Gorlick, R.
(2004). The patterns of relapse in osteosarcoma: the Memorial Sloan-Kettering
experience. Pediatr Blood Cancer. 1, 46-51.

Eckfeldt, C.E., Mendenhall, E.M., and Verfaillie, C.M. (2005). The molecular repertoire of the
'almighty' stem cell. Nat Rev Mol Cell Biol. 9, 726-737.

Feldman, N., Gerson, A., Fang, J., Li, E., Zhang, Y., Shinkai, Y., Cedar, H., and Bergman, Y.
(2006). G9a-mediated irreversible epigenetic inactivation of Oct-3/4 during early
embryogenesis. Nat Cell Biol. 2, 188-194.

Galli, R., Binda, E., Orfanelli, U., Cipelletti, B., Gritti, A., De Vitis, S., Fiocco, R., Foroni, C.,
Dimeco, F., and Vescovi, A. (2004). Isolation and characterization of tumorigenic, stem-
like neural precursors from human glioblastoma. Cancer Res. 19, 7011-7021.

Gerrard, L., Zhao, D., Clark, A.J., and Cui, W. (2005). Stably transfected human embryonic stem
cell clones express OCT4-specific green fluorescent protein and maintain self-renewal
and pluripotency. Stem Cells. 1, 124-133.

Gibbs, C.P., Jr., Weber, K., and Scarborough, M.T.(2002). Malignant bone tumors. Instr Course
Lect. 51, 413-428.

Gronthos, S. and Simmons, P.J. (1995). The growth factor requirements of STRO-1-positive
human bone marrow stromal precursors under serum-deprived conditions in vitro. Blood.
4, 929-940.

Hanahan, D. and Weinberg, R.A. (2000). The hallmarks of cancer. Cell. 1, 57-70.

Hattori, N., Nishino, K., Ko, Y.G., Hattori, N., Ohgane, J., Tanaka, S., and Shiota, K.( 2004).
Epigenetic control of mouse Oct-4 gene expression in embryonic stem cells and
trophoblast stem cells. J Biol Chem. 17, 17063-17069.

Hemmati, H.D., Nakano, I., Lazareff, J.A., Masterman-Smith, M., Geschwind, D.H., Bronner-
Fraser, M., and Komblum, H.I. (2003). Cancerous stem cells can arise from pediatric
brain tumors. Proc Natl Acad Sci U S A. 25, 15178-15183.









Ignatova, T.N., Kukekov, V.G., Laywell, E.D., Suslov, O.N., Vrionis, F.D., and Steindler, D.A.
(2002). Human cortical glial tumors contain neural stem-like cells expressing astroglial
and neuronal markers in vitro. Glia. 3, 193-206.

Jhappan, C., Gallahan, D., Stahle, C., Chu, E., Smith, G.H., Merlino, G., and Callahan, R.
(1992). Expression of an activated Notch-related int-3 transgene interferes with cell
differentiation and induces neoplastic transformation in mammary and salivary glands.
Genes Dev. 3, 345-355.

Jones, P.A. and Baylin, S.B. (2002). The fundamental role of epigenetic events in cancer. Nat
Rev Genet. 6, 415-428.

Katsetos, C.D., Legido, A., Perentes, E., and Mork, S.J. (2003). Class III beta-tubulin isotype: a
key cytoskeletal protein at the crossroads of developmental neurobiology and tumor
neuropathology. J Child Neurol. 12, 851-866; discussion 867.

Lessard, J. and Sauvageau, G. (2003). Bmi-1 determines the proliferative capacity of normal and
leukaemic stem cells. Nature. 6937, 255-260.

Li, C.D., Zhang, W.Y., Li, H.L., Jiang, X.X., Zhang, Y., Tang, P.H., and Mao, N. (2005).
Mesenchymal stem cells derived from human placenta suppress allogeneic umbilical cord
blood lymphocyte proliferation. Cell Res. 7, 539-547.

Meissner, A., Wernig, M., and Jaenisch, R. (2007). Direct reprogramming of genetically
unmodified fibroblasts into pluripotent stem cells. Nat Biotechnol. 10, 1177-1181.

Meyers, P.A., Schwartz, C.L., Krailo, M., Kleinerman, E.S., Betcher, D., Bernstein, M.L.,
Conrad, E., Ferguson, W., Gebhardt, M., Goorin, A.M., Harris, M.B., Healey, J., Huvos,
A., Link, M., Montebello, J., Nadel, H., Nieder, M., Sato, J., Siegal, G., Weiner, M.,
Wells, R., Wold, L., Womer, R., and Grier, H. (2005). Osteosarcoma: a randomized,
prospective trial of the addition of ifosfamide and/or muramyl tripeptide to cisplatin,
doxorubicin, and high-dose methotrexate. J Clin Oncol. 9, 2004-2011.

Miller, C.W., Aslo, A., Won, A., Tan, M., Lampkin, B., and Koeffler, H.P. (1996). Alterations of
the p53, Rb and MDM2 genes in osteosarcoma. J Cancer Res Clin Oncol. 9, 559-565.

Ohlstein, B., Kai, T., Decotto, E., and Spradling, A. (2004). The stem cell niche: theme and
variations. Curr Opin Cell Biol.6, 693-699.

Ohnishi, K. (2007). PML-RARalpha inhibitors (ATRA, tamibaroten, arsenic troxide) for acute
promyelocytic leukemia. Int J Clin Oncol. 5, 313-317.

Okita, K., Ichisaka, T., and Yamanaka, S. (2007). Generation of germline-competent induced
pluripotent stem cells. Nature. 7151, 313-317.

Pardal, R., Clarke, M.F., and Morrison, S.J. (2003). Applying the principles of stem-cell biology
to cancer. Nat Rev Cancer.12, 895-902.









Pasca di Magliano, M. and Hebrok, M. (2003). Hedgehog signalling in cancer formation and
maintenance. Nat Rev Cancer.12, 903-911.

Piccirillo, S.G. and Vescovi, A.L. (2006). Bone morphogenetic proteins regulate tumorigenicity
in human glioblastoma stem cells. Ernst Schering Found Symp Proc. 5, 59-81.

Qiang, Y.W., Endo, Y., Rubin, J.S., and Rudikoff, S. (2003). Wnt signaling in B-cell neoplasia.
Oncogene.10, 1536-1545.

Ragland, B.D., Bell, W.C., Lopez, R.R., and Siegal, G.P. (2002). Cytogenetics and molecular
biology of osteosarcoma. Lab Invest.4, 365-373.

Reya, T., Morrison, S.J., Clarke, M.F., and Weissman, I.L. (2001). Stem cells, cancer, and cancer
stem cells. Nature. 6859, 105-111.

Reynolds, B.A. and Weiss, S. (1992). Generation of neurons and astrocytes from isolated cells of
the adult mammalian central nervous system. Science. 5052, 1707-1710.

Sandberg, A.A. and Bridge, J.A. (2003). Updates on the cytogenetics and molecular genetics of
bone and soft tissue tumors: osteosarcoma and related tumors. Cancer Genet Cytogenet.
1, 1-30.

Simmons, P.J., Gronthos, S., Zannettino, A., Ohta, S., and Graves, S. (1994). Isolation,
characterization and functional activity of human marrow stromal progenitors in
hemopoiesis. Prog Clin Biol Res. 271-280.

Singh, S.K., Clarke, I.D., Terasaki, M., Bonn, V.E., Hawkins, C., Squire, J., and Dirks, P.B.
(2003). Identification of a cancer stem cell in human brain tumors. Cancer Res.18, 5821-
5828.

Singh, S.K., Hawkins, C., Clarke, I.D., Squire, J.A., Bayani, J., Hide, T., Henkelman, R.M.,
Cusimano, M.D., and Dirks, P.B. (2004). Identification of human brain tumour initiating
cells. Nature. 7015, 396-401.

Steindler, D.A. (2007). Stem cells, regenerative medicine, and animal models of disease. Ilar J.4,
323-338.

Stewart, K., Walsh, S., Screen, J., Jefferiss, C.M., Chainey, J., Jordan, G.R., and Beresford, J.N.
(1999). Further characterization of cells expressing STRO-1 in cultures of adult human
bone marrow stromal cells. J Bone Miner Res. 8, 1345-1356.

Thomas, D. and Kansara, M. (2006). Epigenetic modifications in osteogenic differentiation and
transformation. J Cell Biochem. 4, 757-769.









van de Wetering, M., Sancho, E., Verweij, C., de Lau, W., Oving, I., Hurlstone, A., van der
Horn, K., Batlle, E., Coudreuse, D., Haramis, A.P., Tjon-Pon-Fong, M., Moerer, P., van
den Born, M., Soete, G., Pals, S., Eilers, M., Medema, R., and Clevers, H. (2002). The
beta-catenin/TCF-4 complex imposes a crypt progenitor phenotype on colorectal cancer
cells. Cell. 2, 241-250.

Varga, A.C. and Wrana, J.L. (2005). The disparate role of BMP in stem cell biology. Oncogene.
37, 5713-5721.

Wernig, M., Meissner, A., Foreman, R., Brambrink, T., Ku, M., Hochedlinger, K., Bernstein,
B.E., and Jaenisch, R. (2007). In vitro reprogramming of fibroblasts into a pluripotent
ES-cell-like state. Nature. 7151, 318-324.

Wuisman, P. and Enneking, W.F. (1990). Prognosis for patients who have osteosarcoma with
skip metastasis. J Bone Joint Surg Am. 1, 60-68.

Yin, T. and Li, L. (2006). The stem cell niches in bone. J Clin Invest. 5, 1195-1201.

Yu, J., Vodyanik, M.A., Smuga-Otto, K., Antosiewicz-Bourget, J., Frane, J.L., Tian, S., Nie, J.,
Jonsdottir, G.A., Ruotti, V., Stewart, R., Slukvin, II, and Thomson, J.A. (2007). Induced
pluripotent stem cell lines derived from human somatic cells. Science. 5858, 1917-1920.









BIOGRAPHICAL SKETCH

Thomas Currie was born in 1980, in Cleveland, Ohio. Shortly thereafter, his family

moved to Florida, where his father worked as an Anesthesiologist and his mother raised him and

his three brothers and sister. After receiving his high school degree in the international

baccalaureate program from Sebastian River High School, Thomas began his undergraduate

career at the University of Florida in 1999 on a 100% Florida bright futures scholarship. In 2004

he received his bachelor degree in psychology and completed the pre-requisites needed for dental

school. In 2006, before applying to dental school, Thomas decided to pursue a master of science

degree from the University of Florida's College of Medicine. He focused his studies on

molecular biology and aimed his research toward understanding human osteosarcoma. During

this time, he was accepted into the University of Florida's College of Dentistry, for the 2008

entering class. In the future, Thomas wishes to work as a dentist in the state of Florida, while

enjoying outdoor activities in his spare time.





PAGE 1

1 DIFFERENTIATION OF PUTATIVE CANCER STEM CELLS AND ITS EFFECT ON TUMORIGENICITY IN OSTEOSARCOMA By THOMAS PATRICK CURRIE A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2008

PAGE 2

2 2008 Thomas Patrick Currie

PAGE 3

3 To my Mother

PAGE 4

4 ACKNOWLEDGMENTS First and forem ost, I would like to thank my parents for supporting me through all of my educational endeavors. Without them, the road w ould have been impossible. They always pushed me to be the best that I could be, and to pur sue my dreamspromising that they would support my decisions along the way. To this day they have kept their word, and because of this I have been able to further my education with uncompromised focus. As I begin dental school in the fall of 2008, I am still comforted by their prom ise that they are behind me 100%. Most importantly I would like to thank Steven Ghivizzani PhD, Parker Gibbs MD, and Padraic Levings PhD for guiding me through my classes, research, and requirements for the masters program. They graciously included me in th eir project, allowed me to focus my thesis on one of their specific aims, and provided directio n along the way. They truly mentored me in the laboratory, ensuring my success through their ad vice and wisdom. They understood my mishaps and mistakes, offering constructive criticism an d providing learning expe riences, rather than acting in a belittling manner. All of the long hour s and late nights have not gone unnoticed and I appreciate them for their dedication in helping me pursue my degree. I must recognize the importance of my committee, which included Steven Ghivizzani PhD, David Bloom PhD, and Ammon Peck PhD. They amicably critiqued my work and steered me down the path to success, while ensuring my data would be scientific ally sound. They raised the bar by challenging me to investigate question s that I was otherwise naive to, allowing me to see things from new perspectives. I know they have dedicated numerous hours to reviewing and helping me and with my work, and I cannot forget that. I truly appreciate their assistance and commitment to my education. I am also grateful to Steve McClellan, who ta ught me how to use the flow cytometer. He spent a great deal of time edu cating me on its usage and answering my questions when things

PAGE 5

5 went awry. He was very patient with me, and be sides being an excellent teacher, kept the flow cytometry lab interesting as well with his wit and humor. I would also like to thank all of my professors that have been gracious enough to lecture with enthusiasm and expertise, maintaining a stimulating learning environment along the way. I can truly say that I have learne d a great deal from them, vas tly increasing my understanding of the biological sciences and improving my view on the mechanisms of life. I must include special thanks to Mrs. Conners for her help in all aspects of the programfrom the time I applied, right up until graduation. Without her help, even registering for classes would have been a confusing process. Her advice wa s always in my best interest, and she always urged me to seek her help before going things alone. I know that she trul y cares for all of the students in the program and that their well bei ng is at the forefront of her priorities. I would like to acknowledge the rest of the Ghivizzani lab an d their contribution to my success. Carrie, Anthony, Rachael, Marsha, Jesse Kristen, Olga and Celine were all very helpful, but more importantly I think that they made working in the lab a fun and exciting place to be. Spending long hours performing bench top procedures would have been unnerving had it not been for their company and comp anionship throughout the past year. Of all of my mentors, I must express my greatest gratitude to Padraic Levings, PhD. Paddy. Through his selflessness and benevolence, he has provided me with more assistance than anyone else in attaining my goals. He patiently walked me step by step through the paces of laboratory work, making sure that I was compet ent enough before I ventured on my own. His upbeat demeanor kept me motivated and eager to learn; making this program a very exciting and fun experience for me, while also allowing us to form a great friendship. Overshadowing all of this was his unrelenting efforts to ensure my success in the program. I know that he has spent

PAGE 6

6 uncountable hours helping me with my experime nts, presentations, and manuscripts; none of which he was required to do. I must say that I owe many of my accomplishments to him and his desire to see me succeed. Lastly I would like to thank some miscellaneous people. First I would like to thank the knuckle-headed patrons who frequent the Swamp Re staurant during late ni ght hours. They have kept my desire to further my education salient, while also reminding me that I do not want to be a bouncer for the rest of my life. Secondly I would like to thank my two dogs, Moose and Legend, for keeping me sane and tranquil during the stressful parts of typing my thesis. Lastly, I would like to thank the Discovery channel, sp ecifically Dirty Jobs and Myth Busters, for entertaining my brain during the time when I could not concentrate on my school work.

PAGE 7

7 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF FIGURES.........................................................................................................................9 ABSTRACT...................................................................................................................................10 CHAP TER 1 INTRODUCTION..................................................................................................................12 Aim 1: To Identify the Origin of the Hete rogeneous Expression Of Oct-4/GFP in the OS521Oct-4GFP Cell Line .................................................................................................14 Aim 2: To Address the Role of Different iation on Tum origenic Capacity of OS521Oct4/GFP Clones......................................................................................................................14 2 METHODS.............................................................................................................................15 Derivation of Osteosarcoma Cell Lines from Patient Biopsies and Description of phOct4/GFP ..................................................................................................................................15 Xenotransplantation and Tumorigenicity Assays................................................................... 16 Fluorescence Activated Cell So rting and Flow Cytom etry.................................................... 17 Clonal Expansion of OS521Oct-4/GFP Cells......................................................................... 18 In vitro Osteogenic Differentiation .........................................................................................18 Treatment with Bone Morphogenic Proteins.......................................................................... 18 3 LITERATURE REVIEW.......................................................................................................20 The Biology of Cancer............................................................................................................20 The Biology of Osteosarcoma................................................................................................21 Stem Cells..................................................................................................................... ..........22 Cancer Stem Cell Theory........................................................................................................ 25 Use of Agents That Induce Different iation for the Treatm ent of Cancer...............................26 Preliminar y Results ............................................................................................................ .....28 Cells Isolated from Bone Sarcoma Cultures Exhibit Stem-Like Attributes.................... 28 Development of an In viv o Model to Examine the Role of Stem-Like cells in the Pathogenesis of Osteosarcoma.....................................................................................32 Expression of an Oct-4 Promoter/GFP Reporter Construct T hat Selectively Identifies Cancer Stem Cells in Osteosarcoma............................................................ 33 4 RESULTS...............................................................................................................................37 Rationale for Aim 1................................................................................................................37 Rationale for Aim 2................................................................................................................38 Clonal Expansion of OS521Oct-4/GFP cells......................................................................... 39

PAGE 8

8 Flow Cytometric Analysis of Cl ones and Clonally Derived T umors..................................... 40 Culture of OS521Oct-4/GFP Clones in Osteogenic Media....................................................41 Culture of OS521Oct-4/GFP Clonally Derived Tumor Cells in BMP Conditioned Media... 42 5 DISCUSSION.........................................................................................................................44 LIST OF REFERENCES...............................................................................................................64 BIOGRAPHICAL SKETCH.........................................................................................................68

PAGE 9

9 LIST OF FIGURES Figure page 5-1 Osteosarcoma: H&E stain of a fe moral osteosarcoma......................................................51 5-2 Cancer Stem Cells..............................................................................................................52 5-3 ESC-specific Genes in Sarcospheres................................................................................. 53 5-4 Immunohistochemical staining for Oct-4 a nd Nanog in sections from tumor biopsies of chondroand osteosarcoma...........................................................................................54 5-5 Analyses of bone sarcoma cultures for e xpression of genes of endoand ectoderm al lineages..............................................................................................................................55 5-6 Multipotent cells in bone sarcoma..................................................................................... 56 5-7 Schematic of phOct-4/GFP................................................................................................ 57 5-8 Expression of phOct-4/GFP and surface an tigen s in OS521 in vitro and in vivo............. 58 5-9 The Oct-4/GFP enriched tumor fraction of OS 521 is highly tum origenic following delivery into NOD/SCID mice........................................................................................... 59 5-10 Serial transplant of Oct-4/GFP clones in vivo ....................................................................60 5-11 The Oct-4/GFP-enriched fr action of clonally derived tum o rs is highly tumorigenic in vivo....................................................................................................................................61 5-12 Culture in osteogenic media induces silencing of the Oct-4/GFP transgene ..................... 62 5-13 Treatment of OS tumor cells with BMPs induced silencing of Oct-4/GFP expression in vitro ................................................................................................................................63

PAGE 10

10 Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science DIFFERENTIATION OF PUTATIVE CANCER STEM CELLS AND ITS EFFECT ON TUMORIGENICITY IN OSTEOSARCOMA By Thomas Patrick Currie August 2008 Chair: Steven Ghivizzani Major: Medical Sciences In previous work, our group had shown that a subpopulation of cells in osteosarcoma, a mesenchymal malignancy, have certain stem ce ll like properties. They expressed embryonic stem cell transcription genes essential in mainta ining pluripotency (Oct-4 and Nanog), they were capable of growing as spherical colonies, and they had th e capacity for self renewal in vitro Additionally, some of our cell lines formed tumors when xenografted subcutaneously into nonobese diabetic/severe combined immune deficient ( NOD/SCID) mice. To investigate the relationship between stemness and tumorigenesis, a cell line (OS521Oct-4/GFP) was transfecte d with a reporter construc t containing the human Oct-4 promoter linked to the gene for green fluores cent protein, and an i ndependently regulated neomycin resistance gene. These transgenic cultures were observ ed to express GFP heterogeneously, as were the tumors that were formed from them. Using fluorescent activated cell sorting, the GFP+ cells were fo und to be 100 fold more tumori genic than GFPcells in terms of penetrance and kinetics, thus indicating that cells capable of expre ssing the Oct-4 gene had enhanced tumorigenic capacity. To further investigate the putative osteosarco ma stem cell, we clonally expanded the Oct4/GFP+ cells and characterized their capacity to form tumors following xenotransplantation in

PAGE 11

11 NOD/SCID mice. Delivery of the cells yielded tumors that were heterogeneous for Oct-4 expression, showing that Oc t-4/GFP+ cells can give rise to GFPcells in vivo We also demonstrated that subsequent passage of these cells gives rise to a population heterogeneous for Oct-4/GFP expression. This suggested that the loss of tumorigenicity was associated with cellular differentiation as indicated by the in ability to activate the Oct-4 promoter. Secondly, because differentiation of tumor cells has proven to be therapeutic in other forms of cancer, we wanted to examine the eff ect of induced differen tiation of the putative osteosarcoma stem cell. OS521Oct-4/GFP clones in monolayer were incubated in commercially available osteogenic differentiation media for 21 days. We observed a 32% reduction in the proportion of cells expressing Oc t-4, suggesting that certain tu mor cells differentiated and no longer possessed a stem like phenotype. The clonally derived tumor cells were then in cubated in the presence of BMP4 or BMP 7 at 200>ng/ml. Flow cytometry showed a 58% an d 36% reduction in GFP expression in the BMP 4 and BMP7 treated cells, respectively. These cells were then xenografted subcutaneously into NOD/SCID mice, and despite apparent differentiation, tumors still formed in both conditions. Analysis of the corresponding BMP4 tumors showed GFP expression at 10%, and the BMP 7 condition showed GFP expression at 70%. These data suggest that a clonally derived population, homogeneous for Oct-4 expression, can undergo differentiation events leadi ng to a loss of expressi on of the gene when passaged in vivo. The resulting tumo rs recapitulate the heterogene ity of the parental line and confirm the notion that GFPcells arise from GFP+ cells. Further induc ed differentiation via BMP4/7 exposure can significantly decrease the proportion of Oct-4 expression, but not to the extent where tumor formation is completely inhibited.

PAGE 12

12 CHAPTER 1 INTRODUCTION Osteosarcom a (OS) is the most comm on primary bone malignancy of childhood and adolescence. Despite advances in surgery a nd chemotherapy, long-term survival rates have stagnated over the last 30 year s, and still 40% of patients diag nosed with osteosarcoma die of their disease. The cancer stem cell theory, though, may account for the lack of effective chemotherapeutics in this and other types of cancer. Recent studies of leukemia, brain and breast cancer, suggest that within a tumor a subpopulation of cells with properties of stem cells, is responsible for tumorigenesis and malignanc y while the remaining tumor mass is comprised of differentiated progeny of the tu mor stem cell that are relatively benign. Thus, agents designed to eradicate the bulk of rapidly dividing cells within a tumor ma ss may not eliminate the actual tumorigenic cell. Following these reports, we ini tiated studies to explor e the participation of stem-like cells in bone sarcomas. Identificati on of such a cancer stem cell could provide valuable insight into the process of tumorigenesi s in osteosarcoma and perhaps a more effective target for therapeutic intervention. In previous and ongoing work we have de monstrated that a s ubpopulation of cells in bone sarcomas share attributes with certain stem-, or stem-like cells: 1) They demonstrated the capacity for self-renewal in vitro in clonal expansion assays (they formed sarcospheres in anchorage-independent seru m starved conditions), and in vivo following serial xenotransplantation in immunocompromised mice; 2) cells from tumors were capable of being induced to differentiate along mesenchymal lineage s, and 3) they differentially expressed the embryonic stem (ES) cell-specific genes Oct-4 an d Nanog. We saw a direct correlation between the expression of Oct-4 and the functional behavi or of cells isolated from bone sarcomas. We found that in vitro as our sphere cultures en larged and differentiated, the percentage of cells

PAGE 13

13 expressing these genes decreased. In at least one patient derived os teosarcoma cell line (OS521Oct-4/GFP), tumorigenesis, as indicated by the capacity to form tumors following xenotransplantation in immunocompromised mice, appeared to be functionally linked to the ability of the cells to express an exogenous reporter construct. This construct contained the full length Oct-4 promoter linked to the coding region for green fluor escent protein GFP) and an independent SV40-promoter driven neomycin resistance cassette. This allowed positive selection of cells that acquired th e plasmid, irrespective of their ca pacity to transcribe the Oct-4 promoter. Thus, in medium supplemented with G418, all surviving cel ls would contain the plasmid and express the neomycin resistance gene ; however, only the subset of cells that were capable of activating the Oct-4 promoter se quence would fluoresce green. We found that following stable transfection of the osteosarcoma line, expression of Oct4 was heterogeneous. Only about 24% of the G418 resistant cells we re GFP+. This subpopulation of GFP+ cells had greater than 100 fold enhanced tumorigenicity relative to populations that did not transcribe the Oct-4 promoter. Those cells that in vivo spontaneously lost the capacity to express the Oct-4 reporter construct also lost their ability to form tumors following xenotransplantation. These data suggest that the tumorigenic ce lls in osteosarcoma, a somatic malignancy, have been reprogrammed to an ES cell-lik e phenotype and that this phenotype drives tumorigenicity. The goal of my research was to inve stigate the relationship between cellular differentiation and tumorigenic capacity in os teosarcoma. We hypothesi zed that a subset osteosarcoma cells, which expressed the embryonic transcription factor Oc t-4, retained a stem cell like phenotype; and because of this, these cells were re sponsible for tumor initiation. Cellular differentiation of these cells should silence the expression of this gene, resulting in a

PAGE 14

14 loss of their stem cell like qualities, thereby ne gatively impacting their tumorigenic capacity. We addressed this hypothesis through two specific aims. Aim 1: To Identify the Origin of the Heterogeneous Expression Of Oct-4/GFP in the OS521Oct-4GFP Cell Line We isolated and expanded indi vidual GFP+clones from xenograft tumors to determine if the heterogeneity in Oct4/GFP expression observed in vivo (e.g. 67% Oct-4/GFP+ vs. 33% GFP-) arose from expansion of pre-existing GFP+ and GFPcell popu lations in the cell inoculums, or represented changes in expres sion of the Oct-4/GFP reporter arising from proliferation/differentiation of Oct-4/GFP+ tumor initiating cells. We have previously shown that activity of the Oct-4/GFP reporter identifies a hi ghly tumorigenic sub-population of cells in osteosarcoma tumors. We reasoned that any changes in expression of the exogenous Oct-4 promoter could be attributed to changes in the bi ology of the cells and not to pre-existing heterogeneity in the in jected cell population. Aim 2: To Address the Role of Differentiat ion on Tumorigenic Capacity of OS5 21Oct4/GFP Clones We examined the ability of culture co nditions and agents that induce cellular differentiation of stem cells to alter expression of the Oct-4/GFP reporter in vitro, and inhibit tumorigenesis in vivo As indicated in Aim 1, we hypothesize d that change in expression of the Oct-4/GFP reporter and loss of tumorigenic potential in vivo are the results of proliferation/differentiation of Oct-4/GFP+ tumor initiating cells.

PAGE 15

15 CHAPTER 2 METHODS Derivation of Osteosarcoma Cell Lines from Patient Bio psies and Description of phOct4/GFP For each tumor sample, cells were isolated from ~1 cubic cm of tumor from an open biopsy taken from patients that have not prev iously received any form of chemotherapy. Histologic analysis of a frozen section taken from an adjacent area of the biopsy specimen was used to confirm the diagnosis of osteosarcoma prior to cell culture prepara tion. From the biopsy, the tissue matrix was enzymatically digested with 0.06g each of collagenase type I and II, and 0.3g of neutral protease for thirty minutes under agitation. The isolat ed cells were then expanded in monolayer culture. To assess the tumorigenic ity of the respective osteosarcoma cell lines, the cells were first tryps inized, washed and counted. Th e cells were re suspended in a minimal volume of saline solution, and 3x104 cells were injected subcutaneously into the backs of separate groups of 6 NOD/SCID mice. (From pr evious experience with this model system we have found that the incidence of tumor formation does not increas e significantly at cell doses beyond 3 x 104/mouse.) The animals were then be monito red for tumor growth over a period of 8 weeks. For this initial screen we worked to iden tify cell lines that formed tumors in at least 3 animals/group at 6 weeks post-injection. From cell lines that were capable of tumorigenesis at a sufficient frequency, the resulting tumors were harvested, enzymatically digested and the isolated cells pooled. The tumor cultures were then expanded and sp lit into thirds. One portion of the cells were frozen in liquid nitrogen and archived. The se cond was reseeded into culture and remained unmodified. The third portion was transfected with the ph Oct-4/GFP construct, and 24 hours later seeded into media containing G418 to positiv ely select for cells containing the plasmid. Following the initial selection (typically 70-80% su rvive), the transfected cel ls were then split on

PAGE 16

16 alternate days for 14-21 days at a ratio of 1:3 to ensu re stable transfection. The cells in culture were then expanded to obtain sufficient numbers fo r evaluation of the relati ve tumorigenicity of GFP-enriched and GFP-depleted cell populations in NOD/SCID mice. These cells were also used to establish clonal populations. Xenotransplantation and Tumorigenicity Assays The anim als used in this project were main tained according to University of Florida IACUC guidelines for tumor studies in experimental animals. All animal studies were performed according to IUCUC protocol number D780. Cells to be xenografted in to NOD/SCID mice from monolayer were first trypsinized, washed and counted, then resuspended in Optimem. The desired concentration of cells was calculated so that injection volumes equaled 100L/mouse. Fresh tumors, from which cells were to be xenografted into NOD/SCID mice, were first enzymatically digested as described above. The isolated cells were then expanded for a period of 24 hours in monolayer with standard culture media (DMEM/F-12, 500g/ml G418, 10% fetal bovine serum, 3% penstrep). Preparation of these cells for injection then followed monolayer preparation protocol. Injection protocols were as follows. Mice we re lightly anesthetized via isoflourane inhalation and, using 1ml insulin syringes, were subcutaneously inoculat ed on the dorsal side between the scapulas. Mice recovered from anes thesia on their own, and were then monitored daily for tumor formation. The skin of the back was palpated for tumor growth, and the numbers of animals with tumors, days to tumor onset, and tumor size was scored for tumorigenicity assays. When tumors reached 1.0 cm in diameter or 12 weeks post injection, they were harvested and prepared as necessary for desired assays. The lungs of a ll animals with tumors were also harvested and examined for metastases.

PAGE 17

17 Fluorescence Activated Cell Sorting and Flow Cytometry Cells to be analyzed via FACS and/or flow cytom etry were first plated at a density of 1x106 cells in Corning 25cm2 flasks, 18-24 hours prior to the assay. The cells were allowed to expand in 7.5ml of Gibco DMEM F-12 media, w ith the addition of 500 g/ml G4-18. Directly before analysis, the cells were trypsinized and washed two ti mes in a freshly prepared 0.05% PBS/BSA solution. Cells were recovered by centrifugation at 1,000RPM for 4 minutes and resuspended at a density of 1x105 cells per 500l of the PBS/BSA solution. Analysis was then performed on the BD FACS ARIA for sorting, or the BD LSR II Flow Cytometer to determine the number of cells expressing GFP, as we ll as the intensity of the fluorescence. If antibodies were used to assay for the presence of specific ce ll surface antigens, following the two initial washes, additional steps are as follows. (First, a solution of IgG was used to block nonspecific binding using 40l / 1x106 cells for 15 minutes.) Immediately following this, 10l of the desired specific phycoerythrin conjugated antibody was incubated with an aliquot of the cells Following two more 0.05% PBS/BSA solution washes, a 20 minute incubation was performed. The flow cytometer was set to record 10,000 even ts, at the same settings for all tests; FSC Voltage: E-1, FSC Amp Gain: 868, SSC Volta ge: 304, SSC Amp Gain: 1.0, FL1 Voltage: 412, FL1 Amp Gain: 1.0, FL2 Voltage: 427, FL2 Am p Gain 1.0. Compensation was adjusted accordingly for each sample, yet FL2 compensati on consistently ranged from 4.0% FL1 to 12% FL1 for optimal data acquirement. Analysis of the data included setting the is otype or negative cont rol at no more than a 3.0% error, in order to compensate for any auto fluorescence or particulate matter that may have been contaminating the sample. No gates were used to isolate any sp ecific populations in the analysis of the data, thus all recorded even ts have been included in the histograms.

PAGE 18

18 Clonal Expansion of OS521Oct-4/GFP Cells Cells from an OS521Oct-4/GFP tumor that we re heterogeneous for GFP expression were fractionated into enriched and depleted populat ions using FACS. The resulting populations were then counted, serially diluted, and plated in separate 96 well dish es at a density of one cell per well. Two hours post plating, allowing the cells e nough time to adhere to the dish, the wells were visually inspected to confirm the presence of only a single cell per well. Wells with multiple cells or those without cells were discarded at th at time. The cells were then visually checked every 24 hours to monitor proliferation. Additiona lly, if cells began to form colonies in two different areas in the well, they would be discarded, ensuring that clones used for experimentation proliferated from a single cell. The cells were incubated in standard culture media and allowed to expand for two weeks. As the populations reached confluence, they were expanded to larger wells, until th ey could be maintained in 25cm2 flasks. In vitro Osteogenic Differentia tion OS521Oct-4/GFP clones from monolayer were seeded at a density of 4x104 cells/cm2 in 24 well platesin either standa rd culture media or in StemXV ivo complete osteogenic base media (R&D Systems). The active ingredients in this media includes -glycerol-phosphate, Lascorbic acid, and dexomethasone. Samples were then cultured for 21-28 days, during which time the cells were passaged and analyzed for G FP expression by flow cytometry every 7 days. Treatment with Bone Morphogenic Proteins Two different m ethods were used to examine the effect of BMPs on Oct-4/GFP expression and tumorigenicity of osteosarco ma tumor cells: adenovi rus (Ad.) mediated transduction of BMP 2, 4, or 7, or culture in conditioned media derived from transduced cells. Adenovirus-mediated transduction of BMP 2, 4, or 7 into osteosarcoma tumor cells: For adenovirus infection, cells were isolated by enzy matic dissociation of freshly harvested tumors

PAGE 19

19 and cultured overnight. The following day cells were trypsinized and reseed ed at a density of 104 cells/cm2 in 24 well dishes in standard culture media and cultured for an additional 12-16 hours. The media were then aspirated, and the cells were washed 3 times with PBS. Cultures were then transduced in Optimem for 4 hours with Ad.BMP 2, 4, or 7 at an estimated multiplicity of infection of 10 infectious particles per cell. In control experiments using Ad.GFP this dose was found to result in transduction of 90-100% of the cells as determined by flow cytometry. Samples were then cultured for an additional five days. Cell media was collected from each sample on days 2 and 5 to determine approximate BMP concentrations using commercially available ELISAs according to the manufacturers protocol (Quantikine, R&D Systems). The respective BMP conditioned media were then stored at -200C for later use (see below). On day 7 post-transduction samples were trypsini zed and the cells were counted. 1x105 cells were transplanted into NOD/SCID mice and monitored as previously indicated for tumorigenicity experiments. Aliquots of the samples were analyzed by flow cytometry for Oct-4/GFP expression. Treatment of osteosarcoma tumor cells with BMP 2, 4, or 7 conditioned media: Additional aliquots of cells isolated from xenotransplanted tumors were init ially seeded as described above, but instead of transduction with adenovirus, th ey were incubated in the BMP conditioned media from adenovirus infected tumor cell cultures, in a 1:1 ratio with standard culture media, for 7-10 days. BMP treated and untreated samples were th en analyzed by flow cytometry for Oct-4/GFP expression and transplanted into NOD/SCID mice (1X105 cells/mouse) and monitored for tumor formation as before.

PAGE 20

20 CHAPTER 3 LITERATURE REVIEW The Biology of Cancer Cancer is typified by a culm ination of genot ypic changes in otherwise normal functioning cells that lead to malignant beha vior. These genetic mutations can be characterized as either loss or gain of function which then lead to observed cancer phenotypes. Loss of function occurs in tumor suppressor genes, which normally function to repress proliferation, repair DNA damage, or promote apoptosis. Gene silencing of tumor suppressors such as p53 is seen in a majority of human cancers (Meyers et al. 2005 ) and often arises from epigenetic changes such as hypermethylation of the CpG islands in the proxima l promoter regions of the gene (Jones and Baylin 2002). Proto-oncogenes are genes whose aberrant acti vity causes normal cells to become cancerous either because they ar e mutated or expressed at the wrong time in development. The gene products can cause increased cellular proliferation independent of extrinsic growth signals as well as resistance to apoptotic mechanisms; these in turn dr ive cancer progression (Hanahan and Weinberg 2000). Tumorigenesis generally does not occur after just a single mutation. It has been suggested that several mutations, typically between four and seven rate-limiting stochastic events, must occur in normal human cells for th em to acquire tumorigenic properties (Gibbs et al. 2002). According to Hananan and Weinberg (Hanahan and Weinberg 2000), the evolution from a normal cell to a cancer cell is the culminat ion of at least six essential alterations in cell physiology that collectively resu lt in the capacity for malignant growth. These alterations include: self sufficiency in growth signals, inse nsitivity to growth-inhibitory signals, evasion of apoptosis, limitless replicative potential, sust ained angiogenesis, and tissue invasion and metastasis.

PAGE 21

21 These are generalizations about changes that have been observed in all types of cancer. Most cancer types have hallmark genes that have been mutated, and each of these changes results in alterations in essential biological mechanis ms. Understanding these mutations can change cancer treatments into a rational science with practitioners aiming treatments a specific molecular targets, thus improving patient co mfort during treatment but more importantly increasing patient survival rate s (Hanahan and Weinberg 2000). The Biology of Osteosarcoma Osteosarcom a is a highly malignant mesenchym al tumor of bone in which the malignant cells produce osteoid. It can arise in any bone, but occurs primar ily in the juxta-epiphyseal regions of rapidly growing long bones (Figure 51). The histopathologic appearance of highgrade intramedullary osteosarcoma is one of malignant spindle cells producing osteoid and immature bone. The bone structure is disorganized and can appear as a fine lacey trabecular pattern or as irregula r clumps of osteoid, distinctly unlike normal bone formation. Classic osteosarcoma may also appear to be predominan tly fibrous or chondroid with only small areas of osteoid formation (Gibbs et al. 2002). Grossly, osteosarcoma begins as a process destructive of medullary bone which progresses to destroy corti cal bone, often with a large associated soft tissue component. The natural history of osteosarcoma is one of relentless local progression with loss of the function of the affected extremity and distant metastasis most often to the lung(Chi et al. 2004; Meyers et al. 2005).A small percentage of patients de velop bone metastases which are almost always fatal(Wuisman and Enneking 1990). Despite advances in the understanding of cancer biology, the characterization of the events that lead to the development of conventional osteosarcoma have yet to be defined. Difficultie s impeding the molecular study of osteosarcoma include obtaining living samples that can be decalcified, as well as samples that are viable after

PAGE 22

22 aggressive chemotherapies. Thus far, unlike many other cancers, there have not been specific translocations or chromosomal rearrangement events that can typify this malignancy. Two genetic lesions that ar e often seen in osteosarco ma inactivate the p53 and retinoblastoma (Rb) tumor suppr essor genes; however, because of the variety of genetic mutations involved in osteosarcoma the relative contributions of these genes have yet to be determined (Sandberg and Bridge 2003). Alterations of the RB1 gene have been shown in up to 70% of reported cases, and loss of heterozygosity for RB1 has been shown to be a marker of poor prognosis (Li et al. 2005). Over expression of the oncogene MDM2 has also been implicated as an important genetic altera tion (Miller et al. 1996). Althoug h this information has provided insight into aspects of the molecular dysregulatio n of osteosarcoma and its heterogeneous nature, to date these types of studies have been of limited value in establishing the molecular determinants of tumorigenesis or in the development of effectiv e therapies(Ragland et al. 2002). Additionally, drug resistance, primarily to methotrexate, by osteosarcoma cells, only increases the difficulty of treati ng patients. Two proposed reasons for drug resistance point to a defect in the reduced folate carrier, or alterati ons and/ or amplificati ons of the dihydrofolate reductase target. Likewise met hotrexate resistance can stem from the functional loss of the RB protein. Resistance to other chemo therapeutic drugs can also be attributed to overexpression of p-glycoprotein, an ATP depende nt transmembrane protein (Sandberg and Bridge 2003) Stem Cells Stem cells are considered to be a rare group of cells with the capacity to self-renew and generate a developmental hierarchy of differen tiating progeny. Pluripotent stem cells derived from the inner cell mass (ICM) of the blastocyst expand and differentiate to ultimately generate all the cells that comprise the adult individual. ES cells are derived from the ICM and retain the widest developmental capacity. They are pluripoten t, indicative of few epigenetic changes due to

PAGE 23

23 methylation, allowing them to differentiate into ectodermal, mesodermal, and endodermal lineages. Additionally, they maintain a stable karyotype and grow indefinitely in culture (Varga and Wrana 2005). At the time of embryonic differentiation, alle ged permanent silencing of embryonic genes is thought to occur via epigenet ic modifications of their resp ective promoter regions. Histone modifications and DNA methylati on serve to block th e reactivation of th ese genes and thus prevent dedifferentiation of somatic cells (Hattori et al. 2004; Feldman et al. 2006). Interestingly, several recent studies have challenged the apparent permanency of embryonic gene silencing by demonstrating the abili ty of differentiated mouse and human skin fibroblasts to be reprogrammed back to a pluripotent undifferentiated stat e (Meissner et al. 2007; Okita et al. 2007; Wernig et al. 2007; Yu et al. 2007). In these studies, murine fibroblasts were retrovirally transduced to express four transcrip tion factors associated with pluripotency. Ectopic expression of these proteins was sufficient to induce epigenetic reprog ramming of the somatic genome into an embryonic pluripotent state. Th e induced pluripotent stem (iPS) cells were shown to form viable chimeras and generate live late-term embryos when injected into tetraploid blastocysts (Meissner et al. 2007; Wernig et al. 2007). More recently, similar results have been obtained using human dermal fibrobl asts lentivirally transduced to express transcription factors associated with pluripotency (Yu et al. 2007). Ea ch of the generated iPS cell clones displayed human ES cell morphology, normal karyotype, and expressed cell surface markers and genes characteristic of human ES cells. All of the cl ones analyzed exhibited a demethylation pattern similar to that of human ES cells. These cells also demonstrated the capacity to differentiate into advanced cell types of each of the three primary germ layers (Okita et al. 2007; Yu et al. 2007).

PAGE 24

24 These experiments indicate that the silencing of embryonic genes in ad ult cells may not be permanent. Adult stem cells generate mature, differentiate d cells that form specific tissues (Eckfeldt et al. 2005). Adult stem cells are characterized by th eir ability to self rene w as well as the ability to differentiate into mature cells of a particular lineage (Reya et al. 2001). These characteristics allow these cells to maintain their population, wh ile also allowing a more specialized response to enable repair of injured tissues. In vivo stem cells reside in a niche, which is a specific location in a tissue where the cells can reside for an indefinite period of time and produce progeny ce lls while self-renewing (Ohlstein et al. 2004). Cellular di vision within the niche can take two different directions and must be carefully regulated. Too lit tle proliferation may l ead to a depletion of the necessary stem cell population; yet, it is possible for unchecked proliferation to result in tumorigenesis. Normally, self renewal or symmet ric division occurs when a stem cell divides into two identical daughter cells. These cells remain in the niche and serve as functioning stem cells, maintaining pluripotency and sustaining the population. Asym metric division occurs where two daughter cells are formed, but one cell stays within the niche to act as a stem cell, and the other cell leaves the niche in order to differentiate and proliferate into additional progeny. Due to different physiologic conditions, both symmetric and asymmetr ic division can occur within two different cells residing in the same niche (Yin and Li 2006). Stem cells are often identified through in vitro studies, where cell surface/cytoplasmic proteins, transcription factors, and proliferative behaviors are often the defining features observed. In these studies, the natural in vivo environments are mimicked as closely as possible through exposure to important growth factors an d morphogenetic molecules. Yet, these attempts

PAGE 25

25 at reproducing the in vivo environment fail to include many im portant events and interactions. Cellular communication through cy tokines, nonphysiological amounts of growth factors, and morphogens can have potent effects on stem cell behavior not typically observed in vivo (Steindler 2007). Cancer Stem Cell Theory The cancer stem cell theory holds that there is a sub-population of cells within a tumor which, like normal stem cells, has the ability to self-renew. These cells can divide asymmetrically, producing an identical daughter stem-like cell and a mo re differentiated cell which upon subsequent divisions generates the vast majority of the tumor bulk, which is essentially benign. This stem-like cell is responsible for initiating and maintaining the growth of the tumor and if not completely eradicated by surg ical extirpation or chem otherapy is responsible for local and distant recurrence (Figure 5-2) (Pardal et al. 2003). Along these lines, Weismann, drawing parallels between cance r stem cells and normal stem cells, has suggested that tumorigenesis can be viewed as aber rant organogenesis (Reya et al. 2001) The first definitive work describing a cancer stem cell was performed by John Dick and colleagues in studies of acute myeloid leukemi a (AML) (Bonnet and Dick 1997). They identified a rare population of human SCID leukemia initiating cells that were able to propagate AML in a xenograft transplant system. The leukemic grafts ge nerated were representa tive of the patients original disease phenotype. They demonstrated that the human AML stem cells purified from patient samples were CD34+ CD38-, resembling the normal hematopoietic stem cell phenotype. Cells from the CD34+CD38+ fraction could not tr ansfer the disease despite having a leukemic blast phenotype. This suggested that the normal hematopoietic stem cell was the target of leukemic transformation. Others have subse quently implicated stem-like cells in the

PAGE 26

26 pathogenesis of brain and breast malignancies sugge sting a broader involvement of stem cells in carcinogenesis (Ignatova et al. 2002 ; Al-Hajj et al. 2003; Hemmati et al. 2003; Galli et al. 2004). It has been suggested that cancer is a dis ease of unregulated self-renewal in which abnormal stem cells utilize the machinery of self-re newal to drive neoplastic proliferation (Pardal et al. 2003). That cancer could arise from a primitive stem-like cell or other precursor seems reasonable as it would require far fewer genetic or epigenetic alterations to effect a malignant change in a cell already equipped with the capacity for self-renew al. Several of the genes shown to play a role in the regulation of normal stem cell self renewal (WNT, Sonic Hedgehog, Notch) have been found to be active in cancer (Jhappan et al. 1992; van de Wetering et al. 2002; Lessard and Sauvageau 2003; Pasca di Magliano and Hebrok 2003; Qiang et al. 2003). Use of Agents That Induce Differentia tion for the Trea tment of Cancer Several lines of evidence sugge st that aberrant expression of the key regulatory proteins of ES cell p luripotency can direct ly contribute to tumorigenesis in several cell types. Therefore, agents that serve to inhibit the activity, or alternatively block the expression of these proteins could be beneficial in the treat ment of cancer. Along th ese lines alterations in the differentiation programs of cancer cells often resu lt in changes to their phenotype related to survival, rate of growth, loss of differentiation and the ability to proliferate and invade surrounding tissue (Hanahan and Weinberg 2000)Anaplasi a, the loss of differentiation, is associated with aggressive clinical behavior, suggesting differentiation co nfers a restraint on tumorigenesis (Thomas and Kansara 2006). In fact, apart from the pres ence of metastatic disease, the degree of differentiation of a sarcoma is the most powerful negative prognos tic indicator. Interestingly, differentiation therapy is already in clinical use. The most striki ng example may be the use of alltrans-retinoic acid (ATRA) in acute promyelocytic leukemia, felt my most to be a stem cell

PAGE 27

27 malignancy. 90% of patients can expect remission of their disease with combination therapy consisting of both conventional chemotherapy and ATRA (Ohnishi 2007). The role of BMP signaling in cellular biology is far reaching and diverse. One of these roles lies in determining cell fate choices during diffe rentiation. Data from work with BMPs in mouse ES cells shows that their role sharply cont rasts with that of anal ogous human cells, where neural differentiation is blocke d by BMPs in mice, yet induced in humans (Varga and Wrana 2005). BMPs are secreted proteins and can di rect mesenchymal stem cells (MSCs) to chondrogenic and osteogenic cell lineages, and in the presence of fibroblast gr owth factors, they can direct ES cells to differentia te into the trophoblast lineage in humans. BMPs are part of the TGFsuperfamily, and transmit sign als through a defined pathwa y. The BMP binds tightly to BMPI/II receptor heterodimer upon which the type II receptor phosphorylates the type I receptor. In the cytoplasm, the downstream effect of BMP binding causes pho sphorylation of BMP RSmads, specifically 1, 5, and 8, which complex with Smad4. The heteromeric complexes then translocate to the nucleus where they can regulate transcription either directly or in concert with other transcription factor s (Varga and Wrana 2005). BMPs may have an important role in controll ing the biology of stem cell cancers as they play a crucial role in early stem cell development, as well as self-renewal; losing the ability to regulate these functions may lead to tumorigenesis (Varga and Wrana 2005). In support of this are studies showing that mutation of the BMP I receptor is a huge risk factor in developing gastrointestinal cancers. Expa nding the potential of differen tiation therapy beyond leukemias, a recent report by Piccirillo et al. (Piccirillo and Vescovi 2006) showed that stem-like, tumorinitiating cells isolated from human glioblastoma s when incubated with certain BMPs increased

PAGE 28

28 the expression of markers of neural different iation, and showed decreased proliferation and tumor formation when transplanted into experime ntal animals. Their work implies that certain populations of tumor stem cells retain an abil ity to respond to normal signals of maturation induction, and efforts to devise therapies to differentiate cancer cells might be fruitful. Histone deacetylases (HDAC) play a major role in the epigenetic ch anges which regulate gene expression within a cell. HDACs catalyze the removal of acetyl groups, and thus stimulate chromatin condensation and promote transcriptional repression. Transcript ional repression can translate to the loss ex pression of tumor suppressor genes, contributing to the formation of cancer. HDAC over-expression has been observed in co lon, breast, prostate, and other cancers. Due to their wide-spread involvement in cancer, HDACs have become a novel target for therapy. HDAC inhibitors, which include a variety of therapeutic agents, can lead to the reversal of epigenetic silencing. HDAC inhib itors come in the form of s hort chain fatty acids, hydroxamic acids, benzamides, and cyclic tetrapeptides. The act ions of these drugs a ffect cellular processes such as inducing cell cycle a rrest, stimulating tumor cell deat h, and promoting differentiation. HDAC inhibitors induce cellular differentiation and promote cell cycle arrest at the G1/S checkpoint. Clinical trials have shown that HDAC inhibitors ha ve anti cancer activity and are being tested as either monothera pies or in combination with chemotherapy (Carew et al. 2008). Preliminary Results Cells Isolated from Bone Sarcoma Cult ures Exhibit Stem-Like Attributes W e initiated a preliminary series of experiments to explore the existence of stem-like cells in bone sarcomas. The culture system or iginally used by Reynolds and Weiss (Reynolds and Weiss 1992) to generate neurosphere clone s from adult mammalian brain, has since been found to isolate cells possessing attributes of stem and progen itor cells. The stressful growth conditions of this system were found to positiv ely select for primitive cells by eliminating the

PAGE 29

29 differentiated cells, which are unable to surviv e. Similarly, suspending dissociated cancerous tissue in semi-solid media without serum select s primitive clonogenic cells that can be expanded and give rise to different classes of cells. This system enables isolation of stem-like cells from malignant human brain tumors by exploiting anchorage independence, serum starvation and necessary pleiotropic growth fact ors(Ignatova et al. 2002). Simila r sphere culture systems have been used to identify tumor stem cells from both brain and breast maligna ncies that are capable of self-renewal in mouse models (Al-Hajj et al. 2003; Hemmati et al. 2003; Singh et al. 2003; Singh et al. 2004). We found that all bone sarcoma cultures formed spherical colonies (sarcospheres) at a frequency of 10-2 10-3 similar to that reported by others for brain and breast malignancies(Ignatova et al. 2002; Al-Hajj et al 2003; Hemmati et al. 2003). Sarcospheres were also generated at a similar frequency from fresh tumor dissociates produced at the time of biopsy. Furthermore, all cultures tested dem onstrated the capacity for self-renewal by the formation of secondary spheres at a similar or increased frequency of approximately 10-2. We also examined these cultures for expression of Oct-4 and Nanog transcription factors found to be indispensible for the maintenance of pl uripotency and self-renew al in ES cells. Using semi-quantitative RT-PCR we found detectable transcripts in each culture type for both transcription factors (Figure 5-3). Immunohistochemical staining of sarcospheres from paraffin sections enabled detection of both Nanog and Oct-4 in similar patterns. Inte restingly, the staining of the smaller spheres showed a high proportion of cells that were positiv e for both transcription factors. As the spheres increased in size and cell number, the cells s howed increasingly greater heterogeneity of immunostaining with a lesser percentage expre ssing these ES cell markers. These observations

PAGE 30

30 suggested a situation whereby during early sphere formation relatively primitive cells undergo a transition from symmetric to asymmetric di vision and thereupon produce daughter cells of various states of differentiation. Following these observations, we addressed whether Oct-4 and Nanog are expressed in actual tumor tissue. For this, paraffin sections from eight bone sarcoma patients were evaluated using immunohistochemistry (Figure 5-4). Nanog and Oct-4 nuclear staining was observed in seven of the eight tumors studied. In each case, as determined by histologic criteria, the stained nuclei were of malignant cells, and not from infiltrating normal cells. Among different tumor specimens, the number of Oct-4 and Nanog positive cells varied considerably. Positive Oct-4 staining ranged from a few percent of the cells in some tumors to up to 25% in others. Nanog staining was also quite variable ranging from ~1% to nearly 50% in certain samples. These results showed that subpopulations of cells in bone sarcomas express regulatory proteins typically restricted specifi cally to embryonal cells. We reasoned that if bone sarcomas express some of the molecular machinery of ES cells, they might, in addition to mesodermal genes, express genes from endodermal and ectodermal lineages. RT-PCR analyses of mRNA from adhe rent and sarcosphere cultures revealed expression of Gata-4, Gata-6 and alpha fe toprotein (AFP), indicative of endodermal differentiation (Figure 5-5a). Expression of -III tubulin RNA was seen as well, which is believed to be a marker of neural-ectoderm, bu t has also been demonstrated in some poorly differentiated malignancies (Katse tos et al. 2003). Using Western blot analyses, expression of AFP and -III tubulin was demonstrated at the protein level in each of seven cultures tested (Figure 5-5b). -III tubulin was also detected in paraffin sections of tumors using immunohistochemistry and in cell culture sample s by immunocytochemistry (Figure 5-5c-f). At

PAGE 31

31 this point, we do not know the position that a putative bone-sarcoma stem cell might occupy in the stem cell hierarchy. It is possi ble that the expression of genes such as Gata-4 and Gata-6 is simply a consequence of a larger global pattern of dysregulated gene expression in these tumor cells. An alternative explanati on is that these genes are indi cative of aberrant pluripotent differentiation of cancer stem cells. Regardless, the detection of expression of ectodermal and endodermal genes implies that the tumor cells ar e not effectively lineage restricted and suggest epigenetic reorganizat ion of the genome in bone sarcomas. Although the bone sarcomas we have studied appear to express transcription factors associated with pluripotent ES cells as well as genes of endodermal and ectodermal lineages, the histologic phenotype of these tumors is by definition one of ar rested mesenchymal differentiation. Thus, it seems reasonable to exp ect that a cancer stem cell in bone sarcoma would arise from a mutagenic lesion in a mesenc hymal progenitor. Therefore, we examined the respective cultures for the presence of cells wi th characteristics of mesenchymal stem cells. Although the precise cell-surface phenotype of an MSC has not been determined, MSCs have been found to variously express the cell surf ace proteins Stro-1, CD90, CD29, CD73 CD44 and CD105(Simmons et al. 1994; Stewart et al. 1999; Li et al. 2005). Using immunocytochemistry and flow cytometry, we screened our tumor cell cultures and found that al l the cultures expressed these antigens, but to different degrees. Although differences in the intensity of immunoreactivity of specific surface antigens we re seen between cell cultures, within any particular culture the cells appeared largely homogenous for expression of these antigens. Unfortunately the apparent uniformity of the cu ltures with regard to expression of surface antigens precluded fractionation of the cells base d on surface antigen, and thus could not be used strategically as a potential method to explore markers of tumorigeni c cells. To determine if cells

PAGE 32

32 within the cultures were indeed multipotent, we attempted to induce adherent cultures of bone sarcoma cells to differentiate along two distinct mesenchymal lineages by culture in osteogenic and adipogenic media. Within each adherent culture tested, we observed discrete foci of mineralization in cells grown in osteogenic medium, and fields of lipid laden cells in those grown in adipogenic medium (Figure 5-6). Cumulatively, the data from the previous s ections suggested that at least subpopulations of cells in bone sarcomas were capable of self -renewal, expressed transc ription factors of ES cells as well as expression of ectoand end odermal genes, and showed the capacity for differentiation along multiple mesenchymal lin eages. Altogether these data support the involvement of stem like cells in bone sarcomas. Development of an In vivo Model to Examine the Role of Stem-Like cells in the Pathogenesis of Osteosarcoma To explore in m ore detail the functi onal relationship between stemness and tumorigenicity of osteosarcoma, we focused our studies on the OS521 line, derived from a high grade, poorly differentiated human osteosarcoma. This cell line was f ound to cause robust tumor formation following subcutaneous xenograft into the backs of NOD/SCID mice. In initial experiments we found that de livery of as few as 3 x 104 cells in saline suspension reproducibly produced tumors of >1 cm diamet er in 4-6 weeks following inje ction. Similar to that observed from the biopsy of the original patient, tumors arising from the xenograft showed clear evidence of osteoid, recapitulating the characteristic pheno type of an osteosarcoma. Characterization of the OS521 culture for several cell surface markers showed that the cells in monolayer were comprised of a largely homogenous population without striking differences with regard to cell surface antigens. The cells were MHC class I+, CD90+, and NCAM+. Interestingly they were uniformly strongly positive for expression of CD 44, a marker of breast cancer stem cells, and

PAGE 33

33 negative for the presence of CD133 (Figure 5-8b ) a marker of stem cells in colon, brain and prostate cancer. Expression of an Oct-4 Prom oter/GFP Reporter Constru ct That Selectively Identifies Cancer Stem Cells in Osteosarcoma In an effort to selectively visualize and tr ack living cells in culture that express ES cellspecific genes and determine their relative partic ipation in tumorigenesis, we transfected the OS521 cells in monolayer with the plasmid construct phOct-4/GFP, (a generous gift from Dr. Wei Cui of the Rosalin Institute, UK) containing the human Oct-4 promoter sequence (spanning -3917 to +55, relative to the transcription start site) linked to the coding sequence for enhanced green fluorescent protein (Figure 5-7). This plasmid also contains an independent SV40promoter driven neomycin resistance cassette, which allows positive selection of cells that have acquired the plasmid construct, irrespective of th eir capacity to transcri be the Oct-4 promoter. Thus in medium supplemented with G418 all surviving cells will contain the plasmid and express the neomycin resistance gene; however, onl y the subset of cells that are capable of activating the Oct-4 promoter sequence will fluoresce green. Ge rrard et al.(Gerrard et al. 2005)showed that in human ES cells stably tran sfected with this construct, GFP expression driven by the Oct-4 promoter faithfully represen ted expression of Oct-4 in undifferentiated ES cells and during their differentiation. In thes e studies, GFP expression co-localized with endogenous Oct-4 protein as well as surf ace antigens SSEA-4 and Tra-1-60. Neural differentiation of the cells, as well as targ eted knockdown of endogenous Oct-4 expression by RNAi, down-regulated GFP and correlated closel y with the reduction in endogenous Oct-4 protein. Following transfection of the OS521 cells with phOct-4/GFP and positive selection of transfectants in media containing G418 we characterized Oct-4 driven-GFP expression of cells in

PAGE 34

34 monolayer using fluorescence microscopy and flow cytometry. Somewhat surprisingly, despite the apparent homogeneity of the cells in culture with regard to cell surf ace proteins, we found that following stable transfecti on of the OS521 line only about 24% of the G418 resistant cells were GFP+ (Figure 5-8a). To begin to determin e the relative participation of the GFP+ cells (those that transcribe the Oct-4 promoter) in tumor initiation, we injected 3 x 104 cells from the total neo resistant cell populati on (all cells both GFP positive and ne gative) subcutaneously into the backs of 6 NOD/SCID mice; the same dose of untransfected OS521 cells were injected into a separate group of 6 animals. At 3-5 weeks post-injection, tumors >0.5 cm diameter had formed in 5/6 and 4/6 animals of the two respective groups, which indicated that transfection of the phOct-4/GFP plasmid did not adversely influen ce the tumor-initiating potential of the OS521 cells. Tumors that were 1.0 cm diameter were harvested. Tumors from animals receiving the phOct-4/GFP transfected cells were brightly fluorescent under UV light, while tumors from animals receiving untransfected OS521 cells showed no evidence of GFP expression (not shown). Histologic section showed large clusters or foci of GFP+ cells distributed throughout the tumor mass (Figure 5-8a). Fo llowing harvest, the tumors were dissociated, and the cells recovered were charac terized for GFP expression by flow cytometry. The proportion of GFP+ cells isolated from the tumors had increased to ~67%, nearly 3-fold over that observed in monolayer culture. There was no a pparent change, however, in th e expression of the various surface antigens with respect to the GFP positiv e and negative cell populations (Figure 5-8b). Altogether, these results suggest ed a selective amplification of cells that express the Oct-4 promoter during tumorigenesis. To determine the relative tumor initiating cap acity of the cells that expressed the Oct-4 promoter construct versus those that did not (i .e. GFP+ cells vs. GFPcells, respectively), the

PAGE 35

35 cells recovered from the harvested tumors were pooled and fractionated by FACS into GFPenriched and GFP-depleted popula tions, as shown in Figure 5-9a-c Subsequent flow cytometry analysis of a portion of the respective fractions s howed that in the enriched fraction ~92% of the cells were GFP+; in the GFP-depleted fraction the number of GFP+ cells was reduced to about 3%. From a starting dose of 3 x 104 cells, we injected ten-fold dilutions of the respective fractions, as well as equivalent numbers of unf ractionated, phOct-4/GFP transfected cells, into individual groups of NOD/SCID mice at 8 animals/group and examined the rate of tumor formation. The GFP-enriched fraction proved to be signif icantly more tumorigenic (>100-fold) than the GFP-depleted fraction (F igure 5-9d). At the 3 x104 cell dose it produced tumors in all mice with a mean time to onset of 22 days, while the GFPdepleted fraction only produced tumors in ~60% of the mice with a mean time to onset of over 6 weeks. At 3 x 103 cells, again, all 8 of the animals receiving cells from the GFP-enriched fraction developed tumors, with a mean time to onset of 34 days. For the GFP-depleted group, on ly 1 of 8 mice develope d a tumor over the 90day time course. At the 3 x 102 cell dose none of the mice from the GFP-depleted group developed tumors, while all of the animals receiving the GFP-en riched cells developed tumors, with a mean time to onset of 42 days. At this dose only 3 of 8 animals receiving unsorted/phOct4/GFP transfected cells formed tumors, with a mean time to onset of ~60 days. Visualization of the freshly excised tumors using inverted fl uorescence microscopy showed that all tumors formed in all groups were highly GFP+. Following dissociation, flow cytometric analysis of the recovered cells showed that tumors formed from the GFP-enriched fractions were comprised of ~70-80% GFP+ cells, while those from the GFPdepleted fractions were ~50-55% GFP+ (not shown).

PAGE 36

36 Interestingly, we passaged the GFP+ cells th rough at least 3 rounds in mice, whereby the cells were injected, harvested from tumors, enri ched and reinjected. At the 300 cell dose (~the lowest dose that we can reasonably expect to deliver) we found that the tumors appeared to increase in virulence with passa ge, producing tumors with shorte r time to onset and more rapid growth rate. We also noted the formation of multiple local tumor nodules following a single injection. Analysis of the lungs of these mice using inverted fluorescence microscopy showed clear evidence of metast ases, with clusters of GFP+ cells readily identified throughout. These data demonstrated that the cells that expresse d the Oct-4 promoter c onstruct displayed selfrenewal in vivo and further supported the participation of a cancer stem cell in tumorigenesis of osteosarcoma.

PAGE 37

37 CHAPTER 4 RESULTS Rationale for Aim 1 Based on previous data, we found that our e xogenous Oct-4/GFP repor ter could identify a subpopulation of cells with enhanced tum or fo rming capacity. These cells were capable of activating the Oct-4 promoter whos e gene product is responsible for maintaining a pluri-potency in ES cells, hinting to the exis tence of a stem like tumor cell in human osteosarcoma. These cells were 100 fold more tumorigenic than those not expressing the Oct-4 prom oter, and the resulting tumors were heterogeneous in nature for GFP expression. We questioned whether this heterogeneity was due to existing differences in the proliferation rates of the GFP+/subpopulations, since fractions from FACS contai ned up to 8.0% cross contamination (Figures 59b, 5-9c), or if the differences represented change s in the ability of certa in cells to express the Oct4 promoter. We aimed to investigate this by creating clonal populatio ns of the respective phenotypes and repeating the tumorigenicity assa ys. Clonal expansion w ould create a population of cells that were homogeneous in nature, allowing any changes in Oct4/GFP expression to be attributed to changes in the bi ology of the progeny cells and not to selective amplification of GFP+ and GFPsubpopulations. We reasoned that in vivo tumor initiation occurs from GF P+ cells, which proliferate and differentiate to form cells that have lost Oct-4 expression as well as th eir tumorigenic potential. These GFP+ cells might be thought of as m oving away from the stem like phenotype and becoming more specialized, as the transform from GFP+ to GFP-. Along these lines, we would not expect to see cells lacking the expression of the Oct-4 promot er to spontaneously gain the ability to express the gene either in vivo or in vitro because this would signify an act of dedifferentiation towards a more stem like state. Du e to epigenetic silencing, this is usually an

PAGE 38

38 irreversible phenomenon. Using cells of a pure clonal population would allow us to observe any shifts in Oct4/GFP expression and to determine if the GFPcell populatio ns arise from GFP+ populations and not vice versa. A dditionally, it would show that the appearance of a GFPpopulation was a direct result of biological changes in the tumorigenic cell and not a shift in the dynamics of the population due to proliferative differences. Rationale for Aim 2 If in fact, cancer stem cells in osteosarcoma rely on the molecular machinery of ES cells that function to maintain pluripotency, then it would be rational to e xpect that the forced induction of cellular differentiati on in these cells should either reduce the expression of these proteins, or alternatively modulat e their effects, and in so doing reduce tu morigenicity. Along these lines, Feinberg and others have suggested that cancer stem cells from certain types of tumors retain the ability to respond to extrinsi c differentiation signals (Jones and Baylin 2002). In exploring the effects of differentiati on agents on osteosarcoma stem cells, we examined if exposure of OS521Oct-4/GFP clones in monolayer to osteogenic differentiation media would stimulate cellular differentiati on and change the expression of osteogenicassociated genes in these cells. If we could associate an increase in osteogenic associated gene transcription with a lack of expression of Oct4, then we could assume that loss the stem like phenotype has occurred along with the differentiation event. This would be important in linking the lack of Oct-4 expression with a cell that has e xperienced differentiation. We also chose to use BMPs 2, 4 and 7, which are well known inducers of osteogenic differentiation in human mesenchymal stem cells, to perform tumorigenici ty assays with. If treatment with these agents induc es a loss of Oct-4 expression via differentiation, resulting in a loss of tumorigenic capacity, we could assume that the tumor initiating potential of osteosarcoma depends on the retention of a stem like phenotype.

PAGE 39

39 If we find that any of these differentiating agents decrease tumorigenicity, these data could form the platform for the development of novel therapies for this devastating disease. Clonal Expansion of OS521Oct-4/GFP cells In order to investigate our hypothesis that heterogeneity can be derived from a clonal population, we had to first create clones from our patient derived OS521 cell line. After completing this task, we observed the following characteristics of the different clones. For the GFP+ fractions, we observed that 90% of the 96 wells held viable cells following serial dilutions and allowing cellu lar adherence. Any wells that were observed to contain more or less than one cell were immediately discarded. Over the course of two w eeks, all of the GFP+ cells began to divide, yet three of the clones di splayed more aggressive proliferation and had reached confluency. These clones, named A1, S1, and T1, were allowed to expand for experimental purposes. At six weeks the S1 and T1 clones had divided enough to populate a 25cm2 flask. The A1 clone inexplicably stopped dividing and expired four weeks into the expansion process. Fluorescence microscopy and flow cytometry were used to confirm the homogeneous expression of G FP in both clones. Figure 5-10a shows that the S1 clonal population in monolayer was 98% positive for GFP expression, which was also comparable to that seen in the T 1population in culture. Likewise, 90% of the GFPpopulation surviv ed initial plating, ye t after two weeks the cells had failed to achieve pr oficient division rates. Additi onal time was allotted for the expansion of the GFPclones, but none of the cells succeeded and eventually died by 8 weeks. Fluorescent microscopy confirmed the absence of GFP in any of the cells, but there were not enough cells in any of the wells to perform a flow cytometry assay.

PAGE 40

40 Flow Cytometric Analysis of Clon es and Clonally Deriv ed Tumors We wanted to determine if the tumors formed from our largely homogeneous clonal populations would recapitulate the heterogeneity observed from the parental OS521 lines. This would allude to the ability of the cells to lose the expression of Oct-4 and proceed towards a more differentiated phenotype. Following delivery of 3 x 104 cells of the respective clon es into NOD/SCID mice, tumors readily formed in animals within 2-3 weeks. An alysis by flow cytometry showed that the cells recovered from the tumors were markedly hete rogeneous with respect to GFP expression, as shown in Figure 5-10b, with a clear reduction in the mean intensity of fluorescence. On first passage (Figure 5-10b), tumors e xhibited fluorescence over a range of three logarithmic orders. The histogram shows a clear shift to the left in fluorescence intensity compared to that of the parental in vitro population, though there is no clear distin ction between the two populations in the histogram. Interestingly, serial passage of unsorted cells from fres hly dissociated tumors derived from these clones resulted in the pr oduction of discrete GFP+ and GFPpopulations. They were composed of approximately 60% G FP+ and 40%GFPby the third passage (Figure 510c). The GFP+ positive fraction more closely re sembled that of the parent clonal population, with the majority of the cells havi ng a fluorescence intensity greater than104. We questioned whether these more discrete GFP+ and GFPpopulations might also be distinct with regards to tumori genic capacity. To this end we harvested passage three tumors, which are composed of the most segregated populations, from one of the clones (S1) and fractionated the cells by FACS into GFP enriched and depleted populations. We then transplanted them as previously described for tumorigenicity e xperiments. We observed results similar to those shown for our parental cultures /tumors using these clonally derived populations regarding the frequency and kine tics of tumor formation by the two fractions at the indicated

PAGE 41

41 doses (compare Figure 5-9d and 5-11b). Analysis of the resulting tumor cell populations for Oct4/GFP expression by flow cytometry showed th at GFP enriched fractions formed tumors composed of 85% GFP+ cells, and GFP depleted fractions formed tumors composed 55% GFP+ cells (Figure 5-11c). The data to this point show that Oct4/GFP+ and Oct-4/GFPcells generate heterogeneous tumors upon transplant, and are capable of extensive self-renewal in vivo : all of which are traits associated with cancer stem cells. However, in this instance the cancer stem cell is not a rare, slowly dividing stem-like cell but a highly proliferative cell whose active cell division directly supports the growth of the tumor. Culture of OS521Oct-4/GFP Clon es in Osteogenic Media To attem pt to link differentiation to the loss of the expression of Oct-4, we took OS521Oct-4/GFP clones from monolayer and atte mpted to stimulate differentiation with commercially available osteogenic media for a pe riod of 21 days. Culturing of human ESCs and MSCs in these conditions has been shown to resu lt in the activation of genes associated with osteogenic differentiation and matrix deposition (Gronthos and Simmons 1995). We believe that a loss of Oct-4 expression during the incubation would show that Oct-4/ GFPcells are indeed more differentiated than their fluorescent counterparts. In order to prevent rolling up of the confluent monolayers, the cultures were trypsinized and reseeded every 7 days to remove dead cells and for analysis by flow cytometry. We were unable to detect significant matrix de position or changes in the expression of genes associated with osteogenesis in samples cultured in osteogenic media by staining of the confluent monolayers with Alizarin Red S and RT-PCR, re spectively (not shown). As shown in Figure 512, OS521Oct-4/GFP clones cultured in osteogenic media for 14 and 21 days were composed of both GFP+ and GFPcells. On day 14 there were 57% GFP+ and 43% GFPcells, and by day 21

PAGE 42

42 the population was composed of 64% GFP+ and 46% GFPcells. Controls grown under standard culture conditions remained uncha nged for GFP expression (95%). These cells were not used in tumorigenicity assays. Culture of OS521Oct-4/GFP Clonally Derive d Tumor Cells in BMP Conditioned Media We hypothesized that induci ng differentiation through various agents, nam ely BMPs, would adversely affect the tumorigenic capacity of the OS521 clonall y derived tumors. We attempted to stimulate differentiation using ei ther adenovirus transduction or exposure of the cells to BMP treated media. Subsequent inject ion of the treated cells yielded the following. BMP conditioned media was generated by infect ion of cells from tumors derived from clone S1, with adenovirus encoding the cDNAs for either BMP 4, or 7 as described. BMP 2 appeared to be toxic to the cells and they did not survive the tran sduction; as a result this portion was omitted from the experiment. Analysis by ELISA for BMP concentrations of cell free supernatants from both infections showed hi gh concentrations of BMP protein (200>ng/ml). Approximately 3 days following the start of induction we first noted by visual observation that BMP-treated cultures adopted a flatter cellular morphology and took longer to grow to confluence than untreated controls. Unsorted tumor cell populations cultured in BMP 4 or -7 conditioned media for 7-10 days also resulted in alterations in the proportion of the GFP+ and GFPfractions, with a clear decrease in mean fluorescence intensity (compare 13A and B). Interestingly this effect was more pr onounced in the BMP 4 treated samples. Tumors formed by transplantation of these cells (1.0X105 cells/mouse) also showed differences in the relative proportions of GFP+ and GFPcell fractions (Figure 5-13c). We observed that for the BMP 4 treated sample 10% of the cells were GFP+ following transplantation. This result is in stark contrast to the BMP 7 treated sample, where 70% of the cells were GFP+. This result contradicted what we had consistently observed for untreated

PAGE 43

43 samples following transplantation; the GFP+ cell fraction usually expands or remains constant (compare Figures 5-13c and 5-11c). Time to tumor onset, penetrance and kinetics were all comparable to tumors formed by untreated controls.

PAGE 44

44 CHAPTER 5 DISCUSSION Our group has identified a putative CSC popul ation in osteosarcoma using an Oct-4 promoter driven GFP reporter. We observed that the GFP enriched fract ion of xenotransplanted tumor cells was roughly 100-fold more tumorige nic than the GFP depleted fraction when transplanted subcutaneous ly into NOD/SCID mice. We believe that silencing of the Oct-4/GFP transgene and concomitant loss of tumorigenic ability is the result of differentiation/proliferation of these cells during tumor initiation and growth. In the present study we have focused on tw o distinct aims. We first showed that individual Oct-4/GFP+ clones we re capable of generating heter ogeneous tumors composed of both GFP+ and GFPcells. This confirms that this cell retains a stem cell like phenotype and possesses key stem cell attributes, such as the abi lity to self renew and recapitulate heterogeneity of the initial population. Additi onally it supports our hypothesis th at changes observed in our experiments were due to biological effects and not the heterogeneous na ture of the collective population. Secondly, we cultured cells from these tu mors in osteogenic media or in the presence of BMP 4 or 7 and found that we were able to induce silencing of the Oct-4 promoter. We then took these treated cells and performed tu morigenicity assays. At a dose of 105 unsorted cells tumors formed within 4 weeks however there were distinct differences between tumors formed by BMP 4 treated samples compared to samples exposed to BMP 7 regarding days to tumor onset and the proportions of G FP+ and GFPcells fractions. In order to obtain a truly ho mogeneous population to which variation could be attributed to biological changes, we clonally expande d the cells from OS521 Oct-4/GFP cell line. Successful expansion of the population expressing the Oct-4 gene product can be attributed to the fact that these cells maintain their pluripotency in vitro and thus their stem like qualities. The

PAGE 45

45 ability of stem cell like cancer cells to divide both symmetrically and as ymmetrically can explain the expansion and restoration of the cell population. Symmetric division, which is what we saw in our clones in vitro contributed to the homogeneous nature of the clonal population. This makes sense since the cells still expressing Oct-4 maintain an ES cell li ke phenotype. We did not observe any asymmetric division in vitro, leading to more differentiat ed cells, until the cells were exposed to the in vivo environment in our tumorigenicity assays. The in vitro environment apparently lacks the necessary cues to induce change or gene silenc ing in the population. The lack of Oct-4 expression in the GFPcells allude to th e reason why they cells were not able to clonally expand. Loss of Oct-4 expre ssion points directly to a cell which no longer transcribes the Oct-4 gene, which is crucial in ma intaining pluripotency in ES cells. Along these lines, a cell that is no longer behaving as a st em cell would be unable to divide in the manner required to reinstate a stable self maintain ing population. Phenotypic characteristics of the clones that were established in vitro showed that they were 98% positive for GFP expression. Upon multiple passages of these cells in vitro the relative homogeneity remained at the highly uniform levels. Loss of Oct-4 promoter activity was not observed in our cultures, which we attribute to the absence of the proper cues that a cell would normally be exposed to in an in vivo environment. Basic media and cell/cell contacts do not provide stimulation for the cells to differentiate or initiate any of the associated epigenetic changes, and they symmetrically divide to maintain their population and existence. Changes in GFP expression can be induced through serial passage in vivo in the NOD/SCID mouse model. Tumor progression is often viewed as driv en by epigenetic plasticity and genomic instability. Epigenetic heterogeneity within stem and progenitor populations could in part account for tumor cell heterogeneity. From these observations we postulate that one

PAGE 46

46 possible explanation for our results is that Oct-4/GFP activity iden tifies tumor cells with a stemlike epigenetic signature with the capacity for infinite self-renewal. Loss of Oct-4/GFP expression may be the result of differentiation of the cancer stem cell population during tumor growth and then an adoption of a more differentiate d epigenetic signature. This leads to a loss of stem-associated activities and the ability to initiate and sustain tumorigenesis in vivo As the number of serial passages increased, we observed that the gap between GFP+ and GFPpopulations increased as we ll (Our results clearly show th at from initial injection and a nearly pure GFP+ cell populati on, we can form tumors that are comprised of a substantial number of GFPcells.) We believe that this phenomenon occurs when the tumor initiating GFP+ cells are exposed to the in vivo micro-environment providing contact with the extracellular matrix, cellular cues of prolifer ation, and growth and proliferati on factors. These cells are now stimulated to divide asymmetrically, where ch anges induce a loss of Oct-4/GFP expression, and thus increase the number of GFPcells. These di fferentiated cells accumulate and contribute to the bulk of the tumor mass, while the less differen tiated GFP+ cells maintain their primitive stem like qualities and their tumorigenic potential. Add itionally, these changes cause the GFPcells to lose their tumor initiating potential. With passage in vivo the clonally expanded Oct-4/GFP+ cells diverge, eventually adopting a profile of GFP expression simila r to the parental line (Figure 5-10). They display a populati on composed of both GFP+ and GFPcells. This observation supports two concepts consistent with a cancer stem cell. First, this shows that tumors formed from the clonally expanded line are capable of recapitulating the heterogeneity of the Oct-4 expression of the original population. More importa ntly though, this suggests that GFPcells are derived from GFP+ cells, suggesting that any GFPcells arise as a result of cellular differentiation.

PAGE 47

47 FACS of clonally derived tumor cells into GFP enriched and GFP depleted fractions, followed by xenotransplantation over a range of cell doses, showed that cells that continued to express the Oct-4 promoter sequence were significantly more tumori genic than cells that had lost that capacity (Figure 5-11). Consistent with results from the non-clonally derived cell population, the GFP+ cell fraction produced tumors in all anim als at all doses. Administration of as few as 300 cells was sufficient to cause tumor formation in all animals tested. The GFPfraction in contrast formed tumors in only 50% of the mi ce at 30,000 cells and 25% at 3,000 cells. Analysis of the resultant tumors from each cell fraction showed that first passage tumors from the GFP+ fraction were comprised of 85% fluorescent cells Tumors from the GFPfraction were also highly GFP+ with about 55% of the cel ls showing fluorescence above background. As before, we believe the tumors arising from the GFPpopulation in all probability are the result of low level co ntamination of GFP+ cells present in the depleted fraction. Foremost, the majority of the cells in tumors generated from the GFP depleted fraction are GFP+. In our exprerience, we have never observed cells reac quire the capacity to express the Oct-4 reporter once lost. Furthermore, as seen in Figure 5-11b a dose of only 300 GFP+ cells is sufficient to form tumors in all mice. Since typically between 3-8% of the cells in the respective fractions are contaminants of the opposing fraction, we would expect between 900-2400 GFP+ cells in the 30,000 cell dose and 90-240 in the 3,000 dose, numbers at both dilutions that approach or exceed the minimum established tumorigenic dose. Cells from the GFPfraction only formed tumors in doses of 3x103 or greater, and in 50% or less of the mice. Tumor formation in this c ondition is most likely due to cross contamination of GFP+ cells in the GFP depleted fraction. Figure 5-9a-c shows that the level of cross contamination can reach 8.0% in the fractiona ted samples. Along these lines, a dose of 3x103

PAGE 48

48 GFPcells contains enough GFP+ cells to initiate tumor formati on, and doses of 300 cells or less do not contain enough cells with tumor forming cap acity. Additionally, we see that the tumor is composed of 55% GFP+ and 45% GFPcells (Fi gure 5-11c). This hetero geneity is also most likely due to the cross contamina tion of GFP+, putative tumor init iating cells that were injected. This small subpopulation would be required to di vide more times to form a tumor, when compared to tumors formed from a GFP enriched fraction. As this small number of GFP+ cells divide, they will lose the ability to express the Oct-4 promoter, resulting in a larger population of GFPcells in the tumor. As the population of the serial passed clones regressed back to resemble the Oct-4/GFP expression of the parental line, we found that we could diverge the tw o populations further by exposure of the cells to osteogenic differen tiation media as well as BMP 4 and BMP 7. Exposure to osteogenic media resulted in an overall reduction in GFP expression from the Oct-4 promoter producing a heterogeneous population of approximately 60% GFP+ and 40% GFPcells. Despite these changes in transcriptio n of the Oct-4 promoter were unable to detect significant matrix deposition by stai ning of the confluent monolayers with Alizarin Red S, or changes in the expression of genes associated with osteogenesis using RT-PCR. This may have occurred for several reasons. First, OS521 is classified histologically as a highly dedifferentiated osteosarcoma and thus may be relatively resi stant to certain differentiation cues. Along these lines, it is possible that this cell line may require a more extended incubation to stimulate the cells to begin secreting matrix. It is also possible that deposited extracellular matrix components were lost during passage of the confluent cult ures during the extended incubation. Additionally, changes in the expression of genes associated wi th osteogenesis may not have been detectable

PAGE 49

49 because of the heterogeneity of the treated cell population. Fract ionating the cells into their respective GFP+ and GFPpopulations, before RT-PCR analysis, may have resolved this issue. .Differentiation from BMP stimulation occurs when the protein causes the transmembrane BMPI receptor to phosphorylate the BMPII receptor and form a heteromeric complex. Intracellularly, the BMPII receptor phos phorylates the SMAD1/5 proteins, which are cytoplasmic. These, in turn, induce the upregul ation of the transcription factors RUNX2 and Osterix, which are critical to osteoinduction. BMPs are also known to activate MAPKs which upregulates RUNX2. We postula te that by activating these pathways, we have induced osteogenic differentiation which causes a loss of the expression of our transgenic reporter gene. This in turn would allow a popul ation of more differentiated but less tumorigenic cells to increase in number, consistent w ith the increased GFPpopulation. Upon xenografting the cells at a dose of 100k cells/mouse, we observed tumorigenesis in 2/3 of the mice inoculated in both BMP conditions. We attribute this to th e large number of cells injected. It should be noted th at in previous work we esta blished that in GFP enriched populations, tumors readily formed at doses of three hundred cells. Along these lines it is highly probable that at least three hundr ed Oct-4/GFP+ cells were deliv ered that maintained their tumorigenic potential. Even if a smaller dose was administered ( 30k cells/mouse), we believe we would most likely observe tumor formation b ecause we would still be injecting a sufficient dose of tumorigenic cells. Though a large portion of the cells fr om the sample were indeed induced to differentiate (loss of GFP expression), and thus had lo st their tumorigenic potential, the underlying fact that a small dose of cells reta in their tumorigenic capacity does not change. The remaining GFP+ population, st ill retaining th eir tumorigenic potential, had perhaps not been given enough time to respond to the BMP signaling. It may also be the case that under the

PAGE 50

50 conditions used; they did not re ceive a sufficient amount of stimulation from the BMPs. As a treatment option, differentiation th erapy using agents such as BM P4 or BMP7 exposure might do little to halt tumor formation due to the very small number of cells required to initiate a tumor. Even though these molecules do induce differe ntiation, the relatively small subpopulation of cells that can form a tumor persists. If this were to be used as an effective therapy, stimulatory effects must induce virtually every cell in the tumor to differentiate. What is the nature of the relationship betw een Oct-4/GFP expression, tumorigenesis, and differentiation at the molecular level? This is a difficult question to address since tumorigenic ability is a complex phenotype involving multiple factors and pathways and BMPs have pleiotropic effects. It is likely that Oct-4/GFP activity is a product of the activity of a pathway or group of factors critical to the tumorigenic phenotype. Future work on this project will involve furt her characterization of the isolated cancer stem cells to identify differen ces at the molecular level betwee n the GFP+ and GFPpopulations. These studies should provide insight into th e pathways and mechanisms responsible for conferring tumorigenicity and provi de clues to the development of therapeutic strategies that target them.

PAGE 51

51 Figure 5-1. Osteosarcoma: H&E stain of a femoral osteosarcoma.

PAGE 52

52 Figure 5-2. Cancer Stem Cells. Treatment of the primary tumor with chem otherapeutic agents kills the rapidly dividing cells of the tumor bulk (blue to black) but fails to eradicate the tumor stem cells (green) which can then cause re-growth of the primary tumor or disseminate to form distal metastases. Metastases Recurrent growth or Chemotherapy Primary Tumor

PAGE 53

53 Figure 5-3. ESC-specific Genes in Sarcospheres. Genes specific to ESCs show increased expression in sarcosphere cultures derived from bone sarcomas. (A) Monolayer and sarco-sphere (SP) cultures fr om five osteosarcoma (OS) and three chondrosarcomas (CS) were analyzed for Oct-4, Nanog and STAT3 mRNA by RT-PCR; -actin expression was used as a positive control. Sphere cultures de monstrate increased transcription of both Oct-4 and Nanog over adherent cultures; STAT3 expression was uniform between both culture types. (B) Relative band intensities for Oct-4 and Nanog for each culture from (A) were quantitated by densitometry, normalized relative to -actin and plotted on the graph shown (Oct-4, x-axis; Nanog, y-axis). As indicated by the grouping, the sphere cultu res of each sarcoma showed significantly greater expression of both Oct-4 and Na nog than adherent monolayer cultures (p<0.05, Pearsons correlation). (C) Wester n blot analysis of lysates from representative bone sarcoma cell cultures for protein expression of Oct-4, STAT3 and activated (phosphorylated, p) STAT3. -actin was used as a positive control for loading, membrane transfer and immunoblotting. All cultures showed positive staining of protein bands of the appropriate sizes as indicated. (D) Small and large sarcospheres were embedded in fibrin a nd then paraffin section and stained using immunohistochemistry for Nanog and Oct-4 as indicated. Small spheres show intense staining in cells in the periphery. Large spheres show similar numbers of darkly staining cells with dramatic ally increased numbers of poorly staining cells in the interior of the sphere.

PAGE 54

54 Figure 5-4. Immunohistochemical st aining for Oct-4 and Nanog in sections from tumor biopsies of chondroand osteosarcoma. One repr esentative osteosarcoma (OS-154) and chondrosarcoma (CS-187) and a positive control, human fetal testis, are shown as indicated. CS-187 shows a single nuclei positive (brown) for Oct-4 and multiple nuclei positive (brown) for Nanog in a lung metastasis from a chondrosarcoma. OS 154 sections demonstrate scattered Oct-4 nuc lear staining and near complete nuclear Nanog staining in a primary fibular osteosarco ma. Twenty-six week fetal testis with scattered Oct-4 and Nanog nuclear stai ning are shown as positive controls.

PAGE 55

55 Figure 5-5. Analyses of bone sarcoma cultures for expression of genes of endoand ectodermal lineages. (A) RT-PCR analyses of adherent and sarcosphere cultures for transcription of endoderm-associated genes (Gata-4, Gata-6, and alpha fetoprotein [AFP]) and the neuro-ectoderm marker, -III tubulin. Primers for b-actin were used as positive reaction controls as indicated. Control la ne represents parall el RT-PCR reactions performed without reverse transcriptase. (B ) Western blot analyses for expression of b-III tubulin and AFP from lysates of adherent cultures shown in panel A, demonstrating protein expression of endoderm and neuro-ectoderm-associated genes. (C and D) Expression of b-III tubulin in tissue specimens from bone sarcomas as detected by immunohistochemistry, and in adhe rent cultures (E a nd F) demonstrated by immunocytochemistry. In panels C and D arrows indicate regions of positive staining. In panels E and F areas of b-III t ubulin staining are seen in red. Nuclei were counterstained blue us ing Hoechsts stain.

PAGE 56

56 Figure 5-6. Multipotent cells in bone sarcoma Following incubation in osteogenic or adipogenic media to induce differentiation along mesenc hymal lineages, the respective cultures were analyzed for mineralization by Von Kossa staining or for lipid vacuoles by staining with Oil-red O. As shown, bot h cultures showed focal staining for osteogenic and adipogenic differentiation.

PAGE 57

57 Figure 5-7. Schematic of phOct-4/GFP The human Oct-4 promoter (~4kb) was cloned upstream of the gene for Green Fluorescent Protein (GFP). The plasmid also contains an SV40promoter-driven Neomycin resistance gene allowing for selection using G418 (NeoR). Human Oct-4 GFP Neo R phOct-

PAGE 58

58 Figure 5-8. Expression of phOct-4/GFP and surf ace antigens in OS521 in vitro and in vivo. (A) Visualization Oct-4/GFP tran sfected monolayer cultures using inverted fluorescence microscopy (top panel) and in tumors by immunofluorescence (bottom panel). (B) Analysis of OCT-4/GFP and surface antigen expression by flow cytometry in monolayer culture and xenotranplanted tumors of OS521. A

PAGE 59

59 Figure 5-9. The Oct-4/GFP enriched tumor fracti on of OS 521 is highl y tumorigenic following delivery into NOD/SCID mice. (A) Xenotranplanted tumo rs were harvested and fractionated by FACS into GFP-enriched a nd depleted populations for tumorigenicity experiments. The resulting populations were 92% and 97% pure as shown (B and C, respectively). (D) The frequency and rate of tumor formation of the two fractions.

PAGE 60

60 Figure 5-10. Serial transp lant of Oct-4/GFP clones in vivo Flow cytometric analysis for Oct4/GFP expression in clones isolated from xenotranspl ants showing the emergence of a GFPpopulation follo wing serial passage in vivo (A) Clone S1 in vitro prior to xenotransplantation (B) following 1 passage in vivo (C) following 3 passages in vivo tumor (D and E) Clone T1 in vitro (r ed) and passage-1 tumor (blue) (E) third passage tumor. A B C D E

PAGE 61

61 Cell Dose GFP + GFP 30,000 4/4 23 days2/4 47 days 3,000 4/4 36 days1/4 50 days 300 4/4 44 days0/4 90 days Cell Dose GFP + GFP 30,000 4/4 23 days2/4 47 days 3,000 4/4 36 days1/4 50 days 300 4/4 44 days0/4 90 days 3X104GFP+ AB C 85%GFP+ 55%GFP+ 3X104GFPFigure 5-11. The Oct-4/GFP-enriched fraction of clonally derived tumors is highly tumorigenic in vivo. (A) Transplanted tumrs were fractionated by FACS and the Oct-4/GFP enriched and depleted fractions were transplanted into NOD/SCID mice. (B) The clonal Oct-4/GFP-enriched fraction is more tumorigenic than th e depleted fraction. (C) Analysis by flow cytometry of tumor tr ansplants shows the tumors arising from the GFP+ and GFPfractions are com posed primarily of Oct-4/GFP+ cells.

PAGE 62

62 Control Day 14 Day 21 ABC Figure 5-12. Culture in osteogeni c media induces silencing of the Oct-4/GFP transgene. FACS analysis of cells isolated from transplanted tumors cultured in standard culture media for 21 days (A) or in osteogenic media for 14 or 21 days (B and C respectively).

PAGE 63

63 25% EGFP+ 38% EGFP+ 25% EGFP+ 38% EGFP+ 10% EGFP+ 70% EGFP+ 10% EGFP+ 70% EGFP+ 60%EGFP+ ABC Figure 5-13. Treatment of OS tu mor cells with BMPs induced s ilencing of Oct-4/GFP expression in vitro. (A) FACS analysis of tumor sample s cultured in standard culture media. (B) Tumor samples cultured in BMP-4 or -7 conditioned media for 10 days. (C) Tumors generated by transplantation of 1X10 5 treated tumor cells.

PAGE 64

64 LIST OF REFERENCES Al-Hajj, M., W icha, M.S., Benito-Hernandez, A. Morrison, S.J., and Clarke, M.F. (2003). Prospective identification of tu morigenic breast cancer cells. Proc Natl Acad Sci U S A. 7, 3983-3988. Bonnet, D. and Dick, J.E. (1997). Human acute myeloid leukemia is organized as a hierarchy that originates from a primitive hematopoietic cell. Nat Med. 7, 730-737. Carew, J.S., Giles, F.J., and Nawrocki, S.T. (2008). Histone deacetylase inhibitors: Mechanisms of cell death and promise in combin ation cancer therapy. Cancer Lett. Chi, S.N., Conklin, L.S., Qin, J., Meyers, P.A ., Huvos, A.G., Healey, J.H., and Gorlick, R. (2004). The patterns of relapse in osteos arcoma: the Memorial Sloan-Kettering experience. Pediatr Blood Cancer. 1, 46-51. Eckfeldt, C.E., Mendenhall, E.M., and Verfaillie, C.M. (2005). The molecular repertoire of the 'almighty' stem cell. Nat Rev Mol Cell Biol. 9, 726-737. Feldman, N., Gerson, A., Fang, J., Li, E., Zhang, Y ., Shinkai, Y., Cedar, H., and Bergman, Y. (2006). G9a-mediated irreversible epigenetic inactivation of Oct-3/4 during early embryogenesis. Nat Cell Biol. 2, 188-194. Galli, R., Binda, E., Orfanelli, U., Cipelletti, B., Gritti, A., De Vitis, S., Fiocco, R., Foroni, C., Dimeco, F., and Vescovi, A. (2004). Isolation and characterization of tumorigenic, stemlike neural precursors from huma n glioblastoma. Cancer Res. 19, 7011-7021. Gerrard, L., Zhao, D., Clark, A.J., and Cui, W. (2005). Stably transfecte d human embryonic stem cell clones express OCT4-specific green fluor escent protein and maintain self-renewal and pluripotency. Stem Cells. 1, 124-133. Gibbs, C.P., Jr., Weber, K., and Scarborough, M.T. (2002). Malignant bone tumors. Instr Course Lect. 51, 413-428. Gronthos, S. and Simmons, P.J. (1995). The gr owth factor requirements of STRO-1-positive human bone marrow stromal precursors under se rum-deprived conditions in vitro. Blood. 4, 929-940. Hanahan, D. and Weinberg, R.A. (2000). The hallmarks of cancer. Cell. 1 57-70. Hattori, N., Nishino, K., Ko, Y.G., Hattori, N., Ohgane, J., Tanaka, S., and Shiota, K.( 2004). Epigenetic control of mouse Oct-4 gene expression in embryonic stem cells and trophoblast stem cells. J Biol Chem. 17, 17063-17069. Hemmati, H.D., Nakano, I., Lazareff, J.A., Masterman-Smith, M., Geschwind, D.H., BronnerFraser, M., and Kornblum, H.I. (2003). Can cerous stem cells can arise from pediatric brain tumors. Proc Natl Acad Sci U S A. 25, 15178-15183.

PAGE 65

65 Ignatova, T.N., Kukekov, V.G., Laywell, E.D., Suslov, O.N., Vrionis, F.D., and Steindler, D.A. (2002). Human cortical glial tumors contain neural stem-like cell s expressing astroglial and neuronal markers in vitro. Glia. 3, 193-206. Jhappan, C., Gallahan, D., Stahle, C., Chu, E., Smith, G.H., Merlino, G., and Callahan, R. (1992). Expression of an activat ed Notch-related int-3 transgene interferes with cell differentiation and induces neoplastic transf ormation in mammary and salivary glands. Genes Dev. 3, 345-355. Jones, P.A. and Baylin, S.B. (2002). The fundament al role of epigenetic events in cancer. Nat Rev Genet. 6, 415-428. Katsetos, C.D., Legido, A., Perentes, E., and Mor k, S.J. (2003). Class III beta-tubulin isotype: a key cytoskeletal protein at the crossroads of devel opmental neurobiology and tumor neuropathology. J Child Neurol. 12, 851-866; discussion 867. Lessard, J. and Sauvageau, G. (2003). Bmi-1 determ ines the proliferative capacity of normal and leukaemic stem cells. Nature. 6937, 255-260. Li, C.D., Zhang, W.Y., Li, H. L., Jiang, X.X., Zhang, Y., Tang, P.H., and Mao, N. (2005). Mesenchymal stem cells derived from human placenta suppress allogeneic umbilical cord blood lymphocyte proliferation. Cell Res. 7, 539-547. Meissner, A., Wernig, M., and Jaenisch, R. ( 2007). Direct reprogramming of genetically unmodified fibroblasts into plur ipotent stem cells. Nat Biotechnol. 10, 1177-1181. Meyers, P.A., Schwartz, C.L., Krailo, M., Klei nerman, E.S., Betcher, D., Bernstein, M.L., Conrad, E., Ferguson, W., Gebhardt, M., Goor in, A.M., Harris, M.B., Healey, J., Huvos, A., Link, M., Montebello, J., Nadel, H., Nied er, M., Sato, J., Siegal, G., Weiner, M., Wells, R., Wold, L., Womer, R., and Grie r, H. (2005). Osteosarcoma: a randomized, prospective trial of the addition of ifosfami de and/or muramyl tripeptide to cisplatin, doxorubicin, and high-dose met hotrexate. J Clin Oncol. 9, 2004-2011. Miller, C.W., Aslo, A., Won, A., Tan, M., Lampkin, B., and Koeffler, H.P. (1996). Alterations of the p53, Rb and MDM2 genes in osteosarcoma. J Cancer Res Clin Oncol. 9, 559-565. Ohlstein, B., Kai, T., Decotto, E., and Spradli ng, A. (2004). The stem cell niche: theme and variations. Curr Opin Cell Biol. 6, 693-699. Ohnishi, K. (2007). PML-RARalpha inhibitors (ATRA, tamibaroten, arse nic troxide) for acute promyelocytic leukemia. Int J Clin Oncol. 5, 313-317. Okita, K., Ichisaka, T., and Yamanaka, S. ( 2007). Generation of germline-competent induced pluripotent stem cells. Nature. 7151, 313-317. Pardal, R., Clarke, M.F., and Morrison, S.J. (2003). Applying the principles of stem-cell biology to cancer. Nat Rev Cancer. 12, 895-902.

PAGE 66

66 Pasca di Magliano, M. and Hebrok, M. (2003). Hedgehog signalling in cancer formation and maintenance. Nat Rev Cancer. 12, 903-911. Piccirillo, S.G. and Vescovi, A.L. (2006). Bone morphogenetic proteins regulate tumorigenicity in human glioblastoma stem cells. Ernst Schering Found Symp Proc. 5, 59-81. Qiang, Y.W., Endo, Y., Rubin, J.S., and Rudikoff, S. (2003). Wnt signaling in B-cell neoplasia. Oncogene. 10, 1536-1545. Ragland, B.D., Bell, W.C., Lopez, R.R., and Si egal, G.P. (2002). Cytogenetics and molecular biology of osteosarcoma. Lab Invest. 4, 365-373. Reya, T., Morrison, S.J., Clarke, M.F., and Weissm an, I.L. (2001). Stem cells, cancer, and cancer stem cells. Nature. 6859 105-111. Reynolds, B.A. and Weiss, S. (1992) Generation of neurons and astrocytes from isolated cells of the adult mammalian central nervous system. Science. 5052, 1707-1710. Sandberg, A.A. and Bridge, J.A. (2003). Updates on the cytogenetics and molecular genetics of bone and soft tissue tumors: osteosarcoma a nd related tumors. Cancer Genet Cytogenet. 1, 1-30. Simmons, P.J., Gronthos, S., Zannettino, A., Ohta, S., and Graves, S. (1994). Isolation, characterization and functional activity of human marrow stromal progenitors in hemopoiesis. Prog Clin Biol Res. 271-280. Singh, S.K., Clarke, I.D., Terasaki, M., Bonn, V.E ., Hawkins, C., Squire, J., and Dirks, P.B. (2003). Identification of a cancer stem cell in human brain tumors. Cancer Res. 18, 58215828. Singh, S.K., Hawkins, C., Clarke, I.D., Squire, J.A., Bayani, J., Hide, T., Henkelman, R.M., Cusimano, M.D., and Dirks, P.B. (2004). Identi fication of human brain tumour initiating cells. Nature. 7015, 396-401. Steindler, D.A. (2007). Stem cells, regenerative medicine, and animal models of disease. Ilar J. 4, 323-338. Stewart, K., Walsh, S., Screen, J., Jefferiss, C.M. Chainey, J., Jordan, G.R., and Beresford, J.N. (1999). Further characterization of cells expr essing STRO-1 in cultures of adult human bone marrow stromal cells. J Bone Miner Res. 8, 1345-1356. Thomas, D. and Kansara, M. (2006). Epigenetic modifications in osteogenic differentiation and transformation. J Cell Biochem. 4, 757-769.

PAGE 67

67 van de Wetering, M., Sancho, E., Verweij, C., de Lau, W., Oving, I., Hurlstone, A., van der Horn, K., Batlle, E., Coudreuse, D., Haramis, A.P., Tjon-Pon-Fong, M., Moerer, P., van den Born, M., Soete, G., Pals, S., Eilers, M., Medema, R., and Clevers, H. (2002). The beta-catenin/TCF-4 complex imposes a crypt progenitor phenotype on colorectal cancer cells. Cell. 2 241-250. Varga, A.C. and Wrana, J.L. (2005). The disparat e role of BMP in stem cell biology. Oncogene. 37, 5713-5721. Wernig, M., Meissner, A., Foreman, R., Brambrink, T., Ku, M., Hochedlinger, K., Bernstein, B.E., and Jaenisch, R. (2007). In vitro repr ogramming of fibroblasts into a pluripotent ES-cell-like state. Nature. 7151, 318-324. Wuisman, P. and Enneking, W.F. (1990). Prognosis for patients who have osteosarcoma with skip metastasis. J Bone Joint Surg Am. 1, 60-68. Yin, T. and Li, L. (2006). The stem cell niches in bone. J Clin Invest. 5, 1195-1201. Yu, J., Vodyanik, M.A., Smuga-Otto, K., AntosiewiczBourget, J., Frane, J.L., Tian, S., Nie, J., Jonsdottir, G.A., Ruotti, V., Stewart, R., Slukvin, II, and Thomson, J.A. (2007). Induced pluripotent stem cell lines derived from human somatic cells. Science. 5858, 1917-1920.

PAGE 68

68 BIOGRAPHICAL SKETCH Thomas Currie was born in 1980, in Clevela nd, Ohio. Shortly thereafter, his family moved to Florida, where his father worked as an Anesthesiologist and hi s mother raised him and his three brothers and sister After receiving his high school degree in the international baccalaureate program from Sebastian River High School, Thomas began his undergraduate career at the University of Fl orida in 1999 on a 100% Florida bri ght futures scholarship. In 2004 he received his bachelor degree in psychology and completed the pre -requisites needed for dental school. In 2006, before applying to dental school, Thomas decided to pursue a master of science degree from the University of Floridas Colle ge of Medicine. He focused his studies on molecular biology and aimed his research towa rd understanding human osteosarcoma. During this time, he was accepted into the University of Floridas College of Dentistry, for the 2008 entering class. In the future, Thomas wishes to wo rk as a dentist in the state of Florida, while enjoying outdoor activities in his spare time.


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E20101107_AAAACW INGEST_TIME 2010-11-08T01:22:29Z PACKAGE UFE0022577_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 17826 DFID F20101107_AABZKE ORIGIN DEPOSITOR PATH currie_t_Page_59.pro GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
b97860b42d068e514d7db2662efe9cdc
SHA-1
3013a72fccbc1762524f17b11201c21c1ebc4720
1051983 F20101107_AABZFH currie_t_Page_38.jp2
3e1f65547c81667f4b24ef1a12b60b1e
b954640ded31fc5c9fceb9f11cd198047fe0e3b1
1051986 F20101107_AABYYK currie_t_Page_18.jp2
8c6a4d322c6b2328835dcbcb3f2e8054
48a15f014ea43a1fdfdfd69ae8b33672931a39d9
56658 F20101107_AABZAK currie_t_Page_60.jpg
d3e473ff0cd88adb8ca0ad6ddb81f828
e0d31033d709e45f8a9d61f7f5eeb147140a8bd1
35779 F20101107_AABZPC currie_t_Page_38.QC.jpg
e5ecd8eecbb56ac79024b8508b92e495
98299d12cc9816fabd197bb05377af1167134b8c
17161 F20101107_AABZKF currie_t_Page_60.pro
b0af0b92b06dfff2e2f6b6d17c471fd4
b5bdff8080ae6025e89b523b41d32ce778ab2eae
1051973 F20101107_AABZFI currie_t_Page_39.jp2
faacb4f88b35a36d242468a821d4ca6b
8aaf320b2dbe328c368c85fdcebc0ead4c1228d0
1051981 F20101107_AABYYL currie_t_Page_09.jp2
a28356d2bc28948bb7f36aff8c157c03
932fdfe1a1dea00ccafab6d591d31f25887f19d5
9498 F20101107_AABZAL currie_t_Page_57.QC.jpg
9cbcc9fe7d254007f0c6e56c9b65a166
85d1e0eda419d36df5a33a27357a0af232ebb6e2
8802 F20101107_AABZPD currie_t_Page_38thm.jpg
fbf014ef0d66d6bdae2e3669197f244f
7d53cc89004876371df12214a3ca094c2e2043cb
23153 F20101107_AABZKG currie_t_Page_61.pro
1efeedd66938b641dcabe0ea777777e9
bd82696f1ea1fc922f9933837fa668bed80017cd
1051985 F20101107_AABZFJ currie_t_Page_40.jp2
3b66b50ed177ed6a8d015da247409cd4
f3d4d75406a549422d90669ee14500a14f996faf
17313 F20101107_AABZAM currie_t_Page_54.QC.jpg
338a6d80007ad6cb67f151488125081b
1f24ebcfe25337853a867d6c40312b8330a88623
33469 F20101107_AABZPE currie_t_Page_39.QC.jpg
e980b69ace9d86c0c34a62b114b33804
cb9c56265422200b3b8da341113c0c05e4c36091
20741 F20101107_AABZKH currie_t_Page_63.pro
4013a98444622859f619663d09f73fa6
7a7e40c75e33765d7b029f6411f2c8bfa02835cf
F20101107_AABZFK currie_t_Page_42.jp2
c899fb1cb95af2d86006ad209cb17c33
b434e94de32975c25b6852240ea680eb4a764624
18317 F20101107_AABYYM currie_t_Page_67.QC.jpg
0458f81450e0e9fbae7db0a1e1e60609
cc0aeeaf1a041b2b6d7b9115da71974c427b0621
38741 F20101107_AABZAN currie_t_Page_65.QC.jpg
f72e1e231d7270c8ce5764418c375f6d
9a696eaa76c65c0561b6f4d4ed9ecbefda8be5dc
8342 F20101107_AABZPF currie_t_Page_39thm.jpg
6f5c70706e7238f663e525216f8d5fe2
27ff21140c8a0430d67ba6578d6f455d981dc8db
70615 F20101107_AABZKI currie_t_Page_65.pro
0a947ab3d2371d49def0e85e7c5b8da1
1d725253da76523683abdad177d873dc0e65f610
149911 F20101107_AABZFL currie_t_Page_43.jp2
6e19295e05e5d636f7f0482895b38e17
ddf6f8f3df0cd9cf856eed9ff1a2916188abdff9
25271604 F20101107_AABYYN currie_t_Page_61.tif
3787fdbd09d4e5b18ece68d71f3f4a6f
43b3130ba1012d7c70d293c05870b5411b55629d
2238 F20101107_AABZAO currie_t_Page_08thm.jpg
4d9c07a3e6731214bd5d33398a847fef
57b97f1e35d218c45bfa1cc505476813dc32310a
37542 F20101107_AABZPG currie_t_Page_40.QC.jpg
1b56c4ab901d878334296cbc376610b1
992ec73b1b3de9679cce0026054384ec25252dee
57033 F20101107_AABZKJ currie_t_Page_66.pro
9e66a8c9d44ba008de7499e8d3bc320b
e0bdd9d8c212cbc5657952b1118acd696f782ed2
F20101107_AABZFM currie_t_Page_44.jp2
a56510a310dbaa416d5eeaecb0c8de2d
6cd234eff4c4613cdf0aa067e4406594192163a5
91456 F20101107_AABYYO currie_t_Page_07.pro
59764d6c70e5314dbd985f6ba9cc3dbe
87ab35d5322135a8fc1431a671b8334d253b26b4
57237 F20101107_AABZAP currie_t_Page_26.pro
de6fe570d214e5eaeffe11fa232473f2
e75e3765ec82e730ac3f183035a79f69b35988aa
36779 F20101107_AABZPH currie_t_Page_41.QC.jpg
be026e915987cf67d3038b71efb51497
3c7996541b87e17952949fe29a660a11f6e50dbe
31842 F20101107_AABZKK currie_t_Page_67.pro
686b89fa3e7f1c44967bc335a0f0334b
8961bf1e2d14888defe5fd5627e1650606f577e9
1051937 F20101107_AABZFN currie_t_Page_45.jp2
8f736186c46fc467bb4d1e0cb3d5d835
0fe74442e6b4ad32bf346d1b564a70c6137ed52c
508273 F20101107_AABYYP currie_t_Page_06.jp2
899ab52c6ebbac9360bf47abdc2d5d17
193ce6ea66a395eb47f6cb82d5430eff218ca4ec
2126 F20101107_AABZAQ currie_t_Page_17.txt
e576fa2171eb6d44250e7bd767488828
73d36fadfc206ac55de83048fbb2534c64820abc
8969 F20101107_AABZPI currie_t_Page_41thm.jpg
067d5a89230937c02cecb5b4a46f4bed
fae8daf239b1ccb02bc16a2a09423548fd91c45b
28573 F20101107_AABZKL currie_t_Page_68.pro
4ec34b95b1c80ae065e31caef8d0a0eb
63038c6e2c22ca881b20928dc7d5a0996839e315
1051901 F20101107_AABZFO currie_t_Page_46.jp2
b6f1ed20d9299a818941107caf3f4623
375096d3537b07118ad74269485eb73958c3d5ce
4897 F20101107_AABYYQ currie_t_Page_68thm.jpg
de0c5a9cb6705729931437dd5acb1bae
da31a5a499be4e149fa60f25f3147bab7cfdb946
1051916 F20101107_AABZAR currie_t_Page_55.jp2
6f901027f8a64ed76c089193b0b5ebc9
5098aac4a065e1f7654d8357fd4460fb2e856803
9098 F20101107_AABZPJ currie_t_Page_42thm.jpg
853cf343312581e764198ef911050355
e2c500547bbf5669585b1682f7ac9b611b013beb
490 F20101107_AABZKM currie_t_Page_01.txt
92e7c41b777b8db0341fd55bfcbceab4
fec7be4368cd626c9c8eac3804c328cd9a781ac3
1051968 F20101107_AABZFP currie_t_Page_47.jp2
57aa6dc1f3c60bc3d3cef911b376362b
0912dae9f4a7b22a5d70ffd3defa2a1bf9f5da5a
6174 F20101107_AABYYR currie_t_Page_55thm.jpg
01567e4447beb0f51d33099897c36b34
5b3a75a67eadb13b831615e809578017afa955bd
1016 F20101107_AABZAS currie_t_Page_02.pro
b9b5cdaa8ca6f02c38799170aebc51be
263fd243af1710a24a5391b546266a2cb6c7d114
5822 F20101107_AABZPK currie_t_Page_43.QC.jpg
cd3e9a483d339458ae6e2b1efe167358
bcd41c34c64b75df318b428d2ab7688876c74aa5
1051975 F20101107_AABZFQ currie_t_Page_48.jp2
a0300028386fb4d2984d49bc318ee769
68467466ca8c4ab0220fea1b6005aa41f9791bd1
1051944 F20101107_AABYYS currie_t_Page_25.jp2
b7a22dd6ff2cbc515b74bc2781145cb6
29e0d5737541b6f40c7edb55091b1f4989267eb9
97 F20101107_AABZKN currie_t_Page_02.txt
bbcb2d1505113701db35b3052503655f
a731b2801725fb26fec5192d56bd50ed6c593e98
1552 F20101107_AABZPL currie_t_Page_43thm.jpg
09ff7467dadd7435e1422d9584735f77
8e11d257ed761993ba8572c6a0a9db7dd982f6f2
F20101107_AABZFR currie_t_Page_49.jp2
57d26db651de63dddd2dfb97dcf07813
f52bde0077d68ff80354acf3d347244e77f29a6f
28588 F20101107_AABYYT currie_t_Page_14.QC.jpg
65f607af450846c9c4d584eb6a8f1365
dcfc2c213256e2d37555332a46f68102f13508e2
14613 F20101107_AABZAT currie_t_Page_59.QC.jpg
774a09ef4866745c13359ce78ba09ecf
fb29d384e3714cff13a1500218f1f9d7731df52e
2179 F20101107_AABZKO currie_t_Page_04.txt
fbafea246a347d30c06eb2ad9125a9f8
cfc9d385c5ff9797e178a1e91774a4da56159994
8725 F20101107_AABZPM currie_t_Page_44thm.jpg
9137d10f4449865295089ea1de7ea7f3
214757b172be2a016afdb51d25fbb284c4a2c834
800766 F20101107_AABZFS currie_t_Page_50.jp2
37b2779f75fed1e5b8722a9ea485e1f9
02d0dbc0739f701cd8ce4cb17116b806279bbc77
3300 F20101107_AABYYU currie_t_Page_62thm.jpg
8197df16f0cf7182c44385974f04a74c
d28a5f9dae77b765e054c4a582699b4106c5c3b0
F20101107_AABZAU currie_t_Page_28.jp2
78a944c47e4fbcd160777db2bd1d119b
3586f59b1afff43af5e6bb5e532dc46444c6f2c2
2168 F20101107_AABZKP currie_t_Page_05.txt
b668996e95ef030ebc396bda07b1eeb8
4a04d0856ad93d0fa55636d912f262a3fc38a717
36038 F20101107_AABZPN currie_t_Page_45.QC.jpg
c0083235baa9a6307f42b884996b9069
3b40869a9fdbf9cedee91b73b4028a4b6fa7c1a4
1051929 F20101107_AABZFT currie_t_Page_51.jp2
71f5a2c721925fdd175af9ecf2dc9bcb
16ed051ba41604b701da65377eb23bf11db3752b
1051958 F20101107_AABYYV currie_t_Page_41.jp2
11ccf9cb6d1ffcb056b2a38db6e18e62
72d7e0c00d441f9a940ab1ddfd8949c1c271ee09
2110 F20101107_AABZAV currie_t_Page_13.txt
330beaf8c73b2d6db6a3599bf0c32f24
e82c55d73265c7401c0116d542ec976edd835a6b
899 F20101107_AABZKQ currie_t_Page_06.txt
61c7e071cf5b4d3423cab9a858157ecf
09c2753c6874a5f35f556bb6892e49942015d51d
8950 F20101107_AABZPO currie_t_Page_45thm.jpg
704528bffe0f5f4b60a08437cf3d3f26
7301fb8fe3c5e451931876ddeb30119a15b452e6
487357 F20101107_AABZFU currie_t_Page_52.jp2
40c1824a85512e922d5b9f522f74cc80
544b62bc491ecf8ce4340ac81393597348e8d8af
F20101107_AABYYW currie_t_Page_33.jp2
e2e8faa69adcae0ea9cb752be9ed9307
531682457830bae340316535edf43bcf81bb29ca
F20101107_AABZAW currie_t_Page_50.tif
6f240184b9439faf82b40ebf07b447f7
ecfde8d785ad1af5570f528b854dbd46ce40c75f
3761 F20101107_AABZKR currie_t_Page_07.txt
178f3b4a1b5d94006554ed442adc973f
4bee0be5c5c1fc624c46fadf4c1a9fcf10394668
36123 F20101107_AABZPP currie_t_Page_46.QC.jpg
6796290eed5533c29cf93bf666600e89
380ce593e278421a0c2ba0b166462a37026edcbb
1051982 F20101107_AABZFV currie_t_Page_53.jp2
0f18e8f999801d478f7aefac528dd7b9
69ef970539157c30dfe4bc834da44bbfc9a2e694
8986 F20101107_AABYYX currie_t_Page_40thm.jpg
343dedb94cd61c3bfe2a0f4b3ed902ab
f5058ba8de6d75bed7587a3400bf8058a56a9988
F20101107_AABZAX currie_t_Page_51.tif
bbe4ce3e765e545630a88e430ae42f95
a6cf7151b1085e44e421bd3271b3c81ee36d1db5
718 F20101107_AABZKS currie_t_Page_08.txt
5f515ed8bade5fac20dbccdbdb77e6bc
07d6e741dc539619b3ea24fd72bbf7a04dbd6147
8845 F20101107_AABZPQ currie_t_Page_46thm.jpg
acdb44f39132801af26418f5c459c42a
bf3734a39f76436c4afae6cf65be25537b3b030f
885073 F20101107_AABZFW currie_t_Page_54.jp2
58337874860a3360fb039ad3de6507cc
24a54cc68d72b3f457d2c48bda5c64b1f1ab23bb
9027 F20101107_AABYYY currie_t_Page_15thm.jpg
da04c1fba57b51ff741af5ec525378d5
d1346908ccacb93ba24fa1e658aa8fb2a64fc22f
110152 F20101107_AABZAY currie_t_Page_05.jpg
7f64b7abccd5fe6eb7e1594118882bb4
c6a5aaa07ab40dacc43bbe7af097465e40856488
2075 F20101107_AABZKT currie_t_Page_09.txt
5ef86266464b49415db72a94b4b1c5a1
ff3541341db43132bd4e103535c9978afea203fc
36223 F20101107_AABZPR currie_t_Page_47.QC.jpg
23f4bb524d8c1a29af7557f5e6956521
72a24d0a8f95a9f4ab81c4dce30a2cd9abc3c72c
622133 F20101107_AABZFX currie_t_Page_56.jp2
8db4edd1dab9480e93842a530755ce15
22c2e0955a868e894f544c80b306c7c618fb9e31
105205 F20101107_AABYYZ currie_t_Page_24.jpg
af6a101de8042b4f3c5cbaf73fe5c653
7ab176683f61e2e381d2ad5a3e9da1cc92268770
114594 F20101107_AABZAZ currie_t_Page_40.jpg
69c445adf5d031e1f5ddcb40a55931da
675fdc367cee91140c2d9a399b6d2e3b827e4a82
2042 F20101107_AABZKU currie_t_Page_11.txt
4ed2f76b3ba07b3f04bc9ce437d06f7f
bb2b1051c008e4ac458afbd9ba462f2c1421ce6d
8915 F20101107_AABZPS currie_t_Page_47thm.jpg
24a613db431a99c7a5a9092d22d70dab
58b4d0c846c569f7fdfc846c0d8b3abd5e5c8bde
2160 F20101107_AABZKV currie_t_Page_12.txt
dc32dbf28b044ff6876f682b2bd12649
6608b1de65d2975b6cdfbed4fd7123800cc29643
35040 F20101107_AABZPT currie_t_Page_48.QC.jpg
11b8b6ca5fa22b6b8dfc11868c3de7b4
f9ca8f5bb3417ffc37a5f951f044de85570e9918
302939 F20101107_AABZFY currie_t_Page_57.jp2
9803d68d39a2e0d3d4ee5592915ea969
ec4a6b147771f7e1c7b65a425fc73cbe1a242600
1697 F20101107_AABZKW currie_t_Page_14.txt
5b271fe1275f2f8008b052f535a5de79
dec65eddcba545eda55961ac9b0a891732e5c86a
110854 F20101107_AABZDA currie_t_Page_33.jpg
e68e26057d11635713ad222dc1f60ae4
568f4f7b5717faf527906a361d0f20ced026b046
8495 F20101107_AABZPU currie_t_Page_48thm.jpg
b83e85c8b4d2a367f367cf37fa2149ac
cafbe2b2e616f010709774548a0bddb6c6008873
518180 F20101107_AABZFZ currie_t_Page_59.jp2
efc412ddf09fc9e505460a56676217cc
069acd7e22223c39b69ab3b8d59f2e61deb586bc
2218 F20101107_AABZKX currie_t_Page_15.txt
1ecdd0bb4f2d548452bcd71865d6bd6d
64cf6b78e36401cc82ffb745a838c43f014e6fe0
115936 F20101107_AABZDB currie_t_Page_34.jpg
4731d2cbcaaeecb91ecbb2bc413f132e
f1518d6fe265d1eba8446d40dd9630facc4a16d8
36712 F20101107_AABZPV currie_t_Page_49.QC.jpg
489a6367938d39a92b7df7d270374c7f
06f3839ccdc578167de8439a3c62caf5aff5e56e
2016 F20101107_AABZKY currie_t_Page_16.txt
7cd2e74ebd03107bcde45f2f0e7c6d8e
197165a686dd25ceffe3a8180abe8888a1462982
110253 F20101107_AABZDC currie_t_Page_35.jpg
e0f967cd90638e96d73e1882da31a7ed
9db92d637d9b03d5274933e02d8d800760f37581
24363 F20101107_AABZPW currie_t_Page_50.QC.jpg
1657cd5f3b82ac3cb33084fb404ecff5
4796b5ad128d5b24d3c6e5c0e1047108c8e6c6b8
F20101107_AABZIA currie_t_Page_53.tif
f4b17944d384812d543f3dd73b50df60
b6703beab076790983052f67096a61966912270b
1858 F20101107_AABZKZ currie_t_Page_19.txt
efdca4e71af963a60693b1540f15705e
97cd38001abeff01796b4bdc708e6ac1e791692d
48732 F20101107_AABZDD currie_t_Page_36.jpg
a9aecc667784fe703c53593d04837860
9f0b9eb62a2b7f00ba853efd27ecd757c4c86518
6027 F20101107_AABZPX currie_t_Page_50thm.jpg
ceee0c3d3e40b4a17ea80c514d4166d9
4eefc205472cff340f7784bd265d05c276d2e45a
F20101107_AABZIB currie_t_Page_54.tif
b5521aac34d1a888ed725a3309cfcdb4
36181c95abcc345231a9df114e108f3e7f4d634d
104461 F20101107_AABZDE currie_t_Page_37.jpg
e196c4c6b7f1eaa4b9195e000ddb686f
7cc52924bc99577adc72643ec2f11def5fcf4b65
35686 F20101107_AABZPY currie_t_Page_51.QC.jpg
d5f5a46c38f02dffc9a71816fcc34211
c4dbf367a1237fef6b4ba54a4fcbbe6a4a635ff1
F20101107_AABZIC currie_t_Page_55.tif
2cbd7c467f6463eec01ba781f937c363
ac55e9825bab6c7302c1da126a6f36b9bea1904a
110053 F20101107_AABZDF currie_t_Page_38.jpg
79ae7aebe339ae9ec3ddff0926d87e99
e074a4cce2ca5393ba00c7edd32f8278c7f2d80b
9007 F20101107_AABZPZ currie_t_Page_51thm.jpg
30dc9dc77f5c98c6795df3707a448ac4
21262d56533cc78c8ad288c46009eef028ae2873
36883 F20101107_AABZNA currie_t_Page_04.QC.jpg
a361e974e882445fddb8137af565e918
d0b7bf3192ffa60a28450d54b4ebcef7c89ff5d8
F20101107_AABZID currie_t_Page_57.tif
c95d7399cb7a6b6a145a9657dc9e0dfe
8d3ae32d41883a9fe6de485ae2ddeeff8070d74a
101552 F20101107_AABZDG currie_t_Page_39.jpg
6be1e6cb77c1c4f725f8f6e00d653d1c
c8fe763e5c53ee7966cbd466004109e595b05cc3
8976 F20101107_AABZNB currie_t_Page_04thm.jpg
edf189576da20dfb0bc9583e681f75fd
52723968d3af8a7dc16371b9b9de13d3345fbb11
F20101107_AABZIE currie_t_Page_58.tif
1bae452477461c26a9b9e122f9161664
5a97292e8d9e6b80ed4e14fb4fa01fbbcb72a5d9
109750 F20101107_AABZDH currie_t_Page_41.jpg
8725a7db7a45ea3fc3ecc29357afcb80
50c8343935d58be938be73d613386b65d145cd54
35564 F20101107_AABZNC currie_t_Page_05.QC.jpg
071537a50adb8f6ee2ea4aedabbb56d5
c27ee4a54ae22c343455f2f67cb0369374083350
F20101107_AABZIF currie_t_Page_59.tif
6dc3eb363f210a51aab07a014b70818f
09b6b62028f755e2aedd252192b8bdde210b1c23
113781 F20101107_AABZDI currie_t_Page_42.jpg
525332f5e3b78392b243bd73b4301dc9
2a35d5ac8fc23becbc74e2d6ac4b50ec749c0a1e
9004 F20101107_AABZND currie_t_Page_05thm.jpg
cb0ffdf4457afd40a7ddca5d95e3d2a8
9b938e5f700af354b55dfbbe144264746b00e8e7
F20101107_AABZIG currie_t_Page_60.tif
545cd7648f3d53d86f196ee372d7cc98
de4fc4fa3d4bfd1f11253882de77f8b8a8f6e90e
15957 F20101107_AABZDJ currie_t_Page_43.jpg
0bf3405b52e3631c59b9216c0f68c55a
e6b51b45e03fc5a2b9ee6819b82b3773bb6e9c7e
16278 F20101107_AABZNE currie_t_Page_06.QC.jpg
71513c0fc358ffe749aea9ce6fb218ae
57a092270da101d4559f7078c6a6b9d5c98b0c81
F20101107_AABZIH currie_t_Page_62.tif
5083f13b1238baf73d8fbb90c836829a
0748c89f65403ea4e86cecc23207cbb668e32d68
106937 F20101107_AABZDK currie_t_Page_44.jpg
4830bbfe079027b17a48e971966d7fca
e7fa68a0d60202a1fa0ed824544f4e30fe030781
4356 F20101107_AABZNF currie_t_Page_06thm.jpg
88fdd0e0faeea7b0dcecc6d47c5afe17
046d8b6dcc0685c0862de21c478ecf3939b89e7d
F20101107_AABZII currie_t_Page_63.tif
07f9c634e385f345ecdc15cf875425b7
861c797b0fb87a7ec31e5241d86590af560ec06a
111563 F20101107_AABZDL currie_t_Page_45.jpg
0d03638002e1ced54062f11c810455a0
ed799b7f79616009b5204ddb78927f4216b5bc45
30463 F20101107_AABZNG currie_t_Page_07.QC.jpg
274175c247cb18e95232f4b67749b5e4
12b2372b2235e76073a6ce7a0801d26080c06cb7
F20101107_AABZIJ currie_t_Page_65.tif
8161d86628a4f35e968f170e9b21c32f
3136c40c9c4e577fcf6f8f308f545d68314294cc
111185 F20101107_AABZDM currie_t_Page_46.jpg
43e1c55fec47e4f9d8f85b6836136af5
f29910d9803976388eac743c6f43cabc6a04b398
8770 F20101107_AABZNH currie_t_Page_08.QC.jpg
5691cae77233ff58bf237f483f4da7d2
d2b334373a006c97496f88c36066bc1b66807ffd
F20101107_AABZIK currie_t_Page_66.tif
bb17b841b133507d37f39fb124b368de
2da342ebb17bceb8e0409443941dce85a92be6f8
111265 F20101107_AABZDN currie_t_Page_47.jpg
fa5c0f7ea212f4c11801c3c86e595f2e
37236556b028558201a6d6847e614c42cfb8e275
6048 F20101107_AABZNI currie_t_Page_09thm.jpg
f177b5e63f8c8c8b933efbca99708d3e
84f0524a01d3fdd657c23da308d5d844e4b333cc
F20101107_AABZIL currie_t_Page_67.tif
71190e077bd52b8c367aedcbab8d1ba8
20083117eb225b85dc63b2af2e03069964ea0fe3
107682 F20101107_AABZDO currie_t_Page_48.jpg
62a8ff6d2df8bd5feb1a5e528afa0e57
79d2bace7413f92ab8bde7d86220443c36482ab7
8536 F20101107_AABZIM currie_t_Page_01.pro
c95d2ca0f6f9e56626e40d8a615707db
d3d7f4cd638ba8a6dabe80a7fdbf8bd9053c6080
112887 F20101107_AABZDP currie_t_Page_49.jpg
5404258e9e4a77c0fdaec21b792a1ea9
6793d6a293c04fa08261ecbd3a3f610c761a566e
30731 F20101107_AABZNJ currie_t_Page_10.QC.jpg
6bb47990b141715bd1749b5f2f61b93b
ba25ff5522d5bd97097d850d24ec4d7eaf4c986c
598 F20101107_AABZIN currie_t_Page_03.pro
84d6c6d5a049bde9fe543a11da5d9418
3906258b36397a4a79f42a3ae64657b03306de21
126514 F20101107_AABZDQ currie_t_Page_51.jpg
4542a897b6f3aec5efc9cc4f27ef129c
872907af136c7318ed4fc0dee7a6b51fc21e5a76
34382 F20101107_AABZNK currie_t_Page_11.QC.jpg
5a7457de4bb8fcbcf446cfddb68c018a
f13c338f5b00ba5d684e91b2f16b8ca66e71ac94
54830 F20101107_AABZIO currie_t_Page_05.pro
3e78ee8a8fcd60f48f8fc96e951b5892
c22b8d45d032d3788e7b29200dd509c5757a624e
122032 F20101107_AABZDR currie_t_Page_53.jpg
5cfe9053b58f8d72dc0b58608ea2ca92
a11cde486716389d958b03d74c8b5611a69ff89c
8773 F20101107_AABZNL currie_t_Page_11thm.jpg
bd85edd708c6f11947d7c592de676296
176893b6ff163fb82785c46ef2a697c44615747c
22470 F20101107_AABZIP currie_t_Page_06.pro
7f5f437c6d02a6e76f3119454693d228
8a4b46dd8caa5292133d0bcdc875e27fac4e83f4
65004 F20101107_AABZDS currie_t_Page_54.jpg
eb06275c45fdc30b48fe3b05459810d9
a05eb6489ffc4915258955e3c06fd2307f14b7c5
34318 F20101107_AABZNM currie_t_Page_12.QC.jpg
b52dde5adfd4c87fba2c2fc5c7a17968
929ea330c91dc0abd816d4a98cd223602d0e941b
17947 F20101107_AABZIQ currie_t_Page_08.pro
4bf0e23e79355e3b5d648d5e8ef3fd10
32a24502ae60ba9b1242f83ec331f7c9b24d2b9a
88046 F20101107_AABZDT currie_t_Page_55.jpg
beda98e49c9c67579bb4dbe8764cc6a6
44efa6aab3e0abc9380acb4643f7f3f4ecdbad0d
8583 F20101107_AABZNN currie_t_Page_12thm.jpg
6411e21f66d7312205e8a755991fb596
6903d5119cff4f53b8c57f9b12ab94b09b006257
50759 F20101107_AABZIR currie_t_Page_09.pro
5491471c9a8b1cfaa9f9b1cdc5bf4d3f
56e877376f0d77f30ef8919c66ff3cf8ea902f93
46947 F20101107_AABZDU currie_t_Page_56.jpg
8a1949ef517ea47d8e6f17f40b104c99
7246c31a61aa18da00abe6d52b2a3407291e11f1
34994 F20101107_AABZNO currie_t_Page_13.QC.jpg
01e6b5222021c0185b84cbd756463baa
bc851659bffdabb8a6756b4e3366dcff4b1cf475
45717 F20101107_AABZIS currie_t_Page_10.pro
2cc885d56190a94677bec5ddb14fc46b
32fd45be620db38fcee534266fe507bb82bd8f7e
29929 F20101107_AABZDV currie_t_Page_57.jpg
31621c82382f82b19ff789d87d4ab8ec
b337ba95aa2e3199b2def2c96517e1a891038d0e
8684 F20101107_AABZNP currie_t_Page_13thm.jpg
32bcbde70234ad39eee67b694c762ca6
edea190c01956b26b9c2491fd175e5b54cbcbd20
51657 F20101107_AABZIT currie_t_Page_11.pro
1c4f8088f63a238ab8c6e11c6a7358f0
d54fe654e0b70060b4b71cb03433d43ae25a5c7b
7179 F20101107_AABZNQ currie_t_Page_14thm.jpg
7329309d7239c47ace746b110d32fd7f
7645d6c349551176e99b1db8102f3352cf0b3f40
53023 F20101107_AABZIU currie_t_Page_12.pro
e9df9b54d8bb31160795d4cc7f414c6b
280b16a2f033505fa906c9d01c2e8d999c35efc3
68095 F20101107_AABZDW currie_t_Page_58.jpg
e3ec1c05a17f038f7e9ba869969a7dc5
075025e4702b2acc645f3d3544304452f1c01a2b
36080 F20101107_AABZNR currie_t_Page_15.QC.jpg
233dd8d183a317fc78509a77dad570af
945750a030b994db38b4e4c19eb05e483d590117
53684 F20101107_AABZIV currie_t_Page_13.pro
674f4482ae6e87379b7241f6ac21c866
8e20320be143a6d34131e43d94b7b732059b4d08
47963 F20101107_AABZDX currie_t_Page_59.jpg
2df6bc21a6ff792204965eb7c29a71f6
6422b865bd591b5615609f5c5c05d67c5a3fc409
8786 F20101107_AABZNS currie_t_Page_16thm.jpg
e0b8dd15ba50797b7a37900572556cd1
9cb1accccdb86cf03b7a244c942d9d0e36d7971b
35870 F20101107_AABZNT currie_t_Page_17.QC.jpg
637230bdcf1664369f68f046cf479fbf
c5d014188c8e8225f676351f1aeb3a81a53838f8
40534 F20101107_AABZIW currie_t_Page_14.pro
797799a73776b814979926ec8f148b29
036ebb04158cf0865c556f69fbf85ac96c13fba3
65666 F20101107_AABYZA currie_t_Page_64.pro
4f8a89fe1cf930668167de6a90ae60ee
4f70cb6237a733a1a375f9795289c157d5afc3ce
468 F20101107_AABZBA currie_t_Page_03thm.jpg
fc3f8917cf21e0546b57bb7d807552b9
42f1b51732ad3e1c429d822c43ad9454dd6549b0
57276 F20101107_AABZDY currie_t_Page_61.jpg
dc8b1e1fa9b3700926ebb0335d50e127
b60917a6c566b96ac91634ad75bce60bcdb47b79
35922 F20101107_AABZNU currie_t_Page_18.QC.jpg
bf4ae8fb212d67ae0782299ed443120a
67f13c3eb9125bd095d3fd7db7a94fce2c1feb9c
50335 F20101107_AABZIX currie_t_Page_16.pro
fc4b21ee96022d900bdb01d5ca77e8ad
3440decb89893ce8d6778ef5b1bcb496abf31503
8879 F20101107_AABYZB currie_t_Page_22thm.jpg
0007065da15d1fd467c5ff9d6dce09a6
1427ceec1177f192599ab465a48a7f9f8ce93c69
100816 F20101107_AABZBB currie_t_Page_20.jpg
c9c898534b25e54ea5746e0f25fa6f3e
c79b78454fb05eca18e804543bfb7f0eaaac44e8
49790 F20101107_AABZDZ currie_t_Page_63.jpg
806518219168005a54bec09c32f7a03f
53679e61820ae8989b2d600bfe1e7f65ad17dc64
8608 F20101107_AABZNV currie_t_Page_18thm.jpg
b3795f43e6d92d679be4581b2039763c
83d5a4e8a405003ea5cc02ec3b2ba43b2c1ce1e2
53258 F20101107_AABZIY currie_t_Page_17.pro
ff9496daffd32d2b67d8c74380943ae8
70ff3c672f46a7163bdd5b49463e779223bd4414
52935 F20101107_AABYZC currie_t_Page_53.pro
1df1626fbb7affa9e7726eb7eb48079a
e1ea408340ffddb717853988e0f551d27148ef4a
F20101107_AABZBC currie_t_Page_15.tif
7b45882d7bee0121920072f12d363f18
1ffa169d066265a8072352a91b3e2381f91862a3
32483 F20101107_AABZNW currie_t_Page_19.QC.jpg
6d291984237b5e5eca078bf61c39297b
a8c14e1b2026732467d4aaae5ae3a7cd13de7ccf
53589 F20101107_AABZIZ currie_t_Page_18.pro
050ee44249ae466019e5de6393f68f41
514479096382c90c87a0521d3c140d1d8ce6d0a5
F20101107_AABYZD currie_t_Page_64.tif
d849659bd65208bc3aebb9125aa0fe9c
bf0bd36737f7f266ce14aa7ac375a674b22be55b
55320 F20101107_AABZBD currie_t_Page_38.pro
a57f8187804472d7a56e9295497b4b09
b673cbf509ea7c3633dfc7e480cbb17b56cedf9e
632288 F20101107_AABZGA currie_t_Page_60.jp2
4f60b01d89990e87cb3e1528ea53c8c2
b1b7c0481b7f955bafd06b68de43a6fbe5b8c9f1
7690 F20101107_AABZNX currie_t_Page_19thm.jpg
8ebe898b825dde088adb62d5a6545c0c
8227666ef687f88d20ea2897d6b1641b659f8349
1051927 F20101107_AABYZE currie_t_Page_31.jp2
99968da64f762fc1df0ab83afc1fa7f9
0e4f6d237438d2a28da961a097edb88396d292aa
8241 F20101107_AABZBE currie_t_Page_37thm.jpg
ff042a019a8b222fd3212e8a6d4847e6
d8ccd7c0ce1872c0ca68f2b3ab2e2deff1ef36e0
601411 F20101107_AABZGB currie_t_Page_61.jp2
aaca3ce1159f14d047a026799f3231e6
2c34158bb6a3553ef4f031e7bb990e78dd95bff2
32437 F20101107_AABZNY currie_t_Page_20.QC.jpg
4e05a5916cf62e742e0397313f243381
0a57b74f37e92efff45f0d826c775e063440e121
54258 F20101107_AABYZF currie_t_Page_04.pro
2ee7a72781b5e9b84fa9948f81d77b1c
f2b3d4f132102b6268fb8558bff5940813395d7d
7677 F20101107_AABZBF currie_t_Page_10thm.jpg
9e3f05501889d9eb1ec249f959a1e23e
e9663fd97d4ff79412b02db4b383583ac778cd52
433554 F20101107_AABZGC currie_t_Page_62.jp2
5c9e2c43543f5c039c1024e60cb45a71
c29a785489e38ff3bbea3632eef35c231c090ccd
7911 F20101107_AABZNZ currie_t_Page_20thm.jpg
4e5487e96ac80568627add71cbecb708
b0b319e089ff221f0b25d2f589d35fd036916198
7480 F20101107_AABYZG currie_t_Page_07thm.jpg
fea87ec6860ec25061c1799cc7f05143
77b788fa639779955e47184bb7f932efd7848ed9
55269 F20101107_AABZBG currie_t_Page_42.pro
686f9d27127f13a5ca62dfcc847867f0
0a02e6d9068527bc3948bede98978a8b53ca551d
2039 F20101107_AABZLA currie_t_Page_20.txt
4ac0933992362d1f7ee2861e8b608573
878fab84bf14d089fa9d47b5667d50a13a443938
579563 F20101107_AABZGD currie_t_Page_63.jp2
6ea817181ff26403c21b93a44d12e240
ca1e3ab8c225cd0a86f9bead233ac97d048f9be4
8754 F20101107_AABYZH currie_t_Page_17thm.jpg
ffbcce9d3cc4b64ba4222bf35ce515cc
ed66d4da16b9638ae11ec6cec2242950050664e1
52623 F20101107_AABZBH currie_t_Page_44.pro
d181eee4851a3a0cb32377d663d1c56e
54ebd89345799dc5d826fb9137a2541057aa774e
2173 F20101107_AABZLB currie_t_Page_21.txt
4178f1a40ca53815a0f189676e237896
127e30bb6e686a9cc63391aa0b8728c6ec858584
F20101107_AABZGE currie_t_Page_64.jp2
3b9f3059a60af578e750e659dd0cb28b
afee5d5226e942d7af15b8f539b9cbf2eeedd265
13068 F20101107_AABZQA currie_t_Page_52.QC.jpg
c972e52fe3682a40e1b65c8b81e41c22
ef3b03a0e331486425ea677fb138ea731d2446f4
34836 F20101107_AABYZI currie_t_Page_16.QC.jpg
75808f38973e22b8632679dcdb9c00be
7ca53a3b835c4d377e58a82281591b75c34f728b
38891 F20101107_AABZBI currie_t_Page_28.QC.jpg
3e4b1c6c856de2f618a6ec073b73714a
3e058c61739b31cce15794d83074f14efa51b7e8
2111 F20101107_AABZLC currie_t_Page_24.txt
93db2a6e4f0ea94fce16790f72477209
9513e7da54f1387e03e1eab4d7df3843897b3c88
1051939 F20101107_AABZGF currie_t_Page_65.jp2
021d1e08314795a8e75ea249a1170ef0
35781e9ab7534e5cebce0a9e3e92041eab46b383
3976 F20101107_AABZQB currie_t_Page_52thm.jpg
854e5c53afa88459644057b637d33abe
b7438bd3867d558d912b651cc3a2e0419a08ea8b
4691 F20101107_AABYZJ currie_t_Page_63thm.jpg
63b7a27ba7e257a9905ad3803be599dd
6352d49def27c64a315cc08e471c1cfd8f85f47c
513347 F20101107_AABZBJ currie_t_Page_36.jp2
d578bbcc2281b98c6775ba0a2661012c
9ca7c62a091a3968fbe32b13271453edb0ba61d8
2073 F20101107_AABZLD currie_t_Page_25.txt
ea2354c30e080ce24e95f999dafc502c
a0c101f9ab1a41c363aad0ed5a39cbf112b8177d
F20101107_AABZGG currie_t_Page_66.jp2
38b422135730ea5a4c41ca8e0e97eced
9a71a51a0cde4b06d196c9d37f16ab095a26bfa2
31967 F20101107_AABZQC currie_t_Page_53.QC.jpg
cba81db6bd34db5a393d15ec3235204c
20100cb5844e0f174d45bb5ac610e2f580cbe4b9
37913 F20101107_AABYZK currie_t_Page_42.QC.jpg
187b639de59436a476a0a603cf50d59b
e71b4e25aff93efafad2ece610711a52f2004cb2
41276 F20101107_AABZBK currie_t_Page_52.jpg
5bcf69727551e3f1ce461b47a19849ad
f176bc4c49205d0db8dfcd29cd5f7e816ac7e9c1
2257 F20101107_AABZLE currie_t_Page_26.txt
8279d03fd5d9b3a218c13390f517a187
d540c82306a038a6c2cfb9fdee5c8a3f85096eed
679045 F20101107_AABZGH currie_t_Page_67.jp2
ccdec9f047cfbdb59d171d06bc908e33
56a9c5de9467323db057c3d4d6f2a9ba270ce18c
F20101107_AABYZL currie_t_Page_68.tif
d33ea7c895206d481da19b8f9fd2efb2
9bc9632a5ad8854a8ed1ae627c83d549aae699e9
53418 F20101107_AABZBL currie_t_Page_15.pro
528fd08f33b852cc8995ff49e410f2ce
43cb1e508952365bd70a805c8429a0140b4e7df0
2092 F20101107_AABZLF currie_t_Page_27.txt
87e129a52c3b7fd8cef0b08647f007d9
d400454354d3edae04923c8d202e6dc03444c961
656844 F20101107_AABZGI currie_t_Page_68.jp2
18202c868aad7df1a6d06b3920256c3c
2df74baf460b3d1e8f0d03414def8e4adcb09929
7986 F20101107_AABZQD currie_t_Page_53thm.jpg
8104e14ed618ec94732b1871876400bb
e48978af29f2eda3f47cca4424820c36703ac786
19868 F20101107_AABYZM currie_t_Page_68.QC.jpg
7c89efc2fddb97e7a0bce7f1746e9dec
a94e5e5ae5d452d9ad0fe68ef9e9bc418e525864
F20101107_AABZBM currie_t_Page_12.tif
1cd40a1540fe0dc1444c2012d8b345c3
de6413973dea453b7792f5cd9dd0ddd5fa31a83c
2272 F20101107_AABZLG currie_t_Page_28.txt
0cccb6ad8e7045fbf240c5aa9f822301
de601c1d12b60aea6b3ccecd233efd3287cae273
F20101107_AABZGJ currie_t_Page_01.tif
eccad1ed93083d153b9350e1ade28f2c
305d4e2f8fc388883966d2ce9e0774a01568ce64
4524 F20101107_AABZQE currie_t_Page_54thm.jpg
857418eef11c0eac4967139137e5058f
92d123cda5fbc59ca833e94f88afeb0f25cf685a
24419 F20101107_AABZBN currie_t_Page_09.QC.jpg
b5e5206b3ea4fa97dd6f3047d166071e
708de80066295438eac3fbd254bcd320277d2051
2099 F20101107_AABZLH currie_t_Page_29.txt
e32605a341dc7fa0f8463d76c91f4f59
f75f2df5203b828d001acbc7e3cb8174133fc553
F20101107_AABZGK currie_t_Page_02.tif
164b1d4255bcfff2e41cfea92b1afd84
19a45385ab18e9381d3328ae8d9ac05d582ae1e3
23936 F20101107_AABZQF currie_t_Page_55.QC.jpg
b40a0370921e189afb58860c02d40ecb
ea43161a8cae824a02bbcf3f92f6ad918205969d
83 F20101107_AABYZN currie_t_Page_03.txt
8e8d8457fd3235a20b25a44d75a1d511
ffad04068decda1a2ede89d3cc8a4b4c883358fd
35828 F20101107_AABZBO currie_t_Page_62.jpg
9208f1396256148587e36e48a51993c1
9edb6f6bf3184e48d0adadacaf24c5684b9bcc69
2085 F20101107_AABZLI currie_t_Page_30.txt
72ce28c3d51e049c64eb298e5331ebb4
138f9306731727717aa7affa9d550b3dc0109213
F20101107_AABZGL currie_t_Page_03.tif
58eba0be24d50597060c5605059ab7b3
3f6f42773ecd33f4447e155a66581dce3e3dc2aa
13514 F20101107_AABZQG currie_t_Page_56.QC.jpg
aa5dc0cda2c6cf473ce85718d3d6346c
330843a906db8f2d7c3c5bea4e80d44a655f13b0
32078 F20101107_AABYZO currie_t_Page_08.jpg
d988e5c78fa379bcc4bfb3dccf14c6ef
c1cd0a6509a961a3b211477cc601cbaa0ffbaf97
9330 F20101107_AABZBP currie_t_Page_65thm.jpg
674249be543ad945fe05469390b9a75b
4dda16286a79e4dbed183c543525da6906c4d3af
2192 F20101107_AABZLJ currie_t_Page_31.txt
b99fd794b3af0c107c7e030dc7f3edcb
bce164a34ac5728ec45bbfdb59fb84cee21c949d
F20101107_AABZGM currie_t_Page_04.tif
22903cb9d284e7e35c2143182f837b27
da85b296e3d937fc840c567a964c872eadff0aed
3709 F20101107_AABZQH currie_t_Page_56thm.jpg
67fa4e905d96f261be9ec661f41f43d9
371ffdbfd41d6e06f51adc5dc5102b01d7e6142f
5078 F20101107_AABYZP currie_t_Page_61thm.jpg
ef4f9764a36365a10eb519c5135bec4c
148e0876f468e7c7049ccf603a235e9bb209f934
F20101107_AABZBQ currie_t_Page_56.tif
2d1c2201f38002f24aaf8bac3320d35c
9347c0d468ebde68c4fe210b9f5f86256f287f1c
2129 F20101107_AABZLK currie_t_Page_32.txt
42ef6771fabd41260ce59b9ff9b1c61a
6f3095e51aed85e8c58eeaac1bd606505e012bec
F20101107_AABZGN currie_t_Page_05.tif
9e5e13e4dcd3889c540c104dfe610544
a19c8c55d0ea8801fab547f36187148aed977fce
2997 F20101107_AABZQI currie_t_Page_57thm.jpg
4bf32882ada8127710f67d9f236a3812
a907e46218c5f214ac52b40898cf13184fba2a5b
8506 F20101107_AABYZQ currie_t_Page_62.pro
14d6bcb7ced9aabe1045860858d454d6
954b6f932169ed6844851156c9ecf46fc6175199
F20101107_AABZBR currie_t_Page_07.tif
4d383ee9a4e4d17bb0c967924db14d07
a4006957bd973d97e89534f838f7d6341159294e
2071 F20101107_AABZLL currie_t_Page_33.txt
e77e6386f4e06c91b8849b5a564cc9cb
2e3fdbd7cf778e77675be464f368f90abfd5d6e8
F20101107_AABZGO currie_t_Page_09.tif
324ed6c638cf3d8bd2bac587ed4044b4
6925bdedbb179c55caf4cfd88dd19f11606e3ec9
5820 F20101107_AABZQJ currie_t_Page_58thm.jpg
5c6b6ee411714554c4dcc972826fadf6
81d5a00ad81c5ff32355682421762c2b12cc0d9c
35231 F20101107_AABZBS currie_t_Page_29.QC.jpg
9a77127d18c1ac439b876971b9abe9a8
94c4a8d2ebc8489d9abd239bfe900f1f0e6b3a3b
2164 F20101107_AABZLM currie_t_Page_34.txt
0bf466380b130509cf11d12c0fa59e9d
99fcb5303ef3fccb68f2a085974b0df943a3b40e
F20101107_AABZGP currie_t_Page_10.tif
ee3b1602d222b1f276a0955fcf7cc263
18607400507b41e27079c6055677bcea9badbf08
965619 F20101107_AABYZR currie_t_Page_58.jp2
ef41938ac3ea15a9455d19de5a441941
d153103281cba7d3e40123ec8f05d00e8c293bfb
4420 F20101107_AABZQK currie_t_Page_59thm.jpg
fe3c785cf5b63559ee605503b9ca89b6
bce46438a989450b8fdaddfc7e920fda8959e0b9
56091 F20101107_AABZBT currie_t_Page_45.pro
bc8deee83b872fdf70127a39e2dc6bd2
1380627e626de685a8da2adc2c4eb38591c87700
2048 F20101107_AABZLN currie_t_Page_35.txt
eacacd0061fb290d36cdd16bff46b513
913dd20fac530190d8b76d510b12b5dff9344702
F20101107_AABZGQ currie_t_Page_11.tif
6ec3b75ad34c23a67b034a70bf7ba62d
95762edd69bf18beeaeaeb2918cc07c28f48de92
2078 F20101107_AABYZS currie_t_Page_23.txt
cefb02099a178a81197172076f1a660d
d4498f8b3573c6047c6b6762a9c3ed46d0bc09c9
17590 F20101107_AABZQL currie_t_Page_60.QC.jpg
bd972278597d82bb9c3145174a6cbf8b
47c17f8103538fb1b71286089d9ac8669d5af4a4
904 F20101107_AABZLO currie_t_Page_36.txt
54dc6fe51e3dabdce701c6c3d198aea3
c7b74a56ea6cae21b424209f73be7696ebb589e5
F20101107_AABZGR currie_t_Page_13.tif
6a43a872c348740df299bacdccf3c8f2
656ae00bb7dff9b864b1688a18deea6d41d90bc3
34920 F20101107_AABYZT currie_t_Page_44.QC.jpg
85766886cbf13d5e37c479a733ea334e
7d3f0ec9fce35a569a61a949a965ee0daa97ad82
5283 F20101107_AABZQM currie_t_Page_60thm.jpg
a8bbff31d1dfc1449f3ec9590d825fa8
dfbb32ed98181b2feab2d7001ad1ab521133e97e
20580 F20101107_AABZBU currie_t_Page_58.QC.jpg
cdb408d4c2b97740c8be8904b3c14ef2
c6f951fb891b11ad1d869f6bdde9ca73feb4effa
2138 F20101107_AABZLP currie_t_Page_37.txt
7964f711ae72550d64d02d2ab06acbfa
715e4970f057776ed7405293085f1c7323dcae3c
F20101107_AABZGS currie_t_Page_14.tif
8c4b12e987edf8362b1dc07879125b46
f8f0fbb2364cde8e1c8ff3a8cf41055062827e91
2012 F20101107_AABYZU currie_t_Page_10.txt
dda1d8436292864b0d6ce18d17be6be2
31bfefc473bbc707639481c35b70421023e1f4bb
15667 F20101107_AABZQN currie_t_Page_63.QC.jpg
478237e2af6e1162f498a7c079004b29
dacd73412d46348eaf13b91c826e67a2ef4ec259
73305 F20101107_AABZBV currie_t_Page_50.jpg
d91f055c974c22f0a15d248392ef0dd2
0f70c31f30bf58b1eeb6876ff72c6b823e3a8d42
2215 F20101107_AABZLQ currie_t_Page_38.txt
f9e4a416e9552bfdb9b8b3884d1227bf
464776cf60b9e93e53e1341431067cee5a8e7775
F20101107_AABZGT currie_t_Page_16.tif
b6b54f4a84ce1f78817db181efe98227
44c0be0c94e82fa1a60b37ca0e29ec6ad5329fdd
17643 F20101107_AABYZV currie_t_Page_61.QC.jpg
60a21577001029d7922ba0cf993c910a
aeb734459de1d8dabfbe5c09f3e0e83b1440c72c
37412 F20101107_AABZQO currie_t_Page_64.QC.jpg
2f142328c6c26241be1e536f61e25d2d
068806c3dc7a537b2e1b86aa0bc1e00c0b4c4144
103911 F20101107_AABZBW UFE0022577_00001.xml FULL
fec6c787672719176de91f1bd49ae916
3191ceae345cdfd0b79813db83769ec1e61e2446
1992 F20101107_AABZLR currie_t_Page_39.txt
a9e2caf8fd3b31dbe19618093c288c3a
4e762bfb6bf5808d6ce3d5281d4cb8a4559d469b
F20101107_AABZGU currie_t_Page_17.tif
82febbdb8b5f8644e3c0f8b6362c3704
b9bd61ca97d22ae1dd4a3358ba557808264c8383
F20101107_AABYZW currie_t_Page_32.tif
7c7b277219d95fc3c7a2e8899c65eb1c
4f702edb77a2a1cac06bc658244e4e1a994987ee
9081 F20101107_AABZQP currie_t_Page_64thm.jpg
35f72d9fa368f8f178087b50051add50
7412aff6ea6eafde4aa6010e2b8158fd457e1442
2219 F20101107_AABZLS currie_t_Page_40.txt
4cca82d939d1fdbaa90495676ecbc130
1094430184310b26ec560b4e0a275cd81f35c534
F20101107_AABZGV currie_t_Page_18.tif
b3431fad8d8d58a8a2b400874bee3afb
38ddc5b39d1ff6cd1d29869b762e2b1b3bb5f4c5
37488 F20101107_AABYZX currie_t_Page_32.QC.jpg
3dccd414678514938ab2156e711adf21
fdc5d84e9743f4fe0524379ee1fdc7fa05f43ae6
32617 F20101107_AABZQQ currie_t_Page_66.QC.jpg
87b1a4ab844798bab85e32920179f2d3
539d2c22712391d3188bb8d2c022ab69a03ba4f2
2120 F20101107_AABZLT currie_t_Page_41.txt
c433f561fe60bee0ee98444cc3550c3d
df4a0924722c6236fad372f1525a78fc9e9a4635
F20101107_AABZGW currie_t_Page_19.tif
37a5ac3ea2c095c2a2b78a97f0f9c590
f024f03af965ba11b0c9baece5392accbbaa292d
2187 F20101107_AABYZY currie_t_Page_18.txt
a494cd8db7c39d4c9030fd57051b3a0f
0296804e0c1a465f47143b603e178e5279b56fd5
8589 F20101107_AABZQR currie_t_Page_66thm.jpg
30def6f8747003d35abcae7d6ea7224c
f73f395145fd79c243f9956d1b5b30d4fa1e62b8
2178 F20101107_AABZLU currie_t_Page_42.txt
76f80e58e9c49789c6d91475bcc5b61a
beb0252c5a0028c9cc3b44fa09220f3df63947c5
F20101107_AABZGX currie_t_Page_20.tif
899200890f44781e6addd5e97d1be710
93d350ac52796187e067b846e78246493c36fa40
52741 F20101107_AABYZZ currie_t_Page_23.pro
4e7f05d03116b4b7dabe3110888465d1
743b67f4a15453fe485d7479d598bf126005abef
27067 F20101107_AABZBZ currie_t_Page_01.jpg
77ce655f67617cd022b99529c2172c21
9a0eee042b2b0a664f3957f4330ba3e8f0ba2269
4581 F20101107_AABZQS currie_t_Page_67thm.jpg
54ca7b7f5f3a090ee4a483985ff67428
4f1abfe735ba76d3288e6cb2720ad984693ea670
257 F20101107_AABZLV currie_t_Page_43.txt
2affef013d39c54f973ad0b78055112d
0c476c00e6fc57d60a4b921cbb26e141a677672a
F20101107_AABZGY currie_t_Page_21.tif
22ad269cd5cc9d8b9ad8464106058217
6435e4c87c7353f7f380125e1d9a24cbf386777e
80606 F20101107_AABZQT UFE0022577_00001.mets
b3b956e15e2367af1736f6773f3de9eb
9c3b5216b678c00c278708ad98fa2d67cfc95960
2154 F20101107_AABZLW currie_t_Page_44.txt
1e07858408cd6bbdc3cfaed104baf5e8
0a7192df63d9640b0e9532acda094a898be6ea65
134945 F20101107_AABZEA currie_t_Page_64.jpg
831399770a7d026da0fab688bbab5740
c0e8f38c5aecd2892bcf5a8bb0693db19c5b3f4b
2211 F20101107_AABZLX currie_t_Page_45.txt
591cce02833b2e41064814fbffe6c407
62d8b3edb4ee5c250a4972965e3a3dfd15c30dc6
138869 F20101107_AABZEB currie_t_Page_65.jpg
c0e1a368c33a78ac5c944de6cd1663c8
a0dc91e6aaf3aa5c53f7610ce9de3b0010aea5a5
F20101107_AABZGZ currie_t_Page_22.tif
d43f2076638d1e2b467eefef5d2c1055
7afe39afa48abb6b8a7e1186feb50f76e8994d84
2176 F20101107_AABZLY currie_t_Page_46.txt
0874da8acd3e7d17222ca295e84596f0
67ff55ab3021f75e66e75515e18878354218ed70
110872 F20101107_AABZEC currie_t_Page_66.jpg
1afd0f10b6f19feef5ab552ca7dd3fcd
b351155804240ba64fc298c88e108087e55adcd2
2193 F20101107_AABZLZ currie_t_Page_47.txt
1ccf06d6040d2ff35d2828bdb22e0dc0
1b879e7977ad20e421606cd7255e8adc62114553
64587 F20101107_AABZED currie_t_Page_67.jpg
bff6c20823d76f76b5c7e3bd9b42b9b3
1539e796eb88ba99c0890e5b39a8b791d18d01c7
47083 F20101107_AABZJA currie_t_Page_19.pro
648273fd5b12f5828694f9ab8f5a7edc
2d67811490e05221aeaee853649477be8c18bcd0
62314 F20101107_AABZEE currie_t_Page_68.jpg
8e3d8f9f9317df0bab711d700a85c863
ab0d49142ae5c1b5b6c199f20b240b5ea4263d3d
54514 F20101107_AABZJB currie_t_Page_21.pro
83019cfa4b00ea511a55aa2b54141466
aaf554352be9a66d1a7d22a0a9fc734b68d43e2b
35834 F20101107_AABZOA currie_t_Page_21.QC.jpg
4424eddd516fca9fbb6bf1b27824d2d5
a15aee76d63599aa7a09f2b78b2a49e45abf4088
256803 F20101107_AABZEF currie_t_Page_01.jp2
fb7765e3e075db2c3f605eff3fd7bd99
af02ef111f06053e4fca4061cb8a609a60f8fe6b
57097 F20101107_AABZJC currie_t_Page_22.pro
1109bdcfb6ce6feb68bbba0e4f4438f2
c4cf0e5121bbf0192b2018eb86c9375bf4788de8
29622 F20101107_AABZEG currie_t_Page_02.jp2
77078bfa7151b18f7d597d269b2d421c
2d18372cd80f8216b06d365a84db1bd68386e224
51804 F20101107_AABZJD currie_t_Page_25.pro
c195c5d997e0f83fbcc9508321594c84
1f3dbc0996ed508e8605ff6c4d215586c60ce0a8
8709 F20101107_AABZOB currie_t_Page_21thm.jpg
337183a4d6c2e20237357776156b2168
6980d3850de9512db7b92cc0e1b12e737be8a47e
18473 F20101107_AABZEH currie_t_Page_03.jp2
c172515196938f013e920bfc0974faa2
084574023647e7e15cce174199989faa99814e4e
53009 F20101107_AABZJE currie_t_Page_27.pro
0b9b0d2481b566601f99141b23729619
83fcc8d347409449445671aae1b1b8af9f3ca806
37021 F20101107_AABZOC currie_t_Page_22.QC.jpg
cce85fee8ea3de72b67423472fde084b
0180b4b575cb0d51e66820919f6092f18aa27185
F20101107_AABZEI currie_t_Page_04.jp2
d0cad2199dd9ba7478c296752633fb95
764fa4a0b443181516b5ebd0e7f9b72c8d43921c
57032 F20101107_AABZJF currie_t_Page_28.pro
d106d90e9a657089c171bb4d7663ba9e
4f00e4c3daeb696cfd95e3afafe3a798ba319f6f
34491 F20101107_AABZOD currie_t_Page_23.QC.jpg
1592abedc6c840d669c9f2f5fca316e3
1aed2a217bb70a6a7b1ef234a74b0bb9af1dd9fb
1051928 F20101107_AABZEJ currie_t_Page_05.jp2
bf74967769150fffe96447afbd88f56d
5e413d8b84b59f7638ed8eb405a4d43b5eda6e04
53219 F20101107_AABZJG currie_t_Page_29.pro
dace02ffa49b665caa43886d3ad61d7e
98220228f6c32aa1c8c3ef497fecbf59119c35f6
8616 F20101107_AABZOE currie_t_Page_23thm.jpg
da4268447b27808c7621a0b9cd594479
24529cd777324a9a3f1176a8dfe6e25eca6ad384
1051984 F20101107_AABZEK currie_t_Page_07.jp2
ee8b572ded30ae9308d09246ade75d24
dbae30a8e3063a9ecc84512d554cdf1ed08c9668
53008 F20101107_AABZJH currie_t_Page_30.pro
cd4a2aa9c13a7bf0884eaddc21c9ec48
b3aa5df16933982145cb8fc8c2b6d80a862b51b5
33596 F20101107_AABZOF currie_t_Page_24.QC.jpg
e143d052089cecfc609cefb657d6f933
d159da323d246911ad9999bc7614a61d086a2c4b
547562 F20101107_AABZEL currie_t_Page_08.jp2
e4824c883c7e00c5cb0e36852e050a2f
12c617f2417743a523f600df8affeca94fe38e0f
55887 F20101107_AABZJI currie_t_Page_31.pro
4f8cb4812091141c5b5e820371ee6a18
23aaf6a98dba1018a61a494fe564db29541ed8c2
8416 F20101107_AABZOG currie_t_Page_24thm.jpg
6fa45fc6788712ddaa6d96dd2ba11376
4111029b8a21559d2273e1ceae9fcee3a5a2f7b6
1051957 F20101107_AABZEM currie_t_Page_10.jp2
1199e69ebab2374dd706dc0ca7f5ba37
516ed6a7cc4bd30e39ae0fdf5a594d5787c147a4
54097 F20101107_AABZJJ currie_t_Page_32.pro
a2f33ad5019427b67a5073807e0406d8
0dfe25288ca14d0f13bbfea788d6b281edb84a93
34219 F20101107_AABZOH currie_t_Page_25.QC.jpg
8c02e8f3730d180fe2d810c50d8b32ff
0eadf7b51fb58b36925556a7be4192545ddbf1d7
1051967 F20101107_AABZEN currie_t_Page_11.jp2
f261bf774f4d1aff6607fade30c450d2
908474614306724fc732613e6b8daa11d1751f0b
52589 F20101107_AABZJK currie_t_Page_33.pro
235bd54c08a66c8d5886cc64307dc04c
6e4535f9062867cbee569b362eb1ccdebc7cd8bc
8657 F20101107_AABZOI currie_t_Page_25thm.jpg
0221c40bc193728404197638e176f463
de07ce393b1dc796866efc313fc93d15227225d1
F20101107_AABZEO currie_t_Page_12.jp2
61654c7579daf45d19ce848585afdc9f
7301dd9e89f45f1a3c08d80447068139cfc85e8c
55221 F20101107_AABZJL currie_t_Page_34.pro
8361aa752d22b2e1b3326065db283086
8d70c501bb2dd477ca8016ec9199ecc87f06889c
38336 F20101107_AABZOJ currie_t_Page_26.QC.jpg
f419f65bfe964c07b83268d6a7442371
83e7ab9ae1e07ce2ef6ecd4df763acdeb502596e
F20101107_AABZEP currie_t_Page_13.jp2
1a9bd1f86041a1fe09baa000af9d3e69
e4521bcac1c73d96d37d54db15052c76e6471c41
52138 F20101107_AABZJM currie_t_Page_35.pro
657252363d9187be1a7ee8fc363181d2
7ce8854c3d50ad8d893acfe5e671e0dc193b7cfd
9545 F20101107_AABZOK currie_t_Page_26thm.jpg
6f8750f1d08a9fd6ae5339a728d62203
263fdf00c7e009f116255de07b75f09be0db4b1c
931009 F20101107_AABZEQ currie_t_Page_14.jp2
9166735013bd95d900a86638de6014b7
03b5ed3ddac9eb8bab59cd33b614c93500dd6965
22626 F20101107_AABZJN currie_t_Page_36.pro
94b6fc5c627a39be902622c1aea745bb
bd699bf853d061f20a895f090eb9df5f29ddc723
36083 F20101107_AABZOL currie_t_Page_27.QC.jpg
e7cb016017489be6d7f25796fc8135a2
1bb9aa36bba5bd6accdaa3d17473d686ad7f37e0
1051980 F20101107_AABZER currie_t_Page_15.jp2
4e93f3f4c50922f10513f05d449ef96b
25c1cf573c728c41458026ebc4944c980e0745f4
51463 F20101107_AABZJO currie_t_Page_37.pro
8a41c1205d6fe6037babf03b96681ce9
771102c54b1800f9a87267e2673beebd1a9015aa
9187 F20101107_AABZOM currie_t_Page_27thm.jpg
81fa050f73d84a0ec18245b822f1e816
382dffd84992603bfef95003c4dcb8ea144a7f40
1051909 F20101107_AABZES currie_t_Page_16.jp2
0e92dd336b3ab5df74795b77e7a6e9d0
f4d3a7f17eb34148ce46a250d5e9ec4507592f26
49400 F20101107_AABZJP currie_t_Page_39.pro
8c9c9fcf6785896895e8c6dc08d45b8d
e2bf86027ad5b7041cf3ffb0e870561fc589dba8
9441 F20101107_AABZON currie_t_Page_28thm.jpg
3bbf3f57f92074b598585e4d80d88da6
9528131b46a5da80a673444d6d9bcf025282d60b
1051965 F20101107_AABZET currie_t_Page_17.jp2
d90b2a36f9748e3a5aee6cd3f7e5e5e6
38acd4701f188acfa5f1434e53920c212f110870
56062 F20101107_AABZJQ currie_t_Page_40.pro
f57b9b1e963f4e480602e9ee27ca50ca
b33ee34eaab200d86a1a4ecaf2cc41965e7521a4
8810 F20101107_AABZOO currie_t_Page_29thm.jpg
591f799ce4965b65416fa32174b18215
8d3faf31b1f31deda3c7cf6ffd398e87d60c2d26
1051947 F20101107_AABZEU currie_t_Page_19.jp2
b510ed70b0e405beb52fd0a920d2edaf
5b95d89a0c7ee44cc7419fe02ace24ab16d7123d
53428 F20101107_AABZJR currie_t_Page_41.pro
3d265d481f5e0022873af0ca2259fd70
f6ff86742f96de86b818274a5dd1d9d0cc94634e
36781 F20101107_AABZOP currie_t_Page_30.QC.jpg
3493282671473289b163c118ed0a64bd
ceb2b2cceaf18e65090ce411f30be69b3869a676
6400 F20101107_AABZJS currie_t_Page_43.pro
0e565086462390839a9d237c96182296
9a310b5466c65d5ae6c62eb33a9185982072c73b
1051978 F20101107_AABZEV currie_t_Page_20.jp2
a47d997a19d65ef6aeccc4b0018e1319
23c3044707ea7d6cfb197216ccc6d376ee623a1d
9440 F20101107_AABZOQ currie_t_Page_30thm.jpg
363e9be3f9c6c6e766506c7529ce2019
eeee90eff4243d53dbdc554f7d4ac0c4a96c0029
55417 F20101107_AABZJT currie_t_Page_46.pro
749992ec02dd30a9253c4804e869bce3
d94661f196ef854876487c725ae3160165bc6e33
F20101107_AABZEW currie_t_Page_21.jp2
08f6681abd6f7092497d20847e49f7f1
704eb9fca00e4f4b7745ac4cb30f706942c0d71d
37750 F20101107_AABZOR currie_t_Page_31.QC.jpg
51caa24e83a2b8eb7d5dbc76e368a906
487d201afdcbad2078e77e23b5551216b00405a4
55637 F20101107_AABZJU currie_t_Page_47.pro
d7a65e1175e0e01f36c9a78783570a25
5c45855fc169970c5bf9a9bada9735c812902808
9568 F20101107_AABZOS currie_t_Page_31thm.jpg
e464ec98b38c43cfd3ef15280e55b006
af68b9986fec420287d7f7e132410ee162d58c4b
52980 F20101107_AABZJV currie_t_Page_48.pro
472b39a5fdadfb3472fbaf3c9b0ed22d
fb9634d52bea526e24436d8f5958868b3408b08d
1051934 F20101107_AABZEX currie_t_Page_22.jp2
3867061775340f50b0115c1dd25c4763
c9b4f876ef6a11bbc1e50efe4b60f750392f669f
9580 F20101107_AABZOT currie_t_Page_32thm.jpg
16b60cd9b936f64e481dad7313ff9bef
80ed62ea2efa4b4279574aae246ef52f1d3d5ce7
55614 F20101107_AABZJW currie_t_Page_49.pro
ce06c62fa457e72b8055f152aa65af52
430605f17982dbb9a7c2a0a952518a961fe56916
4325 F20101107_AABZCA currie_t_Page_02.jpg
0aef360a9d72724e0eb475993db3abf5
b42c77f11ccb7be36a340dca18ce5cfb5548d153
1051952 F20101107_AABZEY currie_t_Page_24.jp2
8487d5a36a59364d01d57aca20eb4b68
ef89c02df72e485ab8ab8789598c332480d0e356
37460 F20101107_AABZOU currie_t_Page_33.QC.jpg
76514f990966a596dc6b390fcca3d75d
2459a4d6ed9c97f5e199072bd708737e1dc9bfb4
35820 F20101107_AABZJX currie_t_Page_50.pro
8cb286e8247f9a7aecb889fef2eeff28
73f14ea068aac599317f3c66bee7f1cd757a5f7e
3262 F20101107_AABZCB currie_t_Page_03.jpg
d2d70108d56dcc9298a17cdd8d6af121
ab26846d217218443f9a6a4b72259c22bc5542fe
1051972 F20101107_AABZEZ currie_t_Page_26.jp2
965b11d82544456dcd89583ffa5a8e7d
bfbfc84573de50ef1efea4ae157ef6a1bf549696
9238 F20101107_AABZOV currie_t_Page_33thm.jpg
62960dada62c32b6fcd174423a4681be
8eb7109a08650ed1d843b0639212f6e6418c00ff
1898 F20101107_AABZJY currie_t_Page_51.pro
2a95a0265311b9f82c1bc2c24a7ee894
661a17322e59b3a098e6764995aa42ed1180b653
112256 F20101107_AABZCC currie_t_Page_04.jpg
2c4cfc0b8153ab7ea47bf9968fab8967
3836b5cc53f5663b1fd8a13819d8b9a36ef11a1e
38057 F20101107_AABZOW currie_t_Page_34.QC.jpg
47203c03b7e3e642e609ffdeccb0422c
151d33e1048eb394205ddfb30174fa638ec47fe5
F20101107_AABZHA currie_t_Page_23.tif
f9fd11b239dd4c56b7cda739d98941ce
132172573b1c41d3dbaf84ba8cf9c9959aaf39dc
9966 F20101107_AABZJZ currie_t_Page_52.pro
bf7a1c38d795f8baf64950b99bce8826
121b08f01043f9b260ba8cc5cec86eeb70dfe603
49062 F20101107_AABZCD currie_t_Page_06.jpg
020cf6714d001c7598eb089fa2780ffe
9d8433cf4bfc9a6452c972191cc79bff12e35514
9524 F20101107_AABZOX currie_t_Page_34thm.jpg
3ffc1c410a79096244da790d9d23c04c
1083fef259bf5ce2174b8b0fdea71eec6c57b70f
F20101107_AABZHB currie_t_Page_24.tif
d05165a3eda945a556511e5265cef1da
d8f2bcbed3814ae65875dfba6be9610b2c17e5ed
127866 F20101107_AABZCE currie_t_Page_07.jpg
dc1267936423cd86bac357355c2dc96c
d355d619130e418fc312fe5b9c68379783d77769
36457 F20101107_AABZOY currie_t_Page_35.QC.jpg
62e5cd32777b084a120b63dc4636d55a
c82583e99943f537315ea218dcc00f5691875b69
F20101107_AABZHC currie_t_Page_25.tif
ee624cdfdbc62c9de62b94f436c8dfe4
c06bcc3ec13fc9b9a1771ff3d5662bdf9a3decd4
89114 F20101107_AABZCF currie_t_Page_09.jpg
b16be049ec53769776eb804ca06c0c8a
2578921aaea4a9fb85e359b37ca74ce0a164157f
9009 F20101107_AABZOZ currie_t_Page_35thm.jpg
27fcc29b332eb99759f6ae3b0a8a9f87
ec9df3e89a1e990e05ebe22a1c56e30c910da09b
F20101107_AABZHD currie_t_Page_26.tif
9e1546170c40121c802afd4791127a7f
eddd695f8b513331d6e6c7fa7f61b2c7a449252d
97153 F20101107_AABZCG currie_t_Page_10.jpg
c983afac0a1d11bcb9846612aba638ed
ccd31a44b5edc56f8ef384fdd417796b83a8ab91
2087 F20101107_AABZMA currie_t_Page_48.txt
9aa6cd82e3832b3aa4d59c3aa2a0bb2c
e18f82d39e0faac835fa6513e3a1cbfebbc1b595
F20101107_AABZHE currie_t_Page_27.tif
9a02ce83f825588582d99d20a36dfe0a
8a5a53d1aef6efc15d81c07693396fd4607804ca
105632 F20101107_AABZCH currie_t_Page_11.jpg
76ee465568624943fc0edede2e2bb0cc
d2e39a3da65c7e4fc9f62bd06d91227a5ce6034b
2184 F20101107_AABZMB currie_t_Page_49.txt
aba05b809be2a06f1f317c29e8874eaa
40b78d0ccb6a742e277c57c4444ba796e6c72893
F20101107_AABZHF currie_t_Page_28.tif
67354196541f15aff36e001ab42b2c03
38fcf9211f06305aae35b5520ba29e90c0e9be8e
107055 F20101107_AABZCI currie_t_Page_12.jpg
cb067e8e20d6721178dfa5c02ba087a2
c9b981eccf09d8cae6885bafa378e0f2ad7b3930
1434 F20101107_AABZMC currie_t_Page_50.txt
dbf64c83f0cf7819e0f33afcea55dc16
8fac31e78945312689a09d8a34ee27216a5f6da0
F20101107_AABZHG currie_t_Page_29.tif
54f004f4548864d4ccb49bc480254414
c9002d8a0fc759533722916a40273bad2782c711
108001 F20101107_AABZCJ currie_t_Page_13.jpg
05108180ee9199283569f0128572831f
3225e3616740e2f8863fa3b30dd8064e6dd1d779
141 F20101107_AABZMD currie_t_Page_51.txt
8542b13c72a67263a7f4f325e90ae465
4745879c8b8b45d41fec5cf77796cc62a2df5070
F20101107_AABZHH currie_t_Page_30.tif
dc2baa2e00f2684be189c1264d193ab9
8b519ff46b5147130056c95fae7a7af0ffa85e23
85682 F20101107_AABZCK currie_t_Page_14.jpg
2a1a2759b2bd66c2db388fdc2fb7e305
ff5e4347d2f2bf368820e8c3311c5165892fbfd7
442 F20101107_AABZME currie_t_Page_52.txt
5988e6bef241f8a78c42ae809428354b
71d20cfa07d4f260e75f755434ced71e8cb71943
F20101107_AABZHI currie_t_Page_31.tif
6824eaa0f0cf289b3651725fa782d9d5
f34446c2c92f163fe6fab02274bccdc7fc47ab3c
109733 F20101107_AABZCL currie_t_Page_15.jpg
dc0ea36861e27c5c0f21e60d3926a595
a4cf44bee21dbd0198b3b83b584badbe5d305a56
2548 F20101107_AABZMF currie_t_Page_53.txt
5721e08632872a1d977440287bc654c5
bac692e3a10b93dfad1d1607d6e87991c5709e9e
F20101107_AABZHJ currie_t_Page_33.tif
c17e11f583b5c30548d101ce69f2208d
3b654bfe05f6d7ff5379a5e09bf592a6cd1cae45
104012 F20101107_AABZCM currie_t_Page_16.jpg
34fa86738df7db5794e063e4f32b4e05
a40c5b791e10d72d81829b0c82aebfbbe2fca126
1123 F20101107_AABZMG currie_t_Page_54.txt
81e4cbd46dcc02b1198ddc2404a59fff
4dabc980aa4d21deb143b036a633b0c2f8ce3caf
F20101107_AABZHK currie_t_Page_34.tif
4cdb63dd647f247363265e5e342db8a4
9d552c654d8d107acfe3e24d542880a243f1bde3
110963 F20101107_AABZCN currie_t_Page_17.jpg
05d5b693d8c5045b8bdefaad425495d9
803227b53ea1b914723725c56cde7447543427b5
F20101107_AABZMH currie_t_Page_55.txt
33e634ebfc18b08a2c7bb95472f15fa5
4857c58228727f1f408d6f32e73ab9274e6c567a
F20101107_AABZHL currie_t_Page_35.tif
e7c39470bcb061085ff6876c2f0a810a
8d1aa23288565a9c1833b09111a70fb1f70fe690
109833 F20101107_AABZCO currie_t_Page_18.jpg
f41247dee592d842cc8838bb367c8fc8
ae9c77e1f1f45d93d3cf4e84e280cf7f5419cb28
707 F20101107_AABZMI currie_t_Page_56.txt
30903edb526827e3268c5d46207e2ddb
582f8e1192e5c445925308b244eb5d995ba7c540
F20101107_AABZHM currie_t_Page_36.tif
bd197e511a4f23c6f053692937f0f9a2
3ab1fce5e69e80d1f1d0e04883df22479b6d2539
98728 F20101107_AABZCP currie_t_Page_19.jpg
2cb34334601adfc88289fe13742a86bb
a985f9a99a9a1f72146f90877e061531c2b02536
461 F20101107_AABZMJ currie_t_Page_57.txt
0c5757a60abea121c0af33f68b38e1ea
5deb3b6b5d2acbcc42bf66de457d528493a73389
F20101107_AABZHN currie_t_Page_37.tif
3cdb4865b3d8f99f04326bc022d62119
0642cf6cfab77dad098a99f4377681f0dbae48d3
110119 F20101107_AABZCQ currie_t_Page_21.jpg
b3a113b14f987aeda15846dbcd4bcf3b
88717b663da0b26c711ec16f214fe71035a06964
713 F20101107_AABZMK currie_t_Page_58.txt
8c78b5df3cc3e2608cce900ff5cd3dd2
7b9fe3683e533858130dd4bf35cf5060c0456b04
F20101107_AABZHO currie_t_Page_38.tif
f68840494e769446e87441eb4efe3cc6
e33ca216d55a917a71544618ea763dee4d78f19f
113225 F20101107_AABZCR currie_t_Page_22.jpg
c99d9e64056eef03014e27a17928c4a9
baf304dc85125c4432d95f84a2837b8e60dc40a2
882 F20101107_AABZML currie_t_Page_59.txt
24f4a6de295faa1742d68967e07e9955
307f63e7fd2e87c5f83abeae5d415c178aec7215
F20101107_AABZHP currie_t_Page_40.tif
87b48c7f1ef362a5a05cabe2b67f14fb
d1657b5987aa448a8ff841d1f3dcdcbd1d0824df
F20101107_AABZCS currie_t_Page_23.jpg
ef22e60ab15db0276c1a7dc67896c16d
acca910ab5f63befe9b7858667feae7b7266f3d2
839 F20101107_AABZMM currie_t_Page_60.txt
eb312fa198f4bd9e44457f8a6316ab98
d97c53ea86019d1122c0a27d213d244cfd832bf4
F20101107_AABZHQ currie_t_Page_41.tif
4d08d20ebc10d9eff81b47fb97534fcc
a01c15044d738497dea585d4b0f15bc3368b52cc
104713 F20101107_AABZCT currie_t_Page_25.jpg
914d091473dae20b07aaeec821454a7f
716f5f708cba68785b94c0f4342c41723da58a5e
1102 F20101107_AABZMN currie_t_Page_61.txt
17944eb0e234fbe5546e33b8a67defab
3da93a9d0076a243ed15760039dbd8e085f8fbd7
111631 F20101107_AABZCU currie_t_Page_27.jpg
6102b1792b16e2900942e503b99d4a15
42368d646d444a5a0679c82d896843a9d476bb35
376 F20101107_AABZMO currie_t_Page_62.txt
2b954abb9ffdc0435a887faa951e0822
ca26d5ebbeddc11f8cd9448d7570e76b7d2b4939
F20101107_AABZHR currie_t_Page_42.tif
c8a528ffede56279172a6632c416c311
6879b2ddc12c03793c8c62bfd206f0877dde808c
1475 F20101107_AABZMP currie_t_Page_63.txt
319994422894965dba58c1617f0e6e0d
cc20639ac97923f7df796d85976036091fcc39b9
F20101107_AABZHS currie_t_Page_43.tif
b326c09b01840642130a789b9fe9ca24
b238a22d8b5bf5177a47fab8ed52f1cadf980d6d
116461 F20101107_AABZCV currie_t_Page_28.jpg
d9d18fb39f20a3d2c3ef3a2af1490c75
b6506b68b2a541c7df1f9881a62f0fa802383a23
2674 F20101107_AABZMQ currie_t_Page_64.txt
61a272a30b96f9da0f08e37b754ed016
490c7b405ff2538fa27f1c253f1a149d8a3a7ea8
F20101107_AABZHT currie_t_Page_44.tif
2b6b3d4186e77cdebdd4a44f3f9634ca
eabf3fb0cd3dd083306c61d46d658a16705d9746
106950 F20101107_AABZCW currie_t_Page_29.jpg
5cee592ac59c77b8ac7249d19c367d56
d7ff0ba5a1eb96c6303eba4e400a38f90237b340
2885 F20101107_AABZMR currie_t_Page_65.txt
a071b48d8034bd5c93ba871aca205c79
013c329ed036009766a26bf05dec8cc14f793358
F20101107_AABZHU currie_t_Page_45.tif
ee5936ac0bbceddafcda6d1de19674b3
3b3435f1534b4cca2444477d1ea915a1ba222031
F20101107_AABZCX currie_t_Page_30.jpg
e87cfe627f3639de1c7bf2b10afe5ef9
360cdd41ea7c8fb1582d6f883b42217ce5acfbae
2349 F20101107_AABZMS currie_t_Page_66.txt
e67b50c53573c7a24757445d6648d29a
aa9870c7046b2472b7ede37166e776ce7c45e783
F20101107_AABZHV currie_t_Page_46.tif
c641d539a43b524447f4d40e635b134c
8e852b8d20c0b0ab1237ddea531914feaddce287
53519 F20101107_AABZAA currie_t_Page_24.pro
38079184b4c68785fc20932a472bfa33
230b98a911864412cf5b9936fae2ebe787029c63
115719 F20101107_AABZCY currie_t_Page_31.jpg
3d97c19c3c6918353e71f880a8f59a69
d9665e1419ff895e8fcd9eee600cab0bb072a584
1317 F20101107_AABZMT currie_t_Page_67.txt
dd49a4c8e5cbaa5d1ce0109ed7fd911e
df902b5cf950b814dff3f936b5128cea9232e261
F20101107_AABZHW currie_t_Page_47.tif
20a0782317bd90861df50dc8a59134d4
3c137cb779cf46b26c4fa51d42a0b0414b12df87
118038 F20101107_AABZAB currie_t_Page_26.jpg
270444c66dc77cdf0c68de5599c91c16
bdc5118d1c9f23d50b4805c14c0abbb9a1875222
114521 F20101107_AABZCZ currie_t_Page_32.jpg
05b459efce96add9bd8338d5a87b6d7c
d0c77d5b05e040d5343d7a5e3db990fefd3ff7f3
1285998 F20101107_AABZMU currie_t.pdf
d5b3d2f73ad13a8d2b61790a42598423
8db45739decd178fa99264586b0e864257326c8b
F20101107_AABZHX currie_t_Page_48.tif
68b569c33c78beb4448a37c3ef933727
99be7acac212e7dc3f191d3d83453ed356cc5521
1173 F20101107_AABZAC currie_t_Page_68.txt
05f9f6a5c1d819a69e7126ba26ba625b
3ab36c447d2764b3ae2f7c9c88731474010d2f82
2060 F20101107_AABZMV currie_t_Page_01thm.jpg
fb454295eeca22d36ff7695b0a05a275
cca4d2582ae008a64547e54373e5c194431c872a
F20101107_AABZHY currie_t_Page_49.tif
7b88043ba73db6649d05127a4f1c944e
d625c2089444e9c01e469e8e134ec8aa6e6b302d
F20101107_AABZAD currie_t_Page_39.tif
3a3275bc9b70fe11772ae0bf68a53864
a6806847823e5b0c2c49da0ef54b8200a6ce7596
8567 F20101107_AABZMW currie_t_Page_01.QC.jpg
84c4da88bba8ec005ffe334722656acb
c0f2206bdfcc4bc1b1076c6c52bd88fe1a8ba180
1051922 F20101107_AABZFA currie_t_Page_27.jp2
867381ab2dafc486bccc4b194e51b3c8
5db929bf67c3e6634bab6c336d1adec036c0c934
F20101107_AABZHZ currie_t_Page_52.tif
db04495f742c9833ce10910a2c2d4ff7
cf941588e6426c6ba89e5c991a34f420b214ef32
1251 F20101107_AABZMX currie_t_Page_02.QC.jpg
2279b7e2fa0788069dfd188761d28347
4e256bf1896294f4c22354a9a25cb6ec3f98641d
F20101107_AABZFB currie_t_Page_29.jp2
7ab240801c00048bcf170d3020b58869
bb59a226f4597dcdafe9a088c3edd2295f744368
2283 F20101107_AABZAE currie_t_Page_22.txt
79286262c253c2bb4939c553d14d7113
cc139f1bc5bcf8b783d8c177702dd71a282ed28d
556 F20101107_AABZMY currie_t_Page_02thm.jpg
a8f31a6588191c6ac090540febf68110
757de2f0251aca70680cd9c38f40921c3c0397d1
1051949 F20101107_AABZFC currie_t_Page_30.jp2
3cef2b172b6032829243033f32d78492
9db5a788916b9a1ea5e624683b75bef2de91144d
4017 F20101107_AABZAF currie_t_Page_36thm.jpg
8074f4ca86777fc560d7500ce36b0a05
b4570a97e7820ee58cc8ce1c8a7594a41c365729
36417 F20101107_AABZKA currie_t_Page_55.pro
8c94339a9dea22165c1da4770095771d
137bce9f50fea4f33f5db3df1f9be415cf22b1b8
998 F20101107_AABZMZ currie_t_Page_03.QC.jpg
45a9550770f47cb8a2e35b97a0080c7b
4db72835ee081db821a8e18748c680c4f7c5ab8c
1051955 F20101107_AABZFD currie_t_Page_32.jp2
8f2b7e2c32a46bac42629848cffcc0a8
adef17a6622246133681c94b8bef782d81f95f9a
F20101107_AABZAG currie_t_Page_06.tif
496794fbefc58b47ca77eeb5fbed8d7a
c12f028dc07f4920702aef818074b01f89d864ed
12660 F20101107_AABZKB currie_t_Page_56.pro
91fb2c4b36efd70be8688137c5ae8ae0
b93f4f0497d6d17acb5bf490a8e09a46ae264869
F20101107_AABZFE currie_t_Page_34.jp2
ec8057eb40feaa18276a0de06e2fdb99
f8b2e2b29f8500f1cf9ce2e9456c6df39cf81b84
49022 F20101107_AABYYH currie_t_Page_20.pro
6126cb3e41fe39d8fda5c745afc5985e
d69906e038a9be6a9d705009d7136a84f9f14f14
11149 F20101107_AABZAH currie_t_Page_62.QC.jpg
0ce5b013110e8f1185064b557a3ad795
80e5d541c410d41b37ee85f6134d8a656292455e
15830 F20101107_AABZPA currie_t_Page_36.QC.jpg
1320995976a01f315845f45a6e6f992a
10e05467d405748d3996ba1df744d1b56713bcf6
8088 F20101107_AABZKC currie_t_Page_57.pro
204960f9285412e687165922b51545ad
9527537519307d67c1df633bb417d7069da6061d
1051979 F20101107_AABZFF currie_t_Page_35.jp2
8f9f00b7ab6879a8d1f16966159d8620
11c5b77d58e37b86dcbccab63fa4797bc071dc7d
F20101107_AABYYI currie_t_Page_08.tif
07bdf0105eece66dbadd4db2ff399844
396c4adc97dc8fc5c436535f884db7dea1791128
9072 F20101107_AABZAI currie_t_Page_49thm.jpg
6d2e6cfb6cc985df295f80f8ae373db6
04aa5e17dcca5dc672f288babcd3fc8c2eb355b0
34322 F20101107_AABZPB currie_t_Page_37.QC.jpg
66e50a3b5b2d571213d9d6cb5dea3532
1ad1464a9b5c1277dc26e67a8fa8eccfba3e8275
12571 F20101107_AABZKD currie_t_Page_58.pro
fb03d8cae7085bd43effff5eb30d2d4c
91bca2f11c8cf2a2157010a555cd9d4fdcbaf831
F20101107_AABZFG currie_t_Page_37.jp2
a0f0f1c69841b56edd10205e843ce911
5647e8ffa00807ad5d4b088061dbcd2c364ca61e
21005 F20101107_AABYYJ currie_t_Page_54.pro
b9801aed53b5770ac5a1a304fe9ddd9d
dac4299bc928528e72428d9eb206bdfe49b7de5f
F20101107_AABZAJ currie_t_Page_23.jp2
cfcc9a16e66eefcad170b1c351a391c9
7afd9fe9c66e22e11baf7b13c77919c43ad4969e