CELLULAR AND MOLECULAR MECHANISMS OF IGF-1 HYPERSENSITIVITY IN

POLYCYTHEMLA VERA

AMER MUSHTAQ MIRZA

A thesis nibrnitted in conformity with the requirements for the degree of Doctor of Philosophy

Graduate Department of Anatomy & Ce1 Biology University of Toronto

O Copyright by Amer Mushtaq Mirza 1997

National Library Bibliothèque nationale du Canada

Acquisitions and Acquisitions et Bibliographic Services services bibliographiques

395 Wellington Street 395. rue Wellington OttawaON K1AON4 OttawaON K1AON4 Canada Canada

Your Me Vofm retarenw

Ovr &J Notre reference

The author has granted a non- exclusive licence allowing the National Library of Canada to reproduce, loan, distribute or sell copies of this thesis in microform, paper or electronic formats.

The author retains ownership of the copyright in this thesis. Neither the thesis nor substantial extracts fiom it may be printed or otherwise reproduced without the author's permission.

L'auteur a accordé une licence non exclusive permettant à la Bibliothèque nationale du Canada de reproduire, prêter, distribuer ou vendre des copies de cette thèse sous la forme de microfiche/nlm, de reproduction sur papier ou sur format électronique.

L'auteur conserve la propriété du droit d'auteur qui protège cette thèse. Ni la thése ni des extraits substantiels de celle-ci ne doivent être imprimés ou autrement reproduits sans son autorisation,

CELLULAR AND MOLECULAR MECHANLSMS

OF IGF-I E l Y P E R S E N S m IN

POLYCYTHEMIA VERA

by

Amer Mushtaq Mirza

Doaor of Philosophy, 1997

Department of Anatomy & Ceii Biology

University of Toronto

We had previously shown that erythroid progenitor cells in PV are hypersensitive to IGF-I

and that this effect occurs through the IGF-I receptor. It was my goal to shed light on the

cellular and molecular mechanisms for this hypersensitivity. We irnmunoprecipitated the

IGF-1 receptor P subunit from penpheral blood mononuclear cells (PBMNC) of PV

patients and nomals. in PV cells, in the absence of added ligand, basal tyrosine

phosphorylation of the P subunit was increased. In the presence of ligand, the receptor

became phosphorylated earlier, to a greater degree, and at lower ligand concentrations.

Then we determined the circulating levels of IGF-1 and two of its binding proteins,

IGFBP-3 and IGFBP- 1, in PV patients and nomais. Plasma IGFBP- 1 in PV was elevated

4-fold. IGFBP- 1 greatly increased the sensitivity of erythroid progenitor cells to IGF-I in

culture as shown by a lefi shifi in its dose-response curve; this enabled us to mimic one

feature of the PV phenotype with normal cells. In PV, we found these cells to be

hypersensitive to the combination of IGF-1 and IGFBP-1. In both PV and normal,

maximai erythroid burst formation occurred when the two were given in approximately

equimolar concentrations. We next investigated receptor intnnsic events that could be

12s responsible for IGF-I hypersensitivity. 1-IGF-1 binding studies performed on

erythrocytes, glycophonn A- erythroid precursors, and PBMNC reveded that the number

of binding sites and their affinity for ligand were not different in PV versus normal. We

then exarnined the intnnsic tyrosine kinase activity (TKA) of the isolated IGF-I receptor.

In the presence of sodium vanadate, an inhibitor of phosphatase activity, the TKA in PV

was substantially increased. In conclusion, the IGF-1 receptor in PV is

hyperphosphorylated, hyperactive, and hypersensitive to ligand. This may be due to: i)

extracellular events that activate the receptor; ii) a receptor intnnsic defect; or, iii) an

intracellular defect in elements involved in receptor desensitization. Therefore, a defect in

either the IGF-1 receptor or its signalling pathway as well as an increase in the circulating

levels of a stimulator of erythropoiesis, IGFBP- 1, may play a central role in the

pathogenesis of PV.

ACKNOWLEDGEMENT

Fint and foremost, 1 would iike to thank my supervisor and mentor, Dr. Arthur

Axelrad. He, more than anyone else, has influenced the way 1 see and do science. 1 would also

k e to thank the members of my supervisory cornmittee, Dr. Alan Bernstein, Dr. Vic Kalnins,

and Dr. Tony Pawson. nieir advise has been excellent, their passion for science infectious, and

both have played a key role in shaping the current work.

1 am indebted to Dr. Dom Arnato for refemng his patients to this ~ ~ d y , and the

patients themselves for volunteering. 1 would also like to thank Dr. Shreen Ezzat, Our

collaborator on the IGFBP- 1 project, and Dr. Bernie Fernandes for his time and support. It

has been a tremendous experience working alongside these gentlemen.

1 am grateful to Denise Eskinazi, Anne Spalding, Val MacDonald, and Angie Banif for

their techanical assistance. 1 am entirely grateful to Maryanne Patterson, Jan Dath, and ail of

the others nurses at the .Ambulatory Hernatology/Oncology Clinic at Mount Sinai Hospital for

al1 their help.

1 would like to recognize Dr. Mike Wiley for his guidance and unwavering support

over these many years. In addition, 1 wodd f ie to thank Joan MacKenzie, Fiona Srnilie, and

Linda Houston would deflect the slings and arrows of outrageous fortune, only on occasion

to, when 1 deserved it, to hit me with the big stick. Also, 1 thank al of my fnends over these

many years who helped keep me sane by forcing me to get out of the lab eveiy now and again.

Then again, this could be the reason it has taken me these many years to finish!

Findy, 1 would k e to recognize the invaluable contribution of my parents. Their love

and devotion has sustained me, and given me strength. Without their unconditional suppon,

none of this would have been possible.

TABLE OF CONTENTS

A bstrac t

.4cknowledgement

Table of Contents

List of Figures

List of Tables

CHAPTER 1: General Introduction

Po&cythemza vera: The Disease

History

Diagnosis

Clinico-pathologicai Evolution

Polycythemia vera: The Search for Mechanisms

Chromosomd Abnormalities

Stem Ce11 Defect

Erythropoietin Independence

Response to innilin-like Growth Factor I

Erythropoietin Receptor Defects

Current Understanding

Course of Investigation

Page *. II

iv

v

viii

X

CaAPTER 2: Increased basal and induced tyrosine phosphorylation 19 of the IGF-1 receptor subunit in circulating rnononuclear cells of patients with Polycythemia vera

Abstract 2 1

Introduction 22

Methods 24

Results 27

Discussion 30

CWPTER 3: Insulin-like growth factor binding protein-l (IGFBP-1) is elevated in patients with Polycythemia vera and stimulates erythroid burst formation in vitro

Abstract

Introduction

Methods

Results

Discussion

CHAPTER 4: A role for insulin-like growth factor binding protein-l in erythropoiesis:Stimulatory effects on circulating erythroid progenitor cells frorn normal individuals and patients with Polycythemia vera

Abstract

Introduction

Methods

Results

Discussion

CHAPTER 5: General Discussion 92

Polycythemza vera: Focusirig the Search for a De fect 93

Polycyhemia vera: Response of Erythroid Progenitor 97 Cells ro Cy tohes

Polycythemia vera: The Possibiliv that Phosphatases 1 0 1 Play a Role

APPENDIX A: Increased tyrosine kinase activity of the insulin-like 105 growth factor receptor in Polycythemia vera

APPENDIX B: Kinetics of red ce11 development in Polycythemia vera: 119 Studies in a strictly serum-free liquid culture system

Abstract

Introducti~ri

Methods

Results

Conclusion

REFERENCES:

List of Figures

Comparison of tyrosine phosphorylation in normal and PV cells at high IGF-1 concentration for similar amounts of IGF-I receptor

Comparison of tyrosine phosphorylation of the 95-kD band fiom PV and normal

Cornparison of tyrosine phosphorylation of the 95-kD band fiom PBMNC of patients with erythrocytosis and normal individuals

Circulating levels of IGF-1, two key binding proteins, and insulin as determined by radioimmunoassay for patients with PV and normal individuais

Plasma levels of IGF-1, IGFBP-3, IGFBP-1 and insulin in normal individuals and patients with PV, pooled data fiom two expenments

Radioirnmunoassay data for the levels of circulating IGF-1 and IGFBP- 1 for normal individuals, patients with secondary erythrocytosis, and patients with PV

Ligand blot of circulating IGF binding protein 1 in nomal individuals and patients with PV.

Plot of the number of erythroid burst component colonies as a percent of maximum versus the molar concentration of IGFBP-1 in the presence of 3x 10 " ' M IGF-1

Action of IGF-1 alone compared to IGFBP- 1 alone on erythroid progenitor cells fiom normal individuais and patients with PV

Titration of IGFBP- 1 from 3x l 0 - l ' ~ to 3x l O-'M in the presence of 3x 1 0-"M IGF-1

Titration of IGF-1 activity from 3x 1 O"'M to 3x 10% with 3x 1 0 - l ~ ~

IGFBP- 1

Plot of the number of erythroid burst component colonies as a percent maximum versus the molar concentration of IGF-1 with and without IGFBP- 1

Figure

Scatchard analysis of binding data of 1 2 ' ~ - ~ ~ ~ - ~ on erythrocytes isolated from normal individuais and patients with PV

Comparison of the affinities of IGF-1 receptors found on normal and PV glycophorin A positive erythroid precursor cells

Data from l z ~ - ~ ~ ~ - ~ binding studies on peripheral blood mononuclear cells frorn 4 normai individuals and 4 patients with PV

Measurement of the intnnsic tyrosine kinase activity of the IGF-1 receptor in normal individuais and patients with PV

Intrinsic tyrosine kinase activity of the normal and PV IGF-1 receptor

Examination of the growth characteristic of various ce11 populations cultured in serum-containing and serum-free liquid culture media

Comparison of serum-containing and strictly serum-fiee liquid culture media with respect to the growth of glycophorin A positive cells

Growth characteristics of Glycophorin A positive cells in normal and PV cultures with 3 U/mL erythropoietin.

Cytospin preparations of PV and normal cells at Day O, Day 7, and Day 12 in culture in the presence of 3UIm.L erythropoietin

Normal and PV erythropoiesis in culture with 3x1 O*"M IGF-1

Pane

1 O9

1 2 1

113

115

117

130

133

135

137

139

List of Tables

Table - 1 PV Study Group Guidelines for diagnosis

2 Densitornetric analysis of the 95-kDa band in normal and PV

3 Burst component colonies produced in vitro by nomai and PV erythroid progenitor cells with increasing arnounts of IGFBP- 1 in the presence of IGF-1

CHAPTER 1

CHAPTER 1

GENERAL INTRODUCTION

Polycythernia vera (PV) is a neoplastic disorder of unknown etiology. Here, 1 shall

discuss the disease itself, reflecting bnefly upon its history, diagnosis, and clinical as well

as pathological evolution. Then, I shall examine avenues which have been explored in an

attempt to shed light on mechanisms in the pathogenesis of this disorder. Finally, 1 shall

outline our understanding of the PV phenotype as 1 began my research, and the course

that was taken to answer some important questions regarding the pathogenesis of this

intriguing hematological disorder in hurnans.

POLYCYTHEMIA VERA: THE DISEASE

HISTORY

Polycythemia vera (PV) was first descnbed by a French physician Vaquez in 1892

as a special form of cyanosis accompanied by a persistent and excessive polycythemia'.

By 1903 Osler had further defined PV, rnaking the distinction between true and relative

polycythernia2, and shortly thereafter, PV was outlined as a hematological disorder in one

of the major medical texts of the day? The condition was reported to be charactenzed by

a chronic cyanosis and polycythemia, in conjunction with an enlargement of the spleen2. It

was judged most likely to be a disease of the bone marrowJ which was at times

accompanied by an elevation in the number of circulating white cells5. These features still

lie at the heart of a clinical diagnosis for this disorder.

In 197 1, the Polycythemia Vera Study Group (Table 1) established stnngent

criteria for the diagnosis of PV which would identiq most individuals with the disorder

and would have a low false-positive rate6. But it is believed that the Study Group

guidelines may exclude some patients who have early disease7. In the latter case, other

criteria such as serum erythropoietin levels, erythropoietin-independent colonies, or bone

marrow morphology, as described below, may aid in the diagnosis8. The subsequent work

on positively diagnosed patients which followed the establishment of the guidelines has

helped to denve a picture of the PV phenotype.

CLINICO-PATHOLOGICAL EVOLUTION

Aithough the clinical consequences of PV are related primarily to erythroid

hyperplasia and the resultant increase in the red ce11 rnass, proliferative activity in the bone

marrow usually affects the granulocytic and megakaryocytic lineages as well. Marrow

fibrosis may also be observed; however, this is thought to be a secondary response to

neoplastic elements which are released by the rapidly proliferating c e ~ l s ~ ~ ' ~ . The disorder

itself evolves through several stages: the erythrocytotic or proliferative stage, the stable

stage, and the end stage which is marked by a spent phase or in some patients acute

leukemia' '.

The proliferative stage of PV, where the red blood ce11 volume and marrow

cellularity rapidly increase, and stable stages of PV, where blood ceIl counts appear to

plateau, are marked by an increased hemoglobin and hematocnt accompanied by a

significant increase in the total volume of circulating red blood cells as compared with

nomal individuals. Yet despite the massive erythropoiesis, red blood ce11 morphology is

usually normal uniess there is iron deficiency due to hemorrhage. If this is the case,

microcytosis and hypochromia may be detected. The total number of circulating white

cells is also increased and neutrophils account for the majority of these cells although

some immature granulocytes may be present. Absolute basophilia is observed in 60% of

al1 patients with neutrophil alkaline phosphatase scores increased in 70% of al1 patients.

Platelet counts are elevated in roughiy 50% of the patient population with counts in excess

of 1 x i o ' ~ /L being not uncornmon. Bone rnarrow biopsy reveais a hypercellular marrow

with erythrocytic, granulocytic, and megakaryocytic hyperp1asia'213, erythrocytic and

rnegakaryocytic proliferation being particularly prominent. Granulopoiesis is

morphologicaily normal with no significant dysplasia. Megakaryocytes in the marrow are

increased in number and appear to be increased in size, with the appearance of some

megakaryocytic colonies. Reticulin fibers may be increased in 33% of ail patients. The

observed splenomegaly is due to sinusoidal and corda1 congestion by red blood cells. At

these stages, extrarnedullary hematopoiesis is at a rninim~rn'~.

The spent phase is marked by the fact that therapeutic intervention is no longer

required to curb the manifestations of the proliferative or stable stages. This phase is often

associated with post-polycythemia myeloid metapiasia where there is a decreasing red

blood ce11 volume, increasing splenornegaly, extramedullary hernatopoiesis, marrow

fibrosis, leukoerythroblastosis, and tex-drop red blood ce11 poikilocytosis'~. There are

reticulin and collagen fibers present in the marrow biopsy, decreased cellularity and dilated

marrow sinusoids. The biopsy may be indistinguishable fiom that of chronic idiopathic

myelofibrosis'5. Post-polycythemia myeloid metaplasia occurs in up to 15% of al1 PV

patients with the incidence increasing to 50% in patients who suMve with PV for over 15

years'6. Only 2% of a11 patients with PV being managed by phlebotomy alone ever enter 7.17-19 the acute leukemia phase in which they develop acute myeloid leukemia . In

contrast, this rate is as high as 15% in patients undergoing treatment with

myelosuppressive agents such as "P (10.3%), alkylating agents (1 3.5% with

chlorambucil), or hydroxyurea (6%)"*? This phase is often preceded by a

myelodysplastic phase23 where there is marked cytopathology and ce11 proliferation that is

nonclonal and not neoplastic.

POLYCYTHEMLA VERA: THE SEARCH FOR MECHANISMS

CHROMOSOMAL ABNORMALITIES

There are a myriad of chromosomal abnormalities known to be associated with

PV'"~', but no one chromosomai defect has yet been linked to the PV phenotype in al1

patients. This would indicate that there is no one lesion which gives rise to the PV

phenotype. However, it is possible that a consistent causative lesion for PV exists but it is

not associated with a gross chrornosomal abnormality and thus may be too subtle to

detect. It has been shown that there is selective involvement of the af5ected Iineages for a

particular chromosomal a b n ~ r m a l i t ~ ~ ~ . In other words, not ail of the affected lineages in

PV will have the same chromosomal abnormality. It is possible that PV may be associated

with a number of lesions. Altematively, the lesions known to exist in PV could arise as

events secondary to the disorder itself A study by Swolin et alz7 indicates that this may

indeed be the case. Upon examination of 64 patients over a period of up to 224 months

following diagnosis, it was noted that initially ody 17% of the patients demonstrated

chromosornal abnormalities. In contrast, abnomai karyotypes were found in up to 80%

of the patients who evolved through to the end stage of PV and developed either myeloid

metaplasia., myelofibrosis, or leukemia. Thus, it appears as if the majority of chromosomal

abnormalities were developed in these patients as the disorder progressed. Another study

arrived at the same conclusion stating that although non-random, the majority of clonal

chromosomal abnormalities were believed to be secondary events in PV patients,

indicating that there is significant chrornosomal instability in the hematopoietic

compartment of these patients. This was supponed by evidence that 3 out of 1 L patients

developed chromosomal abnormalities afier treatment with phlebotomy alone. Those

individuais treated with aikylating agents, '*P, or hydroxyurea demonstrated chromosomal

abnormalities in 8 of 16, 5 of 9, and 4 of 8 patients, r e ~ ~ e c t i v e l ~ ~ ~ . Also, no consistent

relationship could be found between the occurrence of chromosomal malformations and

the development of the various end stage conditionsz7. Thus, chromosomal abnormalities

have been disappointing as etiologicai factors in PV. But this does not nile out genetic

defects entirely. Lesions too small to detect cytologically may still play an important role

in the pathogenesis of PV.

STEM CELL DEFECT

PV is charactenzed by a panhyperplasia of the major myeloid lineages,

erythrocytic, granulocytic, and megakaryocytic2g. This implies that the PV defect is

expressed in a rnultipotential hematopoietic stem ce11 which gives nse to these lineages

The notion that PV arises from a defect in a multipotential stem cell, although intuitively

satisfying, was not conclusively demonstrated until Adamson et al3' used x-chromosome

linked polymorphisms to study the clonality of blood cell populations. Normal women are

rnosaic with respect to the activity of their x-chromosomes. A given population of cells

contains a mixture of cells which will either contain an active matemal x-chromosome or

an active patemal x-chromosome. These authors utilized the x-chromosome-linked alleles

for isoenzymes of glucose-6-phosphate dehydrogenase (G6PD). In women who are

heterozygous for G6PD ( ~ d ~ d ~ ) , a given ce11 will express either the G$ or ~ d 9 phenotype. Thus, neoplasms 60m these individuais contain only one type of G6PD (either

A or B), that is they onginate frorn a single cell, are said to be clonal in ongin, whereas

neoplasms which contain both A and B G6PD types are believed to be of multicellular

origin. Adamson et al demonstrated that in PV red cells, neutrophils, monocytes, and

platelets in the penpherai blood al1 camed the same =PD type and therefore the same

active x-chromosome. This was in contrast to B and T cells in PV which displayed a

mixed pattern of x-inactivation3'. These findings strongly suggested a monoclonal ongin

for the red cell, neutrophil, monocyte, and platelet lineages, whereas the B and T cells

were shown to have a polyclonal origin. This indicated that the lineages affected by the

PV defect al1 anse fiom the sarne cell. This ce11 then gives rise to the PV clone, which

becomes dominant, giving rise to most if not ail of the myeloid cells in the circulation. It

also indicated that this clone is abnorrnal as abnormal quantities of red cells, neutrophils,

platelets and monocytes were produced. The evidence presented by the authors that only

the myeloid lineages were affected indicates that the defective stem ce11 is not the bone

marrow repopulating ce11 but a multipotential myeloid progenitor.

Direct evidence for the involvement of a multipotential myeloid progenitor cell in

PV was reported by Ash et al3'. The multipotential myeloid progenitor ce11 CFU-GEMM

in semi-soiid culture produces mixed colonies of granulocytic, erythrocytic, macrophage,

and rnegakaryocytic cells3*. Ash et al" utiIized these cultures to examine the properties of

this pluripotent hematopoietic stem ce11 Eorn PV patients. They demonstrated that CFU-

GEMM fiom PV patients could be cultured in the absence of added erythropoietin, a

hallmark of the PV phenotype, and had a G6PD type which was consistent with the cell

belonging to the PV clone. This suggested that the PV stem ce11 defect is expressed at the

level of the CFU-GEMM or earlier.

Aithough several authors have shown clearly that PV involves red cells, platelets,

granulocytes, and monocytes/macrophages, others have questioned whether there is

strictly myeloid involvement in this disorder or whether it extends to the lymphoid lineages

as well. Once again using x-chromosome linked G6PD as a marker for cellular mosaicism,

Raskind et al3) investigated the involvement of the B lymphoid lineage in P V Epstein-

Barr virus-transfomed B-ce11 lines were established from a sample of peripheral blood

from a patient with PV. Examination of the resulting transformed B-ce11 lines revealed

that the ratio of type A and type B G6PD was not 1 : 1 as was predicted statistically.

Approximately 40% of the ce11 Iines contained cells which were nonclonal, producing a

mixture of both type A and B G6PD and indicating that the transformed ce11 lines arose

from more than one cell. These latter were excluded from the experiment. Approximately

56% were monoclonal for the M P D type identified as that of the PV clone, i.e. that of the

red cells and granulocytes, while approximately 5% were of the opposite =PD type. The

authors thus concluded that PV stem cells are capable of differentiation to B lymphocytes

as well as myeloid cells. However, the PV clone did not dominate the B cell lineage as it

does the red ce11 lineage. They attributed the lack of complete dominance of B

lymphocytes by the PV clone to the long-lived nature of lymphocytes, suggesting that the

nonclonally denved cells antedated the development of disease. They also claimed that the

normal cells rnay enjoy some selective advantage in vitro. A cntisism of this study is that

it involved a single patient with PV. 1 believe the study, to be convincing, would have to

be extended to include data from several more patients with normal subjects as controls

for lyonization, the random process by which one of the two copies of the x-chromosome

are inactivated. As it stands, the result is intriguing, but 1 feel the question of involvement

of the B lymphoid lineage in PV remains open.

More recently, to determine if T lymphocytes are part of the PV clone, Tsukamoto

et al3" investigated the clonal basis of PV and other myeloproliferative disorders with

restriction fragment length polymorphisrns of x-chromosome linked phosphoglycerate

kinase and hypoxanthine phosphoribosyltransferase genes. Eight patients with PV

revealed monoclonal pattems of x-inactivation in the granulocyte lineage, suggesting a

clonal ongin for the population. However, 1 out of 8 patients with PV demonstrated

apparently monoclonal x-inactivation in the T ce11 lineage. This was interpreted by the

authors to be an example of heterogeneous lineage involvement of T cells in PV,

indicating that at times T cells may originate from the PV clone. This rnay not necessarily

be true. It has previously been shown that constitutive skewing of the x-chromosome

inactivation pattem with >75% expression of one allele may occur in approximately 20-

25% of individual$'. The authors acknowledge this potential problem but appear to have

accounted for it by showing that the pattern of x-inactivation in liver cells is nonclonal

when the hematopoietic cells are monoclonal. However x-chromosome inactivation

pattems are frequently different in different tissues36; consequently, cells of non-

hematopoietic compartments cannot be used as controls for the lyonization pattern of

hematopoietic cells. One way to control for this is to do a population study of PV using

normals as controls. In light of this, 1 believe that T ce11 involvement in PV remains

unproved.

Normal erythropoiesis requires erythropoietin and erythropoiesis fluctuates in

response to erythropoietin levels3'. However in PV. red ce11 production is increased, but

the erythropoietin levels in PV patients are low, and aithough urinary erythopoietin levels

increase in response to phlebotomy, they may not retum to levels seen in normal

individuals3*. Moreover, serum erythropoietin levels in patients with PV are normal39 or

below normalJ0, thus indicating that over-production of erythropoietin, as in cases of

secondary polycythemia 6, is not an underlying cause of the PV defect. This apparent

independence of red ce11 production frorn erythropoietin levels in PV suggested to eariier

investigators that erythropoiesis in PV was autonomous29*41, something which is

characteristic of a neoplasm. There is further evidence to indicate that modulations of

erythropoietin levels have no eEect upon erythropoiesis in PV. Adamson's observation

that urinary erythropoietin levels increased in response to phlebotomy suggests that the

decrease in erythropoietin levels observed in PV may be attnbuted to negative feedback

resulting in the down-regdation of erythropoietin production in the presence of an

increased number of red blood cells'"'. In fact, it is known that PV patients are normal in

their erythropoietic response to hypoxia at altitude and relative hyperoxia after returning

to sea levelJJ. These events would suggest that erythropoiesis is uncoupled fiom

erythropoietin; although, erythropoietin production and response mechanisms are normaf

in this disorder, and thus are not the cause of the increased red cell production.

Stephenson et al demonstrated that normal erythroid progenitor cells fiom the

bone marrow and fetal liver of mice could be induced with erythropoietin to produce

colonies of hemoglobin-synthesizing cells in semi-solid cultureJS. Using a modified version

of this stem^^, Prchal and Axelrad cultured bone marrow cells fiom patients with PV and

were the first to demonstrate a regulatory defect in the response of erythroid progenitor

cells to erythropoietinJ7. When normal human bone marrow cells were plated in the

presence of erythropoietin, they produced colonies of nucleated cells which stained

positively for benzidine, and differentiated into non-nucleated red cells within six to eight

days. These erythropoietic colonies had an obligate requirement for erythropoietin as no

colonies formed in its absence. In contrast, the bone marrow cells of patients with PV

gave rise to erythropoietic colonies in culture when no erythropoietin was added to the

medium. In other words, the PV marrow contained erythropoietin-independent erythroid

progenitor cells which produced colonies without added erythropoietin. These colonies

were described by the authors as "endogenous". The authors concluded that these

endogenous erythroid colonies arose because the progenitor cells i) were either truly

independent of erythropoietin, ii) had dready been infiuenced by erythropoietin in vivo, or

iii) were unusually sensitive to minute amounts of erythropoietin present in these serum-

containing cultures.

Eaves et ala8 removed young developing erythroid colonies from culture and

replated them into secondary cultures, with and without added erythropoietin. They found

that endogenous colonies could be produced from the replated cells, thus indicating that

the ery-thropoietin-independence was inherent to the cells and not the result of the

progenitor cells dready having been exposed to erythropoietin in vivo. Consequently,

their work essentially rules out the second hypothesis as a cause for the endogenous

colonies. However, the problem of whether the erythroid progenitor cells in PV were

tmly independent of erythropoietin, or simply hypersensitive to it was not resolved until

much later.

The observed increase in red ce11 production in the presence of low erythropoietin

levels in PV patients and the formation of endogenous erythroid colonies without

exogenously added erythropoietin, suggested to sorne that PV progenitor cells might be

independent of hormonal regu~ation'~. However, this statement cannot be generalized to

include ail erythroid progenitors in PV as it would ignore the repeated observation that in

PV, there are progenitor cells which are clearly responsive to erythropoietin50'53. Others

suggested that erythroid progenitor cells in PV were hypersensitive to minute quanitities

of erythropoietin known to be present in senim containing culture^'^. Zanjani et al5'

believed that as their cultures contained serum, and semm was known to contain

erythropoietin, then the progenitors in PV could be responding to the minute amounts of

erythropoietin present. They used polyclonal anti-erythropoietin antibodies raised to

semm purified erythropoietin and reported few endogenous erythroid colonies in cultures

so depleted. When serum-purified erythropoietin was added to culture, for a given

amount of erythropoietin, there were more colonies in PV than in normal cultures. Thus.

the conclusion that PV progenitors were "hypersensitive" to erythropoietin was based, not

on a dose-response curve, but on an increase in the number of colonies. However, it had

aiready been demonstrated that the plating efficiency of erythroid progenitors in PV was

higher than normal5z5J. This observation was later extended to include multipotential

myeloid progenitors CFU-GEMM and CFU-GM as well as early erythroid progenitors

BFU-E". In other words, there were simply more erythroid progenitors present in PV

than in normal.

Hypersensitivity is defined as an increased ability to respond to a stimulus, and a

ce11 is said to be hypersensitive when it displays an increased response to a stimulus

compared to another cell. Thus. hypersensitivity is a cornparison of relative sensitivities.

A proper measure of relative sensitivities must be made using normalized data which takes

into account the number of cells present. For example, we measured the sensitivity of

cells of type A and B to a particular ligand by calculating a separate dose-response curve

for the ligand or stimulus being tested for each ce11 type A and B, where the response was

plotted as a percent maximum response for that particular cell. This normalized the data

within a particular population of cells. Then, a calculation of the respective half-maxîmum

value of the dose-response curve of ce11 type A and B could be used to estimate the

sensitivity of each ce11 type to that particular ligand. Consequently, a cornparison of these

half-maximum values ailows comparisons of sensitivities. A significant lefi-shifi in the

dose-response curve thus plotted is indicative of hypersensitivity. Unfortunately, Zanjani

et alJ0 did not follow the above criteria. Their calculations were based on response to

ligand without normalization as a percent maximum. This does not allow for variability in

the number of responsive cells within a particular population. It erroneously assumes that

al1 ce11 populations have the sarne number of responsive cells. Therefore, two ce11

populations that have equivalent sensitivities to ligand, but where one population may

have a greater number of cells whichare responsive to ligand, will appear to be

"hypersensitive" .

Golde et al5* were the first to test the sensitivities of erythroid progenitor cells in

PV and normal individuals with dose-response curves. They used serum containing

cultures and found that the response of PV progenitor cells to erythropoietin was not

different fiom that of normal progenitor cells. Later, Casadevall et al5' using "serum-free"

cultures reported that erythroid progenitor cells in PV had approximately I O-fold

increased sensitivity to erythropoietin. However, they observed an increased sensitivity in

only 2 out of 4 patients tested. Also, the conditions under which the experiments were

carried out were not truly serurn-fiee as the preparation of erythropoietin used was not

recombinant erythropoietin, but serum-purified, and thus Iikely contained undefined

"bioactivities". Therefore, even when dose-response curves were used, serum containing

and incornpletely defined "serum-fiee" media have given equivocal answers to the

question of erythropoietin hypersensitivity in PV.

Al1 of the cultures used to assay erythroid colonies thus far have contained serum

and serum-cornponent proteins. Both include minute quantities of undefined activities

which may stimulate as well as inhibit erythropoietic activity. Also, some cultures

contained leukocyte conditioned medium (LCM) to enhance erythroid colony growth. In

fact, it was known early on that the erythropoietin preparations, fetal calf serum, and

bovine semm aibumin (BSA) al1 contained "LCM-like" factors5%r burst promoting

activity (BPA). BSA has been shown to increase the number of erythroid progenitor cells

and reduce their requirements for semm and erythropoietin in cultures6. This could be

attributed to the BSA itself or factors such as Iipids contaminating the BSA". Therefore,

there is the possibility that some progenitors may have been responding to these other

factors and that the pecuiiar "sensitivity" ascribed to some erythroid progenitor cells in PV

may be attributed to the stimulatory activity of factors distinct fiom erythropoietin.

Consequently, these questions of erythropoietin hypersensitivity or independence could

not properly be addressed until the advent of truly senim-fiee culture conditions.

Correa and ~xelrad" desciibed an improved semm-free medium, with

recombinant cytokines providing defined BPA-like activities, used for the culture of

human erythroid progenitor cells With this improved serum-fiee medium, they were able

to demonstrate that insulin-like growth factor I (IGF-1) could entirely replace

erythropoietin in culture. However, to do so required 100-fold higher molar

concentrations of IGF-I than erythropoietin itself Using this improved serum-free

medium, Correa et al5' were able to investigate the question of erythropoietin

hypersensitivity versus erythropoietin independence as a cause of the endogenous colonies

observed in PV. They found the erythropoietin dose-response curve of erythroid

progenitor celis from PV patients to be statistically not different Corn that of normal

individuals, thus confirming the results of Golde et als2 in a "clean" system. In addition, in

the absence of erythropoietin. they showed that the erythroid progenitor cells in PV were

hypersensitive to IGF-1 by virtue of a dramatic lefi shifi in the IGF-I dose response curve.

Thus, their work provided compelling evidence that the erythroid progenitor cells in PV

are independent of erythropoietin and hypersensitive to IGF-1.

RESPONSE TO LVSULIN-LIKE GROWTH FACTOR I

The fact that the IGF-1 receptor has been identified on erythrocytes60"1 as well as

on peripheral blood mononuclear c e ~ l s ~ ~ suggests that it may play a rote in

e ry th r~~oies i s~~ . IGF-I is in fact known to stimuiate the development of erythroid

progenitor cells in culture6448 and can support erythropoiesis in the presence69*70 or

absence of erythropoietin 58*70. In addition, serum IGF-1 but not serum erythropoietin

levels correlate with erythropoiesis during accelerated growth in rats7'. An -8kDa peptide

isolated ffom an anephric patient whose hemoglobin level was not drarnatically reduced,

showed that this peptide was capable of regulating erythropoiesisn, and in some patients

with anernia and rend failure, it has been suggested that IGF-1 c m replace erythropoietin

as a stimulator of erythropoiesisn. Also, in hemodialysis patients with severe secondary

hyperthyroidism, semm IGF-1 levels, but not erythropoietin levels, were correlated with

hematocrit, indicating that IGF-I could be an important regulator of erythropoiesis in these

patients74. In light of evidence for the signifiant role of IGF-1 in erythropoiesis, the

finding of Correa et al5' that PV erythroid progenitor cells are hypersensitive to IGF-1

becorne even more important. Such a defect could theoretically play a pivotal role in the

pathogenesis of PV.

ERYTHROPOIETIN RECEPTOR DEFECTS

A priori, as erythropoietin plays a key role in normal adult erythropoiesis, it is

theoretically possible that an abnormality in the erythropoietin signalling pathway or the

erythropoietin receptor itself may be a cause of the PV phenotype. It is known that

mutations of the erythropoietin receptor itself5 cm lead to erythropoietin-independent

activation of the receptor. Longmore et al" detennined the effect of this receptor-

activating mutation upon erythropoiesis and hematopoiesis by infecting hematopoietic

progenitor cells and mice with the mutant receptor. They were able to demonstrate that

infected erythroid progenitor cells were factor independent and developed erythropoietin-

independent colonies in culture. Infected mice were polycythemic and had splenomegaly,

two hallmarks of PV. Thus, this interesting paper clearly illustrates that a lesion in the

erythropoietin receptor can lead to polycythemia and raises the question as to whether

such erythropoietin receptor defect could be the cause of PV in humans. Although it is

known that truncation of the receptor in the cytoplasrnic domain produces an exaggerated

response to erythropoietin and is associated with familial p ~ l ~ c ~ t h e m i a ~ ~ ~ ~ ~ , and theories of

erythropoietin receptor mutations in PV are enticing, the simple fact of the matter is that

these mutations are exceedingly rare in erythroid malignancies79 and have not been found

in pv80.

CURRENT UNDERSTANDING

From a synthesis of these data, a picture for PV begins to emerge. PV is not

caused by an increase in erythropoietin production, as circulating levels of erythropoietin

are normal39 or below normalJ0. but is a result of the proliferation of an abnormal stem ceIl 47.8 1.82 c ~ o n e j ~ . ~ ' . The abnormd PV clone is independent of erythropoietin , and we know

that there are at least two distinct phenotypes for erythroid progenitor cells in PV. The

first is essentially normal in that these cells demonstrate normal responsiveness to

erythropoietin, and probably to other growth factors. The second phenotype is one where

the progenitor cells can be independent of erythropoietin and hypersensitive to IGF-1. It is

clear that, in the bone marrow, the PV clone is far From dominant, representing only 6-

37% of al1 erythroid progenitor c e ~ l s " ~ . ~ ~ , but in the circulation it is ovenvhelrningly

dominant, representing almost 100% of al1 myeloid end cells. Thus, it appears that in the

bone marrow, the cradle of hematopoiesis, the early PV clone CO-exists with the normal

early progenitors, neither of which is dependent upon erythropoietin for its s u ~ v a l ~ ' .

But, once these cells differentiate to becorne dependent upon erythropoietin, in PV, where

serum erythropoietin levels are low, the PV clone, being erythropoietin-independent and

hypersensitive to IGF-1, has a selective advantage over the normal erythropoietin-

dependent clone. The fact that there is a higher frequency of clonally derived CFU-E than

BFU-E in G6PD heterozygous individuals with PV supports this notiong". In this way,

cells of the PV clone proliferate unchecked by fdling erythropoietin levels, unlike their

normal counterparts. As the red ce11 mass increases, the erythropoietin levels further

decrease in responseJ3, thus giving the PV clone a hrther advantage, while causing the

normal progenitors in the presence of low erythropoietin levels to quiesce or apoptoseg5.

Thus begins a dangerous downward spiral into hematopoietic dysregulation.

COURSE OF tNVESTIGATION

Recent work in our laboratory had shown that erythroid progenitors cells in PV

are hypersensitive to IGF-I and it was demonstrated that this effect occurred through the

IGF-1 r e ~ e ~ t o ? ~ . It was my goal to shed light on the rnolecular mechanism for this

observed hypersensitivity. The IGF-1 receptor is a member of the family of receptor

tyrosine kinases and tyrosine phosphorylation of the IGF-1 receptor B subunit activates the

receptor and plays a criticd role in signal transductiong6.

In chapter 2, we exarnined tyrosine phosphorylation of the IGF-1 receptor

subunit irnmunoprecipitated fiom penpheral blood mononuclear cells of PV patients and

normal individuals. In the absence of added ligand, we found that there was increased

basal tyrosine phosphorylation of the P subunit in PV cells. In the presence of added

ligand, we also found that the receptor was hypersensitive and hyperresponsive with

respect to tyrosine phosphorylation. As tyrosine phosphorylation plays a key role in

receptor activation and signal transduction, we wished to determine whether this increase

was due to extnnsic versus intrinsic regdation of IGF-1 receptor.

In chapter 3, we examined the levels of IGF-I, and of two insulin-like growth

factor binding proteins, IGFBP-3 and IGFBP- 1. in PV patients and normal controls. We

showed that the levels of plasma IGFBP- 1 in patients with PV are elevated. We also

demonstrated for the first tirne that IGFBP- 1 is a powerful stimulator of erythropoiesis in

vitro. When titrating the effect of IGF-I with IGFBP- 1 on erythroid progenitor cells in

PV, we found that these cells were hypersensitive to a putative IGF-IAGFBP-1 cornplex.

Consequently, elevation of IGFBP- 1 in the circulation of PV patients could underlie the

massive overproduction of red cells in this disorder.

In chapter 4, we exarnined the effect of IGF-1 with IGFBP-1 on erythroid

progenitor cells isolated from PV and normal individuals. We confirmed Our earlier

finding that the IGFBP-1 with IGF-I is a powerfùl stimulator of erythropoiesis and that

progenitor cells in PV are hypersensitive to this activity. We also demonstrated that

IGFBP-1 is able to greatly increase the sensitivity of erythroid progenitor cells to IGF-1.

In this way we were able to mirnic the PV phenotype with normal cells. These

observations provide evidence for extracellular events which rnay rnodulate IGF-1 receptor

activity. The fact that there is observed hypersensitivity to the IGF-MGFBP- 1 "cornplex"

in PV indicates that other events may also underlie the PV defect.

Appendix A addresses our investigation of receptor intnnsic events which rnay

regulate IGF-1 receptor activity. We examined the number of IGF-1 receptors and their

affinity for ligand on red blood cells, nucleated erythroid precursor cells, and peripheral

blood mononuclear cells from PV patients and normal individuais. We determined that the

number of binding sites and their afftnity for IGF-1 was not different in PV compared to

normal. We also examined the intrinsic tyrosine kinase activitiy of the IGF-I receptor

itself over a broad range of ligand concentrations. In the presence of sodium vanadate, an

inhibitor of phosphatase activity, we found the tyrosine kinase acitivity in PV to be

substantiaily increased, thus correlating the increased tyrosine phosphorylation observed in

Chapter 2 to a biologically relevant event, i.e. an increase in tyrosine kinase activity.

Appendix B describes a novel strictly serum-free liquid culture system for the

production of glycophonn A positive erythroid precursor cells from glycophonn A

negative erythroid progenitor cells in vitro. These precursors were then used for Our

binding studies in Appendix A. This system has also been used to study the kinetics of red

ce11 development in PV and normal in mass culture. The system was also shown to be

applicable to the investigation of proliferation and differentiation of cells in other

hematopoietic lineages.

Chapter 5 summarizes Our findings and discusses how they may contibute to Our

understanding of the pathogenesis of this important hematopoietic disorder.

Table - 1

B2 - Leukocytosis Neutophils > t .2xt o'/@

B3 - hcreased neutrophil alkaline phosphatase score > 100

B4 - hcreased serum B12 > 900 pg/mL

Polycythemia vera Study Group Criteria for the Diagnosis of this Myeloproliferative

Disorder. A positive diagnosis of Polycythemia vera (PV) is based upon three major

(Category A) and four minor (Category B) criteria. An individual is considered to have

PV if al1 three of the major criteria are met, or if two of the major critena are met in

combination with any two of the minor criteria. Note that leukocytosis (B2) and increased

neutrophil alkaline phosphatase (B3) must occur in the absence of fever or infection.

CHAPTER 2

CHAPTER 2

Increased basai and induced tyrosine phosphorylation of the IGF-1 receptor P subunit in

circulating mononuclear cells o f patients with Polycythemia vera1

Amer M. Mirza, Paulo N. Correa, and Arthur A. Axelrad

Dept. of Anatomy and Ce11 Biology

University of Toronto, Toronto, Ontario

Canada

' This work \vas ~ p p ~ n e d by a grant from the Medical Research Council of Canada (MA3969).

Portions of this work were presented at the American Society of Hematology meeting, Nashviile. TN. December 1994 (Blood 84 (10) suppl 1. p. 5 17a).

A manuscript derived from ths chapter has been published in Blood 86(3): 877-882, 1995.

ABSTRACT

We have previously shown that circulating progenitor ceiis in patients with

Polycythemia vera (PV) are hypersensitive to insulin-like growth factor 1 (IGF-1) with respect

to eiythroid burst formation in sem-6ee medium, and that this effect occurs through the IGF-

1 receptor. ln normals, no bursts developed at IGF-I concentrations below IO-^ M, whiie in PV

burst formation occurred at IGF-I concentrations as low as IO-'' M. To investigate the

molecular bais of this IGF-I hypersensitivity phenornenon, we examined tyrosine

phosphorylation of the IGF-I receptor P subunit in peripheral blood mononuclear ceils

(PBMNC) fiom 8 PV patients and 6 normals. Cells were exposed to IGF-I at concentrations

of 1 0 ~ and 10'1° M for 0, i , 3, and 10 min, and then lysed. Proteins were resolved by SDS-

PAGE and Western blotted with anti-phosphotyrosine antibody. The identity of the tyrosine

phosphorylated IGF-1 receptor P subunit was confirmed with a polyclonal anti-IGF-1 receptor

antibody. We found that the level of basal tyrosine phosphorylation of the IGF-I receptor P subunit in the absence of exogenous IGF-1 was substantially greater in PV than in normal, for

the same amount of receptor protein. In the presence of 10*M IGF-1, in both nomal and PV,

the level of tyrosine phosphorylation was increased. At 10- 'O M IGF-1 in normais, no evidence

of increased tyrosine phosphorylation was detected; however in PV, a pronounced increase in

tyrosine phosphorylation was observed at both 1 0 ~ and IO-'' M IGF-1. These data were

confirmed by irnrnunoprecipitation of the IGF-1 receptor followed by Western blotting with

anti-phosphotyrosine antibody. In contrast, in PBMNC fiom 3 patients with erythrocytosis, no

signifïcant increase above normal was seen in either basal or induced tyrosine phosphorylation

of the IGF-1 receptor P subunit. Our findings thus reveal two distinctive features of the PV

phenotype in PBMNC: i) in the absence of ligand, an increased basal tyrosine phosphorylation

of the IGF-I receptor P subunit, and ii) in the presence of ligand, a hypersensitive and

hyperresponsive receptor with respect to tyrosine phosphorylation. These features rnay

influence the ability of the receptor to transmit a prouerative signal; they may thus play a role

in the pathogenesis of PV.

INTRODUCTION

Polycythemia vera (PV) is a chronic myeloproliferative disorder of as yet undetermined

etiology characterized by a hyperplasia of all three major myeloid lineages. Due to the clonai

nature of the disease, it is believed that dysregulation of a pluripotential stem cell leads to

increased proliferation and expansion of the affécted rnyeloid cornpartments, but with a

particdar emphaçis on the erythroid lineage3'? PV is distinguished li-om secondary

polycythernia in that the defea is intrinsic to the ceils6, and the relentless overproduction of red

blood ceils in PV occurs in the presence of normal O2 saturation and with levels ofserum

erythropoietin (Epo), the key hormone of nomal adult erythr~~oiesis", often depresseda-88-".

Recently, work in Our laboratory using an irnproved semm-free mediums8 has demonstrated

that circulating erythroid progenitors @FU-E) are not hypersensitive to Epo as previously

believeds'", but exhibit a more than 100-fold increased sensitivity to insulin-like growth factor

1 (IGF-1)". Furthermore, using an anti-IGF-1 receptor antibody', we were able to abolish the

IGF-I hypersensitivity, indicating that the effkt was mediated via the IGF-1 receptor.

The receptor for IGF-I is a member of the tyrosine kinase receptor famiY*92w and is

highly homologous to the insulin r e ~ e ~ t o ? ~ ~ . The IGF-I receptor is encoded by an &A

which is translated to yield a 180kDa receptor precurso? that is glycosylated, dimerized, and

proteolyticaily processed to yield the mature a 2 4 2 heterotetramer consisting of two

extracellular ligand binding a chains disulphide linked to two B chains that span the membrane

once and contain the intrinsic tyrosine kinase a c t i v i e . Extracellular ligand binding to the

receptor stimulates its intrinsic tyrosine specific protein kinase activity, which leads to B subunit 98-10 1 autophosphorylation . Autophosphoryiation of the cytoplasmic domain of the subunit

102-105 leads to a drarnatic increase in its kinase activity and its subsequent phosphorylation of 106.107 intracellular substrates . It is thus believed to have an essentiai role in signal

108-1 10 transduction .

Receptor tyrosine kinases are considered to be important in mediating the effects of

extracellular growth factors on ceii proliferation and ~erentiation*. In fact, an abnorrnai

increase in the activity of receptor tyrosine kinases has been associated with unregulated

gowthl ' l . Studies indicate that an increased tyrosine kinase activity - anributed either to an 112-1 14 increase in the nurnber of tyrosine kinase receptors or their constitutive activation"' -

elicits uicreased ~ignal l in~~ '" '~~. Since we had previously observed hypersensitivity to IGF-I in

PV, and as an increased tyrosine kinase activity had ben shown to be associated with

unregulated proliferation, we compared the kinase aaivities of the IGF-1 receptor in cells from

patients with PV with those from normal individuals. Here, we report that in PV tyrosine

phosphorylation of the IGF-1 receptor B subunit is increased in the absence of added ligand and

its response to ligand is substantially enhanced; we suggest that these changes may provide a

molecular bais for the hypersensitivity to IGF-1 in this myeloproliferative disorder.

METHODS

Patients

Positive selection of PV patients foiiowed the standard guidelines established by the PV

Study (3roup6: The three major criteria were increased total RBC rnass (>35rnL/kg) and

splenomegaiy in the presence of normal O? saturation (>92%). In the absence of one of the

major criteria, a diagnosis of PV was made ifthere was a combination of two of the foilowing

four minor criteria: i)increased platelet count (>400~10~/L); ü) increased white blood ceil count

(> 12x 1 09/L); i) increased neutrophil aikaline phosphatase score (> 120); and iv) increased

s e m BI^ (>700 pmoVL). AU patients were being managed by phlebotomy at the time of

study. Selection of patients with relative erythrocytosis was based on their being negative with

respect to the criteria for PV, but having eievated hematocrits associatecf with decreased

plasma volumes.

Cell Preparations

M e r informed consent, peripheral blood was obtained by venipuncture from healthy

volunteer donors and patients, and was irnrnediately placed in a polypropylene tube containing

1 O U h L preservative-fiee sodium heparin (#820 5077MF;Gibco). The heparinized blood was

layered ont0 1 5m.L of Ficoil-Hypaque (Phmacia, Montreal, Canada) and the Lght-density

mononuclear celis (PBMNC) were coiiected after centrifugation at 450xG for 30 min at room

temperature. CeU suspensions were washed three times (450xG, 10 min at room temperature)

in a-minimal essential medium (aMEM) containing 0.1% fatty acid-fiee and giobulin-£tee

bovine semm albumin (FAF-BS A), and the ceiis were counted.

Stimulation of ceils with ligand

M e r three washes in aMEM+O. 1% FAF-BSA (4SOxG, 10 min at room temperature),

PBMNC were resuspended in basai serum-fke mediums8 at a concentration of 5x 106 cells/mL

for 30 min at room temperature. The ceils were then washed in aMEM+û. 1% FAF-BSA and

resuspended at 5x10~ cells/mL in aMEM+O. 1% FAF-BSA without or with IGF-1 at a final

concentration of 10' or 1 ~ ' ~ M. PBMNC were incubated with and without ligand for 0, 1, 3,

IO, or 30 min at 37°C. The cells were then washed in cold aMEM+O. 1% FAF-BSA,

rnicrohged (Mikroliter, Hettich) at 15000rpm for 5 seconds and irnrnediately lysed in sarnple

b a e r ( M M Tris pH 6.8,4% SDS, 2% P-mercaptoethanol, 5% glycerol, 0.1% Bromophenol

blue) for imrnunobloning with anti-phosphotyrosine (UBI) or ad-IGF-1 receptor antibody

(ZIBI), or in lysis bufEer (1% Triton X- LOO, 1% bovine hemoglobin, 0.1% SDS, and 0.5%

sodium deoxycholate) for irnrnunoprecipitation in the presence of protease inhibitors (Stock in

DMSO: 1 h n o l Aprotinin, 85mol Phosphoramidone, 270nmoI TLCK, 565nrnol TPCK,

370nmoi PMSF, 280mol E-64, IOOnmoi Leupeptin, and 30nmol Pepstatin) used at 1/1000

stock concentration and 2mM sodium vanadate.

Immunoprecipitation of IGF-1 receptor

Prepared cells were lysed in lysis buffer, protease inhibitors, and sodium vanadate at a

final concentration of Sx l~~cells/rnL for 1 hr at 4°C. The lysate was then centrifùged twice at

3000xG for 10 min and again at 1 OOOOxG for 30 min. The supernatant was precleared

overnight at 4°C with 100mL of a prepared 1 : 1 sluny of protein G-sepharose (Sigma) plus

dilution buffer (0.1% Triton X- 100,O. 1% bovine hemoglobin in TSA (lOmM Tris-CI, pH8.0,

140mM NaCl, 0.025% N a 3 ) ) and the supernatant was removed. Supematants in 200mL final

volumes were placed in 1.5m.L microfige tubes precoated with dilution bdfer, 3rng of an&

IGF-1 receptor antibody (aIR3) were added, and the contents of the tube were incubated for

1 hr at 4°C on an orbital shaker. Then, 25mL of a prepared sluny of protein G-sepharose was

added to the mixture and the contents were incubated overnight at 4°C on an orbital shaker.

shaker. The complexed sepharose was then washed four times: twice in dilution buffer, once in

TSA and finally in 0.05M Tris-CI, pH 6.8. M e r the final wash, 50mL sarnple buffer (1 0 mM

Tris-HC1 pH 7.2, 1% SDS, O. 1 M P-mercaptoethanol (P-ME), 1 0% glycerol, and 0.0 1 %

Bromophenol blue) was added to the complexed sepharose and the tubes were incubated at

100°C for Smin.

Immunobloaing with anti-phosphotyrosine or anti-IGF-1 receptor antibody

To analyze tyrosine phosphorylation, ceils prepared in sample buffer were centrifùged

at 5000g for 10 min. The supemantants were subjected to electrophoresis on an 8% SDS-gel.

Proteins were electrophoreticaüy transferred to supported nitrocellulose membrane. The

membrane was blocked with non-specific proteins (Block 1, 5% w/v powdered skirn milis in

PBS). FoUowing primary blocking, the membrane was incubated overnight with fresh Block 1

with pnmary monoclonal antibody (4G 10, UBI) at a dilution of 1 mg/mL, or polyclond anti-

IGF-IR (UBI). Next day, f i e r the appropriate washes (PBS followed by TrisNaCl (50rnM

Tris, 1 50mM NaCl, pH 7.5)) the nitrocellulose membrane was incubated with the secondary

antibody, rabbit anti-mouse horseradish peroxidase (HRP) conjugate (Sigma), or goat anti-

rabbit HRP conjugate (Sigma) in Block II (5% w/v powdered skim rnilk in Tns/NaCI). The

blotted nitrocellulose membrane was treated with enhanced cherniluminescence substrates

(ECL - Amersharn) and the membrane was used to expose x-ray film.

Densitometric Analysis

An LKB Ultrascan XL Laser Densitometer was used to scan imrnunoblots.

Absorbance of the 95kDa band, the IGF-1 receptor B subunit, was quantified as the area under

the peak for the band in arbitrary absorbance units. This absorbance value was then normalized

as a percent of the maximum absorbance for each individual. The means and standard errors of

these percent values were then used to plot cuves.

RESULTS

Tyrosine phosphorylation of the IGF-1 receptor in normal and PV

It has been shown that in cells expressing the IGF-1 receptor, stimulation with

IGF-1 rapidly induces autophosphorylation of the IGF-1 receptor P subunit on specific

tyrosine residues that play an essentid role in the cells' response to IGF-1." Using

PBMNC isolated from normal individuals, we first showed that, in the presence of IO-'

moVL IGF-1, autophosphorylation of the P subunit occurred within 1 minute, with

phosphorylation peaking between 3 and 10 minutes (data not shown), thus indicating that

PBMNC have IGF-1 receptors which respond to IGF-1 in the expected way. Next, we

compared tyrosine phosphorylation of the receptor P subunit in PBMNC obtained from

normal individuals and patients with PV. Figure 1 compares tyrosine phosphorylation of

the IGF-1 receptor P subunit in PV with that of the normal in high concentration of ligand.

It illustrates that at time zero, in the absence of added IGF-1, the amount of tyrosine

phosphorylation in PV cells was substantially greater than in normal. With time, in the

presence of 1 0 ' ~ m o n IGF-1. there was an increase in tyrosine phosphorylation in both

PV and normal cells, but the P subunit was phosphorylated earlier and to a much higher

level in P V In contrast, the amount of IGF-1 receptor f3 subunit was similar in PV and

normal, as shown by immunoprecipitation of the IGF-1 receptor P subunit from

metabolically labeled cells with an anti-IGF-I receptor antibody (aIR3) (Fig 1).

Therefore, in PV cells the increased ability of the receptor to respond to IGF-1, as

evidenced by increased tyrosine phosphorylation, appeared to be specific, and was not

simply caused by an increase in the amount of receptor protein.

We had previously shown that circulating progenitor cells in patients with PV are

hypersensitive to IGF-1 with respect to erythroid burst formation in serum-free medium.

In nomais, no bursts developed at IGF-1 concentrations beiow 10" moliL, whereas in PV

burst formation occurred at IGF-1 concentrations below 1 O-" moK. To determine

whether the IGF-1 receptor is hypersensitive to IGF-1 with respect to tyrosine

phosphorylation, PV and normal cells were exposed to IGF-1 at concentrations of 10 '~

moUL and IO-'' mol& and then lysed. Proteins were resolved by SDS-polyacrylarnide gel

electrophoresis (SDS-PAGE) and Western blotted with antiphosphotyrosine antibody. To

quantitate tyrosine phosphorylation of the IGF-1 receptor in response to ligand,

densitometric analysis was perfomed. Figure 2 shows data on Western blots from eight

PV patients and six nomals, normalized as percent of maximum + SEM. The

densitometric data demonstrate an increased tyrosine phosphorylation in PV over normal

cells in the absence of added IGF-I(O.O1 > P > 0.001) and in the presence of low

concentration of ligand, 10'" molL IGF-1, at 1 and 3 minutes (0.02 > P > 0.0 1 and 0.00 1

> P, respectively). Tyrosine phosphorylation in PV appeared to peak at 3 minutes and

then decrease; however, in the normal. the apparent increase in tyrosine phosphorylation

at 10 minutes did not reach the Ievel of statistical significance (P > 0.05). In the presence

of high concentration of ligand, 10" moVL IGF-1 in normal, tyrosine phosphorylation was

significantly increased at 10 minutes over time O (P < 0.00 1); in PV, tyrosine

phosphorylation was significantly increased at 3 minutes (0.0 1 > P > 0.00 1). Together

rhese data indicate that not only is the IGF-1 receptor in PV hyperresponsive to IGF-1 as

compared with the normal in that, at a given concentration of ligand, the magnitude andor

rate of the response are substantially greater, but it is also hypersensitive in that it

responds to ligand concentrations 100-fold less than required to activate the receptor in

normal cells (Fig 2). This mirrors the increased sensitivity to IGF-1 previously observed in

PV with respect to erythroid burst formation in vitro."

Specificity of increased tyrosine phosphorylation of the IGF-I receptor in PV

To test the specificity of the observed increase in tyrosine phosphorylation, we

examined three patients with erythrocytosis who had elevated hematocritslRBC mass, but

did not fuifil the cnteria for PV. As Fig 3 illustrates, there appeared to be no significant

difference (P > 0.05 for al1 points) with respect to tyrosine phosphoqlation of the 95-kD

band, either in the absence of IGF-1, or in the presence of IGF-I at low or high

concentrations. These results were confirmed when tested against a larger pool of

nomals (data not shown). Thus, the phenornenon of a receptor that had increased basal

tyrosine phosphoryIation, and was hypersensitive and hyperresponsive in the presence of

ligand, appears to be specific to PV.

DISCUSSION

We have show that in patients with the myeloproliferative disorder PV i) the IGF-I

receptor p subunit of circulating mononuclear cels, in the basai unstimulateci state,

demonstrated an increased level of tyrosine phosphorylation over that of the normal. This was

not due to a simple increase in the amount ofreceptor protein in PV (Fig. 1). ü) The induction

of tyrosine phosphorylation by IGF-I in PV occurred at approximately 100-fold lower ligand

concentration than in normal, i. e. it was hypersemitive to ligand. It also peaked eariier and

reached higher intensity, Le. it was &perrepmive to ligand. These findings are consistent

with, extend and provide a molecular basis for Our previous observation that in PV, erythroid

progenitor cells arnong circulating mononuclear ceIls are hypersensitive to IGF-I with respect

to erythroid burst formation in vitro5g. IGF-1 is known to play a role in emopoiesis in vitro

and in vivo and in adults whose erythropoietin driven erythropoiesis is k ~ ~ a i r e d ~ ~ .

Whether the enhanced basal tyrosine phosphorylation observed in our cultures of

PBMNC from patients with PV is due to a constitutively activated receptor tyrosine kinase or

simply represents a hypersensitive response to minute arnounts of ligand which may be released

by the PBMNC during the course of the experiment is unknown. Furthermore, whether the

observed increase in tyrosine phosphorylation actually represents an increase in biological

activity of the IGF-1 receptor in PV still remains to be formally established. Experirnents

designed to resolve these questions are at present underway in Our laboratory.

Either of two hypotheses could account for our findings: The k s t is a selective

hypothesis at a ceII population level, and essentiaiiy represents amphfication of a pre-existing

rninority sub-population of normal ceIis that possess an enhanced basal tyrosine kinase activity

and an enhanced sensitivity and response to IGF-1. This hypothesis postdates that the PV

phenotype results from the abnormal expansion of an otherwise normal subpopulation, but it

laves unanswered the question of what is responsible for the self-renewal and selective

amp tifkation of this particular subpopulation in PV. For example, expansion of a normal celi

population having fetal IGF-I receptor'l6 with increased tyrosine kinase aaivity would fit this

model. The second is a mutational hypothesis, and postdates that a mutation in the IGF-I

receptor or in its signalling pathway, which rendered the proliferative signal constitutively

active in the absence of ligand and hypersensitive as well as hypemesponsive in the presence of

ligand, could be the acquired abnorrnaiity of the PV hematopoietic progenitor clone that lads

to the PV phenotype. However, where such a defect would lie is open to discussion. An

altered IGF-1 receptor with the subunit having a relative molecuiar weight of 105kDa has

been found in Ieukemic tells"'. But from Figs. 1 above, as weii as from data not shown, we

found no evidence of an abnomality in the IGF-1 receptor P subunit that wouid be manifest as

a change in its relative rnolecular weight. However, a lesion in the a subunit of the receptor

which affects its tyrosine kinase activity, or a discrete lesion in the B subunit which may not

alter its relative molecular weight, is of course possible and would not be detected by the

methods used here. If the lesion does not lie within the receptor, but lies within the IGF-I

receptor signal transduction pathway or regulatov proteins, then it mua be somehow closely

associated with the receptor, since the defect appears to have a dUea effect on its kinase

activity, as evidenced by increased autophosphorylation in PV. Among the prirnary candidates

for such a defect would be those elements involved directly in IGF-1 receptor activation and/or

desensitization.

Table 2

Densitometric Analysis of the 95-kD Band in Normal and PV

Time (min) O I 3 1 O O 1 3 10

Absorbancc (arbitrary units) 0.140 0.201 0.270 0.452 0.248 0.581 0.961 0.500 Percent maximum absorbance 14.6 20.9 28.1 47.0 25.8 60.5 100.0 52.0 Antiphosphotyrosiric

Absorbancc (arbitrary uniis) 0.345 0.545 0.286 0.344 0 . 3 0 8 0.301 0.276 0.385 Pcrccnt maximum absorbancc 63.3 100.0 52.5 63.1 56.5 55 .2 50.6 70.6 Ariti-IGF-1 rcccptor

Representative time course of absorbance in arbitrary units in the presence of 1 0 ' ~ mollL IGF-1 in a normal individual and a PV patient. PBMNC were isolated and placed in serum-free culture. They were then exposed to 1 0 . ~ molL IGF-1 for 0, 1, 3, and 10 minutes. The cells were lysed and the lysates were resolved on duplicate 8% gels by SDS-PAGE. The proteins were then transferred to nitrocellulose membranes. (A) Tyrosine phosphorylation of the 95-kD band (IGF- I p subunit as confirmed by Western blot with anti-IGF-1 receptor antibody) The nitrocellulose membrane was Western blotted with a monoclonal antiphosphotyrosine antibody (4GIO). (B) Amount of IGF-1 receptor protein A duplicate nitrocellulose membrane was Western blotted with a polyclonal anti-IGF-l receptor antibody (UBI). The highest absorbance value, whether normal or PV, was taken as 100.0% and percent maximum for each time point was calciilated.

Figure 2.1 - Cornparison of tyrosine phosphorylation in normal and PV cells at high IGF-1

concentration for similar amounts of IGF-1 receptor. Time course of tyrosine

phosphorylation after immunoprecipitation of the IGF-1 receptor B subunit in the presence

of 10.~ moK IGF-1. PBMNC were metabolically labeled with 3 5 ~ - ~ e t and ? S - C ~ S

(translabel; ICN) and placed in serum-fiee culture. They were then exposed to 10 '~ mol/L

IGF-1 for 0. 1. 3 and 10 minutes. The cells were lysed, and the lysates were used to

immunoprecipitate the IGF-1 receptor P subunit with a monoclonal anti-IGF-1 receptor

antibody (uIR2). Irnmunoprecipitates were resolved on an 8% gel by SDS-PAGE and the

proteins were then transferred to nitrocellulose membrane and Western blotted with a

monoclonal antiphosphotyrosine antibody (4G10). As a control for the arnount of IGF-1

receptor, a duplicate gel was drïed and used to expose x-ray film.

------ N --- ---- pv --------

Time (min) 0 1 3 1 0 0 1 3 1 0

Anti-Phosphotyrosine ,

IGF-1 Receptor P Subunit H)ilH#WHP)H,

Figure 2.2 - Cornparison of tyrosine phosphorylation of the 95-kD band fiom PV and

normal cells at 10'" mol/L and 10-~ m o n IGF-1. Western blots derived fiom eight PV

patients and six normal individuals were used for quantitation of tyrosine phosphorylation

of the 95-kD band by densitometric analysis. The data are presented as percent maximum

+ SEM. (A) Tyrosine phosphorylation of the 95-kD band in response to the low -

concentration of IGF-1 in cells fiom patients with PV (O) and normal individuals (a).

(B) Tyrosine phosphorylation of the 95-kD band in response to the hgh concentration of

IGF-I in cells from patients with PV (0) and normal individuals (R).

Figure 2.3 - Comparkon of tyrosine phosphorylation of the 95-kD band fiom PBMNC of

patients with erythrocytosis and normal individuais at 10-'O m o n and lu8 mol/L IGF-1.

Western blots representing three patients with erythrocytosis @,O) and three normal

individuals (a,.) were used for quantitation of tyrosine phosphorylation of the 95-kD

band by densitometric analysis. Data are presented as percent maximum 5 SEM. (A)

Tyrosine phosphorylation of the 95-kD band in response to the low concentration of IGF-

1. (B) Tyrosine phosphorylation of the 95-kD band in response to the high concentration

of IGF-1.

Time (min)

Time (min)

CHAPTER 3

CHAPTER 3

Insulin-like growth factor binding protein-l (IGFBP- 1 ) is elevated in patients with

Polycythemia vera and stimulates erythroid burst formation in vitro2

Amer M. Mirza', Shereen Ezzat*, and Arthur A. Axelrad'

Department of Anatomy and Ce11 Biology', and

the Division of Endocrinology and Metabolism'

University of Toronto, Toronto, Ontario

Canada

: This work was supported by a grant from the Medical Research Council of Canada (MA3969).

Portions of this work were presented at the Amcrican Sociel of Hernatology meeting, Seattle. WA. December 1995 (Blood 86 (10) suppf 1. p. 15 la)

A manuscript derived from this chapter has been accepted for publication in the March 17, 1997 issue of Blood.

ABSTRACT

Previously, we found that in the myeloproliferative disorder Polycythemia vera

(PV) circulating erythroid progenitor cells were hypersensitive to insulin-like growt h

factor I (IGF-1), an effect s h o w to occur through the IGF-1 receptor. Also, in ceils of PV

patients, the IGF-1 receptor was hyperphosphorylated on tyrosine residues under basal

conditions, and its tyrosine phosphorylation in response to exogenous IGF-1 was strongly

augmented. Thus, as IGF-1 appeared to play a role in the pathogenesis of PV, we wished

to assess its level in the circulation of these patients. Normally, most of the circulating

IGF-1 is bound to specific high affinity IGF binding proteins which can regulate its activity.

We determined the circulating levels of IGF-1 and two of its key binding proteins, IGFBP-

I and IGFBP-3. In two separate expenments, plasma samples from a total of 23 PV

patients age- and sex- matched with 41 normals were compared by radioirnmunoassay.

The levels of IGFBP-1 in patients with PV (37.80 +/-4.33pg.L) were more than four-fold

higher than in normals (9.34 +/- 1.34pgL) or patients with secondary erythrocytosis (9.47

+/- 1.96pg/L) , while the plasma concentrations of IGFBP-3 and IGF-1 in these patients

were simiiar to those of normal subjects. As circulating IGFBP-1 levels may be influenced

by insulin, we measured the concentrations of insulin in the same samples. Our data

showed that the elevation of circulating IGFBP- 1 in PV could not be attributed to low

levels of insulin in these patients. The substantial increase in concentration of IGFBP- 1

was confirmed on ligand blots performed with ' 2 5 ~ - ~ ~ ~ - ~ . IGFBP-1 can be either

inhibitory or stimulatory to the action of IGF-1 under different conditions. We reasoned

that if IGFBP- 1 were stimulatory for erythropoiesis, an elevated IGmP- 1 level could help

to explain the increased sensitivity to IGF-1 observed in PV. If IGFBP-1 were inhibitory,

it might suggest a compensatory mechanism where a hyperphosphorylated IGF-1 receptor

in PV might induce a negative modulator of IGF-1 action, in this case IGFBP-1 . To

distinguish between these two hypotheses, we titrated the effect of IGFBP-1 in the

presence of IGF-1 with respect to erythroid burst formation and found that IGFBP- 1 was

strikingly stimulatory. The elevated level of IGFBP- 1 coupled with its ability to stimulate

erythroid burst formation provide an attractive mechanism to account for the increased

sensitivity of erythroid progenitor cells to IGF-1 and the consequent overproduction of red

cells charactenstic of Polycythernia vera.

INTRODUCTION

Polycythernia vera (PV) is a chronic rnyeloproliferative disorder characterized by

hyperplasia of the three major myeloid lineages, but with particular emphasis on the

erythroid lineage118. A relentless overproduction of red blood cells occurs in the presence

of normal oxygen saturation and with levels of serum erythropoietin often depressed?

Like erythropoietin (Epo), insulin-like growth factor 1 (IGF-1) can stimulate erythroid

progenitor ce11 proliferation and differentiation, but it normally requires a much higher

concentration to bnng about this effect5' We have previously shown that in PV,

circulating erythroid progenitor cells are hypersensitive to IGF-I with respect to burst

formation in serum-free culture, and that this effect occurred through the IGF-1 receptorSg

Recently, we have found that tyrosine phosphorylation of the IGF-1 receptor P subunit

was increased in the cells of PV patients in the absence of exogenous ligand. In the

presence of ligand, tyrosine phosphorylation in PV cells occurred more rapidly, at lower

concentrations of IGF-1, and attained a higher level of phosphorylation than in n~rma l s ' ' ~

These findings strongly suggest that IGF-1 andior its receptor play a role in the

pathogenesis of PV.

IGF-I is a highly conserved 70 amino acid 7.5 kDa protein which is produced

ubiquitously120. It exens acute anabolic effects on protein and carbohydrate metabolism in

a wide variety of tissues, as well as more long term effects on ce11 proliferation and

differentiation"' .

There is very little fiee IGF-I in the circulationlu. Most of the IGF-I circulates

bound to specific high affinity binding proteins. These homologous but distinct IGF

binding proteins (IGFBP- 1 to -6) regulate both the bioavailability and the bioactivity of

IGF-I in extracellular fluidslu. The binding proteins exhibit tissue and developmental

specificity with respect to their expression, and thus are thought to play an important

physiological role in IGF transport, localization, and action, but the exact mechanisms

remain to be elucidated. Two key IGF binding proteins are IGFBP-3 and IGFBP- 1 . Most

of the IGF-1 in circulation is bound by IGFBP-3, whose circulating levels are more than

IO-fold higher than any of the other binding proteins'u. IGFBP-3, regulated by growth

hormone, is known to inhibit the biologicai activity of I G F - I ~ * ~ ' ~ ~ . IGFBP-1, regulated by

a variety of factors12', is believed to be an acute modulator of IGF-1 action, under different 129.130 conditions either enhancinglza or attenuating its activity .

The potential importance of IGF-1 in PV prompted us to investigate the levels of

IGF-I and its binding proteins IGFBP- 1 and IGFBP-3 in the circulation of patients with

this disorder. In the present study, we found that the level of circulating IGFBP- 1 was

substantiaily and specifically increased in PV, an observation that led us to ask whether

this binding protein was inhibitory or stimulatory for erythroid progenitor cell

proliferation.

METHODS

Patients and Normal controls

Patients cllliically diagnosed as having PV with the help of the PV Study Group

guidelines6 were used in this study. AU patients were being managed by phlebotomy at the tirne

of study; one PV patient had previously been treated with hydroxyurea. Patients categorized

as having secondary erythrocytosis in this snidy had an increase in the number of circulating red

cells, but were clinicaily mled out as having PV according to the PV Study Group guidelines6

Norrnai volunteers were healthy individuals who were age- and sex- matched to PV patients.

Radioimmunoassay (RIA)

The levels of IGFBP- 1, IGFBP-3, IGF-1, and insulin were determined by

radioirnmunoassay as previously descnbedl". The assays for IGFBP- I and IGFBP-3 were

perforrned with commercially available kits (Diagnostic Systems Laboratories, Webster,

TX). The kits had an intraassay variability of 5.2% and 6.7%, respectively. Their

interassay variability was 4.3% for IGFBP-1 and 3.5% for IGFBP-3. Semm IGF-1 levels

were determined with a kit from Nichols Institute Diagnostics (San Juan Capistrano, CA)

which had an intraassay variability of 3.0% and an interassay variability of 5.2%. Serum

insulin levels were determined with a commercially available kit (ICN Biomedicals, Costa

Mesa, CA) with a sensitivity of 1.4 mU/L and intraassay variations of 13.1% and 2.5% for

5.0 and 26.0 mU/L, respectively. Al1 assays were performed in duplicate, and the same

plasma sample was used for IGFBP- 1, IGFBP-3, IGF-1, and insulin determinations.

Irnmunoprecipitation of IGFBP-1 and Ligand Blotting

In briec following informeci consent, patient and normal blood samples were coilected

in EDT A vacutainers (Becton, Dickson, Rutherford, NJ) and centrifuged at 45ûxG for 1 0 min.,

the plasma was removed and was either aored at -20°C or used imrnediately. Sarnples of

plasma were centrifugeci at 3000xG and 200pL of the resulting supernatant were precleared

overnight at 4°C with 100p.L of a prepared 1 : 1 slurry of protein G-Sepharose (Sigma) plus

dilution buffer (0.1% Triton X- 100,O. 1% bovine hemoglobin in TSA (1 OmmoVL Tris-HCI, pH

8.0, 140 rnmoi/L NaCl, 0.025% N&)) and the supernatant removed. Supernatants in a

300p.L final volume were placed in 1 SmL microfuge tubes precoated with dilution buffer, a

1 500 dilution of rabbit anti-human IGFBP- 1 polyclonal antibody V I , Lake Placid, NY) was

added, and the contents of the tube were incubated at 4°C on an orbital shaker for 1 hour.

Then 25p.L of a prepared slurry of protein G-Sepharose was added to the tubes and the

samples were incubated at 4°C ovemight on an orbital shaker. The complexed Sepharose was

then washed four times: &ce in dilution baer , once in T S 4 and finally in 50 mmoVL Tris.CI,

pH 6.8. Mer the final wash, the imunoprecipitated sarnples were prepared with non-

denaturing loading buffer and boiled for 5 min. The ligand blots were othenvise performed as

previously describedl3'. The molecular weight markers (Bio-Rad, Hercules, C.4) and prepared

samples were subjected to electrophoresis on a 12% SDS-gel under non-denaturing conditions.

Proteins were electrophoreticaily transferred to supporteci nitrocellulose membrane (Bio-Rad,

Hercules, CA). The membrane was blocked with non-specific protein (1% fatty acid-fiee and

globulin-free bovine semm aibumin, 0.1% Tween 20 (T), in Tris buffered saline (TBS)) for 1

hour. Following primary blocking, the membrane was incubated overnight with &esh blocking

solution plus 2pCi '%IGF-1 (ICN, Montreal, Quebec, Canada). The nitroceilulose membrane

was subsequently washed with TTBS three times for 15 min. each and the membrane was used

to expose x-ray film at -70°C. The x-ray film was analyzed with an LKB Ultrascan XL Laser

Densitometer &KI3 Biochrom, Cambridge, UK) to quantitate the IGFBP- 1 bands found in PV

and normal individuals, as well as of the recombinant IGFBP- 1 (UBI, Lake Placid, NY) used

later in this study.

Cell preparations and colony assay in vitro

M e r informeci consent, peripheral blood was obtained by venipuncture fiom healthy

voiunteer donors and patients, and was immediately placed in a polypropylene tube containing

1 0Ulrn.L preservative-fiee sodium heparin (#820 5077MF;Gibco). Within 2-4 hours, the

heparinized blood was Iayered ont0 15mL of Ficoii-Hypaque (Phamcia, Montreai, Canada)

and the light-density mononuclear cells (PBMNC) were collecteci afler centrifugation at 450g

for 30 min at room temperature. Ce1 suspensions were washed three times (450g 10 min at

room temperature) in a-minirnal essential medium (aMEM) containhg 0.1 % fatty acid-fkee

and globulin-&ee bovine serum aibumin (FAF-BSA), and the ceUs were counted.

The culture assays for erythroid bursts were performed in serum-free medium

(SeroZeroTM Stem Ceil Medium (SCM); US Patent Number 5,397,706 Mar. 14,1995) with

methylcellulose as previously described with the following changes. Recombinant human

IGFBP- 1 (UBI, Lake Placid, NY) was titrated (3x 1 O-''' M to 3x 1 O-' ' M) in the presence of

3x10-" M recombinant human IGF-1 (R&D Systems, Minneapolis, MN) in aMEM plus

0.1% FAF-BSA, and preincubated for 1 hour at room temperature to allow complex

formation to occur. The IGF-MGFBP- 1 putative complex was added to Ficoll-separated

peripheral blood mononuclear celis and the cells were plated as describedSs.

Hemoglobinized burst component colonies of 250 cells were scored at 14 days.

Statistical analyses

Data on concentrations of circulating IGF-1, IGFBP- I , IGFBP-3 and insulin

obtained for each experiment were tested with the Wilcoxon rank sum test for non-

parametric data. The mean +/- standard error of the mean were calculated for the pooled

data fiom the two experiments and the results were analyzed with Student's unpaired t

test.

RESULTS

Levels of circulating IGF-1 and its binding proteins IGFBP-3 and IGFBP-1 in

patients with Polycythemia vera

Levels of plasma IGF-1, IGFBP-3, and IGFBP- 1 were determined in two separate

experiments for a total of 23 PV patients age- and sex- matched with those of 4 1 normal

individuals. The resdts are presented in Figures 1 and 2. We found that the circulating

level of IGF-1 in patients with PV (1 S 1.30 +/-13.73pgL) was not significantly different

(0. Pp>O.6) from that of normal controls (143.85 +/- 1 1.3 S p a ) . Similarly, when we

compared the circulating levels of IGFBP-3, we found that the level in PV (2.89 +/-

0.24mgL) was not different (0.3>p>0.2) corn that of the normal (2.47 +/-O. 14mgL).

However, the level of circulating IGFBP- I in PV (3 7.80 +/-4.3 3 pgR) was significantly

greater (pc0.00 1) than that of normal individuais (9.34 +/- 1 -34pgL). Thus, in PV, the

circulating levels of IGFBP- 1 were more than four-fold higher than in the normal controls,

whle the plasma concentrations of IGFBP-3 and IGF-I in these sarne patients were not

significantly different From those of normal individuals. As high levels of IGFBP- 1 can be

associated with low levels of insulin13', we determined the concentrations of insulin in

these same plasma samples in PV (144.47 +/-37.22pmoVL) and found them not to be

significantly different (0.1 0>p>0 .OS) from those of normal individuals (222.73 +/-

25.93prnoUL). As an independent test, we compared IGFBP- 1 concentrations in a group

of 5 PV patients and 5 normal individuals with comparable (0.9>p>0.8) levels of insulin

(1 7 1.60 +/- 18.26 and 176.80 +/-19.03pmoVL respectively). We determined that the level

of IGFBP- 1 in these PV patients (23 .O4 +/-Cl3 pglL) was significantly greater than in the

normals (3.86 +/- I.32pgL) (0.02<p<0.05) for comparable levels of insulin. Thus, the

elevation of IGFBP- 1 in PV was independent of insulin and therefore could not be

attributed to low levels of insulin in these patients.

We wished to determine whether IGFBP- 1 was specifically elevated in patients

with PV, or was simply either the cause or the result of a non-specific increase in the

number of red cells. In a separate experiment, we compared the levels of circulating IGF-I

and IGFBP- 1 in 7 patients with secondary erythrocytosis, 3 normal individuals, and 4

patients with PV. The resuits are shown in Figure 3. We hund that the plasma IGF-1

levels in patients with secondary erythrocytosis (237.67 +/- 24.62pgL) were not different

(0.2>p>O. 1) from those of patients with PV (1 76.25 +/- 34.86pgL). Also, the plasma

levels in PV were not different (0.5>p>0.4) from normal (145.00 +/- 25.5 1). In contrast,

IGF-1 levels in patients with secondary erythrocytosis appeared to be slightly elevated in

cornparison to normal (0.05>p>0.02). The plasma IGFBP- I levels in patients with

secondary erythrocytosis (9.47 +/- 1.96pgL) were not significantly different (O. 1 <p<O. 05)

from those of normal individuals (1 S. 17 +/- 2.07pgL). However, the IGFBP- 1 levels in

patients with PV (49.7 +/- 6.56pg/L) were significantly greater (p<O.OOl) than those

observed for either patients with secondary erythrocytosis, or normal individuals. Thus, it

appeared that IGFBP-1 is specifically elevated in PV.

Plasma sarnples obtained from normal individuals and PV patients were used to

imrnunoprecipitate IGFBP- 1 . These immunoprecipitates were then subjected to ligand

blotting with 1 2 5 ~ - ~ ~ ~ - ~ (Figure 4). A representative autoradiograph reveals the relative

amounts of IGFBP-1 in plasma sarnples From normal individuals and patients with PV. It

demonstrates that the arnount of circulating IGFBP- 1 was increased in PV over that in

normals. The increase was detennined to be approximately three-fold by densitometric

analysis. Thus, the ligand blots cofirmed the specific increase of IGFBP-1 in PV

observed by RIA. They aiso showed that there are at least two species of IGFBP- 1

present in circulation of both PV patients and normals. The approxirnately 3OkDa band

represents a putative non-phosphorylated species of IGFBP- 1, whereas the approximately

28kDa band probably represents a phosphorylated ~ ~ e c i e s l ) ~ . The ratio of non-

phosphorylated to phosphorylated IGFBP- 1 (approximately 4: 1 by densitometric analysis)

appeared to be similar in PV and nomal, both of which appeared to be similar to that of

the recombinant IGFBP-1 used later in this study.

Effect of IGFBP-I on erythroid burst formation in vitro

Increased levels of IGFBP- 1 in PV could have strikingly different consequences

depending upon whether this binding protein bad an inhibitor). or stimulatory effect on the

action of IGF-1. To determine the biological effect of IGFBP-1 on erythroid burst

formation in vitro, we titrated IGFBP- 1 in the presence of IGF-1, and assayed its efYects

on erythroid burst formation in semisolid semm-fiee culture. IGFE3P- 1 was titrated in the

presence of 3x 1 O-'' M IGF-1, a concentration previously shown to be stimulatory for

erythroid burst formation in PV but not in the The results are presented in

Figure 5 and Table 1. We found that in both normal individuals and patients with PV,

IGFBP-1 was strikingly stirnulatory for erythroid burst formation in the presence of IGF-1.

This stimulatory action occurred in the normal at IGFBP- 1 concentrations as low as 3x 10'

"M IGF-1. It should be noted that normal erythroid progenitors usually do not respond

to IGF-1 levels less than 1 O-'' M. Therefore, IGFBP- 1 appeared to be able to stimulate

erythroid burst formation by markedly increasing progenitor ce11 sensitivity to IGF-1. In

PV, the same stimulation of erythroid burst formation occurred at IGFBP- 1

concentrations as low as ~ X I O - " M. This indicates that, although IGFBP-1 stimulated

erythroid burst formation in both PV and normal, progenitor ceils in PV appeared to

require considerably less IGFBP- 1. In terms of half-maximum effects, erythroid burst

formation occurred in PV at an IGFBP- 1 concentration -60-fold lower in PV (- 1 x 1 O'"

M) than in normal (-6x 1 0-13 M).

DISCUSSION

We have made two main observations in the present study: 1) in vivo in patients

with PV, the circulating level of IGFBP- 1 was substantially and specifically elevated over

the normal; 2) in vitro in cells of both PV patients and normal individuals, IGFBP-1 in the

presence of IGF-1 acted as a powefil stirnulator of erythroid burst formation.

With MA we found that circulating IGFBP- 1 was more than four-fold higher in

PV patients (37.80 +/-4.33pg/L) than in normals (9.34 +/-1.34pg.L) or in patients with

secondary erythrocytosis (9.47 +/- 1.96pg/L), while the plasma concentrations of IGF-1

and IGFBP-3 in these patients were not significantly different from nomal. Ligand blots

confirmed that the plasma levels of IGFBP- 1 in PV patients were increased. Although

IGFBP- 1 levels are regulated by a number of hormonal and metabolic factors, insulin is 127.135 nonnally considered to be a pnmary physiological regulator of IGFBP-1 in plasma .

Insulin is known to depress IGFBP- 1 in vivo I33.136 and to inhibit its production by human

fetal liver in explant culture, as well as by the human hepatoma-denved ce11 line Fiep~2"'.

Therefore as high levels of IGFBP-1 can be associated with low insulin levels, we

compared IGFBP- 1 concentrations in PV vsrsus normal at comparable levels of insulin.

We demonstrated higher (0.05>p>0.02) IGFBP-1 levels in PV than in normal individuals

for comparable levels of insulin, and thus mled out low insulin in the circulation of these

patients as a causative factor for the elevation of IGFBP-1 in PV. IGF-1 itself has also 138.139 been impiicated in the regulation of IGFBP-1 levels . However, in the present study

we found plasma IGF-1 levels in PV to be normal (Figures 1-3). Thus, the elevation of

IGFBP- 1 appears to be related to the PV defect and is not simply due to the metabolic

fluctuations to which this binding protein is susceptible. Also, the fact that circulating

levels of IGFBP- 1 are not elevated in patients with secondary erythrocytosis indicates that

the increase is specific to PV, and thus may be part of the PV defect, and cannot be

attributed to events which non-specifically increase the number of circulating red cells.

Further work will be necessary to determine whether the elevation of IGFBP-I

concentration in PV is due to its increased biosynthesis, mobilization fiom pools,

diminished clearance fiom the circulation, or a combination of any of these factors,

Interestingly, we found the level of IGF-1 to be slightly elevated in patients with

secondary erythrocytosis (23 7 +/- 24.62) compared to normais ( 145.00 +/- 25.5 1 ) (Figure

3) . As IGF-I is known to play a stimulatory role in erythr~~oiesis'~, it may be that the

increased IGF-I in circulation is in part responsible for the observed erythrocytosisl"O.

IGFBP- 1 has been s h o w to play either a s t i r n u l a t ~ r ~ ~ ~ ~ * ' ~ ~ ~ ~ ~ or an ifibitory134, 1'". 142 role in modulating IGF-I action under different experimental conditions.

In the case of MDA-23 1 cells, IGFBP-I bound IGF-I and augmented its mitogenic effects.

These cells normally proliferate in response to exogenous IGF-1, but exposure to IGFBP- 1

in addition to IGF-1 enhanced their mitogenic response by approximately two-f01d'"~. In

contrast, in the case of MG-63 osteosarcoma cells, which also proliferate in response to

exogenous IGF-1, the addition of IGFBP- 1 in the presence of IGF-1 suppressed

subsequent stimulation of DNA synthesisl"; this effect was believed to be due to

IGFIIGFBP- 1 complex formation which effectively prevented IGF-1 binding to its

receptor.

We reasoned that if IGFBP- 1 were stimulatory to the action of IGF-1, an elevated

IGFBP- 1 level could help to explain the increased IGF-1 sensitivity observed in PV. The

IGF-VIGFBP- 1 complex could facilitate ligand interactions with the IGF-1 receptor and in

such a system, less IGF-1 would be required to activate the receptor. In this way, even

nomal levels of IGF-1 could constitutively activate the IGF-I receptorNg In contrast, if

IGFBP-1 were inhibitory, it might suggest a compensatory mechanism. In PV, the

hyperphosphorylated IGF-1 r e ~ e ~ t o r l ' ~ could result in an amplified mitogenic signal which

might be interpreted by the cells as an increase in the circulating level of IGF-1. The cells

could then respond by increasing the synthesis/release of IGFBP-1, in this case a putative

negative regulator of IGF-1 action. To distinguish between these two hypotheses, we

titrated the effect of IGFBP-1 with respect to its ability to affect the number of erythroid

burst component colonies under serum-fiee conditions in the presence of IGF-1. We

found that the action of IGFBP-1 was clearly stimulatory for erythroid burst formation

(Figure 5 , Table 1). IGFBP-1 increased the number of burst component colonies in cells

obtained from both normal individuais and patients with PV. It also increased the

sensitivity of erythroid progenitor cells in PV and normal by approximately two orders of

magnitude. However, in PV we found that a substantially lower concentration of IGFBP-

1 was required for this effect. Whether it is IGFBP-1 alone or the combination of IGFBP-

I with IGF-1 that is responsible for the increased erythoid burst formation seen in this

work is at present under investigation.

Although our data are consistent with the notion that IGFBP-1 acts as a positive

modulator of IGF-I action, the basis for the stimuiatory nature of IGFBP- 1 seen in the

present work remains to be elucidated. However, previous work may help to shed light

on this issue. Two isofoms of IGFBP- 1 isolated from human arnniotic fluid reveaied that

one was able to potentiate the effects of IGF-1, while the other was inhibitory13'.

Subsequent work has suggested a relationship between the biologicai activity of IGFBP- 1

and its state of phosphorylation. Cultured human endometrial stromai cells from

proliferative phase endometriurn secrete non-phosphorylated IGFBP- 1, while cells from

the secretory nonproliferative phase endometrium secrete a heavily p hosp horylated

IGFBP- 1 lu. HepGZ cells produce an IGFBP-1 which has been used in many studies

showing that it is inhibitory; this IGFBP- 1 is predorninantly phosphorylatediJ5, while a

recombinant IGFBP- 1 which has been show to be stimulatory, is n~n-~hos~horylated~'~.

It is also known that phosphorylated IGFBP- 1 binds IGF-1 with a higher affinity than its

non-phosphorylated i s~ fo r rn '~~ , thus effectively inhibiting IGF-1 activity by preventing its

interaction with the receptor. Non-phosphorylated IGFBP- 1, whch binds IGF-I Iess

tightly, may be able to transport and release IGF-1 to the receptor, thereby localizing IGF-

1 activity and potentiating its action. From our data (Figure 4) it appears that, as in the

case of human arnniotic fluid, there are at least two isoforms of IGFBP-1 in the

circulation, however fùnher work is required to elucidate the biological activities of these

isoforms. Expenrnents are currently undenvay in Our laboratory to determine the

phosphorylation status of circulating IGFBP- 1 in PV patients and normal individuals.

In PV, the marked increase in red ce11 mass is not due to elevated Epo levels, nor

as we have shown in this study, is it due to elevated levels of free IGF-I in circulation. As

our results indicate, plasma IGF-I levels in PV are not different ffom those in normal

individuals. However, erythroid progenitor cells are hypersensitive to IGF-I in this

disorder. This could be due to an intrinsic lesion in the IGF-1 receptor signalling pathway,

or it could be due to an extra-receptor event(s). We have shown that the level of IGFBP-

1 is increased four-fold in patients with PV and that, in the presence of IGF-1, IGFBP- 1 is

stimulatory for erythroid burst formation. Our findings thus raise interesting new

questions regarding the regdatory role of IGFBP- 1 in vivo. The increase in a positive

modulator of IGF-1 action in PV provides an attractive mechanisrn to account for the

heightened sensitivity of erythroid progenitor cells to IGF-1 in this disorder. They also

suggest that the sustained high circulating level of this positive regulator could be an

important factor in the constitutively excessive erythropoiesis characteristic of PV.

Figure 3.1 - Circulating levels of IGF-1, two key binding proteins, and insulin as

determined by radioimrnunoassay for patients with PV and normal individuals. Data fiom

two separate experirnents ( Expt. 1 (O,@) and Expt. 2 (O,.)) compare plasma levels of

IGF-I in pg/L (A), IGFBP-3 in mgL (B), IGFBP-1 in pg/L, (C), and insulin in pmoVL (D)

for age- and sex-matched normal individuals (N) and patients with Polycythemia vera

(PV). Each point represents one individual while the horizontal bars indicate the rnedian

value for each group. Levels of IGF-1, IGFBP-3 and insulin are normal in patients with

PV, but their IGFBP- 1 levels are significantly elevated.

Plasma IGF-1 Plasma IGFBP-3

0 O

Plasma IGFBP-1

Figure 3.2 - Plasma levels of IGF-1, IGFBP-3, IGFBP- 1 and insulin in normal individuals

(Normal) and patients with Polycythemia vera (PV) (pooled data fiorn two expenments).

Values for IGF-I are given in pg/L (n=4 1 for Normal and n=23 for PV), IGFBP-3 in mg/L

x 1 O-* (n=4 1 for Normal and n=23 for PV), IGFBP-1 in pg/L x 1 O-' (n=4 1 for Normal and

n=23 for PV), (IGFBP-1) in pg/L x IO-' (n=38 for Normal and n=23 for PV), and insulin

in prnollL (1140 for Normal and n=23 for PV) for normal individuals (0) and patients

with Polycythemia vera (m). The level of plasma IGFBP- 1 in PV is significantly higher

than that in normal individuais ((*) p<0.0 1 ; (* *)p<O. 00 1). However, the circulating Ieveis

of IGF-1, IGFBP-3 and insulin are not significantly different in PV versus normal. Three

values for IGFBP- 1 above 104pg/L in the normal (IV) of experiment 1 (Figure 1) were

statistically shown to lie outside the confidence lirnits of both the normal (p<O.OO 1 ) and

the PV (pCO.0 1 ) populations sarnpled. The mean +/- standard error of the mean for the

level of IGFBP- 1 in nomals, if these values were to be excluded, is given in brackets.

Figure 3.3 - RIA data for the levels of circulating IGF-1 and IGFBP- 1 for normal

individuals (Normal), patients with secondasr erythrocytosis (E) and patients with

Polycythernia vera (PV) are shown as the rnean +/- standard error of the mean. Values are

given in pgL for IGF-I (n=3 for Normal, n=6 for E, and n=4 for PV), and in pg/L x IO- '

for IGFBP-1 (n=3 for Normal, n=7 for E, and n=4 for PV). The level of IGF-1 in PV was

not different fiom those in patients with secondary erythrocytosis or nomals (0.2>p>0.1

and 0.5>p>0.4 respectively). In contrast, IGFBP-1 Ievels were significantly higher in PV

cornpared to either patients with secondary erythrocytosis or normal individuals

(pc0.00 1). Therefore, IGFBP- 1 appeared to be specifically elevated in patients with PV.

i Normal

GEmlE

PV

IGFBP- 1 pgR. x10-1

Figure 3.4 - Ligand blot of circulating IGF binding protein 1 in normal individuals and

patients with PV. A representative ligand blot shows plasma levels of IGFBP- 1 in

Polycythernia vera (PV) patients and normal individuals (N). Plasma sarnples fiom 2 PV

patients and 2 age- and sex- matched normals were used to irnmunoprecipitate IGFBP- 1

and these were separated by SDS-PAGE under non-denaturing conditions, transferred to

nitrocellulose membrane and ligand blotted with 12'1-IGF-1. The membrane was then

washed and used to expose x-ray film. Recombinant human IGFBP-1 (UBI, Lake Placid,

NY) was run as a positive control and C indicates a negative control for the antibody

used. Relative molecular weights are shown in kDa and were determined by prestained

molecular weight markers (Bio-Rad, Hercules, CA). The position of the putative non-

p hosphorylated (3 0 kDa) IGFBP- I species is indicated. Using densitometric analysis, we

found the arnounts of non-phosphorylated IGFBP-I in the two normal samples to be

1 .O2 1 and 1.167, respectively, in arbitraq units; in the two PV patients the amounts were

found to be 2.897 and 3.1 15, respectively, while in the recombinant control it was 1.067.

The amounts of the phosphorylated IGFBP-I species (28kDa) were 0.188 and 0.256 in

the two normals, respectively; in the two PV patients they were 0.707 and 0.736

respectively, while in the recombinant control it was 0.292. The ligand blots confirm that

the level of circulating IGFBP- I in PV is greater than in normal.

Figure 3.5 - Plot of the number of erythroid burst component colonies as a percent of

maximum versus the molar concentration of IGFBP-1 in the presence of 3x 1 O *" M IGF-1

Representative data obtained for cells of normal individu& (0) and PV patients (0 ) are

shown. The half-maximum value for PV occurs at - 1 x l O -'' M IGFBP- 1, while that for

the normal occurs at -6x 10 - 1 3 M IGFBP- 1. Thus with respect to burst formation,

erythroid progenitor cells in PV are approximately 60-fold more sensitive than normal to

IGFBP-1 in the presence of IGF-1.

Table 3

Burst component colonies produced in vitro by normal and PV erythroid progenitor cells with increasing rmounts of IGFBP-1 in the presence of IGF-I

Data fiom titrations of IGFBP- 1 in the presence of 3x 1 O - " M IGF-I with respect to erythroid burst formation by PV and nonal progenitor cells in vitro. Data for cells of 2 normal individuals and 3 PV patients are given as mean number of burst component colonies +/- standard error of the mean, and were used to plot the curve shown in Figure 5 .

IGF-1 (M) IGFBP-1 (M)

Normal Normal

3x10'" 3 x 1 0 " ~

3x10'" O

89 i l - 7.2 72 +/- 3.3

3 x 1 0 ~ " 3x 1 0 " ~

84 +/- 7.8 92 +/- 8,8

3x 10'" 3x 10'12

3x10'" 3x 10-"

10 1 +/- 8.0 107 +/- 5.0

164 +/- 4.7 244 +/- 24.6

166+/- 12.8 209 +/- 25.4

CHAPTER 4

CHAPTER 4

A role for insulin-like growth factor binding protein- 1 in erythropoiesis:

Stimulatory effects on circulating erythroid progenitor cells from normal individuais and

patients with Polycythernia vera2

Amer M. Mina and Arthur A. Axelrad

Department of Anatomy and Cell Biology

University of Toronto, Toronto, Ontario

Canada

' This work was nipponed by a grant from the Medical Research Council of Canada (MA3969).

67

ABSTRACT

We had previously s h o w that circulating erythroid progenitor cells from patients

with Polycythemia vera (PV) are hypersensitive to insulin-like growth factor 1 (IGF-1),

dose-response curves in vitro being shified to the left some 2 orders of magnitude as

compared to normal. More recently. in the course of investigating the significance of an

elevated circulating level of insulin-like growth factor binding protein 1 (IGEBP- 1) in PV,

we titrated the effect of this IGF binding protein on erythroid burst formation in culture

and found that in the presence of IGF-1, IGFBP-1 is strikingly stimulatory, and that

erythroid progenitor cells in PV were approxirnately 60-fold more sensitive to its effects

than normal. In the present paper, we descnbe the systematic investigation of this

phenornenon. Penpheral blood mononuclear cells were isolated from 5 normal individuals

and 5 patients with PV and used to examine the effect of IGFBP- 1 at various

concentrations compared to IGF-1 alone. We found that IGFBP-1 alone had no effect on

normal erythroid burst formation. In PV, even at very low concentrations of IGFBP-1

(3x 1 0-14 M), erythroid burst formation was significantly stimulated. IGFBP- 1 at 3x 1 0-l4 M

was at least as effective in this regard as 3x10*11 M IGF-I alone. Maximal burst formation

occurred when IGF-1 and IGFBP- 1 were present in approximately equimolar

concentrations or when BP- 1 was in slight excess; when IGFBP- 1 was in more that 10-

fold excess, it had an inhibitory effect on erythroid burst formation. To titrate the effect of

IGF-1 on erythroid progenitor cells in the presence and absence of IGFBP- 1, cells were

taken from 8 normal individuals and 10 patients with PV. We found rhat in the normal

IGF-I alone stimulated erythroid burst formation in a dose-dependent manner with half-

maximal burst production occumng at approximately 1 . 2 ~ 1 O-'' M IGF-1. However, in the

presence of 3x 1 O-'* M IGFBP- 1, the IGF-1 dose-response curve was shifted to the left

with a new half-maximum of approximately 7 . 5 ~ 1 0 " ~ M. Thus, IGFBP- 1 increased the

sensitivity of erythroid progenitor cells approximately 16-fold, showing that, in the

presence of IGF-1, IGFBP- 1 is a powerful stimulator of normal erythropoiesis. In PV. the

half maximum value for the dose-response curve of IGF-1 alone was shifled from

approxirnately 2 . 4 ~ 1 0'12 M to approxirnately 6x 1 O-" M, thus increasing the sensitivity of

PV progenitor cells by more than 40-fold. As with the IGFBP-1 titration, we noted that

maximal erythroid burst formation occurred when roughly equimolar amounts of IGF-1

and IGFBP- 1 were added to the cultures. In contrast to the IGFBP- 1 titration on normal

cells, greater than 1 O-fold excess of IGF-1 did not result in inhibition on PV cells. We

have thus shown that IGFBP-1 was greatly able to increase the sensitivity of normal

erythroid progenitor cells to IGF-1 and thus mimic the PV phenotype in culture. This

work has helped to shed light on the rote of IGFBP-1 in normal erythropoiesis. It also

provides a mode1 for a possible mechanisrn of IGF-1 hypersensitivity in PV erythroid

progenitor cells, a hallmark of the PV phenotype.

A considerable body of work elucidates the role of IGF-I and its IGFBPs in a 63.157 variety of tissues, and the role of IGF-1 in hematopoiesis is now well documented

However, a role of IGFBP-1 together with IGF-1 in hematopoiesis is a novel concept1"*

and thus merits fbrther investigation.

Insulin-like growth factor I (IGF-1) is a highly conserved 7.5kDa growth factor

which is expressed from a single copy gene on the long a m of human chromosome 12120.

Plasma IGF-1 is produced primanly in the liver under the controi of growth hormone

(GH)~"; however, the synthesis of IGF-1 has been demonstrated in a wide variety of

tissues150 and by many ce11 types, pnmarily of mesenchymal ongin? IGF-1 binds with

high affinity to its receptorg6 to exert a variety of effects. These include acute anabolic

efYects on carbohydrate rnetabolism as well as more long term effects on ce11 proliferation,

differentiation, ce11 cycle progression, and even ce11 deathlu. IGF-1 has been s h o w to

play a role in erythropoiesis" regulating human erythroid progenitor ce11 proliferation147

and apoptosis1s2 in culture.

IGF-1 circulates specifically bound to a farnily of high affinity binding proteins

which modulate its biological effects. To date, six homologous but distinct IGF binding

proteins, numbered IGFBP-I through -6, four of which are found primarily in the

circulation, have been charactenzedlS3. It is believed that the IGFBPs regulate both the

bioavailability and bioactivity of IGF-1 in extracellular fluids'? IGFBP- 1 is a 234 arnino

acid 25kDa protein which is found in the circulation. It, more than any of the other

binding proteins, shows dynamic regulation with rapid modulation of its levels in the

circulation15J, IGFBP-1 is known to be regulated by a variety of factors1" and is believed

to be an acute modulator of IGF-1 action, under different conditions either

enhancing 143.155.156 or attenuating its activity 129,130.157.158

We had previously shown that in Polycythemia vera, a chronic myeloproliferative

disorder in humansl", circulating erythroid progenitor cells are hypersensitive to IGF-1

with respect to burst formation in strictly senim-free culture, and that this effect occurs

through the IGF-I r e ~ e ~ t o r ' ~ . These findings strongly suggested that IGF-1 plays a role in

the pathogenesis of P V However, we have recently shown that although the circulating

level of IGF-1 in these patients was normal, they showed a 4-fold increase in the level of

plasma IGFBP- 1 la. Moreover, IGFBP- 1 in conjunction with IGF-1 was shown to be a

powerful stimulator of erythroid burst production in culture.

Although there is a considerable body of work elucidating the role of IGF-1 and its

IGFBPs in a variety of tissues, their role in normal and abnormal hematopoiesis is largely

unexplored. We wished to investigate the role of IGF-I and IGFBP-1 in normal

erythropoiesis and to determine whether these peptides play a role in the pathogenesis of

PV. Using a strictly senim-fiee semi-solid culture system for the production of erythroid

bursts, we titrated the effects of IGF-I and IGFBP-1 alone, and in combination, on

erythroid progenitor cells fiom normal individuals and patients with PV.

METHODS

Normal Controls and Patients

Patients chcally diagnosed as havhg PV were positively selected for this study with

the help of the PV Study Group guidelines? At the t h e of the study, al1 patients were being

managed by phlebotomy; although, one PV patient had previously been treated with

hydroxyurea. Normal volunteen were healthy individuals who were matched to PV patients.

CeU Preparations

Mer informed consent, peripheral blood was obtained by venipuncture From healthy volunteer

donors and patients, and was irnrnediately placed in a polypropylene tube containing lOU/mL

preservative-free sodium hep& (#820 5077MF;Gibco). Within 2-4 hours, the heparinized

blood was diluted 2-fold with a-minimal essential medium (a-MEM) plus 0.1% fatty acid-free

and globulin-fke bovine serum aibumin PAF-BSA), and layered ont0 15mL of Ficoll-

Hypaque (Pharmacia, Montreal, Canada) in a final volume of 45mL. The interphase containing

the light-density mononuclear cells (PBMNC) were collected d e r centrifugation at 450xG for

30 min at room temperature. Ce11 suspensions were washed three times (450xG, 10 min at

room temperature) in a-MEM containing 0.1% FAF-BSA, and the cels were counted.

Senim-Free ClonaI Cell Culture

The culture assays for erythroid bursts were performed in strictly semm-fiee medium

(SeroZeroTM Stem Ce11 Medium (SCM); US Patent Number 5,397,706 Mar. 14,1995) with

methylcellulose as previously describeds8 with some modifications. In bief, semm-free

culture of PBNMC was performed in flat-bottomed (1.5 x 1 .O cm) plastic wells (Linbro

#76-000-04, Flow Laboratories, MacLeqVA). Recombinant human IGFBP- 1 (UN,

Lake Placid, NY) was titrated From 3x 1 O-'' M to 3x 1 O-' M final concentration in the

presence of 3x 1 O-" M recombinant human IGF-1 (R&D Syderns, Minneapolis, MN) in

aMEM plus 0.1% FAF-BSA. Sirnilarly, recombinant human IGF-Il was titrated fiom

3x IO-"M to 3x 10%4 final concentration in the presence of 3x 1 O-'* M recombinant human

IGFBP- 1 in aMEM plus 0.1% FAF-BSA. The factors were preincubated for I hour at

room temperature to ailow complex formation to occur. The IGF-VIGFBP-1 complex or

control (SCM alone) were added to SCM with 0.8% of 1 500-centripose methylcellulose

(Methocel A4M1 premium grade; Dow Chernical, Midland, MI) and Ficoll-separated

penpheral blood mononuclear cells. The ceils were plated at a final concentration of

1 x 1 O' celVrnL with each well containing 7x 10' cells. Hemoglobinized burst component

colonies of 250 cells were scored at 13- 14 days.

Statistical Analyses

The mean +/- standard error of the mean of the number of burst cornponent colonies for

three wells at each titration point for al1 normals and PVs were determined. The data were

normalized as percent maximum burst component colonies calculated for each individual

within a particular expenment. In the case of Figure 1, the mean of the percent maximum

data +/- standard error of the mean for a set of five normal individuals and five patients

with PV were determined. The data were analyzed with Student's unpaired t test.

IGFBP- 1 action on erythroid progenitor cells from normal individuals and patients

with Polycythemia vera

We exarnined the erythropoietic effect of IGF-I (3x1 O-" M) cornpared with various

concentrations of IGFBP- 1 (3x 1 O-''' M to 3x 1 O-' M) on erythroid burst formation in strictly

semrn-free semi-solid culture. Peripherai blood mononuclear cells (PBMNC) isolated

from 5 normal individuals and 5 patients with PV were cultured with either medium aione,

medium plus IGF-1, or medium plus IGFBP- 1. The nurnber of erythroid burst component

colonies were scored after 14 days and the scores were normalized as percent maximum

values for each individual. The mean +/- standard error of the mean for ail nomals (N)

and patients with PV (PV) were determined. The results are illustrated as a histograrn in

Figure 1. We found that under conditions optimized for erythroid burst f~rmation'~. in the

normal. 3x10-" M IGF-1 had no significant effect on erythropoietic activity cornpared to

the control (0.6>p>0.5). However, this concentration of IGF-1 was stimulatory for

erythroid burst formation in PV (0.02>p>0.0 l), i.e. there was an increase in the

production of erythroid bursts at concentrations of IGF-1 that are not normaily

stimulatory These findings confinned our previous results that erythroid progenitor cells

in PV were hypersensitive to IGF-1". In the presence of IGFBP-1 alone, there was no

significant increase in nomai erythroid burst formation at 3x 1 0-14 M (O. l>p>0.05), 3x IO"*

M (0.9>p>0.8) or 3x 1 O-' M (0.5>p>0.4). In contrast, a marked increase in erythroid burst

formation was observed in PV even at very low concentrations of IGFE3P- 1 (3x10-'' M)

(0.0 l>p>0.00 1). There was a further increase in erythroid burst formation at 3xl0-'* M

IGFBP- 1. However, at a relatively high concentration of IGFEP- 1 (3x 1 O-' M), there was

no stimulatory effect on PV erythroid burst formation above background. Our data

dernonstrate that in PV, while erythroid stimulation was achieved with either exogenous

IGFBP- 1 or IGF-I alone, 3x 1 0*12 M IGFBP- 1 provided stimulation which was significantly

greater than that seen at a IO-fold rnolar excess of IGF-1 aione (0.0 1>p>0.00 1).

Therefore in PV, it appears that at low and moderate concentrations, IGFBP- 1 aione is at

least as effective a stirnulator of erythropoietic burst formation as IGF-1, but not so at high

concentrations. However, IGFBP- I alone appears to have no significant activity in the

normal.

Stimulatory effects of IGFBP-2 in the presence of IGF-1 in normal individuah and

patients with Polycythemia vera

The effects of IGFBP- 1 in the presence of IGF-1 on erythroid progenitor cells were

next examined in cells fiom 5 normal individuals and 5 patients with PV. IGFBP-1 was

titrated over a range of concentrations fiom 3x 1 0-lJ M to 3x1 M in the presence of

3 i O M F I The dose-response curve for a representative normal and PV patient are

shown in Figure 2. In the normal, we found that 3xl0"'M IGF-I alone was insufficient

for the stimulation of erythroid burst formation. However, in PV, where it did stimulate

burst formation, it was not sufficient for maximal burst production. Maximal stimulation

in both the normal and PV was achieved with a combination of IGF-I and IGFBP- 1, and

this occurred at approximately equimolar concentrations of the two proteins (Figure 2 and

TabIe 1). But, when the molar concentration of IGFBP- 1 was more than 10-fold in excess

of the IGF-1 concentration, IGFBP-1 was no longer able to potentiate the action of IGF-1

in the normal, bringing the number of erythroid bursts back to background levels. In PV,

it inhibited the action of IGF-1, bringing the erythroid burst-forming activity below

background levels with IGF-1 aione. It should be noted that IGFBP-1 was able to

stimulate normal erythroid progenitor cells in the presence of 3xl0-" M IGF-I, a

concentration that is not usually stimulatory for normal erythroid burst formation, yet in

the presence of 3x 10" M IGFBP- 1, it was able to provide maximal erythropoietic

stimulation. Thus, it appears that IGFBP- 1 was able to potentiate erythropoietic activity

in both normal and PV. When haif maximum values for the IGFBP- 1 dose-response

curves were calculated for the normal (approxirnately 1 x 1 O-' ' M) and PV (approximately

6x 10-l~ M), PV erythroid progenitor cells were found to be hypersensitive to a

combination of IGFBP- I/IGF-I compared to normal cells. These data provide evidence

for a role of IGFBP- 1, in the presence of IGF-1, as a powerful stimulator of erythroid

burst formation both in normal and in PV. It appears that when IGFBP- 1 is up to 1 O-fold

molar excess, it potentiates the action of IGF-1; however, beyond this point, it inhibits

erythroid burst formation under the influence of IGF-1.

Effects of IGF-1 with and without [GFBP-1 on erythroid burst formation in cells

from normal individuals and patients with Polycythemia vera

To test the hypothesis that IGFBP-1 was able to potentiate the action of IGF-1

with respect to erythroid burst formation, and to determine the degree to which this

occurs, the effects of IGF-1 were titrated (3x 1 O-'' M to 3x1 O-' M)on erythroid progenitor

cells from a total of 8 normal individuals and 10 patients with PV. Dose-response curves

of IGF-1 alone were compared to dose-response curves of IGF-1 plus 3x 1 0 " ~ M IGFBP- 1 .

A representative IGF-I dose-response curve with and without IGFBP-1 is given in Figure

3A for normal and Figure 3B for PV. In the normal (Figure 3A), there is no significant

effect on erythroid burst formation with 3 x 1 O-'* M IGFBP- 1 alone as compared to medium

alone (0.2>p>0.1). When titrating IGF-1 alone, the first significant increase in burst

activity above background occurred at 3 x 10'1° M IGF-1. However, when titrating the

effects of IGF-1 in the presence of IGFBP-1, the first significant increase in erythroid burst

formation occurred at 3x 10-l2 M IGF-1. Thus IGFBP- 1, which appeared to have no

erythropoietic activity on its own in the normal, was able to increase the responsiveness of

normal erythroid progenitor cells to IGF-1 by at least 10-fold. Half maximum values were

calculated for each dose-response curve with (7 .5~10* '~ M) and without ( 1 . 2 ~ 1 O-'' M)

3x 1ûL2 M IGFBP- 1. We determined that the presence of IGFBP- 1 in culture was able to

cause a left shift in the IGF-1 dose-response curve in the normal, indicating that it was able

to increase the sensitivity of normal erythroid progenitor cells to IGF-1 by about 16-fold.

In erythroid progenitor cells fkom patients with PV (Figure 3B), we observed that

with 3x l 0*12 M IGFBP- 1 alone, there was significant stimulation of erythroid burst

formation compared to medium alone. When we titrated the effect of IGF-I alone, the

first significant increase in burst activity occurred at 3x 1 O-'* M IGF-1, with a half-

maximum value for the dose-response curve being approximately 2 . 4 ~ 1 0-12 M. Compared

with nomal, this would indicate that erythroid progenitor cells in PV are approximately

200-fold more sensitive to IGF-1 than their normal counterparts. in the presence of 3x 10- 12 M IGFBP- 1, the first significant increase in erythroid burst formation above background

occurred at 3x1 0'14 M IGF-1, with a half-maximum value for the dose-response curve

calculated at about 6x 1 O-'' M. This indicates that IGFBP- 1 is able to potentiate the action

of IGF-I by approximately 40-fold. These data indicate that IGFBP-1 is able to

powerfully potentiate IGF-1 activity with respect to erythroid burst formation by

increasing the sensitivity of erythroid progenitor cells to IGF-1 in both nomal individuals

and patients with PV. The degree of stimulation, between 10 and 100-fold, appears to be

similar in both normal and PV, with PV perhaps being approximately twice as sensitive to

IGF-1 in combination with IGFBP- I .

Mimicking the IGF-1 hypersensitivity of PV with IGFBP-I

As IGFBP-I was able to "sensitize" normal erythroid progenitor cells to IGF-I and

thus potentiate IGF-I induced erythropoietic activity, we wished to determine if we could

rnimic one feature of the PV phenotype, i.e. that of hypersensitivity to IGF-1. with IGFBP-

1. Figure 4 illustrates the IGF-1 dose-response curve with and without IGFBP- 1 for

progenitor cells fiom normal individuals and patients with PV. Cornparison of these

curves shows that 3x10''~ M IGFBP- 1 in the normal caused a left shift which allowed its

dose-response curve to essentially overlap the PV curve (without IGFl3P-1). Thus we

demonstrated that IGFBP- 1 could render normal erythroid progenitor cells as

hypersensitive to IGF-1 as PV erythroid progenitor cells. It should be noted that under the

sarne conditions, the IGF-I dose-response curve in PV undenvent a sirnilar left shifi. The

data thus show that PV erythroid progenitor cells, despite being already hypersensitive to

IGF-1, can still be made to substantially augment their response to IGF-I by exposure to

IGFBP- 1 .

DISCUSSION

It has previously been shown that IGF-I is a stimulator of erythropoiesis; it cm

replace erythropoietin for the production of red blood cells in vivo in anephric patients74,

and in culture for the production of erythroid colonies. In the latter case, a molar

concentration of IGF-1 approximately 100-fold greater than that of erythropoietin was

required to achieve the sarne level of activation? It had previously been shown in our

laboratory that erythroid progenitor cells from patients wirh PV are hypersensitive to IGF-

I with respect to their ability to induce erythroid burst formation in strictly serum-fiee

cu~ture'~ That is, PV erythroid progenitor cells require a lower concentration of IGF-1 in

order to elicit erythroid burst formation in culture. For example, where 3x10'" M IGF-1

was permissive for erythroid burst formation in PV, normal burst formation did not occur

until a concentration of approximately 3x 1 O*'' M IGF-1 or higher was reached. In the

current study, we have confirmed these findings and extended them to show that PV

erythroid progenitor cells are hypersensitive to a combination of IGF-1 and IGFBP-1.

When we exarnined the ability of IGFBP- 1 to promote erythroid burst formation,

we found that in PV concentrations as low as 3x10-'' M could be stimulatory, but in the

normal, IGFBP- 1 alone did not have an effect, even at a relatively high concentration.

Assuming that the microenvironments of the normal and PV cultures were identical, this

meant that either i) IGFBP- 1 alone could stimulate both normal and PV erythropoietic

activities. but the concentration of IGFBP- 1 required to stimulate normal erythroid

progenitor cells had not been reached in the present expenment, or that ii) IGFBP- 1 could

stimulate erythroid burst formation only in P V As even 3x 10" M IGFBP- 1 was not

stimulatory in the norrnai, according to the first hypothesis, this would mean that PV

progenitor cells would have to display a hypersensitivity to IGFBP-1 more than five orders

of magnitude greater than that of the normal; this seems unlikely. As for the second

hypothesis, it remains a possibility, or else we must conclude that the normal and PV

rnicroenvironments in culture are not the same. IGF-1 is pnmarily produced in the liver in

response to growth hormone (GH) which is able to modulate its levels in the ~irculation"'~

However, IGF-1 is also produced by a variety of ce11 types including myeloid ~ e l l s ~ ~ ~ and

its synthesis rnay be induced by cytokines such as IL-1 16', m l 6 ' , CSF-1 and IL-3'62,

not only GH. Although IGFBP-1 has been shown to bind to the a@!-integrin receptor in

an IGF-1 independent manner to stimulate migration of CHO c e l ~ s ' ~ ~ , there is no evidence

that we are aware of to suggest that IGFBP- 1 induces a rnitogenic effect in the absence of

IGF- 1. Thus, the simplest explanation for the stimulatory effects of IGFBP- 1 alone in PV

culture is to suggest that PV PBMNC produce endogenous IGF-1 which is secreted into

culture and that this, together with the exogenous IGFBP- 1, induced erythroid burst

formation. Therefore, despite evidence that circulating levels of IGF-1 in PV are

nomal14*, the arnounts of IGF-1 in the hematopoietic microenvironment, which can be

modulated by IGFBPs, also known to be produced by certain myeloid ~ells '~' , may be

sufficient to bring about the observed stimulation.

Figure 2 illustrates that when a fixed dose of IGF-1 is given with a varying dose of

IGFBP- 1, IGFBP- 1 is able to potentiate the action of IGF- 1 on erythroid progenitor cells

to produce erythroid burst, in the nonnal as well as in PV. We found that maximal burst

formation occurred in the presence of IGF-I when the ratio of IGF-1:IGFBP-1 was

approximately equimolar up to a point where IGFBP- 1 was in approximately 10-fold

molar excess. However, when IGFBP-1 was present in greater that 10-fold molar excess,

it was no longer able to potentiate IGF-1 action in the nonnal; in PV, it actually had an

inhibitory effect on IGF-I induced erythropoietic activity, reducing the number of bursts

below levels seen for IGF-I alone. We interpreted this to mean that both IGF-1 and

IGFBP- 1 are required for maximal stimulation. When IGFBP- 1 was in more than 100-

fold rnolar excess, IGFBP- 1 was no longer stimulatory for erythroid burst formation. The

1 : 1 ratio would suggest that a putative IGF-VIGFBP- 1 complex is required for action.

Also, in the range where IGFBP-1 was stimulatory, PV erythroid progenitor cells were

hypersensitive to the putative IGF-I/IGFBP-I complex.

As IGFBP- 1 was able to potentiate the effect of 3x10*'' M IGF-1 and thus allowed

erythroid burst formation to occur in the normal, we wished to determine if IGFBP-1 was

able to increase the sensitivity of erythroid progenitor cells to IGF-1. We titrated IGF-I in

the presence of a fixed concentration of IGFBP- 1. Figure 3 demonstrates that the

presence of IGFBP- 1 was able to increase the sensitivity of erythroid progenitor cells in

both normal and PV. Thus, it appears that IGFBP-1 stimulates IGF-1 mediated

erythropoietic activity by reducing the cellular requirement for IGF-I. How it is able to do

this is at the moment unclear

It had previously been suggested that a putative complex could facilitate the

interaction of IGF-1 with its receptor, or perhaps work through an as yet undiscovered

IGFBP- I receptor. Both of these hypotheses require that IGFBP- 1, directly or through

IGF-1, bind to the ce11 surface in order to potentiate the mitogenic action of IGF-I'*~.

Although IGFBP-1 has been shown to potentiate the rnitogenic action of IGF-1 on human

fibroblasts and BHK-2 1 cells, radiolabelled IGFBP- 1 alone or in complex with IGF-I was

not found to associate with these ce~ls'~'. Aiso, if a putative complex were indeed the

active agent, then factors which facilitated complex formation, such as IGFBP- I affinity

for ligand, should increase the mitogenic activity of IGF-1 in combination with IGFBP- 1.

However, this is not the case. It was found that those IGF-1 analogs in which residues in

the B-region of IGF-1 are substituted with the analogous residues in the B-chah of insulin,

which does not bind IGFBPs, have normal affinity for IGF-I receptor, but result in

approximately 100-fold reduced afinity for IGFBPS'? These analogs which had a

reduced affinity for IGFBP- 1 were more potent than IGF-1 in stimulating DNA synthesis

of porcine aortic smooth muscle cells.

Also, several other pieces of evidence suggest that the IGFBP- 1 affinity for IGF-1

plays a role in its biological activity. Two isofoms of IGFBP-1 isolated £Yom human

arnniotic fluid revealed that one was able to potentiate the eEects of IGF-1, while the other

was inhibit~ry'"~ Cultured human endometrial stroma1 cells from proliferative phase

endometrium secrete a non-phosphorylated isoform of IGFBP- 1, while cells from the

secretory non-proliferative phase endometnum secrete a heavily phosphorylated IGFBP-

1 lu. It is known that the phosphorylated IGFBP-1 binds IGF-1 with a higher af£inity than

its non-phosphorylated isofor~n'~'. Thus, a model suggesting indirect action of IGFBP- 1

is more likely. Experimentai evidence suggests that forms of IGFBP- 1 which have a low

affinity for IGF- I have the greatest biological activity. From a synthesis of these data, we

cm construct a model for IGF-IAGFBP- I action. An isoform of IGFBP- 1 which had a

high affnity for IGF-1 would effectively inhibit IGF-1 activity by preventing its release

from the IGF-VIGFBP- I complex and thus its subsequent interaction with the IGF-I

receptor. IGFE3P- 1, which binds IGF-I less tightly, may be able to retain complexed but

still biologically active IGF-1, in that this fom of IGF-1 has not been taken up by ce11 or

destroyed, and gradually release it to the receptor. In this way, the "complex" would

extend the half-life of active IGF-1 in culture, thereby potentiating its action.

Knauer et ai167, using mitogenic response of fibroblasts to epidermal growth factor

(EGF) as a model, have suggested that the response to ligand is linearly related or

proportional to the steady state number of occupied receptors. They demonstrated that

the continued occupancy of the receptor after a relatively short cornrnitment time could be

the rate-Iimiting step in mitogenesis. Blum et a116' have since s h o w that the continuous

introduction of free IGF at a low dose (O. lng/ml/hour) stimulated DNA synthesis more

efficiently than a single one-time dose of a high concentration of IGF-1. Although the

total IGF-1 adrninistered over time was IO-fold less than that given in the single dose, they

both produced the sarne degree of stimulation. The notion of IGFBP-1 binding and

releasing IGF-I to target cells over time is supported by Our curent work. This model

may also help to explain the negative effects of IGFBP- 1 in large excess. A large molar

excess of IGFBP- 1 may scavenge the free IGF-I, and when it is released, the probability of

its binding to another free IGFBP- 1 rnolecule would be greater. Thus, IGF-1 would be

prevented from interacting with the IGF-I receptor. Theoreticaily, in such a model,

IGFBP-1 would reversibly bind IGF-1 and over time deliver it to the IGF-1 receptor in

minute quantities, providing conditions for steady activation of the IGF-I receptor.

Activation of the IGF-1 receptor leads to autophosphorylation of the receptor B subunit on tyrosine residues. These events are critical for subsequent phosphorylation of

intracellular substrates and signal transduction. Steady and continous activation of the

IGF-I receptor by small physiologically active doses of IGF-I would lead to a receptor

which would appear to be hyperphosphorylated on tyrosine residues. This is consistent

with Our previous worklLg which demonstrates a hyperphosphorylated IGF-1 receptor in

PV.

Such a mode1 for the biologicai action of IGFBP-1 in hematopoiesis, coupled with

the finding that the levels of circulating IGFBP-1 are elevated in patients with PV"* and

the fact that our current study indirectly suggests that PV PBMNC may produce

endogenous IGF-1 in culture, rnay, at least in part, provide a mechanism for the observed

hypersensitivity of PV erythroid progentor cells to IGF-1. The fact that we were able to

mirnic the PV phenotype with normal erythroid progenitor cells and IGFBP- 1 (Figure 4)

suggests that this may at least in theory be true. However, it should be noted that under

the sarne conditions, PV erythroid progenitor cells were more sensitive to the combination

of IGF-1 and IGFBP- 1 than their normal counterparts, suggesting that there may also be

another cause for the IGF-1 hypersensitivity in PV.

Our work provides strong evidence for the role of IGF-1 and at least one of its

binding proteins, IGFBP- 1, in normal and PV erythroid burst formation in culture.

Dysregulation of the IGF-VIGFBP- 1 regdatory pathway may thus lie at the heart of a

mechanism for the pathogenesis of PV.

Figure 4.1 - Action of IGF-I alone compared to IGFBP-1 alone on erythroid progenitor

cells fiom normal individuais and patients with PV. Basal Sero-Zero stem ce11 medium

optimized for erythroid burst formation in semi-solid methylcellulose culture was used to

examine erythroid burst formation induced by medium alone (Negative Control), medium

plus IGF-1 (3x 10" ' M), medium plus three different concentrations of IGFBP- 1 (3x 1 O-"

M, 3 x 1 O"* M, and 3 x 1 O-' M). The number of erythroid burst component colonies for 5

normal individuals and 5 patients with PV were normalized as percent of maximum

erythroid burst production for each individual. The mean +/- standard error of the mean

Bras determined for the percent maximum values of al1 normals and al1 patients and

presented.

IGF-1 IGFBP-1 IGFBP-1 IGFBP-1 Control (3xlO-IIM) (3x10-14M) (3x10-12M) (3s 1 O - 9 ~ )

Figure 4.2 - Titration of IGFBP- 1 From 3x 1 O-'' M to 3x 1 o - ~ M in the presence of 3x 1 O-''

M IGF-1. The figure illustrates the number of erythroid burst component colonies as

percent maximum values versus the molar concentration of IGFBP-1. Representative data

obtained for cells of nomal individuals (0) and patients with PV are shown. The

solid lines represent mean erythroid burst formation with 3x10-"M IGF-1 alone for the

normal (N) and PV, respectively. The half-maximum for normal occurs at -3xl O-'' M

while the half-maximum for PV occurs at -6x 1 O-'* M. Thus, with respect to burst

formation, PV erythroid progenitor cells are about 50-fold more sensitive to the

combination IGF-1 with IGFBP-1 than normal. Burst formation in the absence of IGF-1

appears to be submaxirnai. This experiment was performed for a total of 5 normal

individuais and 5 patients with PV.

Figure 4.3 - Titration of IGF-1 activity fiom 3x 1 0'14 M to 3x 1 O-' M in the presence of

3x 1 0-l2 M IGFBP- 1. Percent maximum erythroid burst formation of a representative

normal individual (A) and a patient with PV (B) are given. The solid line represents mean

erythroid burst formation in the presence of 3x 1 0*12 M IGFBP- 1 alone. The data illustrate

that IGFBP- 1 is able to sensitize erythroid progenitor cells to IGF-1, allowing even small

concentrations of IGF-1 to be effective in burst formation. In the nomai (A), the binding

protein was able to shifi the half-maximum of the dose-response curve fiom - 1 2 x 1 O-'" M

to 7 . 5 ~ 1 O-'' M, increasing the sensitivity of normal erythroid progenitor cells by - 16-fold.

In PV (B), IGFBP- 1 shifts the haif-maximum fiom - 2 . 4 ~ 1 O-'* M to -6x 1 0-14 M, increasing

the sensitivity of erythroid progenitor cells by about 40-fold. This experiment was

performed for a total of 8 normal individuals and 10 patients with PV.

Percent Maximum Bunt comp4nent Colonies Percent Maxlmum Burst Component Colonies

Figure 4.4 - Plot of the number of erythroid burst component colonies as a percent

maximum versus the molar concentration of IGF-1 with and without IGFBP- 1 . The

presence of IGFBP- 1 cause a lefi shifi in the IGF-I dose-response curve of the normal

which essentially overlaps the PV dose-response curve with IGF-I alone. Thus, the PV

phenotype can be mirnicked by providing normal cells with IGFBP-I .

Molar Concentration of IGF-I

CHAPTER 5

CHAPTER 5

GENERAL DISCUSSION

POYLCYTHEMIA VERA: FOCUSING THE SEARCH FOR A DEFECT

Polycythemia vera (PV) is a human chronic myeloproliferative disorder with no

obvious cause". It is beiieved that in this condition an aquired somatic stem ce11 mutation

leads to changes that perturb the behavior of the ce11 such that the stem ceil becomes

dominant, eventually producing most or al1 peripheral blood cells, with the possible

exception of lymphocytes; and the proIiferation/differentiation of the stem ce11 is

quantitatively abnormal, generating greatly increased numbers of red blood cells with a

concomitant rise in neutrophils and platelets. The cellular and molecular basis of these

aiterations remains tantalizing, but poorly understood.

The work of Prchal and ~xelrad" demonstrating exythropoietin-independent

colony formation by PV bone marrow cells in plasma culture was the first hnctional

hematopoietic defect ascnbed to the PV phenotype. The involvement of erythropoietin in 59,168. the development of these colonies has essentially been niled out , however, other

factors may play a aiticai role8'. Subsequent work in Our laboratory has provided

convincing evidence that the erythropoietin-independent colonies were due to an

approximately 100-fold increase in the sensitivity of PV erythroid progenitor cells to IGF-

1. It was shown that monoclonal antibodies directed against the IGF-1 receptor were able

to abrogate this hypersensitivity'g, in a way converting PV cells to the normal phenotype.

Therefore, there is strong evidence that IGF-1 and/or the IGF-1 receptor signal

transduction pathway play a key role in the pathogenesis of PV.

In the course of the current study, we undertook a systematic investigation of

elements involved in both extnnsic and intrinsic regdation of the IGF-1 pathway in

patients with PV in cornparison to nomal individuals. It was hoped that any differences

found would help to shed light on defects that could give nse to the PV phenotype.

As cells in the hematopoietic microenvironment are exposed to numerous

cytokines, the specificity of cellular responses is dictated, to a great extent, by

transmembrane receptors. These receptors specificaily bind ligand, thus generating a

cascade of intracellular signals to modulate cellular a~tivities'~'. The receptor for IGF-1 is a

member of the tyrosine kinase receptor fdyPw and is highly homologous to the insuiin

recepto?596. The human IGF-1 receptor cDNA consists of a 4 10 1 nucleotide open reading

fiame which encodes a 1367 amino acid protein170. This 18OkDa receptor precurso? is then

giycosyiated, dimerized, and proteolyticaiiy processed to yieid the mature a242

heterotetramer considng of two extracellular ligand binding a chains disulphide linked to two

p chains that span the membrane once and contain the intrinsic tyrosine specific protein kinase

activityg5. Extracellular ligand binding to the receptor stimulates its i n t ~ s i c tyrosine specifk 98S.lO 1 protein kinase activity, which lads to P subunit autophosphorylation

Autophosphorylation of the cytoplasmic domain of the P subunit leads to a dramatic increase in 102-105 its kinase activity and its subsequent phosphorylation of intracellular substrates 1o0.106.107 . It

108-1 10 is thus be!ieved to have an essentiai role in signal transduction .

As the hypersensitivity exhbited by PV erythroid progenitor cells was manifested

through the IGF-1 receptor5', we wished to investigate tyrosine phosphorylation of the IGF-1

receptor in PV versus normal. We have demonstrated1lg that in PV, tyrosine phosphorylation

of the IGF-I receptor P subunit occurs and is increased in the absence of added ligand as

compared to normal individuals and patients with secondq erythrocytosis. Also, in the

presence of ligand, the induction of tyrosine phosphorylation by IGF-1 in PV occurred at an

approhately 100-fold Iower ligand concentration than in the normal. In other words, the

receptor in PV was found to be hypersensitive with respect to tyrosine phosphorylation. Thus,

our findings are consistent with, extend, and provide a molecular bais for the observed

hypersensitivity of erythroid progenitor cells in PV. An abnormal increase in the activity of

receptor tyrosine kinases has been associated with umegdated growth"l. Studies indicate that

an increased tyrosine kinase activity - attnbuted either to an increase in the number of tyrosine 112-1 14 kinase receptors or their constitutive a~tivation"~ - elicits increased signalhglll'lls Such

a defect in a myeloid stem ceil in PV could result in increased ceil proliferation and the

observed erythrocytosis, granulocyt osis, and thrombocytosis whic h are c haractenstics of the

PV phenotype. However, whether the enhanced tyrosine phosphorylation of the IGF-1

receptor p subunit observed in PV, with and without IGF-1, represents an extrinsic modulation

of the receptor by either increased extraceiiular IGF-1 activity or is due to a constitutively

activated receptor tyrosine kinase was not known. Furthemore, whether the observed

increase in tyrosine phosphorylation actuaüy represents an increase in biological activity of the

IGF-1 receptor in PV stiii remained to be formaily established. Answering these questions

would be crucial in establishing the importance of this phenornenon.

IGF-I and its receptor are potentially very imporiant in the pathogenesis of PV.

However, the circulating levels of IGF-I in PV had not been investigated. But, there is

normally very little free IGF-1 in the circulationlu. Most of the IGF-1 circulates bound to

specific high affinity binding proteins. Therefore, in order to investigate IGF-1 activity in

the circulation, we would have to also investigate the circulating levels of the binding

proteins. IGF binding proteins (IGFBP-1 to -6) are homologous but distinct and are

known to modulate both the availability of IGF-1 to target tissues and the activity of IGF-1

in extracellular f l ~ i d s ' ~ ~ . The binding proteins exhibit both tissue and developmental

specificity with respect to their expression, and thus are thought to play an important

physiological role in IGF transport, localization, and action. IGFBP- 1 through -4 are

found in the circulation. Therefore, we examined the level of circulating IGF-1 as well as

two of its binding proteins, IGFBP- 1 and IGFBP-3, in normal individuals and compared

them to those in patients with PV. At the time, IGFBP-2 and IGFBP-4 could not be

investigated for technical reasons. We found that aithough the circulating levels of IGF-I

and IGFBP-3 in PV were normal, the levels of plasma IGFBP-1 in PV (37.80 +/-

4.33 pg/L) were increased more than 4-fold in cornparison to normal (9.34 +/- 1.34pg/L)

or patients with secondary erythrocytosis (9.47 +/-1.96pgL). The significance of this

increase was not known. In the literature, IGFBP-I had been s h o w to have both

stimulatory and inhibitory effects. To determine its effect on erythroid burst formation we

cultured peripheral blood mononuclear cells with IGFBP-1 and IGF-1. We found that

IGFBP- 1 in the presence of IGF-1 was strikingly stimulatory. Dose-response curves of

IGF-I in the presence of IGFBP- 1 revealed that IGFBP- 1 caused a left shifi in the dose-

response curve, indicating that IGFBP- 1 increased the sensitivity of normal and PV

erythroid progenitor cells to IGF-1. Thus, IGFBP-1 rnay play an important physiological

role in the modulation of IGF-I action in normal erythropoiesis. The increase in a positive

modulator of IGF-1 action in PV provides an attractive mechanisrn to account for the

heightened sensitivity of erythroid progenitor cells to IGF-1 in this disorder. These data

also suggest that the sustained high circulating level of this positive regulator could be an

important factor in the constitutiveiy excessive erythropoiesis which is a characteristic of

the PV phenotype. In addition, due to its powehl stimulatory effects, it may be a good

candidate for an extracellular event that results in increased tyrosine phosphorylation of

the IGF-I receptor and thus rnay also help to explain PV hypersensitivity to IGF-I.

However, erythroid progenitor cells in PV are still hypersensitive to the effects of IGF-1

with IGFBP- 1. Therefore, the increase in plasma IGFBP- 1 is insufficient on its own to

account for the whole of the PV phenotype.

Alternatively, a mutation in the IGF-1 receptor or in its signalling pathway, which

rendered the proliferative signal constitutively active in the absence of ligand and hypersensitive

as well as hyperresponsive in the presence of ligand, couid be the acquired abnorrnality of the

PV hematopoietic progenitor clone that leads to the PV phenotype. However, the question of

where mch a defect would lie is open. We found no evidence of a gross abnomalit. in the

IGF-1 receptor P subunit that would be manifest as a change in its relative molecular weightllg.

However, a lesion in the a subunit of the receptor that af5ects its tyrosine kinase aaivity, or a

discrete lesion in the p subunit which may not alter its relative molecular weight, is of course

possible and would not be detected by the methods used in our investigation thus far. Also,

point mutations in the a subunit, could affect the receptor's affinity for ligand and thus increase

PV IGF-1 receptor tyrosine phosphorylation. Therefore, we tested the binding capacity and

affinity of IGF-1 receptor for 1 2 5 ~ - ~ ~ ~ - ~ in erythrocytes, glycophorin A- erythroid

precursor cells, as well as penpheral blood mononuclear cells isolated fiom normal

individuals and patients with PV (Appendix A). We found that binding capacity and

afiïnity for ligand in PV were normal. Therefore, if there was a mutationai defect within

the IGF-1 receptor in PV, it did not appear to affect the receptor's affinity for ligand. To

determine whether the intrinsic tyrosine kinase activity in PV had been afTected by a

putative lesion in the IGF-1 receptor, we examined the ability of IGF-1 receptor to tyrosine

phosphorylate a substrate in vitro (Appendix A). IGF-1 receptors were isolated from the

erythrocytes of normal individuals and patients with PV, and their intrinsic tyrosine kinase

activity assayed. For receptors isolated in the presence of sodium vanadate, an inhibitor of

tyrosine kinase activity, we found that the tyrosine kinase activity of the receptor in PV

was increased in the absence of added IGF-1, and hypersensitive in response to ligand.

These data were consistent with the observed increase in tyrosine phosphorylation of the

IGF-1 receptor and the hypersensitivity of erythroid progenitor cells to IGF-1 in PV.

However, to conciusively determine if this is due to an intrinsic Iesion within the PV IGF-I

receptor, or if it is a reflection of receptor extrinsic events that lead to receptor activation,

we will have to sequence the IGF-1 receptor in PV.

POLYCYTHEMIA VERA: RESPONSE OF ERYTHROID PROGENITOR

CELLS TO CYTOKINES

There is also some evidence from long-term cultures of PV bone marrow cells

indicating that these cells exhibited abnormal proliferation characteristics and cycling

behavior"'. These findings suggested to the authors that PV cells displayed an altered

susceptibility to regulatory factors. Subsequent investigation of this phenornenon by

others has revealed that PV erythroid progenitor cells do have an aitered sensitivity to

certain cytokines.

Granulocyte-macrophage colony stimulating factor (GM-CSF) and Interleukin-3

(IL-3) are factors that are known to play a role in the proliferation of early erythroid

progenitor ~ e l l s ' ~ ~ ' ~ , but normally require a differentiation factor such as erythropoietin

for erythroid progenitor ce11 differentiationl7' de Wolf et als2 have demonstrated that both

GM-CSF and IL-3 enhance the proliferation of erythropoietin-independent colonies in the

absence of added erythropoietin and in the presence of an ami-erythropoietin antibody. In

contrast, normal erythroid progenitor cells were strictly dependent upon added

erythropoietin for the growth promoting effects of GM-CSF and IL-3. These results

indicated that either the erythroid progenitor cells in PV displayed abnormal responses to

GM-CSF and IL-3, or that, as these cells were independent of erythropoietin for their

differentiation, unlike their normal counterparts, they did not require a differentiation

activity in order to be assayed. However, the normal erythroid progenitor cells, which

may have been as responsive to these factors as PV cells, simply could not be assayed

without the differentiation promoting effects of erythropoietin. Further work would have

to be done to determine the relative sensitivities of PV progenitor cells to these and other

factors, and in fact these studies have been camed out.

Dai et ai investigated the response of PV erythroid progenitor cells to IL-3 175.176.

They reasoned that as PV displayed a trilineage hyperplasia30 and as IL-3 was a cytokine

that acted upon early hematopoietic cells in order to provide trilineage growth

enhancement ln, it rnay play a role in the pathogenesis of PV. Using semi-punfied

populations of erythroid progenitor ce~ls'~', they found that whereas the growth of normal

erythroid progenitor cells in the absence of exogenous IL-3 declined by approximately

60% after 48 hours, the growth of PV erythroid progenitor cells only declined by

approximately 30%, thus demonstrating a reduced dependence on IL-3 in PV"' In dose-

response experiments with IL-3, they were able to demonstrate an IL-3 hypersensitivity in

PV in excess of 100-fold in one case17' and approximately 38-fold in a n ~ t h e r ' ~ ~ . Similarly,

they were able to demonstrate an approximately 48-fold increased sensitivity of PV

erythroid progenitor cells to GM-CSF"~.

The direct effects of various cytokines on erythroid progenitor cells in PV are

relatively simple to comprehend when compared to the web of direct and indirect effects,

paracrine and autocnne, which potentially exist for heterogeneous ce11 populations in ce11

culture. Therefore, it should not be surprising that other cytokines such as interleukin- 1

(IL- 1 ) have aiso been shown to have indirect effects on erythroid burst formation in

culture. Using non-adherent penpherd blood mononuclear cells as well as sorted CD34'

cells from normal individuals and PV patients, de Wolf and co-worker~~~' dernonstrated

stimulation of erythroid burst formation with exogenous IL- 1 in the absence of

erythropoietin in PV, but not in normal. However, this effect was abrogated by the

addition of anti-GM-CSF antibodies. Thus, it appeared that the effect of IL- 1 on burst

formation was an indirect one. It was hypothesized that IL4 acted on target cells which

in tum secreted GM-CSF into culture and it was the GM-CSF that was able to potentiate

erythroid burst formation in PV, but not in the normal. These events would be consistent

with existing data on PV erythroid progenitor ce11 hypersensitivity to GM-CSF'~~.

Although IL-3 hypersensitivity in PV, in the range of 2 orders of magnitude, has

been demonstrated, there is evidence that this too may be an indirect effect. Recently,

Arkins et al162 have shown that murine bone marrow cells cultured in the presence of a

variety of cytokines such as colony stimulating factor-l (CSF-l), IL-3, GM-CSF, and

granulocyte colony stimulating factor (G-CSF) each stimulated the production of insulin-

like growth factor4 (IGF-1) mRNA to varying degrees, with the rnost potent stimulation

occurring with CSF-1 and IL-3. These cytokines increased the production of IGF-1

transcnpts by approximately 50-75% over the virtually negligible amounts present in

freshly isolated marrow cells. In contrast, after addition of these cytokines to culture,

there was an approximately 50% dom-replation of IGF-I receptor mRNA dunng

proliferation and differentiation of the bone marrow cultures. These data strongly suggest

that IGF-1 was being produced in culture in response to the cytokines, and was acting on

cells canying the IGF-I receptor. Thus, an argument could be made for IGF-1 induction

of bone marrow ce11 proliferation in response to CSF- 1 and IL-3. This would suggest that

CSF-1 and IL-3 may promote hematopoiesis in much the same way as growth hormone'78- 180 , i.e. through IGF-1. It has previously been shown that hematopoietic ~ e l l s ' ~ ' , and

perhaps more relevant to us, erythroid progenitor celldg2 display a profound requirement

for IGF-1. This casts a new and different light on the role of IGF-1 in hematopoiesis, and

on the observed hypersensitivity of erythroid progenitor cells to cytokines such as GM-

CSF and especially IL-3. If indeed they are working indirectly through IGF-1, then are PV

erythroid progenitor cells tmly hypersensitive to IGF-1 alone?

The colony assay system has made profound and incalculable contributions to the

field of hematopoiesis. But, the assay system we have learned to trust is rife with

complexities both subtle and obvious, which cm make results difficult to interpret. Cell

cultures are drarnatically affected by the purity of ce11 populations under study, as well as

other ce11 populations which are present, but are not being studied at the time, and thus are

often ignored or simply tolerated. At another level of complexity, these cultures are

affected by the direct and indirect action of intentionally added growth factors, as well as

the presence of unknown quantities of unintentionaily added growth factors present in

reagents used to support colony growth. These events reflect the iudden complexity of

hematopoiesis in culture, and perhaps hematopoiesis in general.

From a synthesis of the current data, we can Say with some degree of certainiy that

PV erythroid progenitor cells are hypersenitive to IGF-1. The finding that circulating

levels of IGFBP- 1 are increased in patients with PV supports the notion that the IGF-1 in

combination with IGFBP- 1 rnay be a key player which modulates this hypersensitivity.

However, the observation that PV cells are hypersensitive to the combination of IGF-1

with IGFBP-1 and the finding that the IGF-1 receptor itself was hyperphosphorylated on

tyrosine residues in the absence of exogenous ligand, and hypersensitive as well as

hyperresponsive with respect to tyrosine phosphorylation in the presence of ligandiL9,

would suggest a defect in the IGF-1 signal transduction pathway. A number of other

functional PV defects such as thromboxane A2 receptor ~ i ~ n a l l i n ~ ' ~ ~ , as well as

significantly reduced diacylglycerol (DAG) production by polymorphonuclear

granulocytes184, would also suggest a defect(s) in one or more aspects of PV signal

transduction, while other pathways are

POLYCYTHEMIA VERA: THE POSSIBLITY TEMT PHOSPEIATASES PLAY A

ROLE

The coordinate action of cytokines is mediated t hrough transmembrane receptors.

Upon binding ligand, these growth factor receptors regulate protein function through

tyrosine phosphorylation of intracellular substrates by the now activated receptor's

intrinsic or recruited tyrosine kinase activity. Thus, receptor tyrosine kinases, as well as

receptor associated tyrosine kinases, play an important role in the control of ce11 100.186 proliferation and differentiation dunng hematopoiesis . Equally important are the

protein tyrosine phosphatases which serve to dephosphorylate these intracellular 187.188 substrates , and are thought to be critical in maintairing the homeostatic balance

required for efficient tyrosine phosphorylation-mediated signalling. Intracellular proteins

may contain Src homology 2 (SH2) domains which bind phosphotyrosine residueslg9. In

this way, phosphotyrosine residues serve to recruit SH2 domain-containing proteins such

as tyrosine phosphatases to their tyrosine phosphorylated targets, thus providing

specificity of actionlgO for modulation of the phosphotyrosine-mediated signal transduction

pathway'gl.

The subfamily of cytoplasmic protein tyrosine phosphatases which contain S H 2

domains include the Src homology protein tyrosine phosphatase 1 (SH-PTP 1), also known

as hematopoietic ce11 phosphatase ( H C P ) ' ~ ~ or SHP"~, Src homology protein tyrosine

phosphatase 2 (SH-PTPZ), also known as ~ ~ p ' ~ ' , and the Drosophila homolog of SH-

PTPZ the corkscrew (csw) gene product195. SH-PTP2 and corkscrew appear to be

ubiquitous in their distribution, interacting with receptor tyrosine kinases to potentiate 1%-199 their signai transduction . In contrast, SH-PTP 1 is pnmarily expressed in cells of

hematopoietic origin. Yi et ai demonstrated that SH-PTPI (SHP) associates with the IL-3

receptor and overexpression of the anti-sense SHP cDNA resulted in increased

proliferation of cells in response to ~ - 3 ~ ~ ~ . Thus, in contrast to SH-PTP2, SH-PTP 1

appears to be a negative rnodulator of signal transduction, involved prhxily in receptor

desensitizationM1. SH-PTP 1 has aiso been s h o w to associate with the stem ce11 factor

(SCF) receptor d e r ligand stirnulated tyrosine autophosph~r~lation~~~, and thus rnay also

regulate its activity.

Consequently, there is strong evidence to suggest that SH-PTP 1 plays an

important role in the negative regulation of cytokine receptors. One of the most

rigorously studied has been the erythropoietin receptor. The erythropoietin receptor lacks

an intrinsic tyrosine kinase activity, but ligand binding tnggers recmitment and tyrosine

phosphorylation of Janus kinase 2 (JAK2) as well as phosphorylation of the erythropoietin

receptor itselP3. Klingmuller et al2'' demonstrated that there was erythropoietin induced

recruitment of SH-PTP 1 to the erythropoietin receptor. The authors believed that

tyrosine residue 429 was responsible for binding the SH2 domain of the phosphatase as

site-directed mutation which converted the tyrosine to a phenylalanine prevented the

recmitment and interaction of SH-PTP 1 with the erythropoietin receptor. In 32D cells

carrying the mutant (tyrosine 429+phenylaianine 429) receptor, tyrosine phosphorylation

of the erythropoietin receptor was persistent and remained high for a longer penod of

tirne. Also interesting, particularly with respect to cytokine sensitivity, was the

observation that erythropoietin dependent growth was achieved with 1/10 the level of

erythropoietin required to stirnulate 32D cells transfected with the wild type erythropoietin

receptor. Thus, a defect in a hematopoietic cell phosphatase cm lead to cytokine

hypersensitivity . We have demonstrated a hyperphosphorylated IGF-1 recept or' '' and

increased sensitivity to IGF-1'' as well as IGF-1 in conjunction with IGFBP-1'" in PV.

This raises the strong possibility that a functional defect in a phosphatase may lie at the

heart of the PV phenotype.

SH-PTP1 also appears to play a critical role in negative regulation of a number of

hematopoietic signalling pathways. Mutations in the murine SH-PTP 1 gene (PTP 1 C) are v 205,206 responsible for the phenotype known as motheaten (me) or viable motheaten (me) .

Mice homozygous for these mutations fail to express a Functional PTP 1C and display 205-207 severe abnormalities involving al1 hematopoietic ce11 lineages . Macrophages fiom

meme mice hyperproliferate in response to CSF- 1 and the CSF- 1 receptor in these cells

was found to be hyperphosphorylated in response to ligand2os, recailing once again the

hyperphosphorylated IGF-I receptor in PV"~. Also, of interest to the study of PV was the

finding that these mice displayed increased numbers of the bipotential myeloid progenitor

cells, granulocyte-macrophage colony-forming units (CRI-GM), as well as increased

numbers of erythroid progenitor cells (CFU-E) which displayed hypersensitivity to

erythropoietin. It would be interesting to know if meme erythroid progenitor cells display

a hypersensitivity to IGF-1.

Thus several aspects of the PV phenotype are at least theoreticdly reflected in the

motheaten mouse, and are consistent with a defect in a tyrosine phosphatase. Therefore,

future investigations dong these lines may prove to be fniitful. Indeed, at a recent

meeting of the Arnerican Society of Hematology, a paper presented by LoCicero et al2''

suggested that the Ievel of the SH-PTP 1 protein, but not mRNA was decreased in

peripheral blood mononuclear cells of PV patients. At the same meeting, Hinshelwood et

al"' demonstrated that the SH-PTP 1 protein in PV was intact. Therefore, a putative

phosphatase defect in PV rnay lie, not within the phosphatase itself, but within elements

that regulate its activity post-transcnptionally.

CONCLUSION

The pathogenesis of PV is noteworthy for a number of reasons: first, and perhaps

foremost, because the condition is a hematological disorder which affects the Iives of

approximately 1 in 100,000 individuals; second, investigation of the PV defect rnay

provide insight into a fundamental biologicai process i-e. the mechanism by which a stem

ce11 can give rise to multiple lineages which are phenotypically very different fiom the

parent and fiom each other. 1 sincerely hope that my work has provided some insights

into the PV phenotype as well as modulation of IGF-I action. It has raised many

questions, some of which may even be worth pursuing, not only because answenng thern

rnay help ameliorate the condition of PV patients, but for the fundamental insights that the

answers may provide with regard to hematopoietic developrnent.

APPENDIX A

Increased tyrosine kinase activity of the insulin-like growth factor 1 receptor in

Polycythernia veraJ

Amer M. Mirza, and Arthur A. Axelrad

Dept. of Anatomy and Ce11 Biology

University of Toronto, Toronto, Ontario

Canada

- -

a This work was supponed by a grant from the Medical Research Council of Canada (MA3969).

106

Polycythernia vera (PV) is a myeloproliferative disorder in humans in which

erythroid progenitor cells exhibit increased sensitivity to insulin-like growth factor I (IGF-

1). We had previously s h o w that penpheral blood mononuclear cells isolated fiom

patients with PV exhibited an increase in tyrosine phosphorylation of the IGF-I receptor.

We reasoned that the increase in tyrosine phosphorylation could be a result of an intrinsic

lesion within the IGF-I receptor in PV cells, either affecting its ability to bind ligand, or

increasing the tyrosine kinase activity of the IGF-I receptor itself A gross mutation in the

IGF-I receptor was essentially niled out as the mobility of the IGF-1 receptor by SDS-

PAGE in PV was normal. This however did not rule out lesions within the receptor, such

as a point mutation, which were beyond the resolution of this technique. We tested the

binding capacity and affinity of IGF-1 receptor for L 2 5 ~ - ~ ~ ~ - ~ in erythrocytes, glycophorin

A- erythroid precursor cells, as well as peripheral blood rnononuclear cells (PBMNC)

isolated 6om normal individuals and patients with PV. We found that the average number

of binding sites decreased from the PBMNC through differentiation to the RBC in both

PV and normal but the binding capacity and affinity for ligand in PV were normal.

Therefore, if there was a mutational defect within the IGF-I receptor in PV, it did not

appear to affect the receptor's affinity for ligand. To determine whether the intrinsic

tyrosine kinase activity in PV had been affected by a putative lesion in the IGF-1 receptor,

we exarnined the ability of the IGF-I receptor to tyrosine phosphorylate a substrate in

vitro. IGF-1 receptors were isolated fiom the erythrocytes of normal individuais and

patients with PV, and their intrinsic tyrosine kinase activity was assayed. Receptors

isolated fiom PV and normal cells in the absence of sodium vanadate, an inhibitor of

tyrosine phosphatase activity, showed no difference with respect to their intrinsic tyrosine

kinase activity. When receptors were isolated in the presence of sodium vanadate, we

found that the tyrosine kinase activity of the IGF-1 receptor in PV was increased in the

absence of added IGF-1, and markedly hypersensitive in response to the ligand. These

data are consistent with the increase in tyrosine phosphorylation of the IGF-I receptor in

PV that we observed earlier and they suggest an explanation for the hypersensitivity of

erythroid progenitor cells to IGF-1 in this disorder.

The molecular basis for the increased tyrosine kinase activity of the IGF-I receptor

in PV remains to be elucidated. We hypothesize that it could be the result of either a

mutation in the IGF-1 receptor itself which has a direct or indirect affect on its kinase

activity, or a diminuation of the activity of a putative hematopoietic tyrosine phosphatase

which is recruired to the IGF-I receptor to down-regulate its activity.

Figure A.1 - Scatchard analysis of binding data of 1 2 5 ~ - ~ ~ ~ - ~ on erythrocytes isolated

from normal individuals and patients with PV. Erythrocytes were isolated from a total of

9 normal individuals and 9 patients with PV. Data are presented from 3 normal

individuals and 3 patients with PV for the concentration of bound ' U ~ - ~ ~ ~ - ~ over free

versus bound in h o l e s per 2x10' cells. The data indicate that there was no difference in

IGF-1 receptor affinity for l Z 5 ~ - I ~ ~ - I on erythrocytes of normal individuals compared to

those isolated from patients with PV.

Scatchard Analysis of 1251-IGF-I Binding to Erythrocytes

PV

0 Normal

Figure A.2 - Cornparison of the affinhies of IGF-1 receptors found on normal and PV

glycophonn A positive erythroid precursor cells. The cells were produced in serum-fiee

mass culture seeded with peripheral blood mononuclear cells (Appendix B). Data from

binding studies performed for 4 normal and 4 PV cultures are presented as a Scatchard

plot of the concentration of bound 1 2 5 ~ - ~ ~ ~ - ~ over fiee versus the concentration of '*'I-

IGF-I bound to 1x10~ cells. These data indicate that both the binding capacity and the

affinity For ligand of the IGF-I receptor on these cells was sirnilar in PV cells compared to

normal.

Scatchard Analysis of 1251-IGF-1 Binding to Erythroid Precursors

I 1 I I 1 1 I 4 I I

0.03125 0.06250 0.12500 0.25000 0.50000 1.00000 2.00000 4.00000 8.00000 16.00000

Bound

Figure A.3 - Data fiom 1 2 5 ~ - ~ ~ ~ - ~ binding studies on peripheral blood mononuclear cells

(PBMNC) fiom 4 normal individuals and 4 patients with PV are s h o w as a Scatchard

plot. Data fiorn 2 normal individuals and 2 patients with PV are given as bound 1 2 5 ~ - ~ ~ ~ - ~

over fiee versus bound 1 2 5 1 - ~ ~ ~ - ~ in holes for 2x 10'cells. The data indicate that the 12'1-

IGF-I binding capacity on PBMNC in normal and PV are sirnilar and the IGF-1 receptor

on these cells have equivaient affinities of IGF-1.

Scatchard Analysis of 1251-IGF-1 Binding to PBMNC

0.30-

0.25-

0.20-

C) U L

$ 0.15- E 3

Cs 0.10-

0.05 -

0.00 1

1 1 I 1 1 I 1

2 1

4 8 16 32 64 128 256

Bound

Figure A.4 - Measurement of the intrinsic tyrosine kinase activity of the IGF-1 receptor in

receptor of normal individuals and patients with PV. The intrinsic activity of IGF-I

receptors in response to increasing concentrations of IGF-1 was assessed on isolated IGF-1

receptors isolated fiom erythrocytes in the absence of sodium vanadate for 4 normal

individuals and 4 patients with PV in an in vitro kinase assay. The amount of y - 3 2 ~

transferred to a polyglycine-polytyrosine substrate (Gly (8O):Tyr (20)) was measured with

scintillation counting. The activity for a particular concentration of IGF-I(3x 1 O-" to

3x 10.~ M) is given in counts per minute. The data demonstrate that the tyrosine kinase

activity increases with increasing concentrations of IGF-1 and appeared to be sirnilar for

normal and PV IGF-I receptors.

IGF-1 Receptor Tyrosine Kinase Activity

a PVI PV2

v PV3 PV4

0 NI N2

A N3 v N4

Figure A S - Intnnsic tyrosine kinase activity of the normal and PV IGF-I receptor. IGF-I

receptor was isolated on Ricin II columns by lectin affinity chromatography in the

presence of sodium vanadate. The intnnsic kinase activity in the presence of increasing

concentrations of IGF-I(3x 1 to 3x 1 was measured as counts per minute by

scintillation counting. Data from 3 normal individuals and 3 patients with PV are given as

the mean counts per minute +/- the standard error of the mean. We found that the PV

receptor has an increased basal tyrosine kinase activity in the absence of ligand and there

was a lefi shfl in the IGF-1 dose response curve indicating that the IGF-I receptor in PV is

hypersensitive to IGF-1 with respect to the recept or's tyrosine kinase activity .

IGF-1 Receptor Tyrosine Kinase Activity

APPENDIX B

Kinetics o f red ce11 development in Polycythemia vera:

Studies in a stnctly serum-free liquid culture system5

Amer M. Mirza, Denise Eskinazi, Bernard J. Fernandes*, and Arthur A. kvelrad

Dept. of Anatorny & Ce11 Biology

University o f Toronto

*Dept. of Pathology

Mount Sinai Hospital

Toronto, Ontario

Canada.

5 This work $vas supportai by a grant from the Medical Research Council of Canada (MA3969).

Portions of this work wre presented at the Amencan Society of Hernatology meeting, Orlando. FL. December 1996 (Blood 88 (10) suppl 1, p. 98a).

ABSTRACT

The major clinico-pathological emphasis in Polycythernia vera (PV) is clearly upon

the erythroid lineage. Despite this, the kinetics of red ce11 development in PV has yet to be

elucidated. Currently popular liquid culture systems rely on semm and complex sources

of cytokines to support the growth and development of human erythroid cells in vitro.

Unknown or unmeasured growth factors andor inhibitors in these preparations make

systernatic and repeatable investigations of lineage developrnent difficult, and render

interpretation of findings questionable. Therefore in order to compare red ce11

development in PV and normal, we used Our Basal serozeroM Stem Ce11 Medium

supplernented with 11-3 (10 ng/mL), SCF (50 ng/mL), Epo (3U/mL), Hemin (lOO@J and

ATRA (30 nM). Peripheral biood mononuciear cells from PV patients and normal

individuais were seeded into culture and followed for 14 days. At intervals during the

culture period, cells were harvested, the total number of viable cells was determined,

differential counts were obtained on Giemsa-stained cytospin preparations, and the

component ce11 populations of the cultures were delineated by Bow cytometry. Cultured

cells from PV patients and normal individuals were assessed for their expression of CD3,

CD 19, CD 1 3, CD34, CD4 1, CD45, HLA-DR, and glycophorin A (glycoA) surface

antigens. We were thus able to follow the developrnent of the various ce11 lineages under

strictly semm-fiee conditions. The kinetics of the granulocyte, monocyte, megakaryocyte,

T, B, and stemlprogenitor ceIl populations did not differ in normal versus PV. The

erythroid ce11 populations behaved differently. In normal, while no glycoA- cells were

detected at day 0, by day 13 approximately 16.1% of the cells were glycoA-, an absolute

increase from O to 4 . 8 ~ 1 O" cells/mL (-30% of al1 cells by morphology). In PV, the

percentage of glycoA- cells increased from 0.3% to 26.8%, an absolute increase from

7x1 o2 to 6.7~10' cells/mL (-40% of al1 cells by rnorphology). In normal cultures, glycoA'

cells were first detected on day 7 and reached a peak on day 13. In contrast in PV

cultures, glycoA- cells were detectable in small nurnbers at day 0, and had already peaked

by day 7. Thus, in PV the rate of accumulation of glycoA* cells is accelerated over that of

the normal. This could occur a) if cornmitted erythroid progenitors in PV left the marrow

earlier than their normal counterparts and continued to differentiate norrnally in culture, or

b) if the rate of differentiation of PV progenitors to glycoA- cells, with or without

proliferation, were greatly incrcased. Expenments are currently underway to distinguish

between these hypotheses, either one of which would funher Our understanding of the PV

phenotype.

Polycythemia vera (PV) is a chronic myeloproliferative disorder of as yet undetermineci

etiology characterised by a hyperplasia of aii three major myeloid lineages. This disorder is

ciininguished from secondary polycythemia in that the def- is intrinsic to the cells6. and the

relentless overproduction of red blood cells in PV occurs in the presence of normal 0 2

saturation and with levels of s e m erythropoietin (Epo), the key hormone of normal adult

erythropoiesiss7, often depressedJ0*88*89. Due to the clonal nature of the disease, it is believed

that dysregdation of a pluripotential stem ce1 lads to increased promeration and expansion of

the affected myeloid compartments, leading to granulocytosis, thrombocytosis, and

erythrocytosis30''. However, the major clinico-pathological emphasis in PV is clearly upon

the erythroid lineage. Despite this, the kinetics of red ce11 development in PV has yet to be

elucidated.

21 1.212 Currently popular liquid culture systems rely on semm and cornplex sources

of cytokines to support the growth and development of human erythroid cells in vitro.

Unknown or undefined quantities of growth factors andor inhibitors in these reagent

preparations have made systematic and repeatable investigations of hematopoietic lineage

development in culture difficult, and have rendered the interpretation of findings

questionable. Therefore in order to compare red ce11 development in PV and normal in

rnass culture, we developed a novel strictly serum-free liquid mass culture systern.

METHODS

Patients and Normal Controls

Positive selection of PV patients followed the standard guidelines established by the PV

Study &oup6: The three major cnteria were increased total RBC mass (>3 5mUkg) and

splenomegaly, in the presence of normal 0 2 saturation (>92%). In the absence of one of the

major cnteriq a diagnosis of PV was made ifthere was a combination of two of the following

four minor criteria: i)increased plateiet count (>400x lo9/'L); ü) increased white blood cell count

(> 12x 1 09/L); üi) increased neutrophil aikaline phosphatase score (> 120); and iv) increased

serum BI2 (>700 pmoüL). AU patients were being managed by phlebotomy at the t h e of

study. Normal blood was obtained fi-om healthy volunteers.

Ceii Preparations and Culture

M e r informed consent, penpherai blood was obtained by venipuncture from heaithy

wiunteer donors and patients, and was immediately placed in a polypropylene tube containing

1 OU/rnL preservative- fiee sodium heparin (H20 5077MF;Gibco). The heparinized blood was

layered ont0 I5mL of Ficoll-Hypaque (Pharmacia, Montreal, Canada) and the light-density

mononuclear cells (PBMNC) were collected f i e r centrifugation at 450xG for 30 min at room

temperature. CeU suspensions were washed three times (450xG, 10 min at room temperature)

in a-minimal essentiai medium (aMEM) containing 0.1% fatîy acid-free and globulin-free

bovine senim albumin (FM-BSA), and the ceils were counted. Cells were resuspended in the

arictly sem-&ee liquid culture medium (Basal sero2eroA' Stem Ce11 Medium

supplemented with 11-3 (10 ng/mL), SCF (50 ng/rnL), Epo (3U/mL) or IGF-I(3xl0-" M),

Hemin ( 1 OOpM) , and ATRA (3 0 nM)) to a final concentration of 5x 1 O' cells/rnL. The

mixture was used to seed culture weiis to a £inal volume of 2mL. The ceiis were cultured for

up to 14 days. At intervals dunng the culture period, cells were harvested, the total

number of viable ceils was detemiined, differential counts were obtained on Giemsa-

stained cytospin preparations, and the component ceil populations of the cultures were

delineated by flow cytometry. Cultured cells fkom PV patients and nomai individuals

were assessed for their expression of CD3, CD 1 9, CD 13, CD34, CD4 1, CD45, HLA-DR

and glycophorin A (glycoA) surface antigens.

RESULTS

Comparison of strictly serum-free and serurn containing liquid culture media

We followed the development of the various ce11 lineages under our one-step

strictly serum-free (SF) conditions and a two-step serum-containing system SC)^'*.

Comparing these two systems, the kinetics of the hematopoietic ce11 development did not

differ for the majority of lineages exarnined (Figure 1). Figure 2 shows the production of

glycophorin A positive cells in the SF versus SC culture for normal PBMNC seeded at a

concentration of 5 x 1 0 ~ cells/mL. The graph illustrates the number of glycophorin A

positive cells produced as a function of time in culture. There appears to be no difference

in the number or rate of ce11 production in SF versus SC mass culture. Having established

that our SF mass culture system was efficient for the production of erythroid cells, we

wished to compare the production of erythroid cells in normal individuals and patients

with PV.

Production of glycophorin A positive cells in SF mass culture in the presence of

3 U h L Epo

PBMNC isolated from normal individuals (N) and patients with PV were seeded

into culture at a concentration of 5x 1 o5 celis/mL. The cultures were followed for up to 14

days and the relative number of glycophorin A positive cells determined by flow cytornetry.

The mean number of g lycopho~ A positive cells +/- standard error of the mean for duplicate

cultures of 2 normal individuals and 2 patients with PV (representative of 5 of each) are given.

In the normal 0, while no glycophorin A positive cells were present at Day O, they were

produced at approxhnately 12-14 days in culture at which point approxhnately 16-39% of al1

ceils in culture were glycophorin A positive . This was i i a r to our finding with the serum

containing system as weii (Figure 2). in contrat, in PV, the first glycophorin A positive ceiis

were produced d e r approximately 7 days in culture representing 5om 13- 16% of d cens in

culture with the majority being produced at approximately 12- 14 days in culture representhg

from 26-56% of ail cells. These data indicate that erythroid ceii production in PV occurs

eariier than normal.

The absolute number of glycophorin A positive cells followed over 14 days in

culture (Figure 3 ) was converted to a percent maximum value for erythroid ce11

production the normal (N) and PV. The photornicrographs (3 12.SX magnification)

illustrate erythroid ce11 production over the 14 days in culture with representative fields of

normal (N) and PV cultures for Day O, Day 7, and Day 14 in culture with Epo (3U/rnL)

(Figure 4). These data illustrate that in normal, erythroid cells are produced only after 12-

14 days in culture. However in PV, terminally differentiated erythrocytes are present as

early as Day 7 in culture and again at Day 14.

IGF-1 stimulated eryt hropoiesis in liquid culture.

There is now considerable evidence in the literature to support the role of insulin-

like growth factor I (IGF-1) in e r y t h r ~ ~ o i e s i s ~ ~ . We have previously s h o w that IGF-1 is

able to support the proliferation and differentiation of normal erythroid progenitor cells in

serurn-free semi-solid culture5' and that erythroid progenitor cells in PV are hypersensitive

to IGF-1". Thus, where a relatively low concentration of IGF-1 is not permissive for

normal erythroid burst formation, it does stimulate the production of ery-throid bursts in

PV. We wished to determine if these findings were also true for the production of mature

erythroid cells in mass culture. Accordingly, we investigated the production of

glycophorin A positive cells over 14 days in SF mass culture in the presence of ~ X I O - " M

IGF-1. The mean nurnber of g lycopho~ A positive cens +/- standard error of the mean for

duplicate cultures of 2 normal individuals and 2 patients with PV are given. In the normal (N),

no glycophorin A positive celis were present at Day 0, and none were found throughout the 14

days in culture. This was in contrast to our finding with Epo (Figures 3 and 4). In marked

contrast PV cultures once again produced glycophorin A positive celis alter approximately 7

days with these cells representing approximately 24-77% of ail cells in culture. By Day 14,

glycophorin A positive cells in culture represented 1349% of ail celis (Figure 5). These data

indicate that temiinally dserentiated erythrocytes are produced in SF mas culture by PV

PBMNC in response to 3x10-" M IGF-1. Thug we have confirmeci at the level of mature red

cells our earlier hdings that erythroid progenitor celis in PV are sensitive to dramaticaiiy Iower

concentrations of IGF-1 which are not n o d y stimulatory for erythropoiesis.

CONCLUSION

Thus, in PV the rate of accumulation of glycophorin A positive cells cells is

acceierated over that of the normal. This could occur a) if committed erythroid

progenitors in PV (e-g. CFU-E) lefi the marrow earlier than their normal counterparts and

continued to differentiate normally in culture, or b) if the rate of differentiation of PV

progenitors to glycophorin A positive cells, with or without proliferation, were greatly

increased. Liquid mass culture under stnctly semm-fiee conditions provides a powerfùl

tool for the investigation of proliferation and differentiation to end cells during nomai and

neoplastic hematopoiesis.

Figure B. 1 - Examination of the growth charactenstic of CD3, CD 19, CD%. CD4 1.

CD45 and Glycophorin A positive ce11 populations in cultures established fiom normal

individuals under serum-containing (1 a) and stnctly serum-free (SF) (1 b) conditions. We

found that the growth characteristics of rnany of the ce11 types exarnined were not

significantly different in the two media. However, there did seem to be a significant

difference with respect to CD 19 positive cells. Serum-free conditions appeared to be

better for expansion of this lineage.

COL - CO.!

Figure B.2 - Comparison of serum-containhg and strictly serum-fiee liquid SF culture

media with respect to the growth of glycophonn A positive cells. The number of

glycophorin A positive celis produced fiom normal penpheral blood mononuclear cells in

culture in the presence of 3U/mL erythropoietin are given as a function of time. There

was on difference between cultures with respect to the rate or nurnber of glycophonn A

positive cells produced.

Num ber of Glycophorin A Positive Cells in Semrn-Containiog Culture

Num ber of G1ycophonn A Positive Cells in Serum-Free CuIture

Figure B.3 - Growth characteristics of Giycophorin A positive cells in normal and PV

cultures with 3UIrn.L erythropoietin. Penpherai blood rnononuclear cells isolated from

normal individuals and patients with PV were seeded at 5x10~ celUrnL in strictly semrn-

fiee culture. The maximum nurnber of glycophorin A positive produced in culture over

time are given. We found that in PV glycophonn A positive cells are produced in culture

after only seven days. Normally, it takes cells 12-1 4 days.

Production of Glycophorin-A Positive Cells with 3UlmL Epo

O 1 2 3 4 5 6 7 8 9 1 0 1 1 f 2 1 3 1 4 I S

Days in Culture

Figure B.4 - Cytospin preparations of PV and normal cells cultured in the presence of

3 U h L erythropoietin. Peripheral blood mononuclear cells isolated from normal

individuals and patients wit h PV were seeded at 5x 1 o5 ce1Vm.L in strictly serum-free

culture. The cultures were harvested and cytospin preparations made and stained with

Giemsa. A representative field of normal and PV cultures at DayO (A and D respectively),

Day 7 (B and E respectively), and Day 12 (C and F respectively) are shown. These

photomicrographs (3 1 ZSx magnification) may be correlated to time points in culture with

the inset graph. These data illustrate that in the presence of erythropoietin, glycophorh A

positive cells are produced earlier than in the normal, with peak production for both

normal and PV occurring at approximately Day 12.

Production of Clycophorin-A Pasitivc Cdis with 3x10-114I IGF-I

- Jr ' a - *

Figure B.5 - Normal and PV erythropoiesis in culture with 3x10"~ M IGF-1. The mean

number of glycophorin A cells produced for duplicate cultures of 2 normal individuals and

2 patients with PV is given. The photornicrographs (3 1 2 . 5 ~ magnification) are for normal

and PV day 12 (A and B respectively). The graph inset and the photomicrographs

illustrate that in the normal, in contrast to cultures containing erythropoietin, no

glycophorin A positive cells are produced in culture. However in PV, erythroid

progenitor cells give rise to glycophonn A positive cells by Day 7. Thus, 3x10"' M IGF-1

is insufficient to induce glycophorin A ce11 production in the normal.

REFERENCES

1. Vaquez MH: Sur une forme speciale de cyanose s'accompagnant d'hypergloblulie excessive et persistante. Comptes Rendus des Seances de la Societie de Biologie et ses Filiales 44: 3 84, 1 892

2. Oder W: Chronic cyanosis with polycythemia and enlarged spleen: a new clinical entity. Am J Med Sci 126: i 87, 1903

5 . Naegeli O : Blutkrankheiten und Blutdiagnostik. Leipzig, Verlag von Veit Comp, 1 908

4. Parkes-Weber F, Watson JH: Chronic polycythemia with enlarged spleen, probably a disease of the bone marrow. Br Med J 1:729, 1904

5. Turk W: Beitrage m r Kentniss des Symptombildes: Polycythamie mit Milztumor und Zyanose. Wien Klin Wochenschr 17: 1 53, 1904

6. Berlin NI: Diagnosis and classification of the polycythemias. Semin Hematol 12:339, 1975

7. Berk PD, Goldberg JD. Donovan PB, Fruchtman SM, Berlin NI: Therapeutic recommendations in polycythemia vera based on Polycythemia Vera Study Group protocols. Semin Hematol23 : 132. 1986

8. Murphy S: Polycythemia vera. Dis Mon 3 : 1 59, 1 992

9. Adamson JW, Fialkow PJ: The pathogenesis of myeloproliferative syndromes. Br J Haematot 3 8 :2W, 1978

10. Castro-Maiaspina Hl Rabellino EM, Yen A, Nachan RL, Moore MAS: Human megakaryocyte stimulation of proliferation of bone marrow fibroblasts Blood 57:78 1, 1982

1 1. Dickstein JI, Vardiman N : Hematopathological findings in the myeloproliferative disorders. Semin Oncol 22% 5, 1995

I I . Ellis JT, Peterson P: The bone marrow in Polycythemia vera. New York, Appleton- Century-Crofis, 1979

13. Ellis JT, Peterson P, Geller SA: Studies of the bone marrow in Polycythemia vera and the evolution of myelofibrosis and second hematological maiignancies. Semin Hematol 23 : 144, 1986

14. Wolf BC, Banks PM, Mann RB: Splenic hematopoiesis in polycythemia vera: A morphologie and immunohistologic study. Am J Clin Path 89:69, 1988

15. Tefferi A, Silverstein MN, Noel P: Agnogenic myeloid metaplasia. Semin Oncol 22:327, 1995

16. Najean Y, Dresch C, Rain J-D: The very-long-term course of Polycythemia: A cornplement to the previously published data of the Polycythernia vera Study Group. Br J Haematol 86 :Z 3, 1994

1 7. Landaw SA: Acute leukernia in Polycythernia vera. Semin Hematol 23 : 1 56, 1986

18. Nand S, Messmore H, Fisher SG, Bird ML, Schulz W, Fisher EU: Leukemic transformation in Polycythernia vera: Analysis of risk factors. Am I Hematol 3432, 1990

19. Kwong YL: Leukaernic risk in polycythaemia vera and essential thrombocytopenia. Eur J Haematol 56:90, 1996

20. Berk PD, Goldberg JD, Silverstein MN: Increased incidence of acute leukemia in polycythernia vera associated with chlorambucil therapy. N Engl J Med 304:441, 198 1

2 1 . Nand S. Stock W. Godwin I, Fisher SG: Leukemogenic risk of hydroxyurea in therapy in Polycythernia vera, Essentiai Thrombocythernia, and Myeloid Metaplasia with Myelofibrosis. Am J Hematol 52:42, 1996

22. Brandt L, Anderson H: Survival and risk of leukaernia in polycythaemia vera and essential thrombocythaernia treated with oral radiophosphorus: Are safer dnigs available? Eur J Haematol 5412 1, 1995

23. Najean Y, Deschamps A, Dresch C: Acute leukernias and rnyelodyspIasia in polycythemia vera: a clinical study with long-term follow-up. Cancer 6 1 239, 1988

24. Dewald GW, Wright PI: Chrornosome abnormalities in the myeloproliferative disorders. Semin Oncol 22:M 1, 1995

25. Diez Martin JL, Graham DL, Petitt RM, Dewald GW: Chrornosome studies in 104 patients with Polycythemia vera. Mayo Clin Proc 66:287, 199 1

26. Price CM, Kanfer EJ, Colman SM, Westwood N, Barrett AJ, Greaves MF: Simultaneous genotypic and immunophenotypic analysis of interphase cells using dual- color fluorescence: A demonstration of lineage involvement in Polycythemia vera. Blood 80:1033, 1992

27. Swolin B, Weinfeld 4 Westin J: A prospective long-tem cytogenetic study in Polycythemia vera in relation to treatment and clinical course. Blood 72:386, 1988

28. Berger Bemheim A, Le Coniat M, Vecchione D, Flandrin G, Dresch C, Najean Y Chromosome studies in Polycythemia vera patients. Cancer Genet Cytogenet 12:2 17, 1984

29. Dameshek W: Some speculations on the myeioproliferative syndromes. Blood 6:372, 1% 1

30. Adamson JW, Fialkow PI, Murphy S, Prchal E, Steinmann L: Polycythemia vera: Stem ce11 and probable clonai ongin of the disease. N Engl J Med 295913, 1976

3 1. Ash RC, Detnck Rq Zanjani ED: In vitro studies of hurnan plunpotentiai hernatopoietic progenitors in Polycythemia vera. J Clin Invest 69: 1 1 12, 1982

32. Ash RC, Detnck EL4, Zanjani ED: Studies of human plunpotential hemopoietic stem cells in vitro. BIood 58:309, 1981

33. Raskind WH, Jacobson Murphy S, Adamson JW, Fiaikow PI: Evidence for the involvernent of B lymphoid cells in Polycythemia vera and Essential Thrornbocythemia. J Clin Invest 75 : 1388, 1985

34. Tsukamoto N, Monta K. Maehara T, Okamoto K, Sakai H, Karasawa M. Naruse T, Omine M: Clonality in chronic myeloproliferative disorders defined by x-chromosome linked probes: demonstration of heterogeneity in lineage involvement. Br J Haematol 86:253, 1994

35 . Gale RE, Wheadon H, Goldstone AH, Burnett AK, Linch DC: Frequency of clonal remission in acute myeloid leukaernia. Lancet 34 1 : 138, 1993

36. Gale RE, Wheadon H, Boulos P. Linch DC: Tissue specificity of X-chromosome inactivation patterns. BIood 83 :2899, 1994

3 7. Krantz SB: Erythropoietin. Blood 77:4 19, 199 1

38. Adamson W: The erythropoietirdhematocrit relationship in normal and polycythernic man: Implications of marrow regulation. Blood 32597, 1968

39. Koeffler HP, Goldwasser E: Erythropoietin radioimmunoassay in evaluating patients with Polycythemia vera. Ann Intern Med 94:44, 198 1

40. Cotes PM, Dore CJ, Yin JAL, Lewis SM, Messines, M, Pearson TC, Reid C : Determination of serum erythropoietin in the investigation of erythrocytosis. N Engl J Med 315283, 1986

4 1. Gumey CW: Polycythernia vera and some possible pathogenetic mechanisms. Ann Rev Med 16: 169, 1965

42. Cotes PM, Bangham DR: Bioassay of erthropoietin in rnice made polycythaemic by exposure to air of a reduced pressure. Nature 19 1 : 1065, 196 1

43. Goldberg MA, Dunning SB, Bunn HF: Regulation of the erythropoietin gene: Evidence that the oxygen sensor is a heme protein. Science 242: 14 12, 1988

44. Bomchil G, Caxmena AO, Segade A, Cavagnaro F, De Testa NG: Studies on the response to hypoxia and relative hyperoxia in two polycythernia vera patients. Indian J Med Res 55543, 1967

45. Stephenson IR. Axelrad Aq McLeod DL, Shreeve MM: Induction of colonies of hemoglobin-synthesizing cells by erythropoietin in vitro. Proc Nat1 Acad Sci USA 68: 1542, 1971

46. McLeod DL, Shreeve MM, Axelrad AA: Improved plasma culture system for production of erythrocytic colonies in vitro: Quantitative assay method for CFU-E. Blood 44517, 1974

47. Prchal SF, Axelrad AA: Bone-mmow responses in Polycythemia vera. N Engl J Med 290: 1382, 1974

48. Eaves AC, Henkelman DH, Eaves CJ: Abnonnal erythropoiesis in the myeloproliferative disorders: An analysis of underlying cellular and humoral mechanisms. Exp Hematol 8 SuppL235, 1980

49. Gurney CW: Pathogenesis of the polycythemias, in Klein H (ed): Polycythemia: Theory and Management, Springfield, Ill. Charles C. Thomas, 1973, p 42

50. Zanjani ED, Lutton ID, H o h a n R, Wasserman LR: Erythroid colony formation by Polycythemia vera bone marrow in vitro. .J Clin Invest 59:841, 1977

5 1. Prchal IF, Adamson W, Murphy S, Steinmann L, Fialkow PJ: Polycythemia vera: The in vitro response of normal and abnormal ce11 lines to erythropoietin. J Clin Invest 61: 1044, 1978

52. Golde DW, Bersch N, Cline M.i: Polycythernia vera: Hormonal modulation of erythropoiesis in vitro. Blood 49:399, 1977

53. Casadevall N, Vainchenker W, Lacombe C, Vinci G, Chapman J, Breton-Gonus J, Varet B: Erythoid progenitors in Polycythemia vera: Demonstration of their hypersensitivity to erythropoietin using senim-free cultures. Blood 59447, 1982

54. Eaves CJ, Eaves AC : Erythropoietin (Ep) dose-response curves for three classes of erythroid progenitors in normal human marrow and patients with Polycythemia vera. Blood 52: 1 196, 1978

55. Douer D, Fabian 1, Cline MJ: Circulating plunpotent haemopoietic cells in patients with myeloproliferative disorders. Br J Haematol 54:3 73, 1 983

56. Konwalinka G, Geissler D, Peschle C, Breier C, Grunewald K, Odavic R, Braunsteiner H: Human erythropoiesis in vitro and the source of burst-promoting activity in a serum- fiee system. Exp Hematol 14399, 1986

57. Stewart S, Zhu B, Axelrad AA: A 'serum-free' medium for the production of erythropoietic bursts by munne bone marrow cells. Exp Hematol 12:309, 1984

58. Correa PN, he l r ad AA: Production of erythropoietic bursts by progenitor cells fiom adult human peripheral blood in an improved serum-fiee medium: Role of insulin-like growth factor 1. Blood 78:2823, 1991

59. Correa PN, Eskinazi D, Axelrad AA: Circulating erythroid progenitors in Polycythemia vera are hypersensitive to insulin-like growth factor 1 in vitro: Studies in an improved serum-free medium. Blood 83 :99, 1994

60. Cantanese VM, Grigorescu F, King GL, Kahn CR: The human erythrocyte insulin-like growth fator 1 receptor: Characterization and demonstration of ligand stimulated autophosphorylation. J Clin Endocrinol Metab 62:692, 1986

6 1. Immi T, Kasuga M, Kadowaki T, Hinika N, Takaka F, Akanuma Y: Characteristics of human erythrocyte insulin-like growth factor 1 receptors. J Clin Endocnnol Metab 62: 1206, 1986

62. Thorsson AV, Hintz RL: Specific 1- 125 somatomedin receptor on circulating mononuclear cells. Biochem Biophys Res Commun 74: 1566, 1977

6 3 Aron DC: Insulin-like growth factor 1 and erythropoiesis. BioFactors 3 :2 1 1, 1992

64. Golde DW: Growth hormone: Species-specific stimulation of erythropoiesis in vitro. Science 196: 1 I 12, 1977

65. K u m A, Zapf J, Eckhardt K-U, Clemons G, Froesch ER, Bauer C: Insulin-like growth factor 1 stimulates erythropoiesis in hypophysectomized rats. Proc Nat1 Acad Sci USA 85:7825, 1988

66. Sawada K, Krantz SB, Dessypris EN, Koury ST, Sawyer ST: Human colony-forming units-erythroid do not require accessory cells, but do require interaction with insulin-Iike growth factor 1 a d o r insulin for erythroid development. J Clin Invest 83 : 170 1, 1989

6 7 K U ~ Z A, Jelkmann W, Bauer C: A new candidate for the regulation of erythropoiesis. FEBS Lett 149:105, 1982

68. Kunz A, Hart1 W, Jelkrnann W, Zapf J, Bauer C: Activity of fetal bovine serum that stimulates erythroid colony formation in fetal mouse livers in IGF-I. J Clin Invest 76: 1643, 1985

69. Akahane K, Tojo A, Urabe A, Takaku F: Pure erythropoietic colony and burst formation in serum-fkee culture, and their enhancement by insulin-like growth factor I. Exp Hematol I 5: 797, I987

70. Buemi M, Allegra A, Aloisi C, Giacobbe MS, Frisina N: Different answer of recombinant human erythropoietin in anephric and non-nephric haemodialyzed patients. Nephron 56:2 18, 1990

71. Kurtz 4 Matter R, Eckhardt K-U, Zapf J: Erythropoiesis, serum erythropoietin, and IGF-I in rats during accelerated growth. Acta Endocrinologica 122:323, 1990

72. Brox AG, Congote LF, Fafard J, Fauser AA: Indentification and characterization of an 8-kd peptide stimulating late erythropoiesis. Exp Hematol 17:769, 1989

73. Congote LF, Bron A, Lin FK, Lu HS, Fauser AA: The N-terminal sequence of the major erythropoietic factor of an anephric patient is identical to insulin-like growth factor 1. J Clin Endocrinol Metab 72: 727, 199 1

74. Urena P, Bomardeaux A, Eckardt K-U, Kunz A, Drueke TB: Insu!in-like growth factor I: A modulator of erythropoiesis in uraemic patients. Nephrol Dial Transplant 7~40, 1992

75. Yoshirnura A, Longmore G, Lodish HF: Point mutation in the exoplasmic domain of the erythropoietin receptor resulting in hormone independent activation of tumorigenicity. Nature 348547, I990

76. Longmore GD, Pharr P, Lodish HF: Mutation in murine erythropoietin receptor induces erythropoietin-independent erythroid proliferation in vitro, polycythemia in vivo. J Clin Invest 1996

77. DtAndrea AD, Yoshimura A, Youssoufian H, Zon L, Koo J, Lodish HF: The cytoplasmic region of the erythropoietin receptor contains non-overlapping positive and negative regulatory domains. Mol Cell Biol 1 1 : 1980, 199 1

78. De La Chapelle 4 Traskelin AL, Juvonen E: Truncated erythropoietin receptor causes dominantly inherited benign human erythrocytosis. Proc Natl Acad Sci USA 90:4495, 1993

79. Le Couedic J-P, Mitjavila M-T, Villevai J-L, Feger F, Gobert S. Mayeux P, Casadevall N, Vainchenkar W: Missense mutation of the erythropoietin receptor is a rare event in human erythroid malignancies. Blood 87: 1 502, 1996

80. Hess G, Rose P. Gamm H, Papadileris S, Huber C, Seliger B: Molecular analysis of erythropoietin receptor system in patients with Polycythemia vera. Br J Haematol 88:794, 1994

8 1 . Eridani S. Dudley JM, Sawyer BM, Pearson TC: Erythropoietic colonies in a semm- free system: Results in primary proliferative polycythaemia and thrombocythaernia. Br J Haematoi 67:387, 1987

82. de Wolf JTM, Beentjes JAM, Esselink MT, Smit JW, Halie RM, Clark SC, Vellenga E: In Polycythemia vera human interleukin 3 and granulocyte-macrophage colony- stimulating factor enhance erythroid colony growth in the absence of erythropoietin. Exp Hematol 1 7: 98 1, 1989

83. Axelrad LU, McLeod DL, Suzuki S, Shreeve MM: Regulation of population size of erythropoietic progenitor cells, in Clarkson B. Marks PA Till JE (eds): Fifih Cold Spring Harbor Laboratory Conference on Ce11 Proliferation, Differentiation of Hernatopoietic Cells, New York, Cold Spring Harbor Laboratory, 1978, p 155

84. Adarnson SW, Singer JW, Catalano P, Murphy S, Lin N, Steinmann L: Polycythernia vera: Further in vitro studies of hematopoietic regulation. J Clin Invest 66: 1363, 1980

85. Nijhof W, de Hann G. Pietens J, Dontje B: Mechanistic options of erythropoietin- stimulated erythropoiesis. Exp Hemato123:369. 1995

86. LeRoith D, Werner H, Beitner-Johnson D, Roberts CTJ: Molecular and cellular aspects of the insulin-like growth factor receptor. Endocr Rev 16: 143, 1995

87. Spivak JL, Pham J, Isaacs M, Hankins WD: Erythropoietin is both a mitogen and a survival factor. Blood 77: 1228. 199 1

88. de Klerk G, Rosengarten PC, Vet RJ, Goudsmit R: Semm erythropoietin (ESF) titer in polycythemia. Blood 58: 1171, 1981

89. Wognum AW, Lansdorp PM, Eaves AC, Krystal G: An enzyme linked immunosorbent assay for erythropoietin using monoclonal antibodies, tetrarneric immune complexes and substrate amplification. Blood 74:622, 1989

90. Lacornbe C, Casadevall N, Varet B: Polycythemia vera: In vitro studies of circulating erythroid progenitors. Br J Haematol44: 189, 1980

9 1. Ku11 FC, Iacobs S, Su Y-F, Svoboda MF, Van Wyk JJ, Cuatrecasas P: Monoclonal antibodies to receptors for insulin and somatomedin C. .i Bi01 Chem 258:6561, 1983

92. Yarden Y, Wlrich A: Molecuiar analysis of signal transduction by growth factors. Biochernistry 27: 3 1 13, 1988

93. Hanks SK, Quinn AM, Hunter T: The protein kinase family: Conserved features and deduced phylogeny of the catalytic domains. Science 24 1 :42, 1988

94. Yarden Y, Ullrich A: Growth factor receptor tyrosine kinases. Annu Rev Biochem 571443, 1988

95. Ullrich A, Gray A., Tarn AW, Yang-Feng T, Tsubokawa M, Collins C, Henze1 W, Le Bon Tl Kathuria S, Chen E, Jacobs S, Francke U, Rarnachandran J, Fujita-Yamaguchi Y: Insulin-like growth factor4 primary stnicture: Cornparison with insulin receptor suggest structural determinants that define functional specificity. EMBO 1 52503, 1986

96. Adarno M, Roberts CTJ, LeRoith D: How distinct are the insuiin and insulin-like growth factor 1 signialing systems? BioFactors 3 : 15 1, 1992

97. Iacobs S, Ku11 FC, Cuatrecasas P: Monensin blocks the maturation of receptors for insulin and somatomedin C. Proc Natl Acad Sci USA 80: 1228, 1983

98. Iacobs S. Ku11 FC, Earp HS, Svoboda M, Van Wyk IJ, Cuatrecasas P: Somatomedin- C stimulates the phosphorylation of the b subunit of its own receptor. J Bi01 Chem 258:9581, 1983

99. Rubin JB, Shia MA, Pilch PF: Stimulation of tyrosine phosphorylation in vitro by insulin-like growth factor 1. Nature 305:438, 1983

100. Ullnch A., Schlessinger J: Signal transduction by receptor with tyrosine kinase activity. Ce11 6 1 :203, 1990

10 1 . Lopaczynski W. Harris S, Nissley P: Insulin-like growih factor4 (IGF-1) dependent phosphorylation of the IGF-1 receptor in MG-63 cells. Regulatory Peptides 48:207, 1993

102. Zick Y, Sasaki N, Rees-Jones RW, Grunberger G, Nissley SP, Rechler MM: Insulin- li ke growth factor4 (IGF-1) stimulates tyrosine kinase activity in purified receptors fiom a rat liver ce11 line. Biochem Biophys Res Commun 1195, 1984

1 03. Boni-Schnetzler M, Kalgian A, Del Vecchio R Pilch PF: Ligand-dependent intersubunit association within the insulin receptor complex activates its instrinsic kinase activity. J Bi01 Chem 2635822, 1988

104. Felty SM, Swanson ML, Wemmie JA, Pessin JE: Functional properties of an isolated a p heterodimenc human placenta insulin-like growth factor 1 receptor complex. Biochemistry 27:3234, 1988

1 05. C hakrovarty A, loslyn MI, Davis JS : C haracterization of insulin and insulin-like growth factor4 actions in the bovine luteal cell: Regulation of receptor tyrosine kinase activity, phophatidylinositol-)-kinase, and deoxyribonucleic acid synthesis. Endocrinology 133: 133 1, 1993

106. Kadowaki T, Koyasu S. Nishida E, Tobe K, Immi T, Takaku F, Sakai H, Yahara 1, Kasuga M: Tyrosine phosphorylation of comrnon and specific sets of cellular proteins rapidly induced by insulin, insulin-like growth factor 1, and epidemal growth factor in an intact cell. J Biol Chem 262:7342, 1987

107. Meyers MGJ, Sun XJ, Cheatharn B, Iachna BR, Glasheen EM, Backer IM, White MF: IRS-1 is a comrnon element in insulin and insulin-like growth factor-[ signaling to the phosphatidylinositol 3'-kinase. Endocrinology 132: 142 1, 1993

108. Murakami MS, Rosen OM: The role of insulin receptor autophosphorylation in signal transduction. J Bi01 Chem 266:22653, 199 1

109. Kato H, Faria TN, Stannard B, Roberts CTJ, LeRoith D: Role of tyrosine kinase activity in signai transduction by insulin-like growth factor4 (IGF-1) receptor. J Bi01 Chem 2 6 8 2 6 5 5 , 1993

110. Kato K. Faria TN, Stannard B. Roberts CTJ, LeRoith D: Essential role of tyrosine residues 1 13 1, 1 13 5, and 1 136 of the insulin-like growth factor4 (IGF-1) receptor in IGF- 1 action. Mol Endocrino1 8:40, 1994

1 1 1 Resnicoff M, Ambrose D, Coppola D, Rubin R: Insulin-like growth factor-1 and its receptor mediate the autocrine proliferation of human ovarian carcinoma ce11 lines. Lab Invest 69:756, 1993

1 12. Di Fiore PP, Pierce IH, Fleming TP, Hazan R, Ullrich A, King CR, Schlessinger J, Aaronson SA: Overexpression of the human EGF receptor confers an EGF-dependent transfonned phenotype to NM-3T3 celis. Ce11 5 1 : 1063, 1987

1 13. Kaleko M, Rutter WJ, Miller AD: Overexpression of the human insulin-like growth factor 1 receptor prornotes ligand-dependent neoplastic transformation. Mol Cell Bi01 10:464, 1990

1 14. La1 RB, Rudolph DL, Folks TM, Hooper WC: Overexpression of insulin-like growth factor receptor type4 in T-ce11 lines infected with human T-lyrnphotrophic virus type 1 and II. Leukemia Res 17:31, 1993

1 15. Roussel MF, Du11 TJ, Rettenmier CW, Ralph P, üllrich 4 Shen CJ: Transforming potential of the c-fis proto-oncogene (CSF- 1 receptor). Nature 325549, 1987

1 16. Alexandrides TK, Smith RI: A novel fetd insulin-like growth factor (IGF-1) 1 receptor. J Bi01 Chem 264: 12922, 1989

1 17. Kellerer M. Obermaier-Kusser B, Ermel B, Wallner U, Haring H-1, Petndes PE: An altered IGF-1 receptor is present in human leukemia cells. J Bi01 Chem 2659340, 1990

1 18. Bethard WF, Jacobson LO: Definition and Classification of Polycythemia, in Klein H (ed): Polycythemia: Theory and Management, Springfield, Charles C. Thomas, 1973, p 18

1 19. Mina AM, Correa PN, Axelrad AA: Increased basal and induced tyrosine phosphorylation of the insulin-like growth factor 1 receptor b subunit in circulating mononuclear cells of patients with Polycythemia vera. Blood 86377, 1995

120. Daughaday WH, Rotwein P: Insulin-like growth factors 1 and II. Peptide, messenger ribonucleic acid and gene structures, semm and tissue concentrations. Endocr Rev 10:68, 1989

12 1. Cohick WS, Clemmons DR: The insulin-like growth factors. Annu Rev Physiol 55:131, 1993

122. Takada M, Nakanome H, Kishida M, Hirose S. Hasegawa T, Hasegawa Y: Measurement of fiee insulin-like growth factor-1 using irnrnunoradiometric assay. J Imrnunoassay 1 5: 263, 1 994

123. Jones II, Clemmons DR: Insulin-like growth factors and their binding proteins: Biological actions. Endocr Rev I6:3, 1995

124. Schrnid C, Rutishauser J, Schlapfer 1, Froesch E q Zapf J: Intact but not truncated insulin-like growth factor binding protein-3 (IGFBP-3) blocks IGF-1-induced stimulation of osteoblasts: Control of IGF signailing to bone cells by IGFBP-3-specific proteolysis? Biochem Biophys Res Commun 179579, 199 1

125. Liu L, Delbe J, Blat C, Zapf J, Harel L: Insulin-like growth factor binding protein (IGFBP-3), an inhibitor of serurn growth factors other than IGF-I and -II. J Ce11 Physiol I53:15, 1992

126. Okajima T, Iwashita M, Takeda Y, Sakarnoto S, Tanabe T, Yasuda T, Rosenfeld RG: Inhibitory effects of insulin-like growth factor (1GF)-binding proteins- 1 and -3 on IGF-activated glucose consumption in mouse BALB/c 3T3 fibroblasts. J Endocrinol 136:457, 1993

127. Lee PDK, Conover CA, Powell DR: Regulation and function of insulin-like growth factor-binding protein- 1. Proc Soc Exp Biol Med 204:4, 1993

128. Elgin RG, Busby WHJ, Clernrnons DR: An insulin-like g r ~ w t h factor (IGF) binding protein enhances the biologic response to IGF-1. Proc Nat1 Acad Sci USA 84:3254, 1987

129. Campbell PG, Novak IF: Insulin-Iike growth factor binding protein (IGFBP) inhibits IGF action on human osteosarcorna cells. J Ce11 Physiol 149293, 199 1

130. Burch WM, Correa I, Shaveley JE, Powell DR: The 25-kilodalton insulin-like growth factor (1GF)-binding protein inhibits both basal and IGF-1 mediated growth of chick embryo pelvic cartilage in vitro. J Clin Endocnnol Metab 70: 173, 1990

13 1. E u a t S, Song-Guang R, Braunstein GD, Melmed S: Octreotide stimulates insulin- like growth factor binding protein- 1 : A potential pituitary-independent mechanisrn for dmg action. I Clin Endocrinol Metab 75: 1459, 1992

132. Hossenlopp P, Seurin D, Segovia-Quinson B, Hardouin S, Binoux M: Analysis of serum insulin-like growth factor binding proteins using western blotting: use of the method for titration of the binding proteins and cornpetitive binding studies. Anai Biochem 154338, 1986

133. Yki-jarvinen H, Makimattila S, Utnainen T, Rutanen E: Portal insulin concentrations rather than insulin sensitivity regulate semm sex hormone-binding globulin and insulin-like growth factor binding protein 1 in vivo. J Clin Endocnnol Metab 80:3227, 1995

134. Busby W, Klapper DG, Clemmons DR: Purification of a 3 1,000 dalton insulin-like growth factor binding protein corn human amniotic fluid: Isolation of two isoforms with different biologic actions. J Bi01 Chem 263: 14203, 1988

135. Yeoh SI, Baxter RC: Metabolic regulation of the growth hormone independent insulin-like growth factor binding protein in human plasma. Acta Endocrinologica 1 19:465, 1988

136 Suikkari A, Koivisto VA, Rutanen E, Yki-jarvinen H, Karonen SL, Seppaia M: Insulin regulates the semm levels of low molecular weight insulin-like growth factor- binding protein. J Clin Endocnnol Metab 66266, 1988

137. Conover CA, Lee PDK: Insulin regulation of insulin-like growth factor-binding protein production in cultured HepG2 cells. J Clin Endocrinol Metab 70: 1062, 1990

138. Young SCJ, Underwood LE, Ceiniker 4 Clernrnons DR: EfEects of recombinant insulin-like growth factor4 (IGF-1) and growth hormone on serum IGF-binding proteins in calot-ically restncted adults. J Clin Endocnnol Metab 75:603, 1992

139. Baxter RC, Hinika N, Takano K, Holman SR, Asakawa K: Responses of insulin-like growth factor binding protein- 1 (IGFBP- 1 ) and the IGFBP-3 complex to administration of insulin-like growth factor-1. Acta Endocrinologica 128: 10 1, 1993

140. Grellier P, Chanson P, Casadevail N, Abboud S, Schaison G: Remission of Polycythemia vera after surgical cure of acromegaly. AM Intern Med 124:495, 1996

14 1. DeMellow J, Baxter RC : Growth hormone-dependent insulin-like growth factor (IGF) binding protein both inhibits and potentiates IGF-1 stimulated DNA systhesis in human skin fibroblasts. Biochem Biophys Res Commun 156: 199, 1988

142. Knauer DJ, Smith GL: Inhibition of biological activity of multiplication stimulating activity by binding to its c h e r protein. Proc Natl Acad Sci USA 77:7252, 1980

143 Camacho-Hubner C, McCusker RH, Clemmons DR: Secretion and biological actions of insulin-like growth factor binding proteins in two human tumor-derived ce11 lines in vitro. J Ce11 Physiol l48:28 1, 199 1

144. Frost RA, Tseng L: Insulin-like growth factor-binding protein- 1 is phosphorylated by cultured human endometrial stroma1 cells and multiple protein kinases in vitro. J Bi01 Chem 266: 1 8082, 199 1

145. Jones .JI, D'Ercole -91. Camacho-Hubner C, Clemmons DR: Phosphorylation of insulin-like growth factor (IGF-1)-binding protein- 1 in ce11 culture and in vivo: effects on affinity for IGF-1. Proc Natl Acad Sci USA 88:748 1, 199 1

146. Jyung RW, Mustoe TA, Busby WHJ, Clemmons DR: Increased wound breaking strength induced by insulin-like growth factor4 in cornbination with IGF binding protein- 1. Surgery 1 l5 :233, 1994

147. Merchav S, Silvian Drachsler 1, Tatarsky 1, Lake M, Skottner A: Comparative studies of erythroid-potentiating effects of biosynthetic human insulin-like growth factors4 and -II. J Clin Endocrinol Metab 74:447, 1992

148. Mirza AM, Eua t S, Axelrad AA: Insulin-like growth factor binding protein-l (IGFBP-1) stimulates erythroid burst formation in vitro and is elevated in patients with Polycythemia vera. Blood 1 997

149. Schwander JC, Haun C, Zapf J, Froesch ER: Synthesis and secretion of insulin-like growth factor and its binding protein by the perfused rat liver: Dependence on growth hormone status. Endocrinology 1 13 :297, 1983

150. D'Ercole Al, Stiles AD, Underwood LE: Tissue concentrations of somatomedin C: Further evidence for multiple sites of synthesis and paracnne or autocrine mechanisrns. Proc Natl Acad Sci USA 8 1 :93 5, 1984

15 1. Han VKM, DZrcole AJ, Lund PK: Cellular localization of sornatomedin (IGF) messenger RNA in the human fetus. Science 236: 193, 1987

152. Muta K, Krantz SB: Apoptosis of human erythroid colony-fomiing cells is decreased by stem ce11 factor and insulin-like growth factor 1 as well as erythropoietin. .i Ce11 Physiol 156:264, 1993

153. Shimanski S, Ling N: Identification and molecular characterization of insulin-like growth factor proteins (IGFBP- 1, -2, -3, -4, -5, and -6). Prog Growth Factor Res 3 :243, 1992

154. Holly IMP: The physiological role of IGFBP- 1. Acta Endocrinologica l24:5S, 199 1

155. Clemmons DR, Gardner LI: A factor contained in plasma is required for IGF binding protein- 1 to potentiate the effect of IGF-I on smooth muscle ceIl DNA synthesis. I Ce11 Physiol 145: 129, 1990

1 56. Bourner MJ, Busby WHJ, Siegel NR, Krivi GG, McCusker RH, Clemmons DR: Cloning and sequence determination of bovine insulin-like growth factor binding protein-2 (IGFBP-2): Cornparison of its structural and functional properties with IGFBP- 1. J Ce11 Biochem 48:2 15, 1992

157. Figueroa JA, Lee AV, Jackson JG, Yee D: Proliferation of cultured human prostate cancer cells is inhibited by insulin-like growth factor (IGF) binding protein- 1: Evidence for an IGF-II autocrine growth loop. J Clin Endocrinol Metab 80:3476, 1995

1 5 8. MotM A, Rutherford C, Anggard EE, Fems GAA: Insulin-like growth factor binding protein- 1 inhibits artenal smooth muscle ce11 proliferation in vitro but does not reduce the neointimal response to balloon catheter injury. Atherosclerosis 1 1857, 1995

159 Arkins S, Rebeiz N, Biragyn 4 Reese DL, Kelly KW: Munne macrophages express abundant insulin-like growth factor4 class I Ea and Eb transcnpts. Endocnnology 133.2334, 1993

160. Kirstein M, Aston C, Hintz R Vlassara H: Receptor-specific induction of insulin-like growth factor 1 in human monocytes by advanced glycosylation end product-modified proteins. J Clin Invest 90:439, 1992

16 1. Lake FR, Noble PW, Henson PM, Riches DWH: Functional switching of macrophage responses to tumor necrosis factor-a ( m a ) by interferons. J Clin Invest 93:1661, 1994

162. Arkins S, Rebeiz N, Brunke-Reese DL, Minshall C, Kelley KW: The colony- stimulating factors induce expression of insulin-like growth factor4 messenger nbonucleic acid dunng hematopoiesis. Endocrinology 1 3 6: 1 1 53, 1 995

163. Jones JI, Gockerman D, Busby WKT, Wright G, Clemmons DR: Insulin-like growth factor binding protein 1 stimulates ce11 migration and binds to the alpha 5 beta 1 integrin by means of its Arg-Gly-Asp sequence. Proc Natl Acad Sci USA 90: 10553, 1993

164. Li YMI Arkins S. McCusker RH, Donovan SM, Liu Q, Jayaraman S. Dantzer R, Kelly KW: Macrophages synthesize and secrete a 25-kilodalton protein that binds insulin- like growth factor-1. J Immun01 156:64, 1996

165. Blum WF, Jeme EW, Reppin F, Kietmian K, Renke MB, Bierich IR: Insulin-like growth factor I (IGF-1)-binding protein complex is a better mitogen than fiee IGF-1. Endocrinology lî5:766, 1989

166. Clemrnons DR, Cascieri MA, Camacho-Hubner C, McCusker RH, Bayne ML: Discrete alterations of the insulin-like growth factor 1 molecule which alter its affinity for insulin-Iike growth factor-binding proteins result in changes in bioactivity. I Bi01 Chem 265:12210, 1990

167. Knauer DI, Wiley HS, Cunningham DD: Relationship between epidermal growth factor receptor occupancy and mitogenic response. J Bi01 Chem 2595623, 1984

168. Fisher MJ, Prchal JF, Prchal JT, D'Andrea AD: Anti-erythropoietin (EPO) receptor monoclonal antibodies distinguish EPO-dependent and EPO-independent erythroid progenitors in Polycythemia vera. Blood 84: 1982, 1994

169. Golding MB, Golding SR: Cytokines and ce11 growth control. Euk Gene Exp 130 1. 1991

1 70. Shier P, Watt VM: Primary structure of a putative receptor for a ligand of the insulin family. J Bi01 Chem 264: 14605, 1989

17 1. Cashman JD, Eaves CJ, Eaves AC: Unregulated proliferation of primitive neoplastic progenitor cells in long term Polycythemia vera marrow cultures. J Clin Invest 8 1 :87, 1988

172. Sieff CA, Niemeyer CM, Nathan DG, Ekem SC, Bierber FR, Yang YC, Wong GG, Clark SC: Stimulation of human hematopoietic colony formation by recombinant gibbon multi-colony stimulating factor or interleukin-3. I Clin Invest 80: 8 18, 1987

173. Metcalf D, Begley CG, Johnson GR Nicola NA, Vadas MA, Lopez AF, Williamson DJ, Wong GG, Clark SC, Wang EA: Biologic properties in-vitro of a recombinant human granulocyte-macrophage colony-stimuiating factor. Blood 67:37, 1986

174. Sawada K, Krantz SB, Dai CH, Koury ST, Hom ST, Glick AD, Civin CI: Purification of human blood burst-forming units-erythroid and demonstration of the evolution of erythropoietin receptors. .i Ce11 Physiol 14212 19, 1990

175. Dai CH, Krantz SB, Means RTJ, Hom ST, Gilbert HS: Polycythemia vera blood burst-foming units-erythroid are hypersensitive to interleukin-3. J Clin Invest 87 : 3 9 1, 1991

176. Dai CH, Krantz SB, Dessypris EN, Means RTJ, Hom ST, Gilbert HS: Polycythemia vera. II. Hypersensitivity of bone rnarrow erythroid, granulocyte-macrophage, and megarkaryocyte progenitor cells to interleukin-3 and granulocyte-macrophage colony stimulating factor. Blood 80: 89 1, 1992

177. de Wolf JTM, Hendriks DW, Esselink MT, Halie MR, Vellenga E: The effects of CL- 1 and IL-4 on the Epo-independent erythroid progenitor in Polycythaemia vera. Br J Haematol 88:242, 1994

178. Murphy WJ, Dumm SK, Anver MR, Longo DL: Irnmunologic and hematologic effects of neuroendocnne hormones: Studies on DW/J dwarfmice. J Immun01 1483799, 1992

179. Murphy WJ, Tsarfaty G, Longo DL: Growth hormone exens hematopoietic growth- promoting effects in vivo and partially counteracts the myelosuppressive effects of azidothymidine. Blood 80: 1433, 1992

180. Wells J A , de Vos AM: Structure and hnction of human growth hormone: Implications for hematopoietins. Annu Rev Biophys Biomol Struct 22:329, 1993

18 1. Ratajczak MZ, Kuczynski WI, Onodera K, Moore J, Ratajczak J, Kregenow DA, De Riel K, Gewitz AM: A reappraisal of the role of insulin-like growth factor 1 in the regulation of human hematopoiesis. I Clin Invest 94:320, 1994

182. Muta K, Krantz SB, Bondurant MC, Wickrema A: Distinct roles of erythropoietin. insulin-like growth factor 1, and stem ceIl factor in development of erythroid progenitor ceIls. .f Clin Invest 94:34, 1994

1 83. Ushikubi F, Ishibashi T, Narurniya S. Okuma M: Analysis of the defective signal transduction mechanisrn through the platelet thromboxane A2 receptor in a patient with Polycythemia vera. Thromb Haemost 67: 144, 1992

184. Sarnuelsson J, Hansson A, Rosendahl K, Palmblad J: Superoxide anion production and phospholipase D-mediated generation of diacylglycerol are subnormal after N-formyl- methionyl-leucyl-phenylalanine stimulation of polymorphonuclear granulocytes in Polycythemia vera. J Lab Clin Med 12 1 :3 10, 1993

185. Le Blanc K, Berg 4 Palmblad J, Sarnuelsson J: Stimulus-specific defect in platelet aggregation in Polycythemia vera. Eur J Haematol 53 : 145, 1994

186. Van der Geer P, Hunter T, Lindberg RA: Receptor protein tyrosine kinases and their signal transduction pathways. Annu Rev Cell Bi01 10:25 1, 1994

187. Fisher EH, Charbonneau Y Tonks NK: Protein tyrosine phosphatases: A diverse farnily of intracellular and transmembrane enzymes. Science 253 :40 1, 199 1

188. Walton KM, Dixon JE: Protein tyrosine phosphatases. Annu Rev Biochem 62: 10 1, 1993

189. Pawson T, Gish GD: SH2 and SH3 dornains: From structure to hnction. Ce11 71:359, 1992

190. Dechert U, Moiter M, Harder KW, Matthews J, Owen P, Clark-Lewis 1, Thomas ML, Aebersold R, Jink FR: Comparison of the specificity of bactenally expressed cytoplasmic protein-tyrosine phosphatases SHP and SH-PTP2 towards synthetic phosphopeptide substrates. Eur J Biochem 23 1 :673, 1995

19 1 . Feng G-S. Hui C-C, Pawson T: SHZcontaining phosphotyrosine phosphatase as a target of protein-tyrosine kinases. Science 259: 1607, 1993

192. Matthews RT, Bowne DB, Flores E, Thomas ML: Characterization of hematopoietic intracellular protein tyrosine phosphatases: Description of a phosphatase containing an SH2 domain and another enriched in proline-, glutamic acid-, senne-, and threonine-rich sequences. Mol Ce11 Bi01 I2:2396, 1992

193. Lorenz U, Ravichandran KS, Pei D, Walsh CT, Burakoff SJ, Neel BG: Lck- dependent tyrosyl phosphorylation of the phosphotyrosine phosphatase SH-PTP 1 in murine T cells. Mol Ce11 Bi01 14: 1 824, 1994

194. Ahmad S, Banville D, Zhao Z, Fischer EH, Shen S-H: A widely expressed human protein-tyrosine phosphatase containing src homology 2 domains. Proc Nat1 Acad Sci USA90:2197, 1993

195. Perkins LA Larsen 1, Pemmon N: Corkscrew encodes a putative protein tyrosine phosphatase that hnctions to transduce the terminal signal fiom the receptor tyrosine kinase torso. Ce11 70:225, 1992

196. Vogel W, Lammers R, Juang J, Ullrich A: Activation of a phosphotyrosine phosphatase by tyrosine phosphorylation. Science 259: 161 1, 1993

197. Bennett AM, Tang TL, Sugimoto S, Walsh CT, Neel BG: Protein-tyrosine- phosphatase SHPTP2 couples platelet-derived growth factor receptor b to Ras. Proc Nat1 Acad Sci USA 91:7335, 1994

198. Noguchi T, Matozaki T, Horita K, Fujioka Y, Kasuga M: Role of SH-PTPZ, a protein-tyrosine phosphatease with Src homolgy 2 domains, in insulin-stimulated ras activation. Mol Ce11 Bi01 143674, 1994

199. Xiao S, Rose D, Sasaoka T, Maegawa H, Burke T, Rollen P, Shoelson S, Olefsky J: Syp (SH-PTP2) is a positive mediator of growth factor-stimulated mitogenic signal transduction. J Bi01 Chem 269:2 1 244, 1 994

200. Yi T, Mui AL-F, Krystal G, Ihle JN: Hematopoietic ce11 phosphatase associates with the interleukin-3 (IL-3) receptor b chain and dom-regulates IL-3 induced tyrosine phosphorylation and mitogenesis. Mol Ce11 Bi01 13 :7577, 1993

20 1. Spivak L, Fisher J, Isaacs Hankins WD: Protein kinases and phosphatases are involved in erythropoietin-mediated signal transduction. Exp Hematoi 20500, 1992

202. Yi T, Ihle JN: Association of hematopoietic ce11 phosphatase with c-kit after stimulation with c-kit ligand. Mol Ceil Bi01 13 :3350, 1993

203. Witthuhn BA, Quelle FW, Silvennoinen O, Yi T, Tang B. Miura O, Ihle IN: JAK2 associates with the erythropoietin receptor and is tyrosine phosphorylated and activated following stimulation with erythropoietin. Ce11 74:227, 1993

204. Klingmuller U, Lorenz U, Cantley LC, Neel BG, Lodish HF: Specific recruitment of SH-PTP 1 to the erythropoietin receptor causes inactivation of IAK2 and termination of proliferative signals. Ce11 80: 729, 1995

205. Shultz LD, Schweitzer PA, Rajan TV, Yi T, Ihle JN, Matthews RJ, Thomas ML, Beier DR: Mutations at the murine rnotheaten locus are with the hematopoietic cell protein-tyrosine phosphatase (Hcph) gene. Ce11 73 : 1445, 1993

206. Tsui HW, Sirninovich KA, de Souza L, Tsui FW: Motheaten and viable rnotheaten mice have mutations in the hematopoietic ce11 phosphatase gene. Nat Genet 4: 124, 1993

207. Van Zant G, Shultz LD: Hematologic abnormalities of the imrnunodeficient mouse mutant, viable rnotheaten. Exp Hematol 17:8 1, 1989

208 Chen HE, Chang S. Tmb T, Neel BG: Regulation of colony-stimulating factor 1 receptor signaling by SH2 domain-containing tyrosine phosphatase SHPTP 1. Mol Ce11 Bi01 16:3685, 1996

209. LoCicero Shen Q, Harris K: The tyrosine phosphatase SHP-I is deficient in Polycythemia vera peripherai blood mononuclear cells. Blood 88:437a, 1996 (abstr.)

2 10. Hinshelwood S, Asimakopoulos FA, Gilbert JGR. Delibrias CC, Gottgens B, Fearon DT, Green AR: The gene encoding hematopoietic ce11 phosphatase (Shp-1) is stmcturally and transcnptionally intact in Polycythernia vera. Blood 88:96a, 1996 (abstr.)

2 1 1 Fibach E, Manor D, Oppenheim A, Rachmilewitz EA: Proliferation and maturation of human erythroid progenitors in liquid culture. Blood 73 : 100, 1989

2 12. Wada Fi, Suda T, Miura Y, Kajii E, Ikemoto S, Yawata Y: Expression of major blood group antigens on human erythroid cells in a two phase liquid culture system. Blood 75505, 1990

IMAGE NALUATION TEST TARGET (QA-3)

APPLIED IN1AGE. lnc - = 1653 East Main Slreet - - a - Rochester. NY 14609 USA -- -- - - Phone: 71 6/482-0300 -- -- - - Fax: 71 61288-5989

O 1993. Appimd image. lx.. All Rqhts Aeserved