Characterization of the antitumor effects of interferons on prostate cancer cells

MISSING IMAGE

Material Information

Title:
Characterization of the antitumor effects of interferons on prostate cancer cells
Physical Description:
Book
Language:
English
Creator:
Hobeika, Amy Claudine, 1970-
Publication Date:

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
aleph - 28629701
oclc - 38855553
System ID:
AA00026803:00001


This item is only available as the following downloads:


Full Text










CHARACTERIZATION OF THE ANTITUMOR EFFECTS OF
INTERFERONS ON PROSTATE CANCER CELLS








By

AMY CLAUDINE HOBEIKA











A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY




UNIVERSITY OF FLORIDA

1997














ACKNOWLEDGEMENTS


I would like to thank my mentor, Dr. Howard M. Johnson, for taking

me into his laboratory and guiding me through my studies. He has given me

advice that will help me through my future adventures in research and has

been a great mentor. I also want to thank my committee members, Dr.

Edward Hoffmann, Dr. Janet Yamamoto, Dr. Tom Bobik, and Dr. Ammon

Peck, for their time, patience, and effort. Many thanks go to my fellow

graduate students and labmates--George, Mustafa, Joe, Kendra, Karrie, Scott,

Martez, Pedro, Taishi, Tim, and Wiggins. I want to thank them all for their

help and understanding, as well as their comical natures. I know it's not easy

to put up with me, and they make the lab a fun place to be. I want to say

special thanks to Prem and Barbara who have really played a major role in

helping me with everything over the past few years.

I especially want to acknowledge my family and friends. Although my

family has been far away and I rarely get to see them, they have always been

supportive. My parents are always willing to help me out, my sister Janine is

always willing to waste a few hours shooting the breeze with me on the

phone, and my brothers Claude and John still treat me like their little sister. I

would also like to mention Steve and Matt, my two closest friends, who are

always a great break from the world of science and always act impressed with



ii









my accomplishments, no matter how small. Finally, I must thank Adrian

Varela, my significant other, my companion, my boyfriend, or whatever you

want to call him, of the last four years. He has been a great help and is always

willing to tell me everything he knows. And I can't forget to mention

Clayton and Cocoa. They have been the best companions a girl could have-

they are unmatched in their loyalty and love and are a great comfort to me.

Who else would be so happy just to keep me company during late nights in

the lab?

































iii














TABLE OF CONTENTS



ACKNOWLEDGEMENTS ....................................................... .... ii

LIST OF TABLES .............................................................................. vi

LIST OF FIGURES ................................................................................ vii

A BSTRACT .................................................................................................viii

CHAPTERS

1 INTRODUCTION ........................................................................... 1

Discovery of Interferons ............................................ .......... 1
Biological Activities of IFNs .................................................... 3
Interferons and Disease ............................................ ........... 9
Interferons and Signal Transduction ....................................... 13
The Mammalian Cell Cycle ...................................................... 17
Adhesion Molecules and Growth Factor Receptors ............. 26
Experimental Rationale ...................................................... 28

2 MATERIALS AND METHODS .................................... 31

Reagents and Cell Lines ................................................... ... 31
Antiviral Assay ..................................................................... 32
Antiproliferative Assays .......................................... ........... 32
DNA Synthesis Assay ......................................................... 33
Cell Cycle Analysis .............................................................. 33
Immunoprecipitation and Immunoblotting ....................... 34
In Vitro Kinase Assays ............................................................. 35
Cellular Morphology ........................................................... 36
Flow Cytometric Analysis of Surface Receptors ......... .......... 36
Analysis of Growth Factor Production ............................... 37
Invasion Chamber Migration .................................... ....... 37

3 RESU LTS ..................................................................................... 38

IFNa Inhibition of the DU145 Cell Cycle ............................. 38


iv









Inhibition of Cell Growth ............................................. 38
Reduction in 3H-thymidine Incorporation ............... 38
Flow Cytometric Analysis of the Cell Cycle .............. 44
Inhibition of cdk2 Activity ........................................ 48
Analysis of Cyclin E and Cyclin D Dependent cdk2
Activity ....................................................... 50
Induction of CKI p21 ......................................... 50
IFNy Inhibition of the DU145 Cell Cycle ............................ 59
Inhibition of the Cell Cycle .................................... 59
Induction of p21WAF1 ................................................ 62
p21wAF Induction Causes an Increase in p21
Bound cdk2 and PCNA .................................... 62
IFNy Induction of a Change in Cell Phenotype ................. 67
Changes in Cell Morphology ................................. 67
Downregulation of the EGF Receptor .................... 70
Modulation of Cell Adhesion Molecules ................ 79
Reduction in Invasive Potential ............................. 81

4 DISCUSSION ............................................................................ 85

REFERENCE LIST ......................................................... .......... 92

BIOGRAPHICAL SKETCH ........................................................... 108


























V














LIST OF TABLES


Table a

I. An overview of the interferons ....................................... 2

II. General biological activities of IFNs ............................ 4

II. Interferon inducible proteins .................................... 8

IV. IFNs in disease therapy .................................................... 10

V. Cyclin-dependent kinase inhibitors .............................. 21

VI. IFNa inhibition of colony formation of DU145 cells .. 39

VII. Cell cycle analysis of IFNa treated DU145 cells ............. 45

VIII. Effect of IFNa treatment on cdk2 activity ................... 49

IX. Effect of IFNa on cyclin specific cdk2 activity ............. 51

X. Inhibition of DU145 colony formation by IFNy ........... 60

XI. Effects of IFNy on the DU145 cell cycle ....................... 61

XII. Effects of IFNa and IFNy on the expression of ICAM-1
and integrin a3 ................................... .......... 80














vi














LIST OF FIGURES
Figure page

1. The IFN signal transduction pathways ....................... 16

2. The mammalian cell cycle ................................ ...... 19

3. The role of p53 in the G1 checkpoint .......................... 24

4. IFNa inhibition of DU145 cellular proliferation ........ 41

5. Treatment of DU145 cells with IFNa inhibits
[3H]-thymidine incorporation .......................... 43

6. IFNa inhibits the progression of DU145 cells through
G1 and S phase of the cell cycle ....................... 47

7. IFNa does not affect cyclin E-cdk2 complex formation
in DU145 cells .................................... ......... 53

8. IFNa treatment increases and/or maintains p21
levels in synchronized DU145 cells ............... 56

9. IFNa induces p21 expression in DU145 cells ............. 58

10. IFNy induces p21 expression in DU145 cells ............. 64

11. Cdk2 and PCNA levels correspond to p21 induction
by IFN y ............................................ ............. 66

12. IFNy induces morphological changes in DU145 cells 69

13. IFNy downregulates the expression of the EGF
receptor ............................................................... 72

14. IFNa and IFNy reduce EGF production by DU145
cells ..................................................................... 76

15. EGF does not induce cyclin D1 in IFNy treated cells 78

16. IFNy decreases the invasive potential of DU145 cells 83


vii














Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

CHARACTERIZATION OF THE ANTITUMOR EFFECTS OF
INTERFERONS ON PROSTATE CANCER CELLS

By

Amy Claudine Hobeika

December, 1997

Chairperson: Howard M. Johnson
Major Department: Microbiology and Cell Science

Interferons (IFNs) function as important cytokines with a broad range

of effects on cells of various origins. Included in these effects are potent

antitumor capabilities. I investigated the antitumor effects of both type I and

type II IFNs on a human prostate cancer cell line DU145. DU145 cells are a

prostatic adenocarcinoma that have mutations in the tumor suppressor gene

products p53, pRB, KAI1, and PTEN. IFN alpha (IFNa) was found to inhibit

cell replication and colony formation of these cells. Analysis by flow

cytometry suggests that IFNa inhibited the progression of DU145 cells from

the G1 through S phase of the cell cycle. IFNa treatment of DU145 cells

reduced cyclin dependent kinase 2 (cdk2) activity. In particular, cyclin E

dependent cdk2 activity was inhibited by IFNa treatment. Consistent with

these data, IFNa was able to induce expression of the kinase inhibitor p21 in




viii









DU145 cells. These data support a role for p21 in mediating the

antiproliferative action of IFNa and describe a mechanism for IFN action.

I additionally examined the role of IFNy on the cell cycle of DU145

cells. IFNy was able to inhibit DU145 cell proliferation using a similar

mechanism of p21 induction. This induction of p21 correlated to an increase

in p21 bound cdk2 and PCNA. Interestingly, while both IFNa and IFNy were

found to inhibit the DU145 cell cycle, only IFNy was able to induce phenotypic

changes in these cells that resulted in an antitumor effect. IFNy treated cells

exhibited a change in cellular morphology when compared with IFNa and

untreated cells. These changes in morphology corresponded to a change in

several cell surface receptors such as the EGF receptor. Most significantly,

these phenotypic changes correlated with a decrease in the metastatic

potential of DU145 cells. These results suggest that IFNy is a superior

antitumor agent to IFNa for DU145 cells. The overall implications of these

findings describe a mechanism for the antitumor activities of both type I and

type II IFNs in a prostatic adenocarcinoma which has mutations in genes

closely involved in cell cycle control and cell adhesion.















ix














CHAPTER 1
INTRODUCTION


Discovery of Interferons

Interferons (IFNs) were first discovered in 1957 by Isaacs and

Lindenmann. They found that the treatment of chick chorio-allantoic

membrane fragments with heat inactivated influenza virus resulted in the

interference of the ability of fresh influenza virus to replicate in these tissues

(Isaacs and Lindenmann, 1957). These observations led to the identification

of a soluble factor produced in response to a viral challenge which they

termed "interferon". The long term results of these experiments have been

the discovery of several proteins which fall under the category of IFNs. These

proteins have been found in all higher vertebrates including humans and

have molecular weights ranging from 15 to 30 KD (Gastl and Huber, 1988).

The human IFN proteins have been characterized and their genes cloned and

expressed (Gray and Goeddel, 1982; Henco et al., 1985).

IFNs are a family of glycoproteins that are distinguishable based on

their cellular source, immunological reactivity, and induction of biological

responses. Original nomenclature identified the IFNs based on the cellular

sources by which they are produced as well as their antibody reactivity. The

current system of IFN nomenclature is based on the naming convention

agreed upon in 1980 whereby IFNs were named using Greek letters. This


1






2



Table I: An overview of the interferonsa


Type Member Main cellular source



I a & o leukocytesb

P fibroblastsb
T trophoblasts


II y lymphocytes



areviewed in Baron et al., 1991
bother cellular sources include epithelial cells, macrophages, and virally
infected cells






3


system distinguishes two main types of IFN, type I and type II. Type I IFNs

include alpha (a), beta (p), omega (w), and tau (t). Type II IFNs include only

one member, IFN gamma (y). An overview of the different IFNs as well as

their cellular sources is presented in Table 1. Currently, in humans, more

than 18 IFNa genes and pseudogenes have been described while only a single

gene has been found for either IFNp or IFNy (Sen and Lengyel, 1992). Several

genes have been found for IFN(o and IFNT (Bazer and Johnson, 1991; Sen and

Lengyel, 1992).



Biological Activities of IFNs

IFNs are cytokines that are produced and secreted by a variety of cell

types in response to several classes of inducers (Gastl and Huber, 1988). They

trigger a multitude of cellular responses including antiviral actions,

inhibition of cell growth and proliferation, regulation of the expression of

specific genes, modulation of cell differentiation, and immunoregulation

(Gastl and Huber, 1988). The specific effects of IFNs are dependent upon both

the type of IFN and the target cell. A review of the current information

concerning the different IFN functions is discussed below. New studies into

complex network of IFN effects are constantly being carried out and the

discovery of additional IFN regulated genes is ongoing. Table 2 lists the

general biological activities of the different types of IFN.

IFNs were originally discovered as a result of their potent antiviral

activity, and thus, this action of IFNs is perhaps the best understood. Recent






4



Table II: General biological activities of IFNsa



Antiviral
Antiproliferative
Regulation of cell growth
Immunoregulatory activities
Modulation of cell differentiation
Regulation of oncogene expression
Regulation of specific genes



"reviewed in Baron et al., 1991; Sen and Lengyel, 1992






5


experiments using gene knockout mice illustrate the critical role that IFNs

play in the defense against viral infections (Huang et al., 1993). The double-

stranded RNA synthesized as an intermediate in the replication of many

DNA and RNA viruses triggers IFN production in cells which is released into

the surrounding environment (Lengyel, 1982). The binding of released IFN

by specific receptors on neighboring cells protects these cells from viral

infection. Several general themes have emerged as to how IFNs are

responsible for cellular protection from viral infection. The IFN system can

impair various steps of viral replication, including penetration, uncoating,

transcription, translation, and the assembly of progeny viruses (Lengyel, 1982;

Petska et al., 1987; Samuel, 1988; Staeheli, 1990). Among the antiviral IFN

inducible genes are two enzymes that inhibit viral protein synthesis: P1/eIF-2

protein kinase (Samuel, 1979), and a 2'-5' oligoadenylate synthetase (2-5AS)

(Kerr, 1987). The 2-5AS enzymatically degrades viral RNA, reducing its

translation into viral proteins. The eIF-2 protein kinase reduces the

translation of viral proteins by decreasing the efficiency of the initiation of

protein synthesis. IFNs also induce the expression of the Mx family of

proteins (Staeheli et al., 1986). Induction of these proteins blocks replication

of influenza virus in cultured cells and in mice possibly by blocking viral

transcription (Staeheli, 1990). Some of the other antiviral effects of IFNs are

mediated through the activation of different aspects of the immune system,

as will be discussed later.






6

Soon after the characterization of IFNs as antiviral agents, it was

observed that, following exposure to IFN, the replication of some cell types

was inhibited (Pauker et al., 1962). IFNs are now known to exert

antiproliferative effects on a variety of cell types including normal cells,

immortalized cell lines, and tumor cells of various histological origins (Gastl

and Huber, 1988). Much of the current IFN research centers around the

complex system of IFN actions on the various proteins involved with cell

growth. IFNs inhibit cell replication by lengthening the time required for

progression of IFN treated cells through the cell cycle (Fleischman and

Fleischman, 1992). This was later found to be a result of IFNs affecting the G1

and/or S phases of the cell cycle (Creasey et al., 1980; Pontzer et al., 1991; Roos

et al., 1984; Tamm et al., 1987). IFNs exert these inhibitory activities by acting

on multiple cellular pathways. Several key cell cycle proteins are affected by

IFNs including phosphorylation state of the tumor suppressor gene product

retinoblastoma protein (pRB) and the expression of the proto-oncogene c-myc

(Einat et al., 1985a; Einat et al., 1985b; Jonak and Knight, 1984; Melamed et al.,

1993). IFNs may also inhibit cell replication by depleting cells of essential

metabolites. IFNs block the induction of the enzyme ornithine decarboxylase

(Sekar et al., 1983) and induce indoleamine 2,3 dioxygenase (Yasui et al., 1986).

A reduction in ornithine decarboxylase synthesis decreases the biosynthesis of

putrescine and other essential polyamines. Indoleamine 2,3 dioxygenase

causes the degradation of the essential amino acid tryptophan. Although

progress in the area of IFN antiproliferative actions has been made, the






7


mechanisms behind these effects, especially with respect to the cell cycle,

remain to be elucidated.

In addition to antiviral and antiproliferative activities, IFNs are also

important immunomodulatory cytokines and they exert numerous

immunoregulatory effects. IFNs upregulate the surface expression of the

major histocompatibility complex (MHC) class I and class II antigens on a

variety of cell types (Sen and Lengyel, 1992). MHC class I molecules are

required for cytotoxic T lymphocyte (CTL) activity and MHC class II molecules

are necessary for antigen presentation to helper T cells. IFNs can also increase

interleukin-2 (IL-2) receptor expression (Johnson and Farrar, 1983). A number

of immune cells are activated by IFNs. IFNs enhance CTL activity (Chen et

al., 1986; Herberman, 1986), activation of NK cells (Giedlund et al., 1987;

Weigent et al., 1983; Tuo et al., 1993), and activation of macrophage

phagocytosis (Baron et al., 1991). These cytokines induce resting CTL cells to

an activated state and also directly induce NK cells to exhibit enhanced

cytotoxic function. Numerous macrophage functions including tumor cell

cytotoxicity, antimicrobial activity, increase in killing of intracellular

pathogens, and antigen processing and presentation are activated by IFNs

(Degre and Bukholm, 1988; Black et al., 1988; Nathan et al., 1983; Neisel et al.,

1986). They also affect the production of antibodies by B cells and can regulate

the isotypes of the immunoglobulins secreted during the humoral immune

response (Finkelman et al., 1988; Johnson and Torres, 1983; Snapper et al.,






8




Table III: Interferon inducible proteinsa


Designation Characteristics Inducer



(2'-5')(A)nsynthetase (2'-5')(A)nsynthesis a, P > y
p68 kinase protein phosphorylation a, 3 > y
Indoleamine 2,3-dioxygenase tryptophan degradation y> a, p
P56 trp-tRNA synthetase > a, P
GBP/y67 guanylate binding y> a, p
Mx families anti-influenza virus a, P > y
IRF1/ISGF2 transcription factor a, p, y
IRF2 transcription factor a, p
MHC class I immune system a, p, y
MHC class II immune system Y
P2-microglobulin immune system a, p, y
IP10 platelet factor 4 related > a, p
200 family cluster of 6 genes a, p
6-16 unknown a, P > y
1-8/9-27 unknown a, p, y
C56,561 unknown a, p >
ISG54 unknown a, p >
ISG15 unknown a, p> y




areviewed in Sen and Lengyel, 1992






9

1988). In summary, IFNs play an important role in the network of immune

interactions that bring about and regulate immunoreactivity and the local

inflammatory response.

Other properties of IFNs include regulation of cellular differentiation

and antimicrobial and antiparasitic effects. As stated previously, there are two

types of IFNs classes- type I and type II. Although IFNs in both these classes

have similar biological functions and overlap in their effects on different cell

types, there are some differences. Table 3 is a partial list of IFN-inducible

proteins and which IFNs- IFNa, IFNp, and/or IFNy- are the inducers for each.

In general, IFNy appears to have the dominant immunoregulatory role while

IFNa and IFNp tend to be stronger inducers of antiviral proteins. One major

difference is the ability of IFNy, and not the type I IFNs, to upregulate MHC

class II (Houghton et al., 1984; Schwartz et al., 1985). Both type I and type II

IFNs have potent antiproliferative effects.



Interferons and Disease

The various biological properties of both type I and type II IFNs make

them potential therapies for a variety of medical disorders having viral,

malignant, and immune etiologies. IFNs have been studied for their

therapeutic efficacy in a number of conditions, and clinical investigations into

a number of diseases are ongoing (Baron et al., 1991; Johnson et al., 1994;

Stuart-Harris et al., 1992). As a result, IFNs have been approved by the Food

and Drug Administration (FDA) for the treatment of several diseases (Baron





10



Table IV: IFNs in disease therapy"


FDA approved currently in clinical
trials

IFNa chronic hepatitis throat warts caused by
B and C papillomavirus

hairy-cell leukemia chronic myelogenous
leukemia
kaposi's sarcoma
colon tumors
genital warts caused
by papillomavirus kidney tumors

bladder cancer

malignant melanoma



IFNp relapsing/remitting basal cell
multiple sclerosis carcinoma


IFNy chronic kidney tumors
granulomatous
disease leishmaniasis

chronic lymphocytic
leukemia

Hodgkin's disease


areviewed in Dorr, 1993; Johnson et al., 1994






11

et al., 1991; Dorr, 1993; Johnson et al., 1994). For example, due to potent

antiviral properties, IFNa is approved for the treatment of chronic hepatitis

type B and C (Dorr, 1993). The immunomodulatory functions of IFNp has

resulted in its approval for the treatment of one type of the autoimmune

disease multiple sclerosis (MS) (Johnson et al., 1994). Other diseases for which

IFN therapy is currently approved by the FDA include hairy cell leukemia,

Kaposi's sarcoma in the acquired immunodeficiency syndrome (AIDS), and

chronic granulomatous disease. Table 4 lists the current FDA approved IFN

therapies as well as others which are currently in clinical trials.

The antiproliferative and immunoregulatory actions of IFNs combined

with their ability to regulate proto-oncogenes and tumor supressor gene

products make them particularly attractive candidates for the treatment of

various cancers. The immunoregulatory roles of IFNs that result in

antitumor effects include enhancement of tumor cytotoxicity by macrophages,

natural killer cells, and T lymphocytes (Baron et al., 1991). In addition, IFN

enhanced expression of MHC antigens and tumor specific antigens result in

more efficient recognition and killing of tumor cells by cytotoxic leukocytes

(Baron et al., 1991). The induction of antibodies to the tumor cells may also be

enhanced by IFNs (Baron et al., 1991). The antiproliferative capabilities of

IFNs can directly inhibit replication of cells to decrease the growth rate of

tumors and malignant cells (Fleischman and Fleischman, 1992).

Furthermore, the ability of IFNs to modulate the expression of tumor

supressor genes and proto-oncogenes has strong implications concerning






12

their antiproliferative capabilities. The proto-oncogene c-myc has been found

to be overexpressed in a large number of cancers (Steiner et al., 1996). IFNs

capability to downregulate c-myc expression could contribute significantly to

control tumor growth, especially in tumors where c-myc overexpression is

involved in the malignancy. IFNs express potent antitumor effects both by

exerting direct antiproliferative effects on target tumor cells, through the

enhancement of immune responses, and by activating host cytotoxic effector

cells to more efficiently lyse target tumor cells.

One obstacle facing the widespread clinical use of IFNs is the toxic side

effects experienced by some populations of patients (Vial and Descotes, 1994).

Adverse effects associated with IFNs are usually acute effects that involve

"flu-like" symptoms that include fever, malaise, tachycardia, chills, headache,

arthralgias, and myalgias (Gauci, 1987; Quesada et al., 1986; Spiegal, 1987).

However, these symptoms are usually not treatment limiting and are

tolerable using symptomatic treatment (Vial and Descotes, 1994). The

symptoms are also reversible after reduction in IFN dosage. A recently

discovered type I IFN, IFNr, provides a non-toxic alternative. IFNt was

originally recognized as a pregnancy recognition hormone in ruminants

(Bazer and Johnson, 1991). It has now been shown to have similar biological

activities to the other type I IFNs but without the associated toxicity (Pontzer

et al., 1988; Pontzer et al., 1991, Soos and Johnson, 1995a; Soos et al., 1995b).

IFNr has been shown to block experimental allergic encephalomyelitis (EAE)






13

in mice and may provide an alternative therapy to IFNP for the treatment of

MS (Soos et al., 1995b).



Interferons and Signal Transduction

The binding of IFNs to their specific receptor is the first step in evoking

their biological responses. All type I IFNs bind to the same cell surface

receptor while IFNy binds to a separate, although similar, receptor (Branca,

1988). The binding of type I IFNs to their receptor brings together two receptor

chains, the IFNa/p receptor 1 chain and IFNa/p receptor 2.2 chain (Domanski

et al., 1995; Novick et al., 1994; Lutfalla et al., 1996; Uze et al., 1990). These

chains are induced to associate in the presence of ligand resulting in the

formation of a functionally active receptor which mediates type I IFN

signaling (Cohen et al., 1995; Darnell et al., 1994). Additionally, a third

component of the type I receptor is believed to exist due to the differential

effects of the type I IFNs (Croze et al., 1996). A receptor complex consisting of

several subunits may explain how the different type I IFNs elicit the

preferential induction of IFN specific genes while still binding the same

receptor (Petska et al., 1987; Rani et al., 1996). The IFNy receptor consists of

two integral membrane polypeptides that include a and p subunits (Aguet et

al., 1988; Farrar and Schreiber, 1993; Soh et al., 1994). The a subunit is

necessary for ligand binding while the P subunit participates in signal

transduction (Farrar and Schreiber, 1993). IFNy binds its receptor as a

homodimeric ligand which results in rapid dimerization of the receptor and






14

subsequent signal transduction (Fountoulakis et al., 1992; Greenlund et al.,

1993). As a result of ligand binding with both the type I and type II receptors,

signal transduction is surprisingly rapid. Within fifteen minutes,

transcription of chromosomal genes is enhanced without the need for new

protein synthesis (Friedman et al., 1984; Larner et al., 1984; Larner et al., 1986;

Levy and Darnell, 1990; Reich et al., 1987). This rapid transmission of signals

combined with the apparant lack of involvement of second messengers has

created an area of intense research in recent years concerning the signal

transduction of the IFNs.

The signal transduction pathway of the IFNs is direct in nature.

Stimulation of the type I and type II IFN receptors initiate the activation of a

class of tyrosine kinases known as Janus kinases or JAKs (Darnell et al., 1994).

Type I IFNs activate two tyrosine kinase called tyrosine kinase 2 (tyk2) and

JAK1 (Silvennoinen et al., 1993; Velazquez et al., 1992). These activated

kinases subsequently phosphorylate three proteins known as STAT113,

STAT91, and STAT84 (Fu et al., 1992; Silvennoinen et al., 1993). The term

"STAT" stands for signal transducer and activator of transcription while the

number designates the molecular weight of the protein. The

phosphorylation of the STATs causes them to associate with one another to

form a complex which, combined with another protein referred to as p48, acts

as a transcription complex to directly control gene transcription (Fu et al.,

1990; Silvennoinen et al., 1993). The type I IFN STAT complex binds a target

sequence motif known as the interferon-stimulated response element (ISRE)






15




















Figure 1. The IFN signal transduction pathways. IFNs bind their receptors
and initiate intracellular signaling events that involve the JAK/STAT
system. IFNy binds as a homodimer and causes the phosphorylation of JAK1
and JAK2. JAK1 then phosphorylates a latent cytoplasmic transcription factor
STAT91. Phosphorylated STAT91 dimerizes and translocates to the nucleus
where it binds the specific promoter elements known as GAS. This binding
leads to the transcription of IFNy inducible genes. A similar mechanism is
used by type I IFNs but utilizes JAK1 and TYK2 which phosphorylate
STAT84/91 and STAT113. These STATs combine with p48 and enter the
nucleus where they activate transcription of genes containing ISRE sequences
in their promoter regions.







IN IFN



Jak2- Jakl Jakl -- Tyk2
cytoplasm

V V



91 113
4 p48



9i)nucleus GAS 91 SRE p48
nucleus GAS W(84 (84) ISRE P48 (84) F>






17

(Kessler et al., 1990). The IFNy activation pathway is similar in that it utilizes

two JAKs, one of which is shared by the type I system. JAK1 and JAK2

phosphorylate STAT91 which forms a homodimer and acts as a transcription

factor which binds to specific promoter elements designated as gamma

activation sites (GASs) (Decker et al., 1991; Greenlund et al., 1994; Igarashi et

al., 1994; Lew et al., 1991; Shuai et al., 1994). JAK/STAT systems similar to

those used by IFNs have been found to be utilized by a large array of cytokines

(Leaman et al., 1996). An overview of the IFN signal transduction pathways

is presented in Figure 1.



The Mammalian Cell Cycle

As previously mentioned, IFNs are believed to exert their

antiproliferative effects by inhibiting the progression of the cell cycle. This

may be due in part through the modulation of proteins involved in cell cycle

regulation. The mechanism behind the anticellular effects of IFNs is poorly

understood. The regulation of the cell cycle is particularly relevant in the

study of IFNs in cancer therapy.

The mammalian cell cycle consists of five phases: GO, G1, S, G2, and M

(Figure 2) (reviewed in Grana and Reddy, 1995). The GO represents the

quiescent phase where cells are considered to "exit" the cycle and remain in a

non-replicating state. Once the cell is stimulated by growth factors or

mitogens to re-enter the cell cycle, it is in the G1 phase where it prepares for

DNA replication. In the S phase, cellular DNA is replicated. The G2 phase






18






















Figure 2. The mammalian cell cycle. The cell cycle consists of five phases GO,
G1, S, G2, and M. The cdk-cyclin complexes are outlined and listed next to
each cell cycle phase.








Cdkl-cyclin B
Cdkl-cyclin A
M : GO


G2


G1
Cdk2-cyclin D
Cdk4-cyclin D
\ Cdk6-cyclin D
S restriction Cdk6-cyclin D
point

Cdk2-cyclin A


Cdk2-cyclin E






20

prepares the cell for mitosis, which is followed by the M phase where the cell

physically divides. A multitude of cell proteins and biochemical pathways

coordinate the cell cycle and the regulation of the system is complex. The

following provides a condensed overview of the key proteins involved in the

mammalian cell cycle.

The progression of the cell cycle is directly dependent upon the activity

of a set of protein kinases known as the cyclin dependent kinases (cdks). As

the name suggests, cdks are kinases which require regulatory subunits called

cyclins in order to be active (Morgan, 1995). Cyclins are named for their cyclic

nature of expression. They are specifically expressed during each phase of the

cell cycle and are promptly degraded as the cycle advances (reviewed in Koff et

al., 1992; Sherr, 1994). Figure 2 outlines the cdk-cyclin complexes which are

active during each cell cycle phase. Briefly, cdk4/cdk6 binds to cyclin D in G1,

cdk2 to cyclin E in G1 to S phase transition, cdk2 to cyclin A in S phase, and

cdkl (cdc2) to cyclin A/B in M (Koff et al., 1992; Nurse, 1990; Sherr, 1993).

These cdk-cyclin partners phosphorylate proteins which cause a cascade of

events resulting in the coordination of cell cycle progression. Cell cycle

control is carried out in large part by regulating cdk activity using a variety of

mechanisms and feedback loops.

One group of proteins that is directly involved in the regulation of cdk

activity are the cyclin dependent kinase inhibitors (CKIs) (Elledge and Harper,

1994). These inhibitors are induced in response to specific extracellular

signals and play an important role, via the inhibition of cdk activity, in






21




Table V: Cyclin-dependent kinase inhibitorsa


Inhibitors
Cell cycle Cyclin-cdk
phase complexes p15 p16 p18 p19 p21 p27


G1 Cdk4/6-cyclin D + + + + + +/-
G1/S Cdk2-cyclin E -- + +
S Cdk2-cyclin A -- -- + --
G2/M Cdkl-cyclin B +


areviewed in Grana and Reddy, 1995.






22

blocking the cell cycle (Elledge and Harper, 1994; Pines, 1994). CKIs inhibit cdk

function by physically binding cdk-cyclin complexes and interfering with

kinase activity. Table 5 lists the known CKIs and the cdk-cyclin complexes

they block. Currently CKIs are separated into two families. The first includes

p21wAF and p27'IP which share partial identity and are involved primarily

with blocking G1 and S phases of the cell cycle (Harper et al., 1993; Toyoshima

and Hunter, 1994; Xiong et al., 1993). Both p2wAF and p27IaP" are present in

quiescent cells and can cause G1 arrest. p21WlA was the first CKI to be

identified and is considered to be a universal CKI (El-Deiry et al., 1993; Gu et

al., 1993). P27KP has been found to be induced in response to TGFp (Polyak et

al., 1994). The other family of CKIs is the INK family which includes p15, p16,

p18 and p19 (Guan et al., 1994; Hannon and Beach, 1994; Hirai et al., 1995;

Serrano et al., 1993). These CKIs are important for blocking cdk4/cdk6 activity

in early G1 phase and can also cause G1 arrest (Hirai et al., 1995; Serrano et al.,

1993). Interestingly, p16 has been designated a tumor suppressor gene since it

has been found that mice with deletions in the p16 gene have a dramatic

increase in tumor formation (Serrano et al., 1996). Other CKIs are potential

candidates for tumor suppressor genes due to their regulatory impact on the

cell cycle.

The prototypic tumor suppressor gene product p53 is a transcription

factor that has multiple roles in cell cycle regulation, and for this reason is

one of the most commonly mutated proteins found in malignant cells

(reviewed in Levine et al., 1991). Many of the functions of p53 in cell






23






















Figure 3. The role of p53 in the G1 checkpoint. DNA damage induces p53
expression. p53 can then lead to the induction of p21WAF, apoptosis, or
facilitate DNA repair. The induction of p21WAF' results in the inhibition of
cdk-cyclin complexes including cdk2-cyclin E. This delay in the progression of
the cell cycle allows time for DNA repair before DNA replication in the S
phase.








Apoptosis
DNA repair





p53 0 p21WAF1




DNA P21 I--
damage I Cdk2 cyclinE


G ]Cdk2 IcyclinA
M S






G2






25

replication have been determined, one of which is its role in cell cycle

checkpoints (Kastan et al., 1992, Kuerbitz et al., 1992). Cell cycle checkpoints

represent the coordination of the cell cycle machinery with the biochemical

pathways that respond to DNA damage and restore its structure (Kaufman et

al., 1995). Checkpoints at G1 and G2 phases delay the cell cycle and provide

more time for repair before the critical phases of DNA replication and

mitosis. DNA damage in cells induces p53 expression which causes G1 arrest

(Kuerbitz et al., 1992). It has been found that this arrest is, at least in part, due

to p53 induction of the CKI p21WFl (El-Deiry et al., 1993; El-Deiry et al., 1994).

p21waF blocks cdk2-cyclin E activity and stops the cell at the G1 checkpoint

(Xiong et al., 1993). Another function of p21 is to bind the proliferating cell

nuclear antigen (PCNA) and block DNA replication (Flores-Rozas et al., 1994;

Waga et al., 1994). PCNA is an essential component of the DNA replication

machinery (Kelman, 1997). p53 is also believed to play a role in the G2

checkpoint and mitosis (Guillouf et al., 1995). However, the mechanism

behind this checkpoint is not fully understood. p53 has other roles in the cell,

including involvement in apoptosis. An overview is provided in Figure 3.

pRB and c-myc are two other proteins that are commonly found to be

abnormally expressed in malignant cells (reviewed in Levine, 1993; Marcu et

al., 1992). They have also been found to be modulated by IFNs. pRB is a

tumor supressor that acts as the main target for G1 and S phase cdk-cyclin

phosphorylation (Cobrinik et al., 1992). The G1 cdk-cyclin complexes

phosphorylate pRB to a hyperphosphorylated state (Grana and Reddy, 1995).






26

In this state, pRB is unable to bind the E2F transcription factor. E2F is then

free to transcribe a number of genes that are involved in cell replication. C-

myc is an oncogene whose overexpression leads to uncontrolled cell growth

(Marcu et al., 1992). It binds with its partner, max, to form a heterodimeric

transcription factor (Amati et al., 1993). This dimer transcribes a number of

genes that promote cell growth. By blocking c-myc expression and the

phosphorylation of pRB, IFNs can regulate the transcription of proteins that

promote cell replication.



Adhesion Molecules and Growth Factor Receptors

Adhesion molecules encompass another area of intense research due

to their significance in tissue development, tumor development, and the

immune response. These surface receptors play critical roles in cell-cell and

cell-extracellular matrix (ECM) interactions. Pathologic alterations in these

adhesion properties underlie many of the phenotypic changes associated with

tumor progression, including changes in cell morphology, migration, tissue

invasiveness, and metastatic potential (reviewed in Hannigan and Dedhar,

1997). There are several classes of cellular adhesion molecules (CAMs):

integrins, cadherins, selectins, and the immunoglobulin superfamily. The

integrins comprise a family of widely expressed transmembrane receptors that

are expressed on all cell types and mediate cell-cell and cell-ECM interactions

(Hynes, 1992). They are named according to the combination of different a

and P subunits which form a heterodimeric receptor. This pairing of different






27

subunits provides receptor diversity. The expression of certain integrin

subunits is tissue or cell specific, such as with the integrin aLP2 (LFA-1) whose

expression is limited to leukocytes (Stewart et al., 1995). The members of the

immunoglobulin superfamily contain typical immunoglobulin-like domains

in the extracellular portion of the molecule and are expressed on a variety of

cell types (Hannigan and Dedhar, 1997). They play an important role in the

inflammatory immune response and modulate tumor spread by regulating

the interaction of circulating tumor cells with host immunocytes. For

example, the intercellular adhesion molecule-1 (ICAM-1) is a member that

can bind the integrin receptor LFA-1 (Marlin and Springer, 1987). LFA-1 is

found on leukocytes including NK cells and CTLs (Makgoba et al., 1988; Dana

and Arnaout, 1994). ICAM-1 and LFA-1 binding can form stable cytolytic

conjugates between tumor cells and cells of the immune system (Hannigan

and Dedhar, 1997). The selectins and cadherins are involved with cell-cell

interactions via carbohydrate moieties and homophilic binding, respectively

(Hart, 1996).

Growth factors can influence constitutive activation of growth

promoting pathways in cancer cells and can modulate cell phenotype

(reviewed in Aaronson, 1991). A large array of factors have been discovered

that affect the growth of virtually all cell types which can act as positive or

negative modulators of cell proliferation and influence differentiation. Many

growth factors cause cells in the GO phase to re-enter the cell cycle (Pledger et

al., 1977). For this reason, several oncogenes encode growth factors and






28

tyrosine kinase receptors that participate in mitogenic signaling (Bishop,

1991). There is much evidence for genetic aberrations affecting growth factors

and their receptors in human malignancies. Among growth factor receptors,

the most frequently implicated in human cancer are the members of the EGF

receptor (EGFR) family (Aaronson, 1991). The EGFR is a tyrosine kinase

receptor that binds both EGF and TGFa ligands. The EGFR gene is often

amplified or overexpressed in squamous cell carcinomas and glioblastomas

(Libermann et al., 1985; Yamamoto et al., 1986), and EGFR expression has been

linked to poor patient prognosis in other malignancies (Kristensen et al., 1996;

Nakopoulou et al., 1995). Control of the overexpression of growth factor

receptors and their ligands has several implications for cancer intervention.

One is the potential improvement in diagnosis and prognosis of cancer.

Possibilities for cancer therapy include effective means for targeting tumor

cells by blocking signal transduction or ligand function.



Experimental Rationale

IFNs are a group of proteins that trigger a multitude of cellular

responses including inhibition of cell growth, modulation of cell

differentiation, and immunoregulation (Gastl and Huber, 1988). For this

reason, many studies have looked at the potential antitumor effects of IFNs

using both in vitro and in vivo models of cancer. These studies have

resulted in the use of IFNs in the treatment of several cancers such as hairy

cell leukemia, CML, and kaposi's sarcoma (Dorr, 1993; Gutterman, 1994).






29

However, IFNs have not been successful in treating some other types of

malignancies. The lack of patient improvement seen in some clinical trials

may be attributed to inadequate knowledge of the underlying antitumor

effects of IFNs. Thus, elucidation of the antitumor mechanisms of IFNs

against a particular type of cancer cell are important for indicating which

cancers may be susceptible to IFN therapy. The work included in this

dissertation takes a fundamental look at the antitumor effects of type I and

type II IFNs using an in vitro system involving a human prostate

adenocarcinoma cell line, DU145.

Prostate cancer is the most commonly diagnosed malignancy in men

(Parker et al., 1996). Like other types of cancers, prostate cancer results from a

loss or mutation of regulatory factors of the cell cycle such as oncogenes and

tumor suppressor genes (Cavenee and White, 1995; Garick, 1994). Previous

studies have shown that human prostate cancer cell lines are sensitive to the

antiproliferative properties of IFNs (Nakajima et al., 1994; Sica et al., 1989).

Although these effects have been recognized, the mechanism behind this

cellular inhibition remains unclear.

DU145 is an interesting cell line to study the antitumor effects of IFNs

for a number of reasons. This cell line was established from a metastatic

lesion in a patient with advanced prostate cancer (Stone et al., 1978). In

addition, DU145 cells have characteristics associated with undifferentiated

malignant prostate cells. For example, these cells are androgen independent

(Stone et al., 1978), tumorigenic in nude mice (Bookstein et al., 1990), and






30

have mutations in several tumor suppressor gene products including p53,

pRB, PTEN, and KAI1 (Bookstein et al., 1990; Dong et al., 1995; Isaacs et al.,

1991; Li et al., 1997). Examination of the antitumor effects of IFNs on DU145

cells describes a potential mechanism for their regulatory capabilities on an

adenocarcinoma that has mutations in proteins closely linked to the cell cycle

and cell adhesion. Studies into the mechanisms of the antitumor effects of

IFNs give insight into the use of IFNs in cancer therapy. In addition, this

information may indicate which cytotoxic agents and cytokines may produce

synergistic combinations with IFNs and therefore provide successful clinical

therapies.













CHAPTER 2
MATERIALS AND METHODS


Reagents and Cell Lines

Purified human IFNa (specific activity 2 x 108 units/ml) was obtained

from Biosource International (Camarillo, CA). Purified human IFNy (specific

activity 4.75 x 107 units/mg) was obtained from Genzyme Diagnostics

(Cambridge, MA). WISH and DU145 cell lines were obtained from American

Type Culture Collection (ATCC, Rockville, MD). Complete media for DU145

cells consisted of Eagles minimal essential medium (EMEM) supplemented

with 5% fetal bovine serum (FBS), 200 U/ml penicillin, and 200 gg

streptomycin. Starvation medium for cell synchronization contained the

ingredients listed above supplemented with 0.5% FBS. Antibodies to cdk2,

cyclin E, cyclin D, PCNA, p21, p27, and p16 were obtained from Santa Cruz

Biotechnology, Inc. (Santa Cruz, CA). Antibodies to ICAM-1 and EGF receptor

were obtained from Pharmingen (San Diego, CA) for flow cytometric analysis

and from Transduction Laboratories (Lexington, KY) for immunoblotting.

Antibodies to integrin a3 were obtained from Oncogene Research Products

(Cambridge, MA).








31






32

Antiviral Assay

IFN activity is expressed in terms of antiviral units/ml as assessed in a

standard cytopathic effect assay (Familletti et al., 1981). Antiviral activity of

human IFNa was determined using the WISH cell line and vesicular

stomatitis virus (VSV). One antiviral unit caused a 50% reduction in

destruction of the monolayer.



Antiproliferative Assays

For colony inhibition studies, anticellular activity was examined using

a modification of a colony inhibition assay (Blalock et al., 1980). DU145 cells

were plated at 600 cells/well in a 24 well plate using complete medium with

or without various concentrations of IFNa or IFNy. Plates were incubated at

370C for 5-6 days to allow for colony formation. Colonies were stained with

crystal violet and counted.

IFN inhibition of cell number was determined by using DU145 cells

plated in complete medium at 1 x 105 cells/well in 6 well plates with or

without IFN. At various time points, cells were removed from flask using

0.25% trypsin-EDTA solution (Sigma Co., St. Louis, MO) washed 2 times with

phosphate buffered saline (PBS), and counted. Cell counts were performed

using a hemocytometer, and cell viability was assessed by trypan blue dye

exclusion (Blalock et al., 1980).






33

DNA Synthesis Assay

DU145 cells were seeded at 2 x 105 cells/well in 6 well plates using

starvation medium for 24 hours. Wells were then washed and replaced with

complete medium alone or medium containing 2500 units/ml IFNa. At 16,

20, and 24 hours, cells were harvested and counted as described above. Cells

were then reseeded into 96 well plates at 2.5 x 104 cells/well and pulsed with 1

gC [3H]thymidine (specific activity, 21 mCi/mg; 1 Ci= 37 Gbq) (Amersham) for

2 hours at 370C. Cells were then harvested on a model M12 Brandel cell

harvester (Gaithersburg, MD) and incorporation of [3H]thymidine was

determined using a liquid scintillation counter.



Cell Cycle Analysis

DU145 cells were synchronized in GO/G1 phase (sychronization was

assessed by flow cytometry analysis by determining the percentage of cells in

the GO phase of the cell cycle) by culturing in starvation medium for 24 hours.

Cells were seeded into 25 cm2 flasks (Sarstedt, Newton, NC) and grown in the

presence of medium alone or medium containing either IFNa or IFNy.

Plates were incubated at 370C in 5% CO2 for varying lengths of time.

Following incubation, cells were harvested and washed 2 times using sample

buffer (PBS containing 1% glucose) and were then counted. Following

centrifugation, cell pellets were resuspended in 100 pl sample buffer and cells

were fixed by adding 1 ml cold ethanol (95%) (-20 C) dropwise while

vortexing. Cells were left in ethanol at 40C for > 24 hours. For propidium






34

iodide (Sigma Co., St. Louis, MO) staining, the prepared cell samples were

washed 2 times with sample buffer and blotted dry. The cells were then

resuspended in 600 800 pl sample buffer containing 50 gg/ml propidium

iodide and 100 U/ml RNase A (Sigma Co.) and allowed to stain for up to 1

hour at room temperature. Samples were filtered through 44-Aum nylon

mesh and data from 30,000 events was acquired with a FACSort (Becton

Dickinson Immunocytometry Systems, San Jose, CA) using the LYSYS II

software system. Analysis of the cell cycle was carried out using CellFIT

software which determines the percentage of cells in each phase of the cell

cycle based on cells collected using the LYSYS II software system.



Immunoprecipitation and Immunoblotting

For preparation of cell lysates, each experiment used an equal number

of DU145 cells per sample in complete medium stimulated with IFNa or

IFNy were lysed at 4C for 20 minutes in 500 pl of ice-cold lysis buffer that

consisted of 50 mM Tris-HCI (pH 7.4), 150 mM NaC1, 2 mM EGTA, 2 mM

EDTA, 50 mM NaF, 20 mM p-glyceryl phosphate, 2 mM Na3VO4, 2 mM

dithiothreitol, leupeptin (10 gg/ml), pepstatin (10 gg/ml), aprotinin (10

pg/ml), benzamidine (5 pg/ml), 1 mM phenylmethanesulfonyl fluoride, 10%

(vol/vol) glycerol, and 1% (vol/vol) Nonidet P-40. A range of 3-6 x 106 cells

per sample were used with individual replicate experiments. Lysates were

then centrifuged at 14,000 x g for 10 minutes. Equal amounts of protein

(protein levels were determined using the BCA protein assay kit (Pierce,






35

Rockford, IL)) from cell lysates were either directly loaded onto a

polyacrylamide gel (a range of 100-175 gg per sample was used with

individual experiments) and electrophoresed or subsequently treated (a range

of 375-500 gg/500 pl was used with individual experiments) with 1-10 jgg of

antibody to the protein of interest for 1 hour at 4C. The lysate-antibody

complex was allowed to bind for an additional hour using 30-40 mg of protein

A sepharose. Samples were then washed three times with 1 ml lysis buffer

followed by a single wash with 1 ml of 50 mM Tris, pH 6.8. Complexes were

eluted in SDS-PAGE sample buffer (0.03 M Tris-HC1, pH 6.8, 10% glycerol,

2%(w/v) SDS, 5% p-mercaptoetanol, and 2.5% bromophenol blue) and

electroporesed through a Tris-HCl polyacrylamide gel with 4% stacking gel.

Following Western transfer, nitrocellulose membranes were probed with

antibodies to the proteins of interest and developed using the ECL

chemiluminescence system (Amersham). Densitometric analysis of

radiographic film using IA-200 Image Analysis Software was used to

determine percent difference between band intensities based on total pixel

value.



In Vitro Kinase Assays

Treatment of cell lysates with antibodies to cdk2, cyclin Dl, and cyclin E

were performed as described above. Antibody-protein conjugates were

washed 3 times with lysis buffer and twice with kinase buffer (50 mM Hepes,

pH 7.5, containing 1 mM EGTA, 10 mM MgC1,, 1 mM Na3VO4, 20 mM f-






36

glyceryl phosphate, 5 gM ATP) and incubated in kinase buffer containing 5 9g

histone H1, and 20 jCi of [32P]--ATP (specific activity 6000 Ci/mmol; 1 mCi=

37 Mbq) (Amersham) in a final volume of 30 jl at 300C for 10 minutes.

Following centrifugation, supernatant fluids (25 .l) were analyzed for histone

H1 phosphorylation by a filter-binding assay using centrifugal Pierce

phosphocellulose units, SpinZymeTM Format purchased from Pierce

(Rockford, IL) according to the manufacturer's instructions.



Cellular Morphology

DU145 cells in complete medium were treated, with or without IFN,

for 5 days in culture slides (Falcon, Becton Dickinson, Bedford, MA). Slides

were washed 3 times with phosphate buffered saline (PBS), fixed, and stained

with eosin-methylene blue. Slides were then analyzed by light microscopy.



Flow Cytometric Analysis of Surface Receptors

DU145 cells were treated with IFN as described above. Cells were then

harvested and washed 3 times with PBS. Equal number of cells per sample (8-

10 x 105 cells/sample) were washed 3 times with flow cytometry buffer (PBS

containing 0.1% Na azide, 5% FBS) and incubated for 1 hour at 40C with

antibodies specific for EGFR (2 gg/ml), ICAM-1 (1 gg/ml), or integrin a3 (4

gg/ml). After washing, cells were incubated for an additional hour with FITC

labeled secondary antibodies specific for mouse IgG1. Data from 30,000 events






37

were acquired as described above and analyzed using Median Fluorescence

Intensity (MFI) software.



Analysis of Growth Factor Production

DU145 cells were treated as described above for 5 days and reseeded at 1

x 105 cells/well into 6 well plates in EMEM (no FBS). Supernatant fluids were

harvested after 24 hours and used in a Cytokine Total ELISA kit (Intergen,

Purchase, NY) specific for human EGF.



Invasion Chamber Migration

DU145 cells were treated as described above for 5 days and reseeded,

without IFN or FBS, at 1 x 10' cells/well into Biocoat Matrigel invasion

chambers (Falcon, Becton Dickinson, Bedford, MA). The chambers were then

placed into 24 well plates containing EMEM supplemented with 20% FBS.

These plates were then incubated 36-48 hours at 370C. Membranes from each

chamber was then removed, subsequently fixed, and stained with eosin-

methylene blue. Total number of invasive cells was then determined.













CHAPTER 3
RESULTS

IFNa Inhibition of the DU145 Cell Cycle

Inhibition of Cell Growth

The investigation into the effects of IFNa on prostate cancer cells was

started by determining the antiproliferative properties of IFNa on DU145

cells, a human prostate cancer cell line. IFNa inhibited colony formation of

DU145 cells at low cell density (600 cells/well) in a dose dependent manner as

shown in Table VI. The antiproliferative effects of IFNa was also determined

by utilizing direct cell counts. DU145 cells were treated with 2500 units/ml of

IFNa and the overall reduction in cell number was determined (Figure 4). A

reduction in the rate of growth by approximately 50% was observed in IFN

treated cultures versus untreated cultures. These results indicate that IFNa

has antiproliferative activity on DU145 cells.



Reduction in 3H-Thymidine Incorporation

In examining the effects of IFNa on the DU145 cell cycle, the

incorporation of [3H]thymidine by cultures synchronized, by serum

starvation, into GO/G1 was first determined (Figure 5). The incorporation of

thymidine by cells is a measure of chromosomal replication, and is therefore

an indication of cellular activity in the S phase of the cell cycle (Tamm et al.,


38






39




Table VI: IFNa inhibition of colony formation of DU145 cells"



IFNa (units/ml) colonies/ well (mean+SD) inhibition (%)



0 77.7 4.2

625 60.3 + 4.7 22.4
1250 40.3 3.5 48.1
2500 33.7 3.2 56.6
5000 30.6 2.1 60.6



aDU145 cells were plated at 600 cells/well with various doses (units/ml) of
IFNa for 6 days and subsequently stained with crystal violet. Samples were
assessed in triplicate and results are expressed as the mean number of
colonies SD. Statistical significance was shown by Student's t-test between
the number of colonies in the presence or absence of 625 U/ml (p<0.05), 1250
U/ml (p<0.02), 2500 U/ml (p<0.006) and 5000 U/ml (p<0.006) of IFNa.






40





















Figure 4. IFNa inhibition of DU145 cellular proliferation. Synchronized
DU145 cells (1 x 105/well) were incubated with or without 2500 units/ml IFNa
for 48, 72, or 96 hours. Wells were then harvested and the total number of
live cells determined. Data are expressed as total number of cells per sample
SD for three replicates. Statistical significance was shown by Student's t-test
between the number if cells in the presence and absence of 2500 U/ml IFNa
for 48 (p<0.03), 72 (p<0.04) and 96 (p<0.03) hours.






41









12.5

Media

---- IFNa 2500 U/ml
10-




o 7.5-



5





2.5-




0-
0 48 72 96

Time (Hours)






42





















Figure 5. Treatment of DU145 cells with IFNa inhibits [3H]thymidine
incorporation. DU145 cells (1 x 105) synchronized in GO/G1 were incubated
with or without IFNa for 16, 20, or 24 hours. Cells were harvested and
reseeded into 96 well plates at 2.5 x 104 cells/well, incubated with [3H]-
thymidine for 2 hours, and harvested on a cell harvester. Data are expressed
as mean cpm SD for six replicates. Statistical significance was shown by
Student's t-test between [3H]thymidine incorporation by cells in the presence
and absence of 2500 U/ml IFNa for 16 (p<0.0006), 20 (p<0.0003) and 24
(p<0.0006) hours.






43










25000

U Media only

] IFNa 2500U/ml

20000-





15000-


U

10000-





5000





0 o
16 20 24

Time (Hours)






44


1987). DU145 cells were incubated with or without 2500 units/ml IFNa for 16,

20, or 24 hours and pulse-labeled at each time point with [3H]thymidine. At

16 hours, the amount of [3H]thymidine incorporated by IFN treated cells was

only 29% of the incorporation seen with the untreated cells. At 20 and 24

hours, IFN treated cells incorporated about 60% of the [3H]thymidine

incorporated by untreated cells. These reductions show that IFN treated cells

entered the S phase at least 4 hours later than untreated cells. Consistent

with this, at 20 and 24 hours, IFN treated cells reached a level of

[3H]thymidine incorporation seen with untreated cells at 16 hours. These data

suggest that IFNa inhibited the progression of DU145 cells from G1 through S

phase.



Flow Cytometric Analysis of the Cell Cycle

The inhibitory effects of IFNa on the DU145 cell cycle were further

examined using flow cytometry analysis (Table VII, Figure 6). DU145 cells

synchronized by serum starvation into GO/G1 were stimulated to enter the

cell cycle by serum addition in the presence or absence of 2500 units/ml IFNa.

As can be seen from Table VII, untreated cells rapidly advanced through the

GO/G1 and S phases, and, by 40 hours, cells had already completed one full

cycle and were again entering the S phase. IFN treated cells, however,

progressed more slowly. After 24 hours, 54% of the IFN treated cells were still

in the GO/G1 phase, while untreated cells had only 39% of cells in GO/G1.

The flow cytometry histograms (Figure 6) depict the state of the DU145 cell






45




Table VII: Cell cycle analysis of IFNa treated DU145 cellsa


cell cycle phase (%)

Time IFNa
(hours) (units/ml) GO/G1 S G2/M


0 0 63.0 18.5 18.5

16 0 54.9 33.6 11.5
2500 66.9 22.1 11.0

20 0 39.5 46.2 14.2
2500 57.8 30.9 11.3

24 0 39.2 35.4 25.4
2500 53.8 33.7 12.5

40 0 62.1 24.9 13.0
2500 48.7 32.7 18.6



aDU145 cells were treated with 0 or 2500 units/ml for 0, 16, 20, 24, or 40 hours.
Progression through the cell cycle was examined using propidium iodide
staining. Data from 30,000 events are presented as percentage of cells in each
stage of the cell cycle. Similar results were seen with two replicates of this
experiment.






46






















Figure 6. IFNa inhibits the progression of DU145 cells through G1 and S
phase of the cell cycle. DU145 cells were synchronized in GO/G1 by growing in
medium containing 0.5% FBS. Progress through the cell cycle was examined
using propidium iodide. Horizontal axis, relative fluorescence intensity;
vertical axis, number of cells. A, B, and C IFNa treated cells (2500 units/ml);
D, E, and F, untreated controls, at 16, 20, and 24 hours after the initiation of
culture, respectively. Similar results were seen with two replicates of this
experiment.






0 O 0ot, OR .80 0 09 O D v

; ~Lfr~~i
~77L~7J






48

cycle at 16, 20, and 24 hours in the presence and absence of IFN. A similar

pattern was seen with IFN treatment at 1000 units/ml (data not shown).

Consistent with the [3H]thymidine incorporation by IFNa treated DU145 cells,

flow cytometry analysis showed that IFNa inhibited progression of prostate

cancer cells through G1 and early S phase of the cell cycle.



Inhibition of Cdk2 Activity

In order to determine the relationship of the phase of the cell cycle

specifically inhibited by IFNa in the context of cyclin dependent kinase (cdk)

activity, the activity of a cdk, cdk2, that is active during the G1 and S phases of

the cell cycle was examined (Pines and Hunter, 1995). DU145 cells were

cultured with or without IFNa for 16 or 24 hours and subsequently harvested.

For each time point, cdk2 was immunoprecipitated and function assessed by

histone Hi-dependent kinase activity (Table VIII). DU145 cells synchronized

to GO/G1 had low cdk2 activity at 16 hours. However, by 24 hours, these cells

increased their cdk2 activity by greater than 10-fold, while IFN treated cells

showed only a 4-fold increase, resulting in a 74% reduction of cdk2 activity

over the control. The data show that IFNa is able to reduce the activity of a

cdk specific for the G1 and S phases of the cell cycle in prostate cancer cells,

and thus inhibit the progression through the cell cycle.






49




Table VIII: Effect of IFNa treatment on cdk2 activitya



Time IFNa CPM reduction
(hours) (units/ml) (%)



16 0 8042 --
2500 6138 23.7

24 0 94880 -
2500 24570 74.1



aDU145 cells were treated with 2500 units/ml IFNa for 16 or 24 hours. Cyclin
dependent kinase 2 activity was assessed by histone Hi-dependent kinase
activity. Cdk2 activity is represented as cpm with corresponding percent
reduction. Similar data with the same patterns were observed in three
repeats of this experiment.






50

Analysis of Cyclin E and Cyclin D Dependent Cdk2 Activity

To further examine the stage of the cell cycle that is regulated by IFNa,

the cyclin specificity of the inhibition of cdk2 activity was determined. Cdk2

binds cyclin D1 and cyclin E during the G1 phase and G1 to S phase transition,

respectively (reviewed in Pines and Hunter, 1995). Cell lysates from DU145

cells treated with IFNa for 16 or 24 hours were immunoprecipitated using

antibodies specific for cyclin D1 or cyclin E and cdk2 activity was subsequently

assessed (Table IX). IFNa treated cells showed up to a 38% reduction of cyclin

E-cdk2 activity over the control, but did not show consistent inhibition of

cyclin D1-cdk2 activity. Of the 74% IFNa induced reduction in overall cdk2

activity (Table VIII), 38% is apparently due to the reduction of cyclin E

dependent cdk2 activity (Table IX). The remaining inhibition of cdk2 activity

by IFNa is probably due, at least in part, to inhibition of activity in the cyclin

A-cdk2 complex (Tiefenbrun et al., 1996). I next immunoprecipitated cyclin E

and immunoblotted using a cdk2 antibody in order to determine relative

amounts of cdk2 complexed to cyclin E. Figure 7 shows that IFNa treatment

of cells did not affect the levels of cdk2 complexed with cyclin E.

Consequently, IFNa inhibition of the activity of the cyclin E-cdk2 complex did

not affect the formation of the cyclin E-cdk2 complex in DU145 cells.



Induction of CKI p21wal

The IFNa induced decrease in cyclin E-cdk2 activity suggested that a

kinase inhibitor may be involved. The CKI p21 is known to bind cyclin E-






51




Table IX: Effect of IFNa treatment on cyclin specific cdk2 activitya


Cyclin Time IFNa CPM reduction
(hours) (units/ml) (%)


Expt. 1

E 16 0 29507 -
2500 22861 22.5

24 0 28607 --
2500 17742 37.9

D1 16 0 5321 --
2500 3881 27.1

24 0 4642
2500 5291

Expt. 2

E 16 0 15954 -
2500 11761 26.3

24 0 14296 --
2500 9907 30.1

D1 16 0 4786 -
2500 4936

24 0 5666
2500 7818


aDU145 cells were treated with 0 or 2500 units/ml IFNa for 16 or 24 hours.
The cyclin-cdk2 complex was immunoprecipitated with antibodies specific for
either cyclin E or cyclin D1. Cdk2 activity was then assessed by histone H1-
dependent kinase activity. Cdk2 activity is represented as cpm with
corresponding percent reduction. Similar data with the same patterns were
observed in two repeats of this experiment.






52






















Figure 7. IFNa does not affect cyclin E-cdk2 complex formation in DU145
cells. The presence of cdk2 complexed with cyclin E was assessed by
immunoprecipitation of DU145 cell lysates (515 gg protein/ 500 lj for each
sample) with cyclin E antibodies and immunoblotting with antibodies specific
for cdk2, as described in 'Materials and Methods'. Lanes 1 and 2 represent
lysates from cells treated for 16 hours with 2500 units/ml IFNa or medium
alone, respectively. Lanes 3 and 4 are from cells treated for 24 hours with or
without IFN, respectively.






53





















1 2 34
IFNa: + + -

81.0--


41.5-

31.8-






54

cdk2 and block its activity (reviewed in Pines and Hunter, 1995). Serum

starvation of cells increases p21 levels and restimulation by serum a

reduction in p21 levels (Pines and Hunter, 1995). DU145 cells synchronized by

serum starvation into GO/G1 were therefore used to determine whether p21

is involved in inhibiting the G1 to S phase transition in IFN treated cells.

Cell lysates from DU145 cells treated with IFNa for 16 and 24 hours were

immunoprecipitated using p21 antibodies, and the presence of both p21 and

cdk2 in these samples was subsequently assessed by immunoblotting. As

shown in Figure 8A, lysates from cells treated with IFNa appeared to have a

greater expression of p21 than untreated cells. In the same

immunoprecipitates, at 16 and 24 hours, cdk2 protein levels were also higher

in IFNa treated cells than in untreated cells (Figure 8B), supporting the

conclusion that in IFNa treated cells increased levels of p21 were associated

with cdk2. This suggests that the expression of p21 in IFNa treated cells

played an important role in inhibiting the cdk2 activity in these cells.

To establish that IFNa treatment induces the expression of p21, cell

lysates from asynchronous DU145 cells that were treated with IFNa or

medium alone were assessed for p21 as described above. IFNa treatment

progressively induced the expression p21 (Figure 9A) over that seen in

untreated cells. Densitometric analysis (Figure 9B) of the p21 bands in Figure

9A showed that IFN treated cells had approximately twice the levels of p21

compared to untreated cells. Thus, IFNa inhibits the G1 to S phase transition

of the cell cycle by inducing p21 expression in a prostate cancer cells. Similar






55















Figure 8. A. IFNa treatment increases and/or maintains p21 levels in
synchronized DU145 cells. Cell lysates (512 gg protein/ 500 gl for each
sample) were immunoprecipitated and immunoblotted using p21 antibodies,
as described in 'Materials and Methods'. Lanes 1 and 2 represent lysates from
synchronized cells treated with 2500 U/ml IFNa or media alone, respectively,
for 16 hours. Lanes 3 and 4 are from cells treated for 24 hours with or without
IFN, respectively. The percent decrease in p21 levels for lanes 1, 2, 3, and 4
was 40.1%, 62.7%, 68.2%, and 70%, respectively, as determiend using
densitometric scanning of radiographic film. The percent decrease represents
the ratios of band intensities from DU145 cell lysates at initiation of cultures
(0 hours, data not shown) and lanes 1 through 4. B. Cdk2 levels correspond
to p21 expression in IFNa treated cells. The immunoblot from Figure 5A was
reanalyzed using antibodies specific for cdk2. The lane assignments are as
stated for 5A. The percent decrease in cdk2 protein levels was 64.3%, 81.5%,
83.2%, and 91% for lanes 1, 2, 3, and 4 respectively, as determined by
densitometric scanning as described for A.






56








A.
1 2 3 4
1234
IFNo + + -

32.7- --p2 1

17.7-














1 2 3 4 B.
IFNo + + -
7.0- dk2
44.1- I M
32.7- 0 1 --- Cdk2






57




















Figure 9. A. IFNa induces p21 expression in DU145 cells. Cell lysates from
DU145 cells (486 gg protein/ 500 gl for each sample) were immunopreciptated
and immunoblotted with antibodies specific for p21. Lanes 1 and 2 represent
lysates from cells treated for 16 hours with 2500 U/ml IFNa or media alone,
respectively. Lanes 3 and 4 represent lysates from cells treated for 24 hours
with or without IFN, respectively. B. The fold increase in p21 levels for 16
and 24 hours in IFN treated and untreated cells was determined by
densitometric scanning of radiographic film. The fold increase represents the
ratios of band intensities from DU145 cell lysates at initiation of cultures (0
hours, data not shown) and Lanes 1 through 4. Standard deviations represent
three separate densitometric readings from figure 9A.






58


A.




1 2 3 4
IFNa: + + -
44.1-
32.7-
rs p21
17.7-







B.
20

T IFNa 2500 U/ml

[] Media

15-











5-





0
16 24

Time (Hours)






59

experiments looked at the possible roles of two other CKIs, p27 and p16.

Induction of p27 or p16 expression was not seen with IFN treatment of DU145

cells (data not shown) suggesting a unique role for p21 in these effects. IFNa

inhibits the G1 to S phase transition of the cell cycle by inducing p21

expression in a prostate cancer cell line.



IFNv Inhibition of the DU145 Cell Cycle

Inhibition of the Cell Cycle

Studies using IFNy were begun by determining the antiproliferative

effects of IFNy on DU145 cells. IFNy was able to inhibit colony formation of

DU145 cells at low cell density in a dose dependent manner as shown in Table

X. Treatment of DU145 cells with IFNy showed significant inhibition of

colony formation at concentrations as low as 312 U/ml with up to 70%

inhibition at 10,000 U/ml. The effects of IFNy on the DU145 cell cycle were

then analyzed using flow cytometry (Table XI). DU145 cells synchronized by

serum starvation into GO/G1 were stimulated to enter the cell cycle by serum

addition in the presence or absence of 2500 U/ml IFNy. Untreated cells

rapidly advanced through the GO/G1 and S phases of the cell cycle, and, by 24

hours, these cells began returning to the G1. IFNy treated cells, however,

progressed more slowly. At 16 hours, 46% of the IFN treated cells remained

in G1 compared to only 33% of the untreated cells. By 24 hours, in contrast to

the untreated cells, the largest percentage of IFN treated cells were in the S

phase. A similar pattern was seen with 1000 U/ml IFNy (data not shown).






60




Table X: Inhibition of DU145 colony formation by IFNy a


IFNy (units/ml) colonies/well (meanS.D.) Inhibition (%)


0 110.5 8.7

312 74.3 3.8 32.6
625 62.7 + 3.1 43.8
1250 57.3 3.2 48.1
2500 51.3 4.0 53.6
5000 41.7 1.5 62.3
10,000 33.7 3.1 70.6


'DU145 cells were plated at 800 cells/well with various doses (units/ml) of
IFNy for 5 days and subsequently stained with crystal violet. Samples were
assessed in triplicate and results are expressed as the mean number of
colonies S.D. Statistical significance was shown by Student's t-test between
the number of colonies in the presence or absence of 312 U/ml (p<0.05), 625
U/ml (p<0.02), 1250 U/ml (p<0.01), 2500 U/ml (p<0.01), 5000 U/ml (p<0.005),
and 10,000 U/ml (p<0.005) of IFNy.






61




Table XI: Effects of IFNy on the DU145 cell cyclea




cell cycle phase (%)

Time IFNy
(hours) (units/ml) GO/G1 S G2/M



0 0 74.1 15.3 10.6

16 0 33.3 60.5 6.2
2500 45.6 44.3 10.0

20 0 26.9 53.6 19.5
2500 32.0 55.9 12.2

24 0 36.8 33.6 29.6
2500 34.1 40.7 25.2



"DU145 cells were treated with 0 or 2500 units/ml of IFNy for 0, 16, 20, or 24
hours. Progression through the cell cycle was examined using propidium
iodide staining, Data from 30,000 events are presented as percentage of cells
in each stage of the cell cycle. Similar results were seen with two replicates of
this experiment.






62

IFNy inhibits the proliferation of DU145 cells by slowing their progression

through the cell cycle.



Induction of p21wll

As previously mentioned, p21 is known to act at the G1 and S phases of

the mammalian cell cycle by directly binding to cdk-cyclin complexes that are

active at these stages and blocking their activity (Gartel et al., 1997). To

determine whether IFNy treatment induces p21 in DU145 cells, cell lysates

from asynchronous cells that were treated with IFNy or media alone were

assessed for p21 by immunoblotting (Figure 10). At the initiation of

treatment, rapidly dividing cells exhibited low levels of p21. After 20 and 40

hours of IFNy treatment, p21 levels appeared to be greatly increased over that

of the untreated cells. As seen with IFNa, this induction of p21 by IFNy was

unique since other cdk inhibitors such as p27 and p16 did not show similar

increases in response to IFN treatment (data not shown).



p21WAF Induction Causes an Increase in p21 Bound Cdk2 and PCNA

I further assessed the consequences of the induction of p21 by IFNy

using immunoprecipitation. Using antibodies specific for p21, cell lysates

from DU145 cells treated with medium alone or IFNy were

immunoprecipitated. These precipitates were then immunoblotted using

antibodies specific for p21, cdk2, and PCNA. Consistent with previous studies

on the role of p21 in the cell cycle, cdk2 and PCNA levels correlated with the






63






















Figure 10. IFNy induces p21 expression in DU145 cells. DU145 cell lysates (141
gg protein per lane) were immunoblotted with antibodies specific for p21.
Lane 1 represents cell lysate from dividing cells at the initiation of treatment
(time 0), lanes 2 and 3 are lysates from cells treated for 20 hours with 2500
units/ml IFNy or media alone, respectively, and lanes 4 and 5 are lysates from
cells treated 40 hours with or without IFNy, respectively. Densitometric
analysis of radiographic film showed a 2.1-fold and 2.6-fold difference in p21
levels between lanes 2 and 3 and lanes 4 and 5, respectively.






64





















1 2 3 4 5
IFNy: + + -



43---

32.3-


17-






65





















Figure 11. Cdk2 and PCNA levels correspond to p21 induction by IFNy.
DU145 cell lysates (463 gg protein/ 500 pl for each sample) were
immunoprecipitated using antibodies specific for p21 and subsequently
immunoblotted using antibodies for (a) PCNA, (b) cdk2, and (c) p21. For (a),
(b), and (c), lane 1 represents lysates from untreated cells at initiation of IFN
treatment while lanes 2 and 3 are from cells treated with 2500 units/ml IFNy
or media alone for 30 hours, respectively. Densitometric analysis of
radiographic film showed a 6.3-fold difference in PCNA levels (a), a 2.22-fold
difference in cdk2 levels (b), and 2.7-fold difference in p21 levels (c) between
lanes 2 and 3.






66


















123
IFNy: + -

43-
S- --.-PCNA
32.3-
A

i j ( <-cdk2
32.3-



. p2l
^'KJr~tl~l~fcw






67

level of p21 expression (Figure 11). The increase in p21 resulting from IFNy

treatment corresponded to an increase in p21 bound cdk2 and PCNA. Cdk2 is

a cyclin dependent kinase which is active in the G1 and S phases of the cell

cycle, while PCNA is an essential component of the DNA replication

machinery (Gartel et al., 1997; Kelman, 1997). By binding these proteins, p21

acts to inhibit cell replication, and the induction of p21 by IFNy suggests that

these aspects of cellular replication are inhibited. Based on the previous

studies with IFNa, both type I and type II IFNs appear to block the DU145 cell

cycle at the G1 and S phases via a similar mechanism of p21 induction.



IFNr Induction of a Change in Cell Phenotype

Changes in Cell Morphology

In the course of studies on the effects of IFNs on the cell cycle, it was

observed that IFNy treatment induced a change in the appearance of DU145

cells. I therefore looked at the impact of IFNa and IFNy on DU145 cellular

morphology. DU145 cells were treated with medium alone, IFNa, or IFNy for

5 days, subsequently fixed, and stained with eosin-methylene blue. Analysis

by light microscopy showed that while untreated and IFNa treated cells

retained normal tissue culture appearance, IFNy treated cells showed a

distinct morphological change (Figure 12). Untreated and IFNa treated cells

exhibited a rounded morphology typical of tumor cells growing in culture.

IFNy treated cells were less rounded with protuberances resulting in more of

a spindle shape. Changes in cellular morphology are often associated with a






68





















Figure 12. IFNy induces morphological changes in DU145 cells. DU145 cells
were treated with (a) medium alone, (b) 5000 units/ml IFNa, or (c) 5000
units/ml IFNy for 96 hours, subsequently stained with eosin-methylene blue,
and analyzed by light microscopy. Total magnification is 100X.






69



*

SIPp

0 40
46 0 VA








19'
--~stu *






70

change in cellular phenotype and differentiation. The DU145 cell line

displays many features common to undifferentiated metastatic tumor cells,

including loss of several functional tumor suppressor gene products as well

as being tumorigenic in nude mice (Bookstein et al., 1990; Dong et al., 1995;

Isaacs et al., 1991). The effects of IFNy on the morphology of DU145 cells

suggests that IFNy induces a phenotypic change and possibly differentiation of

these cells.



Downregulation of the EGF Receptor

Expression of the EGFR and its ligands has been associated with human

cancer cells of various origins and, in several instances, has been correlated

with the stage of differentiation of tumor cells (Cohen et al., 1994; Ioachim et

al., 1996; Nakopoulou et al., 1995; Stumm et al., 1996). Further, in some

malignancies, increased EGFR expression has been found to be an indicator of

poor patient prognosis (Almadori et al., 1995; Kristensen et al., 1996). The

DU145 cell line has previously been shown to express EGFR, as well as its

ligands EGF and TGFa, and, as a result, possesses an autocrine feedback loop

(Connolly and Rose, 1991). I looked at the effects of type I and type II IFNs on

the expression of the EGFR. DU145 cells were treated with medium alone,

IFNa, or IFNy for 5 days and the expression of the EGFR was analyzed by flow

cytometry using antibodies specific for the EGFR (Figure 13A). IFNy caused

greater than a 50% reduction in EGFR compared to untreated cells. No effect

on EGFR expression was seen with IFNa treatment. These effects were also






71



















Figure 13. IFNy downregulates the expression of the EGF receptor. DU145
cells were treated with medium alone, 5000 units/ml IFNa, or 5000 units/ml
IFNy for 5 days. (a) Cells were stained using antibodies specific for the EGFR
and analyzed by flow cytometry. Similar data with the same pattern were
observed in 3 repeats of this experiment. Data are expressed as mean
fluorescence intensity. (b) DU145 cell lysates (138 gg protein per lane) were
immunoblotted using antibodies specific for EGFR. Lane 1 represents lysate
from cells treated with IFNa, lane 2 is from cells treated with IFNy, and lane 3
is from untreated cells. Densitometric analysis of radiographic film showed a
80% decrease in EGFR levels between IFNy and untreated cells (lanes 2 and 3).






72









A.
20



17.5



15



12.5



S10-



. 7.5-



5-


2.5-



0
IFN alpha IFN gamma untreated






73

















B.



1 2 3
IFN: a Y -

200- *-EGFR
135-




41.9-






74

analyzed by western blot (Figure 13B). Again, while no change was seen with

IFNa, IFNy clearly downregulated the expression of EGFR. I then looked at

the expression of the EGFR ligand EGF (Figure 14). Both IFNa and IFNy

reduced EGF production with IFNa showing the more significant

downregulation. These results are interesting in that both IFNa and IFNy

reduced growth factor production and cellular replication, but only IFNy

affected the receptor expression.

To determine the significance of the decrease in receptor expression,

the impact of EGF on cells pretreated with IFNa and IFNy was examined.

EGF is known to stimulate quiescent cells to enter the cell cycle and increase

cyclin D1 expression (reviewed in Lavoie et al., 1996). Cyclin D1 is a proto-

oncogene that has been found to be overexpressed in a number of human

neoplasms (Lees and Harlow, 1995). Lysates from IFN treated or untreated

cells stimulated with EGF were assessed for cyclin D1 expression. Figure 15

shows that IFNy treated cells exhibited little expression of cyclin D1 even with

the addition of EGF to the cell culture medium. In contrast, IFNa treated cells

expressed higher levels of cyclin D1 and showed a modest increase in the

presence of EGF. A similar pattern was seen with untreated cells. Further,

IFNa treated cells in the presence of EGF increased in number by thirty

percent over those without EGF while no difference in cell number was seen

with IFNy and EGF (data not shown). These findings indicate that the

downregulation of EGFR expression on DU145 cells by IFNy results in a cell

type that is less sensitive to the growth enhancing effects of EGF. Previous






75






















Figure 14. IFNa and IFNy reduce EGF production by DU145 cells. DU145 cells
were treated with medium alone, 5000 units/ml IFNa, or 5000 units/ml IFNy
for 5 days. Cells were then harvested and reseeded into 6 well plates at 1 x 105
cells/well. Supematants were collected at 24 hours and analyzed by ELISA for
EGF. Data are expressed as mean concentration S.D. Statistical significance
was shown by Student's t-test between m.f.i. for cells treated with media alone
and IFNa (p<0.02).






76










7



6-



5-



S4-



Af3

0










O
2-N alpha IFN gamma untreated



1-




IFN alpha IFN gamma untreated






77




















Figure 15. EGF does not induce cyclin D1 in IFNy treated cells. DU145 cells
were treated for 5 days with either 5000 U/ml IFNa, 5000 U/ml IFNy or
medium alone followed by the addition of 0.5 ng/ml EGF to fresh culture
media and incubated for 6 hours. Cell lysates (155 gg protein per lane) were
then immunoblotted for the presence of cyclin D1. Densitometric analysis of
radiographic film showed a 31% increase between lanes 1 and 2, a 6% increase
between lanes 3 and 4, and a 23% increase between lanes 5 and 6 of cyclin D1
levels. In addition, there is a 61% increase in cyclin D1 between lanes 3 (IFNy)
and 5 (untreated).






78




















IFNa IFNY untreated
EGF:- + -+ +

cyclin
32- D1

17.8






79

studies have shown that in vitro invasiveness of DU145 cells and in vivo

progression of DU145 tumors in nude mice are modulated by EGFR mediated

signals (Prewett et al., 1996; Xie et al., 1995). The downregulation of EGFR by

IFNy in DU145 cells suggests that IFNy induces a less malignant phenotype

and decreases the invasive potential of this cell line.



Modulation of Adhesion Molecules

In addition to soluble factors, cell-cell and cell-extracellular matrix

(ECM) adhesion properties underlie many of the phenotypic changes

associated with tumor progression and metastatic potential (Hannigan and

Dedhar, 1997). The effects of IFNy on the expression of the adhesion

molecules ICAM-1 and integrin a3 were analyzed (Table XII). The ability to

evade the host immune response is characteristic of tumor cells. ICAM-1

plays a role in leukocyte adhesion by binding the LFA-1 surface receptor

found on cytotoxic T lymphocytes (CTL), B cells, and natural killer (NK) cells

(Dana and Arnaout, 1994; Makgoba et al., 1988). As a result, cells with high

ICAM-1 expression could be the target of a host immune response in vivo.

ICAM-1 has previously been found to be an immediate response gene

induced by IFNy (Caldenhoven et al., 1994). IFNy was found to induce ICAM-

1 expression in DU145 cells by 239% over that of untreated cells. An increase

in integrin a3 expression was also seen as a result of IFNy treatment (Table

XII). Integrins mediate interactions between cells and ECM, and previous

work with another prostate carcinoma cell line, PC-3, has shown that highly






80




Table XII: Effects of IFNa and IFNy on the expression of ICAM-1
and integrin a3"


ICAM-1 Integrin a3

m.f.i.S.D. increase m.f.i.S.D. increase
(%) (%)


untreated 153.2 4.2 --- 368.2 9.9


IFNa 127.2 2.8 --- 387.7 1.0 5.3%


IFNy 519.7 26 239% 457.2 + 12.7 24.2%



"DU145 cells were treated with either 5000 units/ml IFNa, 5000 units/ml
IFNy, or media alone for 5 days. Cells were stained with either antibodies
specific for ICAM-1 or integrin a3 and analyzed by flow cytometry. Data are
expressed as mean fluorescence intensity (m.f.i.) S.D for three replicates.
Statistical significance was shown by Student's t-test between m.f.i. for cells
treated with IFNy or media alone for ICAM-1 (p<0.03) and integrin a3
(p<0.01).






81

invasive PC-3 variants have low levels of integrin a3 expression (Dedhar et

al., 1993). Thus, upregulation of the ICAM-1 and integrin a3 adhesion

molecules suggests that IFNy induces a less tumorigenic and less metastatic

phenotype in DU145 cells.



Reduction in Invasive Potential

I next tested whether the modulation of surface receptors on DU145

cells actually correlated with a change in invasive potential. Invasion by

tumor cells is a crucial step in the multistage process of tumor spread and the

formation of metastasis (Liotta, 1987). Several in vitro systems have been

established to study the invasiveness of tumor cells. One commonly used

system utilizes insert chambers with separating filters coated with a layer

basement membrane matrix that contains components similar to the ECM.

These invasion chambers are suitable for studying cell invasion of malignant

cells (Albini et al., 1987). DU145 cells were treated for 5 days with or without

IFN. These cells were subsequently reseeded into invasion chambers which

contain membranes coated with a basement membrane matrix. The total

number of cells which migrated across the membrane were then determined

for each treatment group (Figure 15). IFNy treated cells showed a 50%

reduction in the number of invasive cells over that of the control.

Furthermore, only a slight reduction in invasion of IFNa treated cells was

seen. Thus, the reduction in the number of invasive cells for the DU145 cell

line was specific for IFNy compared to IFNa even though both IFNs inhibited






82






















Figure 16. IFNy decreases the invasive potential of DU145 cells. DU145 cells
were treated with medium alone, 5000 units/ml IFNa, or 5000 units/ml IFNy
for 5 days. Cells were then reseeded into invasion chamber inserts containing
matrigel basement membrane matrix. Cells that migrated across the matrix
were stained with eosin-methylene blue and counted. Data are expressed as
the mean number of cells + S.D for three replicates.






83









150-









100









50-









0
IFN alpha IFN gamma untreated






84


cell growth. These results imply that IFNy can both reduce cell growth and

metastasis of DU145 prostate cancer cells.













CHAPTER 4
DISCUSSION


Both type I and type II IFNs possess potent antitumor properties for a

variety of cell types by inducing a number of cellular responses. Previous

studies have looked at the antiproliferative effects of IFNs on cancer cells both

in vivo and in vitro, and have shown that IFNs directly act on tumor cells to

prolong their progression through the cell cycle (Fleischman and Fleischman,

1992). Indirect antitumor functions of IFNs include the activation of different

aspects of the immune system, which leads to the targeting of cancer cells by

cytolytic immune cells (Baron et al., 1991). The combination of these

antitumor properties has led to the use of IFNs in a number of human

malignancies. However, while IFNs are therapeutic with certain cancers, they

are not with some others. Many of the cases where IFNs have not been found

to be clinically useful may be a result of inadequate knowledge of the

mechanisms by which IFNs function. The mechanisms behind the

antiproliferative effects of IFNs are poorly understood. Previous work has

shown with some cell types IFNs inhibit the G1 and S phases of the cell cycle

(Creasey et al., 1980; Pontzer et al., 1991; Roos et al., 1984; Tamm et al., 1987)

and modulate the expression of c-myc and pRB (Einat et al., 1985a; Einat et al.,

1985b; Jonak and Knight, 1984; Melamed et al., 1993). The vast majority of

these studies have been carried out on the Daudi cell line, which is a B cell


85






86

lymphoma. Daudi cells are extremely sensitive to the antiproliferative effects

of IFNa and undergo G1 arrest upon treatment with IFN (reviewed in

Subramaniam et al., 1997). Few studies have used cells of other origins to

study the effects of IFNs on the cell cycle.

The DU145 cell line, as previously mentioned, is an interesting model

to study the mechanism of the regulation of the cell cycle by IFNs. These cells

were isolated from a metastatic lesion and are a poorly differentiated

adenocarcinoma (Stone et al., 1978). Most malignancies, including prostate

cancer, with high metastatic potential and poor differentiation of cells are

associated with poor patient prognosis (Garnick, 1994). Tumors of the

prostate that are well-differentiated are more likely to be confined to the

prostate gland and can be treated by removal of the prostate (Gamick, 1994). A

poorly differentiated prostatic adenocarcinoma is more likely to extend

beyond the prostate and metastasize. Currently, besides radiation, hormonal

therapy is the only common treatment of prostate cancer once metastasis has

occurred (Garnick, 1994).. The drawback to this therapy is that, at some point,

most metastatic tumors become resistant to hormonal therapy and, as a

result, progress rapidly. By studying the DU145 cell line, we can gain insight

into the mechanism of IFNs on a metastatic adenocarcinoma.

DU145 cells are sensitive to the antiproliferative effects of IFNa and

IFNy, and, consistent with previous studies on other cell lines, appear to be

affected at the G1 to S phase transition of the cell cycle. This inhibition of cell

replication is at least partly due to an increase in p21 expression and the






87

subsequent decrease in activity of the cyclin E-cdk2 complex. The interaction

of p21 with cyclin E-cdk2 blocks the phosphorylating activity of this complex

which is necessary for the G1 to S phase transition of the cell cycle and DNA

replication (Gartel et al., 1997; Pines and Hunter, 1995; Sherr, 1993). p21 also

inhibits DNA replication by binding PCNA (Kelman, 1997). By inducing p21

expression, IFNs are able to control different aspects of the cell cycle. These

results describe a possible mechanism for the antiproliferative affects of IFNs

on a prostate cancer cell line.

Many of the proteins involved in controlling the cell cycle have been

designated as products of tumor suppressor genes or oncogenes due to the

association of mutations of these genes with different types of cancers. DU145

cells have mutations in the tumor suppressor gene products p53, pRB, PTEN,

and KAI1 (Bookstein et al., 1990; Dong et al., 1995; Isaacs et al., 1991; Li et al.,

1997). Various oncogenes and tumor suppressor genes have been identified

in prostatic tumors including mutations in p53 and pRB (Netto and

Humphrey, 1994). This provides another characteristic of DU145 cells that

makes them interesting to study. Approximately fifty percent of all cancers

have mutations in the p53 genes, and pRB is frequently deregulated in a

variety of tumors (reviewed in Gangopadhyay et al., 1997; La Thangue, 1997).

The mutant p53 in DU145 cells means that, although p53 is known to induce

p21, the induction of p21 by IFNs in this case is independent of p53 status.

The pRB gene product is directly phosphorylated by cyclin E-cdk2 complexes

that are inhibited by p21 (Hatakeyama and Weinberg, 1995). Again, in the case






88

of DU145 cells, this aspect of p21 regulation of the cell cycle does not

contribute significantly to the inhibition of cell replication, because of the pRB

mutation. Other features of p21 regulation, including the binding of PCNA

and other potential cdk2 targets, are therefore most likely involved.

Although both IFNa and IFNy can induce p21, IFNy induced

biochemical changes associated with changes in cellular phenotype indicate

that it possesses other antitumor effects against DU145 cells not related to p21.

One significant effect that appears to be independent of p21 status is the

downregulation of the EGFR. The EGFR is the most studied of the tyrosine

kinase receptors and has long been associated with human tumorigenesis

(Helden and Ronnstrand, 1997). It is frequently found to be overexpressed or

mutated in several different human tumor types. Like other cells of

epidermal origin, prostate cells are stimulated by EGF to replicate. The DU145

cell line in particular has been shown to have an autocrine stimulatory loop

with the EGFR and its ligands (Connolly and Rose, 1991). Therefore, IFNy

regulation of the EGFR lends insight into the mechanism behind IFNy

control of cell replication and phenotype. Other surface receptors modulated

by IFNy, and not IFNa, are ICAM-1 and integrin a3. Integrin a3 has been

correlated with the metastatic potential of another prostatic adenocarcinoma

cell line PC-3. The direct involvement of the EGFR and integrin a3 in the

invasiveness of these cells remains to be elucidated. However, they could act

as important markers of a metastatic phenotype. The strong upregulation of

the ICAM-1 adhesion molecule by IFNy is another aspect of its antitumor






89

effects. An increase in ICAM-1 could allow these cells to be more easily

targeted by the immune system. ICAM-1 is the receptor for the integrin LFA-

1 which is found on the surface of CTLs, NK cells, and B cells (Dana and

Arnaout, 1994; Makgoba et al., 1988). The significance of this increase in cell-

cell interactions may be seen with in vivo metastasis. IFNy not only reduces

the invasiveness of the DU145 cells, but the few cells that do escape the tumor

site could be more readily attacked by the host immune system.

As stated above, since both IFNa and IFNy upregulate p21, the

phenotypic changes that result from IFNy must be independent of p21. It is

likely that IFNs influence multiple biochemical pathways that may or may

not overlap. Although neither IFNa or IFNy induced the CKIs p27 or p16, it

is possible that IFNs can affect other regulators and inhibitors that influence

tumor cell growth. Previous studies with Daudi cells have shown that IFNs

suppress the phosphorylation of pRB (Melamed et al., 1993). Although this

modulation of pRB is insignificant in DU145 cells, the cell cycle is still

inhibited. The ability of IFNs to affect multiple cellular pathways in order to

regulate cell growth allows for the potential of IFNs to affect a wide range of

cell types with various genetic mutations. For example, regulation of the EGF

production is significant for the regulation of adenocarcinomas as well as

cancers of other origins due to the number of genes and biochemical

pathways stimulated by the EGFR signal transduction. Based on the effects of

IFNs on DU145 cells, cancers with mutations in p21 and/or the EGFR, such as

the erb-B family of oncogenes, may not be the best candidates for IFN therapy.






90

Determining the mechanism behind IFNs antitumor capabilities is also

useful for finding potential combination therapies. By understanding the

pathways influenced by IFNs, cytotoxic agents that can produce synergistic

effects when combined with IFNs may be found.

Even though both IFNa and IFNy inhibit DU145 cell growth, the

phenotypic changes induced by IFNy suggest that it is superior to IFNa as an

antitumor agent for DU145 prostate cancer cells. This is in contrast to studies

carried out using the Daudi cell line (Subramaniam et al., in press). While

IFNa induces a strong G1 arrest, IFNy has no antiproliferative effects on

Daudi cells. There are several potential explanations for the differential

effects of IFNs on various cell lines. Possibilities include IFN receptor

expression, signal transduction pathways, and target genes. For this reason,

studies into the mechanism behind the cellular actions of IFNs are relevant.

The characterization of the antitumor effects of IFNs on DU145 cells

gives insight into the poorly defined mechanism of IFN antiproliferative and

antitumor functions. The DU145 cell line is a metastatic adenocarcinoma

with mutations in key tumor suppressor genes that are commonly defective

in a variety of human cancers. Even with these mutations and metastatic

potential, both IFNa and IFNy are able to exert potent antiproliferative

activities, with IFNy also inducing a phenotypic change that results in an

antitumor effect. Future studies will include further defining the mechanism

of IFNy effects on DU145 cells. These will include determining how IFNy

controls the expression of the EGFR. In addition, the possible involvement of






91

collagenase activity in the IFNy reduction of the DU145 invasive potential,

via the downregulation of the EGFR, may be determined. Finally, work with

nude mice will determine the effects on IFNy on the metastatic potential of

DU145 cells using an in vivo system. These studies will expand on the work

contained in this dissertation and provide further details of the mode of

action of IFNs on an adenocarcinoma cell line.




Full Text

PAGE 1

CHARACTERIZATION OF THE ANTITUMOR EFFECTS OF INTERFERONS ON PROSTATE CANCER CELLS By AMY CLAUDINE HOBEIKA A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1997

PAGE 2

ACKNOWLEDGEMENTS I would like to thank my mentor. Dr. Howard M. Johnson, for taking me into his laboratory and guiding me through my studies. He has given me advice that will help me through my future adventures in research and has been a great mentor. I also want to thank my committee members. Dr. Edward Hoffmann, Dr. Janet Yamamoto, Dr. Tom Bobik, and Dr. Ammon Peck, for their time, patience, and effort. Many thanks go to my fellow graduate students and labmates-George, Mustafa, Joe, Kendra, Karrie, Scott, Martez, Pedro, Taishi, Tim, and Wiggins. I want to thank them all for their help and understanding, as well as their comical natures. I know it's not easy to put up with me, and they make the lab a fun place to be. I want to say special thanks to Prem and Barbara who have really played a major role in helping me with everything over the past few years. I especially want to acknowledge my family and friends. Although my family has been far away and I rarely get to see them, they have always been supportive. My parents are always willing to help me out, my sister Janine is always willing to waste a few hours shooting the breeze with me on the phone, and my brothers Claude and John still treat me like their little sister. I would also like to mention Steve and Matt, my two closest friends, who are always a great break from the world of science and always act impressed with ii

PAGE 3

my accomplishments, no matter how small. Finally, I must thank Adrian Varela, my significant other, my companion, my boyfriend, or whatever you want to call him, of the last four years. He has been a great help and is always willing to tell me everything he knows. And I can't forget to mention Clayton and Cocoa. They have been the best companions a girl could havethey are unmatched in their loyalty and love and are a great comfort to me. Who else would be so happy just to keep me company during late nights in the lab? iii

PAGE 4

TABLE OF CONTENTS ACKNOWLEDGEMENTS ii LIST OF TABLES vi LIST OF FIGURES vii ABSTRACT viii CHAPTERS 1 INTRODUCTION 1 Discovery of Interferons 1 Biological Activities of IFNs 3 Interferons and Disease 9 Interferons and Signal Transduction 13 The Mammalian Cell Cycle 17 Adhesion Molecules and Growth Factor Receptors 26 Experimental Rationale 28 2 MATERIALS AND METHODS 31 Reagents and Cell Lines 31 Antiviral Assay 32 Antiproliferative Assays 32 DNA Synthesis Assay 33 Cell Cycle Analysis 33 Immunoprecipitation and Immunoblotting 34 In Vitro Kinase Assays 35 Cellular Morphology 36 Flow Cytometric Analysis of Surface Receptors 36 Analysis of Growth Factor Production 37 Invasion Chamber Migration 37 3 RESULTS 38 IFNa Inhibition of the DU145 Cell Cycle 38 iv

PAGE 5

Inhibition of Cell Growth 38 Reduction in ^H-thymidine Incorporation 38 Flow Cytometric Analysis of the Cell Cycle 44 Inhibition of cdk2 Activity 48 Analysis of Cyclin E and Cyclin D Dependent cdk2 Activity 50 Induction of CKI p21™ 50 IFNy Inhibition of the DU145 Cell Cycle 59 Inhibition of the Cell Cycle 59 Induction of p21^^" 62 p2;^wAPi Induction Causes an Increase in p21 Bound cdk2 and PCNA 62 IFNy Induction of a Change in Cell Phenotype 67 Changes in Cell Morphology 67 Downregulation of the EOF Receptor 70 Modulation of Cell Adhesion Molecules 79 Reduction in Invasive Potential 81 4 DISCUSSION 85 REFERENCE LIST 92 BIOGRAPHICAL SKETCH 108 V

PAGE 6

LIST OF TABLES Table pag e L An overview of the interferons 2 n. General biological activities of IFNs 4 in. Interferon inducible proteins 8 IV. IFNs in disease therapy 10 V. Cyclin-dependent kinase inhibitors 21 VI. IFNa inhibition of colony formation of DU145 cells .. 39 Vn. Cell cycle analysis of IFNa treated DU145 cells 45 VIII. Effect of IFNa treatment on cdk2 activity 49 DC. Effect of IFNa on cyclin specific cdk2 activity 51 X. Inhibition of DU145 colony formation by IFNy 60 XI. Effects of IFNy on the DU145 cell cycle 61 Xn. Effects of IFNa and IFNy on the expression of ICAM-1 and integrin a3 80 vi

PAGE 7

LIST OF FIGURES Fig ure page 1. The IFN signal transduction pathways 16 2. The mammalian cell cycle 19 3. The role of p53 in the Gl checkpoint 24 4. IFNa inhibition of DU145 cellular proliferation 41 5. Treatment of DU145 cells with IFNa ir\hibits pH]-thymidine incorporation 43 6. IFNa inhibits the progression of DU145 cells through Gl and S phase of the cell cycle 47 7. IFNa does not affect cyclin E-cdk2 complex formation in DU145 cells 53 8. IFNa treatment increases and /or maintains p21 levels in synchronized DU145 cells 56 9. IFNa induces p21 expression in DU145 cells 58 10. IFNy induces p21 expression in DU145 cells 64 11. Cdk2 and PCNA levels correspond to p21 induction by IFNy 66 12. IFNy induces morphological changes in DU145 cells 69 13. IFNy downregulates the expression of the EGF receptor 72 14. IFNa and IFNy reduce EGF production by DU145 cells 76 15. EGF does not induce cyclin Dl in IFNy treated cells 78 16. IFNy decreases the invasive potential of DU145 cells 83 vii

PAGE 8

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy CHARACTERIZATION OF THE ANTITUMOR EFFECTS OF INTERFERONS ON PROSTATE CANCER CELLS By Amy Claudine Hobeika December, 1997 Chairperson: Howard M. Johnson Major Department: Microbiology and Cell Science Interferons (IFNs) function as important cytokines with a broad range of effects on cells of various origins. Included in these effects are potent antitumor capabilities. I investigated the antitumor effects of both type I and type n IFNs on a human prostate cancer cell line DU145. DU145 cells are a prostatic adenocarcinoma that have mutations in the tumor suppressor gene products p53, pRB, KAIl, and PTEN. IFN alpha (IFNa) was found to inhibit cell replication and colony formation of these cells. Analysis by flow cytometry suggests that IFNa inhibited the progression of DU145 cells from the Gl through S phase of the cell cycle. IFNa treatment of DU145 cells reduced cyclin dependent kinase 2 (cdk2) activity. In particular, cyclin E dependent cdk2 activity was inhibited by IFNa treatment. Consistent with these data, IFNa was able to induce expression of the kinase inhibitor p21 in viii

PAGE 9

DU145 cells. These data support a role for p21 in mediating the antiproliferative action of IFNa and describe a mechanism for IFN action. I additionally examined the role of IFNy on the cell cycle of DU145 cells. IFNy was able to inhibit DU145 cell proliferation using a similar mechanism of p21 induction. This induction of p21 correlated to an increase in p21 bound cdk2 and PCNA. Interestingly, while both IFNa and IFNy were found to inhibit the DU145 cell cycle, only IFNy was able to induce phenotypic changes in these cells that resulted in an antitumor effect. IFNy treated cells exhibited a change in cellular morphology when compared with IFNa and untreated cells. These changes in morphology corresponded to a change in several cell surface receptors such as the EGF receptor. Most significantly, these phenotypic changes correlated with a decrease in the metastatic potential of DU145 cells. These results suggest that IFNy is a superior antitumor agent to IFNa for DU145 cells. The overall implications of these findings describe a mechanism for the antitumor activities of both type I and type II IFNs in a prostatic adenocarcinoma which has mutations in genes closely involved in cell cycle control and cell adhesion. ix

PAGE 10

CHAPTER 1 INTRODUCTION Discovery of Interferons Interferons (IFNs) were first discovered in 1957 by Isaacs and Lindenmann. They found that the treatment of chick chorio-allantoic membrane fragments with heat inactivated influenza virus resulted in the interference of the ability of fresh influenza virus to replicate in these tissues (Isaacs and Lindenmann, 1957). These observations led to the identification of a soluble factor produced in response to a viral challenge which they termed "interferon". The long term results of these experiments have been the discovery of several proteins which fall under the category of IFNs. These proteins have been found in all higher vertebrates including humans and have molecular weights ranging from 15 to 30 KD (Gastl and Huber, 1988). The human IFN proteins have been characterized and their genes cloned and expressed (Gray and Goeddel, 1982; Henco et al., 1985). IFNs are a family of glycoproteins that are distinguishable based on their cellular source, immunological reactivity, and induction of biological responses. Original nomenclature identified the IFNs based on the cellular sources by which they are produced as well as their antibody reactivity. The current system of IFN nomenclature is based on the naming convention agreed upon in 1980 whereby IFNs were named using Greek letters. This 1

PAGE 11

Table I: An overview of the interferons^ Type Member Main cellular source I a&co leukocytes'' p fibroblasts'' X trophoblasts II Y lymphocytes ^reviewed in Baron et al., 1991 ''other cellular sources include epithelial cells, macrophages, and virally infected cells

PAGE 12

3 system distinguishes two main types of IFN, type I and type II. Type I IFNs include alpha (a), beta (p), omega (co), and tau (x). Type 11 IFNs include only one member, IFN gamma (y). An overview of the different IFNs as well as their cellular sources is presented in Table 1. Currently, in humans, more than 18 IFNa genes and pseudogenes have been described while only a single gene has been found for either IFNp or IFNy (Sen and Lengyel, 1992). Several genes have been found for IFNco and IFNx (Bazer and Johnson, 1991; Sen and Lengyel, 1992). Biological Activities of IFNs IFNs are cytokines that are produced and secreted by a variety of cell types in response to several classes of inducers (Gastl and Huber, 1988). They trigger a multitude of cellular responses including antiviral actions, inhibition of cell growth and proliferation, regulation of the expression of specific genes, modulation of cell differentiation, and immunoregulation (Gastl and Huber, 1988). The specific effects of IFNs are dependent upon both the type of IFN and the target cell. A review of the current information concerning the different IFN functions is discussed below. New studies into complex network of IFN effects are constantly being carried out and the discovery of additional IFN regulated genes is ongoing. Table 2 lists the general biological activities of the different types of IFN. IFNs were originally discovered as a result of their potent antiviral activity, and thus, this action of IFNs is perhaps the best understood. Recent it

PAGE 13

4 Table II: General biological activities of IFNs^ Antiviral Antiproliferative Regulation of cell growth Immunoregulatory activities Modulation of cell differentiation Regulation of oncogene expression Regulation of specific genes ^reviewed in Baron et al., 1991; Sen and Lengyel, 1992

PAGE 14

5 experiments using gene knockout mice illustrate the critical role that IFNs play in the defense against viral infections (Huang et al., 1993). The doublestranded RNA synthesized as an intermediate in the replication of many DNA and RNA viruses triggers IFN production in cells which is released into the surrounding environment (Lengyel, 1982). The binding of released IFN by specific receptors on neighboring cells protects these cells from viral infection. Several general themes have emerged as to how IFNs are responsible for cellular protection from viral infection. The IFN system can impair various steps of viral replication, including penetration, uncoating, transcription, translation, and the assembly of progeny viruses (Lengyel, 1982; Petska et al, 1987; Samuel, 1988; Staeheli, 1990). Among the antiviral IFN inducible genes are two enzymes that irihibit viral protein synthesis: Pl/eIF-2 protein kinase (Samuel, 1979), and a 2'-5' oligoadenylate synthetase (2-5AS) (Kerr, 1987). The 2-5AS enzymatically degrades viral RNA, reducing its translation into viral proteins. The eIF-2 protein kinase reduces the translation of viral proteins by decreasing the efficiency of the initiation of protein synthesis. IFNs also induce the expression of the Mx family of proteins (Staeheli et al., 1986). Induction of these proteins blocks replication of influenza virus in cultured cells and in mice possibly by blocking viral transcription (Staeheli, 1990). Some of the other antiviral effects of IFNs are mediated through the activation of different aspects of the immune system, as will be discussed later.

PAGE 15

6 Soon after the characterization of IFNs as antiviral agents, it was observed that, following exposure to IFN, the replication of some cell types was iiihibited (Pauker et al., 1962). IFNs are now known to exert antiproliferative effects on a variety of cell types including normal cells, immortalized cell lines, and tumor cells of various histological origins (Gastl and Huber, 1988). Much of the current IFN research centers around the complex system of IFN actions on the various proteins involved with cell growth. IFNs inhibit cell replication by lengthening the time required for progression of IFN treated cells through the cell cycle (Fleischman and Fleischman, 1992). This was later found to be a result of IFNs affecting the Gl and/or S phases of the cell cycle (Creasey et al., 1980; Pontzer et al., 1991; Roos et al., 1984; Tamm et al., 1987). IFNs exert these inhibitory activities by acting on multiple cellular pathways. Several key cell cycle proteins are affected by IFNs including phosphorylation state of the tumor suppressor gene product retinoblastoma protein (pRB) and the expression of the proto-oncogene c-myc (Einat et al., 1985a; Einat et al., 1985b; Jonak and Knight, 1984; Melamed et al., 1993). IFNs may also inhibit cell replication by depleting cells of essential metabolites. IFNs block the induction of the enzyme ornithine decarboxylase (Sekar et al., 1983) and induce indoleamine 2,3 dioxygenase (Yasui et al, 1986). A reduction in ornithine decarboxylase synthesis decreases the biosynthesis of putrescine and other essential polyamines. Indoleamine 2,3 dioxygenase causes the degradation of the essential amino acid tryptophan. Although progress in the area of IFN antiproliferative actions has been made, the

PAGE 16

7 mechanisms behind these effects, especially with respect to the cell cycle, remain to be elucidated. In addition to antiviral and antiproliferative activities, IFNs are also important immunomodulatory cytokines and they exert numerous immunoregulatory effects. IFNs upregulate the surface expression of the major histocompatibility complex (MHC) class I and class n antigens on a variety of cell types (Sen and Lengyel, 1992). MHC class I molecules are required for cytotoxic T lymphocyte (CTL) activity and MHC class n molecules are necessary for antigen presentation to helper T cells. IFNs can also increase interleukin-2 (IL-2) receptor expression (Johnson and Farrar, 1983). A number of immune cells are activated by IFNs. IFNs enhance CTL activity (Chen et al., 1986; Herberman, 1986), activation of NK cells (Giedlund et al, 1987; Weigent et al., 1983; Tuo et al., 1993), and activation of macrophage phagocytosis (Baron et al., 1991). These cytokines induce resting CTL cells to an activated state and also directly induce NK cells to exhibit enhanced ^ cytotoxic function. Numerous macrophage functions including tumor cell cytotoxicity, antimicrobial activity, increase in killing of intracellular pathogens, and antigen processing and presentation are activated by IFNs (Degre and Bukholm, 1988; Black et al., 1988; Nathan et al., 1983; Neisel et al, 1986). They also affect the production of antibodies by B cells and can regulate the isotypes of the immunoglobulins secreted during the humoral immune response (Finkelman et al., 1988; Johnson and Torres, 1983; Snapper et al..

PAGE 17

8 Table III: Interferon inducible proteins^ Designation Characteristics Inducer (2'-5')(A)„synthetase (2'-5')(A)„synthesis a, p>Y p68 kinase protein phosphorylation oc, p > Y Indoleamine 2,3-dioxygenase tryptophan degradation Y> a, p P56 trp-tRNA synthetase Y> a, p GBF/y67 guanylate binding Y> a, p Mx families anti-influenza virus a, p > Y IRF1/ISGF2 transcription factor a, P, Y IRF2 transcription factor a,P MHC class I immune system a, p, Y MHC class n immune system y Pj-microglobulin immune system a, P, Y IPIO platelet factor 4 related Y> a, p 200 family cluster of 6 genes a,P 6-16 unknown a, P > Y 1-8/9-27 unknown a, p,Y C56, 561 unknown a, P>Y ISG54 unknown a, p>Y ISG15 unknown a, p>Y ^reviewed in Sen and Lengyel, 1992

PAGE 18

9 1988). In summary, IFNs play an important role in the network of immune interactions that bring about and regulate immunoreactivity and the local inflammatory response. Other properties of IFNs include regulation of cellular differentiation and antimicrobial and antiparasitic effects. As stated previously, there are two types of IFNs classestype I and type n. Although IFNs in both these classes have similar biological functions and overlap in their effects on different cell t)^es, there are some differences. Table 3 is a partial list of IFN-inducible proteins and which IFNsIFNa, IFNp, and /or IFNyare the inducers for each. In general, IFNy appears to have the dominant immunoregulatory role while IFNa and IFNp tend to be stronger inducers of antiviral proteins. One major difference is the ability of IFNy, and not the type I IFNs, to upregulate MHC class n (Houghton et al., 1984; Schwartz et al., 1985). Both type I and type n IFNs have potent antiproliferative effects. Interferons and Disease The various biological properties of both type I and type n IFNs make them potential therapies for a variety of medical disorders having viral, malignant, and immune etiologies. IFNs have been studied for their therapeutic efficacy in a number of conditions, and clinical investigations into a number of diseases are ongoing (Baron et al., 1991; Johnson et al, 1994; Stuart-Harris et al, 1992). As a result, IFNs have been approved by the Food and Drug Administration (FDA) for the treatment of several diseases (Baron

PAGE 19

10 Table IV: IFNs in disease therapy^ FDA approved currently in clinical trials IFNa chronic hepatitis BandC throat warts caused by papillomavirus hairy-cell leukemia kaposi's sarcoma genital warts caused by papillomavirus chronic myelogenous leukemia rolori tumors kidney tumors bladder cancer malignant melanoma IFNP relapsing / remitting multiple sclerosis basal ceU carcinoma IFNy chronic granulomatous disease kidney tumors leishmaniasis chronic lymphocytic leukemia Hodgkin's disease ^reviewed in Dorr, 1993; Johnsori et al, 1994

PAGE 20

11 et al., 1991; Dorr, 1993; Johnson et al., 1994). For example, due to potent antiviral properties, IFNa is approved for the treatment of chronic hepatitis type B and C (Dorr, 1993). The immunomodulatory functions of IFNp has resulted in its approval for the treatment of one type of the autoimmune disease multiple sclerosis (MS) (Johnson et al., 1994). Other diseases for which IFN therapy is currently approved by the FDA include hairy cell leukemia, Kaposi's sarcoma in the acquired immunodeficiency syndrome (AIDS), and chronic granulomatous disease. Table 4 lists the current FDA approved IFN therapies as well as others which are currently in clinical trials. The antiproliferative and immunoregulatory actions of IFNs combined with their ability to regulate proto-oncogenes and tumor supressor gene products make them particularly attractive candidates for the treatment of various cancers. The immunoregulatory roles of IFNs that result in antitumor effects include erUiancement of tumor cytotoxicity by macrophages, natural killer cells, and T lymphocytes (Baron et al., 1991). In addition, IFN enhanced expression of MHC antigens and tumor specific antigens result in more efficient recognition and killing of tumor cells by cytotoxic leukocytes (Baron et al., 1991). The induction of antibodies to the tumor cells may also be enhanced by IFNs (Baron et al., 1991). The antiproliferative capabilities of IFNs can directly inhibit replication of cells to decrease the growth rate of tumors and malignant cells (Fleischman and Fleischman, 1992). Furthermore, the ability of IFNs to modulate the expression of tumor supressor genes and proto-oncogenes has strong implications concerning

PAGE 21

12 their antiproliferative capabilities. The proto-oricogene c-myc has been found to be overexpressed in a large number of cancers (Steiner et al., 1996). IFNs capability to downregulate c-myc expression could contribute significantly to control tumor growth, especially in tumors where c-myc overexpression is involved in the malignancy. IFNs express potent antitumor effects both by exerting direct antiproliferative effects on target tumor cells, through the enhancement of immune responses, and by activating host cytotoxic effector cells to more efficiently lyse target tumor cells. One obstacle facing the widespread clinical use of IFNs is the toxic side effects experienced by some populations of patients (Vial and Descotes, 1994). Adverse effects associated with IFNs are usually acute effects that involve "flu-like" symptoms that include fever, malaise, tachycardia, chills, headache, arthralgias, and myalgias (Gauci, 1987; Quesada et al., 1986; Spiegal, 1987). However, these symptoms are usually not treatment limiting and are tolerable using symptomatic treatment (Vial and Descotes, 1994). The symptoms are also reversible after reduction in IFN dosage. A recently discovered type I IFN, IFNt, provides a non-toxic alternative. IFNx was originally recognized as a pregnancy recognition hormone in ruminants (Bazer and Johnson, 1991). It has now been shown to have similar biological activities to the other type I IFNs but without the associated toxicity (Pontzer et al., 1988; Pontzer et al, 1991, Soos and Johnson, 1995a; Soos et al., 1995b). IFNx has been shown to block experimental allergic encephalomyelitis (EAE)

PAGE 22

13 in mice and may provide an alternative therapy to IFNp for the treatment of MS (Soos et al., 1995b). Interferons and Signal Transduction The binding of IFNs to their specific receptor is the first step in evoking their biological responses. All type I IFNs bind to the same cell surface receptor while IFNy binds to a separate, although similar, receptor (Branca, 1988). The binding of type I IFNs to their receptor brings together two receptor chains, the IFNa/p receptor 1 chain and IFNa/p receptor 2.2 chain (Domanski et al., 1995; Novick et al., 1994; Lutfalla et al., 1996; Uze et al., 1990). These chains are induced to associate in the presence of ligand resulting in the formation of a functionally active receptor which mediates type I IFN signaling (Cohen et al., 1995; Darnell et al., 1994). Additionally, a third component of the type I receptor is believed to exist due to the differential effects of the type I IFNs (Croze et al., 1996). A receptor complex consisting of several subunits may explain how the different type I IFNs elicit the preferential induction of IFN specific genes while still binding the same receptor (Petska et al., 1987; Rani et al., 1996). The IFNy receptor consists of two integral membrane polypeptides that include a and p subunits (Aguet et al, 1988; Farrar and Schreiber, 1993; Soh et al., 1994). The a subunit is necessary for ligand binding while the p subunit participates in signal transduction (Farrar and Schreiber, 1993). IFNy binds its receptor as a homodimeric ligand which results in rapid dimerization of the receptor and

PAGE 23

14 subsequent signal transduction (Fountoulakis et al., 1992; Greenlund et al., 1993). As a result of ligand binding with both the type I and type II receptors, signal transduction is surprisingly rapid. Within fifteen minutes, transcription of chromosomal genes is enhanced without the need for new protein synthesis (Friedman et al., 1984; Larner et al., 1984; Larner et al., 1986; Levy and Darnell, 1990; Reich et al., 1987). This rapid transmission of signals combined with the apparant lack of involvement of second messengers has created an area of intense research in recent years concerning the signal transduction of the IFNs. The signal transduction pathway of the IFNs is direct in nature. Stimulation of the type I and type II IFN receptors initiate the activation of a class of tyrosine kinases known as Janus kinases or JAKs (Darnell et al., 1994). Type I IFNs activate two tyrosine kinase called tyrosine kinase 2 (tyk2) and JAKl (Silvennoinen et al., 1993; Velazquez et al., 1992). These activated kinases subsequently phosphorylate three proteins known as STATUS, STAT91, and STAT84 (Fu et al., 1992; Silvennoinen et al., 1993). The term "STAT" stands for signal transducer and activator of transcription while the number designates the molecular weight of the protein. The phosphorylation of the STATs causes them to associate with one another to form a complex which, combined with another protein referred to as p48, acts as a transcription complex to directly control gene transcription (Fu et al., 1990; Silvennoinen et al, 1993). The type I IFN STAT complex binds a target sequence motif known as the interferon-stimulated response element (ISRE)

PAGE 24

15 Figure 1. The IFN signal transduction pathways. IFNs bind their receptors and initiate intracellular signaling events that involve the JAK/STAT system. IFNy binds as a homodimer and causes the phosphorylation of JAKl and JAK2. JAKl then phosphorylates a latent cytoplasmic transcription factor STAT91. Phosphorylated STAT91 dimerizes and translocates to the nucleus where it binds the specific promoter elements known as GAS. This binding leads to the transcription of IFNy inducible genes. A similar mechanism is used by type I IFNs but utilizes JAKl and TYK2 which phosphorylate STAT84/91 and STATUS. These STATs combine with p48 and enter the nucleus where they activate transcription of genes containing ISRE sequences in their promoter regions.

PAGE 25

16 1^

PAGE 26

17 (Kessler et al., 1990). The IFNy activation pathway is similar in that it utilizes two JAKs, one of which is shared by the type I system. JAKl and JAK2 phosphorylate STAT91 which forms a homodimer and acts as a transcription factor which binds to specific promoter elements designated as gamma activation sites (GASs) (Decker et al., 1991; Greenlund et al., 1994; Igarashi et al, 1994; Lew et al., 1991; Shuai et al., 1994). JAK/STAT systems similar to those used by IFNs have been found to be utilized by a large array of cytokines (Leaman et al., 1996). An overview of the IFN signal transduction pathways is presented in Figure 1. The Mammalian Cell Cycle As previously mentioned, IFNs are believed to exert their antiproliferative effects by inhibiting the progression of the cell cycle. This may be due in part through the modulation of proteins involved in cell cycle regulation. The mechanism behind the anticellular effects of IFNs is poorly imderstood. The regulation of the cell cycle is particularly relevant in the study of IFNs in cancer therapy. The mammalian cell cycle consists of five phases: GO, Gl, S, G2, and M (Figure 2) (reviewed in Grana and Reddy, 1995). The GO represents the quiescent phase where cells are considered to "exit" the cycle and remain in a non-replicating state. Once the cell is stimulated by growth factors or mitogens to re-enter the cell cycle, it is in the Gl phase where it prepares for DNA replication. In the S phase, cellular DNA is replicated. The G2 phase

PAGE 27

18 Figure 2. The mammalian cell cycle. The cell cycle consists of five phases GO, Gl, S, G2, and M. The cdk-cyclin complexes are outlined and listed next to each cell cycle phase.

PAGE 28

19 0 0 0 C C3 C H tH >^ >^ >^ u u u I I I CS| ^ ^ ^ ^ ^ ^ u u u

PAGE 29

20 prepares the cell for mitosis, which is followed by the M phase where the cell physically divides. A multitude of cell proteins and biochemical pathways coordinate the cell cycle and the regulation of the system is complex. The following provides a condensed overview of the key proteins involved in the mammalian cell cycle. The progression of the cell cycle is directly dependent upon the activity of a set of protein kinases known as the cyclin dependent kinases (cdks). As the name suggests, cdks are kinases which require regulatory subunits called cyclins in order to be active (Morgan, 1995). Cyclins are named for their cyclic nature of expression. They are specifically expressed during each phase of the cell cycle and are promptly degraded as the cycle advances (reviewed in Koff et al., 1992; Sherr, 1994). Figure 2 outlines the cdk-cyclin complexes which are active during each cell cycle phase. Briefly, cdk4/cdk6 binds to cyclin D in Gl, cdk2 to cyclin E in Gl to S phase transition, cdk2 to cyclin A in S phase, and cdkl (cdc2) to cyclin A/B in M (Koff et al., 1992; Nurse, 1990; Sherr, 1993). These cdk-cyclin partners phosphorylate proteins which cause a cascade of events resulting in the coordination of cell cycle progression. Cell cycle control is carried out in large part by regulating cdk activity using a variety of mechanisms and feedback loops. One group of proteins that is directly involved in the regulation of cdk activity are the cyclin dependent kinase inhibitors (CKIs) (EUedge and Harper, 1994). These inhibitors are induced in response to specific extracellular signals and play an important role, via the inhibition of cdk activity, in

PAGE 30

21 Table V: Cyclin-dependent kinase inhibitors^ Inhibitors Cell cycle Cyclin-cdk phase complexes pl5 pl6 pl8 pl9 p21 p27 Gl Cdk4/6-cyclin D + + + + + +/Gl/S Cdk2-cyclinE + + S Cdk2-cyclinA — + G2/M Cdkl-cyclinB + ^reviewed in Grana and Reddy, 1995.

PAGE 31

22 blocking the cell cycle (EUedge and Harper, 1994; Pines, 1994). CKIs inhibit cdk function by physically binding cdk-cyclin complexes and interfering with kinase activity. Table 5 lists the known CKIs and the cdk-cyclin complexes they block. Currently CKIs are separated into two families. The first includes p2;i^wAFi p27*^^ which share partial identity and are involved primarily with blocking Gl and S phases of the cell cycle (Harper et al., 1993; Toyoshima and Hunter, 1994; Xiong et al., 1993). Both p21^^^ and p27'^i are present in quiescent cells and can cause Gl arrest. p21^'^''^ was the first CKI to be identified and is considered to be a universal CKI (El-Deiry et al., 1993; Gu et al., 1993). P27^^has been found to be induced in response to TGFp (Polyak et al., 1994). The other family of CKIs is the INK family which includes pl5, pl6, pl8 and pl9 (Guan et al., 1994; Hannon and Beach, 1994; Hirai et al., 1995; Serrano et al, 1993). These CKIs are important for blocking cdk4/cdk6 activity in early Gl phase and can also cause Gl arrest (Hirai et al., 1995; Serrano et al., 1993). Interestingly, pl6 has been designated a tumor suppressor gene since it has been found that mice with deletions in the pl6 gene have a dramatic increase in tumor formation (Serrano et al., 1996). Other CKIs are potential candidates for tumor suppressor genes due to their regulatory impact on the cell cycle. The prototypic tumor suppressor gene product p53 is a transcription factor that has multiple roles in cell cycle regulation, and for this reason is one of the most commonly mutated proteins found in malignant cells (reviewed in Levine et al., 1991). Many of the functions of p53 in cell

PAGE 32

23 Figure 3. The role of p53 in the Gl checkpoint. DNA damage induces p53 expression. p53 can then lead to the induction of p21^'^^\ apoptosis, or facilitate DNA repair. The induction of p21^^^ results in the inhibition of cdk-cyclin complexes including cdk2-cyclin E. This delay in the progression of the cell cycle allows time for DNA repair before DNA replication in the S phase.

PAGE 33

24

PAGE 34

25 replication have been determined, one of which is its role in cell cycle checkpoints (Kastan et al., 1992, Kuerbitz et al., 1992). Cell cycle checkpoints represent the coordination of the cell cycle machinery with the biochemical pathways that respond to DNA damage and restore its structure (Kaufman et al., 1995). Checkpoints at Gl and G2 phases delay the cell cycle and provide more time for repair before the critical phases of DNA replication and mitosis. DNA damage in cells induces p53 expression which causes Gl arrest (Kuerbitz et al., 1992). It has been foimd that this arrest is, at least in part, due to p53 induction of the CKI p21'^^' (El-Deiry et al, 1993; El-Deiry et al., 1994). p2]^wAFi blocks cdk2-cyclin E activity and stops the cell at the Gl checkpoint (Xiong et al., 1993). Another function of p21 is to bind the proliferating cell nuclear antigen (PCNA) and block DNA replication (Flores-Rozas et al., 1994; Waga et al, 1994). PCNA is an essential component of the DNA replication machinery (Kelman, 1997). p53 is also believed to play a role in the G2 checkpoint and mitosis (Guillouf et al., 1995). However, the mechanism behind this checkpoint is not fully understood. p53 has other roles in the cell, including involvement in apoptosis. An overview is provided in Figure 3. pRB and c-myc are two other proteins that are commonly found to be abnormally expressed in malignant cells (reviewed in Levine, 1993; Marcu et al, 1992). They have also been found to be modulated by IFNs. pRB is a tumor supressor that acts as the main target for Gl and S phase cdk-cyclin phosphorylation (Cobrinik et al., 1992). The Gl cdk-cyclin complexes phosphorylate pRB to a hyperphosphorylated state (Grana and Reddy, 1995).

PAGE 35

26 In this state, pRB is unable to bind the E2F transcription factor. E2F is then free to transcribe a number of genes that are involved in cell replication. Cmyc is an oncogene whose overexpression leads to uncontrolled cell growth (Marcu et al., 1992). It binds with its partner, max, to form a heterodimeric transcription factor (Amati et al., 1993). This dimer transcribes a number of genes that promote cell growth. By blocking c-myc expression and the phosphorylation of pRB, IFNs can regulate the transcription of proteins that promote cell replication. Adhesion Molecules and Growth Factor Receptors Adhesion molecules encompass another area of intense research due to their significance in tissue development, tumor development, and the immune response. These surface receptors play critical roles in cell-cell and cell-extracellular matrix (ECM) interactions. Pathologic alterations in these adhesion properties underlie many of the phenotypic changes associated with tumor progression, including changes in cell morphology, migration, tissue invasiveness, and metastatic potential (reviewed in Hannigan and Dedhar, 1997). There are several classes of cellular adhesion molecules (CAMs): integrins, cadherins, selectins, and the immunoglobulin superfamily. The integrins comprise a family of widely expressed transmembrane receptors that are expressed on all cell types and mediate cell-cell and cell-ECM interactions (Hynes, 1992). They are named according to the combination of different a and p subunits which form a heterodimeric receptor. This pairing of different

PAGE 36

27 subunits provides receptor diversity. The expression of certain integrin subunits is tissue or cell specific, such as with the integrin UlPj (LFA-1) whose expression is limited to leukocytes (Stewart et al., 1995). The members of the immunoglobulin superfamily contain typical immunoglobulin-like domains in the extracellular portion of the molecule and are expressed on a variety of cell types (Hannigan and Dedhar, 1997). They play an important role in the inflammatory immune response and modulate tumor spread by regulating the interaction of circulating tumor cells with host immunocytes. For example, the intercellular adhesion molecule-1 (ICAM-1) is a member that can bind the integrin receptor LFA-1 (Marlin and Springer, 1987). LFA-1 is found on leukocytes including NK cells and CTLs (Makgoba et al., 1988; Dana and Arnaout, 1994). ICAM-1 and LFA-1 binding can form stable cytolytic conjugates between tumor cells and cells of the immune system (Hannigan and Dedhar, 1997). The selectins and cadherins are involved with cell-cell interactions via carbohydrate moieties and homophilic binding, respectively (Hart, 1996). Growth factors can influence constitutive activation of growth promoting pathways in cancer cells and can modulate cell phenotype (reviewed in Aaronson, 1991). A large array of factors have been discovered that affect the growth of virtually all cell types which can act as positive or negative modulators of cell proliferation and influence differentiation. Many growth factors cause cells in the GO phase to re-enter the cell cycle (Pledger et al., 1977). For this reason, several oncogenes encode growth factors and

PAGE 37

28 tyrosine kinase receptors that participate in mitogenic signaling (Bishop, 1991). There is much evidence for genetic aberrations affecting growth factors and their receptors in human malignancies. Among growth factor receptors, the most frequently implicated in human cancer are the members of the EGF receptor (EGFR) family (Aaronson, 1991). The EGFR is a tyrosine kinase receptor that binds both EGF and TGFa ligands. The EGFR gene is often amplified or overexpressed in squamous cell carcinomas and glioblastomas (Libermann et al., 1985; Yamamoto et al., 1986), and EGFR expression has been linked to poor patient prognosis in other malignancies (Kristensen et al, 1996; Nakopoulou et al., 1995). Control of the overexpression of growth factor receptors and their ligands has several implications for cancer intervention. One is the potential improvement in diagnosis and prognosis of cancer. Possibilities for cancer therapy include effective means for targeting tumor cells by blocking signal transduction or ligand function. Experimental Rationale IFNs are a group of proteins that trigger a multitude of cellular responses including inhibition of cell growth, modulation of cell differentiation, and immunoregulation (Gastl and Huber, 1988). For this reason, many studies have looked at the potential antitumor effects of IFNs using both in vitro and in vivo models of cancer. These studies have resulted in the use of IFNs in the treatment of several cancers such as hairy cell leukemia, CML, and kaposi's sarcoma (Dorr, 1993; Gutterman, 1994).

PAGE 38

29 However, IFNs have not been successful in treating some other types of mahgnancies. The lack of patient improvement seen in some clinical trials may be attributed to inadequate knowledge of the underlying antitumor effects of IFNs. Thus, elucidation of the antitumor mechanisms of IFNs against a particular type of cancer cell are important for indicating which cancers may be susceptible to IFN therapy. The work included in this dissertation takes a fundamental look at the antitumor effects of type I and type n IFNs using an in vitro system involving a human prostate adenocarcinoma cell line, DU145. Prostate cancer is the most commonly diagnosed malignancy in men (Parker et al., 1996). Like other types of cancers, prostate cancer results from a loss or mutation of regulatory factors of the cell cycle such as oncogenes and tumor suppressor genes (Cavenee and White, 1995; Gamick, 1994). Previous studies have shown that human prostate cancer cell lines are sensitive to the antiproUferative properties of IFNs (Nakajima et al., 1994; Sica et al., 1989). Although these effects have been recognized, the mechanism behind this cellular inhibition remains unclear. DU145 is an interesting cell line to study the antitumor effects of IFNs for a number of reasons. This cell line was established from a metastatic lesion in a patient with advanced prostate cancer (Stone et al., 1978). In addition, DU145 cells have characteristics associated with undifferentiated malignant prostate cells. For example, these cells are androgen independent (Stone et al., 1978), tumorigenic in nude mice (Bookstein et al., 1990), and

PAGE 39

30 have mutations in several tumor suppressor gene products including p53, pRB, PTEN, and KAIl (Bookstein et al, 1990; Dong et al., 1995; Isaacs et al., 1991; Li et al., 1997). Examination of the antitumor effects of IFNs on DU145 cells describes a potential mechanism for their regulatory capabilities on an adenocarcinoma that has mutations in proteins closely linked to the cell cycle and cell adhesion. Studies into the mechanisms of the antitumor effects of IFNs give insight into the use of IFNs in cancer therapy. In addition, this information may indicate which cytotoxic agents and cytokines may produce synergistic combinations with IFNs and therefore provide successful clinical therapies.

PAGE 40

CHAPTER 2 MATERIALS AND METHODS Reagents and Cell Lines Purified human IFNa (specific activity 2 x 10* units /ml) was obtained from Biosource International (Camarillo, CA). Purified human IFNy (specific activity 4.75 x 10^ units/mg) was obtained from Genzyme Diagnostics (Cambridge, MA). WISH and DU145 cell lines were obtained from American Type Culture Collection (ATCC, Rockville, MD). Complete media for DU145 cells consisted of Eagles minimal essential medium (EMEM) supplemented with 5% fetal bovine serum (FBS), 200 U/ml penicillin, and 200 ^ig streptomycin. Starvation medium for cell synchronization contained the ingredients listed above supplemented with 0.5% FBS. Antibodies to cdk2, cyclin E, cyclin D, PCNA, p21, p27, and pl6 were obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Antibodies to ICAM-1 and EGF receptor were obtained from Pharmingen (San Diego, CA) for flow cytometric analysis and from Transduction Laboratories (Lexington, KY) for immunob lotting. Antibodies to integrin a3 were obtained from Oncogene Research Products (Cambridge, MA). 31

PAGE 41

32 Antiviral Assay IFN activity is expressed in terms of antiviral units/ ml as assessed in a standard cytopathic effect assay (Familletti et al., 1981). Antiviral activity of human IFNa was determined using the WISH cell line and vesicular stomatitis virus (VSV). One antiviral unit caused a 50% reduction in destruction of the monolayer. Antiproliferative Assays For colony inhibition studies, anticellular activity was examined using a modification of a colony inhibition assay (Blalock et al., 1980). DU145 cells were plated at 600 cells/well in a 24 well plate using complete medium with or without various concentrations of IFNa or IFNy. Plates were incubated at 37C for 5-6 days to allow for colony formation. Colonies were stained with crystal violet and counted. IFN inhibition of cell number was determined by using DU145 cells plated in complete medium at 1 x 10^ cells/well in 6 well plates with or without IFN. At various time points, cells were removed from flask using 0.25% trypsin-EDTA solution (Sigma Co., St. Louis, MO) washed 2 times with phosphate buffered saline (PBS), and counted. Cell counts were performed using a hemocytometer, and cell viability was assessed by trypan blue dye exclusion (Blalock et al., 1980).

PAGE 42

33 DNA Synthesis Assay DU145 cells were seeded at 2 x 10^ cells /well in 6 well plates using starvation medium for 24 hours. Wells were then washed and replaced with complete medium alone or medium containing 2500 units/ml IFNa. At 16, 20, and 24 hours, cells were harvested and counted as described above. Cells were then reseeded into 96 well plates at 2.5 x 10^ cells/well and pulsed with 1 \iC [^H]thymidine (specific activity, 21 mCi/mg; 1 Ci= 37 Gbq) (Amersham) for 2 hours at 37C. Cells were then harvested on a model M12 Brandel cell harvester (Gaithersburg, MD) and incorporation of [^H]thymidine was determined using a liquid scintillation counter. Cell Cycle Analysis DU145 cells were synchronized in GO/Gl phase (sychronization was assessed by flow cytometry analysis by determining the percentage of cells in the GO phase of the cell cycle) by culturing in starvation medium for 24 hours. Cells were seeded into 25 cm^ flasks (Sarstedt, Newton, NC) and grown in the presence of medium alone or medium containing either IFNa or IFNy. Plates were incubated at 37C in 5% CO^ for varying lengths of time. Following incubation, cells were harvested and washed 2 times using sample buffer (PBS containing 1% glucose) and were then counted. Following centrifugation, cell pellets were resuspended in 100 ^1 sample buffer and cells were fbced by adding 1 ml cold ethanol (95%) (-20 C) dropwise while vortexing. Cells were left in ethanol at 4C for > 24 hours. For propidium

PAGE 43

34 iodide (Sigma Co., St. Louis, MO) staining, the prepared cell samples were washed 2 times with sample buffer and blotted dry. The cells were then resuspended in 600 800 |il sample buffer containing 50 |ig/ml propidium iodide and 100 U/ml RNase A (Sigma Co.) and allowed to stain for up to 1 hour at room temperature. Samples were filtered through 44-|xm nylon mesh and data from 30,000 events was acquired with a FACSort (Becton Dickinson Immunocytometry Systems, San Jose, CA) using the LYSYS n software system. Analysis of the cell cycle was carried out using CellFIT software which determines the percentage of cells in each phase of the cell cycle based on cells collected using the LYSYS II software system. Immunoprecipitation and Immunoblotting For preparation of cell lysates, each experiment used an equal number of DU145 cells per sample in complete medium stimulated with IFNa or IFNy were lysed at 4C for 20 minutes in 500 )il of ice-cold lysis buffer that consisted of 50 mM Tris-HCl (pH 7.4), 150 mM NaCl, 2 mM EGTA, 2 mM EDTA, 50 mM NaF, 20 mM p-glyceryl phosphate, 2 mM NajVO^, 2 mM dithiothreitol, leupeptin (10 |ig/ml), pepstatin (10 Hg/ml), aprotinin (10 ^ig/ml), benzamidine (5 Hg/ml), 1 mM phenylmethanesulfonyl fluoride, 10% (vol/vol) glycerol, and 1% (vol/vol) Nonidet P-40. A range of 3-6 x 10* cells per sample were used with individual replicate experiments. Lysates were then centrifuged at 14,000 x g for 10 minutes. Equal amounts of protein (protein levels were determined using the BCA protein assay kit (Pierce,

PAGE 44

35 Rockford, IL)) from cell lysates were either directly loaded orito a polyacrylamide gel (a rar\ge of 100-175 |xg per sample was used with individual experiments) and electrophoresed or subsequently treated (a range of 375-500 |ig/500 |il was used with individual experiments) with 1-10 |ig of antibody to the protein of interest for 1 hour at 4C. The lysate-antibody complex was allowed to bind for an additional hour using 30-40 mg of protein A sepharose. Samples were then washed three times with 1 ml lysis buffer followed by a single wash with 1 ml of 50 mM Tris, pH 6.8. Complexes were eluted in SDS-PAGE sample buffer (0.03 M Tris-HCl, pH 6.8, 10% glycerol, 2%(w/v) SDS, 5% p-mercaptoetanol, and 2.5% bromophenol blue) and electroporesed through a Tris-HCl polyacrylamide gel with 4% stacking gel. Following Western transfer, nitrocellulose membranes were probed with antibodies to the proteins of interest and developed using the ECL chemiluminescence system (Amersham). Densitometric analysis of radiographic film using IA-200 Image Analysis Software was used to determine percent difference between band intensities based on total pixel value. ) \ ^ IK In Vitro Kinase Assays I > Treatment of cell lysates with antibodies to cdk2, cyclin Dl, and cyclin E were performed as described above. Antibody-protein conjugates were washed 3 times with lysis buffer and twice with kinase buffer (50 mM Hepes, pH 7.5, containing 1 mM EGTA, 10 mM MgCl^, 1 mM Na.YO,, 20 mM p-

PAGE 45

36 glyceryl phosphate, 5 |xM ATP) and incubated in kinase buffer containing 5 |j.g histone HI, and 20 |iCi of [^^P]--^ATP (specific activity 6000 Ci/mmol; 1 mCi= 37 Mbq) (Amersham) in a final volume of 30 ^1 at 30C for 10 minutes. Following centrifugation, supernatant fluids (25 )il) were analyzed for histone HI phosphorylation by a filter-binding assay using centrifugal Pierce phosphocellulose units, SpinZyme™ Format purchased from Pierce (Rockford, IL) according to the manufacturer's instructions. 'i 3 ,* Cellular Morpholog y DU145 cells in complete medium were treated, with or without IFN, for 5 days in culture slides (Falcon, Becton Dickinson, Bedford, MA). Slides were washed 3 times with phosphate buffered saline (PBS), fixed, and stained with eosin-methylene blue. Slides were then analyzed by light microscopy. Flow Cytometric Analysis of Surface Receptors DU145 cells were treated with IFN as described above. Cells were then harvested and washed 3 times with PBS. Equal number of cells per sample (810 X 10^ cells/sample) were washed 3 times with flow cytometry buffer (PBS containing 0.1% Na azide, 5% FBS) and incubated for 1 hour at 4C with antibodies specific for EGFR (2 |ig/ml), ICAM-1 (1 ng/ml), or integrin a3 (4 Hg/ml). After washing, cells were incubated for an additional hour with FITC labeled secondary antibodies specific for mouse IgGl. Data from 30,000 events

PAGE 46

37 were acquired as described above and arialyzed using Median Fluorescence Intensity (MFI) software. Analysis of Growth Factor Production DU145 cells were treated as described above for 5 days and reseeded at 1 X 10^ cells/well into 6 well plates in EMEM (no FBS). Supernatant fluids were harvested after 24 hours and used in a Cytokine Total ELISA kit (Intergen, Purchase, NY) specific for human EGF. Invasion Chamber Migration ^ DU145 cells were treated as described above for 5 days and reseeded, without IFN or FBS, at 1 x 10^ cells/well into Biocoat Matrigel invasion chambers (Falcon, Becton Dickinson, Bedford, MA). The chambers were then placed into 24 well plates containing EMEM supplemented with 20% FBS. These plates were then incubated 36-48 hours at 37C. Membranes from each chamber was then removed, subsequently fixed, and stained with eosinmethylene blue. Total number of invasive cells was then determined.

PAGE 47

CHAPTERS RESULTS IFNa Inhibition of the DU145 Cell Cycle Inhibition of Cell Growth t., < ^ ; The investigation into the effects of IFNa on prostate cancer cells was started by determining the antiproliferative properties of IFNa on DU145 cells, a human prostate cancer cell line. IFNa inhibited colony formation of DU145 cells at low cell density (600 cells /well) in a dose dependent manner as shown in Table VI. The antiproliferative effects of IFNa was also determined by utilizing direct cell counts. DU145 cells were treated with 2500 units/ml of IFNa and the overall reduction in cell number was determined (Figure 4). A reduction in the rate of growth by approximately 50% was observed in IFN treated cultures versus untreated cultures. These results indicate that IFNa has antiproliferative activity on DU145 cells. >\ Reduction in ^H-Thymidine Incorporation In examining the effects of IFNa on the DU145 cell cycle, the incorporation of [^H]thymidine by cultures synchronized, by serum starvation, into GO/Gl was first determined (Figure 5). The incorporation of thymidine by cells is a measure of chromosomal replication, and is therefore an indication of cellular activity in the S phase of the cell cycle (Tamm et al., 38

PAGE 48

• .* 39 Table VI: IFNa inhibition of colony formation of DU145 cells^ IFNa (units /ml) colonies/ well (meanSD) inhibition (%) 0 77.7 4.2 625 60.3 4.7 22.4 1250 40.3 3.5 48.1 2500 33.7 3.2 56.6 5000 30.6 2.1 60.6 "011145 cells were plated at 600 cells/ well with various doses (units/ml) of IFNa for 6 days and subsequently stained with crystal violet. Samples were assessed in triplicate and results are expressed as the mean number of colonies SD. Statistical significance was shown by Student's t-test between the number of colonies in the presence or absence of 625 U/ml (p<0.05), 1250 U/ml (p<0.02), 2500 U/ml (p<0.006) and 5000 U/ml (p<0.006) of IFNa.

PAGE 49

40 Figure 4. IFNa inhibition of DU145 cellular proliferation. Synchronized DU145 cells (1 x lOVwell) were incubated with or without 2500 units/ml IFNa for 48, 72, or 96 hours. Wells were then harvested and the total number of live cells determined. Data are expressed as total number of cells per sample SD for three replicates. Statistical significance was shown by Student's t-test between the number if cells in the presence and absence of 2500 U/ml IFNa for 48 (p<0.03), 72 (p<0.04) and 96 (p<0.03) hours.

PAGE 50

Time (Hours)

PAGE 51

42 Figure 5. Treatment of DU145 cells with IFNa inhibits ['Hjthymidine incorporation. DU145 cells (1 x 10^) synchronized in GO/Gl were incubated with or without IFNa for 16, 20, or 24 hours. Cells were harvested and reseeded into 96 well plates at 2.5 x 10* cells/well, incubated with pH]thymidine for 2 hours, and harvested on a cell harvester. Data are expressed as mean cpm SD for six replicates. Statistical significance was shown by Student's t-test between [^H]thymidine incorporation by cells in the presence and absence of 2500 U/ml IFNa for 16 (p<0.0006), 20 (p<0.0003) and 24 (p<0.0006) hours.

PAGE 52

Time (Hours)

PAGE 53

M 1987). DU145 cells were incubated with or without 2500 units/ml IFNa for 16, 20, or 24 hours arid pulse-labeled at each time poirit with [^H]thymidirie. At 16 hours, the amount of [^HJthymidine incorporated by IFN treated cells was only 29% of the incorporation seen with the untreated cells. At 20 and 24 hours, IFN treated cells incorporated about 60% of the [^H] thymidine incorporated by untreated cells. These reductions show that IFN treated cells entered the S phase at least 4 hours later than untreated cells. Consistent with this, at 20 and 24 hours, IFN treated cells reached a level of [^H]thymidine incorporation seen with untreated cells at 16 hours. These data suggest that IFNa inhibited the progression of DU145 cells from Gl through S phase. Flow Cytometric Analysis of the Cell Cycle The inhibitory effects of IFNa on the DU145 cell cycle were further examined using flow cytometry analysis (Table VII, Figure 6). DU145 cells synchronized by serum starvation into GO/Gl were stimulated to enter the cell cycle by serum addition in the presence or absence of 2500 units/ml IFNa. As can be seen from Table VII, untreated cells rapidly advanced through the GO/Gl and S phases, and, by 40 hours, cells had already completed one full cycle and were again entering the S phase. IFN treated cells, however, progressed more slowly. After 24 hours, 54% of the IFN treated cells were still in the GO/Gl phase, while untreated cells had only 39% of cells in GO/Gl. The flow cytometry histograms (Figure 6) depict the state of the DU145 cell

PAGE 54

45 Table VII: Cell cycle analysis of IFNa treated DU145 cells' cell cycle phase (%) Time IFNa (hours) (imits/ml) GO/Gl S G2/M 0 0 63.0 18.5 18.5 16 0 54.9 33.6 11.5 2500 66.9 22.1 11.0 20 0 39.5 46.2 14.2 2500 57.8 30.9 11.3 24 0 39.2 35.4 25.4 2500 53.8 33.7 12.5 40 0 62.1 24.9 13.0 2500 48.7 32.7 18.6 "011145 cells were treated with 0 or 2500 ur\its/ml for 0, 16, 20, 24, or 40 hours. Progression through the cell cycle was examined using propidium iodide staining. Data from 30,000 events are presented as percentage of cells in each stage of the cell cycle. Similar results were seen with two replicates of this experiment.

PAGE 55

46 Figure 6. IFNa inhibits the progression of DU145 cells through Gl and S phase of the cell cycle. DU145 cells were synchronized in GO/Gl by growing in medium containing 0.5% FBS. Progress through the cell cycle was examined using propidium iodide. Horizontal axis, relative fluorescence intensity; vertical axis, number of cells. A, B, and C IFNa treated cells (2500 units/ml); D, E, and F, untreated controls, at 16, 20, and 24 hours after the initiation of culture, respectively. Similar results were seen with two replicates of this experiment.

PAGE 56

47

PAGE 57

48 cycle at 16, 20, and 24 hours in the presence and absence of IFN. A similar pattern was seen with IFN treatment at 1000 units /ml (data not shown). Consistent with the [^H]thymidine incorporation by IFNa treated DU145 cells, flow cytometry analysis showed that IFNa inhibited progression of prostate cancer cells through Gl and early S phase of the cell cycle. Inhibition of Cdk2 Activity In order to determine the relationship of the phase of the cell cycle specifically inhibited by IFNa in the context of cyclin dependent kinase (cdk) activity, the activity of a cdk, cdk2, that is active during the Gl and S phases of the cell cycle was examined (Pines and Hunter, 1995). DU145 cells were cultured with or without IFNa for 16 or 24 hours and subsequently harvested. For each time point, cdk2 was immunoprecipitated and function assessed by histone Hl-dependent kinase activity (Table VIII). DU145 cells synchronized to GO/Gl had low cdk2 activity at 16 hours. However, by 24 hours, these cells increased their cdk2 activity by greater than lO-fold, while IFN treated cells showed only a 4-fold increase, resulting in a 74% reduction of cdk2 activity over the control. The data show that IFNa is able to reduce the activity of a cdk specific for the Gl and S phases of the cell cycle in prostate cancer cells, and thus inhibit the progression through the cell cycle. t i t':'

PAGE 58

49 Table VIII: Effect of IFNa treatment on cdk2 activity* Time IFNa CPM reduction (hours) (units/ml) (%) 16 0 8042 2500 6138 23.7 24 0 94880 2500 24570 74.1 "011145 cells were treated with 2500 units/ml IFNa for 16 or 24 hours. Cyclin dependent kinase 2 activity was assessed by histone Hl-dependent kinase activity. Cdk2 activity is represented as cpm with corresponding percent reduction. Similar data with the same patterns were observed in three repeats of this experiment.

PAGE 59

Analysis of Cyclin E and Cyclin D Dependent Cdk2 Activity To further examine the stage of the cell cycle that is regulated by IFNa, the cyclin specificity of the inhibition of cdk2 activity was determined. Cdk2 binds cyclin Dl and cyclin E during the Gl phase and Gl to S phase transition, respectively (reviewed in Pines and Hunter, 1995). Cell lysates from DU145 cells treated with IFNa for 16 or 24 hours were immunoprecipitated using antibodies specific for cyclin Dl or cyclin E and cdk2 activity was subsequently assessed (Table IX). IFNa treated cells showed up to a 38% reduction of cyclin E-cdk2 activity over the control, but did not show consistent inhibition of cyclin Dl-cdk2 activity. Of the 74% IFNa induced reduction in overall cdk2 activity (Table VIII), 38% is apparently due to the reduction of cyclin E dependent cdk2 activity (Table IX). The remaining inhibition of cdk2 activity by IFNa is probably due, at least in part, to inhibition of activity in the cyclin A-cdk2 complex (Tiefenbrun et al., 1996). I next immunoprecipitated cyclin E and immunoblotted using a cdk2 antibody in order to determine relative amounts of cdk2 complexed to cyclin E. Figure 7 shows that IFNa treatment of cells did not affect the levels of cdk2 complexed with cyclin E. Consequently, IFNa inhibition of the activity of the cyclin E-cdk2 complex did not affect the formation of the cyclin E-cdk2 complex in DU145 cells. Induction of CKT p 21 The IFNa induced decrease in cyclin E-cdk2 activity suggested that a kinase inhibitor may be involved. The CKI p21 is known to bind cyclin E-

PAGE 60

51 Table IX: Effect of IFNa treatment on cyclin specific cdk2 activity^ Cyclin Time (hours) IFNa (units /ml) CPM reduction (%) Expt. 1 E 16 0 2500 29507 ^ 22861 22.5 24 0 2500 28607 17742 37.9 Dl 16 0 2500 5321 3881 27.1 24 0 2500 4642 5291 Expt. 2 E 16 0 2500 15954 11761 26.3 24 0 2500 14296 9907 30.1 Dl 16 0 2500 4786 4936 24 0 2500 5666 7818 ^DU145 cells were treated with 0 or 2500 units/ml IFNa for 16 or 24 hours. The cyclin-cdk2 complex was immunoprecipitated with antibodies specific for either cyclin E or cyclin Dl. Cdk2 activity was then assessed by histone Hldependent kinase activity. Cdk2 activity is represented as cpm with corresponding percent reduction. Similar data with the same patterns were observed in two repeats of this experiment.

PAGE 61

52 ; > ;• i ^: M : ^ • — S-.v i^v Figure 7. IFNa does not affect cyclin E-cdk2 complex formation in DU145 cells. The presence of cdk2 complexed with cyclin E was assessed by immunoprecipitation of DU145 cell lysates (515 ^ig protein/ 500 ^1 for each sample) with cyclin E antibodies and immunob lotting with antibodies specific for cdk2, as described in 'Materials and Methods'. Lanes 1 and 2 represent lysates from cells treated for 16 hours with 2500 units/ml IFNa or medium alone, respectively. Lanes 3 and 4 are from cells treated for 24 hours with or without IFN, respectively.

PAGE 62

53 12 3 4

PAGE 63

54 cdk2 and block its activity (reviewed in Pines and Hunter, 1995). Serum starvation of cells increases p21 levels and restimulation by serum a reduction in p21 levels (Pines and Hunter, 1995). DU145 cells synchronized by serum starvation into GO/Gl were therefore used to determine whether p21 is involved in inhibiting the Gl to S phase transition in IFN treated cells. Cell lysates from DU145 cells treated with IFNa for 16 and 24 hours were immunoprecipitated using p21 antibodies, and the presence of both p21 and cdk2 in these samples was subsequently assessed by immunoblotting. As shown in Figure 8A, lysates from cells treated with IFNa appeared to have a greater expression of p21 than untreated cells. In the same immunoprecipitates, at 16 and 24 hours, cdk2 protein levels were also higher in IFNa treated cells than in untreated cells (Figure 8B), supporting the conclusion that in IFNa treated cells increased levels of p21 were associated with cdk2. This suggests that the expression of p21 in IFNa treated cells played an important role in ir\hibiting the cdk2 activity in these cells. To establish that IFNa treatment induces the expression of p21, cell lysates from asynchronous DU145 cells that were treated with IFNa or medium alone were assessed for p21 as described above. IFNa treatment progressively induced the expression p21 (Figure 9A) over that seen in untreated cells. Densitometric analysis (Figure 9B) of the p21 bands in Figure 9A showed that IFN treated cells had approximately twice the levels of p21 compared to untreated cells. Thus, IFNa inhibits the Gl to S phase transition of the cell cycle by inducing p21 expression in a prostate cancer cells. Similar

PAGE 64

55 Figure 8. A. IFNa treatment increases and /or maintains p21 levels in synchronized DU145 cells. Cell lysates (512 protein/ 500 |J,1 for each sample) were immunoprecipitated and immunoblotted using p21 antibodies, as described in 'Materials and Methods'. Lanes 1 and 2 represent lysates from synchronized cells treated with 2500 U/ml IFNa or media alone, respectively, for 16 hours. Lanes 3 and 4 are from cells treated for 24 hours with or without IFN, respectively. The percent decrease in p21 levels for lanes 1, 2, 3, and 4 was 40.1%, 62.7%, 68.2%, and 70%, respectively, as determiend using densitometric scanning of radiographic film. The percent decrease represents the ratios of band intensities from DU145 cell lysates at initiation of cultures (0 hours, data not shown) and lanes 1 through 4. B. Cdk2 levels correspond to p21 expression in IFNa treated cells. The immunoblot from Figure 5A was reanalyzed using antibodies specific for cdk2. The lane assignments are as stated for 5A. The percent decrease in cdk2 protein levels was 64.3%, 81.5%, 83.2%, and 91% for lanes 1, 2, 3, and 4 respectively, as determined by densitometric scanning as described for A.

PAGE 65

56 s12 3 4 IFNa + + 32.7— 17.7— .p21 1 ^ 3 4 g_ IFN
PAGE 66

57 Figure 9. A. IFNa induces p21 expression in DU145 cells. Cell lysates from DU145 cells (486 |ig protein/ 500 |il for each sample) were immunopreciptated and immunoblotted with antibodies specific for p21. Lanes 1 and 2 represent lysates from cells treated for 16 hours with 2500 U/ml IFNa or media alone, respectively. Lanes 3 and 4 represent lysates from cells treated for 24 hours with or without IFN, respectively. B. The fold increase in p21 levels for 16 and 24 hours in IFN treated and untreated cells was determined by densitometric scanning of radiographic film. The fold increase represents the ratios of band intensities from DU145 cell lysates at initiation of cultures (0 hours, data not shown) and Lanes 1 through 4. Standard deviations represent three separate densitometric readings from figure 9A.

PAGE 67

58 12 3 4 IFNa: + — + — 44.132.717.7A. .p21

PAGE 68

59 experiments looked at the possible roles of two other CKIs, p27 ar\d pl6. Induction of p27 or pl6 expression was not seen with IFN treatment of DU145 cells (data not shown) suggesting a unique role for p21 in these effects. IFNa inhibits the Gl to S phase transition of the cell cycle by inducing p21 expression in a prostate cancer cell line. IFNv Inhibition of the DU145 Cell Cycle Inhibition of the Cell Cycle Studies using IFNy were begun by determining the antiproliferative effects of IFNy on DU145 cells. IFNy was able to inhibit colony formation of DU145 cells at low cell density in a dose dependent manner as shown in Table X. Treatment of DU145 cells with IFNy showed significant inhibition of colony formation at concentrations as low as 312 U/ml with up to 70% inhibition at 10,000 U/ml. The effects of IFNy on the DU145 cell cycle were then analyzed using flow cytometry (Table XI). DU145 cells synchronized by serum starvation into GO/Gl were stimulated to enter the cell cycle by serum addition in the presence or absence of 2500 U/ml IFNy. Untreated cells rapidly advanced through the GO/Gl and S phases of the cell cycle, and, by 24 hours, these cells began returning to the Gl. IFNy treated cells, however, progressed more slowly. At 16 hours, 46% of the IFN treated cells remained in Gl compared to only 33% of the untreated cells. By 24 hours, in contrast to the untreated cells, the largest percentage of IFN treated cells were in the S phase. A similar pattern was seen with 1000 U/ml IFNy (data not shown).

PAGE 69

60 Table X: Inhibition of DU145 colony formation by IFNy^ IFNy (units /ml) colonies/well (meanS.D.) Inhibition (%) 0 110.5 8.7 312 :, 74.3 3.8 32.6 625 62.7 3.1 43.8 1250 57.3 3.2 48.1 2500 51.3 4.0 53.6 5000 41.7 1.5 62.3 10,000 33.7 3.1 70.6 'DU145 cells were plated at 800 cells/well with various doses (units/ml) of IFNy for 5 days and subsequently stained with crystal violet. Samples were assessed in triplicate and results are expressed as the mean number of colonies S.D. Statistical significance was shown by Student's t-test between the number of colonies in the presence or absence of 312 U/ml (p<0.05), 625 U/ml (p<0.02), 1250 U/ml (p<0.01), 2500 U/ml (p<0.01), 5000 U/ml (p<0.005), and 10,000 U/ml (p<0.005) of IFNy.

PAGE 70

Table XI: Effects of IFNy on the DU145 cell cycle" cell cycle phase (%) Time IFNy (hours) (units/ml) GO/Gl S G2/M 0 16 20 24 0 2500 0 2500 0 2500 74.1 15.3 10.6 33.3 45.6 60.5 44.3 6.2 10.0 26.9 32.0 53.6 55.9 19.5 12.2 36.8 34.1 33.6 40.7 29.6 25.2 ''DU145 cells were treated with 0 or 2500 units/ml of IFNy for 0, 16, 20, or 24 hours. Progression through the cell cycle was examined using propidium iodide staining. Data from 30,000 events are presented as percentage of cells in each stage of the cell cycle. Similar results were seen with two replicates of this experiment.

PAGE 71

62 IFNy inhibits the proliferation of DU145 cells by slowing their progression through the cell cycle. Induction of p21 ^^^ As previously mentioned, p21 is known to act at the Gl and S phases of the mammalian cell cycle by directly binding to cdk-cyclin complexes that are active at these stages and blocking their activity (Cartel et al., 1997). To determine whether IFNy treatment induces p21 in DU145 cells, cell lysates from asynchronous cells that were treated with IFNy or media alone were assessed for p21 by immunoblotting (Figure 10). At the initiation of treatment, rapidly dividing cells exhibited low levels of p21. After 20 and 40 hours of IFNy treatment, p21 levels appeared to be greatly increased over that of the untreated cells. As seen with IFNa, this induction of p21 by IFNy was unique since other cdk inhibitors such as p27 and pl6 did not show similar increases in response to IFN treatment (data not shown). p2iWAFi it^duction Causes an Increase in p21 Bound Cdk2 and PCNA I further assessed the consequences of the induction of p21 by IFNy using immunoprecipitation. Using antibodies specific for p21, cell lysates from DU145 cells treated with medium alone or IFNy were immunoprecipitated. These precipitates were then immimob lotted using antibodies specific for p21, cdk2, and PCNA. Consistent with previous studies on the role of p21 in the cell cycle, cdk2 and PCNA levels correlated with the

PAGE 72

63 Figure 10. IFNy induces p21 expression in DU145 cells. DU145 cell lysates (141 |xg protein per lane) were immunoblotted with antibodies specific for p21. Lane 1 represents cell lysate from dividing cells at the initiation of treatment (time 0), lanes 2 and 3 are lysates from cells treated for 20 hours with 2500 units /ml IFNy or media alone, respectively, and lanes 4 and 5 are lysates from cells treated 40 hours with or without IFNy, respectively. Densitometric analysis of radiographic film showed a 2.1-fold and 2.6-fold difference in p21 levels between lanes 2 and 3 and lanes 4 and 5, respectively.

PAGE 73

64

PAGE 74

65 Figure 11. Cdk2 and PCNA levels correspond to p21 induction by IFNy. DU145 cell lysates (463 |xg protein/ 500 |il for each sample) were immunoprecipitated using antibodies specific for p21 and subsequently immunoblotted using antibodies for (a) PCNA, (b) cdk2, and (c) p21. For (a), (b), and (c), lane 1 represents lysates from untreated cells at initiation of IFN treatment while lanes 2 and 3 are from cells treated with 2500 units/ml IFNy or media alone for 30 hours, respectively. Densitometric analysis of radiographic film showed a 6.3-fold difference in PCNA levels (a), a 2.22-fold difference in cdk2 levels (b), and 2.7-fold difference in p21 levels (c) between lanes 2 and 3.

PAGE 76

67 level of p21 expression (Figure 11). The increase in p21 resulting from IFNy treatment corresponded to an increase in p21 bound cdk2 and FCNA. Cdk2 is a cyclin dependent kinase which is active in the Gl and S phases of the cell cycle, while PCNA is an essential component of the DNA replication machinery (Cartel et al., 1997; Kelman, 1997). By binding these proteins, p21 acts to inhibit cell replication, and the induction of p21 by IFNy suggests that these aspects of cellular replication are inhibited. Based on the previous studies with IFNa, both type I and type II IFNs appear to block the DU145 cell cycle at the Gl and S phases via a similar mechanism of p21 induction. IFNy Induction of a Change in Cell Phenotype Changes in Cell Morpholog y In the course of studies on the effects of IFNs on the cell cycle, it was observed that IFNy treatment induced a change in the appearance of DU145 cells. I therefore looked at the impact of IFNa and IFNy on DU145 cellular morphology. DU145 cells were treated with medium alone, IFNa, or IFNy for 5 days, subsequently fixed, and stained with eosin-methylene blue. Analysis by light microscopy showed that while untreated and IFNa treated cells retained normal tissue culture appearance, IFNy treated cells showed a distinct morphological change (Figure 12). Untreated and IFNa treated cells exhibited a rounded morphology typical of tumor cells growing in culture. IFNy treated cells were less rounded with protuberances resulting in more of a spindle shape. Changes in cellular morphology are often associated with a

PAGE 77

68 Figure 12. IFNy induces morphological changes in DU145 cells. DU145 cells were treated with (a) medium alone, (b) 5000 units/ml IFNa, or (c) 5000 units /ml IFNy for 96 hours, subsequently stained with eosin-methylene blue, and analyzed by light microscopy. Total magnification is lOOX.

PAGE 79

70 change in cellular phenotype and differentiation. The DU145 cell line displays many features common to undifferentiated metastatic tumor cells, including loss of several functional tumor suppressor gene products as well as being tumorigenic in nude mice (Bookstein et al., 1990; Dong et al., 1995; Isaacs et al., 1991). The effects of IFNy on the morphology of DU145 cells suggests that IFNy induces a phenotypic change and possibly differentiation of these cells. Downregulation of the EGF Receptor Expression of the EGFR and its ligands has been associated with human cancer cells of various origins and, in several instances, has been correlated with the stage of differentiation of tumor cells (Cohen et al., 1994; loachim et al, 1996; Nakopoulou et al, 1995; Stumm et al, 1996). Further, in some malignancies, increased EGFR expression has been found to be an indicator of poor patient prognosis (Almadori et al., 1995; Kristensen et al., 1996). The DU145 cell line has previously been shown to express EGFR, as well as its ligands EGF and TGFa, and, as a result, possesses an autocrine feedback loop (Connolly and Rose, 1991). I looked at the effects of type I and iype U IFNs on the expression of the EGFR. DU145 cells were treated with medium alone, IFNa, or IFNy for 5 days and the expression of the EGFR was analyzed by flow cytometry using antibodies specific for the EGFR (Figure 13A). IFNy caused greater than a 50% reduction in EGFR compared to untreated cells. No effect on EGFR expression was seen with IFNa treatment. These effects were also

PAGE 80

71 Figure 13. IFNy downregulates the expression of the EGF receptor. DU145 cells were treated with medium alone, 5000 units/ml IFNa, or 5000 units/ml IFNy for 5 days, (a) Cells were stained using antibodies specific for the EGFR and analyzed by flow cytometry. Similar data with the same pattern were observed in 3 repeats of this experiment. Data are expressed as mean fluorescence intensity, (b) DU145 cell lysates (138 |J,g protein per lane) were immunoblotted using antibodies specific for EGFR. Lane 1 represents lysate from cells treated with IFNa, lane 2 is from cells treated with IFNy, and lane 3 is from untreated cells. Densitometric analysis of radiographic film showed a 80% decrease in EGFR levels between IFNy and untreated cells (lanes 2 and 3).

PAGE 81

72

PAGE 82

73

PAGE 83

analyzed by western blot (Figure 13B). Again, while no change was seen with IFNa, IFNy clearly downregulated the expression of EGFR. I then looked at the expression of the EGFR ligand EOF (Figure 14). Both IFNa and IFNy reduced EGF production with IFNa showing the more significant downregulation. These results are interesting in that both IFNa and IFNy reduced growth factor production and cellular replication, but only IFNy affected the receptor expression. To determine the significance of the decrease in receptor expression, the impact of EGF on cells pretreated with IFNa and IFNy was examined. EGF is known to stimulate quiescent cells to enter the cell cycle and increase cyclin Dl expression (reviewed in Lavoie et al., 1996). Cyclin Dl is a protooncogene that has been found to be overexpressed in a number of human neoplasms (Lees and Harlow, 1995). Lysates from IFN treated or untreated cells stimulated with EGF were assessed for cyclin Dl expression. Figure 15 shows that IFNy treated cells exhibited little expression of cyclin Dl even with the addition of EGF to the cell culture medium. In contrast, IFNa treated cells expressed higher levels of cyclin Dl and showed a modest increase in the presence of EGF. A similar pattern was seen with untreated cells. Further, IFNa treated cells in the presence of EGF increased in number by thirty percent over those without EGF while no difference in cell number was seen with IFNy and EGF (data not shown). These findings indicate that the downregulation of EGFR expression on DU145 cells by IFNy results in a cell type that is less sensitive to the growth enhancing effects of EGF. Previous

PAGE 84

75 Figure 14. IFNa and IFNy reduce EGF production by DU145 cells. DU145 cells were treated with medium alone, 5000 units/ml IFNa, or 5000 units/ml IFNy for 5 days. Cells were then harvested and reseeded into 6 well plates at 1 x 10^ cells/well. Supernatants were collected at 24 hours and analyzed by ELISA for EGF. Data are expressed as mean concentration S.D. Statistical significance was shown by Student's t-test between m.f.i. for cells treated with media alone and IFNa (p<0.02).

PAGE 85

76 IFN alpha IFN gamma untreated

PAGE 86

77 Figure 15. EGF does not induce cyclin Dl in IFNy treated cells. DU145 cells were treated for 5 days with either 5000 U/ml IFNa, 5000 U/ml IFNy or medium alone followed by the addition of 0.5 ng/ml EGF to fresh culture media and incubated for 6 hours. Cell lysates (155 |ig protein per lane) were then immunoblotted for the presence of cyclin Dl. Densitometric analysis of radiographic film showed a 31% increase betv/een lanes 1 and 1, a 6% increase between lanes 3 and 4, and a 23% increase between lanes 5 and 6 of cyclin Dl levels. In addition, there is a 61% increase in cyclin Dl between lanes 3 (IFNy) and 5 (untreated).

PAGE 87

78

PAGE 88

79 studies have shown that in vitro invasiveness of DU145 cells and in vivo progression of DU145 tumors in nude mice are modulated by EGFR mediated signals (Prewett et al., 1996; Xie et al., 1995). The downregulation of EGFR by IFNy in DU145 cells suggests that IFNy induces a less malignant phenotype and decreases the invasive potential of this cell line. Modulation of Adhesion Molecules In addition to soluble factors, cell-cell and cell-extracellular matrix (ECM) adhesion properties underlie many of the phenotypic changes associated with tumor progression and metastatic potential (Harmigan and Dedhar, 1997). The effects of IFNy on the expression of the adhesion molecules ICAM-1 and integrin a3 were analyzed (Table XII). The abihty to evade the host immune response is characteristic of tumor cells. ICAM-1 plays a role in leukocyte adhesion by binding the LFA-1 surface receptor fotmd on cytotoxic T lymphocytes (CTL), B cells, and natural killer (NK) cells (Dana and Arnaout, 1994; Makgoba et al., 1988). As a result, cells with high ICAM-1 expression could be the target of a host immune response in vivo. ICAM-1 has previously been found to be an immediate response gene induced by IFNy (Caldenhoven et al, 1994). IFNy was found to induce ICAM1 expression in DU145 cells by 239% over that of untreated cells. An increase in integrin a3 expression was also seen as a result of IFNy treatment (Table Xn). Integrins mediate interactions between cells and ECM, and previous work with another prostate carcinoma cell line, PC-3, has shown that highly

PAGE 89

80 Table XII: Effects of IFNa and IFNy on the expression of ICAM-1 and integrin a3^ ICAM-1 Integrin a3 m.f.i.+S.D. increase m.f.i.S.D. increase (%) (%) untreated 153.2 + 4.2 — 368.2 9.9 IFNa 127.2 2.8 — 387.7 1.0 5.3% IFNy 519.7 26 239% 457.2 12.7 24.2% "011145 cells were treated with either 5000 units/ml IFNa, 5000 units/ml IFNy/ or media alone for 5 days. Cells were stained with either antibodies specific for ICAM-1 or integrin a3 and analyzed by flow cytometry. Data are expressed as mean fluorescence intensity (m.f.i.) S.D for three replicates. Statistical significance was shown by Student's f-test between m.f.i. for cells treated with IFNy or media alone for ICAM-1 (p<0.03) and integrin a3 (p<0.01).

PAGE 90

81 invasive PC-3 variants have low levels of integrin a3 expression (Dedhar et alv 1993). Thus, upregulation of the ICAM-1 and integrin a3 adhesion molecules suggests that IFNy induces a less tumorigenic and less metastatic phenotype in DU145 cells. f Reduction in Invasive Potential I next tested whether the modulation of surface receptors on DU145 cells actually correlated with a change in invasive potential. Invasion by tumor cells is a crucial step in the multistage process of tumor spread and the formation of metastasis (Liotta, 1987). Several in vitro systems have been established to study the invasiveness of tumor cells. One commonly used system utilizes insert chambers with separating filters coated with a layer basement membrane matrix that contains components similar to the ECM. These invasion chambers are suitable for studying cell invasion of malignant cells (Albini et al., 1987). DU145 cells were treated for 5 days with or without IFN. These cells were subsequently reseeded into invasion chambers which contain membranes coated with a basement membrane matrix. The total number of cells which migrated across the membrane were then determined for each treatment group (Figure 15). IFNy treated cells showed a 50% reduction in the number of invasive cells over that of the control. Furthermore, only a slight reduction in invasion of IFNa treated cells was seen. Thus, the reduction in the number of invasive cells for the DU145 cell line was specific for IFNy compared to IFNa even though both IFNs inhibited

PAGE 91

82 h* r'U' ^r; Figure 16. IFNy decreases the invasive potential of DU145 cells. DU145 cells were treated with medium alone, 5000 units/ml IFNa, or 5000 units/ml IFNy for 5 days. Cells were then reseeded into invasion chamber inserts containing matrigel basement membrane matrix. Cells that migrated across the matrix were stained with eosin-methylene blue and counted. Data are expressed as the mean number of cells S.D for three replicates.

PAGE 92

83 150 IFN alpha IFN gamma untreated

PAGE 93

84 cell growth. These results imply that IFNy can both reduce cell growth and metastasis of DU145 prostate cancer cells.

PAGE 94

CHAPTER 4 DISCUSSION Both type I and type II IFNs possess potent antitumor properties for a variety of cell types by inducing a number of cellular responses. Previous studies have looked at the antiproliferative effects of IFNs on cancer cells both in vivo and in vitro, and have shown that IFNs directly act on tumor cells to prolong their progression through the cell cycle (Fleischman and Fleischman, 1992). Indirect antitumor functions of IFNs include the activation of different aspects of the immune system, which leads to the targeting of cancer cells by cytolytic immune cells (Baron et al., 1991). The combination of these antitumor properties has led to the use of IFNs in a number of human malignancies. However, while IFNs are therapeutic with certain cancers, they are not with some others. Many of the cases where IFNs have not been found to be clinically useful may be a result of inadequate knowledge of the mechanisms by which IFNs function. The mechanisms behind the antiproliferative effects of IFNs are poorly understood. Previous work has shown with some cell types IFNs inhibit the Gl and S phases of the cell cycle (Creasey et al., 1980; Pontzer et al., 1991; Roos et al., 1984; Tamm et al., 1987) and modulate the expression of c-myc and pRB (Einat et al., 1985a; Einat et al., 1985b; Jonak and Knight, 1984; Melamed et al., 1993). The vast majority of these studies have been carried out on the Daudi cell line, which is a B cell 85

PAGE 95

86 lymphoma. Daudi cells are extremely sensitive to the antiproliferative effects of IFNa and undergo Gl arrest upon treatment with IFN (reviewed in Subramaniam et al., 1997). Few studies have used cells of other origins to study the effects of IFNs on the cell cycle. The DU145 cell line, as previously mentioned, is an interesting model to study the mechanism of the regulation of the cell cycle by IFNs. These cells were isolated from a metastatic lesion and are a poorly differentiated adenocarcinoma (Stone et al., 1978). Most malignancies, including prostate cancer, with high metastatic potential and poor differentiation of cells are associated with poor patient prognosis (Garnick, 1994). Tumors of the prostate that are well-differentiated are more likely to be confined to the prostate gland and can be treated by removal of the prostate (Garnick, 1994). A poorly differentiated prostatic adenocarcinoma is more likely to extend beyond the prostate and metastasize. Currently, besides radiation, hormonal therapy is the only common treatment of prostate cancer once metastasis has occurred (Garnick, 1994). The drawback to this therapy is that, at some point, most metastatic tumors become resistant to hormonal therapy and, as a result, progress rapidly. By studying the DU145 cell line, we can gain insight into the mechanism of IFNs on a metastatic adenocarcinoma. DU145 cells are sensitive to the antiproliferative effects of IFNa and IFNy, and, consistent with previous studies on other cell lines, appear to be affected at the Gl to S phase transition of the cell cycle. This inhibition of cell replication is at least partly due to an increase in p21 expression and the

PAGE 96

87 subsequent decrease in activity of the cyclin E-cdk2 complex. The interaction of p21 with cyclin E-cdk2 blocks the phosphorylating activity of this complex which is necessary for the Gl to S phase transition of the cell cycle and DNA replication (Cartel et al., 1997; Pines and Hunter, 1995; Sherr, 1993). p21 also inhibits DNA rephcation by binding PCNA (Kelman, 1997). By inducing p21 expression, IFNs are able to control different aspects of the cell cycle. These results describe a possible mechanism for the antiproliferative affects of IFNs on a prostate cancer cell line. Many of the proteins involved in controlling the cell cycle have been designated as products of tumor suppressor genes or oncogenes due to the association of mutations of these genes with different types of cancers. DU145 cells have mutations in the tumor suppressor gene products p53, pRB, PTEN, and KAIl (Bookstein et al., 1990; Dong et al., 1995; Isaacs et al., 1991; Li et al, 1997). Various oncogenes and tumor suppressor genes have been identified in prostatic tumors including mutations in p53 and pRB (Netto and Humphrey, 1994). This provides another characteristic of DU145 cells that makes them interesting to study. Approximately fifty percent of all cancers have mutations in the p53 genes, and pRB is frequently deregulated in a variety of tumors (reviewed in Gangopadhyay et al., 1997; La Thangue, 1997). The mutant p53 in DU145 cells means that, although p53 is known to induce p21, the induction of p21 by IFNs in this case is independent of p53 status. The pRB gene product is directly phosphorylated by cyclin E-cdk2 complexes that are inhibited by p21 (Hatakeyama and Weinberg, 1995). Again, in the case

PAGE 97

88 of DU145 cells, this aspect of p21 regulation of the cell cycle does not contribute significantly to the inhibition of cell replication, because of the pRB mutation. Other features of p21 regulation, including the binding of PCNA and other potential cdk2 targets, are therefore most likely involved. Although both IFNa and IFNy can induce p21, IFNy induced biochemical changes associated with changes in cellular phenotype indicate that it possesses other antitumor effects against DU145 cells not related to p21. One significant effect that appears to be independent of p21 status is the downregulation of the EGFR. The EGFR is the most studied of the tyrosine kinase receptors and has long been associated with human tumorigenesis (Helden and Ronnstrand, 1997). It is frequently found to be overexpressed or mutated in several different human tumor types. Like other cells of epidermal origin, prostate cells are stimulated by EGF to replicate. The DU145 cell line in particular has been shown to have an autocrine stimulatory loop with the EGFR and its ligands (Connolly and Rose, 1991). Therefore, IFNy regulation of the EGFR lends insight into the mechanism behind IFNy control of cell replication and phenotype. Other surface receptors modulated by IFNy, and not IFNa, are ICAM-1 and integrin a3. Integrin a3 has been correlated with the metastatic potential of another prostatic adenocarcinoma cell line PC-3. The direct involvement of the EGFR and integrin a3 in the invasiveness of these cells remains to be elucidated. However, they could act as important markers of a metastatic phenotype. The strong upregulation of the ICAM-1 adhesion molecule by IFNy is another aspect of its antitumor

PAGE 98

89 effects. An increase in ICAM-1 could allow these cells to be more easily targeted by the immune system. ICAM-1 is the receptor for the integrin LFA1 which is found on the surface of CTLs, NK cells, and B cells (Dana and Amaout, 1994; Makgoba et al., 1988). The significance of this increase in cellcell interactions may be seen with in vivo metastasis. IFNy not only reduces the invasiveness of the DU145 cells, but the few cells that do escape the tumor site could be more readily attacked by the host immune system. As stated above, since both IFNa and IFNy upregulate p21, the phenotypic changes that result from IFNy must be independent of p21. It is likely that IFNs influence multiple biochemical pathways that may or may not overlap. Although neither IFNa or IFNy induced the CKIs p27 or pl6, it is possible that IFNs can affect other regulators and inhibitors that influence tumor cell growth. Previous studies with Daudi cells have shown that IFNs suppress the phosphorylation of pRB (Melamed et al., 1993). Although this modulation of pRB is insignificant in DU145 cells, the cell cycle is still ii\hibited. The ability of IFNs to affect multiple cellular pathways in order to regulate cell growth allows for the potential of IFNs to affect a wide range of cell types with various genetic mutations. For example, regulation of the EGF production is significant for the regulation of adenocarcinomas as well as cancers of other origins due to the number of genes and biochemical pathways stimulated by the EGFR signal transduction. Based on the effects of IFNs on DU145 cells, cancers with mutations in p21 and/or the EGFR, such as the erb-B family of oncogenes, may not be the best candidates for IFN therapy.

PAGE 99

90 Determining the mechanism behind IFNs antitumor capabilities is also useful for finding potential combination therapies. By understanding the pathways influenced by IFNs, cytotoxic agents that can produce synergistic effects when combined with IFNs may be found. Even though both IFNa and IFN7 inhibit DU145 cell growth, the phenotypic changes induced by IFNy suggest that it is superior to IFNa as an antitumor agent for DU145 prostate cancer cells. This is in contrast to studies carried out using the Daudi cell line (Subramaniam et al., in press). While IFNa induces a strong Gl arrest, IFNy has no antiproliferative effects on Daudi cells. There are several potential explanations for the differential effects of IFNs on various cell lines. Possibilities include IFN receptor expression, signal transduction pathways, and target genes. For this reason, studies into the mechanism behind the cellular actions of IFNs are relevant. The characterization of the antitumor effects of IFNs on DU145 cells gives insight into the poorly defined mechanism of IFN antiproHferative and antitumor functions. The DU145 cell line is a metastatic adenocarcinoma with mutations in key tumor suppressor genes that are commonly defective in a variety of human cancers. Even with these mutations and metastatic potential, both IFNa and IFNy are able to exert potent antiproliferative activities, with IFNy also inducing a phenotypic change that results in an antitumor effect. Future studies will include further defining the mechanism of IFNy effects on DU145 cells. These will include determining how IFNy controls the expression of the EGFR. In addition, the possible involvement of

PAGE 100

91 J; coUagenase activity in the IFNy reduction of the DU145 invasive potential, via the downregulation of the EGFR, may be determined. Finally, work with nude mice will determine the effects on IFNy on the metastatic potential of DU145 cells using an in vivo system. These studies will expand on the work contained in this dissertation and provide further details of the mode of action of IFNs on an adenocarcinoma cell line.

PAGE 101

REFERENCE LIST Aaronson, S.A. (1991) Growth factors and cancer. Science 254: 1146-1153. Aguet, M., Dembic, Z., and Merlin, G. (1988) Molecular cloning and expression of the human interferony receptor. Cell 55: 273-280. Albini, A., Iwamoto, Y., Kleinman, H.K., Martin, G.R., Aaronson, S.A., Kozlowski, and McEwan, R.N. (1987) A rapid in vitro assay for quantitating the invasive potential of tumor cells. Cancer Res. 46: 32393245. Almadori, G., Cadoni, G., Maurizi, M., Ottaviani, F., Paludetti, G., Cattani, P., and Scambia, G. (1995) Oncogenes and cancer of the larnyx. EGFR, p21 ras and HPV-DNA infections. Acta Otorhinolaryngol. Ital. 15: 1-22. Amati, B., Brooks, M.W., Levy, N., Littlewood, T.D., Evan, G.I., and Land, H. (1993) Oncogenic activity of the c-myc protein requires dimerization with Max. Cell 72: 233-245. Baron, S., Tyring, S.K., Fleischmann, W.R., Coppenhaver, D.H., Niesel, D.W., Klimpel, G.R., Stanton, J., and Hughes, T.K. (1991) The interferons: mechanisms of action and clinical applications. J. Amer. Med. Assoc. 266: 1375-1383. Bazer, F.W., and Johnson, H.M. (1991) Type I conceptus interferons: Maternal recognition of pregnancy signals and potential therapeutic agents. Am. J. Reprod. Immunol. 26: 19-22. Bishop, J.M. (1991) Molecular themes in oncogenesis. Cell 64: 235-248. Black, CM., Remington, J.S., and McCabe, R.E. (1988) Toxoplasma. In: G.I. Byrne and J. Turco (eds.) Interferon and Nonviral Pathogens. New York: Marcel Dekker Inc., pp. 237-261. Blalock, J.E., Georgiades, J.A., Langford, M.P., and Johnson, H.M. (1980) Purified human immune interferon has more potent anticellular activity than fibroblast or leukocyte interferon. Cell. Immunol 49390394. 92

PAGE 102

93 Bookstein, R., Shew, J.-Y., Chen, P.-L., Scully, P., and Lee, W.-H. (1990) Suppression of tumorigenicity of human prostate carcinoma cells by replacing a mutated RB gene. Science 24: 712-715. Branca, A.A. (1988) Interferon receptors. In vitro Cell. Devel. Biol. 24: 155165. Caldenhoven, E., Coffer, P., Yuan, J., van de Stolpe, A., Horn, F., Kruijer, W., and van der Saag, P.T. (1994) Stimulation of the human intercellular adhesion molecule-1 promoter by interleukin-6 and IFN-y involves binding distinct factors to a palindromic response element. J. Biol. Chem. 269: 21146-21154. Cavenee, W.K., and White, R.L. (1995) The genetic basis of cancer. Sci. Am. 272: 72-79. Chen, L., Tourvieille, B., Burns, G.F., Bach, F.H., Mathieu-Mahul, D., Sasportes, M., and Bensussan, A. (1986) Interferon: a cytotoxic T lymphocyte differentiation signal. Eur. J. Immunol. 16: 767-770. Cobrinik, D., Dowdy, S.F., Hinds, P.W., Mittnacht, S., and Weinberg, R.A. (1992) The retinoblastoma protein and regulation of cell cycling. Trends Biochem. Sci. 17: 312-315. Cohen, B., Novick, D., Barak, S., and Rubinstein, M. (1995) Ligand-induced association of the type I interferon receptor components. Mol. Cell. Biol. 15: 4208-4214. Cohen, D.W., Simak, R., Fair, W.R., Melamed, J., Scher, H.I., and CordonCardo, C. (1994) Expression of transforming growth factor-a and the epidermal growth factor receptor in human prostate tissues. J. Urol. 152: 2120-2124. Connolly, J.M., and Rose, D.P. (1991) Autocrine regulation of DU145 human prostate cancer cell growth by epidermal growth factor related polypeptides. The Prostate 19: 173-180. Creasey, A.A., Bartholomew, J.C., and Merigan, T.C. (1980) Role of GO-Gl arrest in the inhibition of tumor cell growth by interferon. Proc. Natl. Acad. Sci. USA 77: 1471-1475. Croze, E., Russell-Harde, D., Charis Wagner, T., Haifeng, P., Pfeffer, L.M., and Perez, H.D. (1996) The human type I interferon receptor. J. Biol. Chem. 271: 33165-33168.

PAGE 103

94 Dana, N., and Arnaout, M.A. (1994) The role of |i2 integrins in leukocyte adhesion. In: Y. Takada (ed.) Integrins the biological problems. Boca Raton: CRC Press, pp. 37-54. Darnell, J., Kerr, I., and Stark, G. (1994) JAK-STAT pathways and transcriptional activation in response to IFNs and other extracellular signaling proteins. Science 264: 1415-1421. Decker, T., Lew, D.J., Mirkovitch, J., and Darnell, J.E. (1991) Cytoplasmic activation of GAP, an IFN-gamma regulated DNA-binding factor. EMBO J. 10:927-932. Dedhar, S., Saulnier, R., Nagle, R., and Overall, CM. (1993) Specific alterations in the expression of a3bl and a6b4 integrins in highly invasive and metastatic variants of human prostate carcinoma cells selected by in vitro invasion through reconstituted basement membrane. Clin. Exp. Metastasis 11: 391-400. Degre, M., and Bukholm, G. (1988) Interferon effects on infection with enteroinvasive bacteria. In: G.I. Byrne and J. Turco (eds.) Interferon and Nonviral Pathogens. New York: Mercel Dekker Inc., pp. 295-310. Domanski, P., Witte, M., Kellum, M., Rubinstein, M., Hackett, R., Pitha, P., and Colamonici, O.R. (1995) Cloning and expression of a long form of the beta subunit of the interferon alpha beta receptor that is required for signaling. J. Biol. Chem. 270: 21606-21611. Dong, J.-T., Lamb, P.W., Rinker-Schaeffer, C.W., Vukanovic, J., Ichikawa, T., Isaacs, J.T., and Barrett, J.C. (1995) KAIl, a metastasis suppressor gene for prostate cancer on human chromosome llpll.2. Science 268: 884886. Dorr, R.T. (1993) Interferon-a in malignant and viral diseases. Drugs 45: 177211. Einat, M., Resnitzky, D., Kimchi, A. (1985a) Close link between reduction of c-myc expression by interferon and GO/Gl arrest. Nature 313: 597-600. Einat, M., Resnitzky, D., Kimchi, A. (1985b) Inhibitory effects of interferon on the expression of genes regulated by platelet-derived growth factor. Proc. Natl. Acad. Sci. USA 82: 7608-7612. El-Deiry, W.S., Harper, J.W., O'Connor, P.M., Velculescu, V.E., Canman, C.E., Jackman, J., Pietenpol, J.A., Burrell, M., Hill, D.E., Wang, Y., Wiman, K.G., Mercer, W.E., Kastan, M.B., Kohn, K.W., Elledge, S.J., Kinzler,

PAGE 104

95 K.W., and Vogelstein, B. (1994) WAFl/CIPl is induced in p53mediated Gl arrest and apoptosis. Cancer Res. 54: 1169-1174. El-Deiry, W.S., Tokino, T., Velculescu, V.E., Levy, D.B., Parsons, R., Trent, J.M., Lin, D., Mercer, W.E., Kinzler, K.W., and Vogelstein, B. (1993) WAFl, a potential mediator of p53 tumor suppression. Cell 75: 817-825. EUedge, S.J., and Harper, J.W. (1994) Cdk inhibitors: on the threshold of checkpoints and development. Curr. Opin. Cell. Biol. 6: 847-852. Familletti, P. C, Rubinstein, S., and Petska, S. (1981) A convenient and rapid cytopathic effect inhibition assay for interferon. Methods Enzymol. 78: 387-394. Farrar, M.A., and Schreiber, R.D. (1993) The molecular cell biology of interferon-Y and its receptor. Annu. Rev. Immunol. 11: 571-611. Finkelman, F.D., Katona, I.M., Mosmann, T.R., and Coffman, R.L. (1988) IFN gamma regulates the isotypes of IgG secreted during in vivo humoral immune responses. J. Immunol. 140: 1022-1027. Fleischman, W.R., and Fleischman, CM. (1992) Mechanisms of interferons' antitumor actions. In: S. Baron, D.H. Coppehaver, F. Dianzani, W.R. Fleischman Jr., T.K. Hughes Jr., G.R. KHmpel, D.W. Niesel, G.J. Stanton, S.K. Tyring (eds.) Interferon: principals and medical applications. Galveston: The University of Texas Medical Branch, pp. 299-309. Flores-Rozas, H., Kelman, Z., Dean, F.B., Pan, Z.Q., Harper, J.W., EUedge, S.J., ODormell, M., and Hurwitz, J. (1994) Cdk-interacting protein 1 directly binds with proliferating cell nuclear antigen and inhibits DNA replication catalyzed by the DNA polymerase delta holoenzyme. Proc. Natl. Acad. Sci. USA 91: 8655-8659. Friedman, R.L., Manly, S.P., McMahon, M., Kerr, I.M., and Stark, G.R. (1984) Transcriptional and posttranscriptional regulation of interferoninduced gene expression in human cells. Cell 38: 745-755. Fountoulakis, M., Zulauf, M., Lustig, A., and Garotta, G. (1992) Stoichiometry of interaction between interferony and its receptor. Eur. J. Biochem. 208: 781-787. Fu, X.-Y., Kessler, D.S., Veals, S.A., Levy, D.E., and Darnell Jr., J.E. (1990) ISGF3, the transcriptional activator induced by interferon alpha, consists of multiple interacting polypeptide chains. Proc. Natl. Acad. Sci. USA 87: 8555-8559.

PAGE 105

96 Fu, X.-Y., Schindler, C, Improta, T., Aebersold, R., and Darnell Jr., J.E. (1992) The proteins of ISGF-3, the interferon alpha-induced transcriptional activator, define a gene family involved in signal transduction. Proc. Natl. Acad. Sci. USA 89: 7840-7843. Gangopadhyay, S.B., Abraham, J., Lin, Y.P., and Benchimol, S. (1997) The tumor suppressor gene p53. In: G. Peters and K.H. Vousden (eds.) Oncogenes and tumor suppressors. New York: Oxford University Press, pp. 261-291. Gamick, M.B. (1994) The dilemmas of prostate cancer. Sci. Am. 270: 72-81. Gartel, A.L., Serfas, M.S., and Tyner, A.L. (1997) p21Negative regulator of the cell cycle. Proc. Soc. Exp. Biol. Med. 213: 138-149. Gastl, G., and Ruber, C. (1988) The biology of interferon actions. Blut 56: 193199. Gauci, L. (1987) Management of cancer patients receiving interferon alpha2a. Int. J. Cancer supp 1: 21-30. Giedlund, M.A., Om, H., Wigzell, H., Senik, A., and Gresser, I. (1987) Enhanced NK cell activity in mice injected with interferon and interferon inducers. Nature 273: 759-763. Grana, X., and Reddy, E.P. (1995) Cell cycle control in mammalian cells: role of cyclins, cyclin dependent kinases (cdks), growth suppressor genes and cyclin-dependent kinase inhibitors (CKIs). Oncogene 11: 211-219. Gray, P. W., and Goeddel, D. V. (1982) Structure of the human immune interferon gene. Nature 298: 859-863. Greenlund, A.C., Farrar, M.A., Viviano, B.L., and Schreiber, R.D. (1994) Ligand induced interferon gamma receptor tyrosine phosphorylation couples the receptor to its signal transduction system. EMBO J. 13: 15911600. Greenlund, A.C., Schreiber, R.D., Goeddel, D.V., and Pennica, D. (1993) Interferon-Y induces receptor dimerization in solution and on cells. J. Biol. Chem. 268: 18103-18110. Gu, Y., Turck, C.W., Morgan, D.O. (1993) Inhibition of CDK2 activity in vivo by an associated 20K regulatory subunit. Nature 366: 707-710.

PAGE 106

97 Guan, K.L., Jenkins, C.W., Li, Y., Nichols, M.A., Wu, X., O'Keefe, C.L., Matera, A.G., and Xiong, Y. (1994) Growth suppression by pl8, a pl6INK4/MTSland pl4INK4B/MTS2 -related CDK6 inhibitor, correlates with wild-type pRB function. Genes Dev. 8: 2939-2952. Guillouf, C., Rosselli, F., Krishnaraju, K., Moustacchi, E., Hoffman, B., and Liebermarm, D.A. (1995) p53 involvement in control of G2 exit of the cell cycle: role in DNA damage-induced apoptosis. Oncogene 10: 22632270. Gutterman, J.U. (1994) Cytokine therapeutics: lessons from interferon a. Proc. Natl. Acad. Sci., USA 91: 1198-1205. Hanningan, G.E., and Dedhar, S. (1997) Adhesion molecules in tumor growth and metastasis. In: L.C. Paul and T.B. Issekutz (eds.) Adhesion molecules in health and disease. New York: Marcel Dekker Inc., pp. 445-481 Harmon, G.J., and Beach, D. (1994) pl5INK4B is a potential effector of TGFbetainduced cell cycle arrest. Nature 371: 257-261. Harper, J.W., Adami, G.R., Wei, N., Keyomarsi, K., and EUedge, S.J. (1993) The p21 cdk-interacting protein Cipl is a potent inhibitor of Gl cyclindependent kinases. Cell 75: 805-816. Hart, I.R. (1996) Adhesion receptors and cancer. In: M.A. Horton (ed.) Molecular biology of cell adhesion molecules. New York: John Wiley & Sons Ltd., pp. 87-98. Hatakeyama, M., and Weinberg, R.A. (1995) The role of RB in cell cycle control. In: L. Meijer, S. Guidet, H.Y.L. Tung (eds.) Advances in cell cycle research vol. I. New York: Plenum Press, pp. 9-19. Heldin, C.-H., and Ronnstrand, L. (1997) Growth factor receptors in cell transformation. In: G. Peters and K.H. Vousden (eds.) Chicogenes and tumor suppressors. New York: Oxford University Press, pp. 55-85. Henco, K., Brosius, J., Fujisawa, A., Fujisawa, J.I., Haynes, J.R., Hochstadt, J., Kovacic, T., Pasek, M., Schambock, A., Schmid, J., Todokoro, K., Walchli, M., Nagata, S., and Weissman, C. (1985) Structural relationship of human interferon alpha genes and pseudogenes. J. Mol. Biol. 185: 227-260. Herberman, R.B. (1986) Effects of biological response modifiers on effector cells with cytotoxic activity against tumors. Semin. Oncol. 13: 195-204.

PAGE 107

98 Hirai, H., Roussel, M.F., Kato, J.-Y., Ashmun, R.A., and Sherr, C.J. (1995) Novel INK4 proteins, pl9 and pl8, are specific inhibitors of the cyclin D-dependent kinases CDK4 and CDK6. Mol. Cell. Biol. 15: 2672-2681. Houghton, A.N., Thomson, T.M., Gross, D., Oettgen, H.F., and Old, L.J. (1984) Surface antigens of melanoma and melanocytes: specificity of induction of la antigens by human y-interferon. J. Exp. Med. 160: 255269. Huang, S., Hendriks, W., Althage, A., Hemmi, S., Bluethmann, H., Kamijo, R., Vilcek, J., Zinkernagel, R.M., and Aguet, M. (1993) Immune response in mice that lack the IFN gamma receptor. Science 259: 17421745. Hynes, R.O. (1992) Integrins: versatility, modulation, and signaling in cell adhesion. Cell 69: 11-25. Igarashi, K., Garotta, G., Ozmen, L., Ziemiecki, A., Wilks, A.F., Harpur, A.G., Lamer, A.C., and Finbloom, D.S. (1994) Interferon gamma induces tyrosine phosphorylation of interferon gamma receptor and regulated association of protein tyrosine kinases, JAKl and JAK2, with its receptor. J. Biol. Chem. 269: 14333-14336. loachim, E., Kamina, S., Athanassiadou, S., and Agnantis, N.J. (1996) The prognostic significance of epidermal growth factor receptor (EGFR), CerbB-2, Ki-67 and PCNA expession in breast cancer. Anticancer Res. 16: 3141-3147. Isaacs, A., and Lindenmann, J. (1957) Virus interference. I. The interferon. Proc. Roy. Soc. London Ser. B. 147: 258-267. Isaacs, W.B., Carter, B.S., and Ewing, CM. (1991) Wild-type p53 suppresses growth of human prostate cancer cells containing mutant p53 alleles. Cancer Res. 52: 4716-4720. Johnson, H.M., Bazer, F.W., Szente, B.S., and Jarpe, M.A. (1994) How interferons fight disease. Sci. Am. 270: 68-75. Johnson, H.M., and Farrar, W.L. (1983) The role of an interferon gamma-like lymphokine in the activation of T cells for expression of interleukin 2 receptors. Cell. Immunol. 75: 154-159. Johnson, H.M., and Torres, B.A. (1983) Recombinant mouse gamma interferon regulation of antibody production. Infect. Immun. 41: 546548.

PAGE 108

99 Jonak, G.J., and Knight, J.E. (1984) Selective reduction of c-myc RNA in Daudi cells by human beta interferon. Proc. Natl. Acad. Sci. USA 81: 1747-1750. Kastan, M.B., Zhan, Q., El-Deiry, W.S., Carrier, F., Jacks, T., Walsh, W.V., Plunkett, B.S., Vogelstein, B., and Fomace Jr., A.J. (1992) A mammalian cell cycle checkpoint pathway utilizing p53 and GADD45 is defective in ataxia-telangiectasia. Cell 71: 587-597. Kaufmann, W.K. (1995) Cell cycle checkpoints and DNA repair preserve the stability of the human genome. Cancer Metastasis Rev. 14: 31-41. Kelman, Z. (1997) PCNA: structure, function, and interactions. Oncogene 14: 629-640. Kerr, I.M. (1987) The 2-5A system. A personal view. J. Interferon Res. 7: 505510. Kessler, D.S., Veals, S.A., Fu, X.-Y., and Levy, D.E. (1990) Interferon-alpha regulates nuclear translocation and DNA binding affinity of ISGF3, a multimeric transcriptional activator. Genes Dev. 4: 1753-1763. Koff, A., Giordano, A., Desai, D., Yamashita, K., Harper, J.W., Elledge, S., Nishimoto, T., Morgan, D.O., Franza, B.R., and Roberts, J.M. (1992) Formation and activation of a cyclin E-cdk2 complex during the Gl phase of the human cell cycle. Science 257: 1689-1694. Kristensen, G.B., Holm, R., Abeler, V.M., and Trope, C.G. (1996) Evaluation of the prognostic significance of cathepsin D, epidermal growth factor receptor, and c-erbB-2 in early cervical squamous cell carcinoma. Cancer 78: 433-440. Kuerbitz, S.J., Plunkett, B.S., Walsh, W.V., and Kastan, M.B. (1992) Wild-type p53 is a cell cycle checkpoint determinant following irradiation. Proc. Natl. Acad. Sci. USA 89: 7491-7495. Larner, A.C., Chaudhuri, A., and Darnell Jr., J.E. (1986) Transcriptional induction by interferon. New protein(s) determine the extent and length of the induction. J. Biol. Chem. 261: 453-459. Larner, A.C., Jonak, G., Cheng, Y.S., Korant, B., Knight, E., and Darnell Jr., J.E. (1984) Transcriptional induction of two genes in human cells by beta interferon. Proc. Natl. Acad. Sci. USA 81: 6733-6737.

PAGE 109

100 La Thangue, N.B. (1997) The retinoblastoma gene product and its relatives. In: G. Peters and K.H. Vousden (eds.) Oncogenes and tumor suppressors. New York: Oxford University Press, pp. 233-260. Lavoie, J.N., Rivard, N., L'AUemain, G., and Pouyssegur, J. (1996) A temporal and biochemical link between growth factor-activated MAP kinases, cyclin Dl induction and cell cycle entry. In: L. Meijer, S. Guidet, and L. Vogel (eds.) Progress in cell cycle research vol. H. New York: Plenum Press, pp. 49-58. Leaman, D.W., Leung, S., Li, X., and Stark, G.R. (1996) Regulation of STATdependent pathways by growth factors and cytokines. FASEB J. 10: 1578-1588. Lees, E.M., and Harlow, E. (1995) Cancer and the cell cycle. In: C. Hutchison and D.M. Glover (eds.) Cell cycle control. New York: Oxford University Press, pp. 228-263. Lengyel, P. (1982) Biochemistry of interferons and their actions. Annu. Rev. Biochem. 51: 251-282. Lengyel, P. (1987) Double-stranded RNA and interferon action. J. Interferon Res. 7: 511-519. Levine, A.J. (1993) The tumor suppressor genes. Annu. Rev. Biochem. 62: 623-651. Levine, A.J., Momand, J., and Finlay, C.A. (1991) The p53 tumor suppressor gene. Nature 351: 453-455. Levy, D., and Darnell Jr., J.E. (1990) Interferon-dependent transcriptional activation: signal transduction without second messenger involvement? New Biol. 2: 923-928. Lew, D., Decker, T., Strehlow, I., and Darnell Jr., J.E. (1991) Overlapping elements in the guanylate binding protein gene promoter mediate transcription induction by alpha and gamma interferons. Mol Cell Biol. 11: 182-191. Li, J., Yen, C, Liaw, D., Podsypanina, K., Bose, S., Wang, S., Puc, J., Miliaresis, C, Rodgers, L., McCombie, R., Bigner, S., Giovanella, B., Ittmann, M., Tycko, B., Hibshoosh, H., Wigler, M., and Parsons, R. (1997) PTEN, a putative protein tyrosine phosphatase gene mutated in human brain, breast, and prostate cancer. Science 275: 1943-1947.

PAGE 110

101 Libermann, T.A., Nusbaum, H.R., Razon, N., Kris, R., Lax, I., Soreg, H., Whittle, N., Waterfield, M.D., Ullrich, A., and Schlessir\ger, J. (1985) Amplification and overexpression of the EGF receptor gene in primary human glioblastomas. J. Cell. Sci. Suppl. 3: 161-172. Liotta, L.A. (1987) Biochemical mechanism of tumor invasion and metastsis. Clin. Physiol. Biochem. 5: 190-199. Lutfalla, G., Holland, S.J., Cinato, E., Monneron, D., Reboul, J., Rogers, N.C., Smith, J.M., Stark, G.R., Gardiner, K., Mogensen, K.E., Kerr, I.M., and Uze, G. (1996) Mutant USA cells are complemented by an interferonalpha beta receptor subunit generated by alternative processing of a new member of a cytokine gene cluster. EMBO J. 14: 5100-5108. Makgoba, M.W., Sanders, M.E., Ginther Luce, G.E., Gugel, E.A., Dustin, M.L., Springer, T.A., and Shaw, S. (1988) Functional evidence that intercellular adhesion molecule-1 (ICAM-1) is a ligand for LFA-1 dependent adhesion in T cell-mediated cytotoxicity. Eur. J. Immunol. 18: 637-640. Marcu, K.B., Bossone, S.A., and Patel, A.J. (1992) Myc function and regulation. Annu. Rev. Biochem. 61: 809-860. Marlin, S.D., and Springer, T.A. (1987) Purified intercellular adhesion molecule (ICAM-1) is a ligand for lymphocyte-associated antigen 1 (LFA-1). Cell 5: 813-819. Melamed, D., Tiefenbrun, N., Yarden, A., and Kimchi, A. (1993) Interferons and interleukin-6 suppress the DNA-binding activity of E2F in growthsensitive hematopoietic cells. Mol. Cell. Biol. 13: 5255-5265. Morgan, D.O. (1995) Principles of CDK regulation. Nahire 374: 283-287. Nakajima, Y., Konno, S., Perruccio, Y., Chen, J.M., Wu, S., An, J., Chiao, M., Choudhury, M., Mallouh, C, and Muraki, J. (1994) The effects of IFNbeta on growth of human prostatic JCA-1 cells. Biochem. Biophys. Res. Commun. 200: 467-474. Nakopoulou, L., Zervas, A., Constantinides, C, Deliveliotis, C, Stefanaki, K., and Dimopoulos, C. (1995) Epithelial differentiation antigens and epidermal growth factor receptors in transitional cell bladder carcinoma: correlation with prognosis. Urol. Int. 54: 191-197. Nathan, C.F., Murray, H.W., Wiebe, M.E., and Rubin, B.Y. (1983) Identification of interferon y as the lymphokine that activates human

PAGE 111

102 macrophage oxidative metabolism and antimicrobial activity. J. Exp. Med. 158: 670-689. ,„ ^. ^ ^w ^ Neisel, D.W., Hess, C.B., Cho, Y.-J., Klimpel, K.D., and Klimpel, G.R. (1986) Natural and recombinant IFN inhibit tissue culture cell invasion by Salmonella and Shigella. Infect. Immun. 52: 828-833. Netto, G.J., and Humphrey, P.A. (1994) Molecular biological aspects of human prostatic carcinoma. Am. J. Clin. Pathol. 102: S57-S64. Novick, D., Cohen, B., and Rubinstein, M. (1994) The human interferon alpha/beta receptor: characterization and molecular cloning. Cell 77: 391-400. Nurse, P. (1990) Universal control mechanism regulating onset of M-phase. Nature 344: 503-508. Parker, S.L., Tong, T., Bolden, S., and Wingo, P.A. (1996) Cancer statistics, 1996. CA Cancer J. Clin, 46: 5-27. Pauker, K., Cantrell, K., and Henle, W. (1962) Quantitative studies on the viral interference in suspended L-cells: III. Effect of interfering viruses and interferon on the growth rate of cells. Virology 17: 324-334. Petska, S., Langer, J.A., Zoon, C.K., and Samuel, C.E. (1987) Interferons and their actions. Annu. Rev. Biochem. 56: 727-777. Pines, J. (1994) Arresting developments in cell-cycle control. Trends Biochem. Sci. 19: 143-145. Pines, J., and Hunter, T. (1995) Cyclin dependent kinases: an embarrassment if riches? In: C. Hutchinson and D.M. Glover (eds.) Cell Cycle Control. New York: Oxford University Press, pp. 144-176. Pledger, W.J., Stiles, CD., Antoniades, H.N., and Scher, CD. (1977) Induction of DNA synthesis in BALB/c 3T3 cells by serum components: reevaluation of the commitment process. Proc. Natl. Acad. Sci 744481-4485. Polyak, K., Kato, J.-Y., Solomon, M.J., Sherr, CJ., Massague, J., Roberts, J.M., and Koff, A. (1994) p27^\ a cyclin-cdk inhibitor links transforming growth factor-p and contact inhibition to cell cycle arrest. Genes Dev 89-22.

PAGE 112

103 Pontzer, C.H., Bazer, F.W., and Johnson, H.M. (1991) Antiproliferative activity of a pregnancy recognition hormone, ovine trophoblast protein-1. Cancer Res. 51: 5304-5307. Pontzer, C.H., Torres, B.A., Vallet, J.L., Bazer, F.W., and Johnson, H.M. (1988) Antiviral activity of the pregnancy recognition hormone ovine trophoblast protein-1. Biochem. Biophys. Res. Commun. 152: 801-807. Prewett, M., Rockwell, P., Rockwell, R.F., Giorgio, N.A., Mendelsohn, J., Scher, H.I., and Goldstein, N.I. (1996) The biological effects of C225, a chimeric monoclonal antibody to the EGFR, on human prostate carcinoma. J. Immunother. 19: 419-427. Quesada, J.R., Talpaz, M., Rios, A., Kurzrock, R., and Gutterman, J.U. (1986) Clinical toxicity of interferons in cancer patients: a review. J. Clin. Oncology 4:234-243. Rani, S., Foster, G.R., Leung, S., Leaman, D., Stark, G.R., and Ransohoff, R. (1996) Characterization of beta-Rl, a gene that is selectively induced by interferon beta (IFN-beta) compared with IFN-alpha. J. Biol. Chem. 271: 22878-22884. Reich, N., Evans, B., Levy, D., Fahey, D., Knight Jr, E., and Darnell, J.E. (1987) Interferon-induced transcription of a gene encoding a 15 KDa protein depends on an upstream enhancer element. Proc. Natl. Acad. Sci. USA 84:6394-6394. Roos, G., Leanderson, T., and Lundgren, E. (1984) Interferon-induced cell cycle changes in human hematopoietic cell lines and fresh leukemic cells. Cancer Res. 44: 2358-2362. Samuel, C.E. (1979) Mechanisms of interferon action: phosphorylation of protein synthesis initiation factor eIF-2 in interferon-treated cells by a ribosome-associated kinase possessing site specificity similar to heminregulated rabbit reticulocyte kinase. Proc. Natl. Acad. Sci. USA 76:600607. Samuel, C. E. (1988) Mechanisms of the antiviral action of interferons. Prog. Nucleic Acid Res. Mol. Biol. 35: 27-72. Schwartz, R., Momburg, F., Moldenhauer, G., Dorken, B., and Schirrmacher, V. (1985) Induction of HLA class-II antigen expression on human carcinoma cell lines by IFN-gamma. Int. J. Cancer 35: 245-250.

PAGE 113

104 Sekar, V., Atmar, V.J., Joshi, A.R., Krim, M., and Kuehn, G. (1983) Inhibition of ornithine decarboxylase in humans fibroblast cells by type I and type II interferons. Biochem. Biophys. Res. Commun. 114: 950-954. Sen, G. C., and Lengyel, P. (1992) The interferon system. J. Biol. Chem. 267(8): 5017-5020. Serrano, M., Hannon, G.J., and Beach, D. (1993) A new regulatory motif in cell-cycle control causing specific inhibition of cyclin D/CDK4. Nature 366: 704-707. Serrano, M., Lee, H., Chin, L., Cordon-Cardo, C., Beack, D., and DePinho, R.A. (1996) Role of the INK4a locus in tumor suppression and cell mortality. CeU 85: 27-37. Sherr, C.J. (1993) Mammalian Gl cyclins. Cell 73: 1059-1065. Sherr, C.J. (1994) Gl phase progression: cycling on cue. Cell 79: 551-555. Shuai, K., Horvath, CM., Tsai Huang, L.H., Qureshi, S.A., Cowburn, D., and Darnell Jr., J.E. (1994) Interferon activation of the transcription factor STAT91 involves dimerization through SH2-phosphotyrosyl peptide interactions. Cell 76: 821-828. Sica, G., Fabbroni, L., Castagnetta, L., Cacciatore, M., and Pavone-Macaluso, M. (1989) Antiproliferative effect of interferons on human prostate carcinoma cell lines. Urol. Res. 17: 111-115. Silvennoinen, O., Ihle, J.N., Schlessinger, J., and Levy, D.E. (1993) Interferoninduced nuclear signaling by JAK protein tyrosine kinases. Nature 366: 583-585. Snapper, CM., Peschel, C, and Paul, W.E. (1988) IFN-gamma stimulates IgG2a secretion by murine B cells stimulated with bacterial lipopolysaccharide. J. Immunol. 140: 2121-2127. f ;• • Soh, J., Donnely, R.O., Kotenko, S., Mariano, T.M., Cook, J.R., Wang., N., Emanuel, S., Schwartz, B., Miki, T., and Petska, S. (1994) Identification and sequence of an accessory factor required for activation of the human IFN-gamma receptor. Cell 76: 793-802. Soos, J.M., and Johnson, H.M. (1995a) Type I interferon inhibition of superantigen stimulation: implications for treatment of superantigen associated disease. J. IFN Cytokine Res. 15: 39-45.

PAGE 114

105 Soos, J.M., Subramaniam, P.S., Hobeika, A.C., Schiffenbauer, J., and Johnson, H.M. (1995b) The IFN pregnancy recognition hormone IFNx blocks both development and superantigen reactivation of experimental allergic encephalomyelitis without associated toxicity. J. Immunol. 155: 2747-2753. Spiegel, R.J. (1987) The alpha interferons: clinical overview. Semin. Oncol. 14:1-12. Staeheli, P. (1990) Interferon-induced proteins and the antiviral state. Adv. Virus Res. 38:147-200. Staeheli, P., Danielson, P., Haller, O., and Sutclife, J.G. (1986) Transcriptional activation of the mouse Mx gene by type I interferon. Mol. Cell Biol. 6: 4770-4774. Steiner, P., Rudolph, B., Muller, D., and Filers, M. (1996) The functions of myc in cell cycle progression and apoptosis. In: L. Meijer, S. Guidet, L. Vogel (eds.) Progress in cell cycle research vol. 2. New York: Plenum Press, pp. 73-82. Stewart, M., Thiel, M., and Hogg, N. (1995) Leukocyte integrins. Curr. Opin. CeU. Biol. 7: 690-696. Stone, K.R., Mickey, D.D., Wunderli, H., Mickey, G.H., and Paulson, D.F. (1978) Isolation of a human prostate carcinoma cell line (DU145). Int. J. Cancer 21: 274-281. Stuart-Harris, R.C., Lauchlan, R., and Day, R. (1992) The clinical applications of the interferons: a review. Med. J. Aust. 156: 869-872. Stumm, G., Eberwein, S., Rostock, W.S., Stein, H., Pomer, S., Schlegel, J., and Waldherr, R. (1996) Concomitant overexpression of the EGFR and erbB-2 genes in renal cell carcinoma (RCC) is correlated with dedifferentiation and metastasis. Int. J. Cancer 69: 17-22. Subramaniam, P.S., Cruz, P.E., Hobeika, A.C., and Johnson, H.M. Type I interferon induction of the cdk-inhibitor p21WAFl is accompanied by ordered Gl arrest, differentiation and apoptosis of the Daudi B-cell line. Oncogene, in press. Subramaniam, P.S., and Johnson, H.M. (1997) A role for the cyclindependent kinase inhibitor p21 in the Gl cell cycle arrest mediated by the type I interferons. J. Interferon Cytokine Res. 17: 11-15.

PAGE 115

Tamm, I., Jasny, B.R., and Pfeffer, L.M. (1987) Antiproliferative action of interferons. In: L.M. Pfeffer (ed.) Mechanisms of interferon actions volume n. Boca Raton: CRC Press, pp. 25-57. Tiefenbrun, N., Melamed, D., Levy, N., Resnitzky, D., Hoffmann, L, Reed, S.I., and Kimchi, A. (1996) Alpha interferon suppresses the cyclin D3 and cdc25A genes, leading to a reversible GO-like arrest. Mol. Cell. Biol. 16: 3934-3944. Toyoshima, H., and Hunter, T. (1994) p27, a novel inhibitor of Gl cyclin-cdk protein kinase activity, is related to p21. Cell 78: 67-74. Tuo, W., Ott, T.L., and Bazer, F.W. (1993) Natural killer cell activity of lymphocytes exposed to ovine, type I, trophoblast interferon. Am. J. Reprod. Immunol. 29: 26-34. Uze, C, Lutfalla, G., and Gressor, I. (1990) Genetic transfer of a functional human interferon alpha receptor into mouse cells: cloning and expression of its cDNA. Cell 60: 225-234. Velazquez, L., Fellous, M., Stark, G.R., and Pellegrini, S. (1992) A protein tyrosine kinase in the interferon alpha /beta signaling pathway. Cell 70: 313-322. Vial, T., and Descotes, J. (1994) Clinical toxicity of the interferons. Drug Safety 10: 115-150. Waga, S., Hannon, G.J., Beach, D., and Stillman, B. (1994) The p21 inhibition of cyclin-dependent kinases controls DNA replication by interaction with PCNA. Nature 369: 574-578. Weigent, D.A., Stanton, G.J., and Johnson, H.M. (1983) Recombinant gamma interferon enhances natural killer cell activity similar to natural gamma interferon. Biochem. Biophys. Res. Commun. Ill: 525-529. Xie, H., Turner, T., Wang, M.H., Singh, R.K., Siegel, G.P., and Wells, A. (1995) In vitro invasiveness of DU145 human prostate carcinoma cells is modulated by EGF receptor-mediated signals. Clin. Exp. Metastasis 13407-419. Xiong, Y., Hannon, G.J., Zhang, H., Casso, D., Kobayashi, R., and Beach, D. (1993) p21 is a uiuversal inhibitor of cyclin dependent kinases. Nature 366: 701-704. Yamamoto, T., Kamata, N., Kawano, H., Shimizu, S., Kuroki, T., Toyoshima, K., Rikimaru, K., Nomura, N., Ishizaki, R., Pastan, I., Gamou, S., and

PAGE 116

Shimizu, N. (1986) High incidence of amplification of the epidermal growth factor receptor gene in human squamous carcinoma cell lines. Cancer Res. 46: 414-416. Yasui, H., Takai, K., Yoshida, R., and Hayaishi, O. (1986) Interferon enhances tryptophan metabolism by inducing pulmonary indoleamine 2,3dioxygenase: its possible occurrence in cancer patients. Proc. Natl. Acad. Sci. USA 83: 6622-6626.

PAGE 117

BIOGRAPHICAL SKETCH Amy Claudine Hobeika, a first-generation American, was born in Cincinnati, Ohio, on December 10, 1970, to Claude and Terry Hobeika. Her father Claude Hobeika was bom and raised in Alexandria, Egypt and came to the United States in 1959 to study medicine. While studying in Lexington, KY, he met a nursing student named Terry Taphorn, and they married in 1961. In 1963, they moved to Gainesville, Florida, where Dr. Hobeika became a resident at the University of Florida teaching hospital. Amy is the youngest of four children. Her sister Janine is the eldest and her two brothers John and Claude are in the middle. Amy attended Ann Weigel Elementary School and Seven Hills Middle and Upper Schools while growing up in Cincinnati, although spending her vacation time to a large extent in Sarasota, Florida. In 1988, she graduated from high school, and much to her fathers dismay, attended college at the University of the South in Sewanee, Tennessee, instead of at the University of Florida. After a rocky freshman year, she decided to study a field that actually held her interest and majored in biology. In 1992, Amy recieved her Bachelor of Science and, to her fathers great joy, moved to Gainesville, Florida, to further her education. In 1993, she was accepted into the Department of Microbiology and Cell Science where she was kindly taken into the laboratory of Dr. Howard Johnson. After graduation. Amy will remain in the Johnson laboratory as a postdoctoral fellow until her 108

PAGE 118

109 "significant other", Adrian, finishes medical school. Then, hopefully, it is on to a successful future.

PAGE 119

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Howard M. JohnsorT, Chair Graduate ResearclyTrofessor of Microbiology and Cell Science I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Edward M. Hofffnai Professor of Microbiology and Cell Science I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Thomas A. Bobik Assistant Professor of Microbiology and Cell Science I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. ^\e.\. K. Yamam^td^ Associate Professor of Veterinary Medicine

PAGE 120

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. ^ — — • Professor of Pathology and Laboratory Medicine This dissertation was submitted to the Graduate Faculty of the College of Agriculture and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. December, 1997 Dean, College of Agriculture Dean, Graduate School