Citation
Study of the anticancer effects of the vascular targeting agent combretastatin A-4 disodium phosphate (CA4DP)

Material Information

Title:
Study of the anticancer effects of the vascular targeting agent combretastatin A-4 disodium phosphate (CA4DP)
Alternate title:
Studies of the anticancer effects of the vascular targeting agent combretastatin A-4 disodium phosphate (CA4DP)
Creator:
Li, Lingyun, 1972-
Publication Date:
Language:
English
Physical Description:
vi, 126 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Angiogenesis ( jstor )
Apoptosis ( jstor )
Blood vessels ( jstor )
Cancer ( jstor )
Cell growth ( jstor )
Cells ( jstor )
Dosage ( jstor )
Endothelial cells ( jstor )
Heterologous transplantation ( jstor )
Tumors ( jstor )
Antineoplastic Agents -- pharmacology ( mesh )
Antineoplastic Combined Chemotherapy Protocols ( mesh )
Cisplatin ( mesh )
Department of Pharmacology and Therapeutics thesis Ph.D ( mesh )
Dissertations, Academic -- College of Medicine -- Department of Pharmacology and Therapeutics -- UF ( mesh )
Mice ( mesh )
Research ( mesh )
Sarcoma, Kaposi -- drug therapy ( mesh )
Sarcoma, Kaposi -- radiotherapy ( mesh )
Stilbenes -- pharmacology ( mesh )
Transplantation, Heterologous ( mesh )
Tubulin -- drug effects ( mesh )
Vinblastine ( mesh )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 2000.
Bibliography:
Bibliography: leaves 113-125.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Lingyun Li.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright [name of dissertation author]. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
025615104 ( ALEPH )
52305575 ( OCLC )

Downloads

This item has the following downloads:


Full Text














STUDIES OF THE ANTICANCER EFFECTS OF THE
VASCULAR TARGETING AGENT COMBRETASTATIN
A-4 DISODIUM PHOSPHATE (CA4DP)












By

LINGYUN LI


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


2000




























This dissertation is dedicated to my husband, Guoyi, and my parents for
their unlimited love and support.















ACKNOWLEGMENTS


I would like to express my sincere gratitude to Dr. Dietmar W. Siemann for

providing me with this very precious opportunity to work in a wonderful laboratory,

and for his immeasurable support and encouragement. This work would not have

been possible without his guidance and understanding.

I would also like to extend my gratitude to the members of my supervisory

committee, Dr. Thomas C. Rowe, Dr. Kathleen T. Shiverick, and Dr. James R.

Zucali, for their valuable advice and continuous encouragement in the completion of

my study. I am also grateful to Dr. Amyn Rojiani at the University of South Florida

for his assistance and collaboration in my graduate study.

In addition, I would like to express my appreciation to the past and present

members of Dr. Siemann's laboratory, including Dr. Kenneth Warrington, Jr.,

Wenyin Shi, Sharon Lepler, Destry Taylor, Heather Newlin, Emma Mercer, and

Howard Salmon, for their help and providing a pleasant working atmosphere. I am

indebted to all faculty and staff and student groups at the Department of

Pharmacology and Therapeutics for the countless help they have rendered and the

stimulating intellectual environment they provided.















TABLE OF CONTENTS

page

ACKNOWLEGMENTS ........................................ ............... ii

ABSTRACT .................. .. ............................... v

CHAPTERS

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

The Unique Physiology and Microenvironment of Solid Tumors .............. 2
Tumor Angiogenesis and Vasculature Targeting ................................. 6
Combretastatin A-4 Disodium Phosphate ........................................ 13
Significance .............. ........................................... 17

2 COMPARISON OF THE DIFFERENT RESPONSES OF NEOPLASTIC
AND NORMAL CELLS TO CA4DP IN VITRO ............................. 26

Introduction ..................... ...................................... 26
Material and Methods ................... ................................... 28
Results ..................... ................ ... ....... .. ........... 31
Discussion ...................... ...... ................. 32

3 EFFECTS OF CA4DP ON HUMAN MICROVASCULAR
ENDOTHELIAL CELLS IN VITRO .......................................... 40

Introduction ....................... .... .. ............. ............ .... 40
Material and Methods ....................... .................. ................. 43
Results ......................................................... ......... 46
Discussion ........ .. .. ..... ...................... ........... ............. 49

4 STUDY OF THE EFFECTS OF CA4DP IN THE MODEL
OF KAPOSI'S SARCOMA ...................................... 62

Introduction .................... ................... .... ....... ....... 62
M material and Methods .................... ................... ................. 65
Results ............... ..... ...... .... ... ...... ............... 68
Discussion .................. ................. ................ 71











5 STUDY OF THE EFFICACY OF CA4DP IN COMBINATION WITH
CONVENTIONAL ANTI-CANCER THERAPIES IN
KS XENOGRAFTS ..................... .................. .............. 89

Introduction ........................ ... ...... .......... 89
Material and Methods .................................................. .............. 92
Results .......................................... ............... 94
D discussion ..................... .. .... ................... ........ ...... 96

6 SUMMARY AND PERSPECTIVE ..................................... 107

REFERENCES ............ ...................................... 113

BIOGRAPHICAL SKETCH ......................... ............... ................ 126







































iv















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


STUDIES OF THE ANTICANCER EFFECTS OF THE
VASCULAR TARGETING AGENT COMBRETASTATIN
A-4 DISODIUM PHOSPHATE (CA4DP)

By

Lingyun Li

December 2000


Chair: Dietmar W. Siemann
Major Department: Pharmacology and Therapeutics

Combretastatin A-4 disodium phosphate (CA4DP) is a tubulin-binding agent

which has been shown to lead to rapid vascular shutdown in a variety of tumor

models. The present studies were undertaken to gain insight into the mechanisms) of

action of CA4DP and to evaluate the antitumor efficacy of CA4DP either alone or in

combination with conventional anticancer therapies in a xenograft model of Kaposi's

Sarcoma (KS).

Initial studies that compared the responses of normal and neoplastic cells to

CA4DP demonstrated that CA4DP had selective activity against proliferating

endothelial cells. Further studies showed that CA4DP treatment resulted in a time

dependent tubulin depolymerization in HMVEC-L cells. Tubulin disruption directly

affected the ability of endothelial cell migration and attachment. Studies carried out










with a clonogenic cell survival assay demonstrated that CA4DP selectively reduced

the number of viable proliferating HMVEC-L cells in a dose dependent manner.

Moreover, CA4DP induced the death of proliferating endothelial cells predominantly

by apoptosis. These findings provide a basis for the in vivo efficacy of CA4DP and its

selective action against the proliferating endothelial cell population found in tumors.

In vivo studies using Hoechst-33342 staining demonstrated that a single 100

mg/kg dose of CA4DP caused a rapid vascular shutdown in KS xenografts.

Histological evaluation showed evidence of morphological damage of tumor cells

within a few hours after drug treatment, followed by extensive haemorrhagic

necrosis. Studies using an in vivo to in vitro clonogenic cell survival assay further

demonstrated that administering increasing doses of CA4DP to tumor-bearing mice

resulted in a dose-dependent increase in tumor cell killing. CA4DP also significantly

enhanced the antitumor effects of radiation and chemotherapeutic agents (cisplatin

and vinblastine) in combination treatment studies. Moreover, repeated doses of

CA4DP treatment either alone or in combination with cisplatin treatment caused

significant growth delay in KS xenografts. These findings suggest that CA4DP ought

to be considered as a candidate agent for therapeutic evaluation in patients with

AIDS-KS in future Phase II clinical studies.















CHAPTER 1
INTRODUCTION



Cancer is characterized by progressive growth of cells that have lost their

proliferative control. These cells ultimately destruct tissue and metastasize to

organs distant from the primary site. In general, only one of three cancer victims

can be cured by a single treatment modality, usually surgery, radio- or

chemotherapy. The main problem with conventional cancer treatments, primarily

chemotherapy and radiation therapy, is that they do not have high specificity for

cancer cells. For radiation therapy, a degree of specificity is achieved by

localizing the radiation to the tumor and its immediate surrounding normal tissue.

For anticancer drugs, it is the rapid proliferation of the cancer cells that makes

them more sensitive to cell killing than their normal cell counterparts. However,

both modalities are limited by their toxicity to normal cells. In the case of

radiotherapy, normal tissue surrounding the tumor limits the radiation dose,

whereas for anticancer drugs, it is usually the killing of rapidly dividing normal

cells, such as those in the bone marrow, hair follicles, and epithelial cells lining

the gastrointestinal tract, that limit the dose that can be given. To achieve more

tumor-specific treatment, differences between normal and malignant cells are

being exploited. The physiology of solid tumors at the microenvironmental level

provides a unique and selective target for cancer treatment.












The Unique Physiology and Microenvironment of Solid Tumors

Abnormal Tumor Vasculature

The physiology of solid tumor differs from that of normal tissues in a

number of important aspects. A critical difference between tumors and normal

tissues is the abnormal nature of the tumor microcirculation compared to the well-

defined microvascular architecture of normal tissues (Konerding et al., 1995).

Solid malignant tumors are composed of both cancerous cells and normal host

component. Tumor growth, resulting from uncontrolled neoplastic cell division, is

absolutely dependent on a parallel proliferation of the nonmalignant cells which

comprise the tumor vasculature. There are two types of vessels in tumor tissues:

the existing vessels in normal tissues into which the tumor has invaded; and tumor

microvessels arising from neovascularization resulting from increased expression

of proangiogenic factors produced by tumor cells (Brown and Giaccia, 1998).

Both types of vessels develop structural and physiological abnormalities that have

become a hallmark of the tumor microvasculature. Studies have shown that tumor

blood vessels are highly irregular, tortuous, have arterio-venous shunts, blind

ends, lack smooth muscle or enervation, and have incomplete endothelial linings

and basement membranes (Grunt et al., 1985; Dewhirst et al., 1989; Shah-Yukich

and Nelson, 1988) (Figure 1-1). As a result, blood flow is often sluggish, highly

irregular, and the vessels are much leakierr" than those in normal tissues. Tumor

blood supply is therefore characterized by both spatial and temporal heterogeneity

in both structure and function.











Hypoxia

In neoplastic tissue, there is a disproportionate relationship between tumor

tissue and its vascular supply. Tumors are said to "outgrow" their blood supply;

neovascularization lags behind the increase in the number of neoplastic cells

(Tannock, 1970). As a consequence, the vascular network fails to provide

adequate nutritional support and leads to heterogeneous tumor microregions

varying in concentrations of oxygen, glucose, and other nutritional factors, as well

as metabolic waste products, both within and among tumors of the same

pathological grade and stage (Vaupel et al., 1989; Vaupel et al., 1996). In most

solid malignancies the tissue 02 status is poorer than in normal tissue at the site of

tumor growth. For example, the medium pO2 measured with electrodes for human

breast tumors was 28 mm Hg, whereas that of normal breast was 68 mm Hg

(Vaupel, 1994).

Due to the irregular blood flow and high interstitial pressure, some

therapeutic agents are poorly delivered to tumors. Cells located distant from the

functional blood supply, often hypoxic cells, could be resistant to drug therapy

because of three factors: (a) they are exposed to lower concentrations of drug than

those adjacent to blood vessels, primarily as a result of the metabolism of such

agents through successive cellular layers; (b) as a result of a decline in nutrient

and 02 availability, cells further away from the vascular system would be dividing

at a reduced rate (Amellem and Pettersen, 1991; Pallavicini et al., 1979); an

important consequence of this hypoxia-induced inhibition of proliferation is that

because most anticancer drugs are primarily effective against rapidly dividing










cells, their effectiveness would be expected to fall off as a function of distance

from blood vessels (Tomida and Tsuruo, 1999; Reynolds et al., 1996); (c)

oxygen-deficient cells may be inherently more resistant (Teicher, 1994).

Tumor hypoxia is also an important factor leading to resistance to

radiotherapy. In many series of human tumors at different sites, roughly half the

tumors have a median value of less than 10 mm Hg (Vaupel et al., 1991;

Nordsmark et al., 1994). This median of 10 mm Hg is significant in that this is the

point at which radiation resistance starts to develop with full resistance at values

of less than 0.5 mm Hg (Brown, 1999). A typical radiation killing curve for

mammalian cells under aerobic and hypoxic conditions is shown in Figure 1-2.

The difference in radiation sensitivity between the aerobic and hypoxic cells,

which is known as the oxygen enhancement ratio, is normally in the range 2.5-3

for mammalian cells. This effect, coupled with the finding that both human and

rodent tumors possess regions of tissue oxygenation below 10 mm Hg, indicates

why tumor hypoxia remains a key focus of research in radiobiology and

radiotherapy.

Recent studies have shown that hypoxia in solid tumors has an important

consequence in addition to conferring a direct resistance to radiation and

chemotherapy. Tumor hypoxia also stimulates tumor progression by promoting

angiogenesis through the induction of proangiogenic proteins such as vascular

endothelial growth factor (VEGF) (Shweiki et al., 1992). Clinical studies with

soft tissue sarcomas (Brizel et al., 1996) and with carcinoma of the cervix

(Sundfor et al., 1998) have shown that hypoxia is an independent and highly










significant prognostic factor predisposing tumors to metastatic spread. Therefore,

tumor hypoxia also is seen as a predisposing factor toward increased malignancy

and metastasis.

Because of their potential importance for treatment outcome, a large effort

has been afforded over the years to identifying strategies that will reduce or

eliminate hypoxic cells within solid tumors. A number of strategies to improving

tumor oxygenation are being investigated, including high oxygen content gas

breathing either alone (Fenton and Siemann, 1995) or coupled with the agent

nicotinamide (Horsman et al., 1994; Siemann et al., 1994), right shifting of the

oxyhemoglobin curve (Siemann and Macler, 1986; Hirst and Wood, 1987), and

use of agents that increase tumor blood flow (Vaupel and Menke, 1989). A

second approach is to use chemical sensitizers that mimic oxygen's ability to

increase the sensitivity of hypoxic cells to radiotherapy and chemotherapy

(Phillips and Wasserman, 1984). Over the past decade, several potent

bioreductive cytotoxins, such as E09 and tirapazamine, agents whose cytotoxic

activity is dramatically enhanced when they are metabolized in a hypoxic

environment, have been identified (Workman, 1992; Brown and Siim, 1996).

Specifically attacking the hypoxic cell subpopulations with bioreductive agents

has a greater therapeutic potential than oxygenating the cells or chemically

sensitizing them to radiation or chemotherapy. Not only is the killing tumor

specific hypoxicc is tumor specific), but the cells killed are the ones resistant to

conventional therapies. This principle of "complementary cytotoxicity" is

illustrated in Figure 1-3. The combined killing of two agents with complementary











cytotoxicity is potentially much greater than that of two agents acting on the same

cell population (Brown and Siim, 1996).

pH

Besides reduced oxygen level, pH is another microenvironmental

characteristic which impacts on the therapeutic outcome of solid tumors. Tumors

have been shown to have an acidic microenvironment compared to normal tissues

(Wike-Hooley et al., 1984). Increased capacity of glycolysis with the resultant

production of lactic acid, lactate production via the breakdown of glutamine, and

CO2 production as a result of cellular respiration may contribute to the shift to

acidic pH in malignant tissues.

The pH status of malignant tissue can also significantly influence drug

activity, especially that of compounds which are weak acids and bases. Studies

have demonstrated that selective decrease of the extracellular pH decreases the

uptake and activity of weak bases such as vinblastine, but increases the uptake

and cytotoxicity of weak acids such as chlorambucil (Gerweck and Seetharaman,

1996; Parkins et al., 1996). Moreover, pH-induced alterations in drug stability,

active transport processes, drug reactivity, activity of enzymes involved in

localized drug activation and the interaction of the drug with its molecular target

could all alter treatment efficacy.

Tumor Angiogenesis and Vasculature Targeting

Tumor Angiogenesis

Angiogenesis plays a significant role during normal growth, in physiological

conditions (e.g., in the placenta and endometrium), and in pathological conditions











such as inflammation, wound healing, and tumor growth. Angiogenesis is thus not

a specific phenomenon in tumors or a pathological condition, but instead an

integral element of numerous different normal and pathological conditions.

Angiogenesis is a complex multistep process involving extracellular matrix

remodelling, endothelial cell migration and proliferation, and capillary

differentiation and anastomosis, which are regulated by angiogenic peptides

(Blood and Zetter, 1990). The newly formed vessels are usually thin-walled

capillaries or sinusoids with little more than an endothelial lining, backed by a

basement membrane. Mobility and remodelling from pre-existing vasculature is

an important component of angiogenesis. Many angiogenic factors that stimulate

proliferation and migration of endothelial cells have been described (Pluda, 1997;

Teicher, 1995). Amongst the most potent and specific factors for vascular growth

is vascular endothelial cell growth factor (VEGF) (Claffey and Robinson, 1996;

Zhang et al., 1995). VEGF exists in several isoforms produced from one gene. It

binds to VEGF receptors of which there are two, VEGFR1 (fit-1) and VEGFR2

(flk-1/KDR). It is generally thought that VEGFR2 is the most important for

angiogenesis while VEGFR1 is expressed on macrophages and stimulates their

migration (Jain et al., 1996).

In normal adults, angiogenesis is limited to specific reproductive organs, and

the growth and turnover of vascular endothelial cells in most tissues is measured

in months and years (Hobson and Denekamp, 1984). Unlike what is found in most

normal tissues, vessels in tumors contain populations of actively dividing

endothelial cells in response to angiogenic factors. In human tumors the number










of dividing endothelial cells may be 50 times greater than in normal tissue (Harris,

1998). The vascularization of solid tumors is a prerequisite if a clinically relevant

size is to be reached. Without sufficient vascular supply, no solid tumor can grow

beyond a few cubic millimeters (Ausprunk and Folkman, 1977; Folkman, 1986).

Further tumor growth depends on nutrient supply via a network of microvessels

(Denekamp, 1993) which can be acquired, in part, by incorporation of existing

host blood vessels. However, it is now well established that the majority of tumor

blood vessels are newly formed as a result of angiogenesis triggered by the

release of stimulators such as VEGF (Siemeister et al., 1998). Thus,

neovascularization is a critical aspect of a tumor's growth and development.

Furthermore, angiogenesis is also essential for systemic metastasis, and recently it

has been shown to be essential for local invasion (Skobe et al., 1997).

Anti-angiogenesis Approach

The utter dependence of the tumor on its induced vessel formation for

growth, survival and spread has created a great deal of enthusiasm for developing

therapeutic approaches to specifically targeting the tumor vasculature (Folkman,

1995; Denekamp, 1993). A variety of approaches are under investigation. One

approach, and most extensively studied, involves attempts to prevent the

development of the vascular supply by inhibiting angiogenesis (Ingber et al.,

1990; O'Reilly et al., 1994; Scott and Harris, 1994). Three options have been

considered so far: (1) inhibition of the turnover from an avascular primary tumor

into a fully vascularized tumor; (2) slowdown of tumor progression by preventing

a tumor from becoming highly vascularized; (3) prevention of neovascularization










of distant metastases. Antiangiogenesis therapy targets a process that, under most

circumstances, is tumor specific and therefore likely to have few normal tissue

side effects.

The antiangiogenesis strategy includes agents which interfere with delivery

or export of angiogenic stimuli (Schweigerer, 1995), antibodies to

inhibit/inactivate angiogenic factors after their release (Mesiano et al., 1998),

antisense therapies (Im et al., 1999), drugs which inhibit receptor action (Witte et

al., 1998), and inhibitors of endothelial cell proliferation (Boehm et al., 1997).

Several of the agents have moved forward to the clinic including

metalloproteinase inhibitors, pentosan polysulphate and TNP-470 (Marshall and

Hawkins, 1995; Denis and Verweij, 1997; Twardowski and Gradishar, 1997).

Because of the major importance of VEGF as an angiogenic factor, numerous

strategies are presently being used to inhibit VEGF activity in tumors. These have

included antisense VEGF mRNA, monoclonal antibodies, and VEGF receptor

inhibitors (Dvorak et al., 1995; Kim et al., 1993).

It is now recognized that angiogenesis is regulated by a balance between pro-

angiogenic and anti-angiogenic factors and that loss of inhibitors may be an early

stage in tumor progression. Angiogenesis inhibitors include thrombospondin

(Taraboletti et al., 1997), several cytokines (IL-4, IL-12) and proteolytic

breakdown products of several proteins, including prolactin (Clapp and

Delaescalera, 1997), plasminogen (Cao et al., 1997) and collagen XVIII (O'Reilly

et al., 1994). It is these inhibitory peptides that have raised hopes that specific

inhibition of tumor angiogenesis may be possible with minimal toxicity and high











efficiency. These angiogenesis inhibitors, particularly angiostatin and endostatin,

are being actively investigated (O'Reilly et al., 1994; Boehm et al., 1997). They

are specific inhibitors of endothelial cell proliferation and have no obvious effect

on resting endothelial cells, nor on a variety of normal, transformed or neoplastic

cells. Studies have shown that systemic administration of endostatin to tumor-

bearing mice resulted in regression of tumors to a microscopic size (Boehm et al.,

1997). A dormant state, without evidence of toxicity, could be maintained for as

long as endostatin was administered. Whether these very encouraging results will

remain to hold true in the future, remains to be seen.

Vascular Targeting Approach

The concept of anti-angiogenic therapy relates to interfering with the

stimulating substances that cause new vessel formation. This should be important

in preventing establishment of small solid tumors or in preventing metastases. By

contrast, another approach, the so-called vascular targeting approach, focuses on

the use of agents that can directly destroy existing tumor vessels (Denekamp,

1990; Denekamp and Hill, 1991). Figure 1-4 illustrates the differences between

anti-angiogenic and vascular targeting approaches. The functioning vascular

network in tumors is pivotal for the survival of the tumor cells. This is confirmed

by the fact that artificial induction of ischaemia by clamping off the tumor-

feeding blood supply results in extensive tumor cell death and, if prolonged,

tumor cures (Denekamp, 1993; Chaplin and Horsman, 1994). Such studies

emphasize the therapeutic potential of strategies that target the tumor vasculature.

The antivascular approach aims to cause a rapid shutdown in the vascular function











of the tumor, leading to extensive secondary tumor cell death. Since thousands of

tumor cells are dependent on each tumor capillary for their metabolic

requirements, an agent which induced even limited damage to these vessels could

produce a cascade of tumor cell death (Denekamp, 1984; Denekamp, 1993).

Several features make tumor vasculature a suitable target in cancer therapy.

While tumor cells are genetically unstable, rapidly mutating, and able to develop

multidrug resistance, vascular endothelial cells are genetically stable and unlikely

to become drug resistant (Folkman et al., 1997). Drug delivery is not a problem,

since the target cells directly line the blood stream. Perhaps most importantly,

tumor vasculature represents an actively growing endothelium whereas the

endothelium in most normal tissues is essentially dormant (Denekamp et al.,

1993). This last feature may provide a key difference between the tumor and

normal tissue which can be exploited.

Development of antibodies to specific epitopes on the tumor vasculature

(Burrows and Thorpe, 1993; Huang et al., 1997) and vascular-targeted gene

therapy (Chaplin and Dougherty, 1999) are two approaches that are receiving

considerable attention. As mentioned previously, the advantages of targeting

endothelium include that it reduces delivery problems and the number of cells that

need to be targeted. These factors make the tumor vasculature a potentially ideal

target for antibody- and gene therapy-based approaches. For antibody-based

strategies, there is a need to identify antibodies that target unique determinants

which are selectively and constitutively expressed on the tumor endothelium.

Several studies are now underway to identify such antibodies. Potential











candidates include TEC-11, which recognized endoglin, and others that recognize

the N-terminal domain of VEGF, avP3 integrin and the receptor tyrosine kinase

Tie-1 (Brooks et al., 1994; Burrows et al., 1995). Gene therapy constitutes a

potentially powerful means of selectively targeting the tumor-associated vascular

endothelial cells, while at the same time minimizing the damage inflicted on

various normal tissues. Candidate genes are expected to either directly kill

vascular endothelial cells or sensitize them to the cytotoxic effects of ionizing

radiation and/or chemotherapeutic agents. The most promising ones are genes

encoding toxic protein inhibitors as well as genes that can convert relatively not-

toxic prodrugs to their biologically active metabolites (Parentesis et al., 1992;

Deonarain et al., 1995).

Besides antibody-based and gene therapy approaches, drug- and cytokine-

based approaches to vascular targeting are also possible. Several agents that elicit

irreversible vascular shutdown selectively within solid tumors have been

identified. These include flavenoids such as flavone acetic acid (FAA) and more

recently DMXAA (Zwi et al., 1994), and tubulin binding agents such as

colchicine and vinblastine. FAA was shown to have a broad spectrum of activity

against solid tumors (Corbett et al., 1986; Hill et al., 1995). The action of FAA

has been attributed in large part to its ability to induce the release of tumor

necrosis factor-alpha (TNF-a) from tumors in situ (Cliffe et al., 1994;

Mahadevanv et al., 1990). In contrast, the tubulin binding agent vinblastine

causes little or no increase in plasma TNF-a levels in tumor-bearing mice (Hill et

al., 1995). Nevertheless, antivascular effects are a common feature of tubulin










binding agents. Chaplin et al. (1996) assessed the effects of vinblastine and four

other tubulin binding agents (dolastatin 10, dolastatin 15, combretastatin Al and

combretastatin A4) on tumor blood flow. It was shown that all five agents induced

a reduction in tumor blood flow range from 50% to 90%. The mechanisms

governing these tumor-selective effects of tubulin binding agents are largely

unknown, but it is possible that the inhibition of tubulin polymerization affects

endothelial cell shape, leading to thrombus formation or changes in permeability

of the endothelium.

Despite the reported antitumor effects, the clinical potential of vascular

targeting strategies ultimately will be largely determined by the selective toxicity

of the reagents. Unfortunately, to date, most of these agents have been reported to

only elicit antivascular effects at doses approaching the maximum tolerated dose

(MTD) and only in the presence of significant morbidity (Chaplin et al., 1996).

For example, vinblastine and colchicine markedly reduced tumor perfusion and

caused necrosis of tumor tissue only when the injected dose was increased to

lethal range (Nihei et al., 1999). To fully appreciate the anti-vascular strategy,

new agents with a large therapeutic window and improved selectivity are needed.

Combretastatin A-4 Disodium Phosphate

The tubulin and actin cytoskeleton are critical mediators for a number of

important endothelial cellular functions other than the mitotic spindle and

chromosome segregation. They facilitate intracellular organization, cell

morphology, cell motility and the intracellular transport of molecules from the site

of synthesis to the cell surface via microtubule motor proteins (Avile, 1992). The











cytoskeleton is a dynamic structure and conformational rearrangements occur in

response to the endothelial cells' environment and exposure to mechanical forces

(Cucina et al., 1995). Because of the pivotal role the cytoskeleton plays with

respect to cell shape in endothelial cells, it is not surprising they have effects on

overall vascular function.

One class of tubulin-binding compounds which has received attention in

recent years is the combretastatins. The African bush willow tree Combretum

Caffrum is the source of 17 natural combretastatins and a further 22 similarly

structured agents have been synthesized, thus making four series of compounds

(named A to D) (Pettit et al., 1987; O'Brien, 1997). Structurally, combretastatins

consist of two substituted benzene rings linked by a saturated, hydroxy-

substituted 2-carbon bridge (Figure 1-5). These combretastatins show structural

similarity to colchicine and are competitive inhibitors of the binding of colchicine

to tubulin (Pettit et al., 1989; Sackett, 1993). They inhibit microtubule activity

and interfere with cell growth and proliferation (Pettit et al., 1989). The

mechanism of their binding to tubulin was examined indirectly for one of them,

combretastatin A-4, by evaluating their effects on the binding ofradiolabeled

colchicine to the protein (Lin et al., 1989). Studies showed rapid binding of

combretastatin A-4 to tubulin even at 0 degrees (binding was complete at the

earliest times examined), in contrast to the relatively slow and temperature-

dependent binding of colchicine. It demonstrated that the effectiveness of

combretastatin A-4 as antimitotic agents appears to derive primarily from the

rapidity of their binding to tubulin.










Combretastatin A-4 showed some concentration-dependent cytotoxicity

against a variety of human tumors (El-Zayat et al., 1993). As combretastatin A-4

itself is poorly soluble in water, a prodrug, combretastatin A-4 disodium

phosphate (CA4DP) was prepared. This extra phosphate group can readily be

cleaved by endogenous nonspecific phosphatases (O'Brien, 1997).

Combretastatin A-4 binds to plasma protein, and this seems to reduce its activity

(Tozer et al., 1999). Therefore, there may be an advantage for using CA4DP

rather than combretastatin A-4 beyond its increased solubility. Figure 1-5 shows

the chemical structures of combretastatin A-4 and CA4DP. Both agents have been

shown to lead to rapid vascular shutdown in several preclinical tumor models

(Dark et al., 1997; Li et al., 1998; Horsman et al., 1998). Recent studies

continued to demonstrate the vascular effect of CA4DP. When assessed in a rat

system, 100 mg/kg CA4DP caused a very large decrease in tumor blood flow,

which by 6 hr, was reduced approximately 100-fold (Tozer et al., 1999).

Calculation of vascular resistance revealed some vascular changes in the heart and

kidney for which there were no significant changes in blood flow. Using magnetic

resonance imaging (MRI), Beauregard et al. (1998) showed that tumor perfusion

decreased significantly in the central region of murine tumors after CA4DP

treatment and it was consistent with the haemorrhage seen in histological sections.

Further preclinical studies have demonstrated that avascular nodules do not

appear to be responsive to this agent, providing additional evidence for a vascular

mechanism of action (Grosios et al., 1999). The rapid, selective and extensive











damage caused to the tumors by CA4DP has highlighted the potential of the agent

as a novel cancer chemotherapeutic agent. More importantly, CA4DP typically

produced these effects at concentrations less than one-tenth of the MTD thus

offering a wide therapeutic window (Dark et al., 1997; Li et al., 1998; Horsman et

al., 1998).

Combretastatin is now in Phase I clinical trials in UK and US. The current

studies are very encouraging (Randal, 2000). A patient with an anaplastic thyroid

tumor, which was unaffected by standard therapies and progressed relentlessly,

was treated with intravenous infusion of combretastatin. Three weeks after the

first infusion, the tumor began to shrink. After six more infusions, CT and MRI

failed to show any trace of the tumor. Later exploratory surgery confirmed the

tumor's total disappearance. There also have been other patients whose previously

very aggressive cancers some of them metastatic stabilized after treatment

with combretastatin for 24 weeks or more (Randal, 2000).

In our laboratory, we have used several animal models to study the antitumor

efficacy of CA4DP. Treatment with CA4DP has been shown to produce extensive

regions of hemorrhagic necrosis in both murine tumors (Li et al., 1998) and

human tumor xenografts. Moreover, as the application of antivascular strategies

will need to be given in conjunction with conventional anticancer therapies, we

also examined the efficacy of combining CA4DP with ionizing radiation. Such

treatment may be required to destroy the remaining rim of tumor cells surviving at

the periphery near normal tissue vessels. Results from our laboratory indicate










superior antitumor efficacy in the combination of CA4DP and radiation (Figures

1-6 and 1-7).

Significance

It has been well established that the vascularization of solid tumors is a

prerequisite if a clinically relevant size is to be reached. The dependence of the

tumor on its induced vessel formation has created a great deal of enthusiasm in

specifically targeting the microcirculation in cancer therapy. The central goal of

this project was to investigate the potential therapeutic utility of vascular targeting

drug CA4DP. Experiments proposed in this project were directed at gaining a

better understanding of the mechanism of its action on tumor and endothelial

cells, and to assess its efficacy in human tumor models either alone or in

combination with conventional anticancer therapies. The thesis focused on a

model of Kaposi's Sarcoma (KS), since AIDS-KS is a fulminant disease that

usually requires aggressive treatment, especially when it involves visceral organs,

which lacks effective therapies. Irradiation, systemic chemotherapy, and

interferon (IFN-a), though helpful, are administered primarily for symptomatic

relief and to prevent disease progression. Conventional treatments do not prolong

survival, and their clinical effectiveness is not satisfying (Sung et al., 1997).

Continued pursuit of more effective agents clearly is needed. Given the possible

future clinical impact of CA4DP, we believe it to be highly worthwhile to

investigate effects of this agent in a preclinical model of KS.

The specific aims of this project are as follows:






18



Specific aim 1. To study the different responses of normal and neoplastic cells

to CA4DP by examining the effects of CA4DP on cell cycle, cell survival, and

apoptosis.

Specific aim 2. To gain further insight into the mechanisms underlying the

antivascular action of CA4DP by examining the effects of CA4DP on human

microvascular endothelial cells (HMVEC-L).

Specific aim 3. To examine the antitumor efficacy ofCA4DP by applying

histological approaches and by measuring its cytotoxic action in KS xenografts.

Specific aim 4. To assess the potential in situ therapeutic benefit of combining

CA4DP with conventional anticancer therapies by evaluating the effects of such

treatments in KS xenografts.

































AV Shunt


Figure 1-1. Diagram showing the principal differences between the vasculature of
normal and malignant tissue. Whereas normal tissues have relatively uniform and
well-ordered blood vessels that are sufficiently close together to oxygenate all of
the tissue, blood vessels in tumors are tortuous, have incomplete vessel walls,
have sluggish and irregular blood flow, and have regions ofhypoxia between the
vessels (Brown and Giaccia, 1998).


I 3Ze3iB























c Mechanism:

LL 1 DNA-H(Chemical Restitution)

\ DNA-DN Cell Death

301 *
*.01 DNA-OO-* DNA-OOH
SAerobic Hypoxic (Damage Fixation)


00) 10 20 30 40
Dose (Gy)




Figure 1-2. Typical survival curves to ionizing radiation for mammlian cells under
aerobic and hypoxic conditions. Most mammalian cells, irrespective of genetic
background, exhibit a survival curve with an Initial "shoulder" region followed by
exponential cell killing. The oxygen enhancement ratio is typically 2.5-3.0. The
dotted vertical line at 14 Gy shows the >2 logs difference in cell kill for aerobic
and hypoxic cells at this dose. Also shown is the mechanism for the greater
sensitivity of aerobic cells as compared to hypoxic cells, ionizing radiation
produces a radical in DNA, which can be either chemically restituted by donation
of hydrogen from nonprotein sulfhydryle (-SH) in the cell or, in the presence of
oxygen, converted into permanent damage that increases the probability of cell
death (Brown, 1999).

























U-
c Radiation / Chem. Drug

.. Combined"



0 50 100 1i

Distance from Capillary (pm)







Figure 1-3. Left, a diagrammatical representation of part of a tumor cord
surrounding a capillary showing decreasing oxygen concentration as well as
decreasing cellular proliferation and drug concentration as a function of distance
from the capillary. Right, the considerations on the left lead to the prediction that
cell killing by radiation or most anticancer drugs will be reduced as a function of
distance from the capillary. The combination of standard treatment with a
hypoxic cytotoxin would be expected to overcome the problem ofhypoxic cells
by producing a relatively uniform cell profile of cell killing as a function of
distance from the capillary (Brown and Siim, 1996).



















































Figure 1-4. Diagram to illustrate the differences between the concepts and the
likely outcome of anti-angiogenic strategies and those designed to produce
ischaemic or haemorrhagic necrosis by vascular targeting (Denekamp, 1993).











a.


CH30
CH30
OH

CH30


b.

CH30O

CHO /
CH3 O I _I
CH 0- ONa

CH3O ONa


Figure 1-5. The chemical structures of combretastatin A-4 (a) and combretastatin
A-4 disodium phosphate (CA4DP) (b)
















10 -
O Combr
0 Radiation + Combr





1
10*1


g .." y////////////////s/////////5
S10
E






10-5 --------------
10,
0 20 40 60 80 100

Combretastatin A-4 prodrug dose (mg/kg)





Figure 1-6. Tumor cell killing in KHT sarcomas treated with increasing doses of
CA4DP either alone of 1 hr after irradiating the tumors with a 15-Gy dose of
radiation. Data are the mean SE of 6-12 tumors.














0 Radiation
* Radiation + Combr






\


\


B





T\
12


0 5 10 15 20 25
Radiation dose (Gy)


Figure 1-7. Tumor cell survival in KHT sarcomas treated with a 100 mg/kg dose
of CA4DP 1 hr after a range of doses of radiation. Results are the mean SE of 3
experiments.















CHAPTER 2
COMPARISON OF THE DIFFERENT RESPONSES OF NEOPLASTIC AND
NORMAL CELLS TO CA4DP IN VITRO


Introduction


The survival and growth of solid tumor deposits depends critically on the

development of a blood vessel network. The functioning vascular network in

tumors provides the tumor cells with oxygen and nutrients, and enables removal

of the toxic waste products of cellular metabolism. The fact that the production of

several angiogenic growth factors can be up regulated by physiological

parameters, including low oxygen or glucose and acidic pH, which are associated

with vascular insufficiency, provides a logical rationale for the strong angiogenic

stimulus in malignant tissue (Chiarotto and Hill, 1999; Namiki et al., 1995). The

continued proliferation of tumor cells will result in deprivation of oxygen and

glucose and production of acidic metabolites, thus stimulating the development of

additional neovasculature (Siemeister et al., 1998). The new vessels facilitate the

further expansion of the tumor cell mass providing a perpetual loop. Clearly, the

cycle can be interrupted by killing or inhibiting the growth of the tumor cells.

However, interventions that compromise the function or growth of the tumor

neovasculature can also be effective. Therefore, tumor blood vessels represent a

central target for the development of new approaches to cancer therapy.










The majority of research work in this area has been focused on preventing

the growth of new tumor vessels, so called anti-angiogenesis. Many agents that

were identified as anti-angiogenic target at least one of the several stages involved

in new vessel formation, i.e. basement membrane degradation, endothelial cell

migration, endothelial cell proliferation and tube formation (Fan et al., 1995). In

contrast to the focus on anti-angiogenic approaches to therapy, there has, until

recently, been relatively little effort afforded to the identification and

development of therapies that specifically compromise the function of the existing

neovasculature in solid tumors.

Interestingly, evidence for the therapeutic potential of vascular targeting

approaches existed a lot earlier than recent studies. It had been reported over 150

years ago that occasionally solid tumors in the clinic could be eradicated when

their circulation was interrupted either by torsion of the vascular pedicle or by

thrombosis of a major feeding vessel (Walsh, 1844). The pivotal role of tumor

vasculature and the effects of its selective destruction were also highlighted by

Woglum over 75 years ago (Woglum, 1923). However, it is only within the last

decade that research has been focused seriously on the development of therapies

that specifically target and damage tumor neovasculature. Despite this limited

development time, many promising approaches, including drug-, antibody- and

gene therapy-based strategies have emerged.

Many tubulin-binding agents have been shown to have antivascular effects.

The vinca alkaloids, for example vincristine and vinblastine, could induce

vascular damage at doses close to the MTD (Baguley et al., 1991; Hill et al.,











1995). However, they also have shown direct cytotoxic effects against a variety of

tumor cells (Zhou and Rahmani, 1992). Although several studies have shown the

antivascular effects of CA4DP in several clinical models (Dark et al., 1997; Li et

al., 1998; Horsman et al., 1998), it is not clear whether the activity of CA4DP is

selectively against vascular endothelial cells or whether it also acts on tumor cells.

In order to shed light on this, in the present studies, we examined the responses of

human tumor cells (KSY-1 and A549), human endothelial cells (HMVEC-L), and

human fibroblasts (FG1522) to CA4DP in vitro.

The tubulin and actin cytoskeleton are critical mediators for mitotic spindle

formation and chromosome segregation. Because CA4DP is a tubulin-binding

agent, studies were initiated by examining its effect on cell cycle distribution in

different cell types. Later we examined the cytotoxicity of CA4DP and whether it

could induce apoptosis in both neoplastic and normal cells.

Material and Methods

Cell Culture

Two human tumor cell lines, Kaposi's Sarcoma cell line KSY-I (ATCC,

Rockville, MD) and lung cancer cell line A549 (ATCC, Rockville, MD) were

used in the study. KSY-1 cells were cultured in positively charged Cell+ tissue

culture flasks from SARSTEDT (Newton, NC). Cell TC flasks provide a

positively charged surface for difficult-to-grow adherent cell cultures. Both KSY-

1 and A549 cells were grown in RPMI 1640 medium (Gibco BRL, Grand Island,

NY) with 100 units/ml penicillin, 0.1 mg/ml streptomycin, and 10% fetal bovine

serum (FBS).










Human microvascular endothelial cells of the lung (HMVEC-L), obtained

from Clonetics (San Diego, CA), were grown in EGM-2-MV medium (Clonetics)

containing 5% FBS and supplements (0.1% hEGF, 0.4% hFGF-B, 0.1% VEGF,

0.1% Ascrobic Acid, 0.04% Hydrocortisone, 0.1% Long R3-IGF-1, 0.1% Heparin

and 0.1% GA-1000). Clonetics Trypsin and Trypsin Neutralization Solution

(TNS) were used for subculture.

Human skin fibroblasts FG1522 (from Dr. Hei's lab, Columbia University)

were grown in Dulbecco's MEM Medium with 10% FBS and 25 pg/ml dose of

gentamycin.

Drug Preparation

CA4DP (Oxigene Inc., Lund, Sweden) was dissolved in 5% sodium

carbonate at a concentration of 10 mM and then subsequently diluted in 0.9%

saline and culture medium immediately before use.

Cell Cycle Studies

A549, KSY-1, HMVEC-L, and FG1522 cells were plated in 60 mm dishes at

2x105 cells/dish. On day 3, the cells were exposed to various concentrations of

CA4DP for a period of 2 hr. The dishes were then washed with PBS and

replenished with fresh media. 22 hr later, the cells were trypsinized, counted, and

fixed in 50% ethanol overnight. Before analyzed by FACS, the cells were treated

with 1 mg/ml RNase (in PBS) for 30 min. The samples were then washed with

PBS twice and resuspended in 25 mg/ml propidium iodide (PI) (in PBS) at a

volume of 1 x106 cells/ml. The cells were stained with PI in darkness for at least

10 min and were then analyzed by FACS for cell cycle distribution on a Becton











Dickinson flow cytometer made available through the University Core Facility for

Flow Cytometry at the University of Florida.

Cell Viability Studies

Cell viability was determined using clonogenic cell survival assay. Briefly,

A549, KSY-1, HMVEC-L, and FG1522 cells were plated in 60 mm petri dishes at

Ix105 cells/dish. On day 3, the cells were exposed to various concentrations of

CA4DP for a period of 2 hr or 24 hr. The dishes were then washed with PBS

twice and the cells were trypsinized and counted. The cells were then mixed with

104 lethally irradiated cells and plated into 60 mm petri dishes. After 2 weeks of

incubation at 370C, colonies of 50 or more cells were counted with the aid of a

dissecting microscope. Cell surviving fractions were calculated as the ratio of

colonies counted in treated versus untreated group. To assess the effect against

quiescent cells, the cells were first grown in 60 mm petri dishes and once

confluent were treated and assayed as described above.

Apoptosis

Four cell lines each grown in 2-well chamber slides were treated with 50 1M

CA4DP for 2 hr. The treated cells were fixed in 4% formaldehyde solution

immediately after treatment or 22 hr later for TdT-mediated dUTP Nick-End

labeling (TUNEL) assay. Basically, the cells were permeabilized in 0.2% Triton

X-100 solution for 5 min. DNA strand breaks were then labeled with fluorescein-

12-dUTP in TdT incubation buffer at 370C for 1 hr. The samples were then

counterstained with 1 pg/ml diamidine-2-phenylindole (DAPI), which binds to

the A-T-rich regions of DNA. Localized green fluorescence of apoptotic cells











(fluorescein-12-dUTP) in a blue background (DAPI) was detected by

fluorescence microscopy. The percentage of apoptotic cell was obtained by

dividing the number of cells with green fluorescence by the total number of cells

with blue fluorescence. A minimum of 300 cells were counted for each condition.

Results

To evaluate the effect of CA4DP on cell cycle distribution, proliferating

normal and neoplastic cells were exposed to a range of doses of CA4DP for 2hr,

and DNA profiles were analyzed by FACS 22 hr later. Results showed that

CA4DP could effectively block all cell types in the G2/M phase beginning at the

doses -0.05-0.1 pLM (Figures 2-1 and 2-2). Increasing the drug dose to higher

concentrations caused either a slight or no increase in the percentage of cells in

the G2/M phase in different cell types (Figure 2-2). It was also observed that, even

at high concentrations, CA4DP arrested only 30-40% HMVEC-L and FG1522

cells in the G2/M phase, whereas in A549 and KSY-1 cells there were almost

complete G2/M blocks.

Clonogenic cell survival assay was used to evaluate cell viability in each cell

line treated with a range of doses of CA4DP. A 2-hr treatment, administered to

exponentially growing cell populations, was found to be ineffective against

FG1522 and A549 cells and showed only slight killing of KSY-1 cells (Figure 2-

3; a-c). In contrast, under those conditions, CA4DP displayed significant dose-

dependent activity against proliferating HMVEC-L (Figure 2-3d). Importantly

this effect was specific to proliferating endothelial cells: quiescent HMVEC-L

cells showed no response to CA4DP even at high doses (Figure 2-3d). With











prolonged drug exposure (24hr), CA4DP caused a clear dose-dependent cell kill

in both A549 and KSY-1 cells, but still showed no toxicity to FG1522 cells.

(Given the exquisite sensitivity of proliferating HMVEC-L to CA4DP, extended

drug exposure times were not evaluated in this cell type.)

To gain further insights into the cytotoxic mechanisms of CA4DP, we also

used TUNEL assay to examine the induction of apoptosis in the different cell

types treated with this agent (Figure 2-4). The results showed that 2-hr treatment

with 50 [iM CA4DP assessed immediately or 22 hr later led to very little

apoptosis in fibroblasts compared to the control group (0.6 and 1.2%,

respectively). A549 cells showed a similar response; there was no induction of

apoptosis immediately after CA4DP treatment, and 22 hr later, 4.6% of cells were

detected as apoptotic cells. CA4DP induced more apoptosis in KSY-1 cells than

in A549 cells. In this cell line the 0.7% apoptotic index in untreated KSY-1 cells

rose to 2.8% immediately after treatment and to 7.9% 22 hr later. The results also

showed that CA4DP induced the highest level of apoptosis in HMVEC-L cells.

Immediately after a 2-hr CA4DP exposure 7.5% of cells underwent apoptosis.

This number increased to almost 20% 22 hr later.

Discussion

The main aim of this study was to investigate whether CA4DP demonstrated

any cell type specificity/selectivity by comparing its effects in normal and

neoplastic cell types. Initial studies with FACS analysis showed that non-toxic

concentrations of CA4DP caused arrests of the cells in the G2/M phase in all the

cells examined, indicating a disruption of mitosis due to a functional deficiency of










the tubulin apparatus, which disables the cells to divide their chromosomes

properly. More importantly, the doses for CA4DP to initiate the cell cycle effect

in different cell types were found to be within the same range (-0.05-0.1 pLM).

This suggested that CA4DP could bind to the tubulin in different types of cells

with the same efficacy, thus resulting in cellular tubulin disruption and similar G2-

arrest effects at the same drug concentration. It was also observed that while

CA4DP could cause almost a complete G2-arrest in A549 and KSY-I cells, only

30-40% of HMVEC-L and FG1522 cells were blocked in the G2/M phase even at

high CA4DP concentrations. We believe that this arises because unlike the tumor

cell lines where the majority of cells are actively proliferating, a significant

proportion of the normal cells may never enter the cell cycle. To attempt to

examine this possibility we are currently utilizing the antibody against Ki67, a

nuclear antigen for proliferating cells, and BUdR labeling to determine the

fractions of proliferating cells in these cell populations.

Although CA4DP exerted similar effectiveness in terms of cell cycle arrest

in both neoplastic and normal cells, results from the clonogenic cell survival assay

showed significantly difference in the responses of the different cell types to this

agent. CA4DP showed dose-dependent activity against proliferating endothelial

cells with short drug exposure (2 hr). For example, a 20 pM drug concentration

reduced viable HMVEC-L cells to less than 10% compared to the untreated cells.

If the endothelial cells were quiescent during the drug exposure, no significant

drug toxicity was observed. In contrast to the endothelial cells, 2-hr exposure to

CA4DP caused no change in cell viability in proliferating FG1522 and A549










cells, and only a slight toxic effect in KSY-1 cells. These observations implied a

selective toxicity of CA4DP toward proliferating endothelial cells. With

prolonged drug exposure (24 hr), CA4DP did show toxicity against tumor cells

(both A549 and KSY-I cells). But over the dose range evaluated FG1522 cells

were still not affected. We believe that the specific killing of dividing endothelial

cells observed after CA4DP treatment might be critical for the in vivo action of

this drug. Parenthetically the observation that normal cells may be less susceptible

to CA4DP treatment may explain the low level of toxicity of CA4DP observed in

the preclinical studies (Dark et al., 1997).

Apoptosis is an active and gene-directed mode of cell death, involved in

embryological development, organ involution, and the response of both normal

and transformed cells to cytotoxic agents (Gorczyca et al., 1993). It is

characterized by rapid nuclear and cytoplasmic condensation and cellular

disintegration into apoptotic bodies (Kerr et al., 1972). Recent work has shown

that apoptosis is controlled by a complex network of positive and negative

signals, which originate either from specific gene products or from the

extracellular environment (Stewart, 1994). The present studies showed that 50

pM CA4DP induced the highest level of apoptosis in endothelial cells compared

to tumor cells and fibroblasts, demonstrating again the selective toxicity of

CA4DP against endothelial cells.

It was observed that CA4DP caused cell cycle arrest in HMVEC-L starting

at very low dose (0.05 pM). However, CA4DP started to show toxicity against

HMVEC-L only at doses higher than 1 pM. The different effects of CA4DP










observed at low and high doses is because at low doses, CA4DP-induced G2-

arrest in the endothelial cells was reversible. Arrested cells recovered at 48-72 hr

after 2-hr CA4DP treatment (data not shown). Only at higher doses (> 1 M),

CA4DP caused irreversible G2-arrest in HMVEC-L, thus leading to subsequent

clonogenic cell death and cell apoptosis.

Taken together, the present results demonstrated that CA4DP displayed a

significantly higher cytotoxicity toward proliferating endothelial cells than to

tumor cells and normal human fibroblasts. This may well explain the selective

antivascular effects of CA4DP observed in preclinical studies in vivo. Still the

fundamental question of "Why CA4DP has such selectivity toward endothelial

cells?" remains largely unknown. Nevertheless the selectivity of the effects are

highly encouraging and clearly warrant more detailed study of the cellular and

molecular mechanisms involved as well as the continued investigation of the

therapeutic potential of the drug as an antivascular agent.















30-G! 61 96%
3PMiase 2839%
i2-M g 65%6


L


ChanneLs











C.anri o'


Figure 2-1. Analysis of cells cycle distributions in A549 cells by FACS. a)
Untreated A549 cells; b) A549 cells treated with 0.1 iM CA4DP for 2 hr and
fixed for FACS analysis 22 hr later.







































CA4DP cell cycle effect


O KSY-1
O HMVEC-L

* FG1522


0.01 0.1 1 10 100

Drug dose (piM)


Figure 2-2. G2-arrest caused by CA4DP treatment in FG1522, A549, KSY-1, and
HMVEC-L cells. The cells were treated with a range of doses of CA4DP for 2 hr
and fixed for FACS analysis 22 hr later.


100

90
80-
70
60-

50-
40

30
20

10-
0
0.001





















KSY-1 -CA4DP


0.01 0.01 0.1 1 10 1( 0
Drug Dose (pM)


FG1522 CA4OP


o. (s )


A549 CA4DP

-I-^- t -. --.


"\


Drg Dose (pM)




HMVC-L-CADP















os,
o eo 75 10 12 tIO
Dose (pM)


Figure 2-3. The cytotoxicity profile of CA4DP against KSY-1 (a), A549 (b),
FG1522 (c), and HMVEC-L cells (d). The cells were exposed to CA4DP for 2 hr
(a-d) or 24 hr (a-c) and the cell killing effects were assessed using clonogenic cell
survival assays. Data are the mean + SE of three experiments.





















Apoptosis 50 IM CA4DP


D Control
0 2 hr treatment
01 2 hr treatment+ 22 recovery


Fibroblast A549 KSY-1 HMVEC-L


Figure 2-4. Effect of CA4DP on cell apoptosis in fibroblasts FG1522, A549,
KSY-1, and HMVEC-L cells. The cells were treated with 50 iM CA4DP for 2 hr
and stained for TUNEL assay either immediately after treatment or 22 hr later.
Data are the mean + SE of three independent experiments.















CHAPTER 3
EFFECTS OF CA4DP ON HUMAN MICROVASCULAR ENDOTHELIAL
CELLS IN VITRO


Introduction

More than 1012 endothelial cells line the inside of blood vessels, covering a

surface area of more than 1000 m2 (Jaffe, 1987). Not only do they form the

structural basis of blood vessels and provide an antithrombogenic surface, but

they also contribute to numerous metabolic functions including coagulation and

thrombolysis, control of vasotonus and antigen presentation, as well as basement

membrane and growth factor synthesis (Pearson, 1991). Endothelial cells in the

adult form a highly heterogenous cell population that varies in different organs.

These cells normally are quiescent. After induction of angiogenesis by angiogenic

cytokines, however, endothelial cells can proliferate as rapidly as bone marrow

cells, which have a turnover time of 5 days (Folkman, 1995). Angiogenesis is a

mode of endothelial cell activation that induces distinct phenotypic changes.

Triggered by paracrine and autocrine mechanisms, it enables endothelial cells to

break away from preexisting vessels to enter a complex morphogenetic cascade

that will ultimately lead to the formation of new vessels with mature endothelial

cells. Endothelial cells released in culture from growth arrest and allowed to

migrate change their adhesive properties, their surface glycosylation pattern, their

cytokine production, and their growth factor receptor expression pattern, as well










as their proteolytic balance (Augustin-Voss et al., 1992; Weich et al., 1991).

Consequently, since angiogenesis is a developmentally regulated process that is

down-regulated in the healthy adult (except for the female reproductive system),

inhibition of angiogenic-specific cell functions might be useful for targeting new

vessels during tumor growth.

One key in the development of treatment strategies is to identify differences

that exist between the tumor vasculature and normal tissues. As mentioned before,

the blood vessels in tumors are proliferating more rapidly than those in normal

tissues (Denekamp, 1990). Thus, targeting features of proliferating endothelium,

or even newly formed vasculature could achieve some selectivity. Another well-

established feature of tumor blood vessels is that, unlike those in normal tissues,

they can be subjected to low oxygen tension (Chaplin et al., 1987; Hill et al.,

1996). A third obvious feature is that tumor endothelium is located adjacent to the

malignant tumor cells, which in turn can alter endothelial cell characteristics.

Exploiting the changes that such microenvironmental stimuli induce both in

endothelial cell function and gene expression will undoubtedly provide the key to

achieving effective and highly selective approach to targeting endothelial cells in

tumors.

Vascular targeting approach aims to destroy the tumor vessels which contain

rapidly proliferating endothelial cells. Antivascular effect is a common feature of

tubulin binding agents. The original interest in the vascular-damaging effects of

such agents was stimulated by studies with colchicine reported in the 1930s and

1940s. Studies clearly demonstrated that colchicine preferentially damaged newly










formed capillaries in tumors with the consequence of inducing hemorrhage and

extensive necrosis (Ludford, 1945). Activity was noted in many different

experimental tumor systems, but significant effects were only achieved at doses

approaching the MTD. Later studies indicate that other tubulin binding agents,

such as vincristine and vinblastine, at doses approximating the MTD could also

induce vascular damage (Baguley et al., 1991; Hill et al., 1995). Recent studies

focused on evaluating a number of inhibitors of tubulin polymerization for their

ability to induce vascular damage in tumors (Chaplin et al., 1996), aiming to

identify agents with a superior therapeutic index for their vascular effects.

Fortunately, combretastatin A-4 and CA4DP have been identified as agents that

can induce vascular damage in tumors at doses much less than the MTD (Dark et

al., 1997; Beauregard et al., 1998).

As with all approaches to cancer therapy, vascular targeting is only realistic

if significant selectivity between tumor and normal tissue response can be

achieved. In vitro studies from Chapter 2 revealed marked cytotoxic effects of

CA4DP against proliferating but not quiescent endothelial cells. A549, KSY-1,

and FG1522 cells were much less affected by CA4DP than the HMVEC-L cells.

Why proliferating endothelial cells show such sensitivity to the in vitro effects of

CA4DP is not known yet. Although the clinical potential of CA4DP has been

recognized by its recent Phase I clinical studies in UK and US, further

information is required regarding its mechanism of action. In light of the potent

antivascular effects of CA4DP the present studies were undertaken to gain insight

into the mechanisms) of action of CA4DP by studying its activity in HMVEC-L.












Material and Methods

Cell Culture

HMVEC-L cells, obtained from Clonetics (San Diego, CA), were grown in

EGM-2-MV medium (Clonetics) containing 5% FBS and supplements (0.1%

hEGF, 0.4% hFGF-B, 0.1% VEGF, 0.1% Ascrobic Acid, 0.04% Hydrocortisone,

0.1% Long R3-IGF-1, 0.1% Heparin and 0.1% GA-1000). Clonetics Trypsin and

Trypsin Neutralization Solution (TNS) were used for subculture.

Drug Preparation

CA4DP (Oxigene Inc., Lund, Sweden) was dissolved in 5% sodium

carbonate at a concentration of 10 mM and then subsequently diluted in 0.9%

saline and culture medium immediately before use.

Indirect Immunofluorescence

The intracellular distribution of microtubules following drug treatment was

determined using indirect immunofluorescent staining (Giannakakou et al., 1998;

Woods et al., 1995). HMVEC-L cells were plated in 35 mm dishes 1 day prior to

treatment with CA4DP. Both treated and untreated cells then were fixed in 1:1

methanol/acetone for 5 min at room temperature and washed with PBS.

Incubation with the primary anti-p-tubulin MAb for 70 min was followed by a 50-

min incubation with the secondary flourescein-conjugated goat anti-mouse IgG

antibody. All antibody incubations and washes were performed at room

temperature. Morphological analysis then was performed by fluorescence

microscopy.










Tubulin Polymerization Assay

HMVEC-L cells grown in 24-well plates were treated with 5 pM CA4DP

for a specified time. After washing each well twice with 1 ml PBS (Ca2+ free), the

cells were lysed at 370C for 5 min in the dark with 100 pl ofhypotonic buffer (20

mM Tris-HC1, pH 6.8, 1 mM MgC12, 2 mM EGTA, 0.5% Nonidet P-40, 2 mM

PMSF, 200 U/ml Approtinin, 100 pg/ml soybean trypsin inhibitor, 5mM E-amino

caproic acid and 1 mM benzamidine) (Giannakakou et al., 1998). The wells were

scraped and the lysates transferred to 1.5 ml Eppendorf tubes. Each well was

rinsed with an additional 100 pl of the hypotonic buffer, and this volume was

pooled with the lysate. Following a brief but vigorous vortex the samples were

centrifuged at 14,000 rpm for 10 min at room temperature. The 200 Pl

supernatants containing soluble or unpolymerized (cytosolic) tubulin were

carefully separated from pellets and transferred to separate tubes. The pellets were

resuspended in 200 pl of hypotonic buffer containing 10 mM Tris, pH 7.5, 1.5

mM MgC12, 10 mM KC1, 0.5% Nonidet P-40 and the protease inhibitors

described above. Each tube containing either the soluble or the polymerized

fraction was mixed with 70 pl of 4x SDS-PAGE sample buffer (0.3 M Tris-HCl,

pH 6.8, 45% glycerol, 20% P-mercaptoethanol, 9.2% SDS and 0.04 g/100 ml

bromophenol blue) and heated at 950C for 5-10 min; 20 pl aliquots of each

sample then were analyzed by SDS-PAGE on a 12% resolving gel and 3%

stacking gel. Following immunoblotting using a primary anti-p-tubulin MAb, the

signal was quantitated by densitometry.










Cell Migration Assay

Confluent cultures of HMVEC-L cells were prepared in 24-well plates. A

scrape wound of uniform width (2 mm) was produced in the monolayers prior to

treatment with CA4DP or drug vehicle. CA4DP exposure was for a period of 2hr

at concentrations of 0.1, 1, and 10 iM. The drug was removed and 24 and 48 hr

later each well was stained with 300 tg/ml neutral red solution for 30 min to help

visualize and localize the cells. The number of cells entering the denuded area

was counted using a phase microscope (Braunhut et al., 1996).

Apoptosis

HMVEC-L cells grown in 2-well chamber slides were treated with CA4DP

at 1-50 [M for 2 hr. After a specified time, the treated cells were fixed in 4%

formaldehyde solution for TUNEL assay. Basically, the cells were permeabilized

in 0.2% Triton X-100 solution for 5 min. DNA strand breaks were then labeled

with fluorescein-12-dUTP in TdT incubation buffer at 370C for 1 hr. The samples

were then counterstained with I jig/ml DAPI, which binds to the A-T-rich regions

of DNA. Localized green fluorescence of apoptotic cells (fluorescein-12-dUTP)

in a blue background (DAPI) was detected by fluorescence microscopy. The

percentage of apoptotic cell was obtained by dividing the number of cells with

green fluorescence by the total number of cells with blue fluorescence. A

minimum of 300 cells were counted for each condition.











Cell Detachment

HMVEC-1 cells were plated in 60 mm petri dishes, and exposed to 0.1-50

pM CA4DP for 2 hr while in logarithmic growth or plateau phase. Cell

detachment rates were determined 22 hr after drug treatment by counting the

detached cells in the media electronically using a Coulter counter.

Results

Initial studies examined the cellular morphology and tubulin organization of

CA4DP-treated cells by immunofluorescence (Figure 3-1). Untreated HMVEC-L

cells demonstrated a well-organized tubulin network with individual microtubule

fibers clearly visible (Figure 3-la). Treatment with 10 gM CA4DP for a 4 hr

period led to a disrupted network of microtubules, which appears as a diffuse

staining pattern in the treated cells (Figure 3-lb). When HMVEC-L cells were

treated with 10 pM CA4DP for 24 hr, the tubulin network disruption was even

more evident (Figure 3-1c). In some treated cells, tubulin components distributed

only in regions around the nucleus, which may account for why cells rounded up

and detached from the monolayer (below and Figure 3-6).

In order to quantitate the effects of CA4DP on tubulin polymerization,

changes in both soluble and polymerized tubulin levels in HMVEC-L cells treated

with CA4DP were examined. Microtubules are an integral part of the cytoskeleton

of eukaryotic cells and are composed of two major soluble proteins, a- and P-

tubulin. Tubulin exists in cells in two forms, soluble (unpolymerized) tubulin and

polymerized (cytoskeletal) tubulin. There exists a dynamic equilibrium between










tubulin polymerization and depolymerization to maintain the normal function of

cells. CA4DP possesses a high affinity for the colchicine binding site on tubulin

and results from SDS-PAGE analysis showed a time-dependent depolymerization

of tubulin when HMVEC-L cells were treated with 5 tM CA4DP (Figure 3-2).

Polymerized tubulin comprised -50% of the total tubulin in HMVEC-L cells

before CA4DP exposure. It decreased to 1.3% 24 hr after CA4DP treatment. It

should be noted that the effect of CA4DP on tubulin polymerization occurred very

rapidly with decreased polymerized tubulin levels being detected within 2 hr after

drug exposure.

As non-toxic doses of CA4DP were found to inhibit cell cycle progression

(Chapter 2), the effect of CA4DP on endothelial cell migration also was

investigated. Endothelial cell migration is an essential endothelial cell function in

both angiogenesis and the wound healing process. In angiogenesis, endothelial

cells are required to proliferate and migrate in response to angiogenic stimuli

(Folkman, 1986). Upon wounding, the cytoskeletal network in a quiescent cell

undergoes dramatic redistribution and reorganization to facilitate the directional

movement of the cell into the injured area (Braunhut et al., 1996). To mimic this

process, a scrape of uniform width (2 mm) was produced in HMVEC-L cell

cultures grown to confluency. Cells migrating into the denuded area were counted

24 and 48 hr later. Results revealed that CA4DP-treated HMVEC-L cells

exhibited a reduced capacity to migrate as seen by the 10-40% reduction in the

number of cells detected in the denuded area (Figure 3-3). In addition,

observations made by phase microscopy also showed that, compared to CA4DP-










treated cells, untreated cells could penetrate a greater distance into the denuded

area over the same time period.

Although studies using the clonogenic cell survival assay (Chapter 2)

revealed a concentration-dependent activity of CA4DP against proliferating

HMVEC-L cells (Figure 2-3), this cell survival assay can not distinguish between

whether the effects of CA4DP result from cell death by necrosis or apoptosis. To

shed light on this issue, HMVEC-L cells were treated with 5 and 50 pM doses of

CA4DP for 2 hr and the induction of apoptosis was assessed at various times after

treatment by TUNEL assay. Under these conditions, HMVEC-L cells showed

clear evidence of nuclear condensation and fragmentation, which are

characteristic of apoptosis. When quantified, a time-dependent increase of

apoptotic cells in the 50 pM-CA4DP-treated cell population from -7%

immediately after treatment to -20% 22 hr later was noted (Figure 3-4a). At the

lower dose (5 pM), increased apoptosis in HMVEC-L cells could be detected 22

hr after drug treatment. Figure 3-4b also showed that the CA4DP-induction of

endothelial cell apoptosis was clearly dose dependent. Compared to only 2%

apoptotic cells in the untreated cell population, HMVEC-L cells treated with 50

pM CA4DP for 2 hr exhibited apoptosis levels -10 fold higher 22 hr after

treatment. The result was consistent with previous studies in which DAPI staining

was used to visualize the apoptotic cells with nuclear condensation and

fragmentation (Table 3-1).











As cells undergoing apoptosis tend to detach from culture dishes, the number

of cells in the supernatant after CA4DP treatment were quantitated using a

Coulter counter. The results showed a dose-dependent increase in cell detachment

when proliferating, but not quiescent, HMVEC-L cell cultures were treated with

CA4DP, again demonstrating the selective activity of CA4DP (Figure 3-5). This

dose dependent endothelial cell detachment correlated closely with the number of

apoptotic cells found on the monolayer (Figure 3-5 vs Figure 3-4b). Furthermore,

when the detached cells were examined by TUNEL assay, more than 90% were

found to be apoptotic.



Discussion

CA4DP is a tubulin-binding agent which has been shown to produce

extensive hemorrhagic necrosis in both rodent and human tumor models (Dark et

al., 1997; Li et al., 1998; Horsman et al., 1998). Indeed animal studies with this

agent used alone or in combination with traditional anticancer therapies have been

sufficiently promising (Li et al., 1998; Chaplin et al., 1999) to initiate Phase I

clinical trials with CA4DP in both the UK and US. The preclinical investigations

from our laboratory and other groups suggested that the agent's selective activity

against proliferating endothelial cells may be of particular importance (Dark et

al., 1997). To explore this further, the present studies were undertaken with

HMVEC-L cells in vitro with the aim of characterizing the effects of CA4DP and

delineating more directly the role of the endothelial cell in its mechanism of

action.










Strategies aimed at targeting the tumor vessel network, particularly

antiangiogenic therapies, have received considerable attention as alternative

cancer therapies (Schweigerer, 1995; Denekamp, 1990; Denekamp, 1993).

Damaging the tumor vessels directly and selectively with antivascular agents

necessitates the existence of key differences between the vessels comprising

tumors and normal tissues. The much higher proliferative index of tumor

associated endothelial cells as compared to those found in normal tissues provides

such a difference (Denekamp, 1993). The development of drugs which are

particularly toxic to dividing endothelial cells aim to exploit this difference and

offer the possibility of significant treatment selectivity.

Earlier results (Chapter 2) demonstrated that CA4DP acts selectively against

proliferating endothelial cells. For example, data in Figure 2-3d illustrated that

CA4DP starts to affect the viability of proliferating HMVEC-L cells at the dose

~1 ptM. Increasing the drug dose to -100-fold higher still showed no effect in

quiescent HMVEC-L cells. This observation was consistent with previous

findings of CA4DP activity in HUVEC cells (Dark et al., 1997). However, as the

clonogenic cell survival assay can not distinguish between necrotic and apoptotic

cell death, based on the studies from Chapter 2 (Figure 2-4), CA4DP-induced

endothelial cell apoptosis was measured in detail. Using the TUNEL assay a dose-

dependent induction of apoptosis in HMVEC-L cells after CA4DP treatment

could be demonstrated (Figure 3-4). This effect was dramatic at the 50 pM dose,

which induced apoptosis in -20% of the treated endothelial cells that were still

attached. Considering the -10% detached cells that were almost all apoptotic after











50 pM CA4DP treatment, the overall apoptotic cell level was as high as 30%,

which was pretty significant. Considering the possibility that only 30-40% of the

HMVEC-L cells are actively dividing cells (Chapter 2), we believe that CA4DP

induces the death of proliferating HMVEC-L cells predominantly by apoptotic

processes. Prior studies using DAPI staining as well as flow cytometric

evaluations of HMVEC-L cells stained with annexin-V and propidium iodide,

also showed 5-10-fold increases in apoptosis in treated compared to untreated

cells (data not shown). These observations are consistent with those of Iyer et al.

who used increased caspase-3 activity to show that CA4DP induced apoptosis in

HUVEC cells (Iyer et al., 1998). It is still possible that the apoptotic cell death

may not show the whole picture of the cytotoxicity of CA4DP. Some endothelial

cells may die because of losing their clonogenicity after CA4DP treatment.

However, since the antivascular effect of CA4DP occurs very rapidly, we believe

that for the in vivo situation, the induction of apoptosis by CA4DP is the major

contributor of cell death compared to clonogenic cell death.

The cytoskeleton of endothelial cells participates in a number of cellular

processes, including not only spindle formation and chromosome segregation, but

also intracellular transport of molecules, cell motility, and angiogenesis

(Giannakakou et al., 1998). The microtubules which form an integral part of the

cytoskeleton therefore provide an attractive molecular target for anti-vascular

drugs such as CA4DP. The present studies showed that CA4DP caused a time-

dependent tubulin depolymerization in HMVEC-L cells. Within 24 hr, most

microtubules in the cells depolymerized into free tubulin subunits (Figure 3-2).










Immunofluorescence studies showed complementary results; a disorganized

pattern of microtubules was evident in CA4DP-treated HMVEC-L cells rather

than a normal cytoskeletal architecture (Figure 3-1). These findings indicate that

CA4DP binds to tubulin and shifts the dynamic equilibrium that normally exists

in cells between polymerized and soluble tubulin.

Disrupted microtubule structure also directly affects the ability of cells to

migrate. Figure 3-3 shows that CA4DP doses of 0.1 10 pM inhibited, by 10 ~

40%, the ability of HMVEC-L cells to migrate into denuded areas in culture

plates. These findings suggest that, although CA4DP may be predominantly

acting as an antivascular agent, it also possesses at least some of the features

typically associated with antiangiogenic agents, namely effects on migration and

proliferation. This conclusion is consistent with the results from Chapter 2 which

showed that CA4DP at doses less than those affecting cell survival in the

clonogenic cell survival assay, blocked HMVEC-L cells at the G2/M phase of the

cell cycle (Figure 2-2).

Another outcome of microtubule structure disruption is endothelial cell

detachment. Results from Figure 3-5 showed that CA4DP caused a dose-

dependent cell detachment in proliferating HMVEC-1 cells. Although not shown,

at high doses, cells began to round up and detach immediately after a 2-hr CA4DP

treatment. If such an effect occurs in vivo, it could explain the vascular shutdown

seen after CA4DP exposure. Considering the irregular, capillary-like vessels in

tumors, even if only a few endothelial cells round up and detach from the










monolayer vessel bed, this might be sufficient to occlude the blood flow and

ultimately shut down the whole vascular supply in the tumor.

In summary, the present results show that CA4DP is specifically toxic to

proliferating endothelial cells predominantly by apoptotic pathways. CA4DP also

inhibits tubulin polymerization, endothelial cell migration and attachment. While

the in vitro results indicate that CA4DP has potent effects on endothelial cells, it

should be recognized that vascular shutdown in CA4DP-treated tumors can occur

within 20 min after treatment (Sackett, 1993; Li et al., 1998). The rapidity of

vascular shutdown observed suggests that more immediate changes are

responsible for the drug effects seen. One possibility is that CA4DP can have

dramatic effects on the three-dimensional shape of newly formed endothelial

cells. It is possible that the early manifestations of cell shape changes brought on

by CA4DP effects on tubulin binding which lead to cell detachment and apoptosis

in vitro, result in similar physical effects in vivo which dramatically alter capillary

blood flow, expose basement membrane and, as a result, induce haemorrhage and

coagulation. Recent studies showed that CA4DP induces endothelial shape

changes with a consequent increase in permeability of an endothelial cell

monolayer to macromolecules (Twardowski and Gradishar, 1997). The increase

in vascular permeability to macromolecules may result in an increase in

interstitial fluid pressure, an increase in blood viscosity, procoagulative effects,

vascular collapse, and the induction of cytokines.

The present findings provide a basis for the selective action of CA4DP

against the proliferating endothelial cell population found in tumors. The reason






54



for the tumor selectivity of CA4DP may relate to differences in proliferation rate

of endothelial cells in tumors and normal tissues. Further investigations of

morphological changes of endothelial cells in vivo need to be pursued.























































Figure 3-1. Indirect immnunofluorescence ofmicrotubules in HMVEC-L cells. a)
No drug treatment: h) 4 hr treatment with 10 pM CA4DP.












































Iigtlurc I --collinuld. c) 24 htr trealimcnt willt 0 pM ('A41).


















Time (hr) 0 2 4 24

%P 49.6 37.0 12.6 1.3

P S P S P S P S







100-
o polymnrzd tubulln
o80 0 soluble tubulln



/Z

40


0- I -

20


0 2 4 24

Time (hr)







Figure 3-2. Effect of CA4DP on tubulin polymerization in HMVEC-L cells. The
cells were treated with 5 pM CA4DP over a 24-hr period. Cells were harvested at
different time points, and tubulin polymerization was assessed. The percent of
polymerized tubulin (%P) was determined by dividing the value of polymerized
tubulin by the total tubulin content (the sum of P and S).






















CA4DP 2 hr treatment


0 0.1 1 10

Drug dose (pM)


Figure 3-3. HMVEC-L cell migration into a 2 mm denuded area in culture plates.
Cells were either untreated or exposed to CA4DP for 2 hr. Migrating cell numbers
were counted 24 and 48 hr later. Data are the mean SE of 6 replicates.
















30-




0 20-



S10-


CA4DP 2 hr treatment


0 50 .M CA4DP
* 5tMCA4DP











5 10 15 20 25
Time after treatment (hr)


CA4DP 2 hr treatment


S20-


. ^
0 10


control 1 5 10 50

Drug dose (pM)


Figure 3-4. Effect of CA4DP on HMVEC-L apoptosis. (a) HMVEC-L cells were
treated with 5 and 50 S M CA4DP for 2 hr and stained for TUNEL assay after a
specified time. (b) HMVEC-L cells were treated with various doses of CA4DP for 2
hr, and the apoptotic nuclei in the monolayer were counted 22 hr later. Data are the
mean SE of three independent experiments.






60












Table 3-1. COMPARIOSN OF APOPTOSIS DATA FROM TUNEL
ASSAY AND DAPI STAINING (CA4DP 2 HR TREATMENT)

APOPTOTIC CELLS (%)
TREATMENT TUNEL DAPI
control 1.8 (+0.2) 1.8 ( 0.2)
1 IM 3.4 (+0.7) 4.9 ( 1.4)
5 pM 5.6 ( 1.9) 7.2 (+2.1)
10I M 11.8 ( 1.4) 13.3 (+1.6)
50 PM 19.6 (+3.0) 19.4 (+2.7)





















CA4DP 2 hr treatment


* Proliferating HMVEC-L

O Quiescent HMVEC-L


0.01 0.1 1 10 10

CA4DP dose (pM)


Figure 3-5. Quantification of cell detachment in CA4DP treated HMVEC-L cells.
Cells were treated with various doses of CA4DP for 2hr and the numbers of
detached cells were counted electronically 22 hr later. Data are the mean SE of 6
replicates.















CHAPTER 4
STUDY OF THE EFFECTS OF CA4DP IN THE MODEL OF KAPOSI'S
SARCOMA


Introduction

Kaposi's Sarcoma (KS) is a highly vascularized neoplasm that primarily

results in raised, highly vascularized lesions (Groopman, 1987; Tappero et al.,

1993). Before the 1980s, KS was a rare disorder that occurred predominantly in

elderly men of Mediterranean or Eastern European Jewish descent. With the

advent of the acquired immunodeficiency syndrome (AIDS) epidemic, its

occurrence has increased dramatically. KS is classified into four different types:

classic, African endemic, iatrogenic or drug-associated, and AIDS-related (Sung

et al., 1997). Classic KS usually follows an indolent and benign clinical course

that rarely requires treatment. In contrast, AIDS-KS is a fulminant disease that

requires aggressive pharmacotherapy, especially when it involves visceral organs.

AIDS-KS is the most common neoplastic disease in patients with AIDS.

Presently, it is the fourth leading clinical manifestation of AIDS (Pluda et al.,

1993). Cutaneous or mucocutaneous lesions may occur. Lesions occurring in the

viscera primarily affect the gastrointestinal tract, lymph nodes, and pulmonary

system. Shortness of breath, dyspnea on exertion, increased respiratory rate, and

decreased oxygen saturation are common in patients with pulmonary

involvement. Involvement of the pulmonary tract occurs in 15-50% of patients










and is estimated to contribute to 25% of deaths in patients with AIDS-KS

(Tappero et al., 1993).

Histopathology of KS reveal highly vascularized lesions with abundant

angiogenesis accompanied by abnormal blood vessel development and leakage of

blood (Gallo, 1998). The predominant cells in the tumor are the spindle-shaped

cells believed to be the tumor cells of KS. These cells express some of the surface

markers of activated endothelium but also contain smooth-muscle actin,

suggesting that the cell of origin may be a primitive vascular cell (Kroll and

Shandera, 1998). Growth factors that support spindle cell proliferation include

interleukin (IL)-l, IL-6, and tumor necrosis factor (TNF) (Sung et al., 1997).

Proteins that regulate neovascularization or angiogenesis such as basic fibroblast

growth factor (bFGF), platelet-derived growth factors, and vascular endothelial

growth factor (VEGF) also promote growth of spindle cells (Nakamura et al.,

1992).

Standard treatments for KS include intralesional injection of vinblastine or

U2-interferon, local radiotherapy and systemic chemotherapy. These treatments

are administered primarily for symptomatic relief and to prevent disease

progression. Cytotoxic chemotherapy is the standard therapy for patients with

extensive lesions and disease involving visceral organs and lymph nodes. A single

agent such as vincristine or a2-interferon is used for mild cases. More advanced

cutaneous or visceral KS is usually treated with combination chemotherapy

including agents such as vincristine, bleomycin, and doxorubicin. However, these

treatments do not significantly prolong survival, and the clinical effectiveness is











not satisfactory (Lilenbaum and Ratner, 1994). Vincristine or a2-interferon

monotherapy give a clinical response in only about 25% of patients, while the

combination of vincristine, bleomycin, and doxorubicin causes regression in

about 40% (Kroll and Shandera, 1998). Continued pursuit of more effective and

less toxic agents is clearly needed.

Given the tumor's histopathology, a variety of new treatment approaches,

particularly those focused on inhibiting tumor angiogenesis are being

investigated. Angiogenesis plays a crucial role in the pathogenesis and

progression of KS (Comali et al., 1996) and antiangiogenesis approaches may

provide a means of arresting the progression of KS. The drug initially tested as an

angiogenesis inhibitor in patients was TNP-470. When administered once weekly

by intravenous infusion, this agent gave partial responses in KS patients (Dezube

et al., 1998). More recently, there has been an interest in exploring the clinical

utility of thalidomide as an anti-KS agent. This was based on evidence that

thalidomide could inhibit angiogenesis, block tumor necrosis factor alpha (TNF-

a), and inhibit intercellular adhesion molecules and basement membrane

formation (Gasc6n and Schwarts, 2000). Preliminary results from two Phase II

clinical trials showed that thalidomide had activity in a subset of patients with KS

(Welles et al., 1997; Bower et al., 1997). Based on the encouraging results

observed with thalidomide and TNP-470, clinical research into the antiangiogenic

activities of these and other agents, including interleukin-12 and angiostatin,

continues (Kroll and Shandera, 1998; Gasc6n and Schwarts, 2000).










Because KS is a highly vasoactive neoplasm, directly targeting the actively

growing vessels of the tumor may be another approach suitable for KS treatment.

Since large numbers of neoplastic cells are directly supported by small numbers

of endothelial cells, damaging the tumor endothelium could have marked impact

on tumor cell survival and growth (Bicknell and Harris, 1992). In the present

investigation we examined the efficacy of CA4DP in KS xenografts by assessing

the extent of both vascular damage and cytotoxic action in these tumors.


Methods and Materials

KS Xenografts and Treatments

KS xenografts were initiated by injecting the flanks of 6-8-week-old athymic

NCR nu/nu mice (Frederick Laboratories, Frederick, MD) with x l06 KSY-1

cells (ATCC, Rockville, MD) (Lunardi-Iskandar et al., 1995) and were serially

passed by subcutaneous transplantation of tumor pieces in the flanks.

Macroscopic tumors were available for experiments 3-4 weeks later. Tumor-

bearing mice were allocated to groups and received either no treatment or

different doses of CA4DP (OXiGENE Inc., Lund, Sweden). CA4DP was

dissolved in 0.9% sterile saline and injected intraperitoneally in a volume of 0.01

ml/g animal body weight.

Hoechst-33342 Studies

Hoechst-33342 (bisBenzimide, Sigma) solution was made up in 0.9% sterile

saline immediately before use. KS-bearing mice were either untreated or treated

with 100 mg/kg CA4DP. Hoechst-33342 then was administered at 40 mg/kg










intravenously (volume 5 ml/kg) at various times after CA4DP injection (Smith et

al, 1988). One minute after Hoechst-33342 injection the mice were killed, the

tumors and the normal tissues (lung, liver, and muscle) of the mice were rejected

and immediately immersed in liquid nitrogen for subsequent frozen sectioning.

For each tumor sample, 10 lim cryostat sections were cut at three different levels

between one pole and the equatorial plane. The sections were air dried and then

studied under UV illumination using a fluorescent microscope. Blood vessel

outlines were identified by the surrounding halo of fluorescent H33342-labelled

cells. Vessel counts were performed using a Chalkley point array for random

sample analysis (Curtis, 1960). Briefly, each section was viewed at x 10 objective

magnification. A 25-point Chalkley grid was positioned randomly over field of

view. Any points falling within haloes of fluorescent cells were scored positive.

Twenty random fields were counted per section and a minimum of six sections

per tumor was examined.

Histological Staining

Histological sections were prepared from KS xenografts 4 and 24 hr after

CA4DP was given. All specimens were fixed in 10% neutral buffered formalin,

routinely processed and embedded in parafin. Sections (4 P) applied to slides

were deparaffinized in xylene and hydrated through graded alcohols. Standard

hematoxylin and eosin (H&E) staining was used for each slide. Necrotic fractions

in KS tumors were quantified by image analysis. Briefly, stained sections were

divided into 4-8 grids and areas of necrosis within each grid were traced on an










Image Pro Plus system. All grid measurements were combined and the percentage

of necrosis relative to the total area of the tumor was calculated.

Clonogenic Cell Survival Assay

Clonogenic cell survival in treated or untreated tumors was assessed using an

in vivo to in vitro clonogenic cell survival assay as previously described

(Allalunis-Tumer and Siemann, 1986; Siemann, 1995). Briefly, 24 hr after

treatment, tumor-bearing mice were killed and their tumors excised and then

dissociated to a single cell suspension using an enzyme cocktail (0.025%

collagenase, 0.05% pronase, and 0.04% DNase). The cells then were mixed with

104 lethally irradiated cells in a-MEM medium containing 10% fetal calf serum

and plated into 60 mm petri dishes. After 2 weeks of incubation at 370C, colonies

of 50 or more cells were counted with the aid of a dissecting microscope. Tumor

surviving fractions were determined by multiplying the calculated fraction of

surviving cells by the ratio of cells recovered in treated versus untreated tumors.

Tumor Growth Delay Assay

Once KS xenografts had reached a minimum size of 200 mm3, the mice were

allocated to different groups for CA4DP treatment. A 100 or a 300 mg/kg dose of

CA4DP was used in the single-dose treatment studies. For the multiple-dose

groups, 100 mg/kg CA4DP was administered either on days 1, 3, and 5 or on days

1, 5, and 9. After treatment, tumors were measured daily using calipers and the

perpendicular diameters were determined. Tumor volumes were estimated using

the equation: Volume = 4cr3/3, r = (a+b)/4, where a and b are the perpendicular

diameters. The time for each tumor to reach a size of 900 mm3 was recorded.










Hypoxia Stress and CA4DP Treatment

KSY-1 cells were plated into 60 mm petri dishes at 2xl05 cells/dish. Once

KSY-1 cells were attached to the dishes, the cells were treated with a range of

doses of CA4DP. Treated KSY-1 cells used for hypoxic conditions were

immediately placed in an airtight chamber and subjected to repeated rounds of

evacuation and replacement with nitrogen gas. The sealed chambers were then

incubated for 24 hr at 370C. KSY-1 cells under aerobic conditions were also

treated with a range of doses of CA4DP at 370C for 24 hr. After CA4DP

treatment, these treated cells were then trypsinized, counted, and plated into 60

mm Cell+ petri dishes for clonogenic cell survival assay. After 2 weeks of

incubation at 370C, colonies of 50 or more cells were counted with the aid of a

dissecting microscope. Cell surviving fractions were calculated as the ratio of

colonies counted in treated versus untreated group.

Results

Initial studies focused on the early effects of CA4DP treatment on tumor

vasculature. Results obtained with the Hoechst-33342 fluorescent dye showed that

a single dose of 100 mg/kg CA4DP caused an almost complete vascular shutdown

in KS xenografts within 4 hr after treatment (Figure 4-1). Compared to the

abundant vasculature in the untreated tumors, KS xenografts in mice treated with

CA4DP showed vessels essentially only near the periphery of the tumors. This

vascular damaging effect of CA4DP occurred rapidly and was detected in KS

xenografts 30 min after drug treatment. Indeed, when functional vessels were

counted, most were found to be shutdown by CA4DP within 0.5 to 2 hr after










treatment (Figure 4-2). The vasculature in normal tissues from untreated and

CA4DP-treated mice also were assessed. It is clearly seen from Figure 4-3 that

none of the vascular networks in the lung, the liver, and the muscle was affected

at 4 hr after a 100 mg/kg dose of CA4DP treatment.

Histological evaluations of KS xenografts showed morphological evidence

of damage in tumor cells within a few hours after CA4DP treatment. By 24 hr

after CA4DP treatment (100 mg/kg), extensive haemorrhagic necrosis could be

seen with viable tumor cells detectable only at the periphery of the tumor adjacent

to the surrounding normal tissues (Figure 4-4).

To quantify the extent of necrosis produced by CA4DP treatment, sections

from KS xenografts removed 24 hr after treatment were assessed using an image

analysis system. The results (Figure 4-5) showed that compared to the -10%

necrosis seen in untreated tumors, treatment with 50 mg/kg CA4DP increased the

extent of necrosis to ~60%. In xenografts treated with 100 mg/kg CA4DP, the

necrotic fraction increased to ~90% 24 hr after drug treatment.

Anti-tumor effectiveness of CA4DP was determined by measuring

clonogenic cell survival in KS xenografts treated with various doses of this agent.

The data demonstrated that administering increasing doses of CA4DP to tumor-

bearing mice resulted in a dose-dependent increase in tumor cell kill (Figure 4-6).

A comparison of results in Figures 4-5 and 4-6 further shows a consistency

between the clonogenic cell survival data and the results of the histological

evaluations, i.e. a 100 mg/kg dose of CA4DP caused -90% tumor cell death and

necrosis 24 hr after treatment. The tumor cell killing effect with a single 100










mg/kg dose of CA4DP treatment also was assessed at various times after

treatment. The results showed that the maximum tumor cell kill (>90%) was

achieved between 1-3 days after CA4DP treatment (Figure 4-7). At later times

cell survival recovered.

Hypoxic tumor cells normally exist in central areas of tumors, far away from

sources of efficient oxygen supply. Since CA4DP causes extensive central

necrosis in KS xenografts, it raised the question of whether part of the effect of

this agent was the consequence of CA4DP killing hypoxic cells more efficiently

than well-oxygenated tumor cells. To answer this question, we examined the

toxicity of CA4DP in both aerobic and hypoxic KSY-I cells in vitro (Figure 4-8).

The results showed that hypoxic KSY-1 cells were not more susceptible to

CA4DP treatment than aerobic KSY-1 cells. Indeed clonogenic cell survival

determinations showed the surviving fraction of hypoxic cells after CA4DP

treatment to be slightly higher than that for aerobic cells. This result further

supports the notion that it is the antivascular effects of CA4DP, not a higher

sensitivity of hypoxic tumor cells to CA4DP, that are responsible for the induced

extensive central necrosis in KS tumors after CA4DP treatment.

In conjunction with the cell survival investigations, studies evaluating the

effects of CA4DP on KS growth also were performed. Tumor-bearing mice were

treated at a size of 200 mm3 with a single dose of either 100 or 300 mg/kg dose of

CA4DP. Such treatments resulted in a slight, but not significant, tumor growth

delay (Figure 4-9, Table 4-1). Indeed the higher dose of CA4DP (300 mg/kg) did

not increase the tumor growth delay compared to that achieved with 100 mg/kg.










In an attempt to apply this agent more efficiently and in particular to impair the

regrowth of vasculature from the surviving rim of tumor cells (Figures 4-1 and 4-

4), studies utilizing repeated injections of CA4DP (100 mg/kg) also were

performed. Treatment commenced when the tumors reached 200 mm3. Two

different treatment schedules were used: CA4DP was administered on days 1, 3

and 5 or on days 1, 5, and 9. The results showed a far superior response of the KS

xenografts to the multiple CA4DP treatment schedules (Figure 4-9). Both

treatments induced significant growth delay compared to untreated tumors and

tumors treated with either single dose of CA4DP (Table 4-1). Administering

CA4DP using the multiple dose schedule resulted in a growth delay of -20 days

(Figure 4-9, Table 4-1).

Discussion

Kaposi's sarcoma is the most common tumor seen in HIV-infected patients.

Several chemotherapeutic agents including vinca alkaloids, etoposide, bleomycin,

and doxorubicin are commonly used to treat AID-KS patients (Sung et al., 1997).

Vinca alkaloids (either vincrinstine, vinblastine, or an alternating regimen of the

two) exert their anticancer effect by binding to tubulin and preventing its

polymerization to form microtubules (Chabner and Collins, 1990; Yarchoan,

1999), thus inhibiting a number of cellular processes, including mitosis. Over the

years there have been several advances in the therapy of this disease, including

the use of liposomal anthracyclines, paclitaxel, and antiangiogenesis agents TNP-

470 and thalidomide (Yarchoan, 1999; McGarvey et al., 1998). Among these,

paclitaxel, which also interferes with microtubule dynamics by promoting the










formation of highly stable microtubules which resist depolymerization, was found

to inhibit the growth of KS-derived spindle cells and to be a potent inhibitor of

endothelial cell proliferation (Saville et al., 1995).

Like vinca alkaloids, CA4DP also binds to tubulin and inhibits tubulin

polymerization. However, unlike the vinca alkaloids, CA4DP has demonstrated

antivascular effects at very low doses (Dark et al., 1997; Li et al., 1998; Chaplin

et al., 1999). It has been well established that the vascularization of solid tumors

is a prerequisite if a clinically relevant size is to be reached (Folkman, 1986). The

dependence of the tumor on its induced vessel formation has created a great deal

of enthusiasm in specifically targeting the microcirculation in cancer therapy

(Denekamp, 1993). Results from our laboratories and those of others have shown

not only that CA4DP has specific effects on actively dividing endothelial cells but

also that this agent can cause rapid vascular shutdown in a variety of preclinical

tumor models (Dark et al., 1997; Li et al., 1998; Horsman et al., 1998). Because

KS is a highly vascularized neoplasm with a cellular origin suggested to be

endothelial cell derived, and because tubulin-binding agents have previously been

found to be active in KS, the present study was undertaken to examine the

efficacy of CA4DP in an in situ model of this disease.

The KSY-1 cell line originated from cells isolated from the pleural effusion

of an AIDS-associated KS patient (Lunardi-Iskandar et al., 1995). KSY-1 cells

promote tumorigenesis, angiogenesis, and metastasis in immunodeficient mice.

The model's similar biological, morphological and immunophenotype make it a

valuable adjunct for studies related to pathogenesis and therapy of AIDS-KS











(Rojiani et al., 2000). In the present study, KSY-1 cells were used to initiate KS

xenografts in athymic mice in order to assess their response to the vascular

targeting agent CA4DP.

The pathophysiological effects of CA4DP observed in the present study in

KS xenografts were similar to those previously reported by our laboratory in the

rodent KHT sarcoma model (Li et al., 1998). CA4DP treatment resulted in a rapid

induction of vascular damage in tumors such that 4 hr after treatment there existed

an almost complete vascular shutdown (Figures 4-1 and 4-2). This was followed

by extensive secondary tumor cell death due to ischemia (Figures 4-4 and 4-5).

Histological assessments showed extensive haemorrhagic necrosis 24 hr after

CA4DP was administered systemically to KS-bearing mice, with only a small rim

of viable tumor cells surviving near the periphery of the tumor (Figure 4-4). These

tumor cells probably survived because they were close to the surrounding normal

tissues where they were supplied with nutrients from the normal tissue

vasculature which was not affected by the action of CA4DP. Studies with

intravital microscopy have shown that peripheral tumor tissue retains some blood

flow after CA4DP treatment but becomes hemorrhagic with dilated blood vessels

(Tozer et al., 1999). This suggests that vascular permeability changes may be

more profound in the periphery, where most of the extravasation of

macromolecules occurs under unperturbed conditions.

Results from the clonogenic cell survival investigations were consistent with

the histological observations. For example, a 100 mg/kg dose of CA4DP which

induced -90% necrosis in KS xenografts also reduced the viable tumor cell










population to -10% of the pretreatment value (Figures 4-5 and 4-6). The cell

survival evaluations established that CA4DP treatments led to a concentration-

dependent killing of KS tumor cells (Figure 4-6). This killing manifested itself

primarily as a rapid loss of viable cells from the cell population within the 24-hr

period after CA4DP treatment. Although normal tissue toxicities were not

measured in the present studies, the antivascular effects of CA4DP were achieved

at doses less than 1/10 of the maximum tolerated dose (MTD) and without

detectable morbidity as previously reported (Dark et al., 1997). While some of the

other tubulin binding compounds, for example the vinca alkaloids, may also

express antivascular action, they exert their effects at doses approaching the MTD

and often only in the presence of significant morbidity (Chaplin et al., 1996).

Indeed vinca alkaloid therapy in AIDS-KS patients is frequently limited by

neutropenia and peripheral neuropathy (Sung et al., 1997). CA4DP's high tubulin

binding affinity (Pettit et al., 1989), selective toxicity in proliferating endothelial

cells, and effectiveness at low doses (Dark et al., 1997) may prove to be of

considerable value in the continuing clinical evaluation of CA4DP as an

antivascular agent.

The present studies showed that a single 100 mg/kg dose of CA4DP had

little effect on the growth of KS xenografts (Figure 4-6) despite the fact that this

dose of CA4DP causes extensive central necrosis in these tumors (Figures 4-4 and

4-5). The most likely explanation is that the remaining viable tumor cells, located

at the periphery of the tumor near the normal tissue, survive and continue to

proliferate. This conclusion is supported by the study illustrated in Figure 4-7.










The viable tumor cell fraction began to increase 2 days after CA4DP treatment

because of the continued proliferation of cells at the rim of the tumors. The lack

of a change in tumor growth following the 100 mg/kg CA4DP treatment probably

is a consequence of a balance between the growth of new cells from the surviving

rim and the removal of necrotic material from the tumor's core. Increasing the

single dose to 300 mg/kg did not result in a greater growth delay in KS xenografts

(Figure 4-9). However this result is not surprising given the extent of cell death

and necrosis caused by a 100 mg/kg CA4DP dose (Figures 4-5 and 4-6).

Increasing the dose further would have little additional effect on the tumor cells

which survive due to their location near normal blood vessels.

The effect of multiple exposures of CA4DP on KS xenograft growth also

was examined. The rationale for these studies was two-fold. First, it was apparent

from the single dose studies that maximum anti-tumor efficacy occurred with

doses of-100 mg/kg (Figures 4-5 and 4-6) and that little could be gained by

increasing the exposure dose further. Second, and more importantly, we reasoned

that administering multiple doses of CA4DP at times when the tumor is regrowing

and the tumor vasculature is recovering and/or re-establishing itself might prove

to be a much more efficient application of this agent. From Figure 4-7 we know

that it was between 2 and 5 days after single dose of CA4DP treatment that the

KS tumors showed recovery and regrowth. Therefore, two different treatment

schedules, either giving CA4DP on days 1, 3, and 5 or on days 1, 5, and 9, were

examined in KS-bearing nude mice. The results showed that unlike the single

dose treatments, both multiple dose schedules caused significant growth delay in










the xenografts (Figure 4-9 and Table 4-1). Multiple doses of CA4DP were clearly

far more effective at inhibiting KS growth than single dose treatments. For

example, administering three 100 mg/kg dose fractions of CA4DP, as opposed to

a single 300 mg/kg treatment, increased the growth delay by -14-17 days (Table

4-1). To date we have not optimized these multiple CA4DP treatment schedules

in the KS model. Whether similar gains can be achieved with lower doses/fraction

and/or greater numbers of multiple treatments is currently under investigation.

Still, the studies described in the present investigations indicate that administering

multiple doses of CA4DP is a very effective way of inhibiting KS growth.

In conclusion, CA4DP treatment can cause vascular shutdown, haemorrhagic

necrosis, extensive tumor cell killing, and growth delay in KS xenografts. These

findings suggest a possible application of the vascular targeting agent CA4DP in

the clinical management of KS.

















































Figure 4-1. KS tumors removed 1 min after i.v. injection of 40 mg/kg H-33342.
Vessels were identified by the surrounding fluorescent tumor cells. Tumors were
either from untreated mice (a) or from mice 30 min (b), 2 hr (c), and 4 hr (d) after
a 100 mg/kg CA4DP treatment. Magnification was x32 for a-d.






















100 mg/kg CA4DP treatment


2 3 4
Time (hr)


Figure 4-2. Vascular counts for KS tumors with Hoechst-33342 staining. Tumor-
bearing mice were treated with 100 mg/kg CA4DP for a specified time before i.v.
injection of 40 mg/kg Hoechst-33342. Vessel counts were performed by using a
Chalkley point array for random sample analysis. Data are the mean + SE.


















A.
















b















Figure 4-3. Normal tissue sections from KS-bearing mice removed I min after i.v.
injection of 40 mg/kg H-33342. Vessels were identified by the surrounding
fluorescent cells. Samples were either from untreated mice (a) or from mice 4 hr
after a 100 mg/kg CA4DP treatment (b). Magnification was xl10. A)Cryostat
sections of the liver.




















































Figure 4-3-continued. B) Cryostat sections of the lung.




















C.








































F gur c4 3 cntinutl. (jC Cy Lia scins tth let. muscic.















a -1 = .
.*- 4 ,E ,-'.. "3.--* L
: -,.i~ .' ... 2-..-


~ s~#B ~ C :!4

.7~.


Figure 4-4. Standard H&E staining of 4- sections from KS tumors. Tumors were
from mice that had received a 100 mg/kg dose of CA4DP 24 hr prior to
assessment. M I,:tlik il, ii was x10 (a) and x40 (b).
























KS xenograft


0 50 100

CA4DP dose (mg/kg)






Figure 4-5. The extent of necrosis in KS xenografts assessed 24 hr after the
administration of single doses (50 or 100 mg/kg) of CA4DP. Data are the mean +
SE.
























.2
0.1





U)


0.01








0.001 ii
0 50 100 150 200

CA4DP dose (mg/kg)






Figure 4-6. Tumor cell killing in KS tumors treated with increasing doses of
CA4DP. Data were determined 24 hr after drug treatment and are the mean SE
of at least 6 tumors.























KS 100mg/kg CA4DP


0 1 2 3 4 5
Time after treatment (day)






Figure 4-7. Tumor cell killing in KS tumors treated with a 100 mg/kg CA4DP.
Clonogenic cell survival assay was performed at a specified time after drug
treatment. Data are the mean SE of at least 6 tumors.
























.o





'I





O Aerobic
0 Hypoxic


0.01
0.001 0.01 0.1 1 10 100

Drug dose (pM)






Figure 4-8. The cytotoxicity profile of CA4DP against KSY-I cells under hypoxic
or aerobic conditions. The cells were exposed to CA4DP for 24 hr and the cell
killing effects were assessed using clonogenic cell survival assay. Data are the
mean SE of three experiments.

























KS tumor growth (w/ CA4DP treatment)

Median value


---
-I-


control SD(100) SD(300) MD(1,3,5) MD(1,5,9)
10(8-13) 13.5(8-19) 12.5(10.5-21) 30(10.42) 27(16-53)


Figure 4-9. Growth delays in KS tumors with single or multiple doses of CA4DP
treatment. Single dose (SD) of CA4DP (100 or 300 mg/kg) was administered on
day 1. Multiple doses (MD) of CA4DP (100 mg/kg) were administered on days 1,
3, and 5 or on days 1, 5, and 9. Each datum point represents an individual animal.


o O







88









Table 4-1. GROWTH DELAY IN KS XENOGRAFTS CAUSED BY SINGLE
OR MULTIPLE DOSES OF CA4DP INJECTION IN TUMOR-BEARING MICE


GROUP MEDIAN TIME
TO 900mm3
(DAYS)
I. Control 10.0
2. S.D.(100)1 13.5
3. S.D.(300)' 12.5
4. M.D. (1,3,5)2 30.0
5. M.D. (1,5,9)' 27.0


GROWTH DELAY SIGNIFICANCE' SIGNIFICANCE3
(DAYS) VS CONTROL


NS
NS
P< 0.025
P< 0.025


4vs2, 5 vs2 P<0.025
4 vs 3, 5 vs 3 P< 0.025


Single dose of 100 or 300 mg/kg CA4DP were used in the treatment.
2100 mg/kg CA4DP was administered to tumor-bearing mice on days 1, 3, and 5 or on days 1,
5, and 9.
3 Determined by Wilcoxon Rank-Sum Test while compared the probability distributions of
growth times of control versus treated tumor (a= .025).















CHAPTER 5
STUDY OF THE EFFICACY OF CA4DP IN COMBINATION WITH
CONVENTIONAL ANTI-CANCER THERAPIES IN KS XENOGRAFTS


Introduction

AIDS-related KS is characterized by a heterogeneous presentation and an

aggressive clinical course. It often presents as multiple, symmetric, cutaneous

lesions. Because of the heterogeneous presentation of AIDS-KS, no single

treatment regimen can be recommended for all patients. Therapy for AIDS-KS is

not curative and, to date, no therapy has been unequivocally proven to impact

survival, with the possible exception ofinterferon-a (Morris and Valley, 1996).

The extent and rate of progression of AIDS-KS and the severity of the underlying

HIV infection are factors used to determine the best treatment approach for

individual patients.

Since cell proliferation and angiogenesis are the two key mechanisms

involved in KS tumor growth, most of the treatments target one or both of them.

Antiproliferative agents commonly used are chemotherapy, interferon and

radiotherapy. AIDS-KS lesions are exquisitely sensitive to radiation therapy (Hill,

1987). Symptoms caused by mass effects (pain and lymphadenopathy) are best

treated with radiation therapy because a response can be more rapidly achieved.

Significant responses are reported in essentially all lesion treated with radiation

therapy, including pulmonary KS lesions refractory to conventional chemotherapy











regimens (Hill, 1987). Complications such as mucositis and ulceration of skin and

tissue may result after radiation therapy.

Both single-agent and combination chemotherapy have been used in patients

with various stages of AIDS-KS. Single-agent chemotherapy produces responses

in ~25% of patients. The duration of response reported has ranged from 1 to 9

months, and all patients relapse eventually after discontinuing therapy (Morris

and Valley, 1996). Several combinations of chemotherapeutic agents have been

investigated in an attempt to increase efficacy and to diminish toxicity by using

lower dosages of individual agents. For example, since vincristine and vinblastine

have exhibited significant activity in AIDS-KS, they were combined in a weekly

alternating schedule in an attempt to reduce toxicity (Sung et al., 1997). Still,

because patients with HIV infection tend to be very susceptible to chemotherapy-

induced toxicity, combination chemotherapy is reserved for treatment in patients

with rapidly progressive or potentially life-threatening disease.

The optimal therapy for AIDS-KS has not yet been determined. Additional

efforts in the management of AIDS-KS are directed at the underlying

pathogenesis of the disease. Several antiangiogenic agents are being evaluated,

including the heparin analog pentosan polysulfate, recombinant platelet factor-4,

fumagillol derivatives, bacterial cell wall complexes, and suramin (Morris and

Valley, 1996). Preclinical studies also examined the effects of antisense

oligonucleotides which target bFGF and VEGF mRNA on KS growth. It was

shown that the antisense oligonucleotides could block angiogenensis and KS

lesion formation in nude mice (Ensoli et al., 1994).











Because KS is a highly vasoactive neoplasm, directly targeting the actively

growing vessels of the tumor may be suitable for KS treatment. Previous studies

in our laboratory have investigated the efficacy of the vascular targeting agent

CA4DP in the model of AIDS-KS. Results have shown that 100 mg/kg dose of

CA4DP caused ~90% tumor necrosis in KS xenografts 24 hr after treatment

(Figure 4-5), and that repeated doses, not single dose, of CA4DP treatment caused

significant growth delay in KS xenografts (Figure 4-9). Previous studies also

indicated that CA4DP alone were unable to eliminate the tumor completely, and a

small, nevertheless viable, rim of tumor remained. These remaining viable tumor

cells continue to proliferate, which may explain the lack of change in tumor

growth following a single dose of CA4DP treatment. Therefore, the application of

antivascular agents will need to be given in conjunction with conventional

anticancer therapies. Our studies have shown that CA4DP can significantly

enhance tumor response to radiation in KHT sarcoma (Li et al., 1998), and others

also demonstrated an effective enhancement of antitumor effects of cisplatin and

5-FU by combining with CA4DP (Chaplin et al., 1999; Grosios et al., 2000).

Radiation and chemotherapy are the standard therapies for patients with various

stages of AIDS-KS. In the present investigations we have examined the efficacy

of combining CA4DP with ionizing radiation or chemotherapeutic agents in the

model of Kaposi's Sarcoma.

Before examining the effects of CA4DP in combination with

chemotherapeutic agents in KS xenografts, we screened the sensitivity ofKSY-I

cells to several agents clinically used to treat KS patients, aiming to find the










suitable agents for the combination study in vivo. These agents are cisplatin,

vinblastine, doxorubicin, and VP-16.

Materials and Methods

Cell Culture and Drug Sensitivity Study

KSY-1 cells (ATCC, Rockville, MD) were cultured in positively charged

Cell+ tissue culture flasks from SARSTEDT (Newton, NC). The cells were grown

in RPMI 1640 medium (Gibco BRL, Grand Island, NY) with 100 units/ml

penicillin, 0.1 mg/ml streptomycin, and 10% FBS, and were passed weekly.

To test the sensitivity of KSY-1 cells to chemotherapeutic agents cisplatin,

vinblastine, doxorubicin, and VP-16, KSY-I cells were plated in 60 mm petri

dishes at I xl05 cells/dish. On day 3, cisplatin (Bristol-Myers Squibb Co.,

Princeton, NJ), vinblastine (Fujisawa USA, Inc., Deerfield, IL), doxorubicin

(Gensia Laboratories, Ltd., Irvine, CA), and VP-16 (Bristol-Myers Squibb Co.,

Princeton, NJ) injection solutions were diluted in 0.9% sterile saline before drug

treatment. KSY-1 cells were then treated with these agents at specified

concentrations for 1 hr. The dishes were then washed with PBS and replenished

with fresh RPMI 1640 medium. 23 hr later, cells in each group were trypsinized,

counted, and plated into 60 mm petri dishes for clonogenic cell survival assay.

After 2 weeks of incubation at 370C, colonies of 50 or more cells were counted

with the aid of a dissecting microscope. Cell surviving fractions were calculated

as the ratio of colonies counted in treated versus untreated group.










KS Xenografts and Treatments

KS xenografts were initiated by injecting the flanks of 6-8-week-old athymic

NCR nu/nu mice (Frederick Laboratories, Frederick, MD) with Ixl06 KSY-1

cells and were serially passed by subcutaneous transplantation of tumor pieces.

Macroscopic tumors were available for experiments 3-4 weeks later. Tumor-

bearing mice were allocated to groups and receive either no treatment or different

treatment strategies. CA4DP, cisplatin, and vinblastine all were injected

intraperitoneally in a volume of 0.01 ml/g animal body weight. In the combination

experiments, cisplatin or vinblastine was administered 1 hr before CA4DP

injection. For radiation treatment, tumors were irradiated in unanesthetized mice

using a 37Cs source operating at a dose rate of 1.5 Gy/min. In the combination

studies, CA4DP was given 0.5-1 hr after radiation treatment.

Measurement of Tumor Response

Clonogenic cell survival in treated or untreated tumors was assessed using an

in vivo to in vitro clonogenic cell survival assay as previously described

(Allalunis-Tumer and Siemann, 1986; Siemann, 1995). Briefly, 24 hr after

treatment, tumor-bearing mice were killed and their tumors excised and then

dissociated to a single cell suspension using an enzyme cocktail (0.025%

collagenase, 0.05% pronase, and 0.04% DNase). The cells then were mixed with

104 lethally irradiated cells in a-MEM medium containing 10% fetal bovine

serum and plated into 60 mm petri dishes. After 2 weeks of incubation at 370C,

colonies of 50 or more cells were counted with the aid of a dissecting microscope.

Tumor surviving fractions were determined by multiplying the calculated fraction




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EBWMOZ46U_17WOU1 INGEST_TIME 2013-01-23T15:13:20Z PACKAGE AA00012962_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES



PAGE 1

STUDIES OF THE ANTICANCER EFFECTS OF THE VASCULAR TARGETING AGENT COMBRETASTATIN A-4 DISODIUM PHOSPHATE (CA4DP) By LINGYUNLI 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 2000

PAGE 2

This dissertation is dedicated to my husband, Guoyi, and my parents for their unlimited love and support.

PAGE 3

ACKNOWLEGMENTS I would like to express my sincere gratitude to Dr Dietmar W. Siemann for providing me with thi s very preciou s opportunity to work in a wonderful laboratory and for hi s immea s urable support and encouragement Thi s work would not have been possible without hi s guidance and understanding I would al s o like to extend my gratitude to the members of my supervisory committee Dr Thomas C. Rowe Dr Kathleen T Shiverick and Dr. Jame s R. Zucali for their valuable advice and continuous encouragement in the completion of my study. I am al s o grateful to Dr Amyn Rojiani at the Uni v ersity of South Florida for his assistance and collaboration in my graduate s tudy. In addition I would like to express my appreciation to the past and present members of Dr. Siemann's laboratory including Dr. Kenneth Warrington Jr. Wenyin Shi Sharon Lepler Destry Taylor Heather Newlin Emma Mercer and Howard Salmon for their help and providing a pleasant working atmosphere. I am indebted to all facult y and staff and student groups at the Department of Pharmacology and Therapeutics for the countless help they have rendered and the stimulating intellectual environment they provided .. 1 1

PAGE 4

TABLE OF CONTENTS page ACKNOWLEGMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 ABST:RACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V CHAPTERS I IN'TRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 The U nique Ph ysiology and Microen v ironment of Solid Tumors . . . . . . . 2 Tumor Angiogen esis and Vasculature Targeting . . . . . . . . . . . . . . . . .. 6 Co mbreta s tatin A4 Disodium Phosphate . . . . . . . . . . . . . . . . . . . . . 13 Sigt1ificance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 17 2 COMPARISON OF THE DIFFERENT RESPONSES OF NEO PLAS TIC AND NORMAL CELLS TO CA 4DP IN VITRO . . . . . . . . . . . . . . . 26 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 Material and M ethods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 Re s ult s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 2 3 EFFECTS OF CA 4DP ON HUMAN MICROV ASCULAR ENDOTHELIAL CELLS IN VIT RO . . . . . . . . . . . . . . . . . . . . . . 40 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 Material an.d Method s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 3 Re su lt s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 Discu ss ion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 4 STUDY OF THE EFFECTS OF CA 4DP IN THE MODEL OF "KAPOSI'S SARCOMA . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Material and Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 Result s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 D . 1scuss1on . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 1 ... 111

PAGE 5

5 STUDY OF THE EFFICACY OF CA4DP IN COMBINATION WITH CONVENTIONAL ANTI-CANCER THERAPIES IN KS XEN OGRAFTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 9 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 Material and Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 D . 96 1scu ss1on ....... .... . .. ... ........... ....... ... .................................. 6 SUMMARY AND PERSPECTIVE . . . . . . . . . . . . . . . . . . . . . . . .. 107 REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 BIOGRAPHICAL SKETCH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126 IV

PAGE 6

Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy STUDIES OF THE ANTICANCER EFFECTS OF THE VASCULAR TARGETING AGENT COMBRETASTATIN A-4 DISODIUM PHOSPHATE (CA4DP) By Lingyun Li December 2000 Chair : Dietmar W. Siemann Major Department: Pharmacology and Therapeutics Combretastatin A-4 disodium phosphate (CA4DP) is a tubulin-binding agent which has been shown to lead to rapid vascular shutdown in a variety of tumor models The present s tudies were undertaken to gain insight into the mechanism( s ) of action of CA4DP and to evaluate the antitumor efficacy of CA4DP either alone or in combination with conventional anticancer therapies in a xenograft model of Kapo s i 's Sarcoma (KS). Initial studies that compared the responses of normal and neoplastic cells to CA4DP demonstrated that CA4DP had selective activity against proliferating endothelial cells. Further s tudies showed that CA4DP treatment resulted in a time dependent tubulin depolymeri z ation in HMVEC-L cells. Tubulin disruption directly affected the ability of endothelial cell migration and attachment Studies carried out V

PAGE 7

with a clonogenic cell survival assay demonstrated that CA4DP selectively reduced the number of viable proliferating HMVEC-L cells in a dose dependent manner. Moreover CA4DP induced the death of proliferating endothelial cells predominantly by apoptosis. These findings provide a basis for the in vivo efficacy of CA4DP and its selective action against the proliferating endothelial cell population found in tumors. In vivo studies using Hoechst-33342 staining demonstrated that a single 100 mg/kg dose of CA4DP caused a rapid vascular shutdown in KS xenografts. Histological evaluation showed evidence of morphological damage of tumor cells within a few hours after drug treatment, followed by extensive haemorrhagic necrosis. Studies using an in vivo to in vitro clonogenic cell survival assay further demonstrated that administering increasing doses of CA4DP to tumor-bearing mice resulted in a dose-dependent increase in tumor cell killing. CA4DP also significantly enhanced the anti tumor effects of radiation and chemotherapeutic agents ( cisplatin and vinblastine) in combination treatment studies. Moreover, repeated doses of CA4DP treatment either alone or in combination with cisplatin treatment caused significant growth delay in KS xenografts. These findings suggest that CA4DP ought to be considered as a candidate agent for therapeutic evaluation in patients with AIDS-KS in future Phase II clinical studies . Vl

PAGE 8

CHAPTER 1 INTRODUCTION Cancer is characterized by progressive growth of cells that have lost their proliferative control. These cells ultimately destruct tissue and metastasize to organs distant from the primary site. In general, only one of three cancer victims can be cured by a single treatment modality, usually surgery, radioor chemotherapy. The main problem with conventional cancer treatments, primarily chemotherapy and radiation therapy, is that they do not have high specificity for cancer cells. For radiation therapy, a degree of specificity is achieved by localizing the radiation to the tumor and its immediate surrounding normal tissue. For anticancer drugs, it is the rapid proliferation of the cancer cells that makes them more sensitive to cell killing than their normal cell counterparts. However, both modalities are limited by their toxicity to nortnal cells. In the case of radiotherapy, normal tissue surrounding the tumor limits the radiation dose, whereas for anticancer drugs, it is usually the killing of rapidly dividing nonnal cells, such as those in the bone marrow, hair follicles, and epithelial cells lining the gastrointestinal tract, that limit the dose that can be given. To achieve more tumor-specific treatment, differences between norn1al and malignant cells are being exploited. The physiology of solid tumors at the microenvironmental level provides a unique and selective target for cancer treatment. 1

PAGE 9

2 The U ni q u e Phys i o l ogy and Microenvironment of Solid Tumors Ab n ormal Tumor Vasc ul ature The physiolog y of s olid tumor differs from that of normal ti ss ues in a number of important as pects A critical difference between tumors and normal tis s ue s is the abnor1n a l nature of the tumor microcirculation compared to the well defined microvascular architecture of normal tissues (Konerding et al., 1995). Solid malignant tumors are composed of both cancerous cells and nor1nal host component. Tumor growth, resulting from uncontrolled neoplastic cell division, is absolutely dependent on a parallel proliferation of the nonmalignant cells which comprise the tumor v asculature. There are two types of vessels in tumor tissues: the existing vessels in normal tissues into which the tumor has invaded ; and tumor microvessels arising from neovascularization resulting from increased expression of proangiogenic factors produced by tumor cells (Brown and Giaccia 1998) Both types of vessel s develop structural and physiological abnorn1alities that ha v e become a hallmark of the tumor microvasculature. Studies have shown that tumor blood vessels are highly irregular tortuous have arterio-venous shunts blind ends lack smooth mu s cle or enervation and have incomplete endothelial linings and basement membrane s (Grunt e t al. 1985 ; Dewhirst et al ., 1989 ; Shah Yukich and N elson 1988) ( Figure 1-1 ) As a re s ult blood flow is often sluggish highly irregular and the ve s sels are much '' leakier '' than those in normal tissues Tumor blood supply is therefore characterized by both spatial and temporal heterogeneity in both structure and function

PAGE 10

3 Hypoxia In neoplastic tissue, there is a disproportionate relationship between tumor tissue and its vascular supply. Tumors are said to "outgrow" their blood supply; neovascularization lags behind the increase in the number of neoplastic cells (Tannock, 1970). As a consequence, the vascular network fails to provide adequate nutritional support and leads to heterogeneous tumor microregions varying in concentrations of oxygen, glucose, and other nutritional factors, as well as metabolic waste products, both within and among tumors of the same pathological grade and stage (Vaupel et al., 1989; Vaupel et al., 1996). In most solid malignancies the tissue 0 2 status is poorer than in normal tissue at the site of tumor growth. For example, the medium p0 2 measured with electrodes for human breast tumors was 28 mm Hg, whereas that of normal breast was 68 mm Hg (Vaupel, 1994) Due to the irregular blood flow and high interstitial pressure, some therapeutic agents are poorly delivered to tumors. Cells located distant from the functional blood supply, often hypoxic cells, could be resistant to drug therapy because of three factors : (a) they are exposed to lower concentrations of drug than those adjacent to blood vessels, primarily as a result of the metabolism of such agents through successive cellular layers; (b) as a result of a decline in nutrient and 0 2 availability, cells further away from the vascular system would be dividing at a reduced rate (Amellen1 and Pettersen, 1991; Pallavicini et al., 1979); an important consequence of this hypoxia-induced inhibition of proliferation is that because most anticancer drugs are primarily effective against rapidly dividing

PAGE 11

4 cells, their effectiveness would be expected to fall off as a function of distance from blood vessels (Tomida and Tsuruo, 1999; Reynolds et al., 1996); (c) oxygen-deficient cells may be inherently more resistant (Teicher, 1994). Tumor hypoxia is also an important factor leading to resistance to radiotherapy. In many series of human tumors at different sites, roughly half the tumors have a median value of less than 10 mm Hg (Vaupel et al., 1991; Nordsmark et al., 1994). This median of 10 mm Hg is significant in that this is the point at which radiation resistance starts to develop with full resistance at values of less than 0.5 mm Hg (Brown, 1999). A typical radiation killing curve for mammalian cells under aerobic and hypoxic conditions is shown in Figure 1-2. The difference in radiation sensitivity between the aerobic and hypoxic cells, which is known as the oxygen enhancement ratio, is normally in the range 2.5-3 for mammalian cells. This effect, coupled with the finding that both human and rodent tumors possess regions of tissue oxygenation below 10 mm Hg, indicates why tumor hypoxia remains a key focus of research in radio biology and radiotherapy. Recent studies have shown that hypoxia in solid tumors has an important consequence in addition to conferring a direct resistance to radiation and chemotherapy. Tumor hypoxia also stimulates tumor progression by promoting angiogenesis through the induction of proangiogenic proteins such as vascular endothelial growth factor (VEGF) (Shweiki et al., 1992). Clinical studies with soft tissue sarcomas (Brizel et al., 1996) and with carcinoma of the cervix (Sundfor et al., 1998) have shown that hypoxia is an independent and highly

PAGE 12

5 significant prognostic factor predisposing tumors to metastatic spread. Therefore, tumor hypoxia also is seen as a predisposing factor toward increased 1nalignancy and metastasis. Because of their potential importance for treatment outcome, a large effort has been afforded over the years to identifying strategies that will reduce or eliminate hypoxic cells within solid tumors. A number of strategies to improving tumor oxygenation are being investigated, including high oxygen content gas breathing either alone (Fenton and Siemann, 1995) or coupled with the agent nicotinamide (Horsman et al., 1994; Siemann et al., 1994), right shifting of the oxyhemoglobin curve (Siemann and Macler, 1986; Hirst and Wood, 1987), and use of agents that increase tumor blood flow (Vaupel and Menke, 1989). A second approach is to use chemical sensitizers that mimic oxygen's ability to increase the sensitivity of hypoxic cells to radiotherapy and chemotherapy (Phillips and Wasserman, 19 84). Over the past decade, several potent bioreductive cytotoxins, such as E09 and tirapazamine, agents whose cytotoxic activity is dramatically enhanced when they are metabolized in a hypoxic environment, have been identified (Workman, 1992; Brown and Siim, 1996). Specifically attacking the hypoxic cell subpopulations with bioreductive agents has a greater therapeutic potential than oxygenating the cells or chemically sensitizing them to radiation or chemotherapy. Not only is the killing tumor specific (hypoxic is tumor specific), but the cells killed are the ones resistant to conventional therapies. This principle of ''complementary cytotoxicity'' is illustrated in Figure 1-3. The combined killing of two agents with complementary

PAGE 13

6 cytotoxicity is potentially much greater than that of two agents acting on the same cell population (Brown and Siim, 1996). pH Besides reduced oxygen level, pH is another microenvironmental characteristic which impacts on the therapeutic outcome of solid tumors. Tumors have been shown to have an acidic microenvironment compared to normal tissues (Wike-Hooley et al. 1984). Increased capacity of glycolysis with the resultant production of lactic acid, lactate production via the breakdown of glutamine, and CO 2 production as a result of cellular respiration may contribute to the shift to acidic pH in malignant tissues. The pH status of malignant tissue can also significantly influence drug activity, especially that of compounds which are weak acids and bases. Studies have demonstrated that selective decrease of the extracellular pH decreases the uptake and activity of weak bases such as vinblastine but increases the uptake and cytotoxicity of weak acids such as chlorambucil (Gerweck and Seetharaman, 1996; Parkins et al. 1996). Moreover, pH-induced alterations in drug stability, active transport processes, drug reactivity, activity of enzymes involved in localized drug activation and the interaction of the drug with its molecular target could all alter treatment efficacy. Tumor Angiogenesis and Vasculature Targeting Tumor Angiogenesis Angiogenesis plays a significant role during norrr1al growth, in physiological conditions ( e.g., in the placenta and endometrium) and in pathological conditions

PAGE 14

7 such as inflammation, wound healing, and tumor growth Angiogenesis is thus not a specific phenomenon in tumors or a pathological condition, but instead an integral element of numerous different normal and pathological conditions. Angiogenesis is a complex multistep process involving extracellular matrix remodelling, endothelial cell migration and proliferation, and capillary differentiation and anastomosis, which are regulated by angiogenic peptides (Blood and Zetter, 1990). The newly forrned vessels are usually thin-walled capillaries or sinusoids with little more than an endothelial lining, backed by a basement membrane Mobility and remodelling from pre-existing vasculature is an important component of angiogenesis. Many angiogenic factors that stimulate proliferation and migration of endothelial cells have been described (Pluda, 1997; Teicher, 1995). Amongst the most potent and specific factors for vascular growth is vascular endothelial cell growth factor (VEGF) (Claffey and Robinson, 1996; Zhang et al., 1995). VEGF exists in severa l isoforms produced from one gene. It binds to VEGF receptors of which there are two, VEGFRl (flt-I) and VEGFR2 (flk-1 / KDR). It is generally thought that VEGFR2 is the most important for angiogenesis wrule VEGFRI is expressed on macrophages and stimulates their migration (Jain et al. 1996). In normal adults, angiogenesis is limited to specific reproductive organs, and the growth and turnover of vascular endothelial cells in most tissues is measured in months and years (Hobson and Denekamp, 1984 ) Unlike what is found in most normal tissues, vessels in tumors contain populations of actively dividing endothelial cells in re s ponse to angiogenic factors. In human tumors the number

PAGE 15

8 of dividing endothelial cells may be 50 times greater than in normal tissue (Harris, 1998) The vascularization of solid tumors is a prerequisite if a clinically relevant size is to be reached. Without sufficient vascular supply, no solid tumor can grow beyond a few cubic millimeters (Ausprunk and Folkman, 1977; Folkman, 1986). Further tumor growth depends on nutrient supply via a network of micro vessels (Denekamp, 1993) which can be acquired, in part, by incorporation of existing host blood vessels. However, it is now well established that the majority of tumor blood vessels are newly formed as a result of angiogenesis triggered by the release of stimulators such as VEGF (Siemeister et al., 1998). Thus, neovascularization is a critical aspect of a tumor's growth and development. Fu11her111ore, angiogenesis is also essential for systemic metastasis, and recently it has been shown to be essential for local invasion (Skobe et al., 1997). Anti-angiogenesis Approach The utter dependence of the tumor on its induced vessel forr11ation for growth, survival and spread has created a great deal of enthusiasm for developing therapeutic approaches to specifically targeting the tumor vasculature (Folkman, 1995; Denekamp, 1993). A variety of app roaches are under investigation. One approach, and most extensively studied, involves attempts to prevent the development of the vascular supply by inhibiting angiogenesis (Ingber et al., 1990; O'Reilly et al., 1994; Scott and Harris, 1994). Three options have been considered so far: (1) inhibition of the turnover from an a vascular primary tumor into a fully vascularized tumor; (2) slowdown of tumor progression by preventing a tumor from becoming highly vascularized; (3) prevention of neovascularization

PAGE 16

9 of distant metastases. Antiangiogenesis therapy targets a process that under most circumstances is tumor specific and therefore likely to have few normal tissue side effects. The antiangiogenesis strategy includes agents which interfere with delivery or export of angiogenic stimuli (Schweigerer, 1995) antibodies to inhibit/inactivate angiogenic factors after their release (Mesiano et al., 1998), antisense therapie s (Im et al., 1999) drugs which inhibit receptor action (Witte et al., 1998) and inhibitor s of endothelial cell proliferation (Boehm et al., 1997) Several of the agents have moved forward to the clinic including metalloproteinase inhibitors, pentosan polysulphate and TNP-470 (Marshall and Hawkins 1995; Deni s and Verweij 1997 ; Twardowski and Gradishar 1997) Because of the major importance of VEGF as an angiogenic factor numerous strategies are pre s ently being used to inhibit VEGF activity in tumors. These have included antisense VEGF mRNA monoclonal antibodies and VEGF receptor inhibitors (Dvorak et al 1995 ; Kim et al., 1993). It is now recogni z ed that angiogenesis is regulated by a balance between pro angiogenic and anti-angiogenic facto1s and that los s of inhibitors may be an early stage in tumor progression Angiogenesis inhibitor s include thrombospondin (Taraboletti et al., 1997), s everal cytokines (IL-4, IL-12) and proteolytic breakdown product s of several proteins including prolactin (Clapp and Delaescalera 1997) plasminogen (Cao et al., 1997) and collagen XVIII (O'Reilly et al., 1994). It i s the se inhibitory peptides that have raised hopes that specific inhibition of tumor angiogenesis may be possible with minimal toxicity and high

PAGE 17

10 efficiency. These angiogenesis inhibitors, particularly angiostatin and endostatin, are being actively investigated (O'Re ill y et al 1994; Boehm et al 1997). They are specific inhibitors of endothelial cell proliferation and have no obvious effect on resting endothelial cells, nor on a variety of normal, transfor1ned or neoplastic cells. Studies have shown that systemic administration of endostatin to tumor bearing mice resulted in regression of tumors to a microscopic size (Boehm et al., 1997). A dormant state, without evidence of toxicity, could be maintained for as long as endostatin was administered. Whether these very enco ur aging results will remain to hold true in the future, remains to be seen. Vascular Targeting Approach The concept of anti-angiogenic therapy relates to interfering with the stimulating substances that cause new vessel formation This should be important in preventing establishment of small solid tumors or in preventing metastases. By contrast, another approach, the so-called vascular targeting approach focuses on the use of agents that can directly destroy existing tumor vessels (Denekarnp, 1990; Denekarnp and Hill, 1991 ). Figure 1-4 illustrates the differences between anti-angiogenic and vascular targeting approaches. The functioning vascular network in tumors is pivotal for the survival of the tumor cells. This is confirmed by the fact that artificial induction of ischaemia by clamping off the tumor feeding blood supply results in extensive tumor cell death and, if prolonged tumor cures (Denekarnp, 1993; Chaplin and Horsman 1994). Such studies emphasize the therapeutic potential of strategies that target the tumor vasculature. The antivascular approach a im s to cause a rapid shutdown in the vascular function

PAGE 18

l l of the tumor, leading to extensive secondary tumor cell death. Since thousands of tumor cells are dependent on each tumor capillary for their metabolic requirements, an agent which induced even limited damage to these vessels could produce a cascade of tumor cell death (Denekamp 1984; Denekamp, 1993). Several features make tumor vasculature a suitable target in cancer therapy. While tumor cells are genetically unstable, rapidly mutating, and able to develop multidrug resistance vascular endothelial cells are genetically stable and unlikely to become drug resistant (Folkman e t al. 1997). Drug delivery is not a problem, since the target cells directly line the blood stream. Perhaps most importantly, tumor vasculature represents an actively growing endothelium whereas the endothelium in most normal tissues is essentially dormant (Denekamp et al., 1993). This last feature may provide a key difference between the tumor and normal tissue which can be exploited. Development of antibodies to specific epitopes on the tumor vasculature (Burrows and Thorpe, 1993; Huang et al., 1997) and vascular-targeted gene therapy (Chaplin and Dougherty, 1999) are two approaches that are receiving considerable attention. As mentioned previously, the advantages of targeting endothelium include that it reduces delivery problems and the number of cells that need to be targeted. These factors make the tumor vasculature a potentially ideal target for antibodyand gene therapy-based approaches. For antibody-based strategies, there is a need to identify antibodies that target unique determinants which are selectively and constitutively expressed on the tumor endothelium. Several studies are now underway to identify such antibodies. Potential

PAGE 19

12 cand idate s include TEC-11, which recognized endoglin, and others that recognize the N-tenninal domain of VEGF, av~ 3 integrin and the receptor tyrosine kinase Tie-1 (Brooks et al., 1994; Burrows et al., 1995). Gene therapy constitutes a potentially powerful means of se lecti vely targeting the tumor-associated vascular endothelial cells, while at the same time minimizing the damage inflicted on various normal tissues. Candidate genes are expected to either directly kill vascular endothelial cells or sensitize them to the cytotoxic effects of ionizing radiation and / or chemotherapeutic agents. The most promising ones are genes encoding toxic protein inhibitors as well as genes that can convert relatively not toxic prodrugs to their biologically active metabolites (Parentesis et al., 1992; Deonarain et al., 1995). Besides antibody-based and gene therapy approaches, drugand cytokine based approaches to vascular targeting are also possible Several agents that elicit irreversible vascular shutdown selective ly within so lid tumors have been identified. These include flavenoids such as flavone acetic acid (FAA) and more recently DMXAA (Zwi et al., 1994), and tubulin binding agents such as col chi cine and vinblastine FAA was shown to have a broad spectrum of activity against solid tumors (Corbett et al., 1986; Hill et al., 1995). The action of FAA has been attributed in large part to its ability to induce the release of tumor necrosis factor-alpha (TNF a) from tumors in situ (Cliffe et al., 1994; Mahadevanv et al., 1990). In contrast, the tubulin binding agent vinblastine causes little or no increase in plasma TNF-a levels in tumor-bearing mice (Hill et al., 1995). Nevertheless, antivascular effects are a common feature of tubulin

PAGE 20

13 binding agents. Chaplin et al. (1996) assessed the effects of vinblastine and four other tubulin binding agents (dolastatin 10, dolastatin 15, combretastatin Al and combretastatin A4) on tu1nor blood flow. It was shown that all five agents induced a reduction in tumor blood flow range from 50% to 903/o. The mechanisms governing these tumor-selective effects of tubulin binding agents are largely unknown, but it is possible that the inhibition of tubulin polymerization affects endothelial cell shape, leading to thrombus forrnation or changes in petmeability of the endothelium. Despite the reported anti tumor effects, the clinical potential of vascular targeting strategies ultimately will be largely deterrr1ined by the selective toxicity of the reagents. Unfortunately, to date, most of these agents have been reported to only elicit antivascular effects at doses approaching the maximum tolerated dose (MTD) and only in the presence of significant morbidity (Chaplin et al., 1996). For example, vinb la stine and colchicine markedly reduced tumor perfusion and caused necrosis of tumor tissue only when the injected dose was increased to lethal range (Nihei et al., 1999). To fully appreciate the anti-vascular strategy, new agents with a large therapeutic window and improved selectivity are needed. Combretastatin A-4 Disodium Phosphate The tubulin and actin cytoskeleton are critical mediators for a number of important endothelial cellular functions other than the mitotic spindle and chromosome segregation. They facilitate intracellular organization, cell morphology, cell motility and the intracellular transport of molecules from the site of synthesis to the cell surface v ia microtubule motor proteins (A vile, 1992). The

PAGE 21

14 cytoskeleton is a dynamic structure and confo11national rearrangements occur in response to the endothelial cells environment and exposure to mechanical force s (Cucina e t al. 1995). Because of the pivotal role the cytoskeleton plays with respect to cell s hape in endothelial cells, it is not s urprising they have effects on overall va s cular function. One class of tubulin-binding compound s which has received attention in recent year s is the combretastatins. The African bush willow tree Combr e tum Caffrum i s the source of 1 7 natural combretastatins and a further 22 similarly s tructured agent s ha v e been synthe s i z ed thus making four s eries of compounds (named A to D) (Pettit et al. 1987 ; O'Brien 1997). Structurally combretastatin s consi s t of two s ubstituted benzene ring s linked b y a s aturated hydroxy sub s tituted 2-carbon bridge (Figure 1-5) The s e combreta s tatin s s how structural s imilarity to col chi cine and are competitive inhibitors of the binding of colchicine to tubulin (Pettit e t al ., 1989 ; Sackett 1993). They inhibit microtubule activity and interfere with cell growth and proliferation (Pettit et al. 1989). The mechani s m of their binding to tubulin wa s examined indirectly for one of them combretastatin A-4, by evaluating their effects on the binding of radio labeled colchicine to the protein (Lin e t al ., 1989). Studies showed rapid binding of combretastatin A-4 t o tubulin even at O degree s (binding wa s complete at the earliest time s examined) in contra s t to the relatively s low and temperature dependent binding of colchicine It demon s trated that the effectiveness of combretastatin A-4 a s antimitotic agent s appear s to derive primarily from the rapidity of their binding to tubulin

PAGE 22

15 Combretastatin A-4 showed some concentration-dependent cytotoxicity against a variety of human tumors (El-Zayat et al., 1993). As combretastatin A-4 itself is poorly soluble in water, a prodrug, combretastatin A-4 disodium phosphate (CA4DP) was prepared. This extra phosphate group can readily be cleaved by endogenous nonspecific phosphatases (O'Brien, 1997). Combretastatin A-4 binds to plasma protein, and this seems to reduce its activity (Tozer et al., 1999). Therefore, there may be an advantage for using CA4DP rather than combretastatin A-4 beyond its increased solubility. Figure 1-5 shows the chemical structures of combretastatin A-4 and CA4DP. Both agents have been shown to lead to rapid vascular shutdown in several preclinical tumor models (Dark et al., 1997; Li et al., 1998; Horsman et al., 1998). Recent studies continued to demonstrate the vascular effect of CA4DP. When assessed in a rat system, I 00 mg/kg CA4DP caused a very large decrease in tumor blood flow, which by 6 hr, was reduced approximately 100-fold (Tozer et al., 1999). Calculation of vascular resistance revealed some vascular changes in the heart and kidney for which there were no significant changes in blood flow. Using magnetic resonance imaging (MRI), Beauregard et al. (1998) showed that tumor perfusion decreased significantly in the central region of murine tumors after CA4DP treatment and it was consistent with the haemorrhage seen in histological sections. Further preclinical studies have demonstrated that avascular nodules do not appear to be responsive to this agent, providing additional evidence for a vascular mechanism of action (Grosios et al., 1999). The rapid, selective and extensive

PAGE 23

16 damage caused to the tumors by CA4DP has highlighted the potential of the agent as a novel cancer chemotherapeutic agent. More importantly, CA4DP typically produced these effects at concentrations less than one -tenth of the MTD thus offering a wide therapeutic window (Dark e t al., 1997 ; Li e t al. 1998; Horsman e t al., 1998). Combretastatin is now in Phase I clinical trials in UK and US. The current studies are very encouraging (Randal, 2000). A patient with an anaplastic thyroid tumor, which was unaffected by standard therapies and progressed relentlessly, was treated with intravenous infusion of combretastatin. Three weeks after the first infusion, the tumor began to shrink. After six more infusions CT and MRI failed to show any trace of the tumor. Later exploratory surgery confitmed the tumor's total disappearance There also have been other patients whose previously very aggressive cancer s some of them metastatic stabilized after treatment with combretastatin for 24 weeks or more (Randal 2000). In our laboratory, we have used several animal models to study the antitumor efficacy of CA4DP. Treatment with CA4DP has been shown to produce extensive regions of hemorrhagic necrosis in both murine tumors (Li e t al. 1998) and human tumor xenografts. Moreover, as the application of antivascular strategies will need to be given in conjunction with conventional anticancer therapies, we also examined the efficacy of combining CA4DP with ionizing radiation. Such treatment may be required to destroy the remaining rim of tumor cells surviving at the periphery near norn1al tissue vessels. Results from our laboratory indicate

PAGE 24

17 superior antitumor efficacy in the combination of CA4DP and radiation (Figures 1-6 and 1-7). Significance It has been well established that the vascularization of solid tumors is a prerequisite if a clinically relevant size is to be reached. The dependence of the tumor on its induced vessel formation has created a great deal of enthusiasm in specifically targeting the microcirculation in cancer therapy. The central goal of this project was to investigate the potential therapeutic utility of vascular targeting drug CA4DP. Experiments proposed in this project were directed at gaining a better understanding of the mechanism of its action on tumor and endothelial cells and to assess its efficacy in human tumor models either alone or in combination with conventional anticancer therapies. The thesis focused on a model of Kaposi's Sarcoma (KS), since AIDS-KS is a fulminant disease that usually requires aggressive treatment especially when it involves visceral organs, which lacks effective therapies. Irradiation, systemic chemotherapy, and interferon (IFN-a), though helpful, are administered primarily for symptomatic relief and to prevent disease progression. Conventional treatments do not prolong survival, and their clinical effectiveness is not satisfying (Sung et al. 1997). Continued pursuit of more effective agents clearly is needed. Given the possible future clinical impact of CA4DP, we believe it to be highly worthwhile to investigate effects of this agent in a preclinical model of KS. The specific aim s of this project are as follows:

PAGE 25

18 Specific aim 1. To study the different responses of nor1nal and neoplastic cells to CA4DP by examining the effects of CA4DP on cell cycle, cell survival, and apoptosis Specific aim 2. To gain further insight into the mechanisms underlying the anti vascular action of CA4DP by examining the effects of CA4DP on human microvascular endothelial cells (HMVEC-L). Specific aim 3. To examine the antitumor efficacy of CA4DP by applying histological approaches and by measuring its cytotoxic action in KS xenografts. Specific aim 4. To assess the potential in situ therapeutic benefit of combining CA4DP with conventional anticancer therapies by evaluating the effects of such treatments in KS xenografts.

PAGE 26

. . , wo rmal , l, . : ~ .' . ., . Temporary Occlusion 19 AV Shunt . Blind Ends Break in Vessel Wall Figure 1-1. Diagram showing the principal differences between the vasculature of no11nal and malignant tissue. Whereas no1n1al tissues have relatively unifor1n and well-ordered blood vessels that are sufficiently close together to oxygenate all of the tissue, blood vessels in tumors are tortuous, have imcomplete vessel walls, have sluggish and irregular blood flow, and have regions of hypoxia between the vessels (Brown and Giaccia, 1998)

PAGE 27

C 0 ..(,) 1 ca at .1 CD C > c= ::, .01 en 00 Aerobic Hypoxic 10 20 30 40 Dose {Gy) 20 DNA-+DNA Mechanism: fd DNA-H(Chemical Restitution) Cell Death DNA-00-+ DNA-OOH (Damage Fixation) Figure 1-2. Typical survival curves to ionizing radiation for mammlian cells under aerobic and hypoxic conditions. Most mammalian cells, irrespective of genetic background, exhibit a s urvival curve with an Initial ''shoulder'' region followed by exponential cell killing. The oxygen enhancement ratio is typically 2.5-3.0. The dotted vertical line at 14 Gy shows the > 2 logs difference in cell kill for aerobic and hypoxic cells at thi s dose. Also shown is the mechanism for the greater sensitivity of aerobic cells as compared to hypoxic cells, ionizing radiation produces a radical in DNA which can be either chemically restituted by donation of hydrogen from nonprotein sulfhydryle (-SH) in the cell or, in the presence of oxygen converted int o per1nanent damage that increases the probability of cell death (Brown, 1999)

PAGE 28

21 C 0 0 l! II. C, C > ::I u, 0 '... . ' ... . '. 50 . ,, 100 I" ' 150 Distance from Capillary (pm) Figure 1-3. Left, a diagrarnmatical representation of part of a tumor cord surrounding a capillary showing decreasing oxygen concentration as well as decreasing cellular proliferation and drug concentration as a function of distance from the capillary. Right, the considerations on the left lead to the prediction that cell killing by radiation or most anticancer drugs will be reduced as a function of distance from the capillary. The combination of standard treatment with a hypoxic cytotoxin would be expected to overcome the proble111 of hypoxic cells by producing a relatively unifo11n cell profile of cell killing as a function of distance from the capillary (Brown and Siim, 1996).

PAGE 29

ANTI ANGIOOENESIS VASCULAR MEDIATED TUMOUR THERAPY VASCULAR TARGETING ..., ... -22 I INHIBIT I TAF PROOUC110N PRfVB\lT BASE>JENT MEMBRANE DISSOWTION INHIBIT ENOOlHB.IAL caL. MIGRATION INHIBIT ENOOlHEUAL ca.L OMSION INHIBIT ANASTAMOSIS OF BUDS KILL ENDOlHELIAL CfllS a .. PREVENT EXPANSION OF _. PREVENT FURTHER TUMOUR VASCULAR NETWORK EXPANSION PREVENT ESTABLISHMENT OF METASTASES DAMAGE ,. FRAGILE ........_ f!l WAUS CAUSE B HAEMORRHAGE -+ fS .............. INCREASE 1.. lmRAVASCULAR / w z I -I a: !5 P RESSURE .... L:_cihi....;E:..;;.ON;;..;.~;... 000 ___, -._1 _~_~_1S_-'I /.___, -....... I _c_
PAGE 30

23 a. b. OH 0 II -ONa I ONa Figure 1-5. The chemical structures of combretastatin A-4 (a) and combretastatin A-4 disodium phosphate (CA4DP) (b)

PAGE 31

C 0 ... u l! .... 0) C > :::, en .. 0 E :::, t10 8 10-4 T ., l T l 10 5 0 20 40 24 ... J. T l 60 Combr Radiation + Combr 80 T D 1 .L 100 Combretastatin A-4 prodrug dose (mg/kg) Figure 1-6. Tumor cell killing in KHT sarcomas treated with increasing doses of CA4DP either alone of 1 hr after irradiating the tumors with a 15-Gy dose of radiation. Data are the mean SE of 6-12 tumors

PAGE 32

25 Radiation Radiation + Combr 10 1 .L C 0 .., u ca '10 2 .... 0, C > T Ei > 'l ::::, u, 10 3 '0 C E ::::, I10 4 T l 1 oS .L 0 5 10 15 20 25 Radiation dose (Gy) Figure 17 Tumor cell survival in KHT sarcomas treated with a 100 mg/kg dose of CA4DP 1 hr after a range of doses of radiation. Results are the mean SE of 3 experiments.

PAGE 33

CHAPTER2 COMPARISON OF THE DIFFERENT RESPONSES OF NEOPLASTIC AND NORMAL CELLS TO CA4DP IN VITRO Introduction The survival and growth of so lid tumor deposits depends critically on the development of a blood vessel network. The functioning vascular network in tumors provides the tumor cells with oxygen and nutrients, and enables removal of the toxic waste products of cellular metabolism. The fact that the production of several angiogenic growth factors can be up regulated by physiological parameters including low oxygen or glucose and acidic pH, which are associated with vascular insufficiency, provides a logical rationale for the strong angiogenic stimulus in malignant tissue (Chiarotto and Hill, 1999; Namiki et al., 1995). The continued proliferation of tumor cells will result in deprivation of oxygen and glucose and production of acidic metabolites thus stimulating the development of additional neovasculature (Siemeister et al., 1998). The new vessels facilitate the further expansion of the tumor cell mass providing a perpetual loop. Clearly, the cycle can be interrupted by killing or inhibiting the growth of the tumor cells However, interventions that compromise the function or growth of the tumor neovasculature can also be effective. Therefore, tumor blood vessels represent a central target for the development of new approaches to cancer therapy. 26

PAGE 34

27 The majority of research work in this area has been focused on preventing the growth of new tumor vessels, so called anti-angiogenesis. Many agents that were identified as anti-angiogenic target at least one of the several stages involved in new vessel formation, i e basement membrane degradation, endothelial cell migration, endothelial cell proliferation and tube formation (Fan et al., 1995). In contrast to the focus on anti-angiogenic approaches to therapy, there has, until recently, been relatively little effort afforded to the identification and development of therapies that specifically compromise the function of the existing neovasculature in solid tumors. Interestingly, evidence for the therapeutic potential of vascular targeting approaches existed a lot earlier than recent studies. It had been reported over 150 years ago that occasionally solid tumors in the clinic could be eradicated when their circulation was interrupted either by torsion of the vascular pedicle or by thrombosis of a major feeding vessel (Walsh 1844) The pivotal role of tumor vasculature and the effects of its selective destruction were also highlighted by Woglum over 75 years ago (Woglum, 1923). However, it is only within the last decade that research has been focused seriously on the development of therapies that specifically target and damage tumor neovasculature. Despite this limited development time many promising approaches, including drug-, antibodyand gene therapy based strategies have emerged. Many tubulin-binding agents have been shown to have antivascular effects. The vinca alkaloids, for example vincristine and vinblastine, could induce vascular damage at doses close to the MTD (Baguley et al. 1991; Hill et al.,

PAGE 35

28 1995). However the y also have s hown direct cytotoxic effects against a variety of tumor cells (Zhou and Rahmani, 1992 ). Although several studies have shown the anti vascular effects of CA4DP in several clinical models (Dark et al., 1997; Li et al., 1998; Horsman et al., 1998), it is not clear whether the activity of CA4DP is selectively against vascular endothelial cells or whether it also acts on tumor cells. In order to shed light on this in the present studies, we examined the responses of human tumor cells (KSY-1 and A549) human endothelial cells (HMVEC-L) and human fibroblasts (FG 1522) to CA4DP in vitro. The tubulin and actin cytoskeleton are critical mediators for mitotic spindle formation and chromosome segregation. Because CA4DP is a tubulin-binding agent, studies were initiated by examining its effect on cell cycle distribution in different cell type s Later we examined the cytotoxicity of CA4DP and whether i t could induce apoptosis in both neoplastic and normal cells. Material and Methods Cell Culture Two human tumor cell lines, Kaposi's Sarcoma cell line KSY-1 (ATCC Rockville MD) and lung cancer cell line A549 (ATCC Rockville, MD) were used in the study KSY-1 cells were cultured in positively charged Cell + tissue culture flasks from SARSTEDT (Newton, NC). Cell+ TC flasks provide a positively charged surface for difficult to-grow adherent cell cultures. Both KSY1 and A549 cells were grown in RPMI 1640 medium (Gibco BRL Grand Island NY) with I 00 units / ml penicillin 0.1 mg/ml streptomycin, and 10 % fetal bovine serum (FBS).

PAGE 36

2 9 Human microvascular endothelial cells of the lung (HMVEC-L), obtained from Clonetics (San Diego CA) were grown in EGM-2-MV medium (Clonetics) containing 5% FBS and supplements (0.1 /o hEGF, 0.4% hFGF-B, 0.1 % VEGF, 0.1 % Ascrobic Acid 0.04% Hydrocortisone, 0.1 % Long R3-IGF-1, 0.1 % Heparin and 0.1 % GA-1000). Clonetics Trypsin and Trypsin Neutralization Solution (TNS) were used for subculture Human skin fibroblasts FG 1522 (from Dr. Hei's lab, Columbia University) were grown in Dulbecco's MEM Medium with 10 % FBS and 25 g/ml dose of gentamycin. Drug Preparation CA4DP (Oxigene Inc., Lund, Sweden) was dissolved in 5% sodium carbonate at a concentration of 10 mM and then subsequent ly diluted in 0.9% saline and culture medium immediately before use. Cell Cycle Studies A549, KSY-1, HMVEC-L and FG1522 cells were plated in 60 mm dishes at 2x10 5 cells / dish On day 3, the cells were exposed to various concentrations of CA4DP for a period of 2 hr. The dishes were then washed with PBS and replenished with fresh media. 22 hr later, the cells were trypsinized, counted, and fixed in 50 % ethanol overnight. Before analyzed by F ACS, the cells were treated with 1 mg/ml RNase (in PBS) for 30 min. The samples were then washed with PBS twice and resuspended in 25 mg/ml propidium iodide {Pl) (in PBS) at a volume of 1 x I 0 6 cells / ml. The cell s were stained with PI in darkness for at least 10 min and were then analyzed by FACS for cell cycle distribution on a Becton

PAGE 37

30 Dickinson flow cytometer made available through the University Core Facility for Flow Cytometry at the University of Florida. Cell Viability Studies Cell viability was determined using clonogenic cell survival assay. Briefly, A549, KSY-1, HMVEC-L and FG1522 cells were plated in 60 mm petri dishes at 1 xl 0 5 cells / dish. On day 3, the cells were exposed to various concentrations of CA4DP for a period of 2 hr or 24 hr. The dishes were then washed with PBS twice and the cells were trypsinized and counted. The cells were then mixed with 10 4 lethally irradiated cells and plated into 60 mm petri dishes. After 2 weeks of incubation at 37 C, colonies of 50 or more cells were counted with the aid of a dissecting microscope Cell surviving fractions were calculated as the ratio of colonies counted in treated versus untreated group. To assess the effect against quiescent cells, the cells were first grown in 60 mm petri dishes and once confluent were treated and assayed as described above. Apoptosis Four cell lines each grown in 2-well chamber slides were treated with 50 M CA4DP for 2 hr. The treated cells were fixed in 4% formaldehyde solution immediately after treatment or 22 hr later for TdT-mediated dUTP Nick-End labeling (TUNEL) assay. Basically, the cells were permeabilized in 0.2% Triton X-100 solution for 5 min. DNA strand breaks were then labeled with fluorescein12-dUTP in TdT incubation buffer at 37 C for 1 hr. The samples were then counterstained with 1 g/ml diamidine-2-phenylindole (DAPI), which binds to the A-T-rich regions of DNA. Localized green fluorescence of apoptotic cells

PAGE 38

31 (fluorescein-12-dUTP) in a blue background (DAPI) was detected by fluorescence microscopy The percentage of apoptotic cell was obtained by dividing the number of cells with green fluore scenc e by the total number of cells with blue fluorescence A minimum of 300 cells were counted for each condition. Res u lts To evaluate the effect of CA4DP on cell cycle distribution, proliferating nor1nal and neopla stic cells were exposed to a range of do ses of CA4DP for 2hr, and DNA profile s were analyzed by F ACS 22 hr later. Re s ults showed that CA4DP could effectively block all cell types in the 0 2 / M phase beginning at the do ses ~ 0.05-0.1 M (Figures 2 -1 and 2-2) Increasing the drug do se to higher concentrations cau sed either a s light or no increa s e in the percentage of cell s in the 0 2/ M phase in different cell types (Figure 2-2). It was also observed that even at high concentrations C A4DP arrested only 30-40% HMVEC-L and FG 1522 cells in the 0 2 /M pha s e whereas in A549 and KSY-1 cells there were almost complete 0 2/ M block s C lonogenic cell s urvival assay was u s ed to evaluate cell viability in each cell line treated with a range of do ses of CA4DP. A 2-hr treatment, administered to exponentially growing cell populati o n s, was found to be ineffective aga in s t FG 1522 and A549 cells and s howed only slight killing of KSY 1 cells (Figure 23; a-c) In contrast, under those conditions, CA 4DP di s played s ignificant do se dependent activity against prolifer a ting HMVEC-L (Figure 2-3d). Importantly thi s effect was s peci fic to proliferating endothelial cells: quiescent HMVEC-L cells s howed no re spo n se to CA4 DP even at high do ses (Figure 2 -3d) With

PAGE 39

32 prolonged drug exposure (24hr), CA4DP caused a clear dose-dependent cell kill in both A549 and KSY-1 cells, but still showed no toxicity to FG 1522 cells. (Given the exquisite sensitivity of proliferating HMVEC-L to CA4DP extended drug exposure times were not evaluated in this cell type.) To gain further insights into the cytotoxic mechanisms of CA4DP we also used TUNEL assay to examine the induction of apoptosis in the different cell types treated with this agent (Figure 2-4 ). The results showed that 2-hr treatment with 50 M CA4DP assessed immediately or 22 hr later led to very little apoptosis in fibrobla s ts compared to the control group (0.6 and 1.2%, respectively). A549 cells showed a similar response; there was no induction of apoptosis immediately after CA4DP treatment, and 22 hr later, 4.6% of cells were detected as apoptotic cells. CA4DP induced more apoptosis in KSY-1 cells than in A549 cells. In thi s cell line the 0.7% apoptotic index in untreated KSY-1 cells rose to 2.8 % immediately after treatment and to 7.9% 22 hr later. The results also showed that CA4DP induced the highest level of apoptosis in HMVEC-L cells. Immediately after a 2-hr CA4DP exposure 7.5% of cells underwent apoptosis. This number increased to almost 20o/o 22 hr later. Discussion The main aim of this study was to investigate whether CA4DP demonstrated any cell type specificity / selectivity by comparing its effects in normal and neoplastic cell types. Initial studies with F ACS analysis showed that non-toxic concentrations of CA4DP caused arrests of the cells in the G 2 /M phase in all the cells examined, indicating a disruption of mitosis due to a functional deficiency of

PAGE 40

33 the tubulin apparatus which disables the cells to divide their chromosomes properly. More importantly, the doses for CA4DP to initiate the cell cycle effect in different cell types were found to be within the same range ( ~ 0.05-0.1 M). This suggested that CA4DP could bind to the tubulin in different types of cells with the same efficacy, thus resulting in cellular tubulin disruption and similar G 2 arrest effects at the same drug concentration. It was also observed that while CA4DP could cause almost a complete G 2 -arrest in A549 and KSY-1 cells only 30-403/o ofHMVEC-L and FG1522 cells were blocked in the G 2/ M phase even at high CA4DP concentrations. We believe that this arises because unlike the tumor cell lines where the majority of cells are actively proliferating, a significant proportion of the normal cells may never enter the cell cycle. To attempt to examine this possibility we are currently utilizing the antibody against Ki67 a nuclear antigen for proliferating cells, and BUdR labeling to determine the fractions of proliferating cells in these cell populations. Although CA4DP exerted similar effectiveness in terms of cell cycle arrest in both neoplastic and normal cells, results from the clonogenic cell survival assay showed significantly difference in the responses of the different cell types to this agent. CA4DP showed dose-dependent activity against proliferating endothelial cells with short drug exposure (2 hr). For example, a 20 M drug concentration reduced viable HMVEC-L cells to less than 10% compared to the untreated cells. If the endothelial cells were quiescent during the drug exposure, no significant drug toxicity was observed. In contrast to the endothelial cells, 2-hr exposure to CA4DP caused no change in cell viability in proliferating FG1522 and A549

PAGE 41

34 cells, and only a slight toxic effect in KSY 1 cells These observations implied a selective toxicity of CA4DP toward proliferating endothelial cells With prolonged drug exposure (24 hr), CA4DP did show toxicity against tumor cells (both A549 and KSY-1 cells). But over the dose range evaluated FG 1 522 cells were still not affected. We believe that the specific kil l ing of dividing endothelial cells observed after CA4DP treatment might be critical for the in vivo action of this drug. Parenthetically the observation that normal cells 1nay be less susceptible to CA4DP treatment may explain the low level of toxicity of CA4DP observed in the preclinical studies (Dark et al., 1997) Apoptosis is an active and gene-directed mode of cell death, involved in embryological development, organ involution, and the response of both nor1nal and transformed cells to cytotoxic agents (Gorczyca et al., 1993). It is characterized by rapid nuclear and cytoplasmic condensation and cellular disintegration into apoptotic bodies (Kerr et al. 1972). Recent work has shown that apoptosis is controlled by a complex network of positive and negative signals, which originate either from specific gene products or from the extracellular environment (Stewart, 1994). The present studies showed that 50 M CA4DP induced the highest level of apoptosis in endothelial cells compared to tumor cells and fibroblasts, demonstrating again the selective toxicity of CA4DP against endothelial cells. It was observed that CA4DP caused cell cycle arrest in HMVEC-L starting at very low dose (0.05 M). However, CA4DP started to show toxicity against HMVEC-L only at doses higher than I M. The different effects of CA4DP

PAGE 42

35 observed at low and high doses is because at low doses, CA4DP-induced G 2 arrest in the endothelial cells was reversible. Arrested cells recovered at 4872 hr after 2-hr CA4DP treatment ( data not shown). Only at higher doses (~ 1 M), CA4DP caused irreversible G 2 -arre s t in HMVEC-L thus leading to subsequent clonogenic cell death and cell apoptosis Taken together the present results demonstrated that CA4DP displayed a significantly higher cytotoxicity toward proliferating endothelial cells than to tumor cells and normal human fibroblasts. This may well explain the selective antivascular effects of CA4DP observed in preclinical studies in vivo. Still the fundamental question of''Why CA4DP has such selectivity toward endothelial cells ?'' remain s largely unknown. Nevertheless the selectivity of the effects are highly encouraging and clearly warrant more detailed study of the cellular and molecular mechanism s involved as well as the continued investigation of the therapeutic potential of the drug as an anti vascular agent.

PAGE 43

a. b. 0 cc ,., .. 0 "' ... 0 0 "' CV .Co E .., .::, z 0 n., ~o 0 50 36 GO.G1 61.96% S-Phase 28 39% G2-M 9 65% 100 1 0 Ctla n ne ls 1 0 0 150 Cha nn ft lS 200 GO.G1 2 15% S-Phase 5 90% G2-M 9195% 200 2 0 2 0 Figure 2-1. Analysis of cells cycle distributions in A549 cells by F ACS. a) Untreated A549 ce l ls; b) A549 cells treated with 0.1 M CA4DP for 2 hr and fixed for F ACS analysis 22 hr later

PAGE 44

37 CA4DP cell cycle effect 100 90 O KSY-1 80 HMVECL A549 0 70 e FG1522 u, 60 Q) CJ Q) 50 u, ca .I:. 40 Q. :E 30 "'c,., (!) 20 10 0 0.001 0 01 0 1 1 10 100 Drug dose (M) Figure 2-2. G 2 -arre st ca u se d by CA4DP tre atme nt in FG1522, A549 KSY-1 and HMVE C -L cells. The cells were treated with a range of do ses ofCA4DP for 2 hr and fixed for F ACS analysis 22 hr later

PAGE 45

a. C 0 ;: u .... C) C > C: :, C/1 QI (,) C. C -B u Ill ... C) C ~ C: :, C/1 = 1 10 KSY-1 -CA4DP 2 hr o 24 hr 10 2 +--~---.-----T----ir-----, 1 10 0 001 0 01 0 1 1 Drug Dose (M) FG1522 CA4DP 2 hr C 24hr 10 10 2 +----.---T-----.-----, 0 01 0 1 1 10 100 Doses (M) 100 38 b. d. C 0 ., u 01 C C: :, C/1 QI (.) C 0 .. u Ill ... .... C) C > > .. :, "' QI u A549CA4DP 1 II -i 1!! T / ... 1 1 J. .i.\T 2 hr 24 hr 1\ T 1\ ... "'T l 10 2 -'---------------, 0.001 0.01 0 1 1 10 1 00 Drug Dose (M) HM VEC-L CA4DP 1 Y--..-
PAGE 46

39 Apoptosis 50 M CA4DP 20 7 Control < 2 hr treatment < 15 ,,_ 0 ,.... ,,,, = ,,, IX) 2 hr treatment+ 22 recovery 9 [> > Q) 0 0 10 < '>., .., ) 0 X .., -r C. 0 C. <( 5 ,X I/ "' I/ r/ X > )< X ,1' )< 0 v ,. t/. X K ,. /' )< '\, -,. V X < > V J< I / .x., < ; :'> / .,, l )c J< I / .x., '/ >< /, r-. ,. / .,, Fibroblast A549 KSY-1 HMVEC-L Figure 2 -4. Effect o f C A4DP on cell apoptosi s in fibrobla s ts FG1522, A549 KSY-1 and HMVE C -L cells. The cells were treated with 50 M CA4DP for 2 hr and stained for T UNE L assay either immediatel y after treatment or 22 hr later Data are the mean S E o f three independent experiments

PAGE 47

CHAPTER3 EFFECTS OF CA4DP ON HUMAN MICROVASCULAR ENDOTHELIAL CELLS IN VITRO Introduction More than 10 1 2 endothelial cells line the inside of blood vessels, covering a surface area of more than 1000 m 2 (Jaffe, 1987). Not only do they form the structural basis of blood vessels and provide an antithrombogenic surface but they also contribute to numerous metabolic functions including coagulation and thrombolysis, control of vasotonus and antigen presentation, as well as basement membrane and growth factor synthesis (Pear s on 1991 ). Endothelial cells in the adult form a highly heterogenous cell population that varies in different organs These cells normally are quiescent. After induction of angiogenesis by angiogenic cytokines however endothelial cells can proliferate as rapidly as bone marrow cells which have a turnover time of 5 days (Folkman, 1995) Angiogenesis is a mode of endothelial cell activation that induces distinct phenotypic changes Triggered by paracrine and autocrine mechanisms it enables endothelial cells to break away from preexisting vessels to enter a complex morphogenetic cascade that will ultimately lead to the formation of new vessels with mature endothelial cells. Endothelial cell s released in culture from growth arrest and allowed to migrate change their adhe s ive properties their s urface glycosylation pattern, their cytokine production and their growth factor receptor expression pattern, as well 4 0

PAGE 48

41 as their proteolytic balance (Augustin-Voss et al., 1992; Weich et al., 1991). Consequently, since angiogenesis is a developmentally regulated process that is down-regulated in the healthy adult ( except for the female reproductive system), inhibition of angiogenic-specific cell functions might be useful for targeting new vessels during tumor growth. One key in the development of treatment strategies is to identify differences that exist between the tumor vasculature and normal tissues. As mentioned before, the blood vessels in tumors are proliferating more rapidly than those in normal tissues (Denekamp, 1990). Thus, targeting features of proliferating endothelium, or even newly formed vasculature could achieve some selectivity. Another well established feature of tumor blood vessels is that, unlike those in normal tissues, they can be subjected to low oxygen tension (Chaplin et al., 1987; Hill et al., 1996). A third obvious feature is that tumor endothelium is located adjacent to the malignant tumor cells, which in turn can alter endothelial cell characteristics. Exploiting the changes that such microenvironmental stimuli induce both in endothelial cell function and gene expression will undoubtedly provide the key to achieving effective and highly selective approach to targeting endothelial cells in tumors. Vascular targeting approach aims to destroy the tumor vessels which contain rapidly proliferating endothelial cells. Antivascular effect is a common feature of tubulin binding agents. The original interest in the vascular-damaging effects of such agents was stimulated by studies with colchicine reported in the 1930s and 1940s. Studies clearly demonstrated that colchicine preferentially damaged newly

PAGE 49

42 formed capillaries in tumors with the consequence of inducing hemorrhage and extensive necrosis (Ludford, 1945). Activity was noted in 1nany different experimental tumor s ystems but significant effects were only achieved at doses approaching the MTD. Later studies indicate that other tubulin binding agents such as vincristine and vinblastine at doses approximating the MTD could also induce vascular damage (Baguley e t al ., 1991; Hill e t al 1995) Recent studies focused on evaluatin g a number of inhibitors of tubulin polymeri z ation for their ability to induce va s cul a r damage in tumors (Chaplin e t al. 1996) aiming to identify agent s with a superior therapeutic index for their vascular effects . Fortunately combretastatin A-4 and CA4DP have been identified as agents that can induce va s cular damage in tumor s at do s es much le s s than the MTD (Dark et al 1997; Beauregard et al 1998). As with all approaches to cancer therapy vascular targeting is only realistic if significant selectivity between tumor and normal tissue response can be achieved. In v itro studies from Chapter 2 revealed marked cytotoxic effects of CA4DP again s t proliferating but not quiescent endothelial cell s A549 KSY-1 and FG 1522 cell s were much less affected by CA4DP than the HMVEC-L cells Why proliferating endothelia l cell s show such sensitivity to the in v itro effects o f CA4DP is not known yet. Although the clinical potential of CA4DP has been recogni z ed by its recent Phase I clinical studie s in UK and US, further information is required regarding its mechani s m of action In light of the potent antivascular effects o f CA4DP the present s tudies were undertaken to gain insight into the mechanism( s ) of action of CA4DP by study i ng its activity in HMVEC-L

PAGE 50

43 Material and Methods Cell Culture HMVEC-L cells, obtained from Clonetics (San Diego, CA), were grown in EGM-2-MV medium (Clonetics) containing 5% FBS and s upplement s (0.1 o/o hEGF, 0.4% hFGF-B, 0.1 o/o VEGF, 0.1 % Ascrobic Acid, 0.04% Hydrocortisone, 0.1 % Long R3-IGF-1, 0.1 % Heparin and 0.1 % GA-1000). C lonetic s Trypsin and Trypsin Neutralization Solution (TNS) were used for subcu lture Drug Preparation CA4DP (Oxigene Inc., Lund, Sweden) was dissolved in 5% sodium carbonate at a concentration of 10 mM and then s ubsequently diluted in 0.9% saline and culture medium immediately bef ore use. Indirect lmmunofluorescence The intracellular distribution of microtubules following drug treatment was deterrr1ined using indirect immunofluorescent staining (Giannakakou et al., 1998; Woods et al., 1995). HMVEC-L cells were plated in 35 mm dishes 1 day prior to treatment with CA4DP. Both treated and untreated cells then were fixed in 1: 1 methanol / acetone for 5 min at room temperature and washed with PBS. Incubation with the primary anti-P-tubulin MAb for 70 min was followed by a 50min incubation with the seco ndary flourescein-conjugated goat anti-mouse IgG antibody. All antibody incubations and washes were perfor1ned at room temperature. Morphological analysis then was performed by fluorescence microscopy.

PAGE 51

44 Tubulin Polymerization Assay HMVEC-L cell s grown in 24-well plates were treated with 5 M CA4DP for a s pecified time After washing each well twice with 1 ml PBS (Ca 2 + free) the cells were ly s ed at 37 C for 5 min in the dark with 100 l ofhypotonic buffer (20 mM Tris-HCl pH 6 .8, 1 mM MgC1 2 2 mM EGT A 0.5 % Nonidet P-40 2 mM PMSF, 200 U / ml Approtinin 100 g/ ml soybean tryp s in inhibitor 5mM e-amino caproic acid and 1 mM benzamidine) (Giannakakou e t al. 1998). The wells were scraped and the ly s ate s transferred to 1.5 ml Eppendorf tubes. Each well was rinsed with an additi o nal 100 l of the hypotonic buffer and this volume was pooled with the lysate Following a brief but vigorou s vortex the samples were centrifuged at 14 000 rpm for 10 min at room temperature. The 200 l s upematants containing soluble or unpolymerized (cytosolic) tubulin were carefully s eparated from pellets and transferred to separate tube s The pellet s were resu s pended in 200 l ofhypotonic buffer containing 10 mM Tri s, pH 7 5 1.5 mM MgCl 2 10 mM KCl 0.5 % Nonidet P-40 and the protease inhibitors described above Ea c h tube containing either the soluble or the polymerized fraction was mixed w ith 70 l of 4 x SDS-PAGE sample buffer (0.3 M Tri s -HCl pH 6.8, 45 % glycerol 20 % P-mercaptoethanol 9.2 % SDS and 0.04 g/ 100 ml bromophenol blue) and heated at 95 C for 5-10 min ; 20 l aliquot s of each sample then were an a l yz ed by SDS-P AGE on a 12 % resolving gel and 3 % s tacking gel Following immunoblotting using a primary anti-P-tubulin MAb the s ignal was quantitat e d b y den s itometry.

PAGE 52

45 Cell Migration Assay Confluent cultures ofHMVEC-L cells were prepared in 24 well plates. A scrape wound of uniform width (2 mm) was produced in the monolayers prior to treatment with CA4DP or drug vehicle. CA4DP exposure was for a period of 2hr at concentrations of 0.1 1, and 10 The drug was removed and 24 and 48 hr later each well was stained with 300 g/ml neutral red solution for 30 min to help visualize and locali z e the cells. The number of cells entering the denuded area was counted using a phase microscope (Braunhut e t al 1996) Apo p tosis HMVEC-L cell s grown in 2-well chamber slides were treated with CA4DP at 1-50 M for 2 hr. After a specified time, the treated cells were fixed in 4% formaldehyde solution for TUNEL assay. Basically the cells were permeabilized in 0.2% Triton X-100 solution for 5 min. DNA strand breaks were then labeled with fluorescein-12-dUTP in TdT incubation buffer at 37 C for 1 hr. The samples were then counterstained with 1 g/ml DAPI which binds to the A-T-rich regions of DNA. Localized green fluorescence of apoptotic cells (fluorescein-12-dUTP) in a blue background (DAPI) was detected by fluo1escence microscopy The percentage of apoptotic cell was obtained by dividing the number of cells with green fluorescence by the total number of cells with blue fluorescence. A minimum of 3 00 cells were counted for each condition

PAGE 53

46 Cell Detachment HMVEC-1 cells were plated in 60 mm petri dishes and exposed to 0.1-50 M CA4DP for 2 hr while in logarithmic growth or plateau phase. Cell detachment rates were determined 22 hr after drug treatment by counting the detached cells in the media electronically using a Coulter counter. Results Initial studies examined the cellular morphology and tubulin organization of CA4DP-treated cells by immunofluorescence (Figure 3 -1 ). Untreated HMVEC-L cells demonstrated a well-organized tubulin network with individual microtubule fibers clearly visible (Figure 3-1 a). Treatment with 10 M CA4DP for a 4 hr period led to a disrupted network of microtubules, which appears as a diffuse s taining pattern in the treated cells (Figure 3-1 b ). When HMVEC-L cells were treated with 10 M CA4DP for 24 hr the tubulin network disruption was even more evident (Figure 3-lc). In some treated cells, tubulin components distributed only in regions around the nucleus which may account for why cells rounded up and detached from the monolayer (below and Figure 3-6). In order to quantitate the effects of CA4DP on tubulin polymerization, changes in both so luble and polymerized tubulin levels in HMVEC-L cells treated with CA4DP were examined. Microtubules are an integral part of the cytoskeleton of eukaryotic cells and are composed of two major soluble proteins aand tubulin. Tubulin exists in cells in two for1ns, soluble (unpolymerized) tubulin and polymeri zed (cytoskeletal) tubulin. There exists a dynamic equilibrium between

PAGE 54

47 tubulin polymerization and depolymerization to maintain the normal function of cells. CA4DP possesses a high affinity for the colchicine binding site on tubulin and results from SDS-P AGE analysis showed a time-dependent depolymerization oftubulin when HMVEC L cells were treated with 5 M CA4DP (Figure 3-2). Polymerized tubulin comprised ~50% of the total tubulin in HMVEC-L cells before CA4DP exposure. It decreased to 1 3% 24 hr after CA4DP treatment. It should be noted that the effect of CA4DP on tubulin polymerization occurred very rapidly with decreased polymerized tubulin levels being detected within 2 hr after drug exposure. As non-toxic doses of CA4DP were found to inhibit cell cycle progression (Chapter 2), the effect of CA4DP on endothelial cell migration also was investigated Endothelial cell migration is an essential endothelial cell function in both angiogenesis and the wound healing process. In angiogenesis, endothelial cells are required to proliferate and migrate in response to angiogenic stimuli (Folkman, 1986). Upon wounding, the cytoskeletal network in a quiescent cell undergoes dramatic redistribution and reorganization to facilitate the directional movement of the cell into the injured area (Braunhut et al., 1996). To mimic this process, a scrape of uniform width (2 mm) was produced in HMVEC-L cell cultures grown to con-fluency. Cells migrating into the denuded area were counted 24 and 48 hr later. Results revealed that CA4DP-treated HMVEC-L cells exhibited a reduced capacity to migrate as seen by the 10 40% reduction in the number of cells detected in the denuded area (Figure 3-3). In addition, observations made by phase microscopy also showed that, compared to CA4DP

PAGE 55

48 treated cells, untreated cells could penetrate a greater distance into the denuded area over the same time period. Although studies using the clonogenic cell survival assay (Chapter 2) revealed a concentration-dependent activity of CA4DP against proliferating HMVEC-L cells (Figure 2-3), this cell survival assay can not distinguish between whether the effects of CA4DP result from cell death by necrosis or apoptosis To shed light on this issue, HMVEC-L cells were treated with 5 and 50 M doses of CA4DP for 2 hr and the induction of apoptosis was assessed at various times after treatment by TUNEL assay Under these conditions, HMVEC-L cells showed clear evidence of nuclear condensation and fragmentation, which are characteristic of apoptosis. When quantified, a time-dependent increase of apoptotic cells in the 50 M-CA4D P-treated cell population from ~ 7o/o immediately after treatment to ~20o/o 22 hr later was noted (Figure 3 -4a) At the lower dose ( 5 M), increased apoptosis in HMVECL cells could be detected 22 hr after drug treatment. Figure 3 -4b also showed that the CA4DP-induction of endothelial cell apoptosis was clearly dose dependent. Compared to only 2% apoptotic cells in the untreated cell population, HMVEC-L cells treated with 50 M CA4DP for 2 hr exhibited apoptosis levels ~ 10 fold higher 22 hr after treatment. The result was consistent with previous studies in which DAPI staining was used to visualize the apoptotic cells with nuclear condensation and fragmentation (Table 3-1).

PAGE 56

49 As cells undergoing apoptosis tend to detach from culture dishes, the number of cells in the supernatant after CA4DP treatment were quantitated using a Coulter counter. The results showed a dose-dependent increase in cell detachment when proliferating, but not quiescent, HMVEC-L cell cultures were treated with CA4DP, again demonstrating the selective activity of CA4DP (Figure 3-5). This dose dependent endothelial cell detachment correlated closely with the number of apoptotic cells found on the monolayer (Figure 3-5 vs Figure 3-4b ). Furthermore, when the detached cells were examined by TUNEL assay, more than 90% were found to be apoptotic. Discussion CA4DP is a tubulin-binding agent which has been shown to produce extensive hemonhagic necrosis in both rodent and human tumor models (Dark et al., 1997; Li et al., 1998; Horsman et al., 1998). Indeed animal studies with this agent used alone or in combination with traditional anticancer therapies have been sufficiently promising (Li et al., 1998; Chaplin et al., 1999) to initiate Phase I clinical trials with CA4DP in both the UK and US. The preclinical investigations from our laboratory and other groups suggested that the agent's selective activity against proliferating endothelial cells may be of particular importance (Dark et al., 1997). To explore this further, the present studies were undertaken with HMVEC-L cells in vitro with the aim of characterizing the effects of CA4DP and delineating more directly the role of the endothelial cell in its mechanism of action.

PAGE 57

50 Strategies aimed at targeting the tumor vessel network, particularly antiangiogenic therapies, have received considerable attention as alternative cancer therapies (Schweigerer, 1995; Denekamp, 1990; Denekamp, 1993). Damaging the tumor vessels directly and selectively with anti vascular agents necessitates the existence of key differences between the vessels comprising tumors and nor1nal tissues. The much higher proliferative index of tumor associated endothelial cells as compared to those found in norrr1al tissues provides such a difference (Denekamp, 1993 ). The development of drugs which are particularly toxic to dividing endothelial cells aim to exploit this difference and offer the possibility of significant treatment selectivity. Earlier results (Chapter 2) demonstrated that CA4DP acts selectively against proliferating endothelial cells. For example, data in Figure 2-3d illustrated that CA4DP starts to affect the viability of proliferating HMVEC-L cells at the dose ~ 1 Increasing the drug dose to ~ 100-fold higher still showd no effect in quiescent HMVEC-L cells. This observation was consistent with previous findings of CA4DP activity in HUVEC cells (Dark et al 1997) However as the clonogenic cell survival assay can not distinguish between necrotic and apoptotic cell death based on the studies from Chapter 2 (Figure 2-4), CA4DP-induced endothelial cell apoptosis was measured in detail. Using the TUNEL assay a dose dependent induction of apopto s is in HMVEC-L cells after CA4DP treatment could be demonstrated (Figure 3-4). This effect was dramatic at the 50 M dose, which induced apoptosis in ~2 0% of the treated endothelial cells that were still attached. Considering the ~ 10 % detached cells that were almost all apoptotic after

PAGE 58

51 50 M CA4DP treatment, the overall apoptotic cell level was as high as 30%, which was pretty significant. Considering the possibility that only 30-40% of the HMVEC-L cells are actively dividing cells (Chapter 2), we believe that CA4DP induces the death of proliferating HMVEC-L cells predominantly by apoptotic processes. Prior studies using DAPI staining as well as flow cytometric evaluations of HMVEC-L cells stained with annexin-V and propidium iodide, also showed 5-10-fold increases in apoptosis in treated compared to untreated cells ( data not shown). These observations are consistent with those of Iyer et al. who used increased ca spase-3 activity to show that CA4DP induced apoptosis in HUVEC cells (Iyer e t al., 1998) It is still possible that the apoptotic cell death may not show the whole picture of the cytotoxicity of CA4DP. Some endothelial cells may die becau s e of losing their clonogenicity after CA4DP treatment. However, since the antivascular effect of CA4DP occurs very rapidly, we believe that for the in vivo situation, the induction of apoptosis by CA4DP is the major contributor of cell death compared to clonogenic cell death. The cytoskeleton of endothelial cells participates in a number of cellular processes, including not only spindle formation and chromosome segregation, but also intracellular transport of molecules cell motility, and angiogenesis (Giannakakou e t al. 1998). The microtubules which forn1 an integral part of the cytoskeleton therefore provide an attractive molecular target for anti-vascular drugs such as CA4DP. The present studies showed that CA4DP caused a time dependent tubulin depolymerization in HMVEC-L cells. Within 24 hr, most microtubules in the cells depolymerized into free tubulin subunits (Figure 3-2).

PAGE 59

52 Immunofluorescence studies showed complementary results; a disorganized pattern of microtubules was evident in CA4D P-treated HMVEC-L cells rather than a nortnal cytoskeletal architecture (Figure 3-1 ). These findings indicate that CA4DP binds to tubulin and shifts the dynamic equilibrium that nonnally exists in cells between polymerized and soluble tubulin. Disrupted microtubule structure also directly affects the ability of cells to migrate. Figure 3-3 shows that CA4DP doses of 0.1 ~ 10 M inhibited, by 10 ~ 40%, the ability of HMVEC-L cells to migrate into denuded areas in culture plates These fmdings suggest that although CA4DP may be predominantly acting as an anti vascular agent, it also possesses at least some of the features typically associated with antiangiogenic agents, namely effects on migration and proliferation. This conclusion is consistent with the results from Chapter 2 which showed that CA4DP at doses less than those affecting cell survival in the clonogenic cell survival assay, blocked HMVEC-L cells at the 0 2 /M phase of the cell cycle (Figure 2-2). Another outcome of microtubule structure disruption is endothelial cell detachment. Results from Figure 3-5 showed that CA4DP caused a dose dependent cell detachment in proliferating HMVEC-1 cells. Although not shown, at high doses, cells began to round up and detach immediately after a 2-hr CA4DP treatment. If such an effect occurs in vivo, it could explain the vascular shutdown seen after CA4DP exposure. Considering the irregular, capillary-like vessels in tumors even if only a few endothelial cells round up and detach from the

PAGE 60

53 monolayer vessel bed, this might be sufficient to occlude the blood flow and ultimately shut down the whole vascular supply in the tumor In summary, the present results show that CA4DP is specifically toxic to proliferating endothelial cells predominantly by apoptotic pathways. CA4DP also inhibits tubulin polymerization, endothelial cell migration and attachment. While the in vitro results indicate that CA4DP has potent effects on endothelial cells, it should be recognized that vascular shutdown in CA4DP-treated tumors can occur within 20 min after treatment (Sackett, 1993; Li et al., 1998). The rapidity of vascular shutdown observed suggests that more immediate changes are responsible for the drug effects seen. One possibility is that CA4DP can have dramatic effects on the three-dimensional shape of newly formed endothelial cells. It is possible that the early manifestations of cell shape changes brought on by CA4DP effects on tubulin binding which lead to cell detachment and apoptosis in vitro, result in similar physical effects in vivo which dramatically alter capillary blood flow, expose basement membrane and, as a result, induce haemorrhage and coagulation. Recent studies showed that CA4DP induces endothelial shape changes with a consequent increase in permeability of an endothelial cell monolayer to macromolecules (Twardowski and Gradishar, 1997). The increase in vascular perrneability to macromolecules may result in an increase in interstitial fluid pressure, an increase in blood viscosity, procoagulative effects, vascular collapse, and the induction of cytokines. The present findings provide a basis for the selective action of CA4DP again st the proliferating endothelial cell population found in tumors. The reason

PAGE 61

54 for the tumor selectivity of CA4DP may relate to difference s in proliferation rate of endothelial cell s in tumor s and normal tissue s. Further investigations of morphological change s of endothelial cell s in vi v o need to be pursued.

PAGE 62

55 a. b. Figure 3-1. Indirect i1111nunofluorescence of microtubules in HMVEC-L cells. a) No drug treattnent; b) 4 l1r treat1ne11t with 10 M CA4DP.

PAGE 63

56 C. l 7 i gl ar c 1 1 -c(> llt i11l1 c<.I. c) 24 l1r trc,1t111c11t wil l1 l () ~l M C'A40P

PAGE 64

Time (hr) %P 100 80 en 60 G) > Q) C :, 40 .c ::, t20 2 37.0 57 0 49.6 p s p s 4 12.6 p s D polymerized tubulln IZJ soluble tubulin 0 2 4 Time (hr) 24 1.3 p s 24 Figure 3-2. Effect of CA4DP on tubulin polymerization in HMVEC-L cells. The cells were treated with 5 M CA4DP over a 24 -hr period. Cells were harvested at different time points and tubulin polymerization was assessed. The percent of polymerized tubulin (%P) was dete1mined by dividing the value of polymerized tubulin by the total tubulin content (the sum of P and S).

PAGE 65

5 8 CA4DP 2 hr treatment 350 D 2 4h r u, 300 t22l 4 8 h r Cl) (J 250 "'C 22 9 % C1) .,_ 3 1 3% cu 200 'C) 37 7% E .... 0 150 'Cl) 100 .c E :, z 50 0 0.1 1 10 Drug dose {M) Figure 3-3 HMVEC-L cell migration into a 2 mm denuded area in culture plates Cel l s were either untreated or exposed to CA4DP for 2 hr Migrating ce l l number s were counted 24 and 48 hr l ater Data are the mean + SE of 6 replicate s.

PAGE 66

5 9 a. CA4DP 2 hr treatment 30 0 50 M CA4DP 0 511M CA4DP Cl) 20 C1) 0 0 0 Q. 0 10 t 0------------------0 5 10 15 20 25 Time after treatment (hr) b. CA4DP 2 hr treatment 30 ';I. -"' 20 Q) (J (J 0 a T 0 10 i '\ ""~ -,A T X ., 0 r 'I X , control 1 5 10 50 Drug dose (M) Figure 3 -4 Effect of C A4DP on HMVEC-L apoptosis. (a) HMVEC-L cells were treated with 5 and 50 M CA4DP for 2 hr and stained for TUNEL assay after a specified time. (b) HMVEC-L cells were treated with various doses of CA4DP for 2 hr, and the apoptotic nuclei in the monolayer were counted 22 hr later. Data are the mean SE of three independent experiments.

PAGE 67

60 Table 3-1. COMP ARIOSN OF APOPTOSIS DATA FROM TUNEL ASSAY AND DAPI STAINING (CA4DP 2 HR TREATMENT) APOPTOTIC CELLS (%) TREATMENT TUNEL DAPI control 1 8 0.2) 1.8 0.2) lM 3.4 0.7) 4.9 1.4) 5M 5.6 1.9) 7 2 (.1) 10 M 11.8 1 4) 13.3 1.6) SOM 19.6 3.0) 19.4 2.7)

PAGE 68

6 1 CA4DP 2 hr treatment 20 18 16 Proliferating HMVEC-L 14 Quiescent HMVEC-L 0 .,. -"' 12 Q) (J "C 10 Q) .t: 8 (J cu .., 6 Q) C 4 2 0 -'-----l'"'T'T'T'ii==::::;::::::;:::::;::;;:;:;:;:;~::::;:::::;::;::;~~:..,._,...,...,.~ 0.01 0 .1 1 10 100 CA4DP dose (M) Figure 3-5 Quantification of cell detachtnent in CA4DP treated HMVEC-L cells. Cells were treated with various doses of CA4DP for 2hr and the numbers of detached cells were counted electronically 22 hr later. Data are the mean + SE of 6 replicates.

PAGE 69

CHAPTER4 STUDY OF THE EFFECTS OF CA4DP IN THE MODEL OF KAPOSI'S SARCOMA Intro du ction Kaposi's Sarcoma (KS) is a highly vascularized neoplasm that primarily results in raised, highly vascularized lesions (Groopman, 1987; Tappero et al., 1993). Before the 1980s, KS was a rare disorder that occurred predominantly in elderly men of Mediterranean or Eastern European Jewish descent. With the advent of the acquired immunodeficiency syndrome (AIDS) epidemic its occurrence has increased dramatically. KS is classified into four different types: classic, African endemic iatrogenic or drug-associated, and AIDS-related (Sung et al., 1997). Classic KS usually follows an indolent and benign clinical course that rarely requires treatment. In contrast, AIDS-KS is a fulminant disease that requires aggressive pharmacotherapy especially when it involves visceral organs. AIDS-KS is the most common neoplastic disease in patients with AIDS. Presently, it is the fourth leading clinical manifestation of AIDS (Pluda et al., 1993 ). Cutaneous or mucocutaneous lesions may occur. Lesions occurring in the viscera primarily affect the gastrointestina l tract, I ymph nodes, and pulmonary system. Shortness of breath, dyspnea on exertion, increased respiratory rate, and decreased oxygen saturation are common in patients with pulmonary involvement. Involvement of the pulmonary tract occurs in 15-50% of patients 62

PAGE 70

63 and is estimated to contribute to 25% of death s in patients with AIDS-KS (Tappero et al., 1993 ). Histopatholog y of KS reveal highl y vascularized le s ions with abundant angiogenesis accompanied by abnormal blood vessel development and leakage of blood (Gallo, 1998) The predominant cells in the tumor are the spindle shaped cells believed to be the tumor cells of KS. These cells express some of the surface markers of activated endothelium but also contain smooth-muscle actin, suggesting that the cell of origin may be a primitive vascular cell (Kroll and Shandera 1998). Growth factors that support spindle cell proliferation include interleukin (IL)I IL-6 and tumor necrosi s factor (TNF) (Sung et al., 1997). Proteins that regulate neovascularization or angiogenesis such as ba si c fibroblast growth factor (bFGF) platelet derived growth factors, and vascular endothelial growth factor (VEGF) also promote growth of spindle cell s (Nakamura et al., 1992) Standard treatment s for KS include intralesional injection of vinblastine or a 2 -interferon local radiotherapy and sys temic chemotherapy. These treatments are administered primarily for symptomatic relief and to prevent disease progression. Cytotoxic chemotherapy is the standard therapy for patient s with extensive lesion s and di seas e involving visceral organs and lymph nodes. A s ingle agent s uch a s vincristine or a 2 -interferon i s used for mild cases. More advanced cutaneous or visceral KS is usually treated with combination chemotherapy including agents such as vincri s tine bleomycin and doxorubicin. However these treatments do not signi ficantl y prolong s urvival and the clinical effectiveness is

PAGE 71

6 4 not satisfactory (Lilenbaum and Ratner, 1994). Vincristine or a. 2 -interferon monotherapy give a clinical response in only about 25% of patients while the combination of vincristine, bleomycin and doxorubicin causes regression in about 40% (Kroll and Shandera 1998). Continued pursuit of more effective and less toxic agents is clearly needed. Given the tumor 's histopathology, a variety of new treatment approaches particularly those focused on inhibiting tumor angiogenesis are being investigated. Angiogenesis plays a crucial role in the pathogenesis and progression of KS (Cornali et al., 1996) and antiangiogenesis approaches may provide a means of arresting the progression of KS The drug initially tested as an angiogenesis inhibitor in patients wa s TNP-470 When administered once weekly by intravenous infusion this agent gave partial responses in KS patients (Dezube et al., 1998). More recently there has been an interest in exploring the clinical utility of thalidomide as an anti-KS agent. This was based on evidence that thalidomide could inhibit angiogenesis block tumor necrosis factor alpha (TNF a ), and inhibit intercellular adhesion molecules and basement membrane formation (Gascon and Schwarts, 2000). Preliminary results from two Phase II clinical trials showed that thalidomide had activity in a subset of patients with KS (Welles et al., 1997; Bower et al., 1997) Based on the encouraging results observed with thalidomide and TNP-470, clinical research into the antiangiogenic activities of these and other agents, including interleukin-12 and angiostatin continues (Kroll and Shandera, 1998 ; Gascon and Schwarts 2000).

PAGE 72

65 Because KS is a highly vasoactive neoplasm, directly targeting the actively growing vessels of the tumor may be another app1oach suitable for KS treatment. Since large numbers of neoplastic cells are directly supported by small numbers of endothelial cells damaging the tumor endothelium could have marked impact on tumor cell survival and growth (Bicknell and Harris, 1992). In the present investigation we examined the efficacy of CA4DP in KS xenografts by assessing the extent of both vascular damage and cytotoxic action in these tumors. Methods and Materials KS Xenografts and Treatments KS xenografts were initiated by injecting the flanks of 6-8-week-old athymic NCR nu/nu mice (Frederick Laboratories, Frederick MD) with 1 x 10 6 KSY-1 cells {ATCC, Rockville MD) {Lunardi-Iskandar et al., 1995) and were serially passed by s ubcutaneous transplantation of tumor pieces in the flanks. Macroscopic tumors were available for experiments 3-4 weeks later. Tumor bearing mice were allocated to groups and received either no treatment or different doses of CA4DP (OXiGENE Inc., Lund Sweden). CA4DP was dissolved in 0.9% sterile saline and injected intraperitoneally in a volume of 0.01 ml/g animal body weight. Hoechst-33342 Studies Hoech s t-33342 (bisBenzimide, Sigma) solution was made up in 0.9% sterile saline immediately before use. KS-bearing mice were either untreated or treated with 100 mg/kg CA4DP. Hoechst-33342 then was administered at 40 mg/kg

PAGE 73

66 intravenously (volume 5 ml/kg) at various times after CA4DP injection (Smith et al, 1988). One minute after Hoechst-33342 injection the mice were killed, the tumors and the normal tissues (lung, liver, and muscle) of the mice were resected and immediately immersed in liquid nitrogen for subsequent frozen sectioning. For each tumor sample, 10 m cryostat sections were cut at three different levels between one pole and the equatorial plane. The sections were air dried and then studied under UV illumination using a fluorescent microscope. Blood vessel outlines were identified by the surrounding halo of fluorescent H33342-labelled cells. Vessel counts were performed using a Chalkley point array for random sample analysis (Curtis, 1960). Briefly, each section was viewed at x 10 objective magnification. A 25-point Chalkley grid was positioned randomly over field of view. Any points falling within haloes of fluorescent cells were scored positive. Twenty random fields were counted per section and a minimum of six sections per tumor was examined. Histological Staining Histological sections were prepared from KS xenografts 4 and 24 hr after CA4DP was given. All specimens were fixed in 10% neutral buffered formalin, routinely processed and embedded in parafin. Sections (4 applied to slides were deparaffinized in xylene and hydrated through graded alcohols. Standard hematoxylin and eosin (H&E) staining was used for each slide. Necrotic fractions in KS tumors were quantified by image analysis. Briefly, stained sections were divided into 4-8 grids and areas of necrosis within each grid were traced on an

PAGE 74

67 Image Pro Plus system. All grid measurements we1e combined and the percentage of necrosis relative to the total area of the tumor was calculated. C l onogenic Cell Survival Assay Clonogenic cell s urvival in treated or untreated tumors was assessed using an in vi v o to in vitro clonogenic cell survival assay as previously described (Allalunis-Tumer and Siemann 1986 ; Siemann 1995) Briefly 24 hr after treatment tumor bearing mice were killed and their tumors excised and then dissociated to a single cell suspension using an enzyme cocktail (0 025 % collagenase 0.05% pronase and 0.04 % DNase). The cells then were mixed with 10 4 lethally irradiated cell s in a-MEM medium containing 10 % fetal calf s erum and plated into 60 mm petri dishes After 2 week s of incubation at 3 7C, colonies of 50 or more cell s were counted with the aid of a dissecting micro s cope Tumor surviving fraction s were determined by multiplying the calculated fraction of surviving cells by the ratio of cells recovered in treated versus untreated tumors. Tumor Growth D elay Assay Once KS xenografts had reached a minimum s ize of200 mm 3 the mice were allocated to different group s for CA4DP treatment. A 100 or a 300 mg/kg dose of CA4DP was used in the s ingle-do s e treatment s tudie s For the multiple-dose groups 100 mg/kg CA4DP wa s administered either on days 1, 3 and 5 or on day s 1 5 and 9. After treatment tumors were mea s ured daily u s ing calipers and the perpendicular diameters were determined. Tumor volumes were estimated u s ing the equation: Volume = 4m 3 / 3 r = (a + b) / 4 where a and bare the perpendicular diameters The time f o r each tumor to reach a s i z e of 900 mm 3 wa s recorded

PAGE 75

68 Hypoxia Stress and CA4DP Treatment KSY-1 cells were plated into 60 mm petri dishes at 2x 10 5 cells / dish. Once KSY-1 cells were attached to the dishes the cells were treated with a range of doses of CA4DP. Treated KSY-1 cells used for hypoxic conditions were immediately placed in an airtight chamber and subjected to repeated rounds of evacuation and replacement with nitrogen gas. The sealed chambers were then incubated for 24 hr at 3 7 C. KSY-1 cells under aerobic conditions were also treated with a range of doses of CA4DP at 3 7 C for 24 hr. After CA4DP treatment, these treated cells were then trypsinized, counted, and plated into 60 mm Cell + petri dishes for clonogenic cell survival assay. After 2 weeks of incubation at 3 7 C, colonies of 50 or more cells were counted with the aid of a dissecting microscope. Cell surviving fractions were calculated as the ratio of colonies counted in treated versus untreated group. Results Initial studies focused on the early effects of CA4DP treatment on tumor vasculature. Results obtained with the Hoechst-33342 fluorescent dye showed that a single dose of 100 mg/kg CA4DP caused an almost complete vascular shutdown in KS xenografts within 4 hr after treatment (Figure 4-1 ). Compared to the abundant vasculature in the untreated tumors KS xenografts in mice treated with CA4DP showed vessels essentially only near the periphery of the tumors. This vascu lar damaging effect of CA4DP occurred rapidly and was detected in KS xenografts 30 min after drug treatment. Indeed, when functional vessels were counted, most were found to be shutdown by CA4DP within 0.5 to 2 hr after

PAGE 76

69 treatment (Figure 4-2) The vasculature in normal tissues from untreated and CA4DP-treated mice also were assessed. It is clearly seen from Figure 4-3 that none of the vasc ular networks in the lung the liver, and the muscle was affected at 4 hr after a 100 mg/kg dose of CA4DP treatment Histological evaluations of KS xenografts showed morphological evidence of damage in tumor cells within a few hours after CA4DP treatment. By 24 hr after CA4 DP treatment (100 mg/kg) extensive haemorrhagic necrosis could be seen with viable tumor cells detectable only at the periphery of the tumor adjacent to the surrounding normal tissues (Figure 4-4). To quantify the extent of necrosis produced by CA4DP treatment, sections from KS xenografts removed 24 hr after treatment were assessed using an image analysis system. The results (Figure 4-5) showed that compared to the ~ 10% necrosis seen in untteated tumors, treatment with 50 mg/kg CA4DP increased the extent of necrosis to ~ 60%. In xenografts treated with 100 mg/kg CA4DP, the necrotic fraction increased to ~ 90% 24 hr after drug treatment. Anti-tumor effectiveness of CA4DP was determined by measuring clonogenic cell s urvival in KS xenografts treated with various doses of this agent. The data demonstrated that administering increa sing doses of CA4DP to tumor bearing mice re sulted in a dose-dependent increa se in tumor cell kill (Figure 4-6). A comparison of results in Figures 4-5 and 4-6 further shows a consistency between the clonogenic cell survival data and the results of the histological evaluations, i.e. a 100 mg/kg do se of CA4DP caused ~ 90% tumor cell death and necrosis 24 hr after treatment. The tumor cell killing effect with a single 100

PAGE 77

70 mg/kg dose of CA4DP treatment also was assessed at various times after treatment. The results showed that the maximum tumor cell kill ( > 90%) was achieved between 1-3 days after CA4DP treatment (Figure 47). At later times cell survival recovered. Hypoxic tumor cells normally exist in central areas of tumors, far away from sources of efficient oxygen supply Since CA4DP causes extensive central necrosis in KS xenografts, it raised the question of whether part of the effect of this agent was the consequence of CA4DP killing hypoxic cells more efficiently than well-oxygenated tumor cells. To answer this question, we examined the toxicity of CA4DP in both aerobic and hypoxic KSY-1 cells in vitro (Figure 4-8). The results showed that hypoxic KSY-1 cells were not more susceptible to CA4DP treatment than aerobic KSY-1 cells Indeed clonogenic cell survival determinations showed the surviving fraction of hypoxic cells after CA4DP treatment to be slightly higher than that for aerobic cells. This result further supports the notion that it is the antivascular effects of CA4DP, not a higher sensitivity of hypoxic tumor cells to CA4DP, that are responsible for the induced extensive central necrosis in KS tumors after CA4DP treatment. In conjunction with the cell survival investigations, studies evaluating the effects of CA4DP on KS growth also were performed. Tumor-bearing mice were treated at a size of 200 mm 3 with a single dose of either I 00 or 300 mg/kg dose of CA4DP. Such treatments resulted in a slight, but not significant tumor growth delay (Figure 4-9, Table 4 1). Indeed the higher dose of CA4DP (300 mg/kg) did not increase the tumor growth delay compared to that achieved with 100 mg/kg.

PAGE 78

71 In an attempt to apply this agent more efficiently and in particular to impair the regrowth of vasculature from the surviving rim of tumor cells (Figures 4-1 and 44), studies utilizing repeated injections of CA4DP (100 mg/kg) also were performed. Treatment commenced when the tumors reached 200 mm 3 Two different treatment schedules were used: CA4DP was administered on days 1 3 and 5 or on days 1, 5 and 9. The results showed a far superior response of the KS xenografts to the multiple CA4DP treatment schedules (Figure 4-9). Both treatments induced significant growth delay compared to untreated tumors and tumors treated with either single dose of CA4DP (Table 4-1 ). Administering CA4DP using the multiple dose schedule resulted in a growth delay of ...., 20 days (Figure 4-9, Table 4-1 ). Discussion Kaposi's sarcoma is the most common tumor seen in HIV-infected patients. Several chemotherapeutic agents including vinca alkaloids, etoposide bleomycin and doxorubicin are commonly used to treat AID-KS patients (Sung et al ., 1997). Vinca alkaloids ( either vincrinstine vinblastine or an alternating regimen of the two) exert their anticancer effect by binding to tubulin and preventing its polymerization to form microtubules (Chabner and Collins 1990 ; Yarchoan 1999) thus inhibiting a number of cellular processes, including mitosis. Over the years there have been several advances in the therapy of this disease, including the use of liposomal anthracyclines, paclitaxel, and antiangiogenesis agents TNP470 and thalidomide (Yarchoan, 1999; McGarvey et al. 1998). Among these paclitaxel, which also interferes with microtubule dynamics by promoting the

PAGE 79

72 formation of highly stable microtubules which resist depolymerization, was found to inhibit the growth of KS-derived spindle cells and to be a potent inhibitor of endothelial cell proliferation (Saville et al., 1995). Like vinca alkaloids, CA4DP also binds to tubulin and inhibits tubulin polymerization. However, unlike the vinca alkaloids, CA4DP has demonstrated antivascular effects at very low doses (Dark et al., 1997; Li et al., 1998; Chaplin et al., 1999). It has been well established that the vascularization of solid tumors is a prerequisite if a clinically relevant size is to be reached (Folkman, 1986). The dependence of the tumor on its induced vessel forrriation has created a great deal of enthusiasm in specifically targeting the microcirculation in cancer therapy (Denekamp, 1993). Results from our laboratories and those of others have shown not only that CA4DP has specific effects on actively dividing endothelial cells but also that this agent can cause rapid vascular shutdown in a variety of preclinical tumor models (Dark et al., 1997; Li et al., 1998; Horsman et al., 1998). Because KS is a highly vascularized neoplasm with a cellular origin suggested to be endothelial cell derived, and because tubulin-binding agents have previously been found to be active in KS the present study was undertaken to examine the efficacy of CA4DP in an in situ model of this disease. The KSY 1 cell line originated from cells isolated from the pleural effusion of an AIDS-associated KS patient (Lunardi-Iskandar et al., 1995). KSY-1 cells promote tumorigenesis angiogenesis and metastasis in immunodeficient mice. The model 's similar biological, morphological and immunophenotype make it a valuable adjunct for studies related to pathogenesis and therapy of AIDS-KS

PAGE 80

73 (Rojiani et al., 2000). In the present study, KSY-1 cells were used to initiate KS xenografts in athymic mice in order to assess their response to the vascular targeting agent CA4DP. The pathophysiological effects of CA4DP observed in the present study in KS xenografts were similar to those previously reported by our laboratory in the rodent KHT sarcoma model (Li et al., 1998). CA4DP treatment resulted in a rapid induction of vascular damage in tumors such that 4 hr after treatment there existed an almost complete vascular shutdown (Figures 4-1 and 4-2). This was followed by extensive secondary tumor cell death due to ischemia (Figures 4-4 and 4-5). Histological assessments showed extensive haemorrhagic necrosis 24 hr after CA4DP was administered systemically to KS-bearing mice, with only a small rim of viable tumor cells surviving near the periphery of the tumor (Figure 4-4). These tumor cells probably survived because they were close to the surrounding normal tissues where they were supp lied with nutrients from the nortnal tissue vasculature which was not affected by the action of CA4DP. Studies with intravital microscopy have shown that peripheral tumor tissue retains some blood flow after CA4DP treatment but becomes hemorrhagic with dilated blood vessels (Tozer et al., 1999). This suggests that vascular pertneability changes may be more profound in the periphery, where most of the extravasation of macromolecules occurs under unperturbed conditions. Results from the clonogenic cell survival investigations were consistent with the histological observations. For example a 100 mg/kg dose of CA4DP which induced ~ 90% necrosis in KS xenografts also reduced the viable tumor cell

PAGE 81

74 population to ~ 10% of the pretreatment value (Figures 4-5 and 4-6). The cell survival evaluations established that CA4DP treatments led to a concentration dependent killing of KS tumor cells (Figure 4-6). This killing manifested itself primarily as a rapid loss of viable cells from the cell population within the 24-hr period after CA4DP treatment. Although normal tissue toxicities were not measured in the present studies, the anti vascular effects of CA4DP were achieved at doses less than 1 / 10 of the maximum tolerated dose {MTD) and without detectable morbidity a s previously reported (Dark et al., 1997). While some of the other tubulin binding compounds, for example the vinca alkaloids, may also express antivascular action, they exert their effects at doses approaching the MTD and often only in the presence of significant morbidity (Chaplin et al., 1996). Indeed vinca alkaloid therapy in AIDS-KS patients is frequently limited by neutropenia and peripheral neuropathy (Sung et al., 1997). CA4DP's high tubulin binding affinity (Pettit et al 1989), selective toxicity in proliferating endothelial cells, and effectiveness at low doses (Dark et al 1997) may prove to be of considerable value in the continuing clinical evaluation of CA4DP as an anti vascular agent. The present studies showed that a single 100 mg/kg dose of CA4DP had little effect on the growth of KS xenografts (Figure 4-6) despite the fact that this dose of CA4DP causes extensive central necrosis in these tumors (Figures 4-4 and 4-5). The most likely explanation is that the remaining viable tumor cells, located at the periphery of the tumor near the nortnal tissue, survive and continue to proliferate. This conclusion is supported by the study illustrated in Figure 47.

PAGE 82

75 The viable tumor cell fraction began to increase 2 days after CA4DP treatment because of the continued proliferation of cells at the rim of the tumors. The lack of a change in tumor growth following the 100 mg/kg CA4DP treatment probably is a consequence of a balance between the growth of new cells from the surviving rim and the removal of necrotic material from the tumor's core. Increasing the single dose to 300 mg/kg did not result in a greater growth delay in KS xenografts (Figure 4-9). However this result is not surprising given the extent of cell death and necrosis caused by a 100 mg/kg CA4DP dose (Figures 4-5 and 4-6). Increasing the dose further would have little additional effect on the tumor cells which s urvi ve due to their location near normal blood vessels The effect of multiple exposures of CA4DP on KS xenograft growth also was examined. The rationale for these studies was two-fold. First, it was apparent from the single dose studies that maximum anti-tumor efficacy occurred with doses of~ 100 mg/kg (Figures 4-5 and 4-6) and that little could be gained by increasing the exposure dose further. Second, and more importantly, we reasoned that administering multiple doses of CA4DP at times when the tumor is regrowing and the tumor vasculature is recovering and/or re-establishing itself might prove to be a much more efficient application of this agent. From Figure 47 we know that it was between 2 and 5 days after single dose of CA4DP treatment that the KS tumors showed recovery and regrowth. Therefore, two different treatment schedules, either giving CA4DP on days 1, 3, and 5 or on days 1, 5, and 9, were examined in KS-bearing nude mice. The results showed that unlike the single dose treatments, both multiple dose schedules caused significant growth delay in

PAGE 83

76 the xenografts (Figure 4-9 and Table 4-1). Multiple doses of CA4DP were clearly far more effective at inhibiting KS growth than single dose treatments. For example, administering three 100 mg/kg dose fractions of CA4DP, as opposed to a single 300 mg/kg treatment, increased the growth delay by ~ 14-17 days (Table 4-1 ). To date we have not optimized these multiple CA4DP treatment schedules in the KS model. Whether similar gains can be achieved with lower doses / fraction and/or greater numbers of multiple treatments is currently under investigation. Still, the studies described in the present investigations indicate that administering multiple doses of CA4DP is a very effective way of inhibiting KS growth. In conclusion, CA4DP treatment can cause va scular shutdown, haemorrhagic necrosis, extensive tumor cell killing, and growth delay in KS xenografts. These findings suggest a possible application of the vascular targeting agent CA4DP in the clinical management of KS.

PAGE 84

77 Figure 4-1. I<.S tumors removed 1 min after i.v. injection of 40 mg/kg H-33342. Vessels were identified by the surrounding fluorescent tumor cells. Tumors were either from untreated mice (a) or from mice 30 min (b), 2 hr (c), and 4 hr (d) after a 100 mg/kg CA4DP treatment. Magnification was x32 for a-d.

PAGE 85

78 100 mg/kg CA4DP treatment 200 "' 1 50 .... C: :::, 0 (J .... ns 100 :::, (J "' ns > 50 0 1 2 3 4 5 Time (hr) Figure 4-2. Vascular counts for KS tumors with Hoechst-33342 staining. Tumor bearing mice were treated with 100 mg/kg CA4DP for a specified time before i.v injection of 40 mg/kg Hoechst 33342. Vessel counts were performed by using a Chalkley point array for random sample analysis Data are the mean SE

PAGE 86

79 A Figure 4-3. Nom1al tissue sec tion s fiom KS-bearing mice removed l tnin after i.v injection of 40 mg/kg H-33342. Vessels were identified by the surroundi11g fluorescent cells. Sa:inples were e ither from untreated mice (a) or from mice 4 hr after a J 00 1ng/kg CA4DP treatment (b ). Magnification was x 10 A) C ryostat sec tion s of the liver.

PAGE 87

8() B. Figure 4-3-continued. B) C ryo s tat sec tion s of the lu11g

PAGE 88

C. "I. IJ ' .. ' \ .._ .. r ,.. .. .. .... 8 1 I I .. .. .. '-~. ,, -~ .... .. t .. -. ? > .. ~, --, '/ .. ;,. a '"' ... . -~. : .ti. . ', ..... '..... ..... .. ,. . ' ... ..,. . .. .. . i .. ~... .. 'C... .. ' _, ,.. IL ..,..,.. -' -~ .~ --.. .... ~ . ..... .... .,..::, -~ -6' ... ', . ... . .. \A .. _....... ~"'" .... -. .... .. -., .... 4 ~ ... .. Figure 4-3-continued. C) Cryostat sections of tl1e leg n1uscle

PAGE 89

\, 82 I .. I ' ,.. ,. ... I ,. ., ,4 .. ...... .. .. J '\. J ... .. ... I ... ,j,~ ... ... Figure 4-4. Standard H&E staining of 4sections from KS tu1nors. Tu1nors were from mice that had received a I 00 mg/kg dose ot CA4DP 24 hr prior to assessment. Magnification was x 10 ( a) and x 40 (b ).

PAGE 90

83 KS xenograft 100 'o' 80 ._, C 0 ... (.) 60 ftS ... ..... (.) ... 0 40 ... (.) Cl) z 20 0 0 50 100 CA4DP dose (mg/kg) Figure 4-5 The extent of necrosis in KS xenografts assessed 24 hr after the administration of single dose s ( 50 or 100 mg/kg) of CA4 DP Data are the mean SE.

PAGE 91

8 4 C 0 0 0 1 cu .:-,C, C -~ > ... ::J u, ... ::J 0.01 t0 50 100 150 200 CA4DP dose (mg/kg) Figure 4 6 Tumor cell killing in KS tumor s treated with increasing doses of CA4DP. Data were determined 24 hr after drug treatment and are the mean SE of at lea s t 6 tumor s.

PAGE 92

85 KS 100mg/kg CA4DP C 0 ... 0 n, .... C') C -~ 0 1 ::::, ,,, 0 E ::::, ..... 0 1 2 3 4 5 Time after treatment (day) Figure 4 7. T u mor cell kil l ing in KS tumors treated with a 100 mg/kg CA4DP C l onogenic cel l survival assay was performed at a specified time after drug treatment Data are the mean+ SE of at least 6 tumo r s.

PAGE 93

C: 0 .., (.) : 1 C, C: "> 0.1 > 1:::s U) Cl) (.J D Aerobic 0 Hypoxic 86 0.01 -+----,----,-----------0.001 0.01 0.1 1 10 100 Drug dose (M) Figure 4-8. The cytotoxicity profile of CA4DP against KSY-1 cells under hypoxic or aerobic conditions. The cells were exposed to CA4DP for 24 hr and the cell killing effects were assessed using clonogenic cell survival assay. Data are the mean + SE of three experiments.

PAGE 94

87 60 KS tumor growth (w / CA4DP trea t ment ) median value 50 C"') E E 40 0 0 a, 0 8 30 l g 8 "' C 0 20 8 a a 1 0 8 0 0 c ontrol S0 ( 100 ) S0 ( 300 ) MD { 1 3 5) MD { 1 5 9 ) 10(8-13) 13 5(8-19) 12 5(10 5-21) 30(10-42) 27(16-53) Figure 4-9 Growth delay s in KS tumors with single or multiple doses of CA4DP treatment Single do se (SD) of CA4DP (100 or 300 mg/kg) was administered on day 1 Multiple do s e s ( MD) of C A4DP (100 mg/kg) were administered on days 1 3, and 5 or on day s 1 5 and 9. Each datum point represents an individual animal.

PAGE 95

88 Table 4-1. GROWTH DELAY IN KS XENOGRAFTS CAUSED BY SINGLE OR MULTIPLE DOSES OF CA4DP INJECTION IN TUMOR-BEARING MICE GROUP 1. Control 2. S.D.(100) 1 3. S.D.(300) 1 4. M.D. (1 3 5) 2 'J 5. M.D. (1 5 9)MEDIANTIME TO 900mm 3 (DAYS) 10.0 13.5 12.5 30.0 27.0 GROWTH DELAY (DAYS) 3.5 2.5 20.0 17.0 SIGNIFICANCE 3 VSCONTROL SIGNIFICANCE 3 NS NS P < 0.025 P < 0.025 4 vs 2 5 vs 2 P < 0.025 4 vs 3 5 vs 3 P < 0.025 1 Single dose of 100 or 300 mg/kg CA4DP were used in the treatment. 2 100 mg/kg CA4DP was administered to tumor-bearing mice on days 1 3 and 5 or on days 1 5, and 9. 3 Determined by Wilcoxon Rank-Sum Test while compared the probability distributions of growth times of control versus treated tumor (a .025).

PAGE 96

CHAPTERS STUDY OF THE EFFICACY OF CA4DP IN COMBINATION WITH CONVENTIONAL ANTI-CANCER THERAPIES IN KS XENOGRAFTS Introd u ction AIDS-related KS is characterized by a heterogeneous presentation and an aggressive clinical course. It often presents as multiple, symmetric, cutaneous lesions Because of the heterogeneous presentation of AIDS-KS, no single treatment regimen can be recommended for all patients. Therapy for AIDS-KS is not curative and, to date, no therapy has been unequivocally proven to impact survival, with the possible exception of interferon-a (Morris and Valley, 1996). The extent and rate of progression of AIDS-KS and the severity of the underlying HIV infection are factors used to deterrnine the best treatment approach for individual patients Since cell proliferation and angiogenesis are the two key mechanisms involved in KS tumor growth, most of the treatments target one or both of them. Antiproliferative agents commonly used are chemotherapy, interferon and radiotherapy. AIDS-KS lesions are exquisitely sensitive to radiation therapy (Hill, 1987) Symptoms caused by mass effects (pain and lymphadenopathy) are best treated with radiation therapy because a response can be more rapidly achieved. Significant responses are reported in essentially all lesion treated with radiation therapy, including pulmonary KS lesions refractory to conventional chemotherapy 89

PAGE 97

90 regimens (Hill 1987). Complications such as mucositis and ulceration of skin and tissue may result after radiation therapy Both single-agent and combination chemotherapy have been used in patients with various stages of AIDS-KS. Sing l e-agent chemotherapy produces responses in ~ 25 % of patients. The duration of response reported has ranged from 1 to 9 months and all patients re l apse eventually after discontinuing therapy (Morris and Valley, 1996) Several combinations of chemotherapeutic agents have been investigated in an attempt to increase efficacy and to diminish toxicity by using lower dosages of individual agents. For example, since vincristine and vinblastine have exhibited significant activity in AIDS-KS they were combined in a weekly alternating schedule in an attempt to reduce toxicity (Sung e t al., 1997). Still because patients with HIV infection tend to be very susceptible to chemotherapy induced toxicity combination chemotherapy is reserved for treatment in patients with rapidly progres s i v e or potentially life-threatening disease The optimal therapy for AIDS-KS has not yet been determined. Additional efforts in the management of AIDS-KS are directed at the under I ying pathogenesis of the di s ease. Several antiangiogenic agents are being evaluated including the heparin analog pentosan po l ysulfate recombinant platelet factor-4, fumagillol derivative s, bacterial cell wall complexe s, and surarnin (Morris and Valley 1996). Preclinical s tudie s al s o examined the effects of antisen s e oligonucleotides which target bFGF and VEGF mRN A on KS growth. It wa s s hown that the anti s en s e oligonucleotides could block angiogenensis and KS le s ion forn1ation in nude mice (Enso l i e t al. 1994)

PAGE 98

91 Because KS is a highly vasoactive neoplasm, directly targeting the actively growing vessels of the tumor may be suitable for KS treatment. Previous studies in our laboratory have investigated the efficacy of the vascular targeting agent CA4DP in the model of AIDS-KS. Results have shown that 100 mg/kg dose of CA4DP caused ~90% tumor necrosis in KS xenografts 24 hr after treatment (Figure 4-5), and that repeated doses, not single dose, of CA4DP treatment caused significant growth delay in KS xenografts (Figure 4-9). Previous studies also indicated that CA4DP alone were unable to eliminate the tumor completely, and a small, nevertheless viable, rim of tumor remained. These remaining viable tumor cells continue to proliferate, which may explain the lack of change in tumor growth following a single dose of CA4DP treatment. Therefore, the application of antivascular agents will need to be given in conjunction with conventional anticancer therapies. Our studies have shown that CA4DP can significantly enhance tumor response to radiation in KHT sarcoma (Li et al., 1998), and others also demonstrated an effective enhancement of antitumor effects of cisplatin and 5-FU by combining with CA4DP (Chaplin et al., 1999; Grosios et al., 2000). Radiation and chemotherapy are the standard therapies for patients with various stages of AIDS-KS. In the present investigations we have examined the efficacy of combining CA4DP with ionizing radiation or chemotherapeutic agents in the model of Kaposi's Sarcoma. Before examining the effects of CA4DP in combination with chemotherapeutic agents in KS xenografts, we screened the sensitivity of KSY-1 cells to several agents clinically used to treat KS patients, aiming to find the

PAGE 99

92 suitable agents for the combination study in vivo. These agents are cisplatin, vinblastine, doxorubicin, and VP-16. Materials and Methods Cell Culture and Drug Sensitivity Study KSY-1 cells (ATCC, Rockville, MD) were cultured in positively charged Cell + tissue culture flasks from SARSTEDT (Newton, NC). The cells were grown in RPMI 1640 medium (Gibco BRL, Grand Island, NY) with 100 units / ml penicillin, 0.1 mg/ml streptomycin, and 1 Oo/o FBS, and were passed weekly. To test the sensitivity of KSY-1 cells to chemotherapeutic agents cisplatin, vinblastine, doxorubicin, and VP-16, KSY-1 cells were plated in 60 mm petri dishes at 1x10 5 cells / dish. On day 3, cisplatin (Bristol-Myers Squibb Co., Princeton, NJ), vinblastine (Fujisawa USA, Inc., Deerfield IL), doxorubicin (Gensia Laboratories Ltd., Irvine, CA), and VP-16 (Bristol-Myers Squibb Co., Princeton, NJ) injection so lutions were diluted in 0.9% sterile saline before drug treatment. KSY-1 cells were then treated with these agents at specified concentrations for 1 hr. The dishes were then washed with PBS and replenished with fresh RPMI 1640 medium. 23 hr later cells in each group were trypsinized, counted, and plated into 60 mm petri dishes for clonogenic cell survival assay. After 2 weeks of incubation at 3 7 C, colonies of 50 or more cells were counted with the aid of a dissecting microscope. Ce ll surviving fractions were calculated as the ratio of colonie s counted in treated versus untreated group.

PAGE 100

93 KS Xenografts and Treatments KS xenografts were initiated by injecting the flanks of 6-8-week-old athymic NCR nu/nu mice (Frederick Laboratories, Frederick MD) with lx10 6 KSY-1 cells and were serially passed by subcutaneous transplantation of tumor pieces. Macroscopic tumors were available for experiments 3-4 weeks later. Tumor bearing mice were allocated to groups and receive either no treatment or different treatment strategies. CA4DP, cisplatin, and vinblastine all were injected intraperitoneally in a volume of 0.01 ml / g animal body weight. In the combination experiments, cisplatin or vinblastine was administered 1 hr before CA4DP injection. For radiation treatment, tumors were irradiated in unanesthetized mice using a 13 7 Cs source operating at a dose rate of 1.5 Gy / min. In the combination studies, CA4DP was given 0.5-1 hr after radiation treatment. Measurement of Tumor Response Clonogenic cell survival in treated or untreated tumors was assessed using an in vivo to in vitro clonogenic cell survival assay as previously described (Allalunis-Turner and Siemann 1986; Siemann, 1995). Briefly, 24 hr after treatment, tumor-bearing mice were killed and their tumors excised and then dissociated to a single cell suspension using an enzyme cocktail (0.025% collagenase, 0.05% pronase, and 0.04% DNase). The cells then were mixed with 10 4 lethally irradiated cells in a-MEM medium containing 10% fetal bovine serum and plated into 60 mm petri dishes. After 2 weeks of incubation at 37 C, colonies of 50 or more cells were counted with the aid of a dissecting microscope. Tumor surviving fractions were determined by multiplying the calculated fraction

PAGE 101

94 of surviving cells by the ratio of cells recovered in treated versus untreated tumors. Tumor Growth Delay Assay Once KS xenografts had reached a minimum size of 200 mm 3 the mice were allocated to different groups for CA4DP and cisplatin treatment. A 100 mg/kg dose of CA4DP given on days 1 3 and 5 or a single dose of 10 mg/kg cisplatin was used in the single-agent treatment studies. For the combination treatment group a 10 mg/kg dose of cisplatin was administered on day one followed by repeated injection of 100 mg/kg CA4DP on days 1, 3, and 5. After treatment, tumors were measured daily using calipers and the perpendicular diameters were determined. Tumor volumes were estimated using the equation : Volume = 4m 3 / 3 r = (a + b) / 4, where a and bare the perpendicular diameters The time for each tumor to reach a size of 900 mm 3 wa s recorded. Results Figure 5-1 illustrates the killing effect of various doses of radiation in KS xenografts either alone or with a 100 mg/kg dose of CA4DP given 0.5-1 hr after radiation The results showed that when KS-bearing 1nice were treated with a combination of radiation and CA4DP, the extent of tumor cell killing was increased significantly, compared with that seen for radiation alone. The addition of CA4DP to the treatment protocol reduced tumor cell survival 10-100-fold below that for radiation alone. Furthermore the combination treatment appeared to alter the shape of the tumor cell survival curve compared to that for radiation

PAGE 102

95 alone, which suggests that the addition of CA4DP to the treatment influenced the radiation-resistant hypoxic cell subpopulation in KS tumors. Since CA4DP alone is unable to eliminate the tumor completely, the possibility of producing even greater anti-tumor effects by adding a second agent to specifically target the rim of viable tumor cells surviving CA4DP treatment was investigated. Because KS xenografts were initiated by injecting KSY-1 cells in nude mice, our initial studies tested the sensitivity of KSY-1 cells to several agents that are clinically used to treat KS patients. Results demonstrated that cisplatin, vinblastine, doxorubicin and VP-16 all showed dose-dependent killing effects in KSY-1 cell s (Figure 5-2) However different dose profiles were observed in different treatment groups. Vinblastine showed toxicity to KSY-1 at the lowest dose range compared to the other agents (Figure 5-2b vs Figures 5-2a 5-2c, 5-2d). Cisplatin showed the strongest tumor cell killing effect at high concentrations Because vinblastine and cisplatin are commonly used agents in KS patients and because of the good response that KSY-1 cells showed to these two agents vinblastine and cisplatin were chosen as the agents to be used in the combination studies with CA4DP in v i v o. The results of treating KS-bearing mice with a range of doses of cisplatin or vinblastine either alone or in combination with 100 mg/kg CA4DP are illustrated in Figures 5-3 and 5-4. In these studies CA4DP was administered 1 hr post chemotherapy and clonogenic cell survi v al was assessed 23 hr later The re s ult s demonstrated that when KS-bearing mice were treated with a combination of cisplatin (or vinblastine) and CA4DP the extent of tumor cell killing could be

PAGE 103

96 increased significantly compared to that seen for either agent alone (Figures 5-3 and 5 4). The addition of CA4DP to the treatment reduced tumor cell survival at least 10 folds below that for cisplatin or vinblastine treatment alone In subsequent studies the combination of cisplatin plus CA4DP was evaluated in KS-bearing mice using tumor growth delay as the response endpoint. Single dose treatment with cisplatin alone ( 10 mg/kg) resulted in significant growth delay in this tumor model (Figure 5-5 Table 5 1 ). This response was further enhanced by combining cisplatin with CA4DP. Indeed tumor-bearing mice treated with this combination showed significantly greater growth delay than that seen for either cisplatin or CA4DP treatment alone (Figure 5-5, Table 5-1 ) It was also observed that CA4DP treatment alone did not cause significant weight loss in the treated mice, and it did not increase the toxic effects of cisplatin in the combination treatment compared to cisplatin treatment alone (Table 5-1 ). D isc u ss i on Kaposi's Sarcoma is the most common malignancy diagnosed in AIDS, occurring in approximately 20% of these patients. Several chemotherapeutic agents including vinca alkaloids, etoposide, doxorubicin and cisplatin are used to treat AIDS-KS patients. Opportunistic infections occur in 30-50% of AIDS-KS patients receiving cytotoxic therapy (Schwartsmann et al., 1998). In addition, drug-related toxicities are observed in almost all patients. In spite of these limitations cytotoxic agents are usually an important component of the therapeutic armament for severe for1ns of symptomatic AIDS-KS

PAGE 104

97 The management of AIDS-KS still presents a challenge to clinicians and scientists. The development of novel therapeutic agents to improve outcomes in patients is badly needed. Because KS is a highly vasoactive neoplasm with a cellular origin suggested to be endothelial cell derived (Kroll and Shandera, 1998), we have investigated the efficacy of CA4DP in an in situ model of this disease. Results from previous studies were encouraging. Single dose of CA4DP treatment resulted in a rapid induction of vascular damage in the tumors such that 4 hr after treatment there existed an almost complete vascular shutdown (Figure 4-1 ). This was followed by haemorrhagic necrosis and extensive tumor cell killing in KS xenografts. Still despite the promising results achieved, cure with an individual modality (monotherapy) is often difficult and the application of vascular targeting approach will undoubtedly be investigated in a neoadjuvant setting. Since CA4DP itself is unlikely to eradicate all of the tumor, in the present studies we have examined the effects of combining CA4DP with radiation and chemotherapeutic agents in KS xenografts. Initial studies focused on the efficacy of combining CA4DP with radiation therapy. Previous studies showed that the vascular shutdown and subsequent induction of necrosis after CA4DP treatment is not complete, thus leaving areas of viable tumor cells from which the tumor could regrow (Li et al., 1998; Figure 4-4). Interestingly, cells surviving treatment with CA4DP tend to be located in areas at the tumor periphery near norrnal tissues, most likely those areas supplied by normal tissue vessels (Figure 4-4). This residual tissue is likely to be well oxygenated and hence responsive to radiation. Results from the clonogenic cell

PAGE 105

98 survival investigations demonstrated that combining CA4DP with radiation treatment could significantly enhance the tumor cell killing effect compared to radiation alone It is well established that the aberrant vascular morphology spatial heterogeneity in vessels, and metabolic microenvironment associated with solid tumors can have significant adverse effects on the efficacy of radiation therapy. Treatment with CA4DP eliminates many of these problem areas by causing extensive hemorrhagic necrosis in the centers of tumors (Li et al., 1998; Horsman et al., 1998). Results showed that when CA4DP was used in conjunction with radiotherapy the tumor's hypoxic cell population appeared to be significantly impacted For radiation treatment alone, the ''break'' in the KS cell survival curve reflects the presence of the hypoxic cell population in this tumor (Figure 5-1 ). It was observed that the inclusion of CA4DP in the treatment strategy altered the radiation dose response curve: the ''break'' seen with radiation treatment alone was eliminated with the combination treatment. It demonstrates that CA4DP may improve the radiation response of tumors by impacting the radiation refractory hypoxic cell subpopulation of tumors. Such a conclusion is consistent with previous histological evaluations in KS xenografts. 100 mg/kg CA4DP could induce massive central necrosis in KS tumors leaving viable cells only at the periphery (Figure 4-4). Taken together, these findings suggest that CA4DP and radiotherapy are acting in a complimentary fashion at the microregional level i e the vascular targeting agent is preferentially eliminating the poorly oxygenated and hence radioresistant tumor cell subpopulations.

PAGE 106

99 The present studies also investigated the efficacy of combining CA4DP with conventional chemotherapeutic agents. The rationale for such combinations was based on two factors First, it is logical to combine drugs or treatment modalities with different mechanisms of action or different dose-limiting toxicities to obtain a therapeutic benefit. Second, as discussed above, our CA4DP studies showed that this agent, though effective at inducing large scale tumor necrosis, fai led to eliminate a viable rim of tumor cells surviving at the tumor's periphery (Figure 44). Consequently, CA4DP treatment was combined with cisplatin or vinblastine therapy. Due to the rapid action of CA4DP against the tumor vasculature (Figure 4-2), in the combination studies, cisplatin and vinblastine treatment preceded CA4DP so as not to interfere with the tumor uptake of the chemotherapeutic agents. The results showed that the CA4DP -chemotherap y combination led to significantly enhanced tumor cell killing (Figures 5-3 and 5-4). Most critically this enhancement could still be achieved at the maximum tolerated single doses of cisplatin (20 mg/kg) and vinblastine (15 mg/kg). Clinically, the application of both vinblastine and cisplatin are limited by their toxic side effects, especially neurotoxicities. Indeed the long terminal elimination half life of vinblastine associated with its low elimination constant may explain its low MTD in patients (Zhou and Rahmani 1992) In contrast, the antivascular effects of CA4DP were achieved at doses le ss than 1 / 10 of the MTD and without detectable morbidity. In the present studies, it is clearly seen that CA4DP potentiates the anticancer effects of cisplatin and vinblastine significantly at very low doses. Therefore,

PAGE 107

100 alternatively, the same anti tumor effects could be achieved at far lower doses of cisplatin and vinblastine when CA4DP was included in the treatment. From previous studies we know that multiple doses of CA4DP were far more effective at inhibiting KS growth than single dose treatments. Still, because of the viable rim of tumor cells left after CA4DP treatment alone, we reasoned that administering a second agent along with CA4DP might prove to be a much more efficient application of this agent. Combining cisplatin with CA4DP proved to be more effective at inhibiting KS growth than either treatment alone (Figure 5-5, Table 5-1 ). The results showed that the combination of cisplatin and CA4DP led to a far superior growth delay and an additive anti-tumor effect in KS xenografts. This growth delay caused by the combination treatment was achieved with no more toxicity detected beyond that seen with cisplatin treatment alone Taken together, these data demonstrated that combining the antivascular agent CA4DP with another agent which kills the tumors cells directly is a feasible and effective way of inhibiting KS growth. In conclusion, the vascular targeting agent CA4DP could significantly enhance the anti tumor effects of radiation therapy and chemotherapeutic agents. These findings from the present studies illustrate that potential utility of employing a treatment strategy that combines a vascular targeting approach with radiation or chemotherapy to elicit increased antitumor efficacy in KS. Phase I clinical trials with CA4DP are nearly complete. The present results suggest that the evaluation of CA4DP in patients with AIDS-KS be considered in future Phase II studies.

PAGE 108

C: 0 ... (.) 0.1 'toC) 0.01 C > :::s u, 0 0.001 E ::s I0.0001 101 m m D Rx 0 Rx+100 mg/kg CA4DP .00001 -t-----------------0 5 10 15 20 25 Rx dose (Gy) Figure 5-1. Tumor cell survival in KS xenogra:fts treated with a 100 mg/kg dose of CA4DP 1 hr after a range of doses of radiation. Results are the mean SE of 3 experiments.

PAGE 109

a. 0.1 2 j 0) C > 0 01 2: i 0 0 0 001 Clsp latl n C 0.0001 -'-------,-----,-----,,----, 0 5 10 15 zo 25 D rug d,0se (M) c. doxorub lcln 1 tl \ C: 0 0 0 1 0 ai C: > 2: iil Cl "i 0 01 0 0.001 -+---------------0 5 10 15 20 Drug Dose (M) 10 2 b. d. 1 C: 0 0 1 t) Cl C: > ::I ii) "i 0 01 u \ c, Vlnbfastlne o --o ____ o 0 001 +-----------,-----,,--0 0.5 1 0 1 5 2 0 2 5 Drug dose (.M) VP16 1 ~""'-c 0 1 ti "' ~ dS .:: CII C: > ; ., "i 0.01 (,) 0 001 -t-----,----...---.--.------,----, 0 S 11) 15 20 30 Drug Dose (M) Figure 5-2. KSY-1 cell survival after 1-hr treatments with chemotherapeutic agents Clonogenic cell survival assay was performed 23 hr after drug treatment. Data are the mean SE of three experiments. a) Cisplatin; b) Vinblastine; c) Doxorubicin; d) VP-16.

PAGE 110

103 1 Cisplatin 0 Cisplatin + 100 mg/kg CA4DP C 0 0.1 ... u cu .... C, C > c= :l u, 0 E 0.01 :l t0.001 -+--------------------, 0 5 10 15 20 Drug dose (mg/kg) Figure 5-3. Tumor cell survival in KS xenografts treated with a 1 00 mg/kg CA4DP 30 min after a range of doses of cis pl atin treatme n t Data are the mean SE of a t l east 6 tumor s

PAGE 111

1 0 4 1 vinblastine Cl o vinblastine + 100 mg/kg CA4DP C \ 0 Q g .., 0.1 \ u I! u. C) C > :, en ... 0 0 01 E :, I0.001 ----------------0 5 10 15 20 Drug Dose (mg/kg) Figure 5-4. Tumor cell survival in KS xenografts treated with a 100 mg/kg CA4DP 30 min after a range of doses of vinblastine treatment. Data are the mean SE of at least 6 tumors.

PAGE 112

105 70 KS tumor growth a 60 median value cisplatin 10mg/kg 50 CA4DP 100 mg/kg C") MD(1 3 5) E E 0 a 0 en 40 0 8 ,,, --lca C 30 + 20 a 10 0-'-------------------------control cisplatin CA4DP CA4DP+cisp Figure 5-5. Growth delays in KS tumors with cisplatin, CA4DP, and CA4DP + cisplatin treatment. 10 mg/kg cisplatin was administered on day 1 for cisplatin treatment. 100 mg/kg CA4DP was given on days 1, 3, and 5 for CA4DP treatment. For the combination treatment. IO mg/kg cisplatin was given on day 1 followe d by three 100 mg/kg dose fractions of CA4DP on days 1, 3, and 5. Each data point represents an individual animal.

PAGE 113

106 Table 5-1. GROWTH DELAY IN KS XENOGRAFTS CAUSED BY CISPLA TIN AND CA4DP TREATMENT GROUP MEDIAN TIME GROWTH TO 900mm 3 DELAY (DAYS) (DAYS) 1. Control 10.0 2. Cisplatin 1 20.5 10.5 3. CA4DP(l 3 5) 2 30.0 20.0 4 CA4DP + cisp 3 38.0 28.0 SIGNIFICANCE 4 SIGNIFICANCE VS GROUP 1 -P < 0.025 4 vs 2 P < 0.025 P < 0.025 4 vs 3 P < 0.025 P < 0.025 1 Single dose of 10 mg/kg cisplatin given on day 1 (tumor size 200 mm 3 ). 2 100 mg/kg CA4DP admini s tered on day s 1 3, and 5. 3 10 mg/kg cisplatin given on day 1 followed by 100 mg/kg CA4DP on days 1 3, and 5. >10% WEIGHT LOSS 0 / 10 2 / 8 0 / 10 2 / 10 4 Determined by Wilcoxon Rank-Sum Test while compared the probability distributions of growth times of control versus treated tumor (a .025).

PAGE 114

CHAPTER6 s Y AND PERSPECTIVE The vascular targeting approach has received considerable attention in recent years. This type of therapy takes advantage of the uniqueness of the endothelium of established tumor capillaries and their supporting structures. The use of antivascular agents offers promise for the treatment of solid tumors as the growth of a tumor above a volume of 1 mm 3 inevitably demands vascularization to supply nutrients for proliferating tumor cells (Folkman, 1986). The current challenge is to develop treatments that are highly selective for the tumor vasculature, thus enabling systemic administration in well-tolerated regimes. The tubulin binding agent CA4DP may provide the lead for developing such selective vascular targeting drugs, based on its promising antivascular effects in preclinical models (Dark et al., 1997; Li et al., 1998; Horsman et al., 1998) and encouraging results from Phase I clinical trials (Randal, 2000). My studies during the last several years have focused on gaining insight into the mechanisms underlying the anti vascular action of CA4DP and examining the antitumor efficacy of CA4DP in a xenograft model of Kaposi's Sarcoma. The present studies den1onstrated that CA4DP not only inhibited tubulin polymerization, endothelial cell migration and attachment, but also induced endothelial cell apoptosis. An important finding from our studies was that CA4DP 107

PAGE 115

108 had selective toxicity against proliferating endothelial cells. By comparing the responses of tumor endothelial, and fibroblast cells to CA4DP, we can clearly see that proliferating endothelial cells are significantly more sensitive to CA4DP treatment Other in vitro studies indicated that CA4DP had the ability to interfere with endothelial cell behavior and this might be a way it exerts its action. Although further studies are needed to fully appreciate why endothelial cells are more sensitive to CA4DP treatment than other types of cells, our findings provide a basis for the in vivo efficacy of CA4DP. Two factors may possibly contribute to the high sensitivity of endothelial cells to CA4DP. First, endothelial cells may contain higher level of alkaline phosphatase than other cell types, resulting in an increased efficacy of CA4DP in endothelial cells. It has been observed that compared to fibroblasts endothelial cells showed higher alkaline phosphatase activity located on the cell membrane (Gallo et al., 1997) Second, endothelial cells may be more sensitive to their cell shape change than other cells. Therefore, with CA4DP treatment they could more easily detach and undergo apoptosis. Recent studies in our laboratory have focused on both these issues Currently we are attempting to measure the activity of alkaline phosphata se in different cell type s. A s tudy of the relationship between phosphatase activity and responsiveness to CA4DP in a range of cell types would provide further insight into the mechani s m of action of this drug. In addition, we have initiated in ves tigations using deconvolution microscopy to observe endothelial cell shape changes after CA4DP treatment on a 3-dimentional level.

PAGE 116

109 These studies will help to further elucidate the mechanisms responsible for the endothelial cell specific effects of CA4DP. The present studies also demonstrated significant antivascular effect of CA4DP in KS xenografts CA4DP caused rapid vascular shutdown which resulted in extensive haemorrhagic necrosis in these tumors. Multiple doses of CA4DP treatment alone already caused long and significant growth delays in KS tumors. Most importantly CA4DP caused these effects at doses less than 1 / 10 of the MTD making CA4DP a promising vascular targeting agent with a wide therapeutic window. The mechanisms of action of CA4DP are l ikely related to its tubulin-binding properties which may be responsible for the rapid endothelial cell damage neovascular shutdown and subsequent haemorrhagic necrosis. Drugs that work by this mechanism are unlikely to eradicate the tumor as a single agent but may enhance the efficacy of other anticancer treatments. That this may indeed be the case was supported by our in vivo investigations which demonstrated that enhanced tumor effects could be achieved by combining CA4DP with standard anticancer therapies such as radiation and chemotherapy. Most critically these enhancements were still achieved at the maximum tolerated doses of cisplatin and vinblastine. Both our in vitro and in vivo studies with CA4DP have demonstrated its strong anti vascular effect which is a common feature of tubulin-binding agents. However, CA4DP appears to be a much better vascular targeting agent compared to some other tubulin-binding agents such as colchicine and vinblastine. Recently

PAGE 117

110 a novel combretastatin A-4 derivative AC7700 (Hori et al., 1999) and another agent which has similar structure to combretastatin A-4, ZD6126 (being developed by AstraZaneca Pharmaceuticals, Wilmington, DE), have also been shown to have strong anti tumor effects due to their rapid shutdown of the tumor blood flow. This made us believe that the similar antivascular properties of CA4DP and the other two agents are closely related to their structural specificity. Moreover, another feature that makes CA4DP unique is its short half-life (30-45 min in nude mice) in vivo compared to that of vinblastine (24 hr) and colchicine (20 hr). The very short half-life of CA4DP allows this agent to act primarily on the vascular endothelial cells, making CA4DP a promising antivascular agent with very low level of toxicity seen in preclinical studies. CA4DP treatn1ent causes massive necrosis in KS tumors, which can induce inflammatory response in cancer patients. Local inflammatory response is helpful in getting rid of the necrotic cell debris in the tumor and also in killing some other survived tumor cells. However when the response mediators, for example, lymphokines and antibodies go into the blood stream, they could cause toxicities due to the non-specific inflammatory response. Therefore anti-inflammatory agents such as corticosteroid horrr1ones should be used along with CA4DP treatment if necessary Treatment for AIDS-KS continues to present a challenge for physicians. The dose-limiting toxicities of radiation and chemotherapy severely limit these therapies in AIDS-KS patients The present studies demonstrate that the combination of CA4DP with conventional anticancer therapies may prove to be

PAGE 118

111 an effective way to achieve a greater therapeutic benefit. The same antitumor effects could be gained at far lower doses of radiation and chemotherapeutic agents when CA4DP was included in the treatment. Alternatively, greater antitumor effectiveness was achieved at maximally tolerated conventional anticancer therapy when combined with CA4DP. Our findings therefore suggest a possible application for CA4DP in the clinical management of KS. Phase I clinical trials with CA4DP are nearly complete. The present results suggest that patients with AIDS-KS might benefit from CA4DP combination stud ie s in future Phase II studies. There are still areas of interests that could be pursued in future preclinical studies. For example our studies have shown an interesting interaction between CA4DP and single dose of radiation in the combination treatment. Future studies might focus on the effects of combining CA4DP with fractionated radiation therapy, since clinically radiation is nonnally given on a multiple-dose schedule to cancer patients. The combinat i on of CA4DP and multiple doses of anticancer agents might also be examined based on the same reason Moreover, the potential effect of CA4DP on tumor metastasis will be worthwhile to test in future preclinical investigations. The work shown in this dissertation has been presented at 5 national and international meetings. A portion of the studies of the in vivo effects of CA4DP has been published in the International Journal of Radiation Oncology Biology and Physics (Li et al. 1998). Another three manuscripts based on the studies from

PAGE 119

112 Chapter 2 to Chapter 5 have been finished and recently submitted for publication to the International Journal of Cancer and British Journal of Cancer

PAGE 120

REFERENCES Allalunis-Turner, M.J., and Siemann, D.W. 1986. Recovery of cell subpopulations from human tumor xenografts following dissociation with different enzymes. Br J Cancer 54:615-622 Amellem, 0., and Pettersen, E.O. 1991. Cell inactivation and cell cycle inhibition as induced by extreme hypoxia: the possible role of cell cycle arrest as a protection against hypoxia-induced lethal damage. Cell Prolif. 24:127-141 Augustin-Voss, H.G. and Pauli, B.U. 1992. Migrating endothelial cells are distinctly hyperglycosylated and express specific migration-associated glycoproteins. J Cell Biol. 119:383-391 Ausprunk, D. and Folkman, J. 1977. Migration and proliferation of endothelial cells in preformed and newly formed blood vessels during tumor angiogenesis. Microvasc Res. 14:53-65 Avile, J. 1992. Microtubule functions. Life Sci. 50:327-334 Baguley, B.C., Holdaway, K.H., Thomasen, L.L., Zhuang, L., and Zwi, L.J. 1991. Inhibition of growth of colon 38 adenocarcinoma by vinblastine and colchicine. Evidence for a vascular mechanism. Eur J Cancer 27:482 -4 87 Beauregard, D.A., Thelwall, E., Chaplin, D.J., Hill, S.A., Adams, G.E., Brindle, K.M. 1998. Magnetic resonance imaging and spectroscopy of combretastatin A4 prodrug-induced disruption of tumour perfusion and energetic status Br J Cancer 77(11): 1761-7 Blood, C.H., Zetter, B.R. 1990. Tumor interactions with the vasculature: angiogenesis and tumor metastasis. Biochem Biophys Acta 1032:89-118 Boehm, T., Folkman, J ., Browder, T., and O'Reilly, M.S. 1997. Antiangiogenic therapy of experimental cancer does not induce acquired drug resistance. Nature 3 90: 404-407 Bower, M., Howard, M. 1997. A phase II study of thalidomide of Kaposi's sarcoma: Activity and correlation with KSHV DNA load [abstract]. J Acquir Immune Defic Syndr Hum Retrovirol. 14:A35 113

PAGE 121

114 Braunhut, S.J., Medeiros D., Lai L. and Bump E.A. 1996. Tempol prevents impairment of the endothelial cell wound healing response caused by ionising radiation Br J Cancer S 157-S 160 Brizel, D.M. Scully S .P ., Harrelson J.M ., Layfield L J ., Bean J.M., Pro s nitz, L.R., and Dewhirst, M W. 1996. Tumor oxygenation predicts for the likelihood of distant metastases in human soft tissue sarcoma. Cancer Res. 56:941-943 Brooks P.C. Stromblad S ., Klemke, R ., Visscher, D., Sarkar F.H., and Cherish, D.A 1995. Anti-integrin in tumor vasculature. Clin Cancer Res. 1: 1623-1634 Burrow F.J., and Thorpe, P.E. 1993 Erad ic ation of large so lid tumors in mice with an immunotoxin directed again s t tumor vasculature Proc Natl Acad Sci 90:8996-9000 Cao, Y.H. Chen, A. An S.S A. Ji R.W.D ., Davidson D. Cao, Y.M. Linas M. 1997. Kringle-5 of pla s minogen is a novel inhibitor of endothelial cell growth. J Biol Chem. 272:22924-22928 Chabner, B.A., and Collins, J A. 1990. Cancer Chemotherapy: Principle s and Practice. JB Lippincott: Philadelphia Chaplin, D J. and Dougherty, G J 1999. Tumor vasculature as a target for cancer therapy. Br J Cancer 80 Suppl 1 : 57-64 Chaplin, D.J. and Horsman M.R. 1994. The influence of tumor temperature on ischemia induced cell death: potential implications for the evaluation of vascular mediated therapies. Radiother Oncol. 30:59-65

PAGE 122

115 Chaplin, D.J., Pettit, G.R., Parkins, C.S., and Hill, S.A. 1996. Antivascular approaches to solid tumor therapy: Evaluation of tubulin binding agents. Br. J. Cancer 74:S86 S88 Chaplin, D.J ., Pettit, G.R., Hill, S.A 1999 Anti-vascular approaches to solid tumor therapy: evaluation of combretastatin A4 phosphate. Anticancer Res. 19(1A): 189-95 Chiarot to J.A., Hill, R.P. 1999 A quantitative analysis of the reduction in oxygen levels required to induce up-regulation of vascular endothelial growth factor (VEGF) mRNA in cervical cancer cell lines. Br J Cancer 80: 1518-1524 Claffey, K.P., Robinson, G.S. 1996. Regulation of vegf/vpf expression in tumor cellsconsequences for tumor growth and metastasis. Cancer Metastasis Rev. 15:165-176 Clapp, C., Delaescalera, G.M. 1997. Prolactins novel regulators of angiogenesis. News Physiol Sci. 12 :231-23 7 Cliffe, S., Taylor, M.L., Rutland, M., Baguley, B.C., Hill, R.P., and Wilson, W.R. 1994. Combining bioreductive drugs (SR 4233 or SN 23862) with the vasoactive agents flavone acetic ac id or 5,6-dimethyxanthenone acetic acid. Int J Radiat Oncol Biol Phys. 29:373-377 Corbett, T.H., Bissery, M.C., Wozniak, A., Plowman, J., Polin, L., Tapazoglou, E., Dieckman, J. and Valeriote, F. 1986. Activity of flavone acetic acid against solid tumors of mice. New Drugs 4:207-220 Comali, E., Zietz, C., Benelli, R., Masiello, L., Breier, G., Tschachler, E., Albini, A., Sturzl, M. 1996 Vascular endothelial growth factor regulates angiogenesis and vascular perrrieability in Kaposi's sarcoma. Am J Pathol. 149:1851-69 Cucina, A., Sterpetti, A.V., Pupelis, G., Fragale, A., Lepidi, S., and Cavallaro, A. 1995. Shear stress induces changes in the morphology and cytoskeleton organisation of arterial endothelial cells. Eur J Vase Endovasc Surg. 9:86-92 Dark, G.G., Hill, S.A., Prise, V.E., Tozer, G.M., Pettit, G.R. and Chaplin, D.J. 1997. Combretastatin A-4, an agent that displays potent and se lective toxicity toward tumor va sculature. Cancer Res. 57: 1829-1834 Denekamp, J 1990. Vascular attack as a therapeutic strategy for cancer. Cancer Metast Re v. 9:267-282 Denekamp, J. 1993. Angiogenesis, neovascular proliferation and vascular pathophysiology as targets for cancer therapy. Br J Radial. 66: 181-196

PAGE 123

116 Denekamp J., and Hill, S. 1991. Angiogenic attack as a therapeutic strategy for cancer. Radiother Oneal. (Suppl.) 20:103-112 Denis, L.J ., and Verweij, J. 1997 Matrix metalloproteinase inhibitors: present achievements and future prospects. Invest New Drugs 15: 175-185 Deonarrain M P Spooner R.A ., and Epenetos A.A 1995. Genetic deli v ery of enzymes of cancer therapy Gene Therapy 24:235-244 Dewhirst, M.W. Tso C.Y. Oliver R. Gustafson, C.S. Secomb, T.W ., Gross, J .F 1989. Morphologic and hemodynamic comparison of tumor and healing normal tissue microvasculature Int J Radiat Oneal Biol Phys 17:91-99 Dezube, B J. Von Roenn J.H. Holden-Wiltse J Cheung T.W. Remick S.C. Cooley, T P. Moore J ., Sommadossi J.P. Shriver S.L. Suckow C W. Gill P.S. 1998. Fumagillin analog in the treatment of Kaposi s sarcoma: a phase I AIDS Clinical Trial Group study. AIDS Clinical Trial Group No 215 Team. I Clin Oneal. 16: 1444-9 Dvorak H.F. Detmar M. Claffey K.P. Nagy J.A. van de Water L ., and Senger D.R. 1995 Vascular permeability factor / vascular endothelial growth factor: an important mediator of angiogenesis in malignancy and inflammation. Int Arch Allergy Immunol. 107:2332 35 El-Zayat, A A., Degen D ., Drabek S ., Clark G M Pettit G.R. and Von Hoff D.D. 1993. In v itro e v aluation of the antineopla s tic acti v ity of combretastatin A-4 a natural product from Combretum Caffrum (arid shrub). Anticancer Drugs 4:1925 Ensoli, B. Markham P Kao V. Barillari G. Fiorelli, V ., Gendelman, R. Raffeld M ., Zon G. Gallo, R.C 1994 Block of AIDS-Kaposi's sarcoma (KS) cell growth angiogenesi s and lesion formation in nude mice by antisense oligonucleotide targeting basic fibroblast growth factor A novel strategy for the therapy of KS. J Clin Invest. 94(5) : 1736-46 Fan T P. Jaggar R. and Bicknell R. 1995 Controlling the vasculature: angiogenesis anti angiogenesis and vascular targeting of gene therapy. Trends Phartnacol Sci. 16:5 7 -66 Fenton B M. Siemann D.W. 1995. Are direct mea s ures of tumor oxygenation reflective of change s in tumor radiosensitivity following oxygen manipulation ? Acta Oneal. 34(3): 30 7 -11

PAGE 124

117 Folkman, J. 1986. How is blood vessel growth regulated in normal and neoplastic tissue? Cancer Res. 46:467 473 Folkman, J. 1995. Seminars in medicine of the Beth Israel Hospital Boston. Clinical applications of research on angiogenesis. N Engl J Med 333 : 1757-1763 Folkman, J. 1997 Addressing tumor blood vessel s Nat Biotechnol. 15(6): 510 Gallo R.L. Dorschner R.A. Takashima S. Klagsbrun M ., Eriksson E and Bemfield M. 1997. Endothelial cell surface alkaline phospl1atase activity i s induced by IL-6 relea s ed during wound repa i r. J Invest Dermatol. 109:597 603 Gallo R C. 1998. The enigmas of Kaposi's Sarcoma Science 282(5395): 1837-39 Gascon, P. and Schwart s, R.A. 2000. Kapo s i's Sarcoma. New treatment modalities. Dermatol C lin. 18(1 ): 16975 Gerweck L .E. Seetharaman K. 1996. Cellular pH gradient in tumor versus nor111al tissue: potential exploitation for the treatment of cancer Cancer Res 56:1194-1198 Giannakakou P Villalba L., Li H. Poruchynsky M., and Fojo T. 1998. Combinations of paclitaxel and vinblastine and their effects on tubulin polymeri z ation and cellular cytotoxicity: characterization of a synergistic schedule Int J Cancer 75:57-63 Gorczyca W ., Gone J. Ardelt B. Tragano F. and Darzynkiewic z, Z. 1993. The cell cycle related differences in su s ceptibility of HL-60 cells to apoptosi s induced by various antitumor agents. Cancer Res. 53:3186-3192 Groopman J.E 1987. Biology and therapy of epidemic Kaposi s sarcoma. Cancer 59:633-637 Grosios K. Holwell S E. McGown A.T ., Pettit, G R. Bibby M.C. 1999 In vivo and in vitro evaluation of combretastatin A-4 and its sodium phosphate prodrug Br J Cancer 81 (8): 1318-27 Grosios K ., Loadman P M. Swaine D.J Pettit G.R ., Bibby M.C. 2000. Combination chemotherapy with combreta s tatin A-4 phosphate and 5-fluorouracil in an experimental murine colon adenocarcinoma. Anticancer Res. 20(1A) : 22933 Grunt T W. Lametschwandtner A. and Staindl 0. 1985. The vascular pattern of basal cell tumor: light microscopy and s canning electron microscopic stud y on vascular corrosion ca s t s. Microvasc Res 2 9:371-386

PAGE 125

118 Harris, A L. 1998. Anti-angiogenesis therapy and strategies for integrating it with adjuvant therapy. Recent Results Cancer Res. 152:341-52 Hill, D.R. 1987. The role of radiotherapy for epidemic Kaposi's sarcoma. Semin Oncol. 14(2 suppl 3): 19-22 Hill, S A., Sampson L E., and Chaplin D.J. 1995 Anti-vascular approaches to solid tumor therapy: Evaluation of vinblastine and flavone acetic acid. Int J Cancer 63:119-123 Hirst D ., Wood, P.J. 1987. The influence of hemoglobin affinity for oxygen on tumor radiosensitivity. Br J Cancer 55:487-91 Hobson, B. and Denekamp, J. 1984. Endothelial proliferation in tumors and nortnal tissue: Continuous labeling studies. Br J Cancer 49 : 405-413 Hori, K., Saito S ., Nihei Y., and Sata, Y. 1999. Antitumor effects due to irreversible stoppage of tumor tissue blood flow : evaluation of a novel combretastatin A-4 derivative, AC7700. Jpn J Cancer Res 90(9) : 1026-38 Horsman M.R., Ehrnrooth, E. Ladekarl M Overgaard J 1998. The effect of combretastatin A-4 di s odium phosphate in a C3H mouse mammary carcinoma and a variety of murine spontaneous tumors. Int J Radiat Oncol Biol Phys 42( 4): 895-8 Horsman, M.R. Nord s mark M. Khalil A.A. Hill S.A. Chaplin, D.J Siemann D.W. Overgaard, J. 1994. Reducing acute and chronic hypoxia in tumors by combining nicotinamide with carbogen breathing. Acta Oncol. 33:371-6 Huang X. Molema G ., King, S ., Satkin s, L. Edgington T.S ., and Thorpe P E 1997 Tumor infarction in mice by antibody-directed targeting of tissue factor to tumor vasculature. Science 24:547-550 Im S.A., Gome z -Man z ano C., Fueyo J. Liu T.J ., Ke L D. Kim, J.S Lee H.Y. Steck P.A ., Kyritsis, A P., Yung W.K. 1999 Antiangiogenesis treatment for gliomas: transfer of antisense-vascular endothelial growth factor inhibits tumor growth in vivo Cancer Res 59(4): 895-900 Ingber D. Fujita T ., Ki s himoto S ., Sudo K ., Kanamaru, T. Brem H. and Folkman J 1990. Synthetic analogues of fumagillin that inhibit angiogene s is and suppress tumor growth Nature 348:555-557 Iyer S. Chaplin D.J ., Rosenthal D.S. Boulares A.H. Li L.Y ., Smulson M E. 1998. Induction of apoptosis in proliferating human endothelial cells by the

PAGE 126

119 tumor-specific antiangiogenesis agent Combretastatin A-4. Cancer Res. 58:45104514 Jaffe, E.A. 1987. Cell biology of endothelial cells. Hum Pathol. 18:234-239 Jain, R.K., Koenig, G.C., Dellian, M., Fukumura, D., Munn, L.L., Melder, R.J. 1996. Leukocyte-endothelial adhesion and angiogenesis in tumors. Cancer Metastasis Rev. 15:195-204 Kerr, J.F.R., Wyllie, A.H., and Currie, A.R 1972. Apoptosis: a basic biological phenomenon with wide ranging implications in tissue kinetics. Br J Cancer 26 : 239 -357 Kim, K.J., Li, B., Winer, J., Armanini, M., Gillett, N., Phillips, H.S., and Ferrara, N. 1993. Inhibition of vascular endothelial growth factor-induced angiogenesis supp res ses tumor growth in vivo. Nature 362:841-844 Konerding, M.A., Miodonski, A.J., and Lametschwandtner, A. 1995. Microvascular corrosion casting in the study of tumor vascu larity: a review. Scanning Microsc. 9:1233-1244 Kroll, M.H., and Shandera, W.X. 1998. AIDS-Associated Kaposi's Sarcoma. Hosp Pract. 33( 4): 85-8, 95-6, 99-102 Li, L., Rojiani, A., and Siemann, D.W. 1998. Targeting the tumor vasculature with combretastatin A-4 disodium phosphate: effects on radiation therapy. Int J Radiat Oncol Biol Phys. 42( 4): 899-903 Lilenbaum, R.C., and Ratner, L. 1994. Systemic treatment of Kaposi's sarcoma: current status and future directions. AIDS 8: 141-151 Lin, C.M., Ho, H.H., Pettit, G R Hamel, E. Antirnitotic natural products combretastatin A-4 and combretastatin A-2: studies on the mechanism of their inhibition of the binding of colchicine to tubulin. Biochemistry 28(17): 6984-91. Ludford, R.J. 1945 Colchicine in the experimental chemotherapy of cancer. J Natl Cancer Inst. 6:89-101 Lunardi-Iskandar, Y., Gill, P., Lam, V.H., Zeman, R.A., Michaels, F., Mann, D.L., Reitz, M.S. Jr, Kaplan, M., Bememan, Z.N., Carter, D. 1995. Isolation and characterization of an immortal neoplastic cell line (KSY-1) from AIDS associated Kaposi's sarcoma. J Natl Cancer Inst. 87(13): 974-981 Mahadevan, V., Malik, S.T., Meager, A., Fiers, W., Lewis, G.P., and Hart, I.R. 1990. Role of tumor necrosis factor in flavone acetic acid induced vascular shutdown. Cancer Res. 50:5537-5542

PAGE 127

120 Marshall, J L., and Hawkins, M.J. 1995. The clinical experience with antiangiogenic agents. Breast Cancer Res Treat. 36:253-261 McGarvey M.E ., Tulpule, A., Cai, J Zheng, T., Masood R., Espina, B ., Arora N., Smith, D.L., and Gill, P .S. 1998 Emerging treatments for epidemic (AIDS related) Kaposi's sarcoma. Curr Opin Oncol. 10(5): 413-21 Mesiano S Ferrara, N., Jeffe R.B 1998. Role of vascular endothelial growth factor in ovaiian cancer: inhibition of ascites formation by immunoneutrali zatio n. Am J Pathol. 53(4): 1249-56 Morris, A.K ., Valley A.W. 1996. Overview of the management of AIDS-related Kaposi's sarcoma. Ann Pharmacother. 30(10): 1150-63 Nakamura, S., Sakurada S., Salahuddin S.Z ., Osada Y., Tanaka, N.G., Sakamoto, N Sekiguchi, M. Gallo, R.C 1992 Inhibition of development of Kaposi's sarcoma related lesions by a bacterial cell wall complex. Science 2 5 5: 14 3 7-40 Namiki, A., Brogi E., Kearney, M. 1995 Hypoxia induces vascular endothelial growth factor in cultured human endothelial cells. J Biol Chem. 270:31189-31195 Nihei, Y. Suzuki, M., Okano, A. Tsuji, T., Akiyama, Y., Tsuruo, T., Saito, S. Hori K. Sato, Y. 1999. Evaluation of antivascular and antimitotic effects of tubulin binding agents in solid tumor therapy. Jpn J Cancer Res. 90(12): 1387-95 Nordsmark, M., Bent zen, S.M. and Overgaard, J 1994. Measurement of human tumor oxygenation status by a polarographic needle electrode Acta Oncol. 33:383-389 O'Brien, C. 1997. Combretastatin A-4 targets vascular damage to tumors Mol Med Today 3:369 O'Reilly, M.S., Holmgren L. Shing, Y., Chen, C., Rosenthal R.A Moses M., Lane, W.S ., Cao, Y., Sage, E.H., and Folkman, J. 1994. Angiostatin: a novel angiogenesis inhibitor that mediates the suppression of metastases by a lewi s lung carcinoma. Cell 79:315-328 Pallavicini M G., Laland, M .E., Millwe, R.G., and Hill, R.P 1979. Cell-cycle distribution of chronically hypoxic cells and determination of the clonogenic potential of cells accumulated in 0 2 -M phases after irradiation of a solid tumor. Cancer Res. 39:1891-1897

PAGE 128

121 Parentesis, J.P., Miller S P ., and Bodley, J.W. 1992. Protein toxin inhibitors of protein synthesis. Biofactors 3: 173-184 Parkins C.S., Chadwick, J A ., Chaplin, D.J. 1996. Inhibition of intracellular pH control and relationship to cytotoxicity of chlorambucil and vinblastine. Br J. Cancer Suppl 74:75-77 Pearson J.D. 1991. Endothelial cell biology. Radiology 179:9 14 Pettit, G.R. Cregg, G.M., and Singh, S.B 1987. Antineoplastic agents 122. Constituents of Combretum Caffrum. J Nat Prod. 50:386-391 Pettit G R., Singh S.B., Hamel, E., Lin, C.M., Alberts, D.S. and Garcia-Kendall D 1989 Isolation and structure of the strong cell growth and tubulin inhibitor combretastatin A-4. Experientia 45:209-211 Phillips T L. Wa sse rman, T H 1984 Promise of radiosensiti ze rs and radioprotectors in the treatment of human cancer. Cancer Treat Rep 68(1) : 291302 Pluda J.M. 1997 Tumor-associated angiogene s i s mechanisms clinical implications, and ther a peutic strategies. Semin Oncol. 24:203-218 Pluda, J.M. Parkinson, D.R. Fiegal E., Yarchoan R. 1993 Non-cytotoxic therapeutic approaches to the treatment of HIV associated Kaposi's sarcoma. Oncology 7: 25-33 Randal J 2000. Antiangiogenesis drugs target s pecific cancers mechani s m s. I Natl Cancer Inst. 92(7): 520-522 Reynolds T.Y., Rockwell S. Glazer, P.M. 1996. Genetic in s tability induced by the tumor microenvironment Cancer Res. 56(24) : 5754-57 Rojiani A Li L. and Siemann D W 2000. Characterization of a xenograft model of AIDS-associated Kaposi 's Sarcoma (AIDS-KS) for preclinical therapeutic investigations. Proceedings of the 91 st AACR annual meeting. p435: # 2768 Rutger s, J.L. Wiec zor ek R ., Bonetti F 1986 The expres s ion of endothelial cell s urface antigen by AIDS-associated Kapo s i 's Sarcoma Evidence for a vascular endothelial cell origin Am J Pathol 122:493-499 Sackett D L. 1993. P o dophyllotoxin steganacin and combretastatin : natural products that bind at the colchicine s ite of tubulin Pharmacol Tuer. 59: 163-2 28

PAGE 129

122 Saville M.W., Lietzau J., Pluda, J M., Feuerstein, I., Odom, J., Wilson W.H ., Humphrey, R W., Feigal, E., Steinberg S.M. Broder S. 1995 Treatment of HIV associated Kaposi's sarcoma with paclitaxel. Lancet 346:26-8 Schwartsmann, G., Stefani, S. and Villarroel R U. 1998. The systemin treatment of AIDS-related Kaposi' s sarcoma. Cancer Treat Reviews 24:415-424 Schweigerer L. 1995. Antiangiogenesis as a novel therapeutic concept in pediatric oncology J Mol Med. 73:497-508 Scott, P ., and Harris, A.L. 1994. Current approaches to targeting cancer using antiangiogenesis therapies. Cancer Treat Rev 20:393-412 Shah-Yukich A.A., and Nelson, A.C. 1988. Characterization of solid tumor microvasculature: a three-dimensional analysis using the polymer casting technique. Lab Invest 58:236-244 Shweiki, D., Itin A., Soffer, D., and Keshet, E. 1992. Vascular endothelial growth factor induced by hypoxia may mediate hypoxia-initiated angiogenesis. Nature (Lond ) 359:843-845 Siemann, D.W. 1995 Chemosensitization of CCNU in KHT murine tumor cells in vivo and in vitro by the agent RB6145 and its isomer PD 144872. Radiother Oncol. 34:47-53 Siemann, D.W. Horsman M R Chaplin, D J. 1994. The radiation response of KHT sarcomas following nicotinamide treatment and carbogen breathing. Radiother Oncol 31: 177-82 Siemann, D.W., Johansen I M., and Horsman M R. 1998. Radiobiological hypoxia in the KHT sarcoma: predictions using the eppendorfHistograph. Int J Radiat Oncol Biol Phy s. 40(5): 1171-1176 Siemann, D.W ., Macle1 L.M. 1986. Tumor radiosensitization through reductions in hemoglobin affmity. Int J Radiat Oncol Biol Phys 12:1295 97 Siemeister G., Martiny-Baron G., Marme D 1998. The pivotal role of VEGF in tumor angiogenesis: molecular facts and therapeutic opportunities Cancer Met Revs 17 :241-248 Skobe M ., Vasilopoulo s, P Goldstein N., Vosseler S ., Fusenig, N.E. 1997 Halting angiogenesis suppresses carcinoma cell invasion Nat Med. 3: 1222 1227 Stewart B.W. 1994 Mechanisms of apoptosis: integration of genetic, biochemical, and cellular indicators. J Natl Cancer Inst. 86:1286-1295

PAGE 130

123 Sundfor, K., Lyng, H., and Rofstad, E.K. 1998. Tumor hypoxia and vascular density as predictors of metastasis in squamous ce l l carcinoma of the uterine cervix. Br J Cancer 78: 822-827 Sung, J.C.Y., Louie S G ., Park, S.Y. 1997. Kaposi 's sarcoma: advances in tumor biology and pharmacotherapy. Pharmacotherapy 17(4): 670-683 Tannock, I.F. 1970. Population kinetics of carcinoma cells capillary endothelial cells and fibroblasts in a transplanted mouse mammary tumour Cancer Res. 30: 2470 2474 Tappero, J.W. Conant, M.A., Wolfe, S.F. 1993. Kaposi's sarcoma: epidemiology pathogenesis, histology clinical spectrum, staging criteria and therapy. J Am Acad De1matol. 28:371-395 Taraboletti G Beloti D Borsotti P ., Vergani, V ., Rusnati M., Presta, M. Giavazzi R. 1997. The 140 kilodalton antiangiogenic fragment of thrombospondin 1 binds to basic fibroblast growth factor. Cell Growth Differ 8:471 479 Teicher, B.A. 1994. Hypoxia and drug resistance. Cancer Metastasis Rev. 13(2): 139-68 Teicher B.A. 1995. Angiogenesis and cancer metastases therapeutic approaches. Crit Rev Oncol Hematol. 20:9-39 Tomida, A Tsuruo T. 1999. Drug resistance mediated by the cellular response to the microenvironment of solid tumors. Anticancer Drug Des 14(2): 16977 Tozer, G M., Prise, V .E Wilson J Locke, R J Vojnovic B Stratford, M.R. Dennis, M F ., Chaplin, D J. 1999. Combretastatin A-4 phosphate as a tumor vascular-targeting agent : early effects in tumors and nortnal tissues. Cancer Res. 59(7): 1626-34 Twardowski P ., and Gradishar, W.J 1997 Clinical trials of antiangiogenic agents. Curr Opin Oncol. 9 : 584-589 Vaupel P 1994. Blood flow oxygenation, tissue pH distribution, and bioenergetic status of tumors. Berlin Ge1rr1any, Bruckeve: Hellmich KG Vaupel, P., Kallinowski F Okunieff P 1989. Blood flow oxygen and nutrient supply, and metabolic microenvironment of human tumors: A review Cancer Res. 49:6449-6465

PAGE 131

124 Vaupel, P ., Menke, H. 1989. Effect of various calcium antagonists on blood flow and red blood cell flux in malignant tumors. Prog Appl Microcirc. 14:88-103 Vaupel, P Schlenger, K., Knoop, C., Hockel, M. 1991. Oxygenation of human tumor: evaluation of tissue oxygen distribution in breast cancers by computerized 0 2 tension measurements. Cancer Res. 51 :3316-3322 Vaupel, P., Thews, 0., Hoeckel, M. 1996. Tumor oxygenation: characterization and clinical implications. In : rhErythropoietin in cancer supportive treatment. Marcel Dekker, Inc. pp. 205 239 Walsh. 1844. The anatomy, physiology, pathology and treatment of Cancer. London Warren, B.A. 1979. The vascular morphology of tumors. In : Tumor Blood Circulation. H.J Petersen, de., CRC Press, Boca Raton, FL. pp. 1-47 Watts, M E., Woodcock, M., Arnold, S., and Chaplin, D.J. 1997. Effects of novel and conventional anti-cancer agents on human endothelial cell permeability: Influence of tumor secreted factors. Anticancer Res 147:329-334 Weich, H.A., Iberg, N., Klagsbrun, M., and Folkman, J. 1991. Transcriptional regulation of basic fibroblast growth factor gene expression in capillary endothelial cells. J Cell Biochem.47: 158-164 Welles, L., Little, R., and Wyvill, K. 1997. Preliminary results of a phase II study of oral thalidomide in patients with HIV infection and Kaposi's sarcoma. J Acquir Immune Defic Syndr Hum Retrovirol. l 4:A21 Wike-Hooley, J.L., Haveman, J., Reinhold, H.S. 1984. The relevance of tumor pH to the treatment of malignant disease. Radiother Oncol. 2:343-366 Witte, L., Hicklin, D.J., Zhu, Z., Pytowski, B., Kotanides, H., Rockwell, P., Bohlen, P. 1998. Monoclonal antibodies targeting the VEGF receptor-2 (Flkl/KDR) as an anti-angiogenic therapeutic strategy. Cancer Metastasis Rev. 17(2): 155-61 Woglum, W H. 1923. A critique of tumor resistance. J Cancer Res. 7:283-311 Woods, J.A., Hadfield, J.A., Rettit, G.R., Fox, B.W., and McGown, A.T. 1995. The interaction with tubulin of a series of stilbenes based on combretastatin A-4. Br J. Cancer 71:705-711 Workman, P. 1994. Enzyme-directed bioreductive drug development revisited: A commentary on recent progress and future prospects with emphasis on quinone

PAGE 132

1 2 5 anticancer agents and quinone metabolizing enzymes particularly DT-diaphorase Oncol Res. 6(10-11): 461-475 Yarchoan R 1999. Therapy for Kaposi' s s arcoma: recent advances and experimental approaches J Acquir Immune Defic Syndr. 21 Suppl 1: S6673 Zhang H-T. Craft P ., Scott P.A. E Ziche M ., Weich H.A. Harri s, A.L. Bicknell R. 1995. Enhancement of tumor growth and v a s cular density by transfection of va s cular endothelial cell growth factor into M C F-7 human brea s t carcinoma cells. J Natl C ancer In s t. 87 : 2132 19 Zhou X.J and Rahmani R 1992. Preclini c al and c linical pharmacology of vinca alkaloids. Drug s 44(Suppl 4): 1 16 Zwi L.J ., Baguley B .C., Gavin J.B ., Wilson W.R 1994 The morphological effects of the anti-tum o r agent s fla v one acetic acid and 5 6-dimethyl xanthenone acetic acid on the col o n 38 mou s e tumor Pathology 26( 2 ): 161-9

PAGE 133

BIOGRAPHICAL SKETCH Lingyun Li was born December 31 1972 in Nenjiang P.R China, to Jingtian Li and Shufen Wang. She received her elementary and secondary education in Beijing P.R.China and graduated high school in 1991. She received a Bachelor of Medicine degree in basic biomedical sciences from Beijing Medical University Beijing P R China in 1996. She began her doctoral studies in the Interdisciplinary Program in College of Medicine University of Florida, Gainesville, Florida, in August 1996. In 1997 s he joined the Department of Pharmacology and Therapeutics and continued her doctoral studies under the mentorship of Dr. Dietmar W Siemann After graduation she will continue her research in anti cancer therapy. 1 26

PAGE 134

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. ~t~~==-----.. Dietmar W Siemann, Chair Professor of Pharmacology and Therapeutics I certify that I h ave read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and i s fully adequate in scope and quality as a dissertation for the degree of Doctor of Philo so phy. --------homas C. Rowe A ssoc iate Professor of Pharmacology and Therapeutics I certify that I ha ve read thi s study and that in my opinion it confortns to acceptable standards of sc holarl y presentation and is fully adequate, in sco pe and quality as a dis s ertation for the degree of Doctor of Philosophy. Kathleen T. Shiverick Professor of Pharmacology and Therapeutics I certify that I have read this study and that in my opinion it conforms to acceptable s tandards of sc holarly presentation and i s fully adequate, in sco pe and quality, as a dissertation for the degree of Doctor of Philosophy. J es R. Zucali Professor of Molevular Genetics and Microbiology This dissertation was submitted to the Graduate Faculty of the College of Medicine and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. December 2000 Dean, Graduate School

PAGE 135

UNIVERSITY OF FLORIDA II I 1111111 II 11 I I 3 1262 08556 97 46