Differential expression of bcl-2 by hematological tumors


Material Information

Differential expression of bcl-2 by hematological tumors anti-apoptosis function and chemotherapy resistance
Alternate title:
Anti-apoptosis function and chemotherapy resistance
Physical Description:
xi, 101 leaves : ill. ; 29 cm.
Guedez, Liliana, 1956-
Publication Date:


Subjects / Keywords:
Genes, bcl-2   ( mesh )
Proto-Oncogene Proteins c-bcl-2   ( mesh )
Drug Resistance, Neoplasm -- physiology   ( mesh )
Hematologic Neoplasms -- physiopathology   ( mesh )
Cytarabine -- pharmacology   ( mesh )
Cytarabine -- drug effects   ( mesh )
Bleomycin -- pharmacology   ( mesh )
Bleomycin -- drug effects   ( mesh )
Apoptosis -- physiology   ( mesh )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph. D.)--University of Florida, 1995.
Includes bibliographical references (leaves 93-99).
Statement of Responsibility:
by Liliana Guedez.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
oclc - 49808345
System ID:

Full Text







This Dissertation is dedicated to my parents' memory,

Luz I. Machado and Oscar A. Guedez.


My family has been my main source of inspiration and

encouragement, and without them I would not have reached my

academic goals. I am deeply grateful to my parents, who are

no longer with me, but who continue to be my best role

models. Most of all, I thank my son Oscar for being so

patient, and for letting me take all this time for my

intellectual growth.

Special thanks go to my mentor James Zucali; without

his dedication, direction and scientific advice, this

dissertation would not have been possible. I would like also

to thank the rest of my committee members, Maureen Goodenow,

Raul Braylan, Saed Khan, and Christopher West.

I also thank my friends Maria Elena Bottazzi, Jayoo

Gokhale and Christy Myrick for being with me in the bad and

good times of my life in Gainesville.



ACKNOWLEDGMENTS ........................................ iii

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

ABSTRACT .................................................. ix


1. BACKGROUND AND SIGNIFICANCE........................ 1

Introduction ....................................... 1
Apoptosis and Cancer............................... 3
Oncogenes Controlling Apoptosis................. 5
Bcl-2 Oncoprotein................................... 6
Antioxidant Function............................ 9
Protein Dimerization........................... 10
Resistance to Cancer Treatment................. 14

2. MATERIALS AND METHODS ............................. 17

Molecular Biology Techniques...................... 17
Protein Extraction and Western Blots............ 17
RNA Extraction and Northern Blots.............. 18
Retroviral Constructs,Electroporation and
G418 Selection ............................... 19
Plasmids and Probes............................ 20
Low Molecular Weight DNA Extraction............ 20
Tissue Culture Techniques.........................22
Cell Lines and Growth Media.................... 22
Drugs ...........................................23
Clonogenic Assay............................... 23
Viability Assay................................ 24
Cell Cloning.... .............................. 24
Flow Cytometric Analysis.......................... 25
Staining of Intracellular Antigen.............. 25

Staining of Cell Surface Antigen............... 26
DNA Fragmentation............................. 26
Cell Cycle Analysis............................ 27

DIFFERENT LEVELS OF BCL-2....................... 29

Introduction...................... ............... 29
Bcl-2 Expression by Hematological Tumors..........30
Ara-C Dose-Response ...............................32
Apoptosis. ........................................ 35
Bcl-2 Expression during Ara-C Action.............. 40
Conclusion........................................ 41


Introduction ..................................... 43
Murine Bel-2 Expression .......................... 44
Bcl-2 Induced Chemoresistance..................... 48
Decreased Apoptosis by Up-regulated Bcl-2......... 49
Conclusion........................................ 52

A FREE RADICAL INDUCER ......................... 57

Introduction. ...................................... 57
Bleomycin Dose Response........................... 58
Bleomycin Effect on the Growth and Morphology
of Transfected U-937 Cells ..................... 60
Cell Death Post-Bleomycin Differentiation.......... 67
Apoptosis.......................................... 72
Conclusion .......................................... 76

6. DISCUSSION........................................ 78

Bcl-2 Levels and the Resistance to Ara-C.......... 79
Up-regulated Bcl-2 in Myeloid Cells Treated
with Bleomycin ............................... 83

Extended viability of Differentiating Myeloid
Leukemia by Up-regulated Bcl-2 ................. 87
Summary............................................ 88
Future Direction.................................. 89

REFERENCE LIST .............................................. 93

BIBLIOGRAPHICAL SKETCH .................................. 100


Figure page

1. The antioxidant mechanism of Bcl-2 ................... 11

2. Dimerization of Bcl-2 and Bax ........................ 13

3. Construct of murine Bcl-2-retroviral vector ..........21

4. Western analysis of human Bcl-2 in three
hematological tumors ............................. 31

5. Ara-C dose-response curves for three
hematological tumors.............................. 33

6. Flow cytometry analysis of Ara-C induced
apoptosis.......................................... 34

7. Effect of Ara-C on the DNA fragmentation of
hematological tumors ............................. 36

8. Flow cytometry analysis of Bcl-2 during Ara-C
action ............................................. 38

9. Western analysis of Bcl-2 during Ara-C action ........39

10. Western and Northern analysis of murine Bcl-2
in U-937 cells ................................... 46

11. Western analysis of murine Bcl-2 in subcloned
U-937 cells ...................................... 47

12. Viability of Ara-C treated U-937 cells ...............50


13. Ara-C dose-response of bcl-2-U937
cell clones ...................................... 51

14. Quantitation of apoptosis in Ara-C treated MNC-
U-937 cells....................................... 55

15. Bleomycin dose-response for colony formation by
U-937 cells...................................... 59

16. Bleomycin effect on the total number of viable
transfected U-937 cells .......................... 61

17. Effect of bleomycin on the cell cycle ................64

18. Effect of bleomycin on cell morphology ...............66

19. Morphology of U-937 cells recovering from
bleomycin treatment .............................. 70

20. The number of viable differentiated cells after
bleomycin treatment .............................. 74

21. Apoptosis after 5 day bleomycin treatment ............75


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



Liliana Guedez

December, 1995

Chairman: Dr. James R. Zucali
Major Department: Pathology and Laboratory Medicine

Apoptosis can be induced by several stimuli including

anti-neoplastic agents. Deregulation of genes controlling

apoptosis has been shown to increase chemotherapy

resistance. One of these genes is bcl-2. Although Bcl-2

oncoprotein has been extensively studied in both T and B-

cells, little is known about its expression and function in

non-lymphoid tumors. The long-term goal of this dissertation

was to study the effect of Bcl-2 on the resistance of

lymphoid and non-lymphoid tumors to chemotherapy and to

determine the role of Bcl-2 in the chemotherapy resistance

of myeloid leukemias. First, the effect of 1-P-d-

arabinofuranosyl-cytosine (Ara-C) was examined on hema-

tological tumors expressing different levels of Bcl-2. The

follicular-lymphoma cell line RL-7 expressing the highest

Bcl-2 levels also showed the highest colony efficiency. The

pre-myeloid leukemia KG-1 expressing intermediate Bcl-2

levels showed the next highest colony efficiency at <10 iM

of Ara-C. In contrast, the leukemic U-937 cells which

expressed the lowest Bcl-2 levels was highly sensitive to


U-937 and KG-1 cells demonstrated DNA fragmentation

when treated with 100 iM Ara-C for one hour or 24 hours

respectively, whereas RL-7 cells displayed no DNA

fragmentation. KG-1 apoptosis was not to due to a complete

Bcl-2 downregulation, since KG-1 cells treated with 100 pM

Ara-C showed a decrease in the Bcl-2 oncoprotein of only 35-

40%. To further examine the direct role of Bcl-2 in the

protection of myeloid tumors against Ara-C, U-937 cells were

transfected with murine bcl-2. The bcl-2-U-937 cells showed

a significant decrease in apoptosis, and an increase in

clonogenic survival. In contrast, U-937 cells carrying the

MNC vector alone underwent apoptosis and showed reduced

colony efficiency at Ara-C <0.01 pM. These results

demonstrate that up-regulation of Bcl-2 in myeloid cells

decreases apoptosis and maintains cell viability.

Treatment of MNC-U-937 and bcl-2-U-937 cells with

sublethal concentrations of bleomycin induced decreased cell

growth and increased differentiation either via accumulation

in GO/G1 or with lack of cell cycle synchronization

respectively. Following differentiation, MNC-U-937 cells die

by apoptosis. In contrast, differentiated bcl-2-U-937 cells

remained viable for two weeks following bleomycin treatment.

This result suggests that Bcl-2 upregulation blocks

apoptosis in cells induced to differentiate.




The long term goal of this investigation was to study

the role of Bcl-2 on the resistance of lymphoid and non-

lymphoid tumors to chemotherapy. In addition, this study

will elucidate the mechanism of Bcl-2 protection in myeloid


Apoptosis, also called programmed cell death, is an

endogenous cellular process whereby an external signal

activates a metabolic pathway that results in cell death.

Cells in apoptosis show common morphological features such

as cell shrinkage, cell surface convolution which results in

the explosion of the cell into a series of membrane-bound

structures called apoptotic bodies, and fragmentation of DNA

into nucleosomal-size fragments (Compton, 1992; Fesus,1993;

Wyllie,1993). Both cell death and cell proliferation are

mandatory for the success of embryonic development and the

renewal of adult tissues. During evolution, apoptosis has

been modified in a cell and tissue specific manner to

accomplish a number of physiological functions. These

include elimination of cells that (1) have no function,(2)

are generated in excess, (3) develop improperly, (4) have

already completed their life span or (5) are harmful. When

aberrant apoptosis occurs, disease arises. For example, the

repertoire of the mammalian immune system is selected by

both positive and negative forces to extinguish the vast

majority of cells which lack appropriate specificities

(McCarthy et al.,1992). Lack of apoptosis during the

development of the immune system favors abnormal conditions

such as autoimmune disorders (Akbar et al.,1993; Garchon,

1992; Mountz and Gause, 1993). Similarly, hematopoiesis is a

tightly regulated process in which cellular proliferation

and differentiation are controlled by several growth

factors. Slow-growing stem cells give rise to progenitors

which then increase the rate of proliferation and more cells

are generated as maturation proceeds. The pool of mature

cells is maintained by balancing the rate of cell


proliferation and the rate of cell death by apoptosis

(Koury, 1992). Induction and inhibition of apoptosis are

assumed to be genetically controlled during hematopoiesis.

When such genetic signals become deregulated, for instance

in altered genes whose products control apoptosis,

illegitimate cell survival may result(Kerr and Harmon,

1991). The extended survival of cells that are destined to

die may prove to be a key event in increasing the chance to

acquire additional genetic defects and eventually contribute

to converting those cells into a malignant cancer. Lack of

apoptosis in tissues such as the bone marrow which displays

a high renewal rate, can also favor the development of


Apoptosis and Cancer

Most studies of oncogenic events have concentrated on

mechanisms of increased cell growth and proliferation.

However, decreased cell death would also result in cell

expansion (Korsmeyer,1993; Wyllie, 1992). Tumor growth can

either be the result of high cell proliferation or

continuous proliferation with the same doubling time and/or

decreased cell death. Thus, in addition to genes whose

products promote cell proliferation, genes which exert their

effects through either repression of apoptosis or failure to

induce correctly programed cell death may also be involved

in tumor growth (Owens and Cohen,1992).

Most of the currently used cytotoxic anticancer

therapies have been shown to induce apoptosis in susceptible

cells (Hickman,1992). The observation that diverse anti-

neoplastic agents with different cellular targets can induce

cell death or apoptosis in a similar fashion (DNA

fragmentation, chromatin condensation) indicates that

cytotoxicity is determinated by the ability of cells to

engage in a common apoptotic program. That program is

controlled by intrinsic factors as shown by the studies of

Fesus et al.,1991 and Bursch et al.,1990. Studies of the

signals which might engage in apoptosis and the genes

modulating it would help to identify new targets for anti-

cancer treatments.

Oncogenes Controlling Apoptosis

Sensitivity of cancer cells to apoptosis-inducing

agents produces an important effect on the outcome to

therapy. Several regulatory genes have been implicated in

the control of apoptosis (Green et al.,1994). Among these

are protooncogenes controlling cell proliferation and signal

transduction such as c-myc, ras, and c-abl, and tumor

suppressor genes such as Rb and p53. When these genes become

deregulated, increased cell survival results and tumors

become resistant to chemotherapy (Owens and Cohen,1992;Lotem

and Sachs,1993; Sugimoto et al.,1992; Fernandez et al.,1994,


Expression of c-myc is of particular interest, since it

seems to determine either continuous proliferation or

apoptosis, depending on the availability of critical growth

factors. In fibroblasts, serum deprivation induces c-myc

downregulation, and the cells remain highly viable and in a

growth-blocked state. However, serum deprivation in cells

which constitutively express c-myc induces apoptosis and no

growth arrest (Bissonnette et al.,1992). On the other hand,


some cellular oncogenes appear to always induce apoptosis

when they are activated. The p53 gene works as a tumor

suppresser by transactivating the WAFl/Cipl gene (El-Deiry

et al.,1993). The protein product of this gene inhibits G1

cyclin-dependent kinases thereby halting cell cycle

progression. In conditions in which DNA repair is necessary,

the cell is blocked in G1 allowing time for repair. If

irreparable DNA results, the cell becomes committed to

apoptosis. Mutant p53 allows escape from this surveillance

mechanism (Ryan et al.,1994). Still other genes appear to

rescue cells from apoptosis. Among these is bcl-2, described

in detail later. Activation of bcl-2 is not unique as an

anti-apoptotic mechanism, because high expression of mutated

genes such as ras or exposure to growth factors induce the

same anti-apoptotic effects as Bcl-2 (Williams et al.,1990;

Fernandez et al.,1994).

Bcl-2 Oncoprotein

A link between the proto-oncogene bcl-2 and the

control of apoptosis has been demonstrated. This gene was

isolated from the breakpoint of the translocation between

chromosome 14 and 18 and is present in a high percentage of

B-cell follicular lymphomas. The translocation puts the

bcl-2 gene under control of the immunoglobulin (Ig)-heavy

chain locus resulting in its aberrant expression. The

breakpoint occurs outside the translated sequences leaving

intact the coding sequences. Two out of three exons code for

the translated region. Bcl-2 mRNA produces two proteins by

RNA splicing (Tsujimoto and Croce,1986; Cleary, et al.,1986;

Bakhshi et al.,1987), a small beta-protein and a larger 26-

kd-alpha protein. The proteins similar in amino terminal

regions, but the alpha-protein contains additional

hydrophobic amino acids in its carboxy terminal end which

allows anchorage of the protein in cell membranes (Tanaka et

al.,1993). The Bcl-2 alpha-protein has been implicated in

the prevention apoptosis (Korsmeyer,1992; Hockenbery et

al.,1990; Tanaka et al.,1993).

Bcl-2 expression has been extensively studied during B

and T lymphocyte maturation. Pre-B cells and mature B cells

at the germinal center are selected during clonal expansion

and show the highest expression, while differentiated cells

downregulate Bcl-2 (McCarthy et al.,1992). T cells also show

differential Bcl-2 expression during thymus development(Veis

et al., 1993). That Bcl-2 contributes to extended survival

of lymphocytes is supported by experiments using bcl-2

transgenic mice. These animals demonstrate high numbers of

viable B cells expressing IgD/IgM on their cell surface in

areas of the lymph node characterized by apoptosis. They

also display an increased population of cortical T cells

abnormally resistant to glucocortocoids and/or gamma-

radiation (McDonnell et al.,1989). These results indicate

that Bcl-2 can prolong the survival of lymphocytes by

preventing apoptosis. Bcl-2 has also been implicated in

oncogenesis by cooperating with oncogenes such as c-myc.

This has been demonstrated in T-lymphoid cell lines

transfected with bcl-2 and c-myc. In mice injected with

cells carrying both genes, accelerated tumorgenicity was

observed. In contrast, mice injected with cells carrying

bcl-2 only do not develop tumors (Reed et al.,1990). Bcl-2

can also contribute to oncogenesis without inducing cell

proliferation. This is supported by experiments in which

bcl-2 is transferred into malignant cells that are growth

factor dependent. These transfected cells show increased

survival without cell proliferation when growth factor is

withdrawn (Nunez et al.,1990).

Antioxidant Function

One of the most notable features of Bcl-2 activity has

been its ability to block gamma-radiation induced cell

death. Ionizing radiation induces hydroxyl radicals (OH.)

from H20 by radiolytic attack (Strasser et al.,1991). The

intracellular sites where these free radicals are generated

include the mitochondria, endoplasmic reticulum, and nuclear

membrane. Bcl-2 has been localized to these sites (Lam et

al,1994; Lithgow et al.,1994;Hockenbery et al.,1990;Akao et

al.,1994). Although it is not known how Bcl-2 works as an

anti-oxidant, it can protect cells against H202 or t-butyl

hydroperoxide in a dose dependent manner. Bcl-2 also

interferes with superoxide inducers such as menadione or

other quinones without preventing free radical production by

these compounds (Hockenbery et al.,1993,Zhong et al.,1993).

In addition, Bcl-2 completely prevents lipid peroxidation of

membranes (Kane et al,1993; Hockenbery et al,1993).

Consistent with these findings is the observation that over

expression of glutathione peroxidase (GSHPx), a known

inhibitor of lipid peroxidation, can partially block

apoptosis (Hockenbery et al., 1993). Another line of

supporting evidence that Bcl-2 is an anti-oxidant comes from

the phenotype of Bcl-2 knock-out mice. These mice display

fulminate lymphocyte apoptosis, polycystic kidney disease

and hypopigmented hair (Veis et al.,1993; Nakayama et

al.,1993). In summary, Bcl-2 seems to block apoptosis by an

unknown anti-oxidant mechanism. This mechanism does not

involve prevention of free radical production, but may

involve the prevention of lipid peroxidation (fig 1) after

hydroxyl radicals have been generated (Bornkamamm and

Richter, 1995).

Protein Dimerarization

In the absence of a clear biochemical function,

researchers began to look for proteins that can interact

with Bcl-2. These findings revealed a multicomponent protein

complex to which several proteins with Bcl-2 homology

belong. The first of these protein described is the 21-Kd

02-: H202






+ OH + OH. OH .


+ OH

+ Fe+++

B cl-2

Lipid Peroxidation
DNA Breaks

Figure 1. The anti-oxidant mechanism of Bcl-2. H202 is
produced by the dismutation of 02. by superoxide dismutase
(SOD). Excess of 02. and H202 will result in the oxidation
of transitional metals and the production of hydroxyl
radicals. In this scheme Bcl-2 prevents peroxidation of
membrane lipids, after OH. radicals are generated from the
metal driven Haber-Weiss and Fenton reactions.

-- H20


- i


partner Bax. This protein has homology with Bcl-2 within two

highly conserved regions, called the Bcl-2 homology 1 and 2

(BH1, BH2) domains (Yin et al.,1993). Bax and Bcl-2 have

been found to form heterodimers by BH1 and BH2 domains, and

to form homodimers between their own molecules (Oltvai et

al.,1993). This activity seems to be dependent on the levels

of both proteins. This finding supports a model in which

Bcl-2 must bind Bax to exert its activity, and suggests that

either Bax/Bcl-2 heterodimer or Bax/Bax homodimer may prove

to be the active pair regulating cell death. Experiments

using double transfection of these genes have shown that the

level of expression of these proteins determines cell

survival. Thus, an excess of Bcl-2 would favor survival;

while an excess of Bax would favor cell death (fig 2). Using

this same approach, other members of the Bcl-2 related

proteins have also been described. One of particular

interest is Bcl-X, a gene initially discovered in chickens.

By screening human libraries, two types of Bcl-X cDNAs were

found. One, a short mRNA (bcl-Xs) where areas of Bcl-2

homology are deleted as well as a long mRNA (bcl-X1) (Boise

et al.1993). Overexpression of Bcl-Xl in an IL-3-dependent

poptosis urviva

Figure 2. Dimerization of Bcl-2 and Bax. After receiving an
apoptotic signal, cells can survive if Bcl-2 is highly
expressed, but will die if Bax predominates.

cell line prevented apoptosis, whereas the short mRNA (Bcl-

Xs) accelerates apoptosis (Boise et al., 1993; Datta et

al.,1995). High expression of Bcl-Xs can override Bcl-2

activity and induce accelerated apoptosis in some cell

systems (Nunez et al.,1994).

Other genes can also indirectly regulate Bcl-2

activity. Recently, P53 has been shown to regulate Bcl-2 by

inducing the bax promoter after an apoptotic stimulus has

been triggered. In this way, P53 can regulate Bcl-2 by

promoting Bax. If an excess of Bax is produced, Bcl-2

activity is abrogated (Miyashita and Reed,1995). This

finding further supports the notion that Bcl-2 activity is

dependent on the level of gene expression.

Resistance to Cancer Treatment

A variety of experimental observations has pointed to

Bcl-2 as a critical regulator of the cell death process

(Hockenbery 1994). Overexpression of this protein can render

cells relatively resistant to a wide spectrum of stimuli.

This broad range of stimuli, with their numerous biochemical

mechanisms, suggests that Bcl-2 functions at a distal point

in the pathway leading to cell death. Thus, despite the

various upstream targets, these stimuli eventually utilize a

common mechanism to kill the cell. This could explain how

Bcl-2 can protect tumor cells from different anti-neoplastic

agents (Fisher et al.,1993 Korsmeyer,1995).

Using gene transfer methods to overexpress bcl-2 in B

cell leukemias that express low Bcl-2 levels, as well as

anti-sense approaches to decrease Bcl-2 levels in t(14;18)-

containing B lymphoma cell lines, it has been demonstrated

that the level of Bcl-2 protein correlates with the relative

resistance or sensitivity to a wide spectrum of

chemotherapeutic drugs and gamma-radiation (Reed et

al.,1990; Miyashita and Reed,1993; Reed,1995). These

observations indicate that prevention of apoptosis by high

Bcl-2 expression may render tumor cells resistant to

chemotherapy similar to what has been shown for B cell

tumors (Nunez et al.,1989). Although Bcl-2 expression has

been extensively studied in both normal and malignant

lymphocytes, little is known about its expression and

function in other non-lymphoid hematological tumors. Myeloid

leukemias represent 80% of all adult leukemias, and these

tumors develop chemoresistance in a high percentage of

patients. Studies to determine how Bcl-2 functions in

myeloid tumors would help to elucidate possible therapies to

treat these tumors.

This dissertation was undertaken to understand not only

the normal expression of Bcl-2 in non-lymphoid hematological

tumors, but to also determine its role in myeloid leukemias

and its clinical implications. In order to accomplish these

main goals, this research dissertation has focused on the

following specific aims: (1) To analyze the effect of

cytosine arabinoside (Ara-C) on the growth of hematological

tumors expressing different levels of Bcl-2 oncoprotein. (2)

To determine the effects of Ara-C on myeloid tumors

transfected with retroviral vectors carrying murine bcl-2 or

vector alone and (3) To determine the function of Bcl-2 in

myeloid tumors treated with bleomycin, a free radical




Molecular Biology Techniques

Protein Extraction and Western Blots

Post-nuclear fractions were prepared by resuspending

107 cells in 1 ml of RIPA buffer containing 150 mM NaC1, 1%

NP-40, 0.1% SDS, 50 mM Tris-HCL (pH 8.0), 0.001% pharaphe-

nylmethylsulphonylfluoride (PMSF) and 10M Leupeptin (Sigma,

St. Louis, MO). Equal amounts of protein, as determined by

protein assay (Bio-Rad, Melville NY), were size separated on

a 12% denaturing polyacrylamide gel. Two gels were run, one

stained with Coomassie blue and used as a control, and the

other was transferred to a nitrocellulose membrane

(Amersham, Arlington, IL) and blocked with 5% non-fat dry

milk in 20 mM tris buffer(TBS) pH 7.5, containing 0.5%

Tween 20. The membrane was incubated overnight at 4 oC with

a 1:200 dilution of mouse anti-human Bcl-2 antibody(Dako,

Carpenteria, CA), in TBS or 5 pg/ml of hamster anti-murine

Bcl-2 IgG (Pharmingen, San Diego CA). After washing 4 times

with TBS, the membranes were incubated for 1 hour at room

temperature with 1:2000 dilution of goat anti-mouse (IgG)

(Amersham, Arlington Heights IL) or goat-anti-hamster

antibody (Cappel, Durham NC) conjugated with horseradish-

peroxidase (HRP). After washing with TBS, the blots were

developed by using an enhanced chemiluminescence detection

system for HRP-labeled second antibody (Amersham Arlington

Heights, IL) and exposed to Hyperfilm for 5 minutes

(Amersham, IL). Films were analyzed in a DTS densitometer

(Pharmacia, Piscataway, NJ).

RNA Extraction and Northern Blots

Total RNA was extracted by the guanidine thiocyanate

(GTC) phenol-chloroform method (Chirwing et al.,1976).

Briefly, cells were rinsed in cold PBS and homogenized in

GTC (4M guanidinium thiocyanate, 0.5% sarkosyl, 25 mM sodium

citrate pH 7.0 and 0.1 M 2-Mercapto-ethanol) and mixed with

2 M sodium acetate (pH 4.0), water saturated phenol and

chloroform-isoamyl alcohol. An equal volume of cold

isopropanol was added and the mixture was centrifuged at

10,000 X g for 20 minutes at 4 C. RNA pellets were

resuspended in water and stored at -70 oC. RNA (10 ig/lane

as determined by absorbance reading) was separated on an

1.2% formaldehyde/MOPS (morpholino-propane-sulfonic acid)

gel and transferred to a nylon membrane using a Turboblot

System (Schleicher & Schuell, Keene NH). After cross-linking

with UV light, the membrane was hybridized at 42 OC with a 32

P-murine bcl-2 probe labeled by random-primer reaction. The

final wash was in 2X SSPE/0.1% SDS for 20 minutes at 42 oC.

Autoradiography was performed at -70 oC with Amersham-


Retroviral Constructs. Electroporation. G418 Selection

A 0.7 kb murine bcl-2 cDNA from the pBMGNEO vector

(Deng and Podack,1993) was subcloned into the Xho I site of

the moloney retroviral vector carrying the neomycin

resistance gene and cytomegalovirus promoter (MNC) (Suresch

et al.,1994) by using standard DNA recombinant methods

(Sambrook et al.,1989). U-937 cells were transfected by

electroporation. First, cells were resuspended in cold RPMI

pH 7.0 without serum and mixed with 10 9g/ml of MNC vector

or MNC-bcl-2 cDNA in the sense orientation (fig 3). Cells

were pulsed for 3 seconds with the electroporator (Bio-Rad)

set at 9.60 mFD of capacitance and 250 mV of current. After

48 hours in culture, stable transfected cells were selected

in 1 mg/ml G418 (Gibco,MD) for 2-3 weeks.

Plasmids and Probes

A murine-specific bcl-2 probe was obtained by isolating

a 200-bp Sma I-Xho I cDNA fragment from bcl-2-pBMGNEO

plasmid. A 1.4 kb human GAPDH (glyceraldehyde-phosphate-

dehydrogenase) specific probe (kindly given by Dr. B. Allen,

University of Florida) was used as a loading control for


Low Molecular DNA Extraction

Cells (5 X 106) were lysed in 5 mM Tris-HCL, pH 7.5,

0.5% TritonX-100, and 20 mM EDTA on ice for 20 min. After

centrifuging at 27,000 X g for 10 min, supernatants were

4.4 Kb



BCL- 2

1.7 Kb

Figure 3. Construct of murine bcl-2 retroviral vector. A
murine bcl-2 retroviral vector was obtained by subcloning a
0.7 kb cDNA of murine bcl-2 into Xho-I site of the
retroviral vector MNC. This places murine bcl-2 under the
control of cytomegalovirus (CMV) promoter. The vector also
expresses the genes for neomycin resistance. Two bcl-2
transcripts are expressed, one full length driven by LTR
(4.4 kb), the other driven by the internal promoter CMV of
1.7 kb.

collected and extracted with phenol-chloroform, followed by

precipitation in cold ethanol. After resuspending in TE-EDTA

(TE), DNA was incubated with 10ig/ml RNAase (Sigma) at 37 C

for 1 hour, extracted with phenol-chloroform and

reprecipated in cold ethanol. DNA (20 l) was loaded in a

1.2% agarose gel containing ethidium bromide (0.5 Jig/ml) in

TAE, and run at 60 V for 2 hours. DNA fragmentation was

recorded under UV light.

Tissue Culture Techniques

Cell Lines and Growth Media

The follicular lymphoma RL-7 (kindly donated by Dr.

Massiah, University of Florida)) and the monocytic leukemic

cell line U-937 (ATCC Rockville, MD) were maintained in RPMI

1640 supplemented with 10% FBS. The early myeloid cell line

KG-1 (ATCC, Rockville, MD) was cultured in IMDM supplemented

with 20% FBS. All cell lines were cultured at 37 oC, in an

atmosphere of 5% and 95% humidity and used in all

experiments when in log phase of growth.

1-P-d-arabinofuranosylcytosine (Ara-C) (Sigma) was

freshly prepared and diluted in complete culture media to

0.001-100 JiM before use. Bleomycin sulfate (Bristol Myer,

NJ) was kindly donated by Dr. Steve Smith (University of

Florida). It was diluted in saline and a stock preparation

(1 mg/ml) was frozen at -20 oC, and used at 0.5-10 pg/ml.

Clonogenic Assay

Tumor cells were incubated in the presence or absence

of bleomycin for 6 hours or with Ara-C for 24 hours,

washed twice with media and plated in 0.35% agar containing

RPMI media or in 0.8% methyl-cellulose (Stem Cell

Technologies, Vancouver, BC) at 1 ml/plate. Various tumor

cell numbers were plated in semi-solid medium with 500 cells

for controls and 500-5000 cells for drug-treated cells.

After 7-10 days of culture, relative colony efficiencies

were recorded by dividing the number of colonies formed from

drug-treated cells by number of colonies formed from

untreated cells. LD50 values were calculated.

Viability Assay

Tumor cells were incubated in the presence or absence

of Ara-C and bleomycin for 24 hours or for 1-10 days

respectively. Absolute number of viable cells were

determined by multiplying the number of cells excluding

trypan blue (0.4% in 0.9% saline) by the volume of culture

media. Percentage of viability was obtained by dividing

number of cells excluding trypan blue by the total number of

cells counted.

Cell Cloning

Stable transfected bcl-2-U-937 cells were cloned by

limiting dilution. Serial dilutions (1/10) of the bulk cell

line were prepared. A starting cell concentration of 10,000

cells/ml was diluted up to 1000 fold and two hundred

microliters of each cell dilution were placed in every well

of a 96-well plate. After one week, plates were checked for

the presence of single colonies. Separate colonies were

expanded and assayed for murine Bcl-2 protein by Western


Flow Cytometric Analysis

Staining of Intracellular Antigen

KG-1 cells were washed in cold PBS pH 7.4. The cells

were fixed in 0.25% paraformaldehyde (PFA) in PBS for 15 min

at room temperature and rinsed with PBS. The cells were then

permeabilized with 500 pl of cold 70% ethanol for 15 minutes

at 4 C. The cells were washed twice in PBS containing 2%

human albumin for blocking and divided equally into two

tubes. Cells were incubated in the dark for 30 min at 4 C

with either 10 [l FITC-conjugated monoclonal mouse anti-

human Bcl-2 antibody (Dako, Carpenteria, CA). FITC-

conjugated mouse IgG was used as a negative isotype control.

Cells were washed twice with cold PBS and analyzed by flow

cytometry using a FacScan (Becton Dickinson, San Jose, CA)

with Lysys software. Fluorescence intensity, measured as

mean channel number, was standardized by using calibrated

fluorescent beads in each experiment (Flow Cytometry

Standards, San Juan, PR) and expressed as Molecules of

Equivalent Soluble Fluorochromes (MESF).

Staining of Cell Surface Antigen

U-937 cells were washed twice with cold PBS and pre-

incubated with 1% human albumin to prevent binding of the

antibodies to Fc receptors. The cells were incubated with 10

pl of FITC-conjugated monoclonal mouse anti-human CD11b

antibody (Biosources, Camarillo CA) or FITC-mouse IgG

(Pharmingen) as isotype control for 30 min at 4 OC in the

dark. Propidium iodide (0.5 pg/ml) was used to exclude dead

cells during flow cytometry analysis performed with a

FacScan (Becton Dickinson) containing Lysis software.

DNA Fragmentation

DNA fragmentation was quantitated by using dUTP

incorporation as previously described by Darzynkiewcz, 1993.

Briefly, 106 RL-7 or wild type and bcl-2 transfected U-937

cells were fixed in 1% paraformaldehyde on ice for 15 min

and rinsed with cold PBS. The cells were permeabilized with

cold 70% ethanol for 15 min. After rinsing twice in cold

PBS, cells were resuspended in 50 il cacodylate buffer

containing 2.5 mM cobalt chloride (Boehringer Mannheim),

0.25 mg/ml BSA, 100 U/ml of terminal deoxynucleotidyl

transferase (TdT) and 0.5 nM biotin-16-dUTP (Boeringer

Mannheim)and incubated for 30 min at 37 OC. After rinsing in

PBS, the cells were incubated at room temperature, in the

dark for 30 min with 100 Li 4X SSC buffer containing 2.5

gg/ml FITC-avidin (Sigma, St.Louis,MO), 5% non-fat dry milk,

and 0.1% Triton X-100. The cells were then resuspended in 1

ml PBS containing 5 [g/ml propidium iodide (PI) and 1%

RNAase A (Sigma) and analyzed in a FacScan (Becton

Dickinson) using bivariate analysis to obtain the level of

DNA fragmentation as dUTP incorporation into 3'-OH DNA

strand-breaks and cell cycle position measured by PI


Cell Cycle Analysis

Nuclei were isolated from U-937 cells as previously

reported by Vindelov et al.,1988. Briefly, 500,000 cells

were rinsed with cold PBS pH 7.4, and resuspended in 50%

cold ethanol for 1 hour at 4 OC. After rinsing, the cells

were resuspended in citrate (1 mg/ml) buffer containing

trisodium citrate, NP-40 (0.01%), spermine and 20 mM tris

(pH 7.6), and incubated with 1% trypsin at room temperature

for 10 min. Soy bean trypsin inhibitor (0.5 mg/ml) and 1%

RNAase (Sigma) were added for an additional 10 min. Nuclei

were centrifuged at 2,000 rpm and resuspended in citrate

buffer (pH 7.6) and propidium iodide (50 jg/ml). Nuclei were

analyzed by flow cytometry using a FacScan equipped with

CellFit software.




Bcl-2 has been extensively studied in B and T

lymphocytes. In contrast, Bcl-2 expression and function in

non-lymphoid hematological tumors is poorly understood. The

objective of this aim was to analyze bcl-2 expression in

myeloid tumors and to correlate Bcl-2 levels with Ara-C


Ara-C is an active antimetabolite commonly used for

both induction and maintenance of aggressive hematological

tumors (Chabner and Myers, 1993). This agent is incorporated

into DNA in a concentration dependent manner, bringing about

inhibition of DNA synthesis. Although the mechanism by which

Ara-C-DNA induces cell death remains largely unknown,

previous studies showed that the treatment of myelogenous

leukemia with Ara-C is associated with apoptosis (Gunji et


al.,1991). Like many other therapies, resistance to Ara-C

can develop (Bhalla et al, 1984; Gunji et al.,1991). In

acute myeloid leukemias(AML), high doses of Ara-C have been

successfully used to avoid resistance to conventional Ara-C

treatment (Heinemann and Jehm,1990; Herzig et al.,1983). To

extend these findings, the present investigation will

attempt to relate the intracellular levels of Bcl-2 with

clonogenic survival and prevention of apoptosis.

Bcl-2 Expression by Hematological Tumors

Bcl-2 protein was measured in three hematological tumor

cell lines, KG-1, U-937 and RL-7 by Western analysis (Figure

4). Scanning densitometry revealed low Bcl-2 levels in U-937

cells with a relative density of RD= 0.008, when compared to

the follicular lymphoma RL-7, which carries the t(14;18)

chromosomal translocation, and expresses high Bcl-2 levels,

RD=3.084. Intermediate Bcl-2 levels were obtained with the

pre-myeloid leukemic cell line KG-1 (RD=0.715).

1 2 3

26 Kd

Lane 1. KG1 Cells
Lane 2. U937 Cells
Lane 3. RL-7 Cells

Figure 4. Western analysis of human Bcl-2 in three
hematological tumors. Fifty micrograms of post-nuclear
protein fraction was analyzed by Western analysis to
determine the level of human 26-Kd-Bcl-2 protein.
Densitometry measurements reveal that the follicular
lymphoma RL-7 cells display an RD=3.084, while monocytic
leukemia U-937 cells express low Bcl-2 levels with and an
RD=0.008. Intermediate Bcl-2 expression was obtained with
the pre-myeloid leukemia KG-1 with a densitometry
measurement of RD=0.72.

Ara-C Dose-Response

These three hematological tumor cell lines were assayed for

their sensitivity to Ara-C by incubating the cells with

different Ara-C concentrations for 24 hours prior to plating

in agar or methylcellulose culture. Figure 5 shows a dose-

response for Ara-C on the clonogenic survival of each of the

cell lines. U-937 cells demonstrate extreme sensitivity to

Ara-C with an LD50 of about 0.005 pM Ara-C and no colony

formation was seen when these cells were treated with J1M

Ara-C. In contrast, the LD5o for RL-7 cells was shifted about

2 logs to the right and measured about 0.5 iM Ara-C. These

cells retain about 20% colony efficiency when treated with

100 (M which was the highest concentration tested. In

contrast, KG-1 cells demonstrate an LDs5 of about 0.1 pM

Ara-C and significant colony formation can still be seen at

10 pM Ara-C. These data suggest a possible relationship

between the level of Bcl-2 expression and the sensitivity of

these hematological tumors to Ara-C.

---- U937

-4-- RL-7

-*- KG-1

0 0.001 0.01 0.1

1 10 100

Figure 5. Ara-C dose response curves for three hematological
tumors. Each cell line was treated with increasing
concentration of Ara-C for 24 hours. After 7-10 days of
culture, relative colony efficiency was obtained. Values are
expressed as mean of triplicates of at least 3 experiments
+/- 1 SD.


RL-7 ARA-C U-937 XAR-C

Figure 6. Flow cytometry analysis of Ara-C induced
apoptosis. Cells were treated with 100 pM Ara-C for one
hour. DNA strand breaks were labeled with biotin-dUTP and
FITC-avidin, while the cell cycle positions were determined
by propidium iodide staining. Untreated cells or Ara-C
treated RL-7 cells do not show apoptosis, while U-937 cells
treated with Ara-C display maximum dUTP incorporation in S
phase and less in GO/G1 and G2/M.

These same cell lines were also examined for apoptosis

using both flow cytometry and DNA ladder formation. In

figure 6, RL-7 and U-937 cells were treated with 100 pM Ara-

C for one hour and DNA fragmentation analyzed by flow

cytometry using dUTP and PI as described in the Methods

Section. Ara-C treated or untreated RL-7 cells do not

display dUTP incorporation, while Ara-C-treated U-937 cells

display high dUTP incorporation in S phase and less dUTP

incorporation in Go/G1 and G2/M phases of the cell cycle.

The data shows maximum Ara-C effect on U-937 cells during

DNA replication since all cells in S phase are in apoptosis.

Apoptosis was also checked by DNA ladder formation in

agarose gels. Each of the cell lines was treated with 100 pM

Ara-C for 24 hours. Figure 7 demonstrates a DNA ladder

typical of cells in apoptosis for U-937 and KG-l cells,

while no DNA fragmentation was observed with RL-7 cells.

1 2 3


Lanes 1,4 KG-1
Lanes 2,5 RL-7
Lanes 3,6 U937

Figure 7. Effect of Ara-C on the DNA fragmentation of
hematological tumors. RL-7, U-937 and KG-1 cells were
treated with 100 pM Ara-C for 24 hours. Low molecular DNA
was extracted from 5 X 106 cells and electrophoresed in
agarose gels. Untreated controls (lanes 1-3) or Ara-C
treated RL-7 cells (lane 5) do not show DNA ladder formation
typical of cells in apoptosis, while Ara-C treated KG-I
cells (lane 4) and U-937 cells (lane 6) demonstrate DNA

am 44 0

H 0 0


n "e
B 4 0I

S01 0

-,IQ 4) (
(d 4 ) 4J

4n 0 >

tw. Om

0 sl
1 n

m d
1 (U O0

- o 4)

0 : 0

H 0 N 4

) ) 4) id
t 9i 4 6
a, a, a, (d (
, $4 4.) 4
P4 4) M M) ))






I- 0





0 0









3 4

-W ,-m



Figure 9. Western analysis of Bcl-2 during Ara-C action. KG-
1 cells were treated as described in figure 8 and 100 pg of
post-nuclear protein fraction analyzed in each lane by
Western blotting. Densitometric data shows a RD=3.0 for
untreated cells, and RD=1.19 for 1 hr, RD=1.21 for 24 hrs
and RD=1.9 for 48 hr of treatment with 100 jiM Ara-C.



Bcl-2 Expression during Ara-C Action

Because KG-1 cells show DNA fragmentation when treated

with 100 p1M Ara-C for 24 hours (see figure 7), these cells

were screened for Bcl-2 expression during Ara-C action to

determine if Bcl-2 was downregulated. Bcl-2 protein was

detected in KG-1 cells by flow cytometry and Western

analysis. Figure 8 depicts changes in the Bcl-2

concentration measured by flow cytometry as MESF during Ara-

C treatment. A 35% decrease of Bcl-2 protein was observed at

1 hour post-treatment. However, no further decrease of Bcl-2

was observed, even after 24 hours when apoptosis had already

occurred. This result was further confirmed by Western blot

analysis. In figure 9, densitometry revealed a relative Bcl-

2 density for control untreated KG-1 cells of RD= 3.0, and

relative densities for Ara-C treated cells of 1.19, 1.21 and

2.00 for 1, 24 and 48 hours postreatment respectively,

indicating that Bcl-2 is not completely downregulated when

cells go into apoptosis.


In the present section, the role of the Bcl-2

oncoprotein on the resistance of non-lymphoid hematological

tumors was investigated. Ara-C was selected because of its

frequent use in the treatment of aggressive hematological


Hematological malignancies expressing high levels of

Bcl-2 such as the follicular lymphoma RL-7 and the pre-

myeloid leukemia KG-1 were found to be markedly resistant

to Ara-C concentrations < 100 and 10 JLM respectively. In

contrast, the low Bcl-2 expressing, myeloid leukemia U-937

cells were sensitive to Ara-C concentrations below 1.0 pM.

These results demonstrate a between the level of Bcl-2 being

expressed and the sensitivity to Ara-C of these

hematological tumors. In addition to colony efficiency, Bcl-

2 levels appear to correlate with the extent of DNA

fragmentation induced by Ara-C. Treatment with the highest

Ara-C concentration (100lM) induced DNA fragmentation in KG-

1 at 24 hrs and in U-937 cells after one hour; whereas RL-7

cells showed no DNA fragmentation at 100 pM. The Bcl-2


oncoprotein was also analyzed during the action of Ara-C in

KG-1 cells. The pre-myeloid cell line KG-1 demonstrated high

sensitivity to 100 pM Ara-C and this increased sensitivity

was not accompanied by complete Bcl-2 down-regulation since

only a 35-40% decrease in the protein level was observed.

Therefore, the sensitivity of these hematological

tumors to Ara-C treatment appears to depend on the levels of

Bcl-2 being expressed. Cells carrying the t(14;18)

translocation express high levels of Bcl-2 and are

resistant to very high concentrations of Ara-C. In cells

expressing very low levels of Bcl-2, such as in the U-937

myeloid tumor, this low Bcl-2 level also correlates with

lack of resistance to Ara-C.




In order to determine whether Bcl-2 has a direct role

in the protection of human myeloid cells from Ara-C, U-937

cells, which express low levels of human Bcl-2, were

transfected with a retroviral vector expressing murine bcl-

2. The use of murine Bcl-2 had the additional advantage of

being able to differentially detect both endogenous human

Bcl-2 and transfected murine proteins with specific probes

and antibodies to the human or murine Bcl-2.

A retroviral vector was constructed by subcloning

murine bcl-2 cDNA into the MNC vector (figure 3). U-937

cells were transfected by electroporation. Cells were

checked for murine Bcl-2 expression, and stable transfected

cells were cloned as explained in Materials and Methods. U-


937 cells were also transfected with the vector MNC alone

and used as a control. Sensitivity of MNC and bcl-2

transfected U-937 cells to Ara-C was tested by using low as

well as high concentrations of Ara-C (Bhalla,et al.,1992),

to determine whether deregulated Bcl-2 can render non-

lymphoid cells resistant to chemotherapy.

Murine Bcl-2 Expression

Transfected U-937 cells were checked for the expression

of murine Bcl-2 by Northern and Western analysis. Figure 10B

demonstrates murine bcl-2 RNA expression only in bcl-2

transfected U-937 cells (lane 3). The molecular size of the

CMV promoter driven transcript is 1.7 kb while the size of

the transcript driven by the LTR is 4.4 kb. Murine Bcl-2 was

also detected by Western analysis. Figure 10A shows the

presence of the 26-Kd band which represents murine Bcl-2

only in the bcl-2 transfected U-937 cells (lane 3), while

MNC and wild type U-937 cells were negative (lanes 1,2).

Stable transfected bcl-2 U-937 cells were also subcloned by

Figure 10. Western and Northern analysis of murine Bcl-2 in
U-937 cells. Cells were transfected with vector MNC alone or
vector carrying the murine bcl-2 gene as described in
Materials and Methods. A) Equal amounts of proteins (50 pg)
were analyzed by Western blots using a murine specific Bcl-2
antibody, control of protein loading was determined as
explained in Methods. Lanes 1,2 are negative controls, while
bcl-2 U-937 cells express the 26-kd oncoprotein as seen in
lane 3. B) The cells were also checked for murine bcl-2 mRNA.
The bcl-2 U-937 cells show the LTR and CMV promoter driven
transcripts of 4.4 and 1.7 kb respectively (lane 3), whereas
wild type and MNC U-937 cells are negative. GAPDH was used as
an RNA loading control as seen in the bottom of the figure.

A Western Blot

-, 26Kd

B Northern Blot

m 0 4

- 4.4Kb



U937 Wild Type
U937 Vector Only
U937 Murine Bcl-2 sense construct

Lane 1.
Lane 2.
Lane 3.


Figure 11. Western analysis of murine bcl-2 in subcloned U-
937 cells. Stable transfected U-937 cells were cloned by
limiting dilution as explained in Materials and Methods.
Equal amounts of proteins (25 Ag) were analyzed by western
blot using an antibody specific for murine Bcl-2. The
relative concentration of murine Bcl-2 was calculated by
densitometry. Murine thymus (lane 1,RD=0.56), MNC U-937
cells (lane 2, RD= 0.002), Clone 8 (lane 3, RD=0.17); Clone
15 (lane 4 RD=1.04); Clone 1 (lane 5, RD=0.42), and uncloned
U-937 cells (lane 6, RD=0.31)


limiting dilution and cell clones checked for murine Bcl-2

expression. Figure 11 shows a representative Western blot of

three U-937 clones (8,15, and 1) and uncloned cells.

Densitometry analysis reveals variations for the three bcl-2

U-937 clones. Clone 15 (lane 4) displays the greatest

amount of Bcl-2 with an RD= 1.04. This is followed by clone

1 with an RD= 0.42 and clone 8 with an RD= 0.17. The

uncloned bcl-2 U-937 cells give an RD= 0.31 which is

significantly greater than the RD= 0.002 value for MNC-U-937


Bcl-2 Induced Chemoresistance

Bcl-2-U-937 cell clones as well as wild type and MNC U-

937 controls were treated for 24 hours with increasing

concentrations of Ara-C and examined for viability using

trypan blue dye exclusion and cloning efficiency. Figure 12

shows that the cell viability of U-937 cells transfected

with vector alone decreased rapidly as the dose of Ara-C

increased (closed triangles). In contrast, the cell

viability of all three of the bcl-2-U-937 clones changed


very little with increasing Ara-C concentrations (open

markers). Figure 13 shows the cloning efficiency of

transfected U-937 cells. MNC U-937 cells (open circles) are

very sensitive to Ara-C displaying an LDs0 of about 0.005 fiM

Ara-C which is similar to that seen for wild type U-937

cells. In contrast, bcl-2 U-937 cell clones 1, 15, and 8

(closed markers) show increased resistance to Ara-C with an

LD50 of 5.0 JiM Ara-C. All three bcl-2-U-937 clones are about

3 log more resistant to Ara-C when compared with wild type

or MNC- U-937 controls. In addition, all three bcl-2 U-937

clones show about 20%-30% colony efficiency when exposed to

100 LM Ara-C, a concentration completely wipes out colony

formation by wild type or MNC U-937 cells.

Decreased Apoptosis by Up-regulated Bcl-2

Apoptosis was measured by flow cytometry using dUTP

incorporation and PI staining for DNA content. Figure 14

depicts the flow cytometric analysis of MNC U-937 cells that

treated for 24 hours with increasing Ara-C concentrations

(0.001-1.0 pM). In panels C, only single cells were gated


-o- CL 1

-V- CL 8

-0- CL 15

0 0.001 0.01 0.1 1 10 100
Ara-C (uM)

Figure 12. Viability of Ara-C treated U-937 cells. Control
U-937 cells and bcl-2 transfected cell clones were treated
with increasing concentrations of Ara-C for 24 hours and
cell survival assessed by their ability to exclude trypan
blue. Results are from a typical experiment that was
repeated three times.



I 40

- 20

100 -- ---

-&-- U987
I U937
\- U937

\CI 15
S \ U937

.2 --- U 897
SCt 8

10 1
0 0.001 0.01 0.1 1.0 10 100

Figure 13. Ara-C dose-response of bcl-2-U-937 cell clones.
Clones 1, 15, and 8 (closed markers) as well as controls
(open markers) were treated with increasing concentrations
of Ara-C for 24 hours. After 7-10 days in culture, relative
colony efficiency was calculated as explained in Materials
and Methods. Data shows that bcl-2 transfected cells
demonstrate higher Ara-C resistance (about 3 logs) as
compared with U-937 controls. Data represents mean +/- 1 SD
of quadruplicates of three separate experiments.


(Rl) according to their DNA. B Panels show almost no changes

in the forward and side scatters of Rl-gated cells as Ara-C

concentrations increase, indicating that these cells are

intact. In A panels, Rl-gated cells demonstrate an increase

in dUTP incorporation as Ara-C concentration was increased.

Maximum dUTP up-take of 58.7% was obtained at 1 |pM Ara-C.

Using the same flow cytometric analysis for all three bcl-2

cell clones reveals very low levels of dUTP incorporation

even at 10 and 100 PM of Ara-C (Table 1).


To further study the role that the Bcl-2 oncoprotein

has on resistance to Ara-C, murine bcl-2 was transfected

into the myeloid U-937 cell line. Bcl-2 U-937 cells

demonstrate increased resistance to Ara-C when analyzing

apoptosis by flow cytometry. In addition, bcl-2 U-937 cells

also demonstrate a 3 log shift in resistance to Ara-C when

comparing LD50 colony forming values with wild type and MNC

U-937 cells. Further analysis reveals that while up-

regulated Bcl-2 protects from programmed cell death with


little or no dUTP incorporation seen at 100 pM Ara-C (table

1) and no change in trypan blue dye exclusion (figure 12),

it cannot prevent a decrease in clonogenicity as seen at 100

pM Ara-C (figure 13), indicating that a cytostatic effect

may occur when employing high doses of Ara-C in cells

altered to express high levels of Bcl-2. Furthermore, bcl-2

U-937 cell clones expressing different levels of Bcl-2 do

not show correlation between the amount of Bcl-2 and the

resistance to Ara-C measured either as trypan blue exclusion

(fig 12), colony efficiency (fig 13) or dUTP uptake (table

1). Whether this lack of correlation using murine bcl-2

transfected cells relates to a threshold Bcl-2 level that is

important or whether murine Bcl-2 shows this phenomenon

cannot be explained.

In summary, up-regulated Bcl-2 expression can render

myeloid cells resistant to Ara-C, the most commonly used

antimetabolite in the treatment of myeloid tumors. The

present results also support the hypothesis that the level

of Bcl-2 expression is an important factor to determine when

analyzing which tumors might respond to chemotherapy.

) O
1 0



a I

rl m



i o


0 )

4 U




0 f

C '
0 d



4. m

* k


0) 4)
-ri a4
Nr 4.

l 0 14 -

*r H- 0 0
Ul JU 414 4
XV 0 (0
H aa

0u d

* rdom
4.) (D V 4 rfn

D0 -- O ,
SM 0 4 4

4.) 1, 4 0< u
St E-1

G 'U *ril 0
0 "
Sag) o

Mr-I 4 41:

I 0 U

A m m
, &

4 )0 0
Z 14
u o

r) r-1

o 0U
0 rl H mu

M 0 0 A
un u m u
0 u oy 0

04 r-l (d P
oU H 04 A
E: a

4) a 1 U

S 0 0
' gO *r -4

0 r-I 0r4 0
4J0H 0) (
o 40 ) a

U 4-1 U 4 M





0 0.78* 0.31 0.96 1.42

0.0001 2.58

0.1 27.93

1.0 58.71 0.26 0.94 1.33

10 0.33 0.61 1.32

100 3.61 3.68 8.82

* Ara-C pM, 24 hour treatment
0 U-937 cells transfected with vector alone
T Cloned BCL-2-U-937 cells
* Percent of dUTP positive cells




Bleomycin is a glycopeptide antibiotic used in the

treatment of cancer. Two main effects have been studied

(Chabner and Meyers,1993). One is the induction of DNA

strand breaks, and another is the destruction of the cell

membrane. Both effects have been shown to be mediated by

free radicals. Bleomycin binds ferrous ions (Fe" ) which can

be oxidized enzymatically by NAD(P)H or non-enzymatically by

H202, and thiols using either the reactions Haber-Weiss and

Fenton (Matkhanov, 1993). The liberated electron can be

accepted by 02 forming superoxides(02. ) or hydroxyl radicals

(OH-) .

Bcl-2 has been reported to function as an anti-oxidant

in lymphoid cells. The mechanism involves blocking the

peroxidation of lipids in the cell membrane by hydroxyl


radicals, without preventing the production of free radicals

(Korsmeyer,1995). The purpose of this investigation was to

determine the effect of up-regulated Bcl-2 in myeloid cells

treated with bleomycin, an anti-neoplastic agent known to

induce free radicals.

Bleomycin Dose-Response

In order to test the effect of up-regulated Bcl-2 in

myeloid cells, sublethal concentrations of bleomycin were

selected to examine the long term effect of the drug on the

cell growth and morphology of myeloid tumor cells.

Sensitivity to bleomycin by wild type U-937 cells was

assayed by incubating the cells at different bleomycin

concentrations for 6 hours prior to plating them in agar.

Figure 15 shows the effect of increasing concentrations of

bleomycin on the clonogenic survival of U-937 cells. High

colony efficiency (100%) is observed up to 1 Ijg/ml

bleomycin. Above 1 gg/ml bleomycin colony efficiency

decreases rapidly.




1 I

0 0.1 1 10
Bleomyoin ug/ml

Figure 15. Bleomycin dose-response for colony formation by
U-937 cells. Wild type U-937 cells (105 cells/ml) were
treated with increasing concentration of bleomycin for 6
hours. After 7-10 days, the colony efficiency was recorded
plotted for each concentration of bleomycin. Data represents
mean +/- 1 SD of triplicates of two experiments.

Bleomycin Effect on the Growth and Morphology of

Transfected U-937 Cells

MNC and bcl-2-U-937 cells were incubated in the

presence or absence of 1 pg/ml bleomycin for at least 10

days and analyzed for cell growth by trypan blue exclusion.

Figure 16 shows the effect of 1 pg/ml bleomycin on the total

number of viable MNC-U-937 and bcl-2-U-937 cells over 10

days of culture. Untreated MNC cells and bcl-2 U-937 cells

demonstrate about 1 log increase in viable cell numbers

through 10 days of culture. In contrast, MNC-U-937 cells

treated with bleomycin show an increase in viable cell

numbers to about 3 X 105 cells/ml through 4 days in culture

followed by a decrease viable cells to the starting levels

of 1 X 105 cells/ml at 7 days of culture. In contrast, bcl-2

U-937 cells show an slow increase in viable cell numbers to

about 5 X 105 cells/ml through 7 days of culture in the

presence of the drug suggesting that Bcl-2 expression can

prevent the loss of viable cells seen in the continuous

presence of the bleomycin.

S 10- N
X Untreated
S-e- MNC

6 -- BCL-2


0 1 3 4 7 10 11
Bleomycin Treatment (days)

Figure 16. Bleomycin effect on the total number of viable
transfected U-937 cells. MNC and bcl-2 transfected U-937
cells were cultured in the presence of bleomycin (1 pg/ml)
for 10 days and the number of viable cells were counted and
compared with viable cell numbers of untreated cells. Data
represents mean of triplicates of three experiments +/- 1


In addition, treated and untreated MNC and bcl-2-U-937

cells were also analyzed for cell cycle progression by flow

cytometry. Figure 17 shows the cell cycle changes of the

transfected U-937 cells through 4 days of culture in the

presence of 1 ig/ml bleomycin. MNC-U-937 cells show an

increase in the percentage of cells in the Go/G1 phase from

39% to 46% at 3 days and 61% at 4 days of culture in the

presence of drug. In contrast, bcl-2-U-937 cells show no

change in Go/G1 or any other cell cycle phases indicating

that Bcl-2 blocks synchronization of cells induced by


Transfected cells were treated with bleomycin as

described above and the effect of bleomycin on cell

morphology was assayed by flow cytometry, using side scatter

to measure cell granularity and CD1b as markers of myeloid

differentiation. Figure 18 shows a 2D analysis of MNC and

bcl-2 U-937 cells after 2 and 4 days of culture in the

presence of 1 Ag/ml bleomcyin. MNC U-937 cells show an

increase from 2% to 41% of cells expressing up-regulated

CDllb and granularity measured by side scatter at 2 days of

Figure 17. Effect of bleomycin on the cell cycle.
Transfected U-937 cells were treated in the presence
or absence of 1 jg/ml bleomycin for 4 days. Nuclei
were isolated and stained with PI as described in
Methods. Percentage of cells in different cell cycle
phases were obtained by using flow cytometry and
CellFit analysis program. MNC U-937 cells treated with
bleomycin show an increase in the percentage of cells
in Go/G1, whereas bcl-2 U-937 cells do not demonstrate
an increase in GO/G1 during the same period of
treatment with bleomycin.




BLM 72H Go/G1:30%

BLM 96H Go/G1:36%


BLM 96H Go/G1:61%



a) ud an ) ( o J
- l ) (d.

r 4H 3 0 (4
S-a) rl m F0

u 14 0 r 4

m )) m M
-. I rn 0 1 fA

0, Q M 0

4J 0) O A-

03 o m A 4

r (d (D I (0
4r-I *

H- m1-H

0 S 4)
*^ rq 0 0 .A
0 0 ri 0 r
o M

S U "-1 0

=n m
r H a

0 o --g 4
0 0
0 o 0 M *

.5 4g a)
"r Q r- -A (1)
g I 0 a -0 <
0 (A 0 0 )4 E

-H Pi D k -ri o e

4 U. -.-)14
o o o u u
4 r-i r A

00 a)
Md (d(U. 4

Sj M Mi41O
b a, ( 1 H
04 r) O Pd en

4D ) aM 4i

Ul 4) 44 (d 14 r-I 0 M
.i $4 44 14 (d 0) 4 (d
4 4 0c tu P4 rO t)


dP dP

H n

w) I .
r-i Ir



mv ..







culture. CD1lb expression increases to 62% of differentiated

cells by 4 days of treatment. In contrast, bcl-2-U-937 cells

demonstrated a delayed increase in granularity and CDllb

expression until 4 days of treatment, with 62% of

differentiated cells.

The above observations indicate that MNC-U-937 cells

stop dividing and start dying by 4 days of culture in the

presence of 1 gg/ml bleomycin. This corresponds to about the

same time that differentiation markers (i.e. granularity and

CDllb) have increased and the cells have synchronized in

Go/G1. In contrast, during this same treatment time, bcl-2

U-937 cells increase in viable cell numbers and show a

delayed increase in cell differentiation with no

synchronization in Go/G1. This indicates that Bcl-2 can

prevent cell death even when cells have differentiated.

Cell Death Post-Bleomycin Differentiation

In order to determine whether dying cells in the MNC U-

937 cell population were terminally differentiated, MNC and

bcl-2-U-937 cells were treated with bleomycin (1 Ag/ml) for


5 days. After this time, bleomycin was removed and the cells

returned to the incubator for 2 weeks. Cells were checked

for viability by trypan blue dye exclusion and morphology by

flow cytometry throughout this 2 week period. Figure 19

shows the flow cytometric analysis immediately after 5 days

of bleomycin treatment (day 0) and at various times after

the drug was removed. Before flow cytometric analysis, cells

were stained with propidium iodide to exclude dead cells

from the analysis and the percentage of differentiated cells

with high side scatter and CD11b expression calculated after

subtracting background flouresence. In panels A, viable MNC

transfected cells demonstrate a significant decrease in the

percentage of cells with up-regulated SSC and CD11b, with

20% at 5 day treatment and down to 0.13% at 2 weeks after

removing the drug (part B), while no drop in the percentage

of differentiated bcl-2-U-937 cells (35.5%-33.2%) is seen

for the same period of time. At this time, cells were once

again treated with bleomycin for 5 days. Top panel in part B

shows re-induction of differentiation in MNC-U-937 cells

from 0.13% to 25%, while bcl-2-U-937 cells increase from 33%

to 42% the percentage of differentiated cells.

Im )EiUH I 0
-- ) tn o 0 .1r 4 o
. 0) 0 I-l W
u0 4 4 -ri0
u 4) UM

U m A
> A $4 J 0
4J M o a m de

0 3 0 O *n 0 0 a
0; 0Ol
0 14 V, 4.)

a H i *0 x- o
.0 r0 4. l

0 44 a, -4 0 44 O 4

0) U 0 0, o

S .-1 4 4 U l O
1P i r- in A M o -.I 44
4 0 0) r4 i 00) n 0
* (D ) !

Sa ro $4 4 M

A __s 0 M I

44 r.1 M 4Ju

M 4 0 >1-Q. 4
|ld H .
.l 4 'Q J > a
. ( d v F3 0 *
M H M ( MO 4 4
)a 0 $ E4 4)

0 w C 4 4) S "
4 )r r *d )
u 0 a 14

H 0 $4 m -
A ) 0) t0 0< 44 H
3a) 4 ro ( d I >, 4
$4 (4) 43 tQ rol g

4) f d 14 4J r d

*rl 4 ) 4) 0 Id Hl I
h E) d is! -rl Q P,













Co U)



~t---- w


The number of viable cells were also determined after 5

days of treatment and daily thereafter after removing the

bleomycin. Figure 20 shows the number of viable cells and

the number of viable cells expressing differentiation

markers for both MNC U-937 cells (fig 20A) and for bcl-2 U-

937 cells (fig 20B) starting after 5 days of bleomycin

treatment (time 0). MNC U-937 cells show an initial decrease

in the number of viable cells 1 day after drug removal,

followed by an increase in viable cell numbers of about 1 X

106 cells/ml after 15 days in the absence of drug (top

curve). The lower curve in panel A shows the number of

viable differentiated cells in MNC-U937 cells. This was

calculated by multiplying the percentage of differentiated

cells obtained by flow cytometry (figure 19) times the

number of viable cells. MNC-U937 cells show a rapid decrease

(about 2 logs) in the number of viable differentiated cells

15 days after removing the drug. This is followed by a rapid

increase in number of differentiated cells in MNC-cells

after readdition of 1 Jg/ml bleomycin for 5 days. The same

analysis was done for bcl-2 U-937 cells and is shown in

panel B. Bcl-2 cells demonstrate an increase in viable cell


numbers to about 1 X 106 cells/ml after 15 days in the

absence of the drug. In contrast to MNC-cells, bcl-2-U-937

cells show little change in the number of viable

differentiated cells present throughout the 2 week period

after removing the bleomycin (fig 20B, lower curve). Fig 20B

also shows an increase of viable differentiated cells after

adding back bleomycin for 5 days. In summary,

differentiated MNC U-937 cells continue dying after removing

bleomycin, and the cells recovering from bleomycin are

undifferentiated, while differentiated bcl-2 U-937 cells

continue to survive after removing the drug.


To determine whether apoptosis was involved in the

death of differentiated cells, MNC and bcl-2 U-937 cells

were treated with 1gg/ml bleomycin for 5 days and checked

for DNA fragmentation by flow cytometry. Figure 21 shows the

percentage of apoptosis of cells cultured in the presence

and absence of bleomycin. Bivariate analysis of DNA strand

breaks, measured as dUTP incorporation and DNA content by

Figure 20. The Number of viable differentiated cells
after bleomycin treatment. MNC and bcl-2 U-937 cells
were treated as in figure 19. The number of viable
cells/ml was determined by trypan blue dye exclusion,
and the number of viable differentiated cells/ml was
obtained by multiplying viable cell number times the
percentage of differentiation obtained by flow
cytometry after excluding dead cells by PI. A) MNC U-
937 cells show an initial decrease in viable cells
followed by an increase to about 106 cells/ml 15 days
following removal of bleomycin (M--U); whereas, the
number of viable differentiated cells decrease
throughout the 15 days after removing the drug to
about 1:100 of initial differentiated cell count (D--
D). B)bcl-2 U-937 cells increased viable cell numbers
throughout 15 days after removing bleomycin (M--M),
and maintained the number differentiated viable cells
(0--D). At day 15, bleomycin was added back for 5
days and an increase in the number of differentiated
cells was seen for both MNC and bcl-2 U-937 cells.




- (A) MNC
Viab. Cells

Diff. Cells

0 1 2 5 15
Days of Culture

1-~ -8

--- (B) BCL-2
Viab. Cells

Diff. Cells








1 2 5 15 2
Days of Culture



Figure 21. Apoptosis after 5 day bleomycin treatment. MNC
and bcl-2 U-937 cells were treated with 1lg/ml bleomycin for
5 days and checked for apoptosis by flow cytometry as
explained in flow cytometry section of Methods. Untreated
cells or treated bcl-2 cells do not show dUTP incorporation
indicating that these cells are not apoptotic, while
bleomycin treated MNC U-937 cells show 47% of cells in
apoptosis, with an apparent accumulation of these cells in






propidium iodide are shown. Untreated MNC and bcl-2-U-937

cells as well as bleomycin treated bcl-2-U-937 cells do not

show DNA fragmentation, while MNC U-937 cells treated for 5

days with bleomycin show 47% of cells in apoptosis with an

apparent accumulation of apoptotic cells in Go/G1 of the

cell cycle.


The effect of up-regulated Bcl-2 in myeloid cell lines

treated with a free-radical inducer was examined. Results

from these studies indicate that sublethal concentrations of

bleomycin decrease cell growth which is accompanied by an

accumulation of cells in Go/G1 of the cell cycle and

increased expression of markers associated with cell

differentiation. The up-regulation of Bcl-2 prevents a

decrease in cell growth and delays but does not block the

action of bleomycin on the induction of differentiation

markers. The cytotoxicity of sublethal concentrations of

bleomycin is mediated by a mechanism in which apoptosis is

involved. The Bcl-2 oncoprotein can block the apoptosis

associated with cell




after bleomycin




The mechanisms by which Bcl-2 prevents apoptosis have

been widely explored, but not yet precisely understood.

Recent studies suggest that the bcl-2 gene product can

result in relative resistance to the cytotoxic effects of

many anti-neoplastic compounds (Hickman et al.,1992; Lotem

and Sachs,1993; Nunez et al.,1990; Miyashita and Reed,

1993). Like most cancer chemotherapy agents, Ara-C induces

apoptosis in cancer cells (Chabner and Myers,1993). Ara-C is

incorporated into DNA bringing about DNA fragmentation and

growth inhibition by an as yet still unknown mechanism

(Major et al.,1981). Ara-C has also been shown to be an

effective agent in the treatment of acute myelogenous

leukemia (Bhalla et al.,1984). In the present study, the

level of Bcl-2 expression in the protection of myeloid


leukemia cells from Ara-C was investigated. In addition, the

effect of up-regulated Bcl-2 in myeloid leukemias treated

with the free radical inducer bleomycin (Matkanov,1993), was

also investigated.

Bcl-2 Levels and the Resistance to Ara-C

Hematological tumor cell lines expressing high Bcl-2

levels such as the follicular lymphoma RL-7 and the pre-

myeloid leukemia KG-1 were found to be markedly resistant to

Ara-C concentrations a 10M; whereas, the low Bcl-2

expressing myeloid leukemia U-937 was sensitive to Ara-C

concentrations 5 0.01 PM. These results demonstrate a

correlation between the levels of deregulated Bcl-2 and the

resistance to chemotherapy. Campos et al., 1993 and 1994

have observed high Bcl-2 expression in acute myelogenous

leukemias (AML) which has been associated with lower

complete remission rates and a shorter survival time for

patients. Similarly, a recent study (Bradbury et al. 1994)

reported that autonomously growing AML blasts which express


high Bcl-2 levels are more resistant to apoptosis than cells

with non-autonomous growth which express low Bcl-2 levels.

Although these studies support a correlation between Bcl-2

levels and chemotherapy resistance, these investigators

reported only qualitative results based on a single dose of

chemotherapy. In contrast, this dissertation clearly

demonstrated a direct correlation between Bcl-2 levels and

Ara-C concentration using multiple doses and also

quantitated the degree of resistance to Ara-C by Bcl-2.

The expression of Bcl-2 was also analyzed during the

action of Ara-C. The pre-myeloid leukemic cell line KG-1 was

found to be sensitive to high Ara-C concentrations (100lM),

and this sensitivity was not accompanied by complete Bcl-2

downregulation, but rather a decrease of only 35-40% in Bcl-

2 concentration. This result has several implications. One,

a narrow threshold concentration of Bcl-2 oncoprotein may

only be required to protect cells from high doses of Ara-C

and any slight decrease in this threshold may be sufficient

to render cells chemotherapy sensitive. Two, the anti-

apoptotic function of Bcl-2 may be decreased by upregulation


of Bax when cells are exposed to high Ara-C by changing the

ratio of the Bcl-2/Bax heterodimer. Three, sensitivity to

very high Ara-C doses may depend on high expression of

proteins such as Bcl-X (short) which can render cells

chemotherapy sensitive regardless of the presence of Bcl-2.

Thus, the sensitivity

of KG-1 cells

to high

concentrations support the observation that a small

decrease in the Bcl-2 levels can render cells

chemosensitive. Results reported by Keith et al., 1995

demonstrate that inhibition of Bcl-2 by antisense

oligonucleotides induces apoptosis and increases the

sensitivity of patient AML blasts to Ara-C. In contrast, the

results reported here using KG-1 cells show that Bcl-2

levels can be downregulated in myeloid leukemias after

treatment with high doses of Ara-C and this results in

increased sensitivity of KG-1 cells to the drug.

Furthermore, these results suggest that increasing the

intensity of the anti-cancer treatment will bring down Bcl-2

levels which may be beneficial to treatment of leukemia. Of

course this will require bone marrow transplant protocols

since increases in chemotherapy dose will induce



myelosuppression. Taken together, the results reported here

further demonstrate that it is the amount of Bcl-2 and not

only the mere presence the important factor in

chemoresistance. Techniques are needed to determine not only

the number of Bcl-2 positive cells, but also the Bcl-2

concentration of these positive cells. In addition, it is

important to determine the relative Bcl-2 concentration in

relation to other Bcl-2 family proteins such as Bax and Bcl-


To further study the effects that the Bcl-2 oncoprotein

has on the resistance of myeloid cells to Ara-C, murine bcl-

2 was transfected into the myeloid leukemic cell line U-937.

Bcl-2 U-937 cells demonstrate increased resistance to Ara-C

when analyzing apoptosis by flow cytometry. In addition,

bcl-2 U-937 cells also demonstrate a 3 log shift in

resistance to Ara-C when comparing LD50 colony forming values

with wild type or MNC U-937 cells. Further analysis of the

results with bcl-2 U-937 cells indicate that up-regulation

of Bcl-2 protects the cells from apoptosis but it cannot

prevent a decrease in clonogenicity observed with Ara-C


concentrations > 10M. This suggests that with high doses of

Ara-C a cytostatic effect is seen in cells expressing high

levels of Bcl-2 oncoprotein. It remains to be determined

whether these cells can continue cell cycle progression

after removing the drug.

The results of this study show that Bcl-2 can protect

non-lymphoid cells from Ara-C. Furthermore, these results

have important implications in the treatment of myeloid

leukemias, since the sensitivity to Ara-C correlates with

intracellular levels of Bcl-2 as has been observed in M4 and

M5 types of AMLs (Bradbury et al.1994; Keith et al.,1995).

In addition, this study shows that the level of Bcl-2

expression is an important parameter to determine when

analyzing which tumors might respond to chemotherapy.

Further studies are warranted to determine how to modulate

Bcl-2 levels to bring about a therapeutic response.

Up-regulated Bcl-2 in Myeloid Cells Treated with Bleomycin

The anti-tumor action of bleomycin has been attributed

mainly to DNA strand scission involving a bleomycin-iron


complex from which hydroxyl radicals can be generated and

DNA damaged (Chabner and Myers,1993). On the other hand,

Bcl-2 protection has been shown to be mediated by an anti-

oxidant mechanism in which the peroxidation of membrane

lipids by hydroxyl radicals is blocked (Korsmeyer, 1995).

In this investigation, the role of Bcl-2 in myeloid

cells treated with bleomycin was analyzed. Sublethal

concentrations of bleomycin induced a decrease in cell

growth in both MNC and in bcl-2 U-937 cells. In addition, a

decrease in the number of viable cells present after 5 days

of bleomycin treatment was seen in MNC-U-937 cells. In

contrast, a decrease in viable cell number for bcl-2-U-937

cells was not seen until 10 days of bleomycin treatment.

The cell growth arrest induced by bleomycin was also

accompanied by an accumulation of MNC U-937 cells in the

Go/G1 phase (60%) of the cell cycle; whereas bcl-2-cells did

not accumulate in any cell cycle phase during 4 days of

bleomycin treatment. These results also show that both MNC

and bcl-2 transfected cells up-regulate the expression of

parameters associated with myeloid differentiation such as

granularity and CDllb antigen. MNC U-937 cells show a rapid


increase of these parameters, while bcl-2-U937 cells

demonstrated a delay of greater than 24 hours in the

expression of myeloid differentiation. These results are

consistent with bleomycin cytotoxicity being mediated after

the induction of myeloid differentiation in MNC U-937

cells. Furthermore, the post-differentiation effect of MNC

U-937 cells is closely related to an increase in Go/G1 phase

of the cell cycle. These results are not consistent with the

cell cycle effects previously described for bleomycin and

other similar DNA damaging agents such as y-radiation and

quinones (Barranco and Humphrey,1986). At toxic

concentrations, these agents exhibit a G2/M accumulation as

a consequence of DNA strand scission. One possible

explanation for these contradictory findings is that

bleomycin affects different cell cycle phases depending on

the concentration of the drug and time of exposure.

Consistent with this hypothesis is the study by Fornari et

al.,(1994) who used sublethal concentrations of the quinone

Doxorubicin, an agent that also damages DNA by free

radicals. The breast cancer cell line MCF-7 was shown to be


induced to differentiate by long exposure to sublethal

concentrations of Doxorubicin. At the same time, these

doxorubicin-treated MCF-7 cells accumulated in Go/G1.

Therefore, cell differentiation appears to be a common

effect when using sublethal concentrations of free radical

inducers. (Fornari et al.,1994). Taken together, these

observations suggest that myeloid leukemias can be induced

to differentiate by products of the Fenton reaction such as

OH., further implicating free radicals in the

differentiation and apoptosis of myeloid leukemias (Nagy et

al.,1993 and 1994; Kobayashi et al.,1994). Cytotoxic

concentrations of these drugs induce DNA strand breaks by

rapidly accumulating OH. radicals resulting in

synchronization of cells in the G2/M of the cell cycle. In

contrast, sublethal drug concentrations induce accumulation

of cells in Go/G1 and the cells differentiate which may be

mediated by a slow production of free radicals.


Extended Viability of Differentiating Myeloid Leukemia by
Up-regulated Bcl-2

To further show that bleomycin induces cell death

through terminal cell differentiation, MNC and bcl-2 U-937

cells were treated with bleomycin for 5 days. At this time,

the drug was removed and the cells incubated without drug

for 2 weeks. Both MNC and bcl-2 transfected U-937 cells were

induced to differentiate during 5 days of bleomycin

treatment. After removing the drug, MNC differentiated cells

continue to die, whereas, the differentiated cells in the

bcl-2 transfected group remain viable through 2 weeks of

culture. This study also demonstrated that the

differentiated MNC-U-937 cells die by apoptosis, further

implicating Bcl-2 in the extended survival of differentiated

myeloid leukemic cells. This is consistent with cell death

occurring during terminal differentiation of myeloid cells

such as granulocytes and macrophages. These cells have a

limited life span, but certain cytokines such as granulocyte

or granulocyte-macrophage colony stimulating factors,

interleukin-l interferon-y, and bacterial lipopoly-


sacharides (LPS) can prevent their death (Klebanoff et

al.,1992, Colotta et al., 1992). When differentiated cells

do not receive these signals, a cell death mechanism is

activated. In the presence of high levels of anti-apoptotic

proteins such as Bcl-2, the cell death pathway is blocked.


The results of this dissertation clearly demonstrate

that up-regulated Bcl-2 can protect myeloid leukemic cells

from the apoptosis induced by chemotherapeutic agents. There

is a direct relationship between the level of Bcl-2 and the

extent of chemoresistance. Hematological tumors which

express high levels of deregulated bcl-2 such as is seen

with follicular lymphoma which carry the t(14;18)

chromosomal translocation show the highest resistance. In

the myeloid leukemias, pre-myeloid cell types, such as KG-1

cells, appear both to express high Bcl-2 levels and to be

more resistant than the more differentiated type of myeloid

leukemia represented by U-937 cells that express low Bcl-2

levels. These observations suggest that Bcl-2 levels are


developmental regulated in myeloid cells and that the level

of Bcl-2 and not the mere presence of the protein is

important as a determining factor in chemoresistance. In

addition, the results obtained in this investigation

demonstrate that Bcl-2 can prevent apoptosis induced by

chemotherapy in rapidly growing tumors as well as

differentiated cell populations. Thus, different cell cycle

perturbations can induce apoptosis and Bcl-2 can protect

against them.

Future Direction

The results of this dissertation will help to

understand how the bcl-2 gene allows hematological tumors to

become resistant to chemotherapy. An immediate application

of this study will be to design clinical protocols to

analyze the relationship between Bcl-2 levels of expression

and the extent of chemoresistance of hematological tumors.

Specifically, this study should help to better understand

bcl-2 expression in non-lymphoid hematological tumors.

Further analysis of bcl-2 expression during chemotherapy