W“WUNW)lWUlllHlllHllNWIHWHWIlHI

   

THESIS

/

JIIIIIIIIIIIIIIIIIIIII‘IIIII’I‘IIIIIIII IIIIIIIIIIIII L

3 1293 01701 9195

This is to certify that the

thesis entitled

Riga/awn 01" Human Breflfc‘ Can (er Cell
(7606/) by {Pho‘njolt‘pt‘df

presented by
Chi Zhary

has been accepted towards fulfillment
of the requirements for

Major professor

Date 8‘21-‘1‘5

0-7539 MS U is an Affirmative Action/Equal Opportunity Institution

 

PLACE IN RETURN BOX to remove thi
To AVOID FINES return on
MAY BE RECALLED with earlier

 

 

LIBRARY

Michigan State
University

 

 

s checkout from your record.
or before date due.
due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

REGULATION OF HUMAN BREAST CANCER CELL DEATH BY
SPHINGOLIPIDS

By

Chi Zhang

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Food Science and Human Nutrition

1998

ABSTRACT

REGULATION OF HUMAN BREAST CANCER CELL DEATH BY
SPHINGOLIPIDS

By
Chi Zhang

Breast cancer is the most common malignant cancer among women and the incidence
is increasing in most countries. Improved treatments are needed including new
chemotherapeutic agents. This study examined the effects of sphingosine and ceramide,
two bioactive metabolites of more complex sphingolipids, on the death of MDA-MB-231
human breast cancer cells which are estrogen receptor-negative and multi-drug resistant.
Both the naturally occurring form of sphingosine, D—erythro, and a cell—permeable form of

ceramide, C2-ceramide, caused cell death in concentration- and time-dependent manners.

Data from terminal deoxynucleotidyl transferase (T dT)-mediated dUTP nick end labeling,
agarose gel electrophoresis of DNA extracts, and flow cytometric analysis suggest that
sphingosine induces apoptosis while ceramide causes death via a non-apoptotic pathway.

Both exogenously added sphingosine and ceramide increased the cellular level of
sphingosine, but the increase caused by sphingosine was much higher than that induced by
ceramide. The results suggest that the cellular sphingosine concentration necessary to
induce breast cancer cell apoptosis is about 5-10 nmol/mg protein. The unnatural synthetic
sphingosine stereoisomers (D-threo, L-threo, L-erythro) also caused death of human breast
cancer cells with potencies greater than or equal to D—erythro-sphingosine. These studies
suggest that D-erythro-sphingosine and its unnatural stereoisomers may provide new

means to treat patients with advanced breast cancer who are estrogen receptor-negative.

TABLE OF CONTENTS

ACKNOWLEDGEMENTS
LIST OF FIGURES
1. LITERATURE REVIEW
A. Breast cancer is a growing problem
1. Breast cancer epidemiology
2. Causes of breast cancer
3. Breast cancer development and diet
B. Chemotherapeutic treatments for estrogen receptor-negative
breast cancer are needed
1. Breast cancer histology
2. Breast cancer detection and therapy
3. Estrogen receptor status and breast cancer chemotherapy
C. The sphingolipids, sphingosine and ceramide inhibit cell growth
and induce cell death
1. Sphingolipid metabolism
2. Sphingosine inhibits cell growth and induces apoptosis
3. Cerarnide inhibits cell growth and induces apoptosis

D. Sphingosine and ceramide may provide a novel means to treat
patients with estrogen receptor-negative breast cancer

11. OBJECTIVES
III. MATERIALS AND METHODS
IV. RESULTS

A. Sphingosine and ceramide differentially affect death of estrogen
receptor-negative MDA-MB-23l human breast cancer cells

B. Structural requirements for sphingosine and ceramide-induced death
V. DISCUSSION
VI. SUMMARY AND FUTURE RESEARCH

VII. REFERENCES

111

Page
iv

WNUJU) Nt-‘t-‘H

fl
Cub-h

12

l3

14

17

17

29

35

4O

42

ACKNOWLEDGMENTS

I would like to express my sincere appreciation to Dr. Joseph J. Schroeder, for his
support, patience and encouragement. I would also like to thank my graduate guidance
committee members, Drs. Maurice Bennink and Dale Romsos, for their kindness and
valuable advice. My sincere gratitude goes to Drs. Les Bourquin, Doyle Lee, Louis King

and Pamela Fraker, for lots of help and advice they gave me during the past two years.

Also, I would like to acknowledge my friends and colleagues who have supported my
effort and made the lab such a pleasant place to work: Eun-Hyun Ahn, Julie Arnold, Min

Sun Kim, Stacey Tremp, Jason Wiesinger and Hong Yang.
My deep love and appreciation goes to my parents and younger brother, for their

love, support and encouragement. Finally, I would like to thank my husband, Bing, for his

love, support and understanding.

iv

Figure 1.

Figure 2.
Figure 3.

Figure 4.

Figure 5.

Figure 6.

Figure 7.

Figure 3.

Figure 9.

Figure 10.
Figure 11.
Figure 12.

Figure 13.

Figure 14.

Figure 15.

Figure 16.

LIST OF FIGURES
Sphingolipid turnover leads to formation of intracellular second
messengers
Sphingosine induces apoptosis
Cerarnide induces apoptosis

Sphingosine and ceramide inhibit growth and induce death of human
breast cancer cells

Sphingosine and ceramide cause DNA strand breaks in human breast
cancer cells

Cerarnide does not cause intemucleosomal DNA fragmentation in human
breast cancer cells

Sphingosine induces intemucleosomal DNA fragmentation in human
breast cancer cells

Flow cytometric analysis of human breast cancer cells cultured with
ceramide

Flow cytometric analysis of human breast cancer cells cultured with
sphingosine

Exogenous sphingosine increased cellular sphingosine concentration
Exogenous Cz-ceramide increased cellular sphingosine concentration
Structures of sphingosine stereoisomers

Sphingosine stereoisomers inhibit growth and induce death of human
breast cancer cells

Sphingosine stereoisomers inhibit growth and induce death of human
breast cancer cells (photographs)

Structures of Cz-ceramide and Cz-dihydroceramide

Cz-dihydroceramide does not cause DNA strand breaks in human breast
cancer cells

1. LITERATURE REVIEW
A. Breast cancer is a growing problem

1. Breast cancer epidemiology-The risk of developing breast cancer is about
one in eight American women (Claus et al., 1991; Marshall, 1993), and the risk of a
woman living in the United States dying of breast cancer is 3.6 percent. Any family history
of breast cancer on either the maternal or paternal side increases risk, but the increase is
fairly small except in women who have first-degree relatives (mothers or sisters) with
breast cancer (Osteen et al., 1986). Some studies indicate that the risk is further increased
if a first-degree relative had either premenopausal or bilateral breast cancer (Osteen et al.,
1986). The risk of breast cancer decreases as the age of the patient at first full term
pregnancy decreases (Osteen et al., 1986). Certain benign, proliferative changes seen in
biopsy specimens are associated with an increased risk for subsequent development of
breast cancer (Osteen et al., 1986).

2. Causes of breast cancer-Mutations in BRCAI , BRCA2 and potentially a few
other genes are responsible for increases in risk of breast cancer incidence on a familial
basis. The BRCAI gene has been localized to chromosome 17q21 (Miki et al., 1994).
Although its normal function is still unknown, it appears to confer tumor suppresser
activity. About five percent of female breast cancer patients may have alteration(s) in
BRCAI (Miki et al., 1994). If their relatives are carriers of the BRCA1 mutation(s), they
have about 85 percent lifetime risk of breast cancer with 50 percent of the breast cancers
occurring before age 50 (Miki et al., 1994). Genetic abnormalities of BRCA2, which have
been localized to chromosome 13q12-13, also confer a high risk of breast cancer (Wooster

er al., 1994). Despite inherited defective genes which predispose women to breast cancer,

about 90 percent of breast cancer incidence is believed to be caused by environmental
factors such as diet, lifestyle, and exposure to environmental contaminants (Wright, 1990;
Lynch et al., 1992).

3. Breast cancer development and diet-Certain dietary factors affect the
development of breast cancer and prevention of breast cancer through dietary modification
is an active area of clinical and epidemiological research. The risk of breast cancer may be
related to the total fat content of the diet (W ynder et al., 1997). As women from oriental
countries with a low fat content in the diet increase their fat intake or migrate to western
countries, the incidence of breast cancer increases, even within the migrating generation
and particularly in later generations (Osteen et al., 1986). Although controversial, some
studies have suggested that obese, postmenopausal women may have a higher risk of breast
cancer than thin women (Cleary and Maihle, 1997). Also overweight or obese breast
cancer patients might have a poorer prognosis (Bastarrachea et al., 1994). Some animal
studies showed that caloric restriction reduced breast cancer incidences (Klurfeld et al.,
1991). High alcohol intake probably increases the risk of breast cancer (Friedenreich et al.,
1993). In addition, other studies suggest that dietary carotenoids (Verhoeven er al., 1997)
and vitamins A (Sankaranarayanan and Mathew, 1996), C (Verhoeven et al., 1997) and E
(Kimmick et al., 1997) might play a preventive role in breast cancer occurrence. Also,
many studies have suggested that a lower risk of breast cancer is associated with higher
intake of dietary fiber and carbohydrates (Stoll et al., 1996). Taken together, current
information suggests that a diet which is low in fat, and high in carbohydrate, fiber,

vegetables and fruit may help to reduce the development of breast cancer. Further,

maintenance of body weight within recommended levels by proper food and energy intake

is also important.

B. Chemotherapeutic treatments for estrogen receptor-negative breast cancer are
needed

1. Breast cancer histology-The majority of breast cancers are poorly
differentiated ductal carcinomas (Osteen et al., 1986). Breast cancers often progress from a
hormone-dependent, nonmetastatic, antiestrogen-sensitive phenotype to a hormone-
independent, antiestrogen- and chemotherapy-resistant phenotype with highly invasive and
metastatic growth properties. This progression is usually accompanied by altered function
of the estrogen receptor or outgrowth of estrogen receptor-negative cancer cells (N akshatri
et al., 1997).

2. Breast cancer detection and therapy-Mammography and, more recently,
breast magnetic resonance imaging are used for breast cancer detection and diagnosis (Orel
et al., 1994). Breast cancer is treated by appropriate combinations of surgery, radiation
therapy, chemotherapy, and hormonal therapy, and is curable if detected in early stages.
The choice of treatment is based on tumor stage, lymph node status, estrogen- and
progesterone-receptor levels in the tumor tissue, menopausal status, general health of the
patient and patient age (de la Rochefordiere et al., 1993). Normally, primary treatment
(mastectomy or lumpectomy with radiation therapy) is followed by hormonal therapy and
adjuvant chemotherapy.

3. Estrogen receptor status and breast cancer chemotherapy-Estrogen
receptor status is a very important criterion to select appropriate hormonal therapy or

adjuvant chemotherapy (e.g. cyclophosphamide, doxorubicinn, menthotrexate, 5-

fluorouracil, prednisone, vincristine and tamoxifen) for breast cancer patients at different
staging categories (Greenspan, 1996). For patients with positive estrogen receptor status,
hormonal therapy (e.g. tamoxifen) is the first choice (Fisher et al., 1989). Chemotherapy is
usually applied for endocrine-resistant patients, estrogen-receptor negative patients, or
patients who have life-threatening metastases (Clavel and Catimel, 1993). Doxorubicin is
one of the most effective single cytotoxic agent, giving a 40 percent response rate in
previously untreated patients (Ahmann et al., 1974). In most cases, combination
chemotherapy regimens are used because they are much more effective than a single agent
(Ahmann et al., 1974; Canellos et al., 1976; Jones et al., 1975). Combination
chemotherapy regimens generally produce higher response rates, ranging from 50 to 80
percent (Canellos et al., 1976; Jones et al., 1975). However, results are still unsatisfactory
because the fraction of complete responders is less that 20 percent, the duration of response
is less than 1 year, and the median survival time of these patients is about 2 years (Legha et
al., 1979; Eddy, 1992). Therefore, the search for new chemotherapeutic agents and more
effective combinations must continue.
C. The sphingolipids, ceramide and sphingosine, inhibit cell growth and induce cell

death

1. Sphingolipid metabolism-The influence of specific dietary factors or drugs
on the development of breast cancer has received substantial attention in medical research.
However, to this point, little has been done to assess the potential of sphingolipids in
treating breast cancer. Sphingolipids are a family of bioactive lipids which regulate cell
behavior (Bell et al., 1993; Merrill et al., 1997). Specifically, ceramide, sphingosine and

sphingosine 1-phosphate are bioactive metabolites of more complex sphingolipids (Figure

1). These bioactive lipids serve as second messengers for the effects of some extracellular
agents on cell growth, oncogenesis, differentiation, and cell death. Studies conducted
using HL-60 human leukemia cells have shown that the addition of 1a, 25-
dihydroxyvitamin D3, an inducer of differentiation, activates a neutral sphingomyelinase
and caused early and reversible hydrolysis of sphingomyelin .and the concomitant
generation of ceramide (Okazaki et al., 1989) (Figure l). The effect of la, 25-
dihydroxyvitamin D3 on cell differentiation is mimicked by cell-permeable ceramide
analogs with shorter N-acyl chains (Okazaki et al., 1994). Tumor necrosis factor-a (TNF-
or) and v-interferon also have been found to induce sphingomyelin turnover and ceramide
production (Kim et al., 1991; Mathias et al., 1991; Dressler et al., 1992). In some cases,
ceramide can be further metabolized to sphingosine by ceramidase (Nikolova-Karakashian
et al., 1997) (Figure 1). Moreover, sphingosine can be phosphorylated by sphingosine
kinase to form sphingosine-l-phosphate (Zhang et al., 1991) (Figure 1). Thus, some
effects previously attributed to ceramide may be mediated via its conversion to sphingosine
or sphingosine- 1 -phosphate.

2. Sphingosine inhibits cell growth and induces apoptosis-Sphingosine is a
bioactive compound. The observation that sphingosine is an inhibitor of protein kinase C
lead to the initial interest in sphingolipids as possible second messengers (Hannun et al.,
1986). Sphingosine added exogenously to tissue culture medium is taken up by the cells
and metabolized to form other lipids (Smith and Merrill, 1995). Cellular sphingosine can
also activate targets to regulate cell behavior. Some recent research shows that sphingosine
plays an important role in apoptosis (Ohta et al., 1994; Ohta et al., 1995; Jarvis et al.,

1996; Nakamura et al., 1996). Conflicting observations on the involvement of sphingosine

OH

Weflzomflc'fiizwicmh
Sphingomyolln

NH

W

O
Sphingomyolinaso

 

Phosphochollno

V on

/W\/\/\/\/\/|\'/CH’OH
\
Cor-amide

NH

MAM/W
A O

Coramldaso
Coramido
syntheses \

Fatty acid

T OH OH

WWW
Sphingosine

NH,

 

 

Sphingosine kinase 0

 

V ' on o—u —on

W AH

phingosino-t phosphate NH,

Figure 1. Sphingolipid turnover leads to formation of intracellular second messengers

in apoptosis point to a great variability depending on sphingosine dosage, cell type, phase
of cell cycle and intracellular signal transduction pathway.

Exogenous addition of sphingosine was demonstrated to induce apoptosis of human
neutrophils (Ohta et al., 1994), human myeloid leukemic HI..-6O cells (Ohta et al., 1995;
Jarvis et al., 1996) and an interleukin-Z-dependent cytotoxic T cell line, CTLL-2 cells
(Nakamura et al., 1996). The intracellular level of sphingosine increased in the cases of
tumor necrosis factor-a (TNF-a) induced apoptosis of human neutrophils (Ohta et al.,
1994) and cardiac myocytes (Ohta et al., 1995) and phorbol ester induced apoptosis of H]..-
60 cells (Ohta et al., 1995; Jarvis et al., 1996), suggesting that sphingosine may function as
an endogenous mediator of apoptosis in these cells. In the experiment conducted with
neutrophils, accumulation of both ceramide and sphingosine was observed after exogenous
addition of TNF-a, with the ceramide concentration increasing prior to that of sphingosine.
Furthermore, exogenous sphingosine (5-15 pM) was capable of inducing apoptosis in
neutrophils, but ceramide and sphingosine-l-phosphate at similar concentrations did not
induce apoptosis. These findings suggested sphingosine derived from deacylation of
ceramide, but not ceramide itself, might mediate cell death in neutrophils after TNF-a
treatment (Ohta et al., 1994). Primary cell cultures appear to be less susceptible to
sphingosine than cancer cells. Sphingosine did not induce apoptosis in normal epithelial
cells such as HUVECs or rat mesangial cells, but did induce apoptosis in their transformed
counterparts (Sweeney et al., 1996).

The cellular concentration of sphingosine increased during apoptosis resulting from
phorbol ester-induced terminal differentiation of HL-60 cells (Ohta er al., 1995; Jarvis et

al., 1996). Apoptosis induced by phorbol ester and sphingosine was accompanied by a

concomitant decrease of expression of bcl—2 proto-oncogene (a suppresser gene of
apoptosis) at both the mRNA and protein levels (Figure 2), while expression of bcl-XL and
bax mRNA did not change. In contrast, expression of bcl-2 did not change in apoptosis
induced by the pharmaceutical protein kinase C inhibitors l—(5-isoquinolinesulfonyl)-2-
methylpiperazine (H7) or staurosporine. These observation suggested that sphingosine
mediates the apoptotic signaling in phorbol ester-induced terminal differentiation of HL-60
cells through down regulation of bet-2, which is probably independent of the protein kinase
C inhibition function of sphingosine (Sakakura et al., 1996). Sphingosine also induces
apoptosis in androgen-independent human prostatic carcinoma DU-45 cells, which express
bcl-XL (another cell death repressor gene) but not bcl-2 at the protein level, through down
regulation of bcl-XL (Figure 2). Again, this probably occurs independently of protein
kinase C inhibition by sphingosine. Furthermore, neither ceramide nor sphingosine-1-
phosphate induce apoptosis in DU-145 cells (Shirahama et al., 1997). On the other hand,
some studies have shown that sublethal concentrations of sphingosine synergistically
augment the apoptotic capacity of ceramide in I-1L—6O and U937 monoblastic leukemia cells
(Jarvis et al., 1996). This effect of sphingosine could also be achieved by acute co-
exposure to highly selective pharmacological inhibitors of protein kinase C or chronic pre-
exposure to the non-tumor-promoting protein kinase C activator, bryostatin 1 (Jarvis et al.,
1996). In addition, both sphingosine and H7 induced apoptosis in neutrophils (Ohta et al.,
1994). The findings demonstrate a protein kinase C-dependent pathway for sphingosine
induced apoptosis. In other words, induction of apoptosis by sphingosine might be related

to inhibition of protein kinase C activity (Figure 2).

 

TNF-a.
Phorbol ester

flng3nyelin

Ceramide

 

 

Sphingosine

// \\

VProtein kinase C “cl/2 bcl-X L Other targets

4Endonuclease

ApOptosis

Figure 2. Sphingosine induces apoptosis

3. Ceramide inhibits cell growth and induces apoptosis-Ceramide has also
been postulated as an intracellular mediator of apoptosis. TNF-a induced apoptosis in
U937 monoblastic leukemia cells. Exposure of U937 cells to TNF-a resulted in
sphingomyelin hydrolysis and ceramide generation. Treatment of cells with either TNF-a
or Cz-ceramide (a short chain, cell-permeable ceramide analog) caused intemucleosomal
DNA fragmentation characteristic of apoptosis (Obeid et al., 1993). In Molt-4 leukemia
cells, serum withdrawal caused a modest apoptotic cell death ”(Jayadev et al., 1995). Serum
deprivation of these cells resulted in significant sphingomyelin hydrolysis and elevation in
endogenous levels of ceramide. A distinct, particulate, membrane-associated and
magnesium-dependent sphingomyelinase was involved in this process. The addition of
exogenous C6-ceramide resulted in more pronounced apoptosis which occurs much sooner
(Jayadev et al., 1995). Moreover, Chouaib’s group found that a TNF-resistant variant (R-
A1) of MCF-7 human breast carcinoma cells could be induced to undergo cell death after
exposure to exogenous sphingomyelinase or cell-permeable C‘s—ceramide (Cai et al., 1997).
In addition, exogenously added cell-permeable ceramides also induced apoptosis in some
other cell lines (Jarvis et al., 1994; Karasavvas et al., 1996; Gill et al., 1997).

Domae’s group showed that c-jun/AP-l was activated by ceramide early in the
process of apoptosis and that the impairment of c-jun/AP-l by curcumin (an inhibitor of
AP-l) or antisense oligonucleotides for c-jun rescued apoptotic cells (Sawai et al., 1995).
This suggested that ceramide was crucially involved in the signal transduction pathway
leading to apoptosis through the activation of c-jun/AP-l (Figure 3).

Neither the specificity nor the mechanism of action of ceramide is fully understood.

Although Cz-ceramide was able to induce differentiation, neither sphingosine nor N-

10

 

TNF-a
Interferon-y

1a., 25-
. Dihydroxy- ,
vitamin D3

/

ICE-related proteases c-jun/AP-l Other targets

 

 

 

 

Endonulease

Apoptosis

Figure 3. Cerarnide induces apoptosis

ethylsphingosine showed any activity (Bielawska et al., 1992); this observation suggested a
critical role for the amide-linked fatty acyl group. Furthermore, Cz-dihydroceramide which
is identical to Cz-ceramide but lacks the 4,5 trans double bond of the sphingoid base
backbone failed to induce DNA fragmentation (Obeid et al., 1993), which demonstrated
that the double bond was obligatory for ceramide to induce apoptosis.
D. Sphingosine and ceramide may provide a novel means to treat patients with
estrogen receptor-negative breast cancer
In preliminary studies, we found that both sphingosine and ceramide inhibited growth and
induced death of MDA-MB-231 cells, an estrogen receptor-negative human breast cancer
cell line. Both sphingosine and ceramide were more cytotoxic for sub-confluent
continuously growing cells than for confluent quiescent cells. This suggests that cancer
cells which are more likely to be dividing in viva may be more susceptible to cell death
induced by sphingosine and ceramide than neighboring normal tissue that may be less
likely to be proliferating. The experiments described herein build upon our preliminary

observations.

12

n. OBJECTIVES

Breast cancer is the most common malignant cancer of women in the western world
(Claus et al., 1991; Marshall, 1993). Approximately 910,000 new cases were diagnosed
worldwide in 1996 (WHO, 1997), which accounted for nine percent of all new cancers.
Many studies are underway to improve human breast cancer prevention, early detection,
and treatment. Our preliminary data show that sphingosine and ceramide inhibit growth
and induce death of MDA-MB-23l human breast cancer cells which lack an estrogen
receptor (Zhang and Schroeder, 1998). The purpose of this project is to study the
mechanism by which sphingosine and ceramide induce human breast cancer cell death.

The specific aims of this project are to:

A. Determine whether sphingosine and ceramide kill cells by apoptosis or via a non-
apoptotic pathway.
B. Determine the structural requirements for the breast cancer cell killing effects of

sphingosine and ceramide.
This study may provide insight into the potential of sphingosine and ceramide to

serve as chemotherapeutic agents for breast cancer patients who are estrogen receptor-

negative.

13

III. MATERIALS AND METHODS

Cell culture-The MDA-MB-231 human breast cancer cell line was purchased from
the American Type Culture Collection. Cells were grown in minimum essential medium
(MEM, GIBCOG’) supplemented with 10% fetal bovine serum (FBS). Sphingolipids were
obtained from Matreya, Inc.. Sphingosine was added as a 1:1 complex with bovine serum
albumin (BSA) while Cz-ceramide was dissolved in ethanol and then added directly to the
medium.

Nucleic acid assay-Total nucleic acid concentration was measure as described by Li
et al. (1990). Cells were seeded at 2.5x105/mL and incubated with treatments in 1% FBS
in MEM for certain times. Then the cells were rinsed twice with phosphate-buffered saline
(PBS, pH 7.4) and lysed with 1N NaOH. The absorbance of the clear cell lysate was read
at 260 nm by spectrophotometry to measure total nucleic acid content.

Detection of DNA strand breaks by the TUNEL reaction-In Situ Cell Detection Kit-
Fluorescein was purchased from Boehringer Mannheim. In this method, terminal
deoxynucleotidyl transferase was used for the incorporation of fluorescent labeled
nucleotides into DNA strand breaks in situ (Gavrieli et al., 1992). Cells were grown and
treated in chamber slides (Nunc, Inc.). Air dried cell samples were fixed with freshly
prepared paraforrnaldehyde solution (4% in PBS, pH 7.4) for 30 minutes at room
temperature. Then cells were incubated in permeablization solution (0.1% Triton' X-100,
0.1% sodium citrate) for 2 minutes on ice (4°C). Each sample was incubated with 50 uL

TUNEL reaction mixture in a humidified chamber for 60 minutes at 37°C in the dark. To

14

identify cells with strand breaks, samples were directly analyzed using a fluorescence
microscope.

Qualitative analysis of intemucleosomal DNA fragmentation by agarose gel
electrophoresis-This method was based on the procedure described by Goruppi et al.
(1994) by using the Genomix Kit purchased from Tel-Test “B”, Inc. Briefly, cells were
collected in PBS, and then incubated with a lysing solution and RNase at 65°C for 30
minutes. Chloroforrn and an acidification solution were then added, and the mixture was
separated with a small gel barrier tube. Sample DNA was contained in the upper, aqueous
phase. Sample DNA was centrifuged with a precipitate solution at 10,000 rpm for 10
minutes. The pellet was washed with ionic exchange solution and then precipitated with
ethanol. The pellet was resuspended in distilled water after a final wash with 70 % ethanol.
DNA samples were run on 1.5 % agarose gel at 60 V for 2.5 hours with 0.5x Tris Borate
EDTA (TBE) running buffer. The DNA was stained with ethidium bromide, and the gel
was photographed under UV light.

Flow cytometric analysis of cell cycle and apoptosis-The cells were prepared based
on the procedure described by Telford et al. (1994). Briefly, cells were harvested by
collecting floating cells in the medium and trypsinization of attached cells. Cells were
precipitated and then resuspended in fresh medium (1% fetal bovine serum in minimum
essential medium) to obtain single cells suspension. Cells were fixed with 70% ethanol in
steps until the final concentration of ethanol was more than 50%. Then cells were stored at
4°C overnight. Cells were pelleted and then resuspended in PBS (pH 7.4) containing 0.1%
Triton X-100 and 0.1 mM EDTA (pH 8.0). Cells in single cell suspension were stained by

DNA staining solution (PBS containing 0.1% Triton X-100, 0.1 mM EDTA, 0.05 mg/ml

15

RNase A, 50ug/ml propidium iodide) for at least 1-2 hours before analysis. The cells were
analyzed by a state-of-the-art Becton-Dickinson Vantage flow cytometer.

Mass measurements of sphingoid bases-Cells were washed with 3 mL of PBS (pH
7.4) and harvested in 0.5 mL PBS. Mass measurements of the long-chain bases were
conducted using a sensitive and reproducible HPLC method (Merrill et al., 1988). Briefly,
the unnatural sphingoid bases, Czo-sphinganine and Czo-sphingosine, were used as internal
standards and the long-chain bases were extracted with chloroform and methanol and then
treated with base to remove interfering glycerolipids. After preparation of the o-
phthalaldehyde (OPA) derivatives, the long-chain bases were separated by reverse-phase
HPLC using a C18-column eluted isocratically with methanol:5 mM potassium phosphate
(pH 7.0), 90:10 (v/v).

Measurement of cell protein-Total cell protein was determined by the method
described by Lowry et al. (1951). The cells were collected in PBS (pH 7.4). Then 1.0 mL
of Lowry reagent (25 mg NaZCuEDTA, 2 g Na2C03, 0.4 g NaOH in 100 mL water) was

added to 100 uL of sample or BSA‘ standard. After 10 minutes, the samples were
incubated with 100 uL of phenol-(FOLIN)-reagent:water (1:1) for 30 minutes at room
temperature. Absorbance was determined at 500 nm.

Statistical Analyses-Two-way factorial analysis of variance (ANOVA) was applied
to the data. Also, two-sample t-test (two tailed, assuming equal variance) was used to
identify significant difference between control and treatment groups at specific time points.
The significance level was adjusted by Bonferroni method. Significant effects were at

P<0.05.

16

IV. RESULTS
A. Sphingosine and ceramide differentially affect death of estrogen receptor-
negative MDA-MB-23l human breast cancer cells

Sphingosine and ceramide cause death of human breast cancer cells-To assess
whether sphingosine and ceramide inhibit growth and cause death of human breast cancer
cells, subconfluent MDA-MB-23l cells were cultured with either D-erythro-sphingosine or
Cz-ceramide (a cell permeable, short chain analog of naturally occurring ceramide). At
various times, dead cells were rinsed away and live cells were harvested and nucleic acid
content was measured as an index of cell number (Figure 4). The nucleic acid
concentration in control cultures decreased ~20-25% within the first 3-6 hours (Figure 4A
& 48). Over the subsequent 18-21 hours, the nucleic acid content doubled. The addition
of D-erythro-sphingosine caused concentration- and time-dependent decreases in total
nucleic acid concentration (Figure 4A). D-erythro-sphingosine at 5 uM significantly
reduced nucleic acid concentration within 24 hours compared to the corresponding control.
In comparison, D-erythro-sphingosine at 10 1.1M reduced nucleic acid concentration to
~40% of the corresponding control within 3 hours. Total nucleic acid was still ~40% of
control at 24 hours.

Like sphingosine, Cz-ceramide also inhibited growth and killed MDA-MB-23l cells
in a concentration- and time-dependent manner (Figure 4B). Cz-ceramide at 5 uM
significantly reduced nucleic acid concentration with 18 hours compared to the
corresponding control. At 10 11M, Cz—ceramide reduced the nucleic acid content of cells to

about 75% of the corresponding control within 3 hours. By 24 hours, the nucleic acid

17

175

 

 

 

 

 

 

 

 

 

 

150 — A
.gg 125 - ..
o S
35.; 100 \ /
w -
3: 75— \.:- ,5
g s .
‘- _ I
mg 50 e *
I
25 — '
o 1 1 r r r 1 r
0 3 6 9 12 15 18 21 24
Culture period (hour)
175
150
§§ 125
o 8
a; 100E
0
as 75
(I 0
B 2
r: a 50
25
00 3 6 9 12 15 18 21 24

means with an
corresponding control.

Culture period (hour)

Figure 4. Sphingosine and ceramide inhibit growth and induce death of human
breast cancer cells. Subconfluent MDA-MB-231 cells were cultured with D-erythro-
sphingosine (A) or Cz-ceramide (B) at 0 (O), 2 (O), 5 (I), or 10 (Cl) 11M. At
various times, cells were harvested and nucleic acid content was measured by
spectrophotometry (#260 nm). Each point is the mean i SD (n=3). At each time,
asterisk (*) are significantly different (P < 0.05) than the

18

concentration for cells cultured with 10 M Cz-ceramide was about 30% of the
corresponding control.

Sphingosine and ceramide difl‘erentially afiect death of human breast cancer cells-
To determine whether sphingosine and ceramide kill breast cancer cells by inducing
apoptosis, subconfluent cultures of MDA-MB-231 cells were cultured with cytotoxic
concentrations of the sphingolipids. Then three techniques were utilized to assess DNA
fragmentation, a hallmark of apoptosis. For the TUNEL reaCtion, cells were cultured with
or without the sphingolipid at 5 [AM for 4 hours. Then DNA strand breaks were labeled
with fluorescein using terminal deoxynucleotidyl transferase and cells were examined via
microscopy under both phase contrast (Figure 5, left panels) and under fluorescent light
(Figure 5, right panels). Cells with DNA strand breaks fluoresce and are considered
TUNEL positive. Control cells were TUNEL negative indicating they had intact DNA
(Figure 5A, right panel). Cells cultured with either 5 uM sphingosine (Figure 5B) or C2-
ceramide (Figure 5C) were TUNEL positive (right panels). These studies indicate that
both sphingosine and ceramide cause DNA strand breaks.

To further investigate the mechanism of cell death, agarose gel electrophoresis was
used to detect DNA fragmentation in MDA-MB-231 cells cultured with or without
sphingolipids. Control cultures had intact, high molecular weight DNA which appeared at
the top of the agarose gels (Figure 6A and 6B, lane 2). Like DNA from control cultures,
DNA from cells cultured with Cz-ceramide at 2-10 M for either 6 hours (Figure 6A) or 24
hours (Figure 6B) was not fragmented. In contrast, DNA extracted from cells cultured for

6 hours with 5 uM D-erythro—sphingosine was fragmented and showed a characteristic

19

 

t2; 1"“? 4‘
i“? ‘

    

Figure 5. Sphingosine and ceramide cause DNA strand breaks in human breast
cancer cells. Subconfluent MDA-MB-23l cells were cultured either without
sphingolipid (A), or with 5 uM D—ervthro-sphingosine (B) or 5 uM Cg-ceramide (C)
for 4 hours and DNA strand breaks were labeled with fluorescein in situ via the
TUNEL reaction. For each pair of photographs. the left panel represents cells under
phase contrast and the right shows cells under fluorescent light.

20

 

Figure 6. Cerarnide does not cause intemucleosomal DNA fragmentation in human
breast cancer cells. Subconfluent MDA-MB-23l cells were cultured with various
concentrations of Cz-ceramide for 6 hours (A) and 24 hours (B). Then DNA was
extracted, stained with ethidium bromide, and analyzed for fragmentation by agarose
gel electrophoresis (1.5% agarose gel, 60V). Figure 6A lane 1, DNA 250 base pair
molecular weight markers; lanes 2 and 3, control; lanes 4 and 5, 2 uM C2-ceramide;
lanes 6 and 7, 511M Cg-ceramide; lanes 8 and 9, 10 uM Cz-ceramide. Figure 6B lane
1, DNA 250 base pair molecular weight markers; lanes 2 and 3, control; lanes 4 and
5, 5M Cg-ceramide.

21

ladder pattern indicative of apoptosis (Figure 7-lanes 2 and 3). These results provide strong
evidence that sphingosine kills MDA-MB-23l cells by inducing apoptosis; whereas,
ceramide causes death of these cells via a non-apoptotic pathway.

The difi‘erential efl‘ects of sphingosine and ceramide on human breast cancer cells
were further examined via flow cytometric analysis. For these studies, MDA-MB-23l cells
were cultured with cytotoxic concentrations of the sphingolipids and then DNA was stained
with propidium iodide. Upon flow cytometric analysis, necrotic cells appear to have
reduced DNA content and form a plateau in the hypodiploid pre-Go/Gl region; whereas,
apoptotic cells appear to have reduced DNA content and form a sharp hypodiploid pre-
Go/G; peak. Flow cytometry showed that control cultures had subpopulations of cells in
each phase of the cell cycle (Go/G1, S, Gle) indicative of subconfluent, proliferating cells
(Figures 8A and 9A). In comparison, Cz-ceramide (2, 5, 10 uM) caused concentration-
dependent increases in the number of cells in the Go/Gl phase and in the pro-Go/Gl region as
a plateau indicative of necrosis (Figure 8B-D). In contrast, D-erythro-sphingosine (5, 10,
15 uM) caused concentration-dependent increases in the number of cells in the pre-Go/G1
region as a peak indicative of apoptosis (Figure 9B-D). The percentage of apoptotic cells in
cultures treated with 10 M and 15 M D-erythro-sphingosine were ~65% (Figure 9C) and
~8 5% (Figure 9D), respectively.

Exogenous sphingosine and ceramide increase cellular sphingosine in human
breast cancer cells-To this point, the results indicate that both D—erythro-sphingosine and
Cz-ceramide kill human breast cancer cells; however, only D-erythro-sphingosine kills cells

by inducing apoptosis, while Cz-ceramide kills cells via a non-apoptotic mechanism. The

22

 

Figure 7. Sphingosine induces intemucleosomal DNA fiagmentation in human breast
cancer cells. Subconfluent MDA-MB-231 cells were cultured without sphingolipid
(lane 1) or with 5 uM D-erythro—sphingosine (lanes 2 and 3) for 6 hours. Then DNA
was extracted, stained with ethidium bromide, and analyzed for fragmentation by
agarose gel electrophoresis (1 .5% agarose gel, 60V).

23

>
U!

 

 

 

 

  

 

 

 

 

 

 

I ;__i A "
E I... E
a .. E
3 3
z a. f
3 - r—--* 3
2.. 1
1 a
DNA Content DNA Content

0
5

 

 

 

 

Cell Number
Cell Number

  

 

 

 

   

64

DNA Content ‘ DNA Content

Figure 8. Flow cytometric analysis of human breast cancer cells cultured with
ceramide. Subconfluent MDA-MB-231 cells were cultured with 0 (A), 2 (B), 5 (C),
or 10 uM (D) D— Cz-ceramide for 24 hours. Subsequently, the cells were fixed with
ethanol and DNA was stained with propidium iodide. Pre-Go/Gl cells, bar I; Go/Gl
cells, bar 2; S and G2/M cells, bar 3; total cells, bar 4.

24

>
a:

 

g ‘ - 12

 

Cell Number
Cell Number

 

 

 

 

 

O
U

 

 

 

 

Cell Number
Cell Number

 

 

 

 

   

DNA Content

Figure 9. Flow cytometric analysis of human breast cancer cells cultured with
sphingosine. Subconfluent MDA-MB-231 cells were cultured with 0 (A), 5 (B), 10
(C), or 15 M (D) D-erythro-sphingosine for 6 hours. Subsequently, the cells were
fixed with ethanol and DNA was stained with propidium iodide. Pre-Go/Gl cells,
bar I ; Go/Gl cells, bar 2; S and G2/M cells, bar 3; total cells, bar 4.

25

ability of cells to metabolize ceramide to sphingosine raised the question as to why ceramide
did not also induce apoptosis similar to sphingosine. To address this question, we examined
the effects of exogenously added cytotoxic concentrations of sphingolipids on the cellular
concentration of sphingosine. In control cells, cellular sphingosine remained at ~250
pmol/mg protein over 60 minutes of culture (Figures 10 and 11). In comparison, the
addition of exogenous D—erythro-sphingosine (2, 5, 10 uM) caused concentration- and
time-dependent increases in cellular sphingosine (Figure 10). Upon addition of D-erythro-
sphingosine at 2, 5, and 10 uM, cellular sphingosine increased to ~15, 5.5, and 10 nmol/mg
cell protein, respectively, afier 30 minutes. Afier 60 minutes of culture, the concentrations
were lower indicating that cellular sphingosine was metabolized. Crceramide (2, 5, and 10
M) also caused concentration- and time-dependent increases in cellular sphingosine
(Figure 11); however, the cellular sphingosine concentrations which were achieved were
lower than that for exogenously added D-erythro-sphingosine. \Vith 10 M Cz-ceramide
treatment, cellular sphingosine only increased to ~1 nmol/mg protein afier 30 minutes and

~1.5 nmol/mg protein after 60 minutes (Figure 11).

26

 

.s .3
O N
I
e-
1.1..

 

Cellular sphingosine
(nmollmg protein)
or

 

 

 

 

 

 

4 ..
*
2 - .:
OK i J
0 30 60

Culture period (min)

Figure 10. Exogenous sphingosine increases cellular sphingosine concentration.
Subconfluent MDA-MB-231 cells were cultured with 0 (O), 2 (O), 5 (I), or 10 M
(D) D-erythro—sphingosine. Subsequently, cellular sphingosine concentrations were
measured at 0, 30, and 60 minutes via HPLC. Each point is the mean i SD (n=3). At
each time, means with an asterisk (*) are significantly difi‘erent (P < 0.05) than the
corresponding control.

27

 

N
or

N
l

 

.5
0|

Cellular sphingosine
(nmollmg protein)

0.5

 

 

 

 

 

0 I 1
O 30 60
Culture period (min)

Figure 11. Exogenous ceramide increases cellular sphingosine concentration.
Subconfluent MDA-MB-231 cells were cultured with 0 (O), 2 (O), 5 (I), or 10
M (El) Cz-cerarnide. Subsequently, cellular sphingosine concentrations were
measured at 0, 30, and 60 minutes via HPLC. Each point is the mean :t SD (n=3).
At each time, means with an asterisk (*) are significantly difi‘erent (P < 0.05) than
the corresponding control.

28

B. Structural requirements for sphingosine and ceramide induced death

The results of the first specific aim indicate that the naturally occurring stereoisomer
of sphingosine, D-erythro—sphingosine, kills MDA-MB-231 human breast cancer cells by
inducing apoptosis. To gain additional information about the molecular mechanism, we also
examined the efl‘ects of the three unnatural stereoisomers (D-threo, L-threo and L-erythro)
(Figure 12).

Like D-erythro-sphingosine (Figure 4A), the unnatural sphingosine stereoisomers
inhibited growth and caused death of MDA-MB-231 human breast cancer cells in
concentration- and time-dependent manners (Figure 13 and 14). All of the unnatural
stereoisomers at 5 uM caused significant reductions in cell number within 3 hours of culture
compared to the corresponding controls (Figure 13). L-erythro-sphingosine had the most
potent cytotoxic effect, with an LDso of ~2-3 M at 3 hours (Figure 133). In comparison,
the LDsos of other sphingosine stereoisomers (D-erythrog D-threo-, L-threo) at 3 hours of
culture were ~5 uM (Figures 4 and 13)

The necessity of the 4,5 trans double bond in the sphingoid base backbone for the
cytotoxic efl‘ect of Cz-ceramide was also examined using Cz-dihydroceramide which lacks
the dOuble bond (Figure 15). At 5 uM, Cz-dihydroceramide did not affect grth of MDA-
MB-231 human breast cancer cells (data not shown) and did not cause DNA strand breaks
(Figure 16). These data indicate that the 4,5 trans double bond of the sphingoid base
backbone of ceramide is obligatory for the cytotoxic action of ceramide in human breast

cancer cells.

29

OH OH

WM

D-erythro-sphingosine (28, 3R)

NH;
OH OH
/\/\/\/\/\/\/\
L-erythro-sphingosine (2R, 38)
NH;
OH OH

 
 
 
 

D-threo-ephingosine (2R. 3R)
NH;

OH OH

 

L-threo-sphingosine (28, 38)

NH,

Figure 12. Structures of sphingosine stereoisomers

30

 

§

N
S

3

 

§

 

Relative growth
(percent of control)
8'

 

 

 

 

 

 

 

 

 

 

 

06512132430354248
Culture period (hour)
300
2%
o
a;
is
.0
32
«a
o ..—_———_—- l 1 1%
0 6.12182430364248
Cuinrreperiodmour)
28
C
8
“*5
ii
his

 

 

O 6 12 18 24 30 36 42 48
Cultureperlodfltour)

Figure 13. Sphingosine stereoisomers inhibit growth and induce death of human
breast cancer cells. Subconfluent MDA-MB-23l cells were cultured with D-threo-
sphingosine (A), L-erythro-sphingosine (B), and L-threo-sphingosine (C) at 0 (O), 2
(O), 5 (I), or 10 (D) M. At various times, cells were harvested and nucleic acid
content was measured by spectrophotometry (7t=260 nm). Each point is the mean :1:
SD (n=3). At each time, means with an asterisk (‘) are significantly difl‘erent (P <
0.05) than the corresponding control.

31

2 “M D-Ilrreo-sphingoaine

    
  

Control
‘a ;7,-< ,_r«‘ ,4 .

  

 

5 uM D-Ilireo—Iphingosine 10 pH D-IIrreo—sphingnsinr

 

 

 

a.
. .. .'>‘- w.
.12.:

r: “awhmrupn - _., ~-

 

B Control 2 HM L-corlllro-rphingosine

 

5 uM L-erythro-sphingnsine l0 11M L-eryllrrmsphingasme
.3 Ilfizfimmmfi'zuvi'
-I'-' i ‘II‘
I: i f
i i
s1 , . t
., 4:.
z .,; i 'r
i ’ 1 i 1‘

 

 

2 uM Mirna-sphingosine

 

5 PM L-Ihrro-sphingosinc 10 uM L-IIlreo—sphingosinc

 

 

I ‘7 I TCSTJZ. maximums-rd

 

 

Figure 14. Sphingosine stereoisomers inhibit growth and induce death of human
breast cancer cells (photographs). Subconfluent MDA-MB-23l cells were cultured
with D-threo—sphingosine (A), L-erjvthro-sphingosine (B), and L-threo-sphingosine
(C) at 0, 2, 5, or 10 (1M for 24 hours.

La.)
N

H

\
Cz-ceramide H
O
H
Cz-dihydroceramide v"
0

Figure 15. Structures of Cz-ceramide and Cz-dihydroceramide

33

 

Figure 16. Cz-dihydroceramide does not cause DNA strand breaks in human breast
cancer cells. Subconfluent MDA-MB-231 cells were cultured with C2-
dihydroceramide for 4 hours and DNA strand breaks were labeled with fluorescein in
situ via the TUNEL reaction. The left panel represents cells under phase contrast and
the right shows cells under fluorescent light.

34

V. DISCUSSION

Breast cancer is one of the most prevalent malignant cancers encountered in the
western world (Marshall, 1993) and currently a great deal of interest is focused on ways to
prevent and treat this disease. Combinations of surgery, radiation therapy, hormone
therapy and chemotherapy are used to treat breast cancer patients. Although chemotherapy
helps to improve survival of patients, the prognoses are far from satisfying, especially for
advanced breast cancer patients who are endocrine resistant, estrogen receptor-negative, or
who have life-threatening metastasis. The present study has utilized MDA-MB-23l human
breast cancer cells as a model system to evaluate the chemotherapeutic potential of several
sphingolipids. MDA-MB-23l cells were derived from a breast adenocarcinoma (Cailleau
et al., 1974). These cells are estrogen receptor-negative and resistant to a variety of drugs.
For example, they are insensitive to antiestrogens (e.g. estradiol, tamoxifen,
benzothiophene) (Thompson et al., 1988; Lippman et al., 1989), DNA-damaging agents
(e.g. preactivated merocyanine 540, merodantoin) (Sharma et al., 1995), anti-polyarnine
(e.g. alpha-difluoromethyl-omithin) (Manni et al., 1992), retinoic acid (Yu et al., 1996)
and some apoptotic inducing agents (Nakshatri et al., 1997). Thus, the MDA-MB-231
cell-line provides an excellent model to study chemotherapy in vitro for those advanced
breast cancer patients who have negative estrogen receptor status. This study demonstrates
that both D-erythro-sphingosine and ceramide inhibit growth and cause death of MDA-
MB-231 human breast cancer cells. Although a number of other studies have shown that
both sphingosine and ceramide can kill cancer cells (Ohta et al., 1995; Jayadev et al., 1995;
Jarvis et al., 1996; Cal et al., 1997), we believe our study is the first to show that these

sphingolipids are cytotoxic for estrogen receptor-negative human breast cancer cells.

35

There are several possible modes by which D-erythro-sphingosine and ceramide
might kill human breast cancer cells. For example, these sphingolipids might induce
apoptosis, a highly-regulated form of physiological cell death which occurs only in targeted
cells. Cellular shrinkage, chromatin condensation, activation of an endogenous
endonuclease which causes extensive DNA fragmentation, and formation of apoptic bodies
are key events in the process of apoptosis (Kerr and Harmon; 1991, Wyllie et al., 1992).
Alternatively, sphingosine and ceramide could cause death of human breast cancer cells via
a non-apoptotic mechanism such as necrosis which is highly unregulated and not targeted
to specific cells. In the present study, though ceramide did induce DNA strand breaks
indicative of DNA damage, it did not cause DNA fragmentation that was detectable via
agarose gel electrophoresis. Moreover, flow cytometric analysis revealed the presence of a

subpopulation of cells in the pre-GolGl region in the form of a plateau indicative of

necrosis. In contrast, D-erythro-sphingosine caused DNA strand breaks, DNA

fragmentation on agarose gel electrophoresis, and a pre-Go/Gl peak indicative of

apoptosis.

Currently, many successful chemotherapeutic treatments involve induction of
apoptosis in the target cancer cells (Kaufmann, 1989; Walker et al., 1991; Huschtscha et
al., 1996). Apoptotic inducing agents are appealing in cancer chemotherapy because they
have two significant advantages: 1) Apoptotic agents eliminate cancer cells from the
human body instead of just inhibiting the growth of cancer cells. Clinically, this is
considered a “true” cure; and 2) Apoptotic agents have low side effects in cancer patients
because they selectively kill cancer cells without having cytotoxic effects on neighboring

normal tissue. The finding that ceramide kills breast cancer cells via a non-apoptotic

36

mechanism suggests that it may not be very effective as a chemotherapeutic agent because
of its potential lack of specificity of action. In contrast, the ability of D-erythro-
sphingosine to induce apoptosis in MDA-MB-23l cells strongly suggests that this
compound has potential as a chemotherapeutic agent either alone or in a combination
regimen. In chemotherapy, combination regimens (e.g. CMFVP, cytoxan, methotrexate,
fluorouracil, vincristine, prednisone) are more commonly used than single cytotoxic agents
(e.g. doxorubicin) because it is often possible to lower the dosage of. the individual agents
(and minimize side effects) while still generating higher response rates (Clavel and
Catimel, 1993).

Our discovery that D-erythro-sphingosine kills human breast cancer cells by inducing
apoptosis; whereas, ceramide kills the cells via a non-apoptotic pathway was puzzling in
light of the ability of cells to metabolize ceramide to sphingosine. Our subsequent results
suggest that this may be due to the inability of ceramide to raise cellular sphingosine high

enough to induce apoptosis. Cz-ceramide at 10 11M only increased the cellular sphingosine

concentration to about 1 nmol/mg protein; whereas, exogenous D-erythro-sphingosine at 5-
10 M (which was apoptotic) increased the cellular sphingosine concentration to about 5-
10 nmol/mg protein. These findings suggest that the critical concentration of cellular
sphingosine necessary to induce apoptosis is 5-10 nmol/mg protein.

The mechanism by which sphingosine induces apoptosis of human breast cancer
cells is not clear. Apoptosis could be due to a direct action of sphingosine on a variety of
systems they are known to affect (Merrill et al., 1997). For example, sphingosine has been
shown to inhibit protein kinase C (Hannun et al., 1986), phosphatidic-acid

phosphohydrolase (Lavie and Liscovitch, 1990; Jamal et al., 1991; Mullmann et al., 1991),

37

and the Na+/K+-ATPase (Oishi et al., 1990) and to activate the epidermal growth factor
receptor kinase (Faucher et al., 1988; Wedegaertner and Gill, 1989) and other sphingosine-
specific protein kinases (Pushkareva et al., 1992). In addition, D-erythro-sphingosine has
been shown to induce apoptosis via suppression of bcl-2 in HL-60 cells (Sakakura et al.,

1996) and via suppression of bcl-xL in androgen-independent human prostatic carcinoma

DU-l45 cells (Shirahama et al., 1997). A recent study has demonstrated that MDA-MB-
231 human breast cancer cells have constitutively active nuclear factor-KB (NF - KB) which
protects the cells from apoptotic agents (Nakshatri et' al., 1997). Therefore, sphingosine
probably induces apoptosis by overcoming or bypassing this protective mechanism.

Our findings that both D-erythro-sphingosine and Cz-ceramide kill estrogen

receptor-negative breast cancer cells raises the possibility that close structural analogs may

be even more effective as chemotherapeutic agents. Our observation that C2-

dihyroceramide does not induce DNA strand breaks or kill cells strongly indicates that the
4-5 trans double bond of the sphingoid base backbone is necessary for the cytotoxic effect
of ceramide. Further, our discovery that the unnatural L-threo and D—threo stereoisomers
of sphingosine are equally as potent as D-erythro-sphingosine and that L-erythro-
sphingosine is more potent indicate that these compounds may also find utility as breast
cancer chemotherapeutic agents. The more potent effect of L-erythro-sphingosine may be
due to poorer cellular metabolism compared to the other sphingosine stereoisomers
(Schroeder et al., 1994).

The results of these studies raise several important new questions: Does sphingosine
bypass a key protection step to cause the cells to undergo apoptosis? What are the

downstream targets of sphingosine in the cell death signal transduction pathway? How is

38

the endonuclease activated during this process? The answers to these questions may lead

[0 new means [0 treat estrogen receptor-negative bl'CflSI cancer.

39

VI. SUMMARY AND FUTURE RESEARCH

The naturally occurring form of sphingosine, D-erythro, caused NHDA-MB-23l
estrogen receptor-negative human breast cancer cells to undergo apoptosis with an LDso of
~5-10 pM. The concentration of cellular sphingosine necessary to induce apoptosis is ~5-
10 nmol/mg protein. Like D-erythro-sphingosine, all of the synthetic, unnatural
stereoisomers, L-erythro, L-threo, and D-threo, also caused death of MDA-MB-23l human
breast cancer cells, though the mechanism of cell death is not yet known. Among the four
stereoisomers, L-erythro-sphingosine was the most potent with an LDso of ~2-3 11M;
whereas, the potencies of L-threo- and D-threo-sphingosine were similar to that of D-
erythro-sphingosine. Cz-ceramide also induced death of MDA-MB-231 human breast
cancer cells, but via a non-apoptotic pathway. In contrast, Cz-dihydroccrarnide did not
cause DNA strand breaks or cell death in breast cancer cells, indicating that the 4,5 trans
double bond of the sphingoid base backbone is critical for ceramide to cause cell death.

The ability of D-erythro-sphingosine to induce apoptosis in estrogen receptor-
negative human breast cancer cells suggests that this molecule has potential as a
chemotherapeutic agent for postmenopausal women who are estrogen receptor-negative.
Moreover, the finding that L-erythro-sphingosine is even more potent than D-erythro-
sphingosine at inducing cell death suggests that L-erythro-sphingosine may be even more
effective as a chemotherapeutic agent. Sphingosine stereoisomers could find roles in
chemotherapy used either alone or in combination with other drugs.

Additional studies are necessary to further evaluate the potential of sphingosine

stereoisomers as chemotherapeutic agents. Future studies should: 1) Determine whether

40

the unnatural stereoisomers (L—erythro, L-threo, and D-threo) cause death by inducing
apoptosis; 2) Determine the molecular mechanism by which sphingosine stereoisomers
induce apoptosis; and 3) Determine the effect of sphingosine stereoisomers on normal
human breast epithelial cells. Studies which take the approach outlined above have the
potential not only to further evaluate the potential of sphingosine stereoisomers as
chemotherapeutic agents, but also to identify targets which are amenable to therapy and to

assess the chemopreventive potential of sphingosine stereoisOmers.

41

VII. REFERENCES

Ahmann, D. L., Bisel, H. F., Eagan, R. T., Edmonson, J. H., and Hahn, R. G. (1974).
Controlled evaluation of adriamycin in patients with disseminated breast cancer. Cancer
Chemother. Rep. 58: 877-882.

Bastarrachea, J., Hortobagyi, G. N., Smith, T. L., Kau, S. W., and Buzdar, A.U. (1994).
Obesity as an adverse prognostic factor for patients receiving adjuvant chemotherapy for
breast cancer. Ann. Intern. Med. 120: 18-25.

Bell, R. M., Hannun, Y. A., and Merrill, A. H., Jr. (Eds) (1993). Advances in Lipids
Research: Sphingolipids and Their metabolites, Vols. 25-26, Academic Press, Orlando,
FL.

Bielawska, A., Linardic, C. M., and Hannun, Y. A. (1992). Cerarnide-mediated biology.
Determination of structural and stereospecific requirements through the use of N-acyl-
phenylaminoalcohol analogs. J. Biol. Chem. 267: 18493-18497.

Cai, Z., Bettaieb, A., Mahdani, N. E., Legres, L. G., Stancou, R., Masliah, J., and Chouaib,
S. (1997). Alteration of the sphingomyelin/ceramide pathway is associated with resistance
of human breast carcinoma MCF-7 cells to TNF-a-mediated cytotoxicity. J. Biol. Chem.
272: 6918-6926.

Cailleau, R., Young, R., Olive, M., and Reeves, W. J., Jr. (1974). Breast tumor cell lines
from pleural effusion. J. Natl. Cancer Inst. 53: 661-674.

Canellos, G. P., DeVita, V. T., Gold, G. L., Chabner, B. A., Schein, P. 8., and Young, R.
C. (1976). Combination chemotherapy for advanced breast cancer: response and effect on
survival. Ann. Int. Med. 84: 389-392.

Claus, E., Risch, N ., and Thompson, W. D. (1991). Genetic analysis of breast cancer in the
cancer and steroid hormone study. Am. J. Hum. Genet. 48: 232-242.

Clavel, M., and Catimel, G. (1993). Breast cancer: Chemotherapy in the treatment of
advanced disease. Eur. J. Cancer 29A: 598-604.

Cleary, M. P. and Maihle, N. J; (1997). The role of body mass index in the relative risk of
developing premenopausal versus postmenopausal breast cancer. Proc. Soc. Exp. Biol.
Med. 216: 28-43.

de la Rochefordiere, A., Asselain, B., Campana, F., and Scholl, S.M., Fenton, J., Vilcoq,
J.R., Durand, J .C., Pouillart, P., Magdelenat, H., and Fourquet, A. (1993). Age as
prognostic factor in premenopausal breast carcinoma. Lancet. 341: 1039-1043.

42

Dressler, K. A., Mathias, S., and Kolesnick, R. N. (1992). TNF-or activates the
sphingomyelin signal transduction pathway in a cell-free system. Science 255: 1715-1718.

Eddy, D. M. (1992). High-dose chemotherapy with autologous bone marrow
transplantation for the treatment of metastatic breast cancer. J. Clin. Oncol. 10: 657-670.

Faucher, M., Girones, N., Hannun, Y. A., Bell, R. M., and Davis, R. J. (1988). Regulation
of the epidermal growth factor receptor phospharylation state by sphingosine in A431
human epidermoid carcinoma cells. J. Biol. Chem. 263: 5319-5327.

Fisher, B., Costantino, J., Redmond, C., and Poisson, R., Bowman, D., Couture, J.,
Dimitrov, N. V., Wolmark, N., Wickerham, D. L., and Fisher, E. R. (1989). A randomized
clinical trial evaluating tamoxifen in the treatment of patients with node-negative breast
cancer who have estrogen-receptor-positive tumors. New England J. Med. 320: 479-484.

Friedenreich, C. M., Howe, G. R., Miller, A. B., and Jain, M. G. (1993). A cohort study of
alcohol consumption and risk of breast cancer. Am. J. Epidemiol. 137: 512-520.

Gavrieli, Y., Sherman, Y., and Ben-Sasson, S. A. (1992). Identification of programmed cell
death in situ via specific labeling of nuclear DNA fragmentation. J. Cell Biol. 119: 493-
501.

Gill, Z. P., Perks, C. M., Newcomb, P. V., and Holly, J. M. (1997). Insulin-like growth
factor-binding protein (IGFBP-3) predisposes breast cancer cells to programmed cell death
in a non-IGF-dependent manner. J. Biol. Chem. 272: 25602-25607.

Goruppi, S., Gustinchin, S., Brancolini, C., Lee, W. M., and Schneider, C. (1994).
Dissection of c-myc domains involved in S phase induction of NII-I3T3 fibroblasts.
Oncogene 9: 1537-1544.

Greenspan, E. M. (1996). Breast cancer chemotherapy after fifty years: More cures and less
mutilation. Mount Sinai J. Med. 65: 250-255.

Hannun, Y.A., Loomis, C.R., Menill, AH. and Bell, RM. (1986). Sphingosine inhibition
of protein kinase C activity and of phorbol dibutyrate binding in vitro and in human
platelets. J. Biol. Chem. 261 : 12604-12609.

Huschtscha, L. I., Bartier, W. A., Ross, C. E. A., Tattersall, M. H. N. (1996).
Characteristics of cancer cell death after exposure to cytotoxic drugs in vitro. Br. J. Cancer
73: 54-60.

Jamal, 2., Martin, A., Munoz, A. G., and Brindley, D. N. (1991). Plasma membrane

fraction from rat liver contains a phosphatidate phosphohydrolase distinct from that in the
endoplasmic reticulum and cytosol. J. Biol. Chem. 266: 2988-2996.

43

Jarvis, W. D., Fomari, F. A., Traylor, R. S., Martin, H. A., Krammer, L. B., Erukulla, R.
K., Bittman, R., and Grant, 8. (1996). Induction of apoptosis and potentiation of ceramide-

mediated cytotoxicity by sphingoid bases in human myeloid leukemia cells. J. Biol. Chem.
271: 8275-8284.

Jarvis, W. D., Kolesnick, R. N ., Fomari, F. A., Traylor, R. S., Gewirtz, D. A., and Grant, 8.
(1994). Induction of apoptotic DNA damage and cell death by activation of the
sphingomyelin pathway. Proc. Natl. Acad. Sci. U. S. A. 91:73-77.

Jayadev, S., Liu, B., Bielawska, A. E., Lee, J. Y., Nazaire, F., Pushkareva, M. Y, Obeid, L.
M., and Hannun, Y.A. (1995). Role for ceramide in cell cycle arrest. J. Biol. Chem. 270:
2047-2052.

Jones, S. E., Durie, B. G., and Salmon, S. E. (1975). Combination chemotherapy with
adriamycin and cyclophosphamide for advanced breast cancer. Cancer 36: 90-97.

Karasavvas, N., Erukulla, R. K., Bittman, R., Lockshin, R., and Zakeri, Z. (1996).
Stereospecific induction of apoptosis in U937 cells by N-octanoyl-sphingosine
stereoisomers and N-octyl-sphingosine. The ceramide amide group is not required for
apoptosis. Eur. J. Biochem. 236: 729-737.

Kaufmann, S. (1989). Induction of endonucleolytic DNA cleavage in human acute
myelogenous leukemia cells by etoposide, camptothecin, and other cytotoxic anticancer
drugs: A cautionary note. Cancer Res. 49: 5870-.

Kerr, J. F. R., Harmon, B. V. (1991). Definition and incidence of apoptosis: An historical
perspective. In Tomei L. D. and Cope F. O. (eds) Apoptosis: The molecular Basis of cell
death. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N. Y., pp. 5.

Kim, M. Y., Linardic, C., Obeid, L., and Hannun, Y. A. (1991). Identification of
sphingomyelin turnover as an effector mechanism for the action of TNF-a and y-
interferon. Specific role in cell differentiation. J. Biol. Chem. 266: 484-489.

Kimmick, G. 6., Bell, R. A., and Bostick, R. M. (1997). Vitamin E and breast cancer: a
review. Nutr. Cancer 27: 109-117.

Klurfeld, D. M., Lloyd, L. M., Welch, C. B., Davis, M. J., Tulp, O. L., and Kritchevsky, D.
(1991). Reduction of enhanced mammary carcinogenesis in LA/N-cp (corpulent) rats by
energy restriction. Proc. Soc. Exp. Biol. Med. 196: 381-384.

Lavie, Y., and Liscovitch, M. (1990). Activation of phospholipase D by sphingoid bases in
NO 108-15 neural-derived cells. Biochem. Biophys. Res. Comm. 167: 607-613.

44

Legha, S. S., Buzdar, A. U., Smith, T. L., Hortobagyi, G. N., Swenerton, K. D.,
Blumenschein, G. R., Gehan, E. A., Bodey, G. P. and Freireich, E. J. (1979). Complete

remissions in metastatic breast cancer treated with combination drug therapy. Ann. Int.
Med. 91: 847-852.

Li, I.-C., Chang, C. C., and Trosko, J. E. (1990). Thymidylate synthetase gene as a
quantitative mutation marker in Chinese hamster cells. Mutation Res. 243:233-239.

Lippman, M. E., and Dickson, R. B. (1989). Mechanism of growth control in normal and
malignant breast epithelium. Rec. Prog. Horm. Res. 45: 383-439.

Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951). Protein
measurement with folin-phenol reagent. J. Biol. Chem. 193: 265-275.

Lynch, H., Watson, P., Conway, T., and Lynch, J. (1992). Natural history and age at onset
of hereditary breast cancer. Cancer Res. 69: 4104-1407..

Manni, A., Badger, B., Martel, J ., Demers, L. (1992). Role of polyamine in the growth of
hormone-responsive and -resistant human breast cancer cells in hude mice. Cancer Lett.
66: 1-9.

Marshall, B. (1993). Search for a killer: Focus shifts from fat to hormones. Science 259:
618-621.

Mathias, S., Dressler, K. A., and Kolesnick, R. N. (1991). Characterization of a ceramide-
activated protein kinase: stimulation by TNF—a. Proc. Natl. Acad. Sci. U. S. A. 88: 10009-
10013.

Merrill, A. H., Jr., Schmelz, E. M., Dillehay, D. L., Spiegel, S., Shayman, J. A., Schroeder,
J. J., Riley, R. T., Voss, K. A., and Wang, E. (1997). Sphingolipids--the enigmatic lipid
class: biochemistry, physiology. and pathophysiology. Toxicol. Appl. Pharmacol. 142: 208-
225.

Merrill, A. H., Jr., Wang, E., Mullins, R. E., Jamison, W. C., Nimkar, S., and Liotta, D. C.
(1988). Quantitation of free sphingosine in liver by high-perforrnance liquid
chromatography. Analy. Biochem. 171 : 373-381.

Miki, Y., Swensen, J., Shattuck-Eiden, D., Futreal, P.A., Harshman, K., Tavtigian, S., Liu,
Q., Cochran, C., Bennett, L. M., and Ding, W. (1994). A strong candidate for the breast
and ovarian cancer susceptibility gene BRCAl. Science 266: 66-71.

Mullman, T. J., Siege], M. I., Egan, R. W., and Billah, M. M. (1991). Sphingosine inhibits

phosphatidate phosphohydrolase in human neutrophils by a protein kinase C-independent
mechanism. J. Biol. Chem. 266: 2013-2016.

45

Nakshatri, H., Bhat-Nakshatri, P., Martin, D. A., Goulet, R. J., Jr., Sledge, G. W., Jr.
(1997). Constitutive activation of NF—KB during progression of breast cancer to hormone-
independent growth. Mol. Cell. Biol. 17: 3629-3639.

Nakamura, S., Kozutsumi, Y., Sun, Y., Miyake, Y., Fujita, T., and Kawasaki, T. (1996).
Dual roles of sphingolipids in signaling of the escape from and onset of apoptosis in a
mouse cytotoxic T-cell line, CI‘LL-2. J. Biol. Chem. 271: 1255-1257.

Nikolova—Karakashian, M., Morgan, E. T., Alexander, C., Liotta, D. C. and Merrill, A. H.
(1997). Bimodel regulation of ceramidase by interleukin-18. Implications for the regulation
of cytochrome p450 2C1 1. J. Biol. Chem. 272: 18718-18724.

Obeid, L. M., Linardic, C. M., Karolak, L. A., and Hannun, Y. A. (1993). Programmed cell
death induced by ceramide. Science 259:1769-1771.

Ohta, H., Sweeney, E. A., Masamune, A., Yatomi, Y., Hakomori, S., and Igarashi, Y.
(1995). Induction of apoptosis by sphinosine in human leukemic I-IL-60 cells: A possible
endogenous modulator of apoptotic DNA fragmentation occurring during phorbol ester-
induction differentiation. Cancer Res. 55: 691-697.

Ohta, H., Yatorrri, Y., Sweeney, E. A., Hakomori, S., and Igarashi, Y. (1994). A possible
role of sphingosine in induction of apoptosis by tumor necrosis factor-alpha in human
neutrophils. FEBS Lett. 355: 267-270.

Oishi, K., Zheng, B., and Kuo, J. F. (1990). Inhibition of Na, K ATPase and sodium pump
by protein kinase C regulators sphingosine, lysophophatidylcboline, and oleic acid. J. Biol.
Chem. 265: 70-75.

Okazaki, T., Bell, R.M., and Hannun, Y. A. (1989). Sphingomyelin turnover induced by
vitamin D3 in HL-60 cells. Role in cell differentiation. J. Biol. Chem. 264: 19076-19080.

Okazaki, T., Bielawska, A., Domae, N., Bell, RM, and Hannun, Y.A. (1994).
Characteristics and partial purification of a novel cytosolic, magnesium-independent,
neutral sphingomyelinase activated in the early signal transduction of la, 25-
dihydroxyvitamin D3-induced HL-60 cell differentiation. J. Biol. Chem. 269: 4070-4077.

Orel, S. G., Schnall, M. D., LiVolsi, V. A., and Troupin, RH. (1994). Suspicious breast
lesions: MR imaging with radiologic-pathologic correlation. Radiology 190: 485-493.

Osteen, R. T., Henderson, I. C., Costanza, M. E., Wood, W. C., and Harris, J. R. (1986).
Cancer Manual (Cady, B. ed.), Seventh edition, pp 151-157.

46

Pushkareva, M. Y., Wasiuddin, A. K., Alessenko, A. V., Sahyoun, N., and Hannun, Y. A.
(1992). Sphingosine activation of protein kinases in Jurkat T cells: in vitro phosphorylation

of endogenous protein substrates and specificity of action. J. Biol. Chem. 267: 15246-
15251.

Sakakura, C., Sweeney, E. A., Shirahama, T., Hakomori, S., and Igarashi, Y. (1996).
Suppression of Bel-2 gene expression by sphingosine in the apoptosis of human leukemic
HL-60 cells during phorbol ester-induced terminal differentiation. FEBS Lett. 379: 177-
180.

Sankaranarayanan, R., and Mathew, B. (1996). Retinoids as cancer-preventive agent. IARC
Sci. Publ. 139: 47-59.

Sawai, H., Okazaki, T., Yamamoto, H., Okano, H., Takeda, Y., Tashima, M., Sawada, H.,
Okuma, M., Ishikura, H., and Umehara, H. (1995). Requirement of AP-l for ceramide-
induced apoptosis in human leukemia HL-60 cells. J. Biol. Chem. 270: 27362-27331.

Schroeder J. J., Crane, H. M., Xia, J., Liotta, D. C., and Merrill, A. H., Jr.. (1994).
Disruption of sphingolipid metabolism and stimulation of DNA synthesis by Fumonisin
Br. J. Biol. Chem. 269: 3475-3481.

Shanna, R., Arnold, L., Gulliya, K. S. (1995). Correlation between DNA topoisomerase II
activity and cytotoxicity in pMC540 and merodantoin sensitive and resistant human breast
cancer cells. Anticancer Res. 15: 295-304.

Shirahama, T., Sakakura, C., Sweeney, E. A., Ozawa, M., Takemoto, M., Nishiyama, K.,
Ohi, Y., and Igarashi, Y. (1997). Sphingosine induces apoptosis in androgen-independent

human prostatic carcinoma DU-145 cells by suppression of bcl-XL gene expression. FEBS
Lett. 407:97-100.

Smith, E. R. and Merrill, A. H., Jr. (1995). Differential Roles of De Novo Sphingolipid
Biosynthesis and turnover in the “burst” of free sphingosine and sphinganine, and their 1-

phosphates and N-acyl-derivatives, that occurs upon changing the medium of cells in
culture. J. Biol. Chem. 270: 1-10.

Stoll, B. A. (1996). Can supplementary dietary fiber suppress breast cancer growth. Br. J.
Cancer 73: 557-559.

Sweeney, E. A., Sakakura, C., Shirahama, T., Masamune, A., Ohta, H., Hakomori, S., and
Igarashi, Y. (1996). Sphingosine and its methylated derivative N, N-dimethylsphingosine
(DMS) induce apoptosis in a variety of human cancer cell lines. Int. J. Cancer 66: 358-366.

Telford, W. G., King, L. E., and Fraker, P. J. (1994). Rapid quantitation of apoptosis in

pure and heterogeneous cell populations using flow cytometry. J. Immuno. Methods 172: 1-
16.

47

Thompson, E. W., Reich, R., Shima, T. B., Albini, A., Graf, G., Martin, G. R., Dickson, R.
B., Lippman, M. E. (1988). Differential regulation of growth and invasiveness of MCF-7
breast cancer cells by antiestrogens. Cancer Res. 48: 6764-6768.

Verhoeven, D. T., Assen, N ., Goldbohm, R. A., Dorant, E., van-’t-Veer, P., Sturmans, F.,
Hermus, KL, and vanden-Brandt, RA. (1997). Vitamins C and E, retinol, beta-carotene

and dietary fiber in relation to breast cancer risk: a prospective cohort study. Br. J. Cancer
75: 149-155.

Walker, P. R., Smith, C., Youdale, T., Leblanc, J., Whitfield, J. F., Sikorska, M. (1991).
Topoisomerase II-reactive chemotherapeutic drugs induce apoptosis in thymocytes. Cancer
Res. 51: 1078-1085.

Wedegaertner, P. B., and Gill, G. N. (1989). Activation of the purified protein tyrosine
domain of the epidermal growth factor receptor. J. Biol. Chem. 264: 11346-11353.

Wooster, R., Neuhausen S. L., Mangion, J., Quirk, Y., Ford, D., Collins, N., Nguyen,
K., Seal, S., Tran, T., and Averill, D. (1994). Localization of a breast cancer
susceptibility gene, BRCA2, to chromosome 13q12-13. Science 265: 2088-2090.

WHO (1997)The World Health Report Geneva: World Health Organization
Wright, K. (1990). Breast cancer: Two steps closer to understanding. Science 191: 1659.

Wyllie, A. H., Kerr, J. F. R., Currie, A. R. (1992). Cell death: The significance of
apoptosis. Int. Rev. Cytol. 68: 251.

Wynder, E. L., Cohen, L. A., Muscat, J. E., Winters, B., Dwyer, J. T., and Blackburn, G.
(1997). Breast cancer: weighing the evidence for a promoting role of dietary fat. J. Natl.
Cancer Inst. 89: 766-775.

Yu, L., Shao, Z., Cai, S. (1996). Growth inhibitory effect of retinoic acid in human breast

cancer cells correlates with retinoic acid receptor (1 expression. Chinese Journal of
Oncology. 18: 429-432.

Zhang, C. and Schroeder, J. J. (1998). Sphingosine and ceramide differentially affect death
of human breast cancer cells. FASEB J. 12: A657.

Zhang, H., Desai, N. N., Olivera, A., Seki, T., Brooker, G., and Spiegel, S. (1991).

Sphingosine-l-phosphate, a novel lipid, involved in cellular proliferation. J. Cell Biol. 114:
155-167.

48

"‘iIIIiIIIIIIIIIII“