~._._...:._:..H....:x . .....:..
..: ..._.._;: f... . .
...:....:...._..
I; 1%,:
.:....‘.. ._.._‘...;
v_..

 

~ .u...-......._‘ . a .

 

 

THE“

MICIH IGAN STATE

III I III

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

0910 1746

This is to certify that the

dissertation entitled

Proliferation, Morphologic and Functional Characteristics
of Alkylnitrosourea Induced Astrocytoma Cells

and Changes Induced by Nerve Growth Factor

presented by

William Russell Hare, Jr.

has been accepted towards fulfillment
of the requirements for

PhD Pathology/Toxicology

degree in

 

 

 

 

Major professor

Date 0C7: ‘30; /?9/

 

MS U is an Affirmative Action/Equal Opportunity Institution 0-12771

 

 

 

' LIBRARY
Michigan State
University

 

I
1

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU Is An Affirmative Action/Equal Opportunity Institution
c:\clrc\datedue.pm&p.t

Proliferation, Morphologic and Functional Characteristics
of Alkylnitrosourea-Induced Astrocytoma Cells
and Changes Induced by Nerve Growth Factor

By

William Russell Hare, Jr.

A Dissertation

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

DOCTOR OF PHILOSOPHY

Department of Pathology
Institute for Environmental Toxicology

1991

ABSTRACT

Proliferation, Morphologic and Functional Characteristics
of Alkylnitrosourea-Induced Astrocytoma Cells
and Changes Inducible by Nerve Growth Factor

By

William Russell Hare, Jr.

Nerve growth factor (NGF) is a potent differentiation promoter of
astrocytes. In response to treatment with NGF, astrocyte morphology changes
and cells take on the end stage appearance of protoplasmic and fibrous
astrocytes. Anaplastic astrocytoma cells also have been reported to respond
with changes suggesting promotion of morphologic differentiation. In
addition, astrocytoma cells have been reported to respond to NGF treatment
by a reduction in growth. NGF has been shown to reduce the size of tumors
in vivo and to control the growth of tumor cells in vitro . Because of these
reverse transforming characteristics, NGF has been suggested as a treatment
for astrocytomas. However, aside from these observations on tumors and
tumor cells, little is known of the molecular mechanisms involved in NGF's
action. There also is very little information available, outside of morphologic
studies, on the effects of NGF on normal astrocytes. Therefore, the purpose of
this dissertation was to examine and compare the action of NGF on neonatal
astrocytes and anaplastic astrocytoma cells in vitro, and with respect to
morphology, function and proliferation. Morphology was studied by
determining the effects of NGF on cellular anatomy, characterized by

formation of cytoplasmic processes and intermediate filaments. These areas

were investigated by using light-microscopy as well as immunofluoresence.
Function was studied by determining the effects of NGF on the uptake of
glutamate (GLU) and y-aminobutyric acid (GABA), the major excitatory and
inhibitory amino acid neurotransmitters, respectively. This area was
investigated by use of 3H-GLU and 3H-GABA and scintillation analysis.
Proliferation was studied by determining NGF induced changes in the cell
cycle of these two cell populations. This was investigated by use of acridine
orange flow cytometry.

NGF treatment of cell cultures in the presence of glutamate resulted in
morphologic differentiation of astrocytoma cells from monotonous spindle
shaped cells to cells resembling protoplasmic and fibrous astrocytes. NGF
increased the intensity of glial fibrillary acidic protein with perinuclear,
nuclear and nucleolar distributions. Amino acid neurotransmitter uptake
was not adversely affected by NGF treatment. NGF was also found to
significantly affect proliferation potential of astrocytoma cells by inducing a

quiescent non-cycling subpopulation of cells.

Copyright by
William Russell Hare, Jr.
1991

 

This dissertation is dedicated to the memory of my wife Jeannette C. Hare.

 

 

 

ACKNOWLEDGMENTS

There are many people who come to mind when thinking of
acknowledgements. A famous 12th century scholar and physician once said,
"honor your teachers because they have brought you into the world of the
future." All of my teachers are gratefully acknowledged for the outstanding
job they have done. If it weren't for them, none of this would have been
possible.

More directly related to the degree at hand are thanks to Dr. Bert
Koestner, Dr. Bob Leader and Dr. Ian Krehbiel who were all instrumental in
making this opportunity available to me. A great deal of thanks also go to Dr.
Chuck Sweeley for allowing me to work in his laboratory and for giving me
scientific support and encouragement, as well as to the Department of
Biochemistry for providing research space and use of their facilities .

Thanks go to Dr. Kathy Lovell and Dr. Keiji Marushige for serving on
my committee and putting up with all the frustrations of a not so typical
graduate student. Thanks, as well, to Dr. Larry Fisher, who had the foresight
to know that I would need additional time and funding to prepare
manuscripts for publication and to complete the writing of this dissertation.
His help with supplemental funding through Dow Chemical Corp. and help
from Dr. Jan Krehbiel and the College of Veterinary Medicine were extremely

vi

 

 

 

 

important for completion of this dissertation. Dr. Annette Kirshner from
NIEHS was also helpful in facilitating an extension of my original NRSA
from NIEHS-NIH and Dr. Charles Mackenzie and Dr. Stuart Sleight were
helpful in securing funds for research.

Technical help and assistance from Mr. Ralph Common with
photography, Dr. Lewis King with flow cytometry, Dr. John Heckman with
immunofluorescence, as well as Dr. Yasuko Marushige, Dorothy Okazaki and
Lyla Melkerson-Watson with tissue culture, are all gratefully acknowledged.
In addition, Donna Craft and Carol Ayala were helpful with preparation of
slides for photomicroscopy.

Graduate students also played an important role in making suggestions
and in overcoming the stress factors, that just seemed to appear out of
nowhere and inflict their pain on graduate students. Thanks, therefore, go to
Dr. Mike Yeager, Dr. Dev Paul and Dr. Jim Fikes from the Department of
Pathology and Doug Wiesner, Lyla Melkerson-Watson and Bao-Jen Shyong
from the Department of Biochemistry. Postdoctoral fellows Dr. Kiyoshi
Ogura, Dr. Misa Ogura, Dr. Zhi-Heng Huang and Dr. Dick Hendry in
Biochemistry were helpful in answering questions and undergraduate
students Amanda, Barb and Amy always generated enthusiasm.

Thanks as well, to those providing special services; to Bessie Bryant for
care of my laboratory animals and to the night custodian George in
Biochemistry for not only helping to keep things spic and span, but for
helping to keep me awake on long nights in the laboratory. I also want to
thank the many secretaries, who assisted me time and time again, Carol,
Mindy and Vickie in Biochemistry and Betty, Bev, Denise, Lezlee, Chris, Lisa,
Anita and Charla in Pathology.

vii

 

 

 

 

Thanks, to my children; Debbie, Bill, Wendy and Becky for their love

and understanding. This could not have been done without their support.
Thanks also, to our good neighbors Geri, Jamie and Jake from
Williamston. And thanks, to all those I've forgotten to mention, to those in
the community and college who supported me, helped me, and assisted me
in so many ways. Its been an education and an experience I shall always
remember and cherish.
Special thanks go to members of my graduate committee:
Dr. Bert Koestner (Chairperson), Department of Pathology
Dr. Bob Leader, Institute for Environmental Toxicology
Dr. Kathy Lovell, Department of Pathology
Dr. Keiji Marushige, Department of Pathology
Dr. Larry Fischer, Institute for Environmental Toxicology

Dr. Chuck Sweeley, Department of Biochemistry
Thanks also to Dr. Henry Bredeck, vice-president for research and
graduate studies, Michigan State University and Dr. Asish Nag, professor of
biological science, Oakland University for their encouragement to follow
through on my scientific endeavors.

THANK YOU ALL!!

Sincerely,

7?»

viii

 

 

List of Tables
List of Figures
Chapter 1. Literature Review

TABLE OF CONTENTS

 

 

 

 

Introduction
Chemical Carcinogenesis
Alkylnitrosoureas and Experimental Brain Tumors ............
Astrocytes and Intermediate Filaments
Growth Factors and Differentiation Promoters ......................
Cell Surface Receptors
Signal Transduction
Endocytosis
Oncogene Involvement
Astrocyte Function
Astrocytoma Treatment
The Question
References

 

 

 

 

 

 

 

 

 

 

Chapter 2. The in vitro Generation of an Interphase, N on-cycling

Go Subpopulation of Anaplastic Astroytoma Cells
Induced by Nerve Growth Factor

Abstract
Introduction

Materials and Methods
materials
neonatal astrocyte cultures
astrocytoma cells
culture conditions
cell culture treatment
cell harvesting and fixation
cell preparation and staining for flow cytometry .......
flow cytometric fluorescent measurements ................

 

 

 

 

 

 

 

 

 

 

 

xiii

VDQQVU‘IOJNHH

I-IHI-IHI-I
\IUIDHO

31
32
33

37
37
37
38
39
39
40
40
41

 

 

 

‘n 11

Discussion

“t

r‘

Chapter 3. Induction of Morphologic Differentiation in Astrocytes

and Astrocytoma Cells by Nerve Growth Factor ........
Abstract
Introduction
Mptbnds

materials

neonatal rat astrocytes

astrocytoma cells

culture cnnditinnc

morphologic effects of NGF and glutamate ................

n In

 

Discussinn

R‘ .--

Chapter 4. The in vitro Effects of Nerve Growth Factor on the

Detection of Intermediate Filaments Expressing
Glial Fibrillary Acidic Protein Epitopes by Rat
Anaplastic Astrocytoma Cells

Abstract

Intrndu ctinn

Materials and Method:
material:
neonatal astrocytes
astrocytoma cells
culture conditions
cell culture :- -—- .
indirect immunofluorescence of vimentin and
GFAP

 

 

 

 

 

 

n It

Discussinn

11‘

Chapter 5. An in vitro Study on the Effects of Nerve Growth

Factor on Gamma-aminobutyric Acid Uptake by
Neonatal Rat Astrocytes and Rat T-9 Astrocytoma
Cells

Abstract

51
55

59
60
61
63

63
63

65
65

67
74
80

85
86
87

91
91
91
92
93
94

94
96
102
107

113
114

 

Chapter 7. Summary and Conclusions

Introduction

 

Material and Methods
materials
neonatal astrocyte cultures
astrocytoma cells
culture conditions
GABA uptake experiments
scintillation analysis
protein concentration
statistical analysis

Results

Discussion

References

 

 

 

 

 

 

 

 

 

 

 

 

Chapter 6. An in vitro Study on Glutamate Uptake by Neonatal

Rat Astrocytes and Rat Astrocytoma Cells ...................
Abstract
Introduction

Materials and Methods
materials
neonatal astrocyte cultures
astrocytoma cells
culture conditions
GLU uptake experiments
scintillation analysis
protein concentration
statistical analysis

Results

Discussion

References

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Oncogene Expression and Cell Proliferation
Morphology and Intermediate Filaments
GLU and GABA Uptake Functions
Conclusions
References

 

 

 

 

xi

 

 

115

118
118
118
119
120
120
122
122

123
126
131

137
138
139

143
143
143
144
145
145
147
147
147

148
151
156

164
164
170
175
181
184

LIST OF TABLES

Chapter 5. An in vitro Study on the Effects of Nerve Growth
Factor on Gamma-aminobutyric Acid by
Neonatal Rat Astrocytes and Rat T-9 Astrocytoma
Cells

Table 1. GABA uptake by neonatal astrocytes
and anaplastic astrocytoma cells cultured under
varying conditions and treatments .................. 125

Chapter 6. An in vitro Study on Glutamate Uptake by Neonatal
Rat Astrocytes and Rat T-9 Astrocytoma Cells

Table 1. GLU uptake by neonatal astrocytes
and anaplastic astrocytoma cells cultured under
varying conditions and treatments ................... 150

xii

 

 

LIST OF FIGURES

Chapter 2. The in vitro Generation of an Interphase, Non-cycling
Go Subpopulation of Anaplastic Astroytoma Cells
Induced by Nerve Growth Factor

Figure 1. Schematic illustration of a
fluorochromatic scatter graph histogram ....... 44

Figure 2. Scatter graph histogram of

proliferating cell pop ‘ ‘innc 46
Figure 2A NR-I neonatal astrocytes
Figure ZB T-9 anaplastic astrocytes

Figure 3. Scatter graph histogram of AO

stained NR-l astrocytes 48
Figure 3A. no treatment
Figure 3B. NGF (500 ng/ml)
Figure 3C. PCS (100 ill/ml)

 

Figure 4. Scatter graph histogram of A0

stained T-9 anaplastic astrocytoma cells ........... 50
Figure 4A. no treatment
Figure 4B. NGF (500 ng/ml)
Figure 4C. PCS (100 Lil/ml)

Chapter 3. Induction of Morphologic Differentiation in Astrocytes
and Astrocytoma Cells by Nerve Growth Factor

Figure 1. Morphologic differentiation of
astrocytes and anaplastic astrocytoma cells by
NGF in vitrn 69

Figure 1A. noeonatal astrocytes
Figure 13. rat anaplastic astrocytes

 

xiii

 

Figure 2. Morphology of rat anaplastic
astrocytoma cells cultured in CDM ................... 71

Figure 2A. no additional treatments
Figure 23. treated with GLU (25 1.1M)
Figure 2C. treated with NGF (500 ng/ml)

Figure 3. Morphology of neonatal rat
astrocytes cultured in CDM 73

Figure 3A. no additional treatments
Figure 33. treated with GLU (2511M)
Figure 3C. treated with NGF (500 ng/ml)

 

Chapter 4. The in vitro effects of nerve growth factor on the
detection of intermediate filaments expressing
glial fibrillary acidic protein epitopes by rat anaplastic
astrocytoma cells

Figure 1. Vimentin indirect immunofluorescent
staining of neonatal rat astrocytes ..................... 99

Figure 2. GFAP indirect immunofluorescent
staining of neonatal rat astrocytes ..................... 99
Figure 2A. no treatment
Figure 23. treated with NGF (500 ng/ml)

Figure 3. Phase contrast photomicrograph
of anaplastic astrocytoma cells 101

 

Figure 4. GFAP indirect immunofluorescent
staining of anaplastic astrocytoma cells ............ 101

xiv

 

 

Chapter 1

Literature Review

Introduction

Primary brain tumors are comprised of an array of diverse neoplastic
cell types, each with its own characteristic morphology and growth pattern.
The most prominent of these brain tumors are those derived from
neuroectodermal origin. Cells of neuroectodermal origin differentiate into
four morphological and functionally distinct cell types. They are neurons,
astrocytes, oligodendrocytes and ependymal cells. Bailey and Cushing were
the first to compare tumor cells with developmental stages of normal cells
(Bailey and Cushing, 1926). Their classification of brain tumor cells resulted
in an orderly classification of what Virchow had earlier categorized
inclusively and termed glioma (Virchow, 1846, 1854). Although the Bailey
and Cushing classification has been subjected to a number of criticisms and
modifications by pathologists through the years, it has remained the major
basis for pathological diagnosis and classification of intracranial tumors .

As a result of this classification scheme, gliomas were found to
constitute half of all intracranial tumors. In addition, this classification
system introduced the thought that tumor cells were actually aberrant cells of
one of the developmental stages of normal cells. Astrocytomas were thought
to be represented by aberrant glioblasts or spongioblasts and therefore were
named and classified as glioblastoma and spongioblastoma. Histologically
cells were also classified by their degree of anaplasia, hence differentiated and

anaplastic astrocytomas became common terminology for astrocytic tumors.

 

Chemical Carcinogenesis

As a result of the classification system introduced by Bailey and
Cushing, comparative histology became the norm for giving a more specific
identity to tumors. However, there was still a lack of information on the
genesis of brain tumors. It wasn't until the 1930's and the advent of
experimentally induced brain tumors in laboratory animals that greater
understanding of the process of neoplastic transformation and carcinogenesis
in general began to unfold. The first‘reports on experimentally produced
brain tumors were the production of brain tumors following intracerebral
application of polycyclic aromatic hydrocarbons (Seligman and Shear, 1939).
Since then, many chemical compounds have been reported to selectively
induce tumors in the nervous system (Kleihues et al, 1976; Bigner and
Swenberg, 1977). These neuro-oncogenic agents have also been reported to
selectively induce tumors of the nervous system through various routes of
administration.

Epidemiological studies in more recent years have provided
convincing evidence that environmental factors including chemicals do play
a major role in the causation of brain tumors (Moss, 1985). These studies
have also brought forth some optimism, since they imply that the incidence
of brain tumors might be reduced by controlling and cleaning up our
environment.

Carcinogenesis is a pathological process that proceeds through defined
stages of initiation, promotion and progression. It is very likely that the
transition from one stage to another is under the influence of separate
environmental or endogenous factors. Therefore, naturally occurring tumors

and experimentally induced tumors may share similar mechanisms of

 

 

3

causation and development. These mechanisms of carcinogenesis can be
studied by examining the biochemical reactions involved in the development
of experimental brain tumors while at the same time considering the
importance of genetic variations.

Several types of chemicals have been found to initiate cancer in animal
models due to the generation of highly reactive intermediates which bind
covalently with the nucleic acids of the target cell nucleus. This covalent
binding to the DNA of cells is an essential step in the initiation of neoplastic
transformation. The changes in DNA structure which result, in turn,
interfere with normal DNA function during replication and transcription.
These changes in DNA function have been found to involve molecular
pathologic processes, which interfere with normal gene expression. These
molecular processes include oncogene activation, inhibitory gene
suppression, DNA amplification, gene transposition, chromosome

translocations and gene deletions.

Alkylnitrosoureas and Experimental Brain Tumors

An ideal experimental brain tumor model can be produced by
alkylnitrosourea compounds in CD or Fischer rats (Koestner et al, 1971;
Schmidek et al, 1970). Alkylnitrosoureas at specific dose dependent
concentrations are found to produce tumors of the brain and nervous system
in more than 90% of the offspring of pregnant rats following prenatal
administration (Druckery et al, 1965). The neurotropic action of these
compounds is believed to be due primarily to the action of their urea radical,
which facilitates their diffusion into the CNS (Laerum et al, 1978). These
compounds are small in size, lipid soluble and extremely reactive. Like other

alkylating carcinogenic agents, their action is mutagenic and results from the

  

 

t ‘ * nmimsé inn-é" mm. .' '" ,‘

 

4

alkylation of specific DNA bases. The carcinogenic action of both
methylnitrosourea (MNU) and ethylnitrosourea (ENU) in turn has been
linked, by some, to a defect in the excision repair mechanism of these
alkylated bases from brain DNA. Nitrosoureas are found to cause alkylation at
N 7- and 06- positions of guanine (Singer, 1975). The production of N 7-alkyl
guanine adducts are usually quickly and efficiently eliminated by chemical,
non-enzymatic depurination. This action in large part is due to the lability of
the glycosyl bonds in the DNA chain. Thus, depurination follows first order
kinetics with a half-life of approximately 155 hours for N 7-a1kylguanosine. In
addition to spontaneous glycosyl bond cleavage, alkylation products may also
be enzymatically removed by the action of specific glycosylase enzymes
(Kleihues and Margison, 1974; Lindahl, 1976; Bigden et al 1981).

Inability or failure of a cell to eliminate the O6-a1kylation of guanine
has been suggested to be the molecular lesion responsible for the carcinogenic
action of the alkylnitrosoureas in rats (Goth and Rajewsky, 1974). This
specific lesion results in the mis-pairing of guanine with thymine instead of
cytosine. This anomalous base-pairing results in a somewhat stable but
defective conformation to the DNA molecule (Goth and Rajewsky, 1974;
Margison and Kleihues, 1975). However, even though 06-methylguanine has
been shown to be the specific lesion linking MNU to its carcinogenic activity,
species and strain-related differences have not been paralleled by differences
in the excision repair capacity of 06-methylguanine adducts (Kleihues et al,
1979).

The gerbil has been shown to have less capacity to repair 06-
methylguanine than the rat. Yet, the gerbil is not susceptible to the neuro-
oncogenic effects of MNU (Swenberg, 1986). However, MN U does induce a

benign melanoma of the skin in the gerbil. Variations in the effects of

 

5

chemical carcinogens in different species such as the gerbil illustrate the

influence of genetics on the process of carcinogenesis.

Astrocytes and Intermediate Filaments

Astrocytes, as their name implies, are star-shaped cells. End-stage
adult-differentiated astrocytes are represented by two morphologic types,
protoplasmic and fibrous. Protoplasmic astrocytes are characterized by having
an epithelioid morphology while fibrous astrocytes are characterized by
having a stellate morphology. Therefore, their morphologic distinction is the
result of their cell process development and ultimate shape. In normal brain,
protoplasmic astrocytes are generally confined to areas of gray matter which
constitute the cerebral and cerebellar cortices and basal ganglia while fibrous
astrocytes inhabit the white matter (Leeson and Leeson, 1981).

Protoplasmic and fibrous astrocytes can also be distinguished by the
presence, location and content of intermediate-sized intracellular filaments.
These intermediate filaments (IFs) measure 7-11 nm in diameter and form
compact bundles in fibrous astrocytes and scantier arrays in protoplasmic
astrocytes (Russell and Rubenstein, 1989). IFs of differentiated astrocytes have
been reported to be formed by polymerization of a biochemically and
immunologically distinct class of proteins, the glial fibrillary acidic protein
(GFAP). Since the discovery of IF’s in the 1970's and their separation into
distinct cell-specific classes, they have been used for cell identification.
Therefore, GFAP immunofluorescence was used for the identification of
astrocytes. Tumors now could be classified not only by morphology but by use
of a cell-specific marker for GFAP.

Another IF protein, vimentin, has also been found in various cells of

mesenchymal origin including astrocytes. In some mammalian species,

 

 

6

vimentin has been found to constitute the major IF protein of immature
developing astrocytes (Bignami et al, 1980,1982; Bignami and Dahl, 1974).
However , in adult, differentiated astrocytes, vimentin has been found to be a
minor component which co-exists with GFAP (Bjorklund et al, 1984).

Under pathological influences, protoplasmic astrocytes have been
reported to be rapidly and permanently converted to fibrous astrocytes
(Russell and Rubenstein, 1989). This process of gliosis is probably the most
frequent and least specific of all the cellular events correlating with
neuropathology and is easily identifiable by the immediate increase in GFAP
expression.

Since the discovery of IFs they have been assumed to be components of
the cytoskeletal system and partly responsible for cell shape (Ishikawa et al,
1968; Lazarides, 1980). However, recent experiments seem to indicate that IFs
either play a more subtle role in the cytoskeletal system or may only represent
a storage or transportation form of their functional subunit-proteins. IFs
have also been demonstrated to be nucleic acid-binding proteins and
susceptible to limited processing by Cath activated neutral thiol proteases
(Nelson and Traub, 1981; Bigbee et al, 1983). They have been further
characterized by their preferential binding to 18$ ribosomal RNA and single
stranded DNA. However, the binding potential of IFs has not only been
determined by structural conformation but by base composition. IF binding
potential has been shown to be greatest by guanine (Traub and Vorgias, 1983;
Traub, 1985). Therefore, IF subunit-proteins have been suggested to carry out a
role in membrane transduction for the transformation of extracellular signals

to intracellular targets by acting as second messengers.

 

 

 

 

7

Growth Factors and Differentiation Promoters

Differentiation promoters have been known to play important roles in
normal neural development. Among these differentiation promoters are
numerous growth factors and hormones. Nerve growth factor (NGF) is a
potent differentiation promoter. It is a polypeptide that is required for the
maturation and maintenance of sympathetic and sensory neurons as well as
certain cholinergic neurons (Levi-Montalcini and Angeletti, 1968; Hefti, 1986;
Williams et al, 1986; Levi-Montalcini, 1987; Greene and Shooter, 1980;
Yankner and Shooter, 1982; Bradshaw, 1978). NGF's mechanisms of action
have been studied largely by use of the rat pheochromocytoma, PC12, cell line.
NGF's promotion of neuronal differentiation has been well studied by using
the PC12 cell model. NGF has been found to be basically a non-mitogenic
growth factor (Cohen et al 1954). It has only been shown to be mitogenic to
certain established cell lines or during specific periods of development (Lillen
and Claude, 1985; Burstein and Greene, 1982). Research involving NGF has
focused heavily on the mechanisms that control cell development and
differentiation.

Previous studies have indicated that NGF controls the movement of
N a”r and K“ across the cell membrane through regulation of the Na+, K+
pump (Varon and Skaper, 1980; 1983). The control of ionic equilibrium has
further been shown to be fundamental to changes in cell proliferation,
extension and elongation of cytoplasmic processes as well as general cell
maintenance and repair (Rozengurt and Mendoza, 1980; Jaffe and N uccitelli,
1977; Becker, 1981). Therefore, ionic control by N GF may be the key to its
action on cell differentiation (Varon and Adler, 1981; Varon and Skaper,

1980,1983).

 

Cell Surface Receptors

Growth factors and hormones share a common dependency on surface
membrane receptors to initiate their actions on responsive cells (James and
Bradshaw, 1984). In addition, molecular characterization of receptors for both
growth factors and hormones has revealed many similarities in their
structural and functional characteristics. There are two broad classes of
growth factor receptors, those with and without intracellular tyrosine kinase
activity. Most of these receptors are divided into three domains: extracellular,
transmembrane and intracellular. The receptors for insulin and insulin-like
growth factor, for example, are composed of four polypeptides. These
polypeptides are symmetrical dimers of 2 or and 2 B-chains. Each chain in
turn is derived from a single precursor polypeptide. This precursor expresses
the same three domains as other growth factor receptors (Ullrich et al, 1985;
Ebina et al, 1985). These receptor proteins are also glycosylated during
processing by O- and N-linked glycosylation. In addition they are each rich in
disulfide bonds which are often found to be clustered into subdomains which
strongly influence their tertiary structure. Both glycosylation and tertiary

structure play important roles in ligand binding and subsequent activation.

Signal Transduction

The molecular responses that result from growth factor-receptor
interaction are for the most part poorly understood. However, they are
capable of inducing transmembrane signals, which are amplified by a variety
of second messengers. Membrane transduction results in immediate and
delayed response. Changes in gene expression are an example of a delayed
response and are thought to be the result of growth factors or altered growth

factors interfering with translation or transcription processes (James and

 

 

 

9

Bradshaw, 1984). Immediate responses are thought to result from changes in
ionic regulation by the membrane or from the direct action of second
messengers like Ca++ and cyclic-AMP.

Cyclic-AMP is a second messenger which is the product of adenyl
cyclase activation. Adenyl cyclase activation results from the binding of
growth factors to G (GTP) binding proteins, which are enzyme-linked
proteins. Other second messengers include inositol trisphosphate (1P3) and
diacylglycerol (DAG) along with Ca++ and phosphorylated proteins.
Phosphorylation takes place as a result of kinase activation. These second
messengers are all generated by the activation of phospholipase enzymes.
Therefore, the combined activation of phospholipase and adenyl cyclase could
also be responsible for many of the immediate cellular responses generated by

growth factors.

Endocytosis

Another consequence of growth factor-receptor complex formation is
induction of endocytosis. The process of endocytosis is thought to be
responsible for the inactivation of many growth factor- receptor complexes. It
also appears possible that this process is necessary for delayed and prolonged
forms of signal transduction. Growth factors can be transported by endocytic
vesicles, either intact or in an altered form due to processing by enzymatic
action, to the nuclear membrane. Nuclear interaction is characterized by
changes in gene expression which result from the interaction of these growth
factor messengers.

IFs also share a role in receptor-mediated endocytosis. IFs and
microtubules undergo a redistribution shortly after cap formation of the

growth factor-receptor complex. These filaments accumulate in the uropod

 

 

 

10

in a parallel fashion to the nucleus-cap axis (Zucker-Franklin et al, 1979;
Butman et al, 1980, 1981; Dellagi and Brouet, 1982). In resting cells, IFs have
been reported to occupy a waiting position in the perinuclear region. When
cells become stimulated IFs radiate out into a more open fibrillar network to
make contact with endocytic vesicles. Therefore, IFs appear to play an
important role in regulation of growth factor-receptor complex inactivation
and may be responsible for their subsequent activity as transducers of delayed

and prolonged membrane stimulated information systems.

Oncogene Involvement

The transformation of cells from a normal physiologic state to a
neoplastic state has been shown in numerous instances to be accompanied by
increased oncogene expression (Fujimoto et al., 1988; Thompson et al., 1986;
Gordon, 1985). This activity usually is thought to relate to the initiation step
in the process of neoplastic transformation. Specific mutations in brain
tumors have been shown to involve the expression of various oncogenes at
different locations of the cellular genome. These oncogene products disrupt
normal membrane transduction through their molecular mimicry of specific
components of the growth factor-receptor interaction. However, oncogene
amplification may not always be an initiating event in carcinogenesis.
Enhanced expression of c-myc and N-myc has been shown to occur during
tumor progression (Little et al, 1983; Schwab et al, 1983). In addition, myc
activation has been shown to be tightly coupled to growth stimulation of
quiescent cells and, therefore, may be related somehow to the entry of cells
into and through the G1 phase of the cell cycle (Campisi et al, 1984; Kelly et al,
1983). It has also been suggested that the myc proteins may regulate the

expression of other gene products by altering their relationship to the nuclear

 

11

matrix or by directly interacting with regulatory sequences (Eisenman et al,
1985). This latter possibility is reflected by the affinity of myc proteins for
DNA. This, as well as other information on nuclear acting oncogene
products, suggests that a definite relationship exists between nuclear structure
and transformation. However, amplified oncogenes have also been shown to
become transcriptionally silent following induction of tumor cell

differentiation (Thiele et al, 1985; Westin et al, 1982).

Astrocyte Function

Until the last few years, very little was known of the function of
astrocytes. They were simply thought to serve as a means of physical support
for CNS neurons. It had been established that they were involved in the
production of scar tissue following mechanical brain injury and, once
established, remained permanently. But, this rather static view of glial cells
relegated them to only a passive and supportive role, suggesting that their
loss would be of no great consequence to a normal functioning nervous
system. However, glial cells are no longer thought to function passively.
There has been a dramatic change in the way glial cells are thought to
function and in the role they play in the homeostasis of the CNS. Numerous
reviews are now available attesting to the active nature and to the importance
of glial function to a normal functioning brain (Hertz, 1978; Schoffeniels et al,
1978; Varon and Somjen, 1979; Stewart and Rosenberg, 1979). Glial cells are
now depicted as modulators of neuronal function. They have been shown to
act as cellular buffers in- the establishment and maintenance of the blood
brain barrier which regulates the extracellular environment as well as to
control electrolyte abnormalities and the levels of amino acid

neurotransmitters through active uptake systems.

 

 

12

The presence of large swollen astrocytes in a number of instances of
brain edema was interpreted by many and suggested by others to mean that
astrocytes play a functional role in the regulation of water and electrolytes
(Gerschenfield et al, 1959; Katzman, 1961; Tower, 1966; Wasterlain and Torack,

1968; Hirano, 1969). It was soon discovered that astrocytes regulated
extracellular K+ (Kuffler and Nicholls, 1966; Orkland et al, 1966; Haljamae

and Hamberger, 1971; Henn et al, 1972; Kimelberg, 1974; Bourke et al, 1975;
Hertz, 1978).

Astrocyte involvement in ammonia metabolism may be one of their
more important functions . Ammonia is constantly produced in the brain
and increases with increased neuronal activity (Richter and Dawson, 1948;
Tsukada, 1971; Quastel, 1974, 1979). In addition, the brain has been reported to
take up ammonia when ammonia levels increase in blood (Webster and
Gabuzda, 1958; Hindfelt, 1975; Lockwood et al, 1979). The role of astrocytes in
regulating ammonia concentrations is directly related to capillary density and
maintenance of the blood brain barrier (Phelps et al, 1977). Reactive and
degenerative changes in astrocytes and the production of Alzheimer type II
astrocytes in conditions of hyperammonemic states has further substantiated
the proposed role of astrocytes in ammonia detoxification (Zamora et al, 1973;
N orenberg, 1977).

Astrocytes also play an important role in the uptake and metabolism of
amino acid neurotransmitters. The astrocytes' role in regulating
neurotransmitter levels had long been suspected because of the extensive
wrapping of astrocytic processes around synaptic nerve endings (Peters and
Palay, 1965; Bunge, 1970). As a result of numerous studies it has become
apparent that astrocytes play a critical role in the uptake, inactivation and

release of the amino acid neurotransmitters (Henn and Hamberger, 1971;

 

 

1 3

Neal and Iversen, 1972; Ehinger, 1972; Henn et al, 1974; Schrier and
Thompson, 1974; Iversen and Kelly, 1975; Harber and Hutchinson, 1976;
Schousboe, 1978; Hertz, 1979). Astrocytes are now thought to modulate
neuronal function by controlling the synaptic levels of glutamate (GLU) and
aspartate excitatory neurotransmitters as well as gamma-aminobutyric acid
(GABA) and glycine inhibitory neurotransmitters (Watkins, 1973; Curtis and
Johnson, 1974; Davidson, 1976; Hertz, 1979; Johnson, 1978). The biochemical
involvement of astrocytes in this regulatory process consists of a glutamate-
glutamine shunt. In this process the neurotransmitters GLU and GABA ,
which are released by presynaptic and postsynaptic nerve endings, are
inactivated by active and passive cellular uptake mechanisms (Van Den Berg
and Garfinkel, 1971; Van Den Berg, 1972; Benjamin and Quastel, 1975; Henn
and Hamberger, 1971; Schousboe et al, 1977, 1978, 1981; Schousboe, 1981 ;
Waniewski et al, 1986). These neurotransmitters are converted to glutamine
following uptake by astrocytes (Schousboe et al, 1977a, 1977b). Glutamine is
passively released by astrocytes for use by neurons, where it is metabolized to
replenish GLU and GABA transmitter pools (Hertz and Schousboe, 1986).

Astrocytes are also believed to play a role in the metabolism of short-
chain fatty acids and glycogen (Cremer et al, 1975; Ibrahim, 1975; Volpe and
Marasa, 1977; Varon and Somjen, 1979). Astrocytic swelling is an early event
which occurs following periods of ischemia (Kimelberg and Ransom, 1986).
Polyunsaturated fatty acids (PUFAs) are rapidly released from membrane
phospholipids during brain eschemia (Brazen, 1970). PUFAs cause numerous
cellular changes including changes in the fluid domain of membranes,
uncoupling of mitochondrial respiration, inhibition of Na+ and K+ATPase
and the generation of oxygen free radicals and other lipid peroxides (Klausner

et al, 1980; Hillered and Chan, 1987; Chan et al, 1983). Therefore, changes in

 

 

14

the metabolism and processing of short-chain fatty acids could have a

detrimental effect on membrane integrity, protein cross-linking and DNA. In
addition, these short chain fatty acids and PUFAs influence astrocyte uptake
of the amino acid neurotransmitters (Chan et al, 1983; Yu et al, 1986, 1987).
Glycogen levels in astrocytes are found to increase when glutamine
generation is inhibited (Folbergrova et al, 1969) The first sign of metabolic
imbalance affecting astrocytes may also be recognized by increased levels of

glycogen storage.

Astrocytoma Treatment

The most essential treatment of astrocytomas and other brain tumors
continues to be surgical removal. However, in recent years radiotherapy and
chemotherapy have also played important roles. There are many reports
indicating improved therapeutic results using combinations of radiotherapy
and chemotherapy (Suzuki, 1988). Chemotherapy is composed of chemical
pharmaceutics which act as either cytotoxic or cytostatic agents. Newer
treatments involving immunotherapy or reverse transforming agents are
presently in the experimental stage. The goal of. immunotherapy is to
stimulate the immune system to attack and destroy tumor cells through the
use of biological response modifiers. The goal of reverse transforming agents
is to control tumor cells by the action of potent differentiation promoters.
NGF, consistent with its action on PC12 cells, has become the most popular
reverse transforming agent for experimental use against astrocytomas.

All treatments of tumors have been directed at completely eliminating
transformed cells. However, reverse transforming agents have been
suggested for potential drug development based on their ability to
differentiate neoplastic cells. It had logically been thought that if tumor cells

#1!

 

 

 

 

15

could be made to resemble differentiated cell types they would also behave
normally. That is, they would cease to proliferate in an uncontrolled fashion
and lose their transformation characteristics. Due to the reverse transforming
effects of these agents on tumor cells, it was suspected that these agents could
also cause tumor regression. However, most of the differentiation effects
produced by reverse transforming agents appeared to be directed only at the
cell membrane and induce a variation of immediate and delayed responses
responses (Pollock et al., 1990). Agents are needed that induce a prolonged
response and stimulate tumor cells to differentiate towards the end-stage of
the normal comparative cell's developmental sequence and become a
quiescent normal functioning cell type. These expectations have made it
necessary to investigate and compare the effects of these agents on the

function of differentiated tumor cells as well as normal astrocytes.

Is N GF just a promoter of morphologic differentiation in astrocytes?

This dissertation investigates the effects of NGF on proliferation and
function as well as morphology of astrocytoma cells. These studies are
undertaken from a comparative view, rat T-9 anaplastic astrocytoma cells are
compared to neonatal rat astrocytes. Proliferation is studied by characterizing
cells in the cell cycle using single-stranded to double-stranded DNA and
analyzing the effects of NGF by flow cytometry. Morphology is studied by
light-microscopy using morphology as well as intermediate filament
expression as a means of evaluating differentiation promotion following
NGF treatment. Function and the effects of NGF on function are studied by
determining the cellular uptake of glutamate (GLU) and y-aminobutyric acid
(GABA) before and after treatment with NGF. GLU and GABA are the major

excitatory and inhibitory amino acid neurotransmitters, respectively. It has

 

 

 

1 6
been suggested that the increased incidence of seizures that accompany
astrocytomas may be related to a disruption of normal function. Therefore,
GLU and GABA uptake was determined and compared between anaplastic
astrocytoma cells and neonatal astrocytes both before and after NGF
treatment. The results presented here support the development of NGF as a
therapeutic agent for the treatment of anaplastic astrocytomas and seizures

related to this condition.

-i-L':_a-_#filflil‘flsltl I— -€ ' ,...

 

17

References

Bailey, P. and Cushing, H. (1926) A Classification of Tumors of the Glioma
Group. Lippincott, Philadelphia.

Bazan, N. (1970) Effects of ischemia and electro-convulsive shock on free
fatty acid pool in the brain. Biochim. Biophys. Acta 218, 1-10.

Becker, R. (1981) Mechanisms of Growth Control, Thomas, New York.

Benjamin, A. and Quastel, J. (1975) Metabolism of amino acids and
ammonia in rat brain cortex slices in vitro: A possible role of ammonia
in brain function. J. Neurochem. 25, 197-206.

Bigbee, J., Bigner, D., Pegram, C. and Eng, L. (1983) Study of glial fibrillary
acidic protein in a human glioma cell line grown in culture and as a
solid tumor. J. Neurochem. 40, 460-467.

Bigden, J., Eastman, A.and Bresnick, E. (1981) A system in mouse liver for
the repair of 05-methylguanine lesions in methylated DNA. Nucleic
Acid Res. 9, 3089-3103.

Bignami, A. and Dahl, D. (1974) Astrocyte-specific protein and neuroglial
differentiation. An immunofluorescence study with antibodies to glial
fibrillary acidic protein. ] Comp Neurol 153, 27-37

Bignami, A., Dahl, D. and Rueger, D. C. (1980) Glial fibrillary acidic (GFA)
protein in normal neural cells and in pathological conditions. In,
Federoff, S. and Hertz, L. (eds) Advances in cellular neurobiology,
Vol. 1. Academic Press, New York, pp 285-310

 

 

1 8
Bignami, A., Raju, T. and Dahl, D. (1982) Localization of vimentin, the non-
specific intermediate filament protein, in embryonal glia and early
differentiating neurons. Dev. Biol. 91, 286-295.

Bigner, D. and Swenberg, J. (1977) In, Experimental Tumors of the Nervous
System. Upjohn Co., Kalamazoo.

Bjorklund, H., Ericksdotter-Nilsson, M., Dahl, D. and Olson, L. (1984)
Astrocytes in smears of CNS tissues as visualized by GPA and
vimentin immunofluorescence. Med. Biol. 62, 38-48.

Bourke, R., Kimelberg, H., West, C. and Bremer, A. (1975) The effects of
HCOg’ on the swelling and ion uptake of monkey cerebral cortex under
conditions of raised extracellular potassium. J. Neurochem. 25, 323-328.

Bunge, R. (1970) Structure and function of neuroglia: Some recent
observations. In, The Neurosciences: Second Study Program (Schmitt,
F., ed.), pp. 782-797, Rockefeller Univ. Press, New York.

Burstein, D. and Greene, L. (1982) Nerve growth factor has both mitogenic
and antimitogenic activity. Dev. Biol. 94, 477-482.

Butman, 3., Bourguignon, G. and Bourguignon, L. (1980) Lymphocyte
capping induced by polycationized ferritin. J. Cell. Physiol. 105, 7-15.

Butman, 3., Jacobsen, T., Cabatu, O. and Bourgiugnon, L. (1981) The
involvement of CAMP in lymphocyte capping. Cell Immunol. 61, 397-
403.

Campisi, J., Gray, H., Pardee, A., Dean, M. and Sonenshein, G. (1984) Cell
cycle control of c-myc but not c-ras expression is lost following chemical
transformation. Cell 36, 241-247.

 

 

 

 

1 9
Chan, P., Kerlan, R. and Fishman, R. (1983) Reductions of gamma-
aminobutyric acid and glutamate uptake and N a+ K+ -ATPase activity

in brain slices and synaptosomes by arachidonic acid. J. Neurochem.
40, 309-316.

Cohen, S., Levi-Montalcini, R. and Hamburger, V. (1954) A nerve growth-
stimulating factor isolated from sarcoma 37 and 180. Proc. Natl. Acad.
Sci. USA 40, 1014-1018.

Cremer, J., Heath, D., Teal, H., Woods, M. and Cavanagh, J. (1975) Some
dynamic aspects of brain metabolism in rats given a postcaval
anastomosis. Neuropathol. Appl. Neurobiol. 1, 293-311.

Curtis, D. and Johnston, G. (1974) Amino acid transmitters in mammalian
CNS. Ergeb. Physiol. Biol. Chem. Exp. Pharmakol. 69, 97-188.

Davidson, N. (1976) Neurotransmitter Amino Acids. Academic Press, New
York.

Dellagi, K and Brouet, J.-C. (1982) Redistribution of intermediate filaments
during capping of lymphocyte surface molecules. Nature 298, 284-286.

Druckery, H., Ivankovich, S. and Preussmann, R. (1966) Teratogenic and
carcinogenic effects in the offspring after single injection of
ethylnitrosourea to pregnant rats. Nature 210, 1378-1379.

Ebina, Y., Ellis, L., Jamagin, K., Edery, M., Graf, L., Clauser, E., Ou, J.-H.,
Masiarz, F., Kan, Y., Goldfine, 1., Roth, R. and Rutter, W. (1985) The
human insulin receptor cDNA: The structural basis for hormone-
activated transmembrane signalling. Cell 40, 747-758.

Ehinger, B. (1972) Cellular location of the uptake of some amino acids into
the rabbit retina. Brain Res. 46, 297-311.

 

 

 

2 0
Eisenman, R, Tachibana, C., Abrams, H. and Hann, S. (1985) v-myc- and c-

myc-encoded proteins are associated with the nuclear matrix. Mol.
Cell. Biol. 5, 114-126.

Folbergrova, J., Passonneau, J. , Lowry, O. and Schulz, D. (1969) Glycogen,
ammonia and related metabolites in the brain during seizures evoked
by methionine sulphoximine. J. Neurochem. 16, 191-203.

Fujimoto, M., Weaker, F., Herbert, D., Sharp, Z., Sheridan, P. and Story, J.
(1988) Expression of three viral oncogenes (v-sis, v-myc, v-fos) in
primary human brain tumors of neuroectodermal origin. Neurology
38, 289-293.

Gerschenfield, H., Wald, F., Zadunaisky, J. and DeRobertis, E. (1959) Function
of astroglia in water-ion metabolism of the central nervous system.
Neurology 9, 412-425.

Gordon, H (1985) Oncogenes. Mayo Clin. Proc. 60, 697-713.

Goth, R. and Rajewsky, M. (1974) Persistence of 05-ethylguanine in rat brain
DNA: Correlation with nervous system specific carcinogenesis by
ethylnitrosoureas. Proc. Natl. Acad. Sci. LISA 71, 639-643.

Greene, L. and Shooter, E. (1980) Nerve growth factor: Biochemistry,
synthesis, and mechanism of action. Annu. Rev. Biochem. 3, 353-402.

Haber, B. and Hutchinson, H. (1976) Uptake of neurotransmitters and
precursors by clonal cell lines of neural origin. In, Transport
Phenomena in the Nervous System (Levi, G., Battistin, L. and Lajtha,
A., eds.), pp. 179-198, Plenum Press, New York.

Haljamae, H. and Hamberger, A. (1971) Potassium accumulation by bulk
prepared neuronal and glial cells. J. Neurochem. 18, 1903-1912.

Hefti, F. (1986) Nerve growth factor promotes survival of septal cholinergic
neurons after fimbrial transections. J. Neurosci. 6, 2155-2162.

 

 

 

 

 

21

Henn, F. and Hamberger, A. (1971) Glial cell functions: Uptake of transmitter
substances. Proc. Natl. Acad. Sci. LISA 68, 2686-2690.

Henn, F., Goldstein, M. and Hamberger, A. (1974) Uptake of the
neurotransmitter candidate glutamate by glia. Nature 249, 663-664.

Henn, F., Haljamae, H. and Hamberger, A. (1972) Glial cell function: Active
control of extracellular K+ concentration. Brain Res. 43, 437-443.

Hertz, L. (1978) Biochemistry of glial cells. In, Cell Tissues and Organ
Cultures in Neurology (Fedoroff, S. and Hertz, L., eds.), pp. 39-71,
Academic Press, New York.

Hertz, L. (1979) Functional interactions between neurons and astrocytes.
Turnover and metabolism of putative amino acid transmitters. Prog.
Neurobiol. 13, 277-323.

Hertz, L. and Schousboe, A. (1986) Role of astrocytes in compartmentation of
amino acids and energy metabolism. In, Astrocytes (Fedoroff, S. and
Vernadakis, A., eds.), vol. 2, pp. 179-208, Academic Press, New York.

Hillered, L. and Chan, P. (1987) Effects of arachidonic acid on respiratory
activities in isolated brain mitochondria. J. Neurosci. Res.

Hindfelt, B. (1975) The distribution of ammonia between extracellular and
intracellular compartments of rat brain. Clin. Sci. Mol. Med. 48, 33-37.

Hirano, A. (1969) The fine structure of brain edema. In, The Structure and
Function of Nervous Tissue (Bourne,G., ed.), vol. 2, pp. 69-135,
Academic Press, New York.

Ibrahim, M. (1975) Glycogen and its related enzymes of metabolism in the
central nervous system. Ergeb. Anat. Entwicklungsgesch. 52, 5-89.

Ishikawa, H., Bischoff, R. and Holtzer, H. (1968) Mitosis and intermediate-
sized filaments in developing skeletal muscle. J Cell Biol 38, 538-555

 

 

22

Iversen, L. and Kelly, J. (1975) Uptake and metabolism of g-aminobutyric acid
by neurones and glial cells. Biochem. Pharmacol. 24, 933-938.

Jaffe, L. and Nuccitelli, R. (1977) Electrical controls of development. Ann.
Rev. Biophys. Bioengin. 6, 445-476.

James, R. and Bradshaw, R. (1984) Polypeptide growth factors. Ann. Rev.
Biochem. 53, 259-292.

Johnson, J. (1978) The excitant amino acid glutamic and aspartic acid as
transmitter candidates in the vertebrate central nervous system. Prog.
Neurobiol. 10, 155-202.

Katzman, R. (1961) Electrolyte distribution in mammalian central nervous
system. Are glia high sodium cells? Neurology 11, 27-36.

Kelly, K., Cochran, 3., Stiles, C. and Leader, P. (1983) Cell-specific regulation
of the c-myc gene by lymphocyte mitogens and platlet-derived growth
factor. Cell 35, 603-610.

Kimberg, H. and Ransom, B. (1986) Physiological and pathological aspects of
astrocytic swelling. In, Astrocytes (Fedoroff, S. and Vernadakis, A., eds.),
vol. 3, pp. 129-166, Academic Press, New York.

Kimelberg, H. (1974) Active potassium transport and [Na++ K+] ATPase
activity in cultured glioma and neuroblastoma cells. J. Neurochem.
22, 971-976.

Klausner, R., Kleinfield, A., Hoover, R. and Karnovsky, M. (1980) Lipid
domains in membranes. Evidence derived from structural
perturbation induced by free fatty acids and life time heterogeneity
analysis. J. Biol. Chem. 255, 1286-1295.

Kleihues, P. and Margison, G. (1974) Carcinogenicity of N -methyl-N-
nitrosourea: Possible role of repair excision of O6-methylguanine from
DNA. J. Natl. Cancer Inst. 53, 1839-

 

 

2 3
Kleihues, P., Doerjer, G., Swenberg, J., Hauenstein, E., Bucheler, J. and Cooper,
H. (1979) DNA repair as a regulatory factor in the organotropy of
alkylating carcinogens. Arch.Taxical. [ Suppl. ] 2, 253-261.

Kleihues, P., Lantos, P. and Magee, P. (1976) Chemical carcinogenesis of the
nervous system. Int. Rev. Exp. Path. 15, 153-232.

Koestner, A., Swenberg, J. and Wechsler, W. (1971) Transplacental
production with ethylnitrosourea of neoplasms of the nervous system
in Sprague-Dawley rats. Am. J. Path. 63, 37-50.

Kuffler, S. and Nicholls, J. (1966) The physiology of neuroglial cells. Ergeb.
Physiol. Biol. Chem. Exp. Pharmakol. 57, 1-90.

Laerum, O. and Rajewsky, M. (1975) Neoplastic transformation of fetal rat
brain cells in culture after exposure to ethylnitrosourea in vivo. J.

Natl. Cancer Inst. 55, 1177-1187.

Laerum, 0., Bigner, D. and Rajewsky, M. (1978) Biology of Brain Tumors.
IUAC monograph, vol 30, Geneva, Switzerland.

Lazarides, E. (1980) Intermediate filaments as mechanical integrators of
cellular space. Nature 283, 249-256

Leeson and Leeson (1981) Histology. W. B. Saunders, Philadelphia.

Levi-Montalcini, R. (1987) Nerve growth factor: thirty-five years later. EMBO
J. 6, 1145-1154.

Levi-Montalcini, R. and Angeletti, P. (1968) Nerve growth factor. Physiol.
Rev. 48, 534-564.

Lillen, L. and Claude, P. (1985) Nerve growth factor is a mitogen for cultured
chromaffin cells. Nature 317, 632-634.

m. I - .'_9'_Ufn'm:' ‘1‘!"‘5- .1 o. u. r . . .

24

Lindahl, T. (1976) New classes of enzymes acting on damaged DNA. Nature
259, 64-66.

Little, C., N au, M., Carney, D., Gazdar, A. and Minna, J. (1983) Amplification
and expression of the c-myc oncogene in human lung cancer cell lines.
Nature 306, 194-196.

Lockwood, A., McDonald, J., Gelbard, A., Reiman, R, Laughlin, J., Duffy, T.
and Plum, F. (1979) The dynamics of ammonia metabolism in man:
Effects of liver disease and hyperammonemia. J. Clin. Invest. 63, 449-

460.

Margison, G. and Kleihues, P. (1975) Chemical carcinogenesis in the nervous
system. Preferential accumulation of 06-methylguanine in rat brain
deoxyribonucleic acid during repetitive administration of N-methyl-N-
nitrosourea. Biochem. J. 148, 521-525.

Moss, A. (1985) Occupational exposure and brain tumors. ]. Toxicol.
Environ. Health 16, 703-711.

Neal, M. and Iversen, L. (1972) Autoradiographic localization of 3H-GABA in
rat retina. Nature: New Biol. 235, 217-218.

Nelson, W. and Traub, P. (1981) Properties of Ca2+ activated protease specific
for the intermediate-sized filament protein vimentin in Ehrlich-
ascites-tumour cells. Eur. J. Biochem. 116, 51-57.

N orenberg, M. (1977) A light and electron microscopic study of experimental
portal-systemic (ammonia) encephalopathy. Lab. Invest. 36, 618-627.

Orkand, R., Nicholls, J. and Kuffler, S. (1966) Effect of nerve impulses on the
membrane potential of glial cells in the central nervous system of
amphibia. J. Neurophysiol. 29, 788-806.

25

Peters, A. and Palay, S. (1965) An electron microscopic study of the
distribution and patterns of astroglial processes in the central nervous
system. J. Anat. 99, 419.

Phelps, M., Hoffman, E. and Rayband, C. (1977) Factors which affect cerebral
uptake and retention of 13NH;:,. Stroke 8, 694-702.

Pollock, J., Krempin, M. and Rudy, B. (1990) Differential effects of NGF, FGF,
EGF, CAMP, and Dexamethasone on neurite outgrowth and sodium
channel expression in PC12 cells. J. Neurosci. 10, 2626-2637.

Quastel, J. (1974) Amino acids and the brain. Biochem. Soc. Trans. 2, 766-
780.

Quastel, J. (1979) The role of amino acids in the brain. Essays Med. Biochem.
4, 1-48.

Richter, D. and Dawson, R. (1948) The ammonia and glutamine content of
the brain. J. Biol. Chem. 176, 1199-1210.

Roscoe, J. and Classe, P. (1976) A sequential in viva/in vitro study of
carcinogenesis induced in the rat brain by ethylnitrosourea. Nature
262, 314-316.

Rozengurt, E. and Mendoza, S. (1980) Monovalent ion fluxes and the control
of cell proliferation in cultured fibroblasts. In, Growth regulation by
ion fluxes (Leffert, H., ed.), Ann. New York Acad. Sci. 339, 175-190.

Russell, D. and Rubenstein, L. (1989) Pathology of Tumors of the Nervous
System. 5th edn., p. 96, Williams & Wilkins, Baltimore.

Schmidek, H. H., Nielsen, S. L., Schiller, A. L. and Messer, J. (1971)
Morphologic studies of rat brain tumors induced by N-
nitrosomethylurea. ]. Neurosurg. 34, 335-340.

26

Schoffeniels, E., Franck, G., Hertz, L. and Tower, D. (1978) Dynamic
Properties of Glial Cells, Pergamon, Oxford.

Schousboe, A. (1978) Glutamate, GABA and taurine in cultured, normal glial
cells. In, Dynamic Properties of Glial Cells (Schoffeniels, E., Franck, G.,
Hertz, L. and Tower, D., eds.), pp. 173-182, Pergamon Press, Oxford.

Schousboe, A. (1981) Transport and metabolism of glutamate and GABA in
neurons and glial cells. Int. Rev. Neurobiol. 22, 1-45.

Schousboe, A., Hertz, L. and Svenneby, G. (1977) Uptake and metabolism of
GABA in astrocytes cultured from dissociated mouse brain
hemispheres. Neurochem. Res. 2, 217-229.

Schousboe, A., Svenneby, G. and Hertz, L. (1977) Uptake and metabolism of
glutamate in astrocytes cultured from dissociated mouse brain
hemispheres. J. Neurochem. 29, 999-1005.

Schrier, B. and Thompson, E. (1974) On the role of glial cells in the
mammalian nervous system. Uptake, excretion and metabolism of
putative neurotransmitters by cultured glial tumor cells. J. Biol. Chem.
249, 1769-1780.

Schwab, M., Alitalo,K., Klempnauer, K., Varmus, H., Bishop, J., Gelbert, F.,
Brodeur, G., Goldstein, M. and Trent, J. (1983) Amplified DNA with
limited homology to myc cellular oncogene is shared by human
neuroblastoma cell lines and neuroblastoma tumour. Nature 305, 245-
248.

Seligman, A. and Shear, M. (1939) Studies in carcinogenesis. VIII:
Experimental production of brain tumors in mice with
methylcholanthrene. Am. J. Cancer 37, 364-395.

Singer, B. (1975) The chemical effects of nucleic acid alkylation and their
relation to mutagenesis and carcinogenesis. Prog. Nucleic Acid Res
Mol. Biol. 15, 219-284.

27

Stewart, R. and Rosenberg, R. (1979) Physiology of glia: Glial-neuronal
interactions. Int. Rev. Neurobiol. 21, 275-309.

Suzuki, J. (1988) Treatment of Glioma, Springer-Verlag, New York.

Swenberg, J. (1986) Brain tumors - problems and perspectives. Pd. Chem.
Toxicol. 24, 155-158.

Thiele, C., Reynolds, C. and Israel, M. (1985) Decreased expression of N-myc
precedes retinoic acid-induced morphological differentiation of human
neuroblastoma. Nature 313, 404-406.

Thompson, C‘., Challoner, P., Neiman, P. and Groudine, M. (1986) Expression
of the c-myb proto-oncogene during cell proliferation. Nature 319, 374-
380.

Tower, D. (1966) Distribution of cerebral fluids and electrolytes in vivo and
in vitro. In Brain Edema (Seitelberger, F. and Klatzo, I., eds.), pp. 303-
332, Springer-Verlag, New York.

Traub, P. (1985) Are intermediate filament proteins involved in gene
expression? In, Intermediate Filaments (Wang, E., Fischman, D., Liem,
R. and Sun, T.-T., eds.), Ann. N. Y. Acad. Sci. 455, 68-78.

Traub, P. and Vorgias, C. (1983) Involvement of the N-terminal peptide of
vimentin in the formation of intermediate filaments. ]. Cell Sci. 63,
43-67.

Tsukada, Y. (1971) Ammonia metabolism. In, Handbook of Neurochemistry
(Lajtha, A., ed.), vol. 5, pp. 215-233, Plenum Press, New York.

Ullrich, A., Bell, J., Chen, E., Herrera, R., Petruzzelli, L., Dull, T., Gray, A.,
Coussens, L., Liao, Y.-C., Tsubokawa, M., Mason, A., Seeburg, P.,
Grunfeld, C., Rosen, D. and Ramachandran, J. (1985) Human insulin
receptor and its relationship to the tyrosine kinase family of oncogenes.
Nature 313, 756-761.

 

28

Van Den Berg, C. (1972) A model of compartmentation in mouse brain based
on glucose and acetate metabolism. In, Metabolic Compartmentation in
the Brain (Balazs, R. and Cremer, J., eds.), pp. 137-166, MacMillan,
London.

Van Den Berg, C. and Garfinkel, D. (1971) A simulation study of brain
compartments: Metabolism of glutamate and released substances in
mouse brain. Biochem. J. 123, 211-218.

Varon, S. and Adler, R. (1981) Trophic and specifying factors directed to
neuronal cells. In, Adv. Cell. Neurobiol (Fedoroff, S. and Hertz, L.,
eds.), vol. 2, pp. 115-163, Academic Press, New York.

Varon, S. and Skaper, S. (1980) In Tissue culture in neurobiology (Giacobini,
E., Vernadakis, A. and Shahar, A., eds.), pp. 333-347, Raven Press, New
York.

Varon, S. and Skaper, S. (1983) In Somatic and autonomic nerve-muscle
interactions (Bumstock, G., Urbova, G. and O'Brien, R., eds.), pp. 213-
252, Elservier/ N orth Holland, Amsterdam.

Varon, S. and Somjen, G. (1979) Neuron-glia interactions. Neurosci. Res.
Program Bull. 17, 1-239.

Varon, S. and Somjen, G. (1979) Neuron-glia interactions. Neurosci. Res.
Program. Bull. 17, 1-239.

Virchow, R. (1846) Ueber das granulierte ansehen der wandungen der
gehirnventrikel. Allgemeine Zeitschrift Psychiatric Psychisch-
gerictliche Medicin 3, 242-250.

Virchow, R. (1854) Ueber eine im gehirin und ruckenmark des menschen
aufgefundene substanz mit der chemischen reaction der cellulose.
Archiv Pathologische Anatomie Physiologic Klinische Medicin 6, 135-
138.

 

29

Volpe, J. and Marasa, J. (1977) Short-term regulation of fatty acid synthesis in
cultured glial neuronal cells. Brain Res. 129, 91-106.

Waniewski, R. A. and Martin, D. L. (1986) Exogenous glutamate is
metabolized to glutamine and exported by rat primary astrocyte
cultures. J. Neurochem. 47, 304-313.

Wasterlain, C. and Torack, R. (1968) Cerebral edema in water intoxication.
Arch. Neurol. 19, 79-87.

Watkins, J. (1973) Metabolic regulation in the release and action of excitatory
and inhibitory amino acids in the central nervous system. Biochem.
Soc. Symp. 36, 33-47.

Webster, L., Jr. and Gabuzda, G. (1958) Ammonium uptake by the extremities
and brain in hepatic coma. 1. Clin. Invest. 37, 414-424.

Westin, E., Wong-staal, F., Gelmann, E., Dalla-Favera, R., Papas, T.,
Lautenberger, J., Eva, A., Reddy, P., Tronick, S., Aaronson, S. and Gallo,
R. (1982) Expression of cellular homologues of retroviral onc genes in
human hematopoietic cells. Proc. Natl. Acad. Sci. USA 79, 2490-2494.

Williams, L., Varon, 5., Peterson, G., Wictorin, K., Fischer, W., Bjorklund, A.
and Gage, F. (1986) Continuous infusion of nerve growth factor
prevents basal forebrain neuronal death after fimbria fornix
transection. Proc. Natl. Acad. Sci. USA 83, 9231-9235.

Yankner, B. and Shooter, E. (1982) The biology and mechanism of action of
nerve growth factor. Annu. Rev. Biochem. 51, 845-868.

Yu, A., Chan, P. and Fishman, R. (1986) Effects of arachidonic acid on
glutamate and gamma-aminobutyric acid uptake in primary cultures of
rat cerebral cortical astrocytes and neurons. J. Neurochem. 47, 1181-
1189.

 

3 0
Yu, A., Chan, P. and Fishman, R. (1987) Arachidonic acid inhibits uptake of
glutamate and glutamine but not of GABA in cultured cerebellar
granule cells. J. Neurosci. Res. 17, 424-427.

Zamora, A., Cavanagh, J. and Kyu, M. (1973) Ultrastructural responses of the
astrocytes to postcaval anastomosis in the rat. J. Neurol. Sci. 18, 25-45.

Zucker-Franklin, D., Liebes, L. and Silber, R. (1979) Differences in the
behavior of the membrane and membrane-associated filamentous
structures in normal and chronic lymphocytic leukemia (CLL)
lymphocytes. Immunol. 122, 97-107.

 

Chapter 2

The in vitro Generation of an Interphase, N on-cycling,
Go Subpopulation of Anaplastic Astrocytoma Cells Induced by

Nerve Growth. Factor

William R. Hare Jr.

Department of Pathology
Michigan State University
East Lansing, MI 48824-1316

Telephone, 51 7-353-91 60

Telefax no. 517-336-1053 -

31

 

32

Abstract

The simultaneous measurement of single stranded and double
stranded DNA by using flow cytometric techniques has made it possible to
distinguish subpopulations of proliferating and non-proliferating cells. The
in vitro growth of neonatal astrocytes in a chemically defined medium
(CDM) and subsequent treatment with fetal calf serum (FCS) induced the
generation of a separate proliferating cell population of astrocytes. The in
vitro growth of anaplastic astrocytoma cells in CDM and subsequent
treatment with nerve growth factor (NGF) (500 ng/ ml) induced the
generation of a separate non-proliferating, quiescent G0 subpopulation of
interphase astrocytoma cells. NGF treatment failed to generate a
subpopulation of neonatal astrocytes and FCS treatment failed to generate a
subpopulation of astrocytoma cells. Therefore, the mechanism of induction
for the generation of subpopulations of astrocytes and astrocytoma cells differ
markedly. As a result, this report supports the potential development of
NGF as a therapeutic agent for the treatment of astrocytomas.

Running Title: NGF Generates N on-cycling G0 Subpopulation in _
Astrocytoma Cells.

Key Words: Nerve growth factor, neonatal astrocytes, astrocytoma
cells, flow cytometry, cell cycle analysis

 

33

Introduction

The concept that tumors are derived from a heterogeneous cell
population composed of both actively cycling and non-cycling cells has been
debated for years (Mendelson, 1962). However, the phenomenon of cell
heterogeneity is shared between tumor cell and normal neonatal precursor
cell populations (Raff et al., 1983; Levi et al., 1986).

Fibrous astrocytes and protoplasmic astrocytes, which differ in their
morphological characteristics, but not in karyotype, have been reported to
derive from a common undifferentiated astrocytic precursor cell population
(Raff et al., 1983; Levi et al., 1986). These astrocytes derived from this
precursor population are also found to differ in their in vitro growth rate,
responsiveness to hormones and growth factors, as well as in antigen
expression (Raff et al., 1983). Likewise, cloned tumor cells from a single
homogeneous appearing tumor have also been reported to be heterogeneous
in their in vitro growth rate, as well as in karyotype, antigen expression,
metastatic and invasive potential, immunogenic potential and
responsiveness to chemotherapeutic agents (Hankansson and Trope, 1974;
Pimm and Baldwin, 1977; Fidler and Kripke, 1977; Steel, 1977).

The ability to grow and reproduce is a fundamental property of most
cell populations. Cells are expected to grow and proliferate in an orderly and
regulated fashion. Normal cells, as well as tumor cells, grow and proliferate
by cells entering a cell cycle. The cell cycle consists of two distinct periods,
interphase and division, and is represented by G1, S, G2 and M phase
components. The interphase and division periods can be distinguished as

separate subpopulations of proliferating cells. The G1 and G2 components

 

34

represent cells in the intervals between DNA replication and mitosis. S phase
is the period of DNA synthesis and M phase is the period of DNA division,
which defines mitosis. There is also a Go component, which represents a
subpopulation of non-cycling cells, whose growth was arrested in interphase.
Cells which make up this non-cycling Go component of a normal
heterogeneous population may also be considered to be a component of a
quiescent pool, a separate and distinct subpopulation of the interphase cell
population. Therefore, cycling cells have an interphase subpopulation which
is represented by G0, G1 and a subpopulation during the period of division
which is represented by G2+M. N on-cycling cells are represented by a distinct
quiescent subpopulation also represented by G0, G1 components which is
derived from the interphase cells.

Unlike normal cells, which can freely enter the G0 phase, tumor cells
continue to divide and are assumed to exhibit only an extended G1 phase.
They have no true non-cycling cell component nor a quiescent
subpopulation.

Subpopulations of cycling cells can be distinguished by their variation
in DNA content. The G2 and M subpopulation have twice the DNA of the Go
and G1 subpopulation. In addition, the Go and G1 components and the G2
and M components of a specific cell type are expected to share the same DNA
content. Therefore, standard cell cycle analysis, utilizing total DNA, can not
necessarily differentiate the subpopulation components G0 and G1 or G2 and
M.

Flow cytometry uses the principle of analyzing a population of cells by
examining single cells, one at a time, following staining with a cell-specific or
cell-component-specific fluorescent labeled dye. Flow cytometry has shown

tremendous advantage in cell cycle analysis due to its speed, sensitivity and

 

 

 

35

precision. Because of these advantages, flow cytometry has become a very
popular method for cytokinetic studies. Several different flow cytometric
techniques incorporating the use of fluorescent dyes have been reported
which are specific for DNA, RNA or protein (Arndt-Jovin and Jovin, 1977;
Darzynkiewciz et al., 1979a; Crissman et al., 1990). It has also been reported
that differential staining of DNA exists between mitotic and interphase cells
(Darzynkiewicz et al., 1977). Continuously dividing cells can also be easily
characterized using flow cytometry by detecting changes in the cellular
distribution of total DNA while cells pass through the cell cycle. In addition,
flow cytometric scatter graph histograms can be generated using two
parameters simultaneously (Darzynkiewicz et al., 1979b; Darzynkiewicz et al.,
1980). These histograms are characteristic and unique for each specific cell
type and are representative of the cell population's proliferation potential
(Dethlefsen et al., 1980).

Acridine orange (A0) is a fluorescent dye which has affinity for both
DNA and RNA. AO intercalates into double-stranded regions of nucleic acids
giving a green fluorescence near 530 nm and contrasts with its stacking on
single-stranded nucleic acids, where it is found to give a red fluorescence near
640 nm. Fixed cells may therefore be analyzed as to their single-stranded (ss)
and double-stranded (ds) DNA characteristics by elimination of cellular RNA
(Darzynkiewicz et al., 1976). This technique should also make it possible to
compare cell populations based on gene expression, since ssDNA content
correlates well with regions of gene expression.

Nerve growth factor (NGF) is a protein necessary for the development,
maintenance and survival of sympathetic, sensory and specific cholinergic
neurons (Levi-Montalcini and Angeletti, 1968; Hefti, 1986; Williams et al.,

1986). NGF has been widely recognized as an in vitro differentiation

 

3 6
promoter of glioma cells (Marushige et al., 1987; Marushige et al., 1989). In

addition, NGF has been shown to retard the growth of anaplastic glioma cells
in vivo and in vitro (Vinores and Koestner, 1980; Marushige et al, 1987).

The purpose of this study was to compare the characteristic AO flow
cytometric histograms of neonatal rat astrocytes and rat astrocytoma cells and
to distinguish induced changes in cell populations as well as subpopulations.
If NGF induced growth retardation was due to an increase in a Go-G1 phase,
AO flow cytometry should be capable of detecting such a change. This study,
therefore, reports on a subpopulation of proliferating cells in neonatal
astrocytes and a subpopulation of non-proliferating cells in anaplastic
astrocytoma cells generated by the in vitro treatment of cell cultures with fetal

calf serum or NGF, respectively.

 

 

37

Materials and methods

Materials

NGF (2.55, grade II), Dispase (grade II) and RNase were purchased from
Boehringer-Mannheim Biochemicals (Indianapolis, IN,USA). Fetal calf
serum (FCS) was purchased from Hyclone Sterile Systems Inc. (Logan, UT,
USA). All culture media were prepared using stock solutions, chemicals and
. supplies from Life Technologies Inc. GIBCO Labs (Grand Island, NE, USA),
Corning Glass Works Inc. (Corning, NY, USA) and Sigma Chemical Co. (St.
Louis, MO, USA). Hl-1 supplement was purchased from Endotronics Inc.
(Coon Rapids, MN, USA).

Neonatal astrocyte cultures

Four-day old Fischer rat pups obtained from Charles River Laboratories
were used to establish cultures of cycling neonatal rat astrocytes (NR-1). The
brains of four sibling rat pups were removed following euthanasia and placed
in a petri dish containing warm DMEM high-glucose media. The meninges
were removed and the brain stem and cerebellum separated. The cerebral
tissues were minced into small pieces (<0.5 mm), combined and placed into a
sterile centrifuge tube containing 5 ml of DMEM high-glucose media. This
suspension of brain tissue was then centrifuged at 750 X g for a period of 10
min. The supernatant fluid was removed and discarded and replaced with 1.5 '
m1 of Collagenase II-S (Sigma) solution (0.8%, w/v). The enzyme treatment
procedure was followed by incubation of the cell mixture on a warm water
bath (37°C) shaker for 30 min, after which the brain tissue was completely

dissociated to a cell suspension by intermittently pipetting with a sterile glass

 

.~_..<.:.'_ v ~ -

38

pasteur pipette supplied with a cotton filter. At no time during preparation of
these cultures were the cells allowed to cool below 30°C. Following the
enzyme treatment the cell suspension was centrifuged at 200 X g for a period
of 10 min and the supernatant fluid removed and discarded. An equal
amount (approx. 3 ml) of DMEM high-glucose medium containing 15% FCS
was gently layered on top of the cell pellet. Then the upper 2 / 3 of the cell
pellet was resuspended in fresh medium using gentle pipetting action. A 1
ml aliquot of this cell suspension was quickly removed and used to seed
primary cultures at approximately a 1:20 split ratio (0.25 ml : 5 ml medium).
These stock cultures were started and maintained in 25 cm2 tissue culture
flasks (Corning) containing 5 ml of complete medium. The medium was
replaced with fresh medium the following day and replaced every 3 days
thereafter. Neonatal astrocytes used in these experiments were at their 3rd
passage. The astrocytic character of these cells was indicated at second and
third passage not only by morphology and their ability to take up 7-
aminobutyric acid, but by their content of the astrocyte-specific, glial fibrillary
acidic protein. Approximately 95% of these cells were GFAP positive.

Astrocytoma cells

The rat anaplastic astrocytoma cells were supplied by Dr. A. Koestner of
the Department of Pathology, Michigan State University, East Lansing, MI.
from his cell storage bank. The T-9 cell line originated from a high grade-
anaplastic astrocytoma induced in Fischer rats by treatment with N-methyl-
N-nitrosourea (MNU) (Schmidek et al., 1971). Stock cultures of T-9 were
established in 25 cm2 tissue culture flasks and maintained in complete
serum-supplemented medium (DMEM and 10% FCS) with replacement
every 3rd day and the splitting of cells every 6th day at a 1:100 split ratio.

3 9

Culture conditions

Cells were cultured in a Hotpack C02 Incubator which was maintained
at 5% C02, 37°C and constant humidity. Both NR-l and T-9 cells were split
into 25 cm2 tissue culture flasks containing 5 ml of complete medium. The
complete medium was changed the following day and replaced with either of
two chemically defined media (CDM). CDM HLIA was composed of DMEM
supplemented with 1% (v/v) HL-l supplement, 400uM glutamine,
gentamycin at 10 ug/ ml, glucose at 1 mg/ ml, CaClz at 175 ug/ ml and MgSO4
at 125 ug/ ml. CDM HLIB was also composed of DMEM and contained the
same supplementation as HLIA but also contained hydrocortisone at 1.6
ug/ ml, prostaglandin F2-alpha at 440 ng/ ml, putrescine at 78 ug/ ml, basic-
fibroblastic growth factor at 8.8 ng/ ml and myelin basic protein at 440 ng/ ml.
This HL1B medium was a slight modification of that originally proposed by
Morrison and De Vellis in 1981 as a CDM that initiates differentiation of
astrocytes. HL-l supplement contained 29 ug/ m1 total protein with 15 ug/ m1

insulin and contained no additional growth factors.

Cell culture treatment

NR-1 and T-9 cells used in these studies were seeded at 6-12,000
cells/ ml into 5 m1 of complete, serum-supplemented medium at initial
plating. The medium was changed the following day to serum-free CDM.
Three days later the CDM was replaced with fresh CDM of the same type and
NGF or serum added. NGF was added at 500 ng/ ml or serum added at 10%
(500 111/5 ml medium). All cell cultures were treated with NGF or serum for
48 hrs and colcemide (0.1 ug/ ml) for 3 hrs prior to cell fixation.

 

 

40

Cell harvesting and fixation

Following cell culture treatment, the medium was removed and cells
were incubated in 2 ml of Dispase (2.4 U/ ml) for a period of <30 sec. Dispase II
was removed and cells were incubated in 2 ml of trypsin (0.05% trypsin,
0.53mM EDTA) for a period of <30 sec. Trypsin was removed and cells were
collected in 4 ml PBS (140mM NaCl, 12mM N a2HPO4, 3.5mM NaH2P04, pH
7.4). The cells were freed from the flask by gentle pipetting action and worked
into a single cell suspension. The contents of two identically treated flasks
were pooled into 15 ml sterile centrifuge tubes and the cell suspension was
centrifuged for 10 min at 750 X g. The supernatant fluid was removed and
discarded. Cells were resuspended into a single cell suspension following the
addition of 1 ml of Hank's balanced salt solution (HBSS) without phenol red.
This cell suspension was again centrifuged for 10 min at 750 X g. The
supernatant fluid was dicarded and cells resuspended in 1 ml HBSS as before.
This single cell suspension was then rapidly fixed in a sterile glass centrifuge
tube by placing cells into ice cold acetone-alcohol fixer (100% acetone, 80%
ethanol; 1:1, v/v). Cells were stored in the cold following fixation.

Preparation and staining of cells was carried out just prior to flow cytometry.

Cell Preparation and staining for flow cytometry

Cells suspended in ice-cold fixative were centrifuged for 10 min at 750 X
g. The fixer was removed and discarded. The cell pellet was resuspended in 1
ml of HBSS without phenol red. This cell suspension was centrifuged and
resuspended in fresh HBSS. The cell suspension was then treated with
RNase (500 Kunitz units) and incubated in a water bath shaker for 60 min at
37°C. The incubation was stopped by centrifuging at 750 X g for 10 min and

pouring off the RNase containing supernatant fluid. The cells were

0' Mk‘ -. -

———fi—.’:—.. .. - :fithfng'ivq-t‘."_

41

resuspended and rinsed one more time in 1 ml of HBSS and then centrifuged
and resuspended in HBSS, as before. An aliquot of 0.2 ml of cell suspension
(510 x 105 NR-l cells and 1-3 x 106 r-9 cells) was then mixed with 0.5 ml of
0.1M KCl-HCl buffer (0.2M KCl, 0.2M HCl; 1:1, v/v, pH 1.4) for 30 sec just prior
to the addition of 2 ml A0 (6 ug/ ml) solution (0.2M NazHPO4, 0.1M citric
acid, pH2.6) at room temperature (20-25°C). Similar techniques have been
described elsewhere (Darzynkiewicz et al., 1976; Trangos et al., 1977;
Darzynkiewicz et al., 1986). Flow cytometric analyses were run as soon as

possible after cell preparation and staining.

Flow cytometric fluorescence measurements

The fluorescence of individual cells was measured in a 50-H
cytofluorograph (Ortho Diagnostics-BD). A blue light (488 nm) source was
used to excite red fluorescence at 640 nm and green fluorescence at 530 nm.
Scatter graph histograms were generated and recorded. All studies were done
on a minimum of three separate cultures and were qualitatively but not
quanitatively analyzed to determine proliferative and non-proliferative
populations and subpopulations. The coefficient of variation in these cell

population studies was consistently <1%.

42

Results

The scatter graph histograms of NR-l neonatal astrocytes were
characteristic of a single, uniform, proliferative cell population following
culture in complete serum-supplemented medium or when cultured in CDM
(Figure 1, 2A, 3A). Cells cultured in CDM and then treated with NGF (500
ng/ ml) did not markedly differ in their histogram from those cultured
without treatment (Figure 3B). However, NR-l cells cultured in CDM and
then treated with serum (100 111/ ml) were found to generate two parallel

populations of cycling cells (Figure 3C).

The scatter graph histograms of T-9 anaplastic astrocytoma cells were
characteristic of a uniform, proliferative tumor cell population when
cultured in complete serum supplemented medium or when cultured in
CDM (Figure 1, 2B, 4A). Cells cultured in CDM and then treated with serum
(100 111/ ml) gave the same scatter graph histogram characteristic of a cycling-
cell population (Figure 4C). However, T-9 cells cultured in CDM and then
treated with NGF (500 ng/ ml) showed two subpopulations of cells, a cycling
and non-cycling population (Figure 4B).

43

Figure 1 Schematic illustration of a fluorochromatic scatter graph
histogram of proliferative and quiescent pools of cycling and non-
cycling cell populations, following fixation and A0 staining. The Y axis
represents dsDNA and the X axis ssDNA. Cycling cell populations
have scatter graph histograms characterized by Go-Gl, S and G2+M
components of the cell cycle. This is illustrated by the sequence of
subpopulations making up the proliferative pools P and P', on the far
left. Interphase cells of the quiescent pool are non-cycling and
therefore, do not have S or G2+M components. They are illustrated by
the Q subpopulation on the right, which contains only non-cycling Go

and G1 components.

 

44

 

 

 

Ran-Fum- vs amt-Fur)

 

45

Figure 2. Scatter graph histograms of A0 stained cycling cell
populations. NR-I neonatal astrocytes (2A) and T-9 anaplastic
astrocytoma cells (28) were cultured in complete FCS-supplemented

(10%) medium and illustrated proliferative populations of cells..

46

 

 

x I l i j- I 1 I 3 r I

RED-FLCX) VS GRN-FL(Y)

d .
o
O
O 0
co
.
0“
. . .
‘Q 0 h
_o
. O
-0

 

 

 

j I s 3 I 3*: I 1 3'

RED-FLtX) VS HRH-FLtY)

47

Figure 3. Scatter graph histograms of A0 stained NR-I astrocytes. NR-
1 cells were cultured in HLlA CDM with no treatment (3A), following
treatment with NGF (500 ng/ml) (3B) or following treatment with FCS
(100 Ill/ml) (3C). Two parallel proliferative cell populations are
illustrated in figure 3C.

crmaa I ’F'

48

 

 

 

 

- -.;.=-'"-'
o 3',
*s
- .435.
.9
CI .:9 '
' .-J'- °
RED-FLtX) VS SRN-FLtY)
- ..':. B
.. 3:2
- ..°fl-
. . 5.-_
- ,2...

 

_ ~ C
’ O
.0
C . . O
a ‘ ° .
c O . -
q
.‘ ;... o. O.
" on ‘5: o {.3 .’~
- i" I; an“
I s .I s .‘._ 5 . g . . fl

 

.7

RED-FLtX) VS SRN-FL(Y)

49

Figure 4. Scatter graph histogram of AO stained T-9 astrocytoma cells
T-9 cells were cultured in HLIB CDM with no treatment (4A),
following treatment with NGF (500 ng/ml) (4B) or following treatment
with FCS (100 111/1111) (4C). The generation of a non-proliferative Go
interphase cell population, induced by NGF treatment, is illustrated by

formation of a quiescent pool of cells in figure 4B.

 

 

50

 

 

 

I I I I I I , I I

RED-FLOO VS ERR-FLU!)

 

 

 

I I I I I I I I

RED-FLO“ VS SRN-FLCY)

 

 

I I I I I I I I

RED-FL (X) VS ERR-FL (Y)

 

 

 

51

Discussion

Neoplastic transformation is well-recognized as a process involving
initiation and promotion which generates an aberrant proliferating cell
population. The results of this two step-process has been termed
carcinogenesis. Carcinogenesis can be further defined as a cellular generated
escape from growth and organizational control. These cellular changes are
recognized by pathologists as changes in phenotype of individual cells and
tissues. The initiation of this process is accepted to be due to genetic lesions
brought about by oncogenic viruses, chemical carcinogens, inherited mutant
genes as well as spontaneous mutations acquired both pre- and post-natally
(Weinstein, 1988). Neoplastic cells generated in this process simply
proliferate, unrestrained by normal physiologic control mechanisms. Cells
continue to grow unrestrained and invade adjacent tissues. Normal cell
communication, necessary for cellular organization and organogenesis,
appears to be interupted (Pitts et al., 1987; Yamasaki, 1990). As a result of
carcinogenesis, organ systems malfunction, aberrant cells invade and colonize
in inappropriate locations and disease processes ensue (Patterson, 1974;
Weinstein, 1988).

The majority of strategies developed for cancer therapy have involved
cytotoxic agents that kill neoplastic cells by interacting directly with their
DNA. Other agents have been developed that are cytostatic and stop cells in a
specific phase of the cell cycle. The problems involved with either of these
therapeutic regimens are their effects on normal cell populations. Cytotoxic

drugs developed to kill proliferative tumor cell populations also kill normal

 

 

52

proliferative cell populations. Cytostatic drugs that stop proliferative cell
populations in a specific phase of the cell cycle also affect the DNA of normal
cells and are responsible many times for the initiation of drug-induced
carcinogenesis. However, the main priority in the development of any
efficacious anti-neoplastic agent is still to control proliferation.

The measurement of DNA using flow cytometric techniques has made
it possible to distinguish subpopulations of cells in various phases of the cell
cycle (Darzynkiewicz et al., 1979; Dethlesen et al., 1980; Nusse et al., 1990). In
addition, cycling cells of proliferative populations can be differentiated from
non-cycling cells of non-proliferative cell populations (Dethlefsen et al., 1980).
However, problems have existed in differentiating closely related cell
populations. Normal cells of the same karyotype and tumor cells expressing
aneuploidy have made it difficult to distinguish pools of proliferative and
non-proliferative populations as well as their subpopulation components.
However, differences in the sensitivity of DNA to acid denaturation and
staining with A0 have made it possible to differentiate subpopulations of
cycling cells, as well as non-cycling cells (Darzynkiewicz et al., 1979b). Separate
differentiated cell types, though even closely related, are known to express
different genes. Gene expression is related to ssDNA regions of the genome:
Therefore, different cell types can be expected to vary in ssDN A content and
location. As a result, flow cytometry may use the dual flurochromatic
properties of A0 staining to differentiate ssDNA from dsDNA, if RNA is
eliminated. As a result, different cell populations can be quickly and easily
characterized by their ss- and dsDNA content and distinguished from one
another (Figure 1).

Neonatal astrocytes have been reported to have the capacity to

differentiate morphologically into two different end-stage types of astrocytes

mun—s— —.-rv —-

53

(Raff et al., 1983; Levi et al, 1986). Therefore, neonatal astrocytes have been
regarded as a precursor cell population which has the capacity to proliferate in
vitro and to respond to differentiation factors. Astrocytes have been found to
differentiate morphologically to either stellate cells with a low proliferation
capacity or to epithelioid cells which continue to proliferate similar to the
precursor cells (Raff et al., 1983). These two differentiated astrocytes have the
characteristics of fibrous and protoplasmic astrocytes, respectively. Since the
cells isolated for use in this study were astrocyte precursor cells we can expect
them to differentiate to either protoplasmic or fibrous astrocytes (Bailey and
Cushing, 1926). This study has characterized these astrocytic cell types based
on their 55 and dsDN A. Therefore, we speculate that protoplasmic astrocytes,
with a greater proliferation potential, are illustrated by the scatter graph
histogram in Figure 3C as the cell population on the left, while the fibrous
astrocytes are illustrated as the population on the right. As a result, A0 flow
cytometry has the ability to differentiate two closely related cell populations of
the same karyotype, but which differ slightly in gene expression and
proliferation potential.

NGF has been reported to retard the growth rate of in vivo and in vitro
astrocytoma cells (Vinores and Koestner, 1980; Marushige et al., 1987).
Although the physiologic consequences of the NGF-cell interaction is known
with regards to morphologic differentiation of sympathetic neurons, very
little is known about the molecular mechanisms responsible for the
retardation of growth of astrocytoma cells. In this study NGF was not found
to markedly affect the proliferating pool of neonatal astrocytes (Figure 3B).
However, NGF was found to induce a quiescent pool of astrocytoma cells
(Figure 4B). Red metachromatic fluorescence of AO-ssDNA complexes has

been reported to be highest in quiescent cells, while the orthochromatic green

54

fluorescence of dsDNA complexes has been found to be greatest in cycling cell
populations (Darzynkiewicz et al., 1978). This suggests that NGF may
preferentially act to increase the sensitivity of chromatin DNA in astrocytoma
cells to acid-denaturation. Therefore, A0 flow cytometry has the potential to
differentiate a proliferative from a non-proliferative tumor cell population.

The significance of this study is that NGF reduces the growth potential
of astrocytoma cells while not affecting neonatal astrocyte populations in
vitro. These results therefore support the potential use of NGF as a

therapeutic agent for the treatment of anaplastic astrocytomas.

55

References

Arndt-Jovin, D. J., and Jovin, T. M. (1977) Analysis and sorting of living cells
according to deoxyribonucleic acid content. ]. Histochem. Cytochem.
25, 585-589.

Bailey, P. and Cushing, H. (1926) A Classification of Tumors of the Glioma
Group. Lippincott, Philadelphia.

Crissman, H. A., Darzynkiewicz, Z., Steinkamp, J. A., and Tobey, R. A.
(1990) Simultaneous fluorescent labeling of DNA, RNA, and protein.
In, Methods in Cell Biology (Darzynkiewicz, Z. and Crissman, H. A.,
eds.), Vol. 33, pp. 305-314, Academic Press, NY.

Darzynkiewciz, Z., Evenson, D. P., Staiano-Coico, L., Sharpless, T. K., and
Melamed, M. L. (1979a) Correlation between cell cycle duration and
RNA content. 1. Cell. Physiol. 100, 425-438.

Darzynkiewicz, Z., Traganos, F, Sharpless, T., and Melamed, M. R. (1976)
Subcompartments of the G1 phase of the cell cycle detected by flow

cytometry. Proc. Natl. Acad. Sci. USA 73, 2881-2884.

Darzynkiewicz,‘Z., Traganos, F., and Kimmel, M. (1986) Assay of cell cycle
kinetics by multivariate flow cytometry using the principle of
stathmokinesis. In, Techniques in Cell Cycle Analysis (Gray, J. W. and
Darzynkiewicz, Z, eds.), pp. 298-301, Humana Press, Clifton, NJ.

Darzynkiewicz, Z., Traganos, F., Sharpless, T., and Melamed, M. R. (1977)
Recognition of cells in mitosis by flow cytometry J. Histochem.
Cytochem. 25, 875-880.

5 6
Darzynkiewicz, Z., Trangos, F., and Melamed, M. R. (1980) New cell cycle
compartments identified by multiparameter flow cytometry.
Cytometry 1, 98-101.

Darzynkiewicz, Z., Trangos, F., Andreeff, M., Sharpless, T, and Melamed
(1979b) Different sensitivity of chromatin to acid denaturation in
quiescent and cycling cells as revealed by flow cytometry. J. Histochem.
Cytochem. 27, 478-485.

Dethlefsen, L. A., Bauer, K. D., and Riley, R. M. (1980) Analytical cytometric
approaches to heterogenous cell populations in solid tumors: A
review. Cytometry 1, 89-97.

Fidler, I. J., and Kripke, M. L. (1977) Metastasis results from preexisting
variant cells within a malignant tumor. Science 197, 893-895.

Hankansson, L., and Trope, C. (1974) On the presence within tumours of
clones that differ in sensitivity to cytostatic drugs. Acta Pathol.
Microbiol. Scand. 82, 35-40.

Hefti, F. (1986) Nerve growth factor promotes survival of septal cholinergic
neurons after fimbrial transections. ]. Neurosci. 6, 2155-2162.

Levi,G., Gallo, V., and Ciotti, M. T. (1986) Bipotential precursors of putative
fibrous astrocytes and oligodendrocytes in rat cerebellar cultures express
distinct surface features and "neuron-like" 7-aminobutyric acid
transport. Proc. Natl. Acad. Sci. USA 83, 1504-1508.

Levi-Montalcini, R. and Angeletti, P. (1968) Nerve growth factor. Physiol.
Rev. 48, 534-564.

Marushige, Y., Marushige, K. and Koestner, A. (1989) Chemical control of
growth and morphological characteristics of anaplastic glioma cells.
Anticancer Res. 9, 1143-1148.

57

Marushige, Y., Raju, N. R., Marushige, K., and Koestner, A. (1987)
Modulation of growth and of morphological characteristics in glioma
cells by nerve growth factor and glia maturation factor. Cancer Res. 47,
4109-4115.

Mendelson, M. L. (1962) Chronic infusion of tritiated thymidine into mice
with tumors. Science 135, 213-215.

Morrison, R. S., and De Vellis, J. (1981) Growth of purified astrocytes in a
chemically defined medium. Proc. Natl. Acad. Sci. USA 78, 7204-7209.

N usse, M., Beisker, W., Hoffmann, C. and Tarnok, A. (1990) Flow cytometric
analysis of G1- and Gz/M-phase subpopulations in mammalian cell

nuclei using side scatter and DNA content measurements. Cytometry
11, 813-821.

Patterson, M., Jr. (1974) Alterations in membrane structure and function
associated with neoplastic transformation in vitro. J. Natl. Cancer Inst.
53, 1493-1498.

Pimm, M. V., and Baldwin, R. W. (1977) Antigenic differences between
primary methylcholanthrene-induced rat sarcomas and post-surgical
recurrences.Int. ]. Cancer 20, 37-43.

Pitts, J., Kam, E. and Morgan, D. (1987) The role of junctional
communication in cellular growth control and tumorigenesis. In, Gap
Junctions (Hertzberg, E. and Johnson, R., eds.), pp. 397-409, Alan R.
Liss, Inc., New York.

Raff, M. C., Miller, R. H., and Noble, M. (1983) A glial progenitor cell that
develops in vitro into an astrocyte or an oligodendrocyte depending on
culture medium. Nature 303, 390-396.

Schmidek, H. H., Nielsen, S. L., Schiller, A. L., and Messer, J. (1971)
Morphological studies of rat brain tumors induced by N-nitrosourea. J.
Neurosurg. 34, 335-340.

58

Steel, G. G. (1977) Growth Kinetics of Tumors, pp. 144-216, Clarendon Press,
Oxford.

Traganos, F., Darzynkiewicz, Z., Sharpless, T., and Melamed, M. R. (1977)
Simultaneous staining of ribonucleic and deoxyribonucleic acids in
unfixed cells using acridine orange in a flow cytofluorometric system.
J. Histochem. Cytochem. 25, 46-56.

Vinores, S. A., and Koestner, A. (1980) The effect of nerve growth factor on
undifferentiated glioma cells. Cancer Letts. 10, 309-318.

Weinstein, B. (1988) The origins of human cancer: Molecular mechanisms of
carcinogenesis and their implications for cancer prevention and
treatment. Cancer Res. 48, 4135-4143.

Williams, L., Varon, 5., Peterson, G., Wictorin, K., Fischer, W., Bjorklund, A.
and Gage, F. (1986) Continuous infusion of nerve growth factor
prevents basal forebrain neuronal death after fimbria fornix
transection. Proc. Natl. Acad. Sci. USA 83, 9231-9235.

Yamasaki, H. (1990) Gap junctional intercellular communication and
carcinogenesis. Carcinogenesis 11, 1051-1058.

Chapter 3

Induction of Morphologic Differentiation in Astrocytes and
Astrocytoma Cells by Nerve Growth Factor.

William R. Hare Jr.

Department of Pathology
Michigan State University
East Lansing, MI 48824-1316

Telephone, 517-353-91 60

Telefax no. 517-336-1053

59

60

Abstract

An in vitro effect of nerve growth factor (NGF) on end-stage
morphologic differentiation of neonatal astrocytes and anaplastic astrocytoma
cells of the rat is reported. NGF (100 ng/ ml) treatment of cell cultures in the
presence of 25uM glutamate (GLU) results in morphologic differentiation of
cells resembling protoplasmic and fibrous astrocytes. These results are
markedly different than previous reports describing cells treated with NGF (5
ug/ ml) under much different conditions. Treatment of astrocytoma cells
with either NGF (500 ng/ ml) or 25 uM GLU alone did not induce morphologic
differentiation. This report concerns the potential use of NGF as a

pharmacologic agent for the differentiation and arrest of tumor growth.

Running Title: The effect of N GF on astrocyte morphology

Key Words: nerve growth factor, glutamate, astrocytes, astrocytoma cells

chemically defined medium.

61

Introduction

Astrocytes are classified on the basis of their morphology as either
epithelioid protoplasmic (Type 1) or stellate fibrous (Type 2) astrocytes. Most
isolates of neonatal astrocytes from brain tissue have been found to contain a
mixture of these two cell types. Most anaplastic astrocytoma cells lose all
morphologic identity with astrocytes and in culture rapidly take on a
monotonous spindle-like shape forming a confluent sheet of cells.

Nerve growth factor (NGF) is a widely recognized potent
differentiation promoter of sympathetic and sensory neuronal cell
populations as well as pheochromocytoma and neuroblastoma cells (Greene
and Tischler, 1976; Reynolds and Perez-Polo, 1981; Sonnefield and Ishii, 1982).
In addition, NGF has been characterized as an important factor in the
maintenance and survival of neuronal cell populations (Whittemore and
Seiger, 1987; Thonen et al., 1987; Springer, 1988). NGF and other maturation
factors have also been found to induce morphologic differentiation in
glioblasts and glioma cells (Lim, 1980; Marushige et al., 1987, 1989). Although
the molecular mechanisms by which NGF promotes neuronal differentiation
are unknown, previous studies have suggested that phosphorylation of
specific proteins occurs during the differentiation process (Hashimoto, 1988;
Koizumi et al., 1988; Hashimoto and Hagino, 1989). The cytoskeleton is
considered to be the major factor responsible for cellular expression of shape.
Variations in cytoskeletal components, therefore, would give rise to
variations in shape, morphologic characteristics and function, since
morphology and function of specific cell populations tend to go hand in

hand.

62

Studies in the past have attempted to reverse the anaplastic character of
neoplastically transformed astrocytes by treatment with various
differentiation promoters (Marushige et al., 1987). These studies were
designed to regain normal end-stage cell differentiation. To obtain this goal
treatment with large doses of promoters was used over a period of 2-4 days.

This study reports on the immediate effects of NGF, at a lower dose, to
induce morphologic differentiation in anaplastic astrocytoma cells cultured in
a chemically defined medium(CDM) in the presence of the excitatory

neurotransmitter glutamate (GLU).

63

Methods

Materials

NGF (2.5S, grade II) was purchased from Boehringer Mannheim
Biochemicals (Indianapolis, IN,USA). Fetal calf serum (FCS) was purchased
from Hyclone Sterile Systems Inc. (Logan, UT, USA). All culture media were
prepared using stock solutions, chemicals and supplies from Life
Technologies, Inc,GIBCO Labs (Grand Island, NE, USA), Corning Glass Works
Inc. (Corning, NY, USA) and Sigma Chemical Co. (St. Louis, MO, USA). Hl-l
supplement was purchased from Endotronics Inc. (Coon Rapids, MN, USA)
and rats from Charles Rivers Labs. (Portage, MI, USA). L-Glutamate (GLU)
was purchased from lCN-Biomedicals Inc. (Costa Mesa, CA, USA).

Neonatal Rat Astrocytes

Neonatal rat astrocytes were generated using 4-day-old Fischer rat pups.
The brains of four rat pups were removed following euthanasia and placed in
a petri dish containing warm DMEM high glucose media. The meninges
were removed and the brain stem and cerebellum were separated from the
cerebrum. The cerebral tissues were minced into small pieces (<05 m),
placed into a sterile centrifuge tube containing 5 ml of DMEM high-glucose
media. This suspension of brain tissue was then centrifuged at 750 X g for a
period of 10 min. The supernatant fluid was removed and discarded and
replaced with 1.5 ml of Collagenase II-S (Sigma) solution (0.8%, w/v). The
enzyme treatment procedure was followed by incubation of the cell mixture

on a warm water bath (37°C) shaker for 30 min, after which the brain tissue

64

was completely dissociated into a cell suspension by intermittently pipetting
with a sterile glass pasteur pipette supplied with a cotton filter. At no time
during preparation of these cultures were the cells allowed to cool below
30°C. Following the enzyme treatment the cell suspension was centrifuged at
200 X g for a period of 10 min and the supernatant fluid removed and
discarded. An equal amount (approx. 3 ml) of DMEM high-glucose medium
containing 15% FCS was gently layered on top of the cell pellet. Then the
upper 2/ 3 of the cell pellet was resuspended in fresh medium using gentle
pipetting action. A 1 ml aliquot of this cell suspension was quickly removed
and used to seed primary cultures at an approximately a 1:20 split ratio (0.25
ml: 5 ml medium). These stock cultures were started and maintained in 25
cm2 tissue culture flasks (Corning) containing 5 ml of complete medium.
The medium was replaced with fresh nutrient medium the following day
and replaced every 3 days thereafter. Neonatal astrocytes used in these
experiments were at their 3rd passage. The astrocytic character of these cells
was indicated at second and third passages not only by morphology, and their
ability to take up y-aminobutyric acid, but by their content of the astrocyte-
specific cell marker, glial fibrillary acidic protein (GFAP).

Astrocytoma Cells

The rat anaplastic astrocytoma cells were supplied by Dr. A. Koestner of
the Department of Pathology, Michigan State University, East Lansing, MI.
from his cell storage bank. The T-9 cell line originated from a high grade-
anaplastic astrocytoma induced in Fischer rats by treatment with N-methyl-
N-nitrosourea (MNU) [31]. Stock cultures were established in 25 cm2 tissue

culture flasks and maintained in complete serum supplemented medium

65

(DMEM and RPMI 1640/ Hams F12) replacement every 3rd day and the
passage of cells every 6th day at a 1:100 split ratio (1-2 X 103 cells).

Culture Conditions

Cells were cultured in a Hotpack CO2 Incubator which was maintained
at 5% C02, 37°C and constant humidity. Both neonatal astrocytes and
astrocytoma cells were split into multi-well (6 well) culture plates (GIBCO
Labs) containing 3 ml of complete medium. This medium helped to assure
cell attachment. The complete medium, serum supplemented, was changed
the following day and replaced with chemically defined medium (CDM). The
CDM was composed of DMEM supplemented with 1% (v/v) HL-l
supplement, 400 uM glutamine, gentamycin at 10 ug/ ml, glucose at 1 mg/ ml,
CaCl2 at 175 ug/ ml and MgSO4 at 125 ug/ ml. and in addition contained
hydrocortisone at 1.6 u g/ ml, prostaglandin F2-alpha at 440 ng/ ml, putrescine
at 78 ug/ ml, basic-fibroblastic growth factor at 8.8 ng/ ml and myelin basic
protein at 440 ng/ ml. This CDM medium was a slight modification of that
originally proposed by Morrison and De Vellis in 1981 as a CDM that initiates
differentiation of astrocytes. HL-l supplement contains 29 ug/ ml total
protein with 15 ug/ ml insulin and contains no additional growth factors or

glutamate.

Morphologic Effects Study of NGF and GLU

Neonatal astrocytes and astrocytoma cells used in this study were
seeded at 6-12,000 cells/ ml into 3 ml of complete, serum-supplemented,
medium at initial plating. The medium was changed the following day to
serum-free CDM. Three days later the CDM was replaced with fresh CDM and
NGF (500 ng/ ml ) or glutamate (25uM) was added to test the single effects of

66

NGF and GLU. Cells were incubated for 48 hr following treatment and then
photomicrographed. Three or more separate cultures were used as replicates
for each study.

To test the immediate combined effects of NGF and GLU on astrocytes
and astrocytoma cells the preceding protocol was followed with the exception
that there were no treatments until the fifth day of culture in CDM. The cells
were treated with 2511M GLU and incubated for 10 min at room temperature
(20-25°C). The cells were treated with NGF (100 ng/ ml) for 3 min near the
end of the GLU incubation (from the 7 min point of the 10 min GLU
incubation). The medium was then removed and the cells were rinsed in a
HEPES buffered saline pH 7.4, also containing 2511M GLU for 10 min. The
buffer was removed and the cells were acid-fixed with 1M perchloric acid for
10 min. The fixer was removed and the cells were covered with Tris buffer
and photographed in the culture dishes using an inverted Nikon TMS

microscope equipped with a 35 mm Nikon FGW camera.

67

Results

The immediate, combined effects of NGF (100 ng/ ml for 3 min) and
GLU (2511M for 10 min) on morphologic differentiation in cultures of
undifferentiated neonatal rat astrocytes and rat anaplastic astrocytoma cells
are illustrated in Figures 1A and 1B, respectively. This combined treatment
resulted in these cells differentiating towards end-stage astrocytes exhibiting
morphologic characteristics of protoplasmic and fibrous astrocytes. The
morphologic characteristics of astrocytoma cells and neonatal rat astrocytes
cultured in CDM are illustrated in Figure 2A and 3A, respectively. In
addition the effects of treatment on these individual cell types with GLU
(2511M for 48 hrs) or NGF (500 ng/ ml for 48 hrs) alone cultured in CDM are
illustrated in Figures 2B, 2C and 3B, 3C, respectively. Neither N GF nor GLU
treatment alone markedly affected morphologic differentiation by
astrocytoma cells. However, there was a difference in the response of
neonatal astrocytes to GLU and NGF. In response to NGF neonatal astrocytes
took on greater epithelioid differentiation, this was in contrast to their
response to GLU, where cells appeared to contract and exhibit a condensed
cytoplasm and nucleus. The plasma membrane and golgi apparatus of the
astrocytoma cells were also quite noticeable in the acid-fixed cell preparations

(Figure 13).

68

Figure 1. Morphologic differentiation of astrocytes and astrocytoma
cells by NGF in vitro. Morphologic differentiation of neonatal rat
astrocytes (1A) cultured in CDM and treated with NGF/GLU (100
ng/ ml 8: 25uM). Solid arrowhead identifies protoplasmic type
astrocyte and open arrowhead identifies fibrous type astrocyte.
Morphologic differentiation of rat anaplastic astrocytoma cells (1B)
cultured in CDM and treated with NGF/GLU (100 ng/ ml 8: 25 uM).
Solid arrowhead identifies protoplasmic type astrocyte and open
arrowhead identifies fibrous type astrocyte. Arrows identify the
carbohydrates, confirmed by PAS staining, of the plasma membrane

and golgi apparatus. Bar in lower right represents 50 um.

 

69

 

70

Figure 2. Morphology of rat anaplastic astrocytoma cells cultured in
CDM. Morphology of rat anaplastic astrocytoma cells cultured in CDM
with no additional treatments (2A), treated with GLU (2511M). (28) or
treated with NGF (500 ng/ml) (2C).

Bar in lower right represents 50 um.

 

 

72

Figure 3. Morphology of neonatal rat astrocytes cultured in CDM.
Morphology of neonatal rat astrocytes cultured in CDM with no
additional treatments (3A), treated with GLU (2511M). (3B) or treated
with N GP (500 ng/ml) (3C).

Bar in lower right represents 50 um.

73

 

74

Discussion

The results presented in this report demonstrate the importance of
early epigenetic events in the determination of morphologic differentiation
by astrocytes. Past reports on morphologic differentiation by these types of
cells have focused on cell density and changes associated with the cellular
content of cytoskeletal proteins (Goldman and Chiu, 1982; Goldman and
Chiu, 1984; Marushige et al., 1989). Regulation of cell morphology and
differentiation have likewise been reported to be associated particularly with
the effects of cyclic AMP. There are many studies describing changes in cell
shape following treatment with dibutyryl cyclic AMP. Many of these studies
have suggested that cyclic AMP is a mediator for the morphologic effects
produced by NGF (Schubert et al., 1978; Garrels and Schubert, 1979; Halegoua
and Patrick, 1980; Cremins et al., 1986). However, most of the alterations in
morphology that have been described in previous studies appear to be most
pronounced in low density cultures where cells are seen to change from a flat
polygonal shape to cells exhibiting a rounded shape with condensed
cytoplasm and many fine cytoplasmic processes (Moonen et al., 1975;
Trimmer et al., 1982). Cells in low density cultures have also been shown to
express a high content of cytoskeletal proteins (Goldman and Chiu, 1984).
This increased production of cytoskeletal proteins by cells in low density
cultures is therefore suggested to be the cause for the change in morphology
and differentiation seen in these cells.

Astrocytes represent a heterogeneous cell population of the

brain. They are heterogeneous with respect to morphology as well as

75

membrane surface properties. Astrocytes are distributed in the brain relative
to their specific function. There are two basic morphologic types of astrocytes,
one is an epithelioid protoplasmic astrocyte (Type 1) while the other is a
stellate-appearing fibrous astrocyte (Type 2) (Raff et al., 1983).

Primary astrocytes in vitro have also been characterized as being a
heterogeneous cell population (Steig et al., 1980; Trimmer et al., 1982).
Questions have always been asked as to whether this heterogeneity resulted
from an intrinsic heterogeneity of astrocytes during initial isolation or was
simply the result of in vitro conditions. Our results suggest that astrocytes
and astrocytoma cells in vitro both exhibit a mixture of two cell types with
morphologic characteristics of protoplasmic and fibrous astrocytes. These cell
types are assumed to be accurate based on previous reports in the literature on
precursor cell populations and developmental sequence (Bailey and Cushing,
1926; Raff et al., 1983). However, the proportion of one morphologic cell type
to the other may be influenced by cell density. In addition, protoplasmic
astrocytes have been reported to respond to growth factors and cyclic AMP
while fibrous astrocytes have not (Herschman, 1986).

NGF is a potent differentiation promoter of pheochromocytoma cells
(Greene and Tischler, 1976). NGF's effect on differentiation by these cells has
been termed early and late occurring events (Greene and Tischler, 1982;
Greene, 1984). Early events are characterized by their rapid onset and are
many times referred to as immediate cell responses which are
transcriptionally independent. Early events appear to peak around 15 min
following treatment. In contrast to early events, late events are observed
between 24 and 48 hr and are frequently found to be transcriptionally
dependent. Early events include the generation of cyclic AMP,
phosphorylation reactions and phosphoinositide hydrolysis. Late events

76

include neurite outgrowth by pheochromocytoma cells as well as their
development of electrical excitability and the induction of several neuron
specific proteins (Basi et al., 1987; Karns et al., 1987; Pollock et al., 1990). Early
events appear to be epigenetic events shared by many different cell types. Late
events appear to be cell specific events regulated by gene expression. N GF has
been found to induce membrane sodium channels and cyclic AMP has been
found to decrease sodium current. This suggests that NGF may not act
through cyclic AMP as a second messenger. Other activators of cyclic AMP
cascades have also been found to inhibit NGF-induced responses (Greene et
al., 1986; Doherty et al., 1987).

Glutamate (GLU) is an excitatory amino acid neurotransmitter that is
ubiquitous to the central nervous system. Astrocytes function to take up
synaptic GLU following neuronal release. GLU is also a very important
constituent of brain metabolism and plays a significant role in regulating
levels of ammonia in the brain. Astrocytes share two ionotrophic GLU
receptors with neurons (Hosli et al., 1979), and in addition express a GLU
receptor associated with the hydrolysis of membrane phosphoinositides
(Nicoletti et al., 1986a; Nicoletti et al., 1986b; Nicoletti et al., 1987). This
receptor functions in membrane signal transduction. Activation of this
receptor generates 1,4,5-inositol triphosphate (1P3) and calcium (Ca‘H') as
second messengers.

Anaplastic astrocytoma cells cultured in CDM do not differentiate to
end-stage astrocytes (Figure 2-A). The treatment of these cells with GLU
(2511M) or NGF (500 ng/ ml) for 48 hrs also causes no significant change in cell
morphology (Figure 2B, 2C). However, the combined treatment of anaplastic

astrocytoma cells with GLU (25uM) for 10 min and NGF (100 ng/ ml) for three

min induces end-stage morphologic differentiation (Figure 1B). This

77

response is markedly different than earlier reports from our department
where cells were treated with NGF (5 u g/ ml) under much different
conditions (Marushige et al., 1987, 1989). The morphologic response in this
study illustrates individual cell variation in differentiation resembeling both
protoplasmic and fibrous astrocytes. Whereas previous studies describe a
uniform population change characterized by a flattened cytoplasmic
projection, lamellipodia, with long slender filamentous processes, filopodia
(Marushige et al., 1987). The treatment of neonatal astrocytes in the same way
also induces their morphologic differentiation (Figure 1A). These results
suggest that N GF may play a permissive role in astrocytes which facilitates the
regulation of differentiation and allows astrocytes and astrocytoma cells to
express end-stage morphologic differentiation in response to other epigenetic
events. In this experimental model these events are represented by the
membrane signal transduction initiated by GLU. Generation of IPB and Ca++
through activation of the GLU receptor could in turn lead to activation of
other protein kinases and the phosphorylation of specific proteins which may
in turn control the organization of the cytoskeletal system. In addition, NGF
may also act to sustain this morphologic differentiation. That is, morphologic
differentiation may not only not take place without NGF but cells may revert
back to their pre-undifferentiated morphology with out NGF's continued
presence. However, the earlier reports from our department indicated that
prolonged treatment with NGF at higher doses, over 4 days, under different
conditions induced the characteristic change in cell morphology previously
described and that this morphologic change appeared to persist even in the
absence of NGF (Marushige et al., 1987).

In contrast to protein phosphorylation, which directly influences

protein function and organization, glycosylation often determines specificity

78

of functional proteins and regulates their turnover and organizational
distribution within the cell. In the past much less attention has been paid to
the functional significance of glycosylation than to the effects of
phosphorylation. One basic type of change in the expression of membrane
carbohydrates and glycosylation, which takes place following neoplastic
transformation, has been shown to be affected by NGF (McGuire and Greene,
1978). In many tumor cell populations high molecular weight fucose-
containing glycoproteins appear to increase (Warren et al., 1973). The
treatment of pheochromocytoma cells with NGF has been found to increase
the incorporation of fucose and glucosamines (McGuire and Greene, 1978).
These reports suggests that NGF may have an effect on glycosylation and
differentiation which is quite different from its effect to facilitate morphologic
differentiation through phosphorylation. NGF may stimulate imperfect
glycosylation of membrane proteins through a cellular mechanism which is
non-specific, since ganglioside composition of neurons and astrocytes are
different (Geissler et al., 1977). This further suggests that morphologic
differentiation may be modulated through two completely different
mechanisms in which NGF could impart both a positive and negative effect.

The acid fixation used to prepare the NGF/GLU combination treatment
of astrocytes and anaplastic astrocytoma cells identifies what may be the large
complex carbohydrates contained in the plasma membrane and the golgi
apparatus of the astrocytoma cells (Figure 1B). The carbohydrates were also
confirmed in the astrocytoma cells by using a PAS stain. However, these
carbohydrates could not be identified by acid fixation in the normal neonatal
astrocytes (Figure 1A).

The significance of this study is that NGF/GLU combination treatment

of astrocytes or astrocytoma cells induces morphologic differentiation towards

79

end-stage prot0plasmic and fibrous astrocytes (Figure 1A, 1B). This response
is suggested to be facilitated by NGF and may be mediated by 1P3 and Ca”
mobilization through activation of GLU-receptor linked hydrolysis of
phosphoinositide. This generation of 1P3 and mobilization of Ca++ further
leads to activation of other protein kinase systems leading to the
phosphorylation of specific proteins controlling cytoskeletal organization. In
addition, it is suggested that NGF may not correct the defect which leads to
aberrant glycosylation of membrane proteins and may actually increase the
incorporation of fucose and glucosamines into large complex membrane
glycoproteins (Figure 1B).

This comparative study of normal and anaplastic cells in vitro has
shown differences and similarities in the induction of morphologic
differentiation by NGF and/ or GLU. The real relevance of such a treatment

regime to in vivo systems remains to be established.

80

References

Basi, G. S., Jacobsen, R. D., Virag, I., Schilling, J. and Skene, J. H. P. (1987)
Primary structure and transcriptional regulation of GAP-43, a protein
associated with nerve growth. Cell 49, 785-791.

Cremins, J., Wagner, J. A., and Halegoua, S. (1986) Nerve growth factor
action is mediated by cyclic AMP and Ca / phospholipid dependent
protein kinase. J. Cell Biol. 103, 887-893.

Doherty, P., Mann, D. A. and Walsh, F. S. (1987) Cholera toxin and
dibutyrl cyclic AMP inhibit the expression of neurofilament protein
induced by nerve growth factor in cultures of naive and primed PC12
cells. J. Neurochem. 49, 1676-1687.

Garrels, J. I. and Schubert, D. (1979) Modulation of protein synthesis by
nerve growth factor. J. Biol. Chem. 254, 7978-7985.

Geissler, D., Martinek, A., Margolis, R. U., Margolis, R. K., Skrivanek, J. A.,
Ledeen, R., Konig, P. and Winkler, H. (1977) Composition of complex
carbohydrates of ox adrenal chromaffin granules. Neuroscience 2, 685-
693.

Goldman, J. E. and Chiu, F.-C. (1982) Growth kinetics, cell shape, and the
cytoskeleton of cultured astrocytes. Trans. Am. Soc. Neurochem.
13, 111-117.

Goldman, J. E. and Chiu, F.-C. (1984) Growth kinetics, cell shape, and the
cytoskeleton of primary astrocyte cultures. J. Neurochem. 42, 175-184.

Greene, L. A. (1984) The importance of both early and delayed-responses
in the biological action of nerve growth factor. Trends Neurosci. 7, 91-
94.

81

.Greene, L. A. and Tischler, A. S. (1976) Establishment of a
noradrenergic clonal line of rat pheochromocytoma cells which
respond to nerve growth factor. Proc. Natl. Acad. Sci. USA 73, 2424-
2428.

Greene, L. A. and Tischler, A. S. (1982) PC12 cultures in neurobiological
research. Adv. Cell Neurobiol. 3, 373-414.

Greene, L. A., Drexler, S. A., Connolly, J. L., Rukenstein, A. and Green, S.
H. (1986) Selective inhibition of responses to nerve growth factor and
of microtubule-associated protein phosphorylation by activation of
adenylate cyclase. J. Cell Biol. 103, 1967-1978.

Halegoua, S. and Patrick, J. (1980) Nerve growth factor mediates
phosphorylation of specific proteins. Cell 22, 571-581.

Hashimoto, S. (1988) K252a, a potent protein kinase inhibitor, blocks nerve
growth factor-induced neurite outgrowth and changes in the
phosphorylation of proteins in PC12h cells. J. Cell Biol. 107, 1531-1539.

Hashimoto, S. and Hagino, A. (1989) Blockage of nerve growth factor action
in PC12h cells by staurosporine, a potent protein kinase inhibitor. J.
Neurochem. 53, 1675-1685.

Herschman, H. R. (1986) Polypeptide growth factors and the CNS.
Trends Neurosci. 9, 53-57.

Hosli, L., Anders, P. F. and Hosli, E. (1979) Action of amino acid
transmitters on cultivated cells of the mammalian peripheral and
central nervous system. J. Physiol (Paris) 75, 655-659.

Karns, L. R., N g, S. C., Freeman, J. A. and Fishman, M. (1987) Cloning of
complementary DNA for GAP-43, a neuron growth-related protein.
Science 236, 597-600.

82

Koizumi, S., Contreras, M. L., Matsuda, Y., Hama, T., Lazarovici, P. and
Guroff, G. (1988) K-252a: a specific inhibitor of the action of nerve
growth factor on PC12 cells. J. Neurosci. 8, 715-721.

Lim, R. (1980) Glia maturation factor, Curr. Top. Dev. Biol. 16, 305-322.

Marushige, Y., Marushige, K., Okazaki, D. L. and Koestner, A. (1989)
Cytoskeletal reorganization induced by nerve growth factor and glial

maturation factor in anaplastic glioma cells. Anticancer Res. 9, 1143-
1148.

Marushige, Y., Raju, N. R., Marushige, K. and Koestner, A. (1987)
Modulation of growth and morphological characteristics in glioma

cells by nerve growth factor and glia maturation factor. Cancer Res.
47, 4109-4115.

McGuire, J. C. and Greene, L. A. (1978) NGF stimulates incorporation of
fucose or glucosamine into an external glycoprotein in cultured rat
PC12 pheochromocytoma cells. Cell 15, 357-365.

Moonen, G., Cam, Y., Sensenbrenner, M. and Mandel, P. (1975)
Variabilityof the effects of serum-free medium, dibutyrl-cyclic AMP or
theophylline on the morphology of cultured new-born astrocytes. Cell
Tissue Res. 163, 365-372.

Morrison, R. S. and De Vellis, J. (1981) Growth of purified astrocytes in a
chemically defined medium. Proc. Natl. Acad. Sci. USA. 78, 7205-7209.

N icoletti, F., Iadarola, M. J., Wroblewski, J. T. and Costa, E. (1986)
Excitatory amino acid recognition sites coupled with inositol

phospholipid metabolism: developmental changes and interaction
with ac 1-adrenoreceptors. Proc. Natl. Acad. Sci. USA 83, 1931-1935.

 

83

Nicoletti, F., Meek, J. L., Iadarola, M. J., Chuang, D. M., Roth, B. L. and Costa,
E. (1986) Coupling of inositol phospholipid metabolism with
excitatory amino acid recognition sites in rat hippocampus. J.
Neurochem. 46, 40-46.

Nicoletti, F., Wroblewski, J. T., Alho, H., Eva, C., Fadda, E. and Costa, E.
(1987) Lesions of putative glutaminergic pathways potentiate the
increase of inositol phospholipid hydrolysis elicited by excitatory
amino acids. Brain Research 436, 103-112.

Pollock, J. D., Krempin, M. and Rudy, B. (1990) Differential effects of NGF,
FGF, EGF, CAMP, and dexamethasone on neurite outgrowth and
sodium channel expression in PC12 cells. J. Neurosci. 10, 2626-2637.

Raff, M. C., Miller, R. H. and Noble, M. (1983) A glial cell progenitor cell
that develops in vitro into an astrocyte or an oligodendrocyte
depending on culture medium. Nature 303, 390-396.

Reynolds, C. P. and Perez-Polo, J. R. (1981) Induction of neurite
outgrowth in the IMR 32 human neuroblastoma cell line by nerve
growth factor. ] Neurosci. Res. 6, 319-325.

Schmidek, H. H., Nielsen, S. L., Schiller, A. L. and Messer, J. (1971)
Morphologic studies of rat brain tumors induced by N-
nitrosomethylurea. J. Neurosurg. 34, 335-340.

Schubert, D., La Corbiere, M., Whitlock, C. and Stallcup, W. (1978)
Alterations in the surface properties of cells responsive to nerve
growth factor. Nature 273, 718-723.

Sonnefield, K. H. and Ishii, D. N. (1982) Nerve growth factor effects and
receptors in cultured human neuroblastoma cell lines. J. Neurosci.
Res. 8, 375-391.

Springer, J. E. (1988) Nerve growth factor receptors in the central nervous
system. Exp. Neurol 102, 354-365.

84

Steig, P. E., Kimelberg, H. K., Mazurkiewicz, J. E. and Banker, G. A. (1980)
Distribution of glial fibrillary acidic protein and fibronectin in primary
astroglial cultures from rat brain. Brain Res. 199, 493-500.

Thonen, H., Brandtlow, C. and Heumann, R. (1987) The physiological
function of nerve growth factor in the central nervous system:
comparison with the periphery. Rev. Physiol. Biochem. Pharmacol.
109, 146-178.

Trimmer, P. A., Reier, P. J., 0h, T. H. and Eng, L. F. (1982) An
ultrastructural and immunocytochemical study of astrocytic
differentiation in vitro. ]. Neuroimmunol. 2, 235-260.

Warren, L., Fuhrer, J. P. and Buck, C. A. (1973) Surface glycoproteins of
cells before and after transformation by oncogenic viruses. Fed. Proc.
32, 80-85.

Whittemore, S. R. and Seiger, A. (1987) The expression, localization and
functional significance of beta-nerve growth factor in the central
nervous system. Brain Res. Rev. 12, 439-464.

Chapter 4

The in vitro effects of nerve growth factor on the detection of
intermediate filaments expressing glial fibrillary acidic protein
epitopes by rat anaplastic astrocytoma cells.

William R. Hare Jr.

Department of Pathology
Michigan State University
East Lansing, MI 48824-1316

Telephone, 51 7-353 -91 60

Telefax no. 517-336-1053

85

86

Abstract

Intermediate filament (IF) proteins of neonatal rat astrocytes are shown
to express glial fibrillary acidic protein (GFAP) and vimentin epitopes detected
by immunocytochemical staining. Stellate astrocytes showed greater staining
intensity than epithelioid astrocytes. Nerve growth factor (NGF) (500 ng/ ml)
increased the intensity of GFAP staining by neonatal astrocytes following
treatment for 48 hrs. NGF (12.5 ng/ ml) also induced IFs showing GFAP
epitope expression in anaplastic astrocytoma cells. This GFAP expression was
fixation-sensitive. Paraformaldehyde fixation preserved GFAP epitopes while
alcohol fixation abolished GFAP epitopes expressed by anaplastic astrocytoma
cells. GFAP and vimentin IFs were found to be perinuclear, nuclear and
nucleolar in distribution. The significance of these findings is that
alkylnitrosourea chemical-induced mutagenesis and carcinogenesis of
astrocytes may be responsible for changes in IF protein. These changes are
suspected to be due to changes in gene structure which may be further
characterized by changes in IF protein solubility and expression of GFAP

epitopes.

Running Title: NGF induced expression of GFAP by astrocytoma cells.

Key Words: NGF, GFAP, intermediate filaments, astrocytoma cells,

immunocytochemistry

87

Introduction

Intermediate filaments (IFs) along with microtubules and actin
microfilaments constitute the three major types of filamentous proteins of
cells. IFs are recognized by their intermediate size of 7-11 nm in diameter,
which is intermediate to the 4-6 nm actin microfilaments and the 22-25 nm
microtubules. IFs are also tissue specific and currently are divided into five
major groups, which include glial fibrillary acidic protein (GFAP) and
vimentin. Specific IFs can be detected in cell populations by the use of labeled
monclonal antibodies. Therefore, knowledge of the specific type of IF
expressed by a cell or cell population could result in a determination of the
specific cell type.

Astrocytes are known to express two types of IFs, GFAP and vimentin
(Eng et al, 1971; Bignami and Dahl, 1974; Liem et al, 1978). Vimentin is
expressed by astrocytes and a variety of mesenchymal cell types such as
fibroblasts, macrophages and endothelial cells. Vimentin has also been
reported to be preferentially expressed by a variety of cell types maintained in
tissue culture (Franke et al,1978, 1982; Pateau et al, 1979). Vimentin, therefore,
is not considered to be an astrocyte-specific IF. Its expression and distribution
have been found to decrease during maturation (Pixley et al, 1984). It is
basically expressed by embryonic non-differentiated astrocytes rather than
mature, differentiated astrocytes (Dahl et al, 1981). There is also a vimentin-
GFAP transition which occurs during astrocyte differentiation. This
transition is initiated at the time of myelination (Dahl, 1981; Yokoyama et al,
1981). Therefore, vimentin IFs are expected to be found in proliferating

populations of astrocytes which in turn are found in greatest numbers in the

88

periventricular germinal layers and the non-myelinated areas of the white
matter in the brain.

GFAP is an astrocyte-specific IF which is expressed by astrocytes from
the early post-natal period of development through end-stage differentiation
(Bock, 1978; Eng and Bigbee, 1978; Bignami et al, 1980; Eng, 1985). GFAP
expression is greatest in fibrous astrocytes of the white matter rather than in
‘ protoplasmic astrocytes of the gray matter (Raff et al,1983). Extensive
expression of GFAP also occurs during the process of reactive astrogliosis.

Astrocytes are easily activated by a number of conditions of the brain,
including physical and chemical injury, infectious diseases and
immunological responses. Astrogliosis and increased expression of GFAP are
also found to occur in cases of brain tumors, being most prominent in
astrocytomas (Russell and Rubenstein, 1989). However, experimentally
produced astrocytomas of the rat induced by methylnitrosourea (MNU)
rapidly lose their ability to express GFAP with increased passage when studied
in cell culture.

Since the discovery of IFs in the 1960's they have been suspected to be a
constituent of the cytoskeleton (Biberfield et al, 1965; Sternlieb, 1965; Ishikawa
et al, 1968). This has been supported by their cellular distribution and their
high degree of insolubility under normal physiologic conditions. However,
more recent studies have found lFs to be associated with membrane-bound
sub-cellular structures and enzyme systems (Lin and Feramisco, 1981;
Klymkowsky et al, 1983). In addition, IFs have been found to bind to DNA
and RNA. These new reports suggest that IFs may not function merely as
mechanical integrators of intracellular space, but may be important regulators
of nuclear and membrane processes (Lazarides, 1980, 1982; Franke et al, 1982;

Osborn et al, 1982).

89

Fixation methods have been shown to play important roles in
immunocytochemistry (Sternberger, 1979; Polak and VanNoorden, 1983;
DeArmond and Eng, 1984). Different fixatives have been implicated in the
preferential immunohistochemical-staining of astrocytes for GFAP IFs
(Shehab et al, 1990). Alcohol fixation caused the preferential staining of GFAP
in fibrous astrocytes, whereas paraformaldehyde fixation caused the
preferential staining of GFAP in protoplasmic astrocytes. These reports
suggest that treatment of astrocytoma cells with different fixatives may affect
the detection of GFAP by immunocytochemical staining techniques.

Growth factors which act as differentiation promoters of astrocytes
have long been suspected to induce characteristic changes in cytoskeletal
proteins. Nerve growth factor (NGF) is a widely recognized differentiation
promoter that has been found to induce morphologic differentiation in
glioma cells (Marushige et al, 1987; Marushige et al, 1989). Although the
molecular mechanisms by which NGF promotes cell differentiation remain
largely unknown; previous studies in our department using different culture
and immunostaining techniques have shown that NGF-induced
differentiation of astrocytoma cells is accompanied by an extension of
cytoskeletal proteins (Marushige et al, 1989). In addition, NGF was not found
to induce GFAP expression. However, other filaments as well as
microtubules were found to gradually extend and fill cytoplasmic processes
during morphologic differentiation induced by NGF. In these studies the
cytoskeleton was considered to be the major factor responsible for the cell's
expression of shape and differentiation. Therefore, variations in cytoskeletal
components could give rise to variations in shape, morphologic
characteristics and function, since morphology and function of specific cell

populations tend to go hand in hand.

90

The purpose for this study is to determine if NGF induces rat anaplastic
astrocytoma cells to express IFs which can be detected by GFAP monoclonal
antibodies and to determine if differences in fixatives affect the detection of
GFAP IFs by immunocytochemical staining. In addition, this report also
addresses IF function and why MNU experimentally-induced astrocytoma

cells may or may not express GFAP IFs.

91

Materials and methods

Materials

NGF (2.5S, grades I and II), anti-vimentin (mouse) and anti-GFAP
(mouse) were purchased from Boehringer-Mannheim Biochemicals
(Indianapolis, IN,USA). Fetal calf serum (FCS) was purchased from Hyclone
Sterile Systems Inc. (Logan, UT, USA). All culture media were prepared
using stock solutions, chemicals and supplies from Life Technologies Inc.
GIBCO Labs (Grand Island, NE, USA), Corning Glass Works Inc. (Corning,
NY, USA) and Sigma Chemical Co. (St. Louis, MO, USA). Hl-1 supplement
and concentrate were purchased from Endotronics Inc. (Coon Rapids, MN,
USA). Anti-mouse-FIT C (goat), mouse serum and L-glutamate were

purchased from ICN-Biomedicals Inc. (Costa Mesa, CA, USA).

Neonatal astrocyte cultures

Neonatal rat astrocytes (NR-1) were generated using 4-day-old Fischer
rat pups. The brains of four rat pups were removed following euthanasia and
placed in a petri dish containing warm DMEM high-glucose media. The
meninges were removed and the brain stem and cerebellum separated. The
cerebral tissues were minced into small pieces (<05 m), combined and
placed into a sterile centrifuge tube containing 5 ml of DMEM high-glucose
media. This suspension of brain tissue was then centrifuged at 750 X g for a
period of 10 min. The supernatant fluid was removed and discarded and
replaced with 1.5 m1 of Collagenase II-S (Sigma) solution (0.8%, w/v). The

enzyme treatment procedure was followed by incubating the cells on a warm

92

water bath (37°C) shaker for 30 min, after which the brain tissue was
completely dissociated to a cell suspension by intermittently pipetting with a
sterile glass pasteur pipette supplied with a cotton filter. At no time during
preparation of these cultures were the cells allowed to cool below 30°C.
Following the enzyme treatment the cell suspension was centrifuged at 200 X
g for a period of 10 min and the supernatant fluid removed and discarded.
An equal amount (approx. 3 ml) of DMEM high-glucose medium containing
15% FCS was gently layered on top of the cell pellet. Then the upper 2/ 3 of
the cell pellet was resuspended in fresh medium using gentle pipetting
action. A 1 ml aliquot of this cell suspension was quickly removed and used
to seed primary cultures at approximately a 1:20 split ratio. These stock
cultures were started and maintained in 25 cm2 tissue culture flasks (Corning)
containing 5 ml of complete medium. The medium was replaced with fresh
medium the following day and replaced every 3 days thereafter. Neonatal
astrocytes used in these experiments were at their 3rd passage. The astrocytic
character of these cells was indicated at second and third passage not only by
morphology and their ability to take up y-aminobutyric acid, but by their
high content of the astrocyte-specific, GFAP. These cells were 95% GFAP
positive. In addition, 95% of the cells were epithelioid-like in morphology.

Astrocytoma cells

The rat anaplastic astrocytoma cells were supplied by Dr. A. Koestner of
the Department of Pathology, Michigan State University, East Lansing, MI.
from his cell storage bank. The T-9 cell line originated from a high grade-
anaplastic astrocytoma induced in Fischer rats by treatment with N-methyl-
N-nitrosourea (MNU) (Schmidek et al, 1971). Stock cultures of T-9a and T-9b

were established in 25 cm2 tissue culture flasks and maintained in complete

93

serum-supplemented medium (DMEM and RPMI 1640/ Hams F12) with
replacement every 3rd day and the splitting of cells every 6th day at a 1:100
split ratio (1-2 X 103 cells). The difference in the two T-9 cell populations was
that T-9a was 20 passages behind T-9b.

Culture conditions

Cells were cultured in a Hotpack C02 Incubator which was maintained
at 5% C02, 37°C and constant humidity. Both neonatal astrocytes and
astrocytoma cells (T-9a and T-9b) were split into multi-well (6 well) culture
plates (GIBCO Labs) containing 3 ml of complete medium and a polylysine-
coated glass coverslip (1 ug/ ml ddH20). The complete medium, serum
supplemented, was changed the following day and replaced with either of
three chemically defined medium (CDM). CDM HLIA was composed of
DMEM supplemented with 1% (v/v) HL-l supplement, 400 uM glutamine,
gentamycin at 10 ug/ ml, glucose at 1 mg/ ml, CaCl2 at 175 ug/ ml and MgSO4
at 125 ug/ ml. CDM HLIB was also composed of DMEM and contained the
same supplementation as HLIA but also contained hydrocortisone at 1.6
ug/ ml, prostaglandin F2-alpha at 440 ng/ ml, putrescine at 78 u g/ ml, basic-
fibroblastic growth factor at 8.8 ng/ ml and myelin basic protein at 440 ng/ ml.
This HLlB medium was a slight modification of that originally proposed by
Morrison and De Vellis in 1981 as a CDM that initiates differentiation of
astrocytes. HL-1 supplement contained 29 ug/ ml total protein with 15 ug/ ml
insulin and contained no additional growth factors or glutamate. CDM HLIC
was purchased as a complete CDM. It contained 5011M glutamate and was

supplemented with gentamycin at 20 ug/ ml and 400 uM glutamine.

9 4

Cell culture treatment

NR-l, T-9a and T-9b cells used in these studies were seeded at 6-12,000
cells/m1 into 3 ml of complete, serum-supplemented medium at initial
plating. The medium was changed the following day to serum-free CDM.
Three days later the CDM was replaced with fresh CDM of the same type and
NGF or serum added. NGF grade II was added at 500 ng/ ml (NR-1 and T-9a),
NGF grade I at 12.5 ng/ ml (T-9b) or serum added at 10% (300 111/ 3 ml
medium). Medium and treatment of T-9b cells were exchanged for fresh
medium and treatment every 12 hrs; this was repeated 4 times. All cell

cultures were treated with NGF or serum for 48 hrs prior to cell fixation.

Indirect immunofluorescence of vimentin and GFAP

Intracellular GFAP and vimentin IF antigens were detected in neonatal
astrocytes and T-9a anaplastic astrocytoma cells following alcohol fixation in
ice-cold ethanol/ acetone (1:1 ; 80%/100%, v/v) for 5 min. GFAP IFs were
detected in T99 anaplastic astrocytoma cells following paraformaldehyde
fixation in 2% paraformaldehyde prepared in lOOmM cacodylate buffer, pH
7.4. Fixer was removed and cells were immersed in 50% methanol for 3 min
followed by 100% methanol for 3 min. Methanol was removed and cells were
rinsed in phosphate-buffered saline (PBS) (140mM N aCl, 12mM N a2HPO4,
3.5mM N aH2PO4), pH 7.2, for 10 min. The cells on coverslips were overlaid
with GFAP or vimentin primary monoclonal antibody (100 ul) and incubated
for 30 min. The anti-GFAP (mouse) and anti-vimentin (mouse) primary
antibodies (20 & 50 u g/ ml) were each used at 1:300 dilutions in PBS. The
cover slips with cells attached were extensively rinsed 3 times for a total of 20
min with 2 ml PBS, pH 7.2 on a rotary shaker (NR-1 and T-9a) or by manually

swirling (T-9b). After extensive rinsing the cells were overlaid with FITC-

95

labeled goat anti-mouse (1:50) secondary antibody (100 ul) and incubated 30
min. These procedures were all carried out in the dark. After incubation

with secondary antibody the cover slips were rinsed once in 2 ml of PBS, pH

7.2 for 10 min on an orbital shaker (NR-1 and T-9a) or by manually swirling
(T-9b). The cells were then overlaid with normal mouse serum (10%) and
incubated for 30 min. Cover slips were rinsed extensively, 3 times, in PBS, pH
7.2, for a total of 20 min, on an orbital shaker (NR-1 and T-9a) or by manually
swirling (T-9b). Cover slips were mounted on glass slides with permamount.
Cells were viewed with a Leitz phase contrast microscope fitted with
epifluorescent light sources and IFs photomicrographed to show FIT C-labeled
specificity. Controls which never showed immunofluorescence included
substitution with normal mouse serum (10%), omission of primary antibody

or elimination of FIT C-labeled anti-mouse secondary antibody.

96

Results

Neonatal astrocytes cultured in HLlB CDM were found to be vimentin-
positive by indirect immunocyto-staining following alcohol fixation (Figure
1). Cells cultured in HL1B CDM were found to be GFAP-positive by indirect
immunocyto-staining following alcohol fixation (Figure 2A). Cells cultured
in CDM and treated with NGF (500 ng/ ml) were found to increase in
fluorescent intensity for GFAP. Intensity was greatest following N GF
treatment, in cells cultured in HLlB. In addition, stellate cells were found to
show greater intensity for GFAP immunocyto-staining than protoplasmic

type cells (Figure 2B).

T-9a anaplastic astrocytoma cells cultured in HLIA or HLIB CDM were
found to be negative for GFAP immunocyto-staining following alcohol
fixation. There was no response to NGF treatment detectable by
immunocytochemistry. In addition, these cells, cultured under these
conditions and fixed in alcohol, were found to show an extremely weak
fluorescence to FITC-labeled secondary antibody following exposure to anti-

vimentin primary monoclonal antibodies.

T-9b anaplastic astrocytoma cells cultured in HLlC CDM were found to

be negative for GFAP immunocyto-staining following paraformaldehyde

fixation. However, cells cultured in HLlC and treated with NGF (12.5 ng/ ml)

97

under different conditions than T-9a anaplastic astrocytoma cells were found
to be GFAP positive after fixation in paraformaldehyde and exposure to goat
anti-mouse-FITC labeled secondary antibody following exposure to mouse
anti-GFAP primary monoclonal antibody (Figure 4).

This treatment with NGF also induced noticeable changes in
mophologic characteristics (Figure 3) which were similar to the morphologic

changes noted in Chapter 3 (Figure 1B).

98

Figure 1. Vimentin indirect immunofluorescent staining of neonatal
rat astrocytes. Cells were cultured in HLIB CDM and treated with NGF
(500 ng/ml) using goat anti-mouse-FITC labeled secondary antibody
and mouse anti-vimentin primary antibody following alcohol fixation.
Notice the diffuse staining of the cytoplasmic regions and the intense

staining of the perinuclear, nuclear and nucleolar regions. X 1225

Figure 2. GFAP indirect immunofluorescent staining of neonatal rat
astrocytes . NR-l cells were cultured in HLlB CDM with no treatment
(2A) and treated with NGF (500 ng/ml) (28). Cells were prepared for
GFAP detection using goat anti-mouse-FITC labeled secondary antibody
and mouse anti-GFAP primary antibody following alcohol fixation.
Notice the intense staining of the perinuclear, nuclear and nucleolar
regions (2A) and the intense staining of the fibrous type astrocyte and
the lack of intensity to its nucleolar region. The protoplasmic type
astrocyte shows diffuse staining of the cytoplasmic regions and greater

intensity to the perinuclear, nuclear and nucleolar regions. X 1225

 

 

 

100

Figure 3. Phase contrast photomicrograph of anaplastic astrocytoma
cells. Cells were cultured in HLIC CDM and treated with NGF (12.5
ng/ ml) every 12 hrs for 48 hrs prior to fixation in paraformaldehyde.
Notice the cellular extensions and cytoplasmic processes. Also notice
that some cells appear paler and more epithelioid (solid arrowhead)
and others appear more dense with condensed cytoplasm being more

characteristic of fibrous astrocytes (open arrowhead). X 269

Figure 4. GFAP indirect immunofluorescencent staining of anaplastic
astrocytoma cells. Cells were cultured in HLlC CDM and treated with
NGF (12,5 ng/ ml) every 12 hrs for 48 hrs. Goat anti-mouse-FITC
labeled secondary antibody and mouse anti-GFAP primary antibody
were used following paraformaldehyde fixation. Notice the diffuse
staining of the cytoplasm and the intense staining of the perinuclear

and nucleolar regions. X 269

102

Discussion

GFAP is considered to be the primary intermediate filament of mature
differentiated astrocytes (Eng, 1985). Today, GFAP is widely accepted to be an
astrocyte-specific intermediate filament (Eng, 1985; Dahl et al, 1986).
Vimentin, however, does constitute the major intermediate filament of
immature astrocytes (Dahl et al, 1981; Schnitzer et al, 1981). The changes in
morphology which accompany cell differentiation in astrocytes have been
shown to be coupled to a shift in IFs from vimentin to GFAP, which begins
about the time of myelination during development (Dahl, 1981; Yokoama,
1981).

It was reported in previous studies in our department that N GF
induces morphologic differentiation in neoplastic astrocytes (Marushige et al,
1987, 1989). In these studies, GFAP was not detected in T-9 astrocytoma cells
using immunocytochemical staining techniques following alcohol fixation.
However, vimentin and GFAP containing IF were detected by IF isolation
. using a standard technique and western blot analysis-1 There also was no
detection of GFAP in cells even after treatment with NGF (5 ug/ ml) using
immunocyto-staining, following alcohol fixation (Marushige et al, 1989). The
data in this study support those findings. That is, GFAP is undetectable in T-9
astrocytoma cells cultured in CDM and fixed in alcohol, even when cells are
treated prior to fixation with NGF. However, because fixation methods have
been previously shown to play an important role in

immunohistocytochemistry (Sternberger, 1979; Polak and VanNoorden, 1983;

 

1 Marushige unpublished personal communication.

1 03
DeArmond and Eng, 1984), alcohol and paraformaldehyde fixatives were

compared in this study. Previous reports had shown preferential
histochemical staining of protoplasmic and fibrous astrocytes in the rat with
GFAP monoclonal antibodies, using a variation of fixatives (Shehab et al,
1990). The results presented here show that GFAP epitopes can be detected by
FITC-labeled secondary antibodies in anaplastic astrocytoma cells, following
NGF treatment, using paraformaldehyde fixation (Figure 2). However, T-9b
astrocytoma cells cultured under similar conditions to those earlier reported
from our department were found, as reported before, to be negative for GFAP
following paraformaldehyde fixation, but without NGF treatment. GFAP
epitopes were detected only after treatment with NGF and paraformaldehyde
fixation. These results, therefore, indicate that alcohol and paraformaldehyde
fixation may have opposite as well as preferential effects on GFAP epitopes
expressed by different morphologic types of astrocytes and astrocytoma cells.
In addition, it appears that NGF may preferentially increase the expression of
IF proteins which share the GFAP epitope common to the GFAP IF protein of
prot0plasmic astrocytes. This has been further supported by other studies on
IFs of neonatal rat astrocytes and astrocytoma cells, which deal with protein
soluability and immunochemical detection of epitopes. These additional
studies include IF protein isolation using a variation of standard technique
and western blots for GFAP, vimentin and protein. These studies suggest that
GFAP epitopes may not be preserved in the IFs of anaplastic astrocytoma cells
following alcohol fixation and could therefore go undetected. These results
further suggest that NGF may have a positive effect on the expression of
GFAP epitopes in IFs of anaplastic astrocytoma cell's. Although, quantitative

analysis was not carried out.

104

This report, therefore, supports previous findings which have reported
positive and/ or negative results on the detection of GFAP by
immunohistochemical staining of protoplasmic and fibrous astrocytes in
serial brain sections (Shehab et al, 1990) and also lends some understanding to
conflicting reports on the expression and detection of GFAP by
immunocytochemical staining of astrocytes and astrocytoma cells grown and
cultured under various conditions and prepared using different methods of
detection. More importantly, these results appear to suggest that a unique IF
may be expressed by T-9 anaplastic astrocytoma cells which shares a GFAP
epitope with neonatal rat GFAP. This GFAP epitope is sensitive to alcohol
fixation but preserved by paraformaldehyde fixation, while the IF differs in
protein solubility characteristics from the definitive GFAP IF expressed by
neonatal rat astrocytes in vitro and adult rats in vivo.

The hypothesis that a unique IF may be expressed by anaplastic
astrocytoma cells which shares a GFAP epitope with IFs expressed by
protoplasmic astrocytes, is supported by the numerous reports on MNU
induced chemical lesions, involving guanine, which are responsible for the
generation of this tumor type (Gercham and Ludlum, 1973; Kleihues and
Magee, 1973; Singer, 1975; Kleihues, 1982). In addition, this hypothesis can
also be supported by the numerous reports dealing with the expression and
function of IFs (Traub, 1983; Traub and Vorgias, 1983, 1984; Vorgias, 1983;
Traub, 1985; Miura et al., 1990). The promoter sequence and the
transcriptional startpoint and promoter function of the mouse GFAP gene
has been characterized (Miura et al., 1990). It has been reported that the cis
elements for astrocyte specific expression is located within 256 base pairs from
the transcription startpoint. Three trans-acting factor binding sites have also

been defined and identified as GF I, GF II, and GF III. These binding sites are

105

high in their guanine content. Mutations in GF 11 have been shown to
drastically reduce promoter activity. Base substitutions in GF I and GF III
have been shown to abolish cell-specific expression. Since 06-methylation of
guanine is reported to be a major result of MNU exposure, it is suggested that
formation of 05-methylguanine in the GFAP gene could lead to an important
mutagenic reaction and affect the expression of GFAP and IFs protein
structures. Future studies should be directed to establish whether a definite
correlation exists between the level of 06-methylguanine incorporation and
the expression of GFAP IFs or other potential mutagenic events in this
experimental brain tumor model.

In summary, GFAP and vimentin IF proteins are shown to be located
in perinuclear, nuclear and nucleolar regions of astrocytes (Figure 1, 2A, 2B)
and astrocytoma cells (Figure 4). IF proteins expressing GFAP epitopes are
induced in astrocytoma cells by NGF. This induction of IF proteins expressing
GFAP epitopes by astrocytoma cells may take place through NGF's generation
of cyclic AMP. Cell specific gene expression has been speculated to be
regulated by a combination of a variety of trans-acting factors also containing
multiple cis-regulatory promoter elements. GF I is a distinct trans-acting
factor which contains a homologue of the AP-2 binding site (Miura et al.,
1990). AP-2 is a transcription factor which mediates transcription activation
in response to at least two signal-transduction pathways, protein kinase C
activation (phorbol esters) and protein kinase A activation (cyclic AMP)
(Imagawa et al., 1987). NGF induced cyclic AMP therefore may act through
activation of protein kinase A or possibly independently and susequently
activate the GFAP promoter region resulting in increased expression of GFAP
IFs. The GFAP epitope is further suggested to be preserved by
paraformaldehyde fixation and abolished by alcohol fixation. IFs of

106

astrocytoma cells expressing this GFAP epitope may also show different
characteristics in IF protein solubility which has been suggested by the
variation in IF protein isolation between neonatal astrocytes and anaplastic
astrocytoma cells. This variation in IF protein may result from secondary
mutational events occurring to guanine nucleotides in the GFAP and
vimentin IF gene, which has an abundant guanine content. IFs have been
shown in this study to have a perinuclear location in the cell. The
significance of IF proteins having a nuclear function may be acknowledged in
future studies by exhibiting their involvement in regulation of mitogenesis,

gene expression and nuclear rRNA transport.

107

References

Biberfield, P., Ericsson, J.-L. E., Perlmann, P. and Raftell, M. (1965) Increased
occurrence of cytoplasmic filaments in in vitro propagated rat liver
epithelial cells. Exp Cell Res 39, 301-305.

Bignami, A. and Dahl, D. (1974) Astrocyte-specific protein and neuroglial
differentiation. An immunofluorescence study with antibodies to glial
fibrillary acidic protein. J Comp Neurol 153, 27-37.

Bignami, A., Dahl, D. and Rueger, D. C. (1980) Glial fibrillary acidic (GFA)
protein in normal neural cells and in pathological conditions. In,
Advances in cellular neurobiology (Federoff, S. and Hertz, L., eds.),
Vol. 1, pp .285-310, Academic Press, New York.

Bock, E. (1978) Nervous system specific proteins. J Neurochem 30, 7-14.

Dahl, D. (1981) The vimentin-GPA protein transition in rat neuroglia
cytoskeleton occurs at the time of myelination. J Neurosci Res 6, 741-
748.

Dahl, D., Bjorklund, H. and Bignami, A. (1986) Immunological markers in
astrocytes. In, Astrocytes (Federoff, S. and Vernadakis, A., eds.), Vol 3,
pp 1-25, Academic Press, New York.

Dahl, D., Rueger, D. C., Bignami, A., Weber, K. and Osborn, M. (1981)
Vimentin, the 57,000 dalton protein of fibroblast filaments, is the major
cytoskeletal component in immature glia. Eur J Cell Biol 24, 191-196.

DeArmond, S. J. and Eng, L. F. (1984) Immunohistochemistry: techniques and
application to neuroncology. Prog Exp Tumor Res 27, 92-117.

1 08
Eng, L. F. (1985) Glial fibrillary acidic protein (GFAP): the major protein of
glial intermediate filaments in differentiated astrocytes.
I Neuroimmunol 8, 203-214.
Eng, L. F. and Bigbee, J. W. (1978) Immunohistochemistry of nervous system
specific antigens. In, Advances in neurochemistry, (Agranoff, B. W.
and Aprison, M. H., eds.), Vol. 3, pp 43-98, Plenum, New York.

Eng, L. F., Vanderhaeghen, J. J., Bignami, A. and Gerst, B. (1971) An acidic
protein isolated from fibrous astrocytes. Brain Res 28, 352-354.

Franke, W. W., Schmid, E., Osborn, M. and Weber, K. (1978) Different
intermediate-sized filaments distinguished by immunofluorescence
microscopy. Proc Natl Acad Sci USA 75, 5034-5038.

Franke, W. W., Schmid, E., Schiller, D. L., Winter, 5., Jarasch, E. D., Moll, R.,
Denk, H., Jackson, B. W. and Illmensee, K. (1982) Differentiation-
related patterns of expression of proteins of intermediate-size filaments
in tissues and cultured cells. Cold Spring Harbor Symp Quant Biol 46,
431-453.

Gerchman, L. L. and Ludlum, D. (1973) The properties of 06-methylguanine
in templates for RNA polymerase. Biochim Biophys Acta 308, 310-316

Imagawa, M., Chiu, R. and Karin, M. (1987) Transcription factor AP-2
mediates induction by two different signal-transduction pathways:
Protein kinase C and cyclic AMP. Cell 51, 251-160.

Ishikawa, H., Bischoff, R. and Holtzer, H. (1968) Mitosis and intermediate-
sized filaments in developing skeletal muscle. J Cell Biol 38, 538-555.

Kleihues, P. and Magee, P. N. (1972) Alkylation of rat brain nucleic acids by N-
methyl-N-nitrosourea and methyl methanesulphonate. J Neurochem
20, 595-606.

/— ,

1 09
Kleihues, P. and Rajewsky, M. F. (1984) Chemical neuro-oncogenesis: role of
structural DNA modification, DNA repair and neural target cell
population. Prog Exp Tumor Res 27, 1-13.

Klymkowsky, M. W., Miller, R. H. and Lane, E. B. (1983) Morphology,
behavior and interaction of cultured epithelial cells after the antibody-
induced disruption of keratin filament organization. J Cell Biol 96,
494-509.

Lazarides, E. (1980) Intermediate filaments as mechanical integrators of
cellular space. Nature 283, 249-256.

Lazarides, E. (1982) Intermediate filaments: A chemically heterogenous,
developmentally regulated class of proteins. Ann Rev Biochem 51,
219-250.

Liem, R. K. H., Yen, S.-H., Salomon, G. D. and Shelanski, M. (1978)
Intermediate filaments in nervous tissue. J Cell Biol 79, 637-645.

Lin, J. J.-C. and Feramisco, J. R. (1981) Disruption of the in vivo distribution of
the intermediate filaments in fibroblasts through the microinjection of
a specific monoclonal antibody. Cell 24, 185-193.

Marushige, Y., Marushige, K., Okazaki, D. L. and Koestner, A. (1989)
Cytoskeletal reorganization induced by nerve growth factor and glial
maturation factor in anaplastic glioma cells. Anticancer Res 9, 1143-
1148.

Marushige, Y., Raju, N. R., Marushige, K. and Koestner, A. (1987)
Modulation of growth and morphological characteristics in glioma
cells by nerve growth factor and glia maturation factor. Cancer Res
47, 4109-4115.

110

Miura, M., Tamura, T.-A. and Mikoshiba, K. (1990) Cell-specific expression of
the mouse glial fibrillary acidic protein gene: Identification of the cis-

and trans-acting promoter elements for astrocyte-specific expression.
I Neurochem 55, 1180-1188.

Morrison, R. S. and De Vellis, J. (1981) Growth of purified astrocytes in a
chemically defined medium. Proc Natl Acad Sci USA 78, 7205-7209.

Osborn, M., Geisler, N ., Shaw, G., Sharp, G. and Weber, K. (1982) Intermediate
filaments. Cold Spring Harbor Symp Quant Biol 46, 413-429.

Paetau, A., Virtanen, I., Stenman, S., Kurki, P., Linder, E., Vaheri,
A.,Westrnark, B.,Dahl, D. and Haltia, M. (1979) Glial fibrillary acidic
protein and intermediate filaments in human glioma cells. Acta
Neuropathol 47, 71-74.

Pixley, S. K., Kobayashi, Y. and De Vellis, J. (1984) A monoclonal antibody
against vimentin: Characterization. Brain Res 317, 185-199.

Polak, J. M. and Van Noorden, S. V. (1983) Immunocytochemistry: practical
applications in pathology and biology. Wright, London.

Raff, M. C., Miller, R. H. and Noble, M. (1983) A glial progenitor cell that
develops in vitro into an astrocyte or an oligodendrocyte depending on
culture medium. Nature 303, 390-396.

Russell, D. S. and Rubenstein,L. J. (1989) Pathology of tumours of the nervous
system. 5th edn. Williams & Wilkins, Baltimore.

Schmidek, H. H., Nielsen, S. L., Schiller, A. L. and Messer, J. (1971)
Morphologic studies of rat brain tumors induced by N -
nitrosomethylurea. J Neurosurg 34, 335-340.

111

Schnitzer, J., Franke, W. W. and Schnitzer,M. (1981) Immunocytochemical
demonstration of vimentin in astrocytes and ependymal cells of
developing and adult mouse nervous system. J Cell Biology 90, 435-
447.

Shehab, S. A. S., Cronly-Dillon, J. R., Nona, S. N. and Stafford, C. A. (1990)
Preferential histochemical staining of protoplasmic and fibrous
astrocytes in rat CNS with GFAP antibodies using different fixatives.
Brain Res 518, 347-352.

Singer, B (1979) N-nitroso alkylating agents: formation and persistence of
alkyl derivatives in mammalian nucleic acids as contributing factors in
carcinogenesis. J Natl Cancer Inst 62, 1329-1339.

Sternberger, L. A. (1979) Immunocytochemistry. Wiley, Berlin.

Sternlieb, I (1965) Perinuclear filaments and microtubles in human
hepatocytes and billary epithelial cells. I Microscopic 4, 551-558.

Traub, P. (1985) Are intermediate filament proteins involved in gene
expression? In, Intermediate Filaments (Wang, E., Fischman, D., Liem,
R. K. H. and Sun, T.-T. , eds.), Ann. N. Y. Acad. Sci. 455, 68-78.

Traub, P. and Vorgias, C. E. (1983) Involvement of the N-terminal peptide of
vimentin in the formation of intermediate filaments. J Cell Sci 63, 43-
67.

Traub, P. and Vorgias, C. E. (1984) Differential effect of arginine modification
with 1,2-cyclohexanedione on the capacity of vimentin and desmin to
assemble into intermediate filaments and to bind to nucleic acids.

] Cell Sci 65, 1-20.

112

Traub, P., Nelson, W. J., Kuhn, S. and Vorgias, C. E. (1983) The interaction in
vitro of the intermediate filament protein vimentin with naturally
occurring RNAs and DNAs among naturally occurring DNA base
sequences, poly (dG) binds vimentin with greatest intensity. J Biol
Chem 258, 1456-1466.

Vorgias, C. E. and Traub, P. (1983) Isolation of glial fibrillary acidic protein
from bovine white matter and its purification by affinity
chromatography on single-stranded DNA-cellulose. Biochem Biophys
Res Comm 115, 68-75.

Yokoyama, K., Mori, H. and Kurokawa, M. (1981) Astroglial filament and
fibroblast intermediate filament proteins in cytoskeletal preparations
from spinal cord and optic nerve. FEBS Lett 135, 25-30.

_.—.a--.- 4.7—4.1.

Chapter 5

An in vitro Study on the Effect of Nerve Growth Factor
on Gamma-aminobutyric Acid Uptake by
Neonatal Rat Astrocytes and Rat T-9 Astrocytoma Cells

William R. Hare Jr.

Department of Pathology
Michigan State University
East Lansing, MI 48824-1316

Telephone, 51 7-353-9160

Telefax no. 517-336-1053

113

114

Abstract

The effect of nerve growth factor (NGF) on the in vitro uptake of
gamma-aminobutyric acid (GABA) by rat astrocytes is reported. NGF's effect
on GABA uptake by neonatal rat astrocytes is compared to NGF's effect on
GABA uptake by rat anaplastic astrocytoma cells. Both of these cell types
originate from the Fischer rat genetic pool. NGF is found not to significantly
increase GABA uptake by neonatal astrocytes or anaplastic astrocytoma cells
cultured in chemically defined media. The experimental uptake data
presented here further show that NGF actually acts to decrease GABA uptake
by neonatal rat astrocytes. We suspect that NGF causes this effect by somehow
affecting the GABAA receptor. Seizures induced as a consequence of
decreased synaptic levels of GABA is well accepted. If NGF were to increase
GABA uptake, it could be expected to potentiate seizures. However, the
present study does not support this hypothesis. This report, therefore,
concerns the potential use of NGF as a pharmacologic agent for the
differentiation and arrest of tumor growth as well as for the reduction of

seizures.

Running Title: The effect of N GF on GABA uptake

Key Words: GABA Uptake, nerve growth factor, basic-Fibroblast growth
factor, GABA receptors, Neonatal astrocytes, Astrocytoma cells.

 

115

Introduction

Gamma-aminobutyric acid (GABA) is widely recognized to be an
important inhibitory amino acid neurotransmitter of the CNS
(ijevic,1974). GABA is inactivated after its release from nerve endings by
uptake processes shared by some neurons and glia (Henn and Hamberger,
1971 ; Iversen 8: Kelly, 1975). These uptake processes are very similar but can
be distinguished by their structural requirements (Bowery ct al., 1974).
Neuronal and glial uptake of synaptic GABA accounts for 70% of total brain
concentrations (Beart, 1976).

Glial cells were first shown to be involved in terminating the action of
GABA in autoradiographic studies using tritiated GABA (Hokfelt 8:
Ljungdahl, 1970). High affinity, energy-dependent GABA uptake was soon
reported in bulk-isolated astroglial cells (Henn 8: Hamberger, 1971; Schousboe
ct al., 1977, 1978). This high affinity uptake was further shown to be sodium-
dependent and optimal at physiologic potassium concentrations (Sellstrom 8:
Hamberger, 1975; Schousboe, 1981).

GABA uptake in astrocytes is coupled to the internal movement of one
atom of sodium (neuronal uptake is coupled to 2 and 3 atoms) and the
external movement of one atom of potassium (Hertz et al., 1978; Martin, 1976;
Schousboe, 1981). This implies that GABA uptake is a function of changes in
membrane potential and the sodium gradient (Sellstrom and Hamberger,
1977; Larsson ct al, 1986). In fact, high concentrations of extracellular
potassium, upon depolarization, initiate the release of accumulated GABA by

neurons in a calcium-dependent fashion (Sellstrom and Hamberger, 1977).

 

116

GABA produces its effects as an inhibitory neurotransmitter by
enhancing the flux of chloride ions across the cell membrane (Krnjevic, 1974).
There are two types of GABA receptors. GABAA is the major inhibitory
receptor and operates a chloride channel. In fact, there are two major types of
chloride channels, those that are GABA operated and those that are voltage
dependent. One type is active, one type is passive, respectively. This voltage-
dependent chloride channel plays an important role in determining the
resting membrane potential as well as in maintaining intracellular pH. The
GABAB receptor regulates Cyclic AMP production in concert with other
stimulatory receptor-ligand complexes (Enna 8: Karbon, 1987). Both GABA
receptors are vital to a normal functioning nervous system. The inhibition of
GABA A receptors has become the molecular target for the development of
many anxiolytic drugs like the benzodiazepines, e.g. Valium (Abalis ct al.,
1983; Fischer 8: Olsen, 1986). In addition, GABAA receptors have been found
to be the neurotoxic target of many potentially toxic compounds including the
chlorinated hydrocarbons and the pyrethroids (Casida 8: Lawrence, 1985;
Abalis ct al., 1985; Stark et al., 1986; Gant et al., 1987). Defects in chloride
permeability relating to imperfect GABA receptors has also been shown in the
hereditary diseases of myotonia and cystic fibrosis (Bryant 8: Morales-
Aguilera, 1971 ; Frizzell et al., 1986).

Nerve growth factor (NGF) is recognized as a potent differentiation and
maturation promoter (Levi-Montalcini 8: Angeletti, 1968; Greene and
Tischler, 1976; Greene 8: Shooter, 1980; Yankner 8: Shooter, 1982; Levi-
Montalcini, 1987; Marushige et al., 1987) as well as an important factor in the
maintenance of selected neuronal cell populations of the nervous system
(Thonen et al., 1987; Whittemore 8: Seiger, 1987; Springer, 1988). Among the

studies on astrocyte morphology are reports that morphology influences

1 1 7
GABA uptake (Wilkin ct al., 1983). It has been reported that stellate-

appearing rat cerebellar astrocytes preferentially take up and concentrate
GABA (Wilkin et al., 1983; Levi et al., 1986). It is hypothesized that NGF
treatment of anaplastic astrocytomas will result in morphologic
differentiation and in an increased uptake of the inhibitory amino acid
neurotransmitter GABA.

As a result of NGF's dramatic ability to promote differentiation, it has
been suggested as a potential therapeutic agent in the treatment of brain
tumors. However, if one assumes that previous reports indicating that
increased morphologic differentiation by astrocytes also increases net GABA
uptake, then NGF's possible therapeutic benefit would be in question, since
increased uptake of GABA would potentially increase the incidence of
seizures. Therefore, the purpose of this study was to determine if NGF
increases the uptake of GABA. [3H]-GABA uptake by neonatal rat astrocytes
and T-9 rat astrocytoma cells was compared under identical in vitro

conditions using two different chemically defined media.

 

118

Materials and Methods

Materials

NGF (2.5 S, grade II) was purchased from Boehringer Mannheim
Biochemicals (Indianapolis, IN,USA). Fetal calf serum (FCS) was purchased
from Hyclone Sterile Systems Inc. (Logan, UT, USA). All culture media were
prepared using stock solutions, chemicals and supplies from Life
Technologies, Inc,GIBCO Labs (Grand Island, NE, USA), Corning Glass Works
Inc. (Corning, NY, USA) and Sigma Chemical Co. (St. Louis, MO, USA). Hl-1
supplement was purchased from Endotronics Inc. (Coon Rapids, MN, USA).
GABA was purchased from ICN-Biomedicals Inc. (Costa Mesa, CA, USA) and
[3H]-GABA (26.9 Ci/mmol) from New England Nuclear, Dupont/NEN
Products. (Boston, MA, USA).

Neonatal Astrocyte Cultures

Neonatal rat astrocytes were generated using 4 day old Fischer rat pups.
The brains of four rat pups were removed following euthanasia and placed in
a petri dish containing warm DMEM high glucose media. The meninges
were removed and the brainstem and cerebellum separated. The cerebral
tissues were minced into small pieces (<0.5 mm), combined and placed into a
sterile centrifuge tube containing 5 ml of DMEM high glucose media. This
suspension of brain tissue was then centrifuged at 750 X g for a period of 10
mins. The upper supernatant fluid was removed and discarded and replaced
with 1.5 m1 of Collagenase II-S (Sigma ) solution (0.8%, w/v). The enzyme

treatment procedure was followed by incubation of the cell mixture on a

 

119

warm water bath (37°C) shaker for 30 mins. The brain tissue was completely
dissociated to a cell suspension by intermittently pipetting with a sterile glass
pasteur pipette supplied with a cotton filter. At no time during preparation of
these cultures were the cells allowed to cool down below 30°C. Following the
enzyme treatment the cell suspension was centrifuged at 200 X g for a period
of 10 mins and the supernatant fluid removed and discarded. An equal
amount (approx. 3 ml) of complete culture medium containing 15% FCS was
gently layered on top of the cell pellet. Then the upper 2/ 3 of the cell pellet
was resuspended in fresh medium using gentle pipetting action. A 1 ml
aliquot of this cell suspension was then quickly removed and used to seed
primary cultures at approximately a 1:20 Split ratio. These stock cultures were
started and maintained in 25 cm2 tissue culture flasks (Corning) containing 5
ml of complete medium. The medium was replaced with fresh medium the
following day and replaced every 3 days thereafter. At the time of second
passage medium was changed replacing DMEM with RPMI 1640/ Hams F12
(1:1, v/v). Neonatal astrocytes used in the GABA uptake experiments were at
their 3rd passage. The astrocytic character of these cells was indicated at
second and third passage not only by morphology but by their content of the
astrocyte specific, glial fibrillary acidic protein.

Astrocytoma Cells

The rat anaplastic astrocytoma cells were supplied by the Department of
Pathology, Michigan State University, East Lansing, MI. from their cell storage
bank. The T-9 cell line originated from a high grade anaplastic astrocytoma
induced in Fischer rats by treatment with N-methyl-N-nitrosourea (MNU)
(Schmidek, et al. 1971). Stock cultures were established in 25 cm2 tissue

culture flasks and maintained by complete serum supplemented medium

 

120

(DMEM and RPMI 1640 / Hams F12) replacement every 3rd day and the
splitting of cells every 6th day at a 1:100 split ratio.

Culture Conditions

Cells were cultured in a Hotpack CO2 Incubator which was maintained
at 5% C02, 37°C and constant humidity. Both neonatal astrocytes and
astrocytoma cells used in the GABA uptake studies were split into multi-well
(6 well) culture plates (GIBCO Labs) containing 3 ml of complete medium.
This medium helped to assure cell attachment. The complete medium,
serum supplemented, was changed the following day and replaced with
chemically defined medium (CDM). CDM HL1A was composed of DMEM
supplemented with 1% (v/v) HL-l supplement, 400 uM glutamine,
gentamycin at 10 ug/ ml, glucose at 1 mg/ ml, CaCl2 at 175 ug/ ml and MgSO4
at 125 ug/ ml. CDM HLlB was also composed of DMEM and contained the
same supplementation as HLIA but also contained Hydrocortisone at 1.6
ug/ m1, Prostaglandin F2-alpha at 440 ng/ ml, Putrescine at 78 ug/ ml, beta-
Fibroblastic Growth Factor at 8.8 ng/ ml and Meylin Basic Protein at 440
ng/ ml. This HLlB medium was a slight modification of that originally
proposed by Morrison and De Vellis as a CDM that initiates differentiation of
astrocytes. HL-I supplement contains 29 ug/ ml total protein with 15 ug/ ml

insulin and no additional growth factors, glutamate or GABA.

GABA Uptake Experiments

Neonatal astrocytes and astrocytoma cells used in these studies were
seeded at 6-12,000 cells / ml into 3 ml of complete, serum-supplemented,
media at initial plating. The medium was changed the following day to

serum free CDM. Three days later the CDM was replaced with fresh CDM of

121

the same type and NGF added. NGF was added at 500 ng/ ml or serum added
at 10% (300 111/3 ml medium). Cells were incubated in NGF or growth factor
containing serum for 48 hrs prior to the addition of GABA. Each GABA
concentration that was studied (2.5 and 25 nmol) contained 3 uCi of [3H]-
GABA per 3 ml of culture medium per petri dish. [3H]-GABA had a specific
activity of 1 uCi/ 40 pmol. GABA specific activity was established at 1 uCi/2.5-
25 nmol. The standard incubation period for GABA uptake was set at room
temperature (RT) (20-25°C) for 10 min. During the uptake incubation, cell
culture plates were placed on a Lab-Line rotary orbit shaker (60 rpm) to assure
dispersion of GABA. The incubation was stopped by removing the [3H]-
GABA containing medium and rinsing the cultures twice in a HEPES
buffered saline solution, pH 7.4, which contained the same concentration of
unlabeled GABA. Each rinse was for a period of 10 min and was also carried
out on a rotary shaker. Following this rinse procedure the cells were extracted
by placing 2 ml of a freshly prepared solution of 1M perchloric acid into each
well. Cells were extracted for 10 min, again utilizing the rotary shaker. Cells
were then given a post-extraction rinse using 2 ml of 0.1M Tris buffer, pH 7.4.
This rinse solution was combined with the extract solution. Throughout this
procedure cells remained attached to the culture plate. GABA uptake data
were recorded as counts per minute (cpm) and then normalized using specific
activity of GABA and cpm/ protein concentration and reported as nmol
GABA/ mg protein/ time (10 min). The experiment was designed to compare
GABA uptake by neonatal rat astrocytes and rat anaplastic glioma cells in two
different CDM and to determine the effects of NGF, serum and hypothermia

(Table 1). Three or more replicate cultures were analyzed for each study.

1 2 2
Scintillation Analysis
The procedure was carried out using a Packard 300 scintillation

counter. Uptake extract solution (0.8 ml) was added to Aquasol H scintillation
fluid (10 ml).

Protein Concentration

The Bio-Rad method of protein analysis was used. Extract uptake
solution (0.8 ml) was combined with 200 ill of reagent and absorbance checked
at 595 nm. This was compared to a standard curve using Bovine IgG and
quantitated accordingly. The average protein concentration for T-9 cultures

was 36.8 ug/ ml. and 38.8 ug/ ml for NR-l cultures.

Statistical Analysis

Determination of statistically significant differences between
experimental groups was formed utilizing the Students-t test. Differences
were considered to be significant when p values = or <0.05 were obtained for
n= or >3. Standard error of the means (SEM) was also determined and

reported where significance was indicated.

123

Results

Effects of Medium on GABA Uptake

The net uptake of GABA (2.5 and 25 nmol) by NR-1 and T-9 cells in
HLlA and HLIB CDM is compared. Cells were cultured in both media and
incubated with GABA for 10 min. There was a significant difference in
GABA uptake by these cells in these different media (Table 1.). GABA uptake
was much greater by cells cultured in HLlA.than in HL1B.

Effects of N GF on GABA Uptake

This study examines the effect of NGF on GABA uptake (2.5 and
25 nmol) following a 10 min incubation at room temperature by NR-1 and T-9
cells cultured in HLlA and HL1B CDM. NGF (500 ng/ ml) was found to
significantly decrease the uptake of GABA (2.5 and 25 nmol) by NR-l
astrocytes cultured in HL1B CDM (Table 1.).

Effects of Fetal Calf Serum on GABA Uptake

This study determined the effects of serum (100 ul/ ml medium) on
GABA uptake following a 10 min incubation at room temperature by NR-l
and T-9 cells cultured in HLlA CDM. The addition of fetal calf serum (FCS)
in place of N GF was not found to significantly increase the uptake of GABA
(25 umol) by T-9 astrocytoma cells nor by NR-1 astrocytes (Table 1).

 

124

Control Study on the Effects of Hypothermia on GABA Uptake.

GABA uptake by NR-l and T-9 cells was determined at temperatures
<4°C. Cell cultures were placed on ice for 90 min prior to a 10 min GABA
incubation at temperature <4°C. Most cellular uptake processes are generally
stopped by low temperatures and recover when temperature is increased.
Hypothermia was found to cause a significant decrease in the uptake of
GABA (25umol) by both NR-l astrocytes as well as T-9 astrocytoma cells
(Table 1.). This suggests that oxidative phosphorylation and the subsequent
generation of ATP is a necessary component for GABA uptake by both cell

types.

125

TABLE 1. GABA uptake by neonatal astrocytes and anaplastic astrocytoma
cells cultured under varying conditions and treatments.

 

 

Culture GABA Uptake by Cell Type
Conditions and Treatment NR-l T-9
(concentration, temp, medium,treatment) (nmol GABA / mg protein/10 min)
25 nmol GABA, RT, HL1A 483.0 (62.5) 571.0 (45.5)
2.5umol GABA, RT, HL1A 34.0 (1.0) 39.7 (3.2)
25umol GABA, RT, HL1B 170.3 (22.7) 311.0 (4.0)
2.5umol GABA, RT, HL1B 15.7 (0.3) 36.33 (6.2)
25 nmol GABA, RT, HL1A, NGF 349.3 (4.7) 494.7 (20.5)
2.5umol GABA, RT, HL1A, NGF 30.5 (1.5) 41.7 (0.8)
25 nmol GABA, RT, HL1B, NGF 122.3 (7.7) 329.7 (11.7)
2.5umol GABA, RT, HL1B, NGF 11.3 (0.3) 32.7 (2.2)
25 nmol GABA, RT, HL1A, SERUM 398.0 (1.2) 780.7 (67.6)
25umol GABA, 4°C, HL1A 101.7 (6.2) 307.0 (13.1)

 

Statistically significant differences in GABA uptake are shown between
NR-l and T-9 cells cultured in HL1B medium, between HL1A and HL1B
medium in both NR-l and T-9 cells, between NGF treated NR-l cells and
non-treated cells cultured in HL1B medium at 2.5 and 25umol GABA and in
both cell types in response to cold temperature. All comparisons were made
using the Students-t test with p<0.05. Standard error of the mean (SEM) is
given in brackets.

126

Discussion

The purpose for this study was to determine if NGF has any significant
effect on GABA uptake. The reason relates to NGF's reported ability to
initiate morphologic differentiation (Greene 8: Tischler, 1976; Marushige et
al., 1989) as well as its reported ability to arrest growth and proliferation
(Marushige et al., 1987). These reports, in addition to many other similar
reports in the literature, suggest that NGF may have benefit in the treatment
of tumors of the nervous system, particularly treatment of tumors of the glial
type. There are other reports in the literature that have suggested that growth
factors, hormones and other chemicals that promote differentiation may also
increase the uptake of the inhibitory neurotransmitter GABA (Wilkin et al,
1983). There are still other reports which state that NGF increases amino acid
transport, including utilization of the gamma-glutamyl cycle (Yankner and
Shooter, 1982). This could be interpreted to mean that NGF, like other
differentiation promoters, could increase the net uptake of GABA. If NGF
were to increase GABA uptake, it's supplementation to specific areas of the
CNS would be expected to potentially induce seizures. The proposition that
seizures arise as a consequence of GABA deficiency was suggested years ago
(Iwama 8: Jasper, 1957; Elliot 8: Jasper, 1959). It has now become universally
accepted that seizure activity and a lowered GABA activity go hand-in-hand
(Krnjevic, 1983) and a net increase in GABA uptake could be expected to bring
about a net decrease in synaptic GABA concentrations. Therefore, if NGF

were to have any hopes of development as a pharmacologic agent in the

127

treatment of tumors of the CNS, it must first be shown that NGF does not
have any detrimental effects on GABA's normal uptake process

Because of the importance of how NGF might affect tumor cells
differently than it does normal cells, this study draws a comparative picture of
the differences in cell type. The study compares two cell types that evolved
genetically identical, both originated from identical lines of Fischer rats,
except that one cell type was transformed genetically by exposure to N-
methyl-N-nitrosourea (MNU) (Schmidek, et al., 1971; Swenberg et al., 1972).
The rat anaplastic astrocytoma (T-9) cell line is well-established and has been
characterized very well (Ko et al., 1980). The neonatal rat astrocytes (NR-1)
were isolated and cultured specifically for this study. The culture methods of
NR-l and T-9 were made to be as identical as possible.

Medium preparation is a very important factor in any study where
morphology is an intricate part of the expected outcome. Different medium
preparations have been shown to initiate differentiation (Morrison 8:
DeVellis, 1981). In this experimental model, the differentiating medium
(HLlB) brought about a significant decrease in the uptake of GABA. This
suggests that one of the factors present in the HL1B medium may
preferentially decrease the uptake of GABA. This decrease could be brought
about by one of the constituents of the HL1B medium interfering with GABA
binding or by initiating a GABA release mechanism. We could speculate that
this action is somehow related to basic fibroblast growth factor (basic-FGF).
Little is known of the membrane receptor-induced activities of basic-FGF. It
has been reported to be mitogenic in the induction of endothelial growth
(Thomas ct al., 1985; Bohlen ct al., 1985; Lobb et al., 1985), but has not been
proven to be mitogenic in astrocytes (Leutz and Schachner, 1981). It has also

been found to be a neurotrophic factor increasing the survival of neurons and

128

growth of neurites in culture, similar to NGF (Morrison, et al., 1986, 1988;
Unsicker ct al., 1987; Walicke et al., 1986). Therefore, FGF, like NGF, may be
more important in the preservation and maintenance of astrocytes cultured
under these conditions. Hl1B medium was found to act synergistically with
NGF in reducing GABA uptake by NR-1 cells. This is shown in Table 1.
Glyc0protein molecules like heparin have been shown to be necessary for
high-affinity binding of basic-FGF to its receptor (Yayon et al., 1991). This
further implies that imperfect glycosylation of membrane protein
constituents of the FGF receptor complex may also be an important factor in
creating differences in the cellular responses of NR-1 and T-9 cells to HL1B
CDM.

NGF causes a significant reduction in the uptake of GABA by NR-1
astrocytes when cells are cultured in HLlB medium and treated with NGF for
48 hrs prior to uptake (Table 1.). The data reported here show that NGF does
not increase the uptake of GABA as other differentiation promoters and
conditions favoring morphologic differentiation have been reported to do
(Wilkin et al., 1983; Levi et al., 1986)

Although GABA uptake is not significantly changed in T-9 astrocytoma
cells, the uptake pattern of GABA in response to NGF treatment suggests a
very heterogeneous population of'NGF receptor complex. In the study of
NR-l neonatal astrocytes, NGF consistently decreased the uptake of GABA.
While in the study of GABA uptake by T-9 astrocytoma cells in the presence
of NGF, GABA uptake showed no statistically significant change overall
(Table 1). It is speculated that NGF may bring about its action relative to the
decrease in GABA uptake by phosphorylation of a GABA binding protein or a
structural change in the binding protein induced by the phosphorylation of
the NGF receptor resulting in the subsequent release or inhibition of GABA

129

binding. These speculations may be supported by the data showing an
increased uptake of GABA by T-9 astrocytoma cells in the presence of serum
(Table 1.). Fetal calf serum is known to contain growth factors whose
receptors are associated with tyrosine kinase activity. At least ten oncogene
products expressed by different tumor cell types are also associated with
tyrosine kinase activity. The availability of high energy phosphate could
explain the over activation of other tyrosine kinase processes. Another
possibility could be that serum induces growth and the expression of NGF
binding sites that are not responsive to tyrosine kinase activation. In either
case, NGF has the ability to modulate the uptake of GABA by NR-l neonatal
astrocytes cultured in HL1B medium and NR-I neonatal astrocytes differ
markedly from T-9 anaplastic astrocytoma cells in their response to N GF.

GABA uptake is well-established to be an active uptake process
associated with ionic movements which are in turn dependent on the
generation of high energy phosphate bonds in the form of ATP. In order to
test the GABA uptake experimental model presentedhere with a control
study, GABA uptake was recorded under conditions of hypothermia (temp
<4°C). NR-1 neonatal astrocytes showed over a 75% reduction in the uptake
of GABA and T-9 anaplastic astrocytoma cells showed over a 50% reduction
in uptake of GABA under hypothermic conditions. These data confirm that
this experimental model testing GABA uptake responds as expected to
conditions of hypothermia.

In summary, N GF has been shown to cause a significant decrease in
GABA uptake by N R-l neonatal astrocytes cultured in a chemically defined
medium. This decrease in GABA uptake may be due to activation of NGF
receptor-linked kinase and subsequent phosphorylation, resulting in

activation of a GABA release mechanism or in the ability of GABA to bind to

1 3O
GABA A receptors. This loss in affinity for GABA A receptors by GABA may
in-turn increase activation of GABAB receptors leading to greater
differentiation by increasing intracellular levels of Cyclic AMP.

This study introduces a new procedure for studying uptake by
astrocytes and astrocytoma cells. GABA uptake has been determined in cells
which were not disturbed prior to initiation of uptake. The determination of
cellular uptake was done by recording [3H]-GABA uptake by these cells using
analyses of cell extract. GABA uptake is reported in relation to protein
extracted. It is known that protein per cell is greater in NR-l neonatal
astrocytes as compared to T-9 anaplastic astrocytoma cells and our results
agree. However, this procedure is new and quite different than other
published procedures. Therefore, caution should be used in making
comparisons of these results to the results of previous uptake studies.

The importance of this study is that NGF does not significantly increase
the net uptake of GABA by astrocytes. This means that NGF and other
differentiation promoters could be expected to influence GABA uptake
independently. These results further show that NGF may potentially
decrease seizures by decreasing GABA uptake and therefore could have
promise for future therapeutic use as a promoter of tumor differentiation as

well as potential use as an anticonvulsant.

131

References

Abalis, I. M., Eldefrawi, M. E. and Eldrfrawi,A. T. (1983) Biochemical
identification of putative GABA/benzodiazapine receptors in house fly
thorax muscles. Pestic. Biochem. Physiol. 20, 39-48.

Albis, I. M., Eldefrawi, M. E. and Eldefrawi, A. T. (1985) High affinity
stereospecific binding of cyclodiene insecticides and gamma-
hexachlorohexane to gamma-aminobutyric acid receptors of rat brain.
Pestic. Biochem. Physiol. 24, 95-102.

Beart, P. M. (1976) The autoradiographic localization of L-(3H)-glutamate in
synaptosomal preparations. Brain Research 103, 350-355.

Bohlen, P., Esch, F., Baird, A., Jones, K. 1. and Gospodarowicz, D. (1985)
Human brain fibroblast growth factor: Isolation and partial chemical
characterization. FEBS Lett. 185, 177-181.

Bryant, S. H. and Morales-Aguilera, A. (1971) Chloride conductance in normal
and myotonic muscle fibers and the action of monocarboxylic aromatic
acids. J. Physiol. (London) 219, 367-383.

Cassida,J. E. and Lawrence, L. J. (1985) Structure activity correlations for
interactions of bicyclophosphorus esters and some
polychlorocycloalkane and pyrethroid insecticides with the brain-
specific t-butylbicyclophosphorothionate receptor. Environ. Health
Perspect. 61, 123-132.

Elliott, K. A. C. and Jasper, H. H. (1959) Gamma-aminobutyric acid. Physiol.
Rev. 39, 383-406.

 

1 32
Enna, S. J. and Karbon, E. W. (1987) Receptor regulation: evidence for a

relationship between phospholipid metabolism and neurotransmitter

receptor - mediated cAMP formation in brain. Trends Pharmacol. Sci. 8,
21-24.

Fischer, J. B. and Olsen, R. R. (1986) Biochemical aspects of

GABA/benzodiazepine receptor function, in Benzodiazepine/GABA

receptors and channels: structural and functional properties.,pp. 241-
259. Liss, New York.

Frizzell, R. A., Halm, D. R., Rechkemmer, G. and Shoemaker, R. L. (1986)

Chloride channel regulation in secretory epithelia. Federation Proc. 45,
2727-2731.

Gant, D. B., Eldefrawi, M. E. and Eldefrawi, A. T. (1987) Cyclodiene insecticides

inhibit GABAa receptor regulated chloride transport. Toxicol. Appl.
Pharmacol. 88, 313-321.

Greene, L. A. and Shooter, E. M. (1980) Nerve growth factor-biochemistry and
synthesis and mechanisms of action. Annu. Rev. Neurosci. 3, 353—402.

Greene, L. A. and Tischler, A. S. (1976) Establishment of a noradrenergic

clonal line of rat pheochromocytoma cells which respond to nerve
growth factor. Proc. Natl. Acad. Sci. USA 73, 2424-2428.

Henn, F. A. and Hamberger, A. (1971) Glial cell function: Uptake of
transmitter substances. Proc. Natl. Acad. Sci. USA 68, 2686-2690.

Hertz, L. Wu, P. H. and Schousboe, A. (1978) Evidence for net uptake of
GABA into mouse astrocytes in primary culture - its sodium
dependence and potassium independence. Neurochem. Res. 3, 313-323.

Hertz, L. and Schousboe, A., (1987) Role of astrocytes in compartmentation of

amino acid and energy metabolism, in Astrocytes Vol. 2 (Federoff,S.
and Vernadakis, A., eds) pp. 179-208. Academic Press, NY

 

1 33
Hokfelt, T. and Ljungdahl, A. (1970) Cellular localization of labeled gamma-
aminobutyric acid (3H-GABA) in rat cerebellar cortex: an
autoradiographic study. Brain Research 22, 391-396.

Iversen, L. L. and Kelly, J. S. (1975) Uptake and Metabolism of gamma-
aminobutyric acid by neurones and glial cells. Biochem. Pharmacol. 24,
933-938.

Iwama, K and Jasper, H. H. (1957) The action of gamma-aminbutyric acid
upon cortical electrical activity in the cat. J. Physiol. 138, 365-380.

Ko, L., Koestner, A. and Wechsler, W. (1980) Morphological characterization
of nitrosourea-induced glioma cell lines and clones. Acta Neuropathol.
(Berl) 51, 23-31.

Krnjevic, K. (1983) GABA-mediated inhibitory mechanisms in relation to
epileptic discharge, in Basic mechanisms of neuronal excitability
(Jasper, H. H. and van Gelder, N. M. eds) p. 249. Alan R. Liss Inc., NY.

Krnjevic, K. (1974) Chemical nature of synaptic transmission in vertebrates.
Physiol. Rev. 54, 418-540.

Larsson, O. M., Hertz, L. and Schousboe, A. (1986) Uptake of GABA and
nipecotic acid in astrocytes and neurons in primary cultures: Changes
in the sodium coupling ratio during differentiation. J. Neurosci. Res.
16, 699-708.

Leutz, A. and Schachner, M. (1981) Epidermal growth factor stimulates DNA-
synthesis of astrocytes in primary cerebellar cultures. Cell Tissue Res.
220, 393-404.

Levi-Montalcini, R. (1987) Nerve growth factor : thirty-five years later. EMBO
J. 6, 1145-1154.

“W‘-

 

134

Levi-Montalcini, R. and Angeletti, P. (1968) Nerve growth factor. Physiol.
Rev. 48, 534-569.

Lobb, R. R., Alderman, E. M. and Fett, J. W. (1985) Induction of angiogenesis
by bovine brain derived class 1 heparin-binding growth factor.
Biochemistry. 24, 4969-4973.

Martin, D. L. (1976) Carrier- mediated transport and removal of GABA from
synaptic regions. in GABA in the nervous system function (Roberts, E.,
Chase, T. N ., and Tower, D. B. eds.), pp. 347-386. Raven Press, New
York.

Marushige, Y., Marushige, K., Okazaki, D. L. and Koestner, A. (1989)
Cytoskeletal reorganization induced by nerve growth factor and glial

maturation factor in anaplastic glioma cells. Anticancer Res. 9, 1143-
1148.

Marushige, Y., Raju, N. R., Marushige, K. and Koestner, A. (1987) Modulation
of growth and morphological characteristics in glioma cells by nerve
growth factor and glia maturation factor. Cancer Res. 47, 4109-4115.

Morrison, R. S., Keating, R. F. and Moskal, J. R. (1988) Basic fibroblast growth
factor and epidermal growth factor exert differential trophic effects on
CNS neurons. J. Neurosci. Res. 21, 71-79.

Morrison, R. S., Sharma, A., De Vellis, J. and Bradshaw, R. A. (1986) Basic
fibroblast growth factor supports the survival of cerebral cortical
neurons in primary culture. Proc. Natl. Acad. Sci. USA. 83, 7537-7541.

Morrison, R. S. and De Vellis, J. (1981) Growth of purified astrocytes in a
chemically defined medium. Proc. Natl. Acad. Sci. USA. 78, 7205-7209.

Schmidek, H. H., Nielsen, S. L., Schiller, A. L. and Messer, J. (1971)
Morphologic studies of rat brain tumors induced by N -
nitrosomethylurea. J. Neurosurg. 34, 335-340.

 

 

135

Schousboe, A. (1981) Transport and metabolism of glutamate and GABA in
neurons and glial cells. Int. Rev. Neurobiol. 22, 1-45.

Schousboe, A., Hertz, L. and Svenneby, G. (1977) Uptake and metabolism of
GABA in astrocytes cultured from dissociated mouse brain
hemispheres. Neurochem. Res. 2, 217-229.

Schousboe, A., Krogsgarrd-Larsen, P., Svenneby, G. and Hertz, L. (1978)
Inhibition of the high affinity uptake, net uptake of GABA into
cultured astrocytes by beta-proline, nipecotic acid and other
compounds. Brain Res. 153, 623-626.

Sellstrom, A. and Hamberger, A. (1975) Neuronal and glial systems for
gamma-aminobutyric acid transport. J. Neurochem. 24, 847-852.

Sellstrom, A. and Hamberger, A. (1977) Potassium stimulated gamma-
aminobutyric acid release from neurons and glia. Brain Res. 119, 189-
198.

Shapiro, D. L. (1973) Morphological and biochemical alterations in foetal rat
brain cells cultured in the presence of monbutyrl cyclic AMP. Nature
(Land) 241, 203-204.

Springer, J. E. (1988) Nerve growth factor receptors in the central nervous
system. Exp. Neurol 102, 354-365.

Stark, L. G., Albertson, T. E. and Joy, R. M. (1986) Effects of
hexachlorocyclohexane isomers on the acquisition of kindled seizures.
Neurobehav. Toxicol. Tcratol. 8, 487-491.

Swenberg, J. A., Koestner, A. and Wechsler, W. (1972) The induction of
tumors of the nervous system with intervenous methylnitrosourea.
Lab Invest. 26, 74-85.

136

Thomas, K. A., Rios-Candelore, M., Gimenez-Gallego, G., DiSalvo,
J.,Bennett,C.,Rodkey, J. and Fitzpatrick, S. (1985) Pure brain derived
acidic fibroblast growth factor is a potent angiogenic vascular

endothelial cell mitogen with sequence homology to interleukin 1.
Proc. Natl. Acad. Sci. USA. 82, 6409-6413.

Thonen, H., Brandtlow, C. and Heumann, R. (1987) The physiological
function of nerve growth factor in the central nervous system:
comparison with the periphery. Rev. Physiol. Biochem. Pharmacol.
109, 146-178.

Unsicker, K., Reichert-Preibsch, H., Schmidt, R., Pettmann, B., Labourdette, G.
and Sensenbrenner, M. (1987) Astroglial and fibroblast growth factors
have neurotrophic functions for cultured peripheral and central
nervous system neurons. Proc. Natl. Acad. Sci. USA. 84, 5459-5463.

Walicke, P. A., Cowan, W. M., Ueno, N ., Baird, A. and Guillemin, R. (1986)
Fibroblast growth factor promotes survival of dissociated hippocampal
neurons and enhances neurite extension. Proc. Natl. Acad. Sci. USA.
83, 3012-3016.

Whittemore, S. R. and Seiger, A. (1987) The expression, localization and
functional significance of beta-nerve growth factor in the central
nervous system. Brain Res. Rev. 12, 439-464.

Wilkin, G. P., Levi, G., Johnstone, S. R. and Riddle, P. N. (1983) Cerebellar
astroglial cells in primary culture: expression of different
morphological appearances and different ability to take up [3H]D-

Aspartate and [3H]GABA. Develop. Brain Res. 10, 265-277.

Yankner, B. A. and Shooter, E. M. (1982) The biology and mechanisms of
action of nerve growth factor. Annu. Rev. Biochem. 51, 845-868.

Yayon, A., Klagsbrun, M., Esko, J. D., Leder, P. and Ornitz, D. M. (1991) Cell
surface, heparin-like molecules are required for binding of basic
fibroblast growth factor to its high affinity receptor. Cell. 64, 841-848.

Chapter 6

An in vitro Study on Glutamate Uptake by Neonatal Rat Astrocytes
and Rat Astrocytoma Cells

William R. Hare Jr.

Department of Pathology
Michigan State University
East Lansing, MI 48824-1316

Telephone, 51 7-353-9160

Telefax no., 517-336-1053

137

138

Abstract

This study reports on glutamate (GLU) uptake by neonatal rat astrocytes
and rat anaplastic astrocytoma cells cultured in insulin-containing chemically
defined media (CDM). Nerve growth factor (NGF) is reported to have no
significant effect on GLU uptake by these cells in CDM. However, GLU
uptake by astrocytes was significantly lower as compared to the astrocytoma
cells when cells were cultured in insulin-containing CDM. The effects of
serum and cold temperature on GLU uptake by these cell types, is also
reported. In addition, GLU uptake by rat astrocytoma cells was found to be an
active uptake process. The potentiation of epileptic-type seizures as well as
the production of neuronal cell death as a consequence of increased levels of
synaptic GLU are well accepted. The experimental data presented here shows
that N GF did not significantly change GLU uptake by neonatal rat astrocytes
or astrocytoma cells. These results suggest that NGF may be useful as a
pharmacologic agent for the treatment of brain tumors without potentiating

an increase of seizures and neuronal cell death.

Running Title: Studies on glutamate uptake by astrocytes.

Key Words: GLU-receptors, GLU-uptake, insulin, nerve growth factor, basic-

fibroblast growth factor, neonatal astrocytes, astrocytoma cells, epilepsy.

 

 

139

Introduction

Glutamate (GLU) is widely recognized as the major excitatory amino
acid neurotransmitter of the central nervous system (CNS). It is now
generally accepted that high levels of GLU are responsible to some degree for
producing seizures and mediating epileptic brain damage and neuronal cell
death (Sloviter et al., 1985; Sloviter, 1986).

The uptake systems of astrocytes provide a means of clearing both GLU
and potassium ions (K+) from the synaptic space following their release by
pre-synaptic and post-synaptic neurons (Schousboe et al., 1981; Schousboe,
1981). The high affinity GLU uptake system is sodium dependent. Sodium
and chloride mediate uptake without activation of a homoexchange
mechanism (Henn et al., 1974; Schousboe et al., 1977; Hertz ct al., 1978;
Waniewski and Martin, 1986). Uptake is not dependent on oxidative
phosphorylation (Waniewski and Martin, 1986). The uptake of GLU is of
particular importance because over-stimulation or persistent stimulation of
neurons through inappropriate release or failure to clear this
neurotransmitter results in the degeneration and death of certain neurons
(Olney, 1969; Rothman, 1984; Simon ct al., 1984; Wieloch, 1985; Choi, 1988;
Faden et al., 1989)-

At least five different types of GLU receptors have been found to exist
in the brain. Many of these receptors are shared by neurons and glia (Hosli et
al., 1979). These include ionotropic receptors which are coupled to the
opening of ion channels in the cell membrane. The ionotropic receptors are

identified by their preferential activation by various analogs as N-methyl-D-

1 40

aspartate (NMDA) receptor, quisqualate (Quis) receptor, ot-amino-B-hydroxy-
5-methyl-4-isoxazolepropionic acid (AMPA) receptor, kainic acid (KA)
receptor and 2-amino-4-phosphonobutyrate (AP4) (Fagg et al., 1986; Johnson
and Koerner, 1988). In addition, a GLU receptor has been reported to be
associated with the hydrolysis of membrane phosphoinositides (Nicoletti et
al., 1986a, 1986b, 1987). This receptor functions in membrane signal
transduction. Activation of this receptor generates 1,4,5-inositol triphosphate
(IP3) and calcium (Ca++) as second messengers. This second messenger-
coupled receptor has been shown to be activated not only by GLU but also by
ibotenate and Quis. This receptor is not however thought to be involved in
the generation of neurotoxicity induced by GLU. Astrocytes are believed to
share only Quis and KA ionotropic receptors as well as the IP3 generating
receptor with neurons (Kettenman and Schachner, 1985; Pearce et al., 1987).

Experimentally, GLU has been shown to produce acute glial and
dendritic swelling and neuronal necrosis following brain treatment (Sloviter
and Dempster, 1985). However, this is in contrast to earlier reports which
supported GLU supplementation not only for improved mental behavior but
for neurologic conditions of epilepsy and mental retardation (Fonnum, 1984).
These earlier reports were supported by GLU's function in the detoxification
of ammonia in the brain (Wiel-Malherbe, 1950). Their is no question that
glutamate functions not only as a neurotransmitter but as an important
constituent of brain metabolism. GLU is consequently thought to be
important in the development of cognitive function and in the maintenance
of mental health. GLU has proven to be particularly important as an
excitatory neurotransmitter transmitter in cortical and hippocampal regions

of the brain (Fonnum, 1984).

141

Insulin is a hormone which serves many functions important to the
maintenance and survival of most mammalian cell systems. Insulin has
been shown to promote the survival and differentiation of fetal neuronal
cells in culture (Hooghe-Peters et al., 1981). However, insulin's most
recognized function is the facilitation of glucose transport.

Insulin does not cross the blood brain barrier. In the brain, glucose
crosses the blood brain barrier by means of a carrier-mediated, non-energy-
requiring facilitated transport. The brain normally maintains a glucose
concentration that is twenty times higher than the concentration necessary to
saturate hexokinase, the first enzyme of the glycolytic pathway. However,
conditions of hypoglycemia could lower the saturation of the glucose carrier
and become the rate-limiting factor for the generation of energy. In addition
to insulin's role in glucose transport it has been found to affect the amino
acid neurotransmitters. Insulin stimulates the sodium-dependent uptake of
the inhibitory amino acid neurotransmitter, y-aminobutyric acid (GABA),
into synaptosomes (Herschman, 1986). Insulin treatment has also been
shown to dramatically reduce cortical and plasma levels of glutamate in
animal studies (Bernasconi et al., 1988)- Glutamate content is reduced 50%
and 90%, respectively, following insulin treatment. In addition several cell
types, with the exception of brain cells, have shown an insulin-dependent
depression of insulin receptor levels (Van Schravendijk ct al., 1984).

Basic-fibroblast growth factor (basic-FGF) has been found to be plentiful
in the central nervous system. Little,is known of the membrane receptor-
induced activities of basic-FGF on astrocytes or astrocytoma cells. It has been
reported to be mitogenic in the induction of endothelial growth (Bohlen et
al., 1985; Lobb et al., 1985; Thomas et al., 1985), but has not been proven to be

mitogenic in astrocytes (Leutz and Schachner, 1981). It has also been found to

 

142

be a neurotrophic factor increasing the survival of neurons and growth of
neurites in culture (Morrison ct al., 1986; Walicke et al., 1986; Unsicker et al.,
1987; Morrison et al., 1988). Therefore, like nerve growth factor, it may be
more important in the preservation and maintenance of astrocytes cultured
under these conditions.

Nerve growth factor (NGF) is recognized as a potent differentiation and
maturation promoter of various neuronal cell types (Levi-Montalcini and
Angeletti, 1968; Greene and Tischler, 1976; Greene and Shooter, 1980; Yankner
and Shooter, 1982; Levi-Montalcini, 1987; Marushige ct al, 1987) as well as an
important factor in the maintenance of selected neuronal cell populations of
the CNS (Thonen et al., 1987; Whittemore and Seiger, 1987; Springer, 1988).
As a result of NGF's dramatic ability to promote differentiation, it has been
suggested as a potential therapeutic agent in the treatment of brain tumors.
Previous reports indicate that increased morphologic differentiation of
astrocytes may also increase neurotransmitter uptake (Wilkin et al., 1983);
therefore, NGF may have therapeutic benefit since increased net uptake of
GLU could potentially decrease the incidence of seizures. Based on these
considerations, the purpose of this study was to determine whether NGF

increases the net uptake of GLU.

 

143

Materials and Methods

Materials

NGF (2.58, grade II) was purchased from Boehringer Mannheim
Biochemicals (Indianapolis, IN,USA). Fetal calf serum (FCS) was purchased
from Hyclone Sterile Systems Inc. (Logan, UT, USA). All culture media were
prepared using stock solutions, chemicals and supplies from Life
Technologies, Inc,GIBCO Labs (Grand Island, NE, USA), Corning Glass Works
Inc. (Corning, NY, USA) and Sigma Chemical Co. (St. Louis, MO, USA). Hl-1
supplement was purchased from Endotronics Inc. (Coon Rapids, MN, USA).
Glutamate was purchased from ICN-Biomedicals Inc. (Costa Mesa, CA, USA)
and [3H]-GLU (25 Ci/mmol) from New England Nuclear, Dupont/NEN
Products. (Boston, MA, USA).

Neonatal Astrocyte Cultures

Neonatal rat astrocytes were generated using 4 day old Fischer rat pups.
The brains of four rat pups were removed following euthanasia and placed in
a petri dish containing warm DMEM high glucose media. The meninges
were removed and the brain stem and cerebellum separated. The cerebral
tissues were minced into small pieces (<0.5 mm), combined and placed into a
sterile centrifuge tube containing 5 ml of DMEM high glucose media. This
suspension of brain tissue was then centrifuged at 750 X g for a period of 10
min. The supernatant fluid was removed and discarded and replaced with 1.5
ml of Collagenase II-S (Sigma) solution (0.8%, w/v). The enzyme treatment

procedure was followed by incubation of the cell mixture on a warm water

 

144

bath (37°C) shaker for 30 min, after which the brain tissue was completely
dissociated to a cell suspension by intermittently pipetting with a sterile glass
pasteur pipette supplied with a cotton filter. At no time during preparation of
these cultures were the cells allowed to cool below 30°C. Following the
enzyme treatment the cell suspension was centrifuged at 200 X g for a period
of 10 min and the supernatant fluid removed and discarded. An equal
amount (approx. 3 ml) of DMEM high glucose medium containing 15% FCS
was gently layered on top of the cell pellet. Then the upper 2/ 3 of the cell
pellet was resuspended in fresh medium using gentle pipetting action. A 1
ml aliquot of this cell suspension was quickly removed and used to seed
primary cultures at approximately a 1:20 split ratio. These stock cultures were
started and maintained in 25 cm2 tissue culture flasks (Corning) containing 5
ml of complete medium. The medium was replaced with fresh nutrient
medium the following day and replaced every 3 days thereafter. Neonatal
astrocytes used in the GLU uptake experiments were at their 3rd passage. The
astrocytic character of these cells was indicated at second and third passage not
only by morphology,and their ability to take up 7-aminobutyric acid, but by

their content of the astrocyte-specific, glial fibrillary acidic protein.

Astrocytoma Cells

The rat anaplastic astrocytoma cells were supplied by Dr. A. Koestner of
the Department of Pathology, Michigan State University, East Lansing, MI.
from his cell storage bank. The T-9 cell line originated from a high grade-
anaplastic astrocytoma induced in Fischer rats by treatment with N-methyl-
N-nitrosourea (MNU) (Schmidek ct al., 1971). Stock cultures were established

in 25 cm2 tissue culture flasks and maintained in complete serum

145

supplemented medium (DMEM and RPMI 1640/ Hams F12) replacement
every 3rd day and the passing of cells every 6th day at a 1:100 split ratio.

Culture Conditions
Cells were cultured in a Hotpack C02 Incubator which was maintained

at 5% C02, 37°C and constant humidity. Both neonatal astrocytes and
astrocytoma cells were split into multi-well (6 well) culture plates (GIBCO
Labs) containing 3 ml of complete medium. This medium helped to assure '
cell attachment. The complete medium, serum supplemented, was changed
the following day and replaced with chemically defined medium (CDM).
CDM HL1A was composed of DMEM supplemented with 1% (v/v) HL-1
supplement, 400 uM glutamine, gentamycin at 10 ug/ ml, glucose at 1 mg/ ml,
CaCl2 at 175 ug/ ml and MgSO4 at 125 ug/ ml. CDM HLIB was also composed
of DMEM and contained the same supplementation as HL1A but also
contained hydrocortisone at 1.6 ug/ ml, prostaglandin F2-alpha at 440 ng/ ml,
putrescine at 78 u g/ ml, basic-fibroblastic growth factor at 8.8 ng/ ml and
myelin basic protein at 440 ng/ ml. This HLIB medium was a slight
modification of that originally proposed by Morrison and deVellis in 1981 as a
CDM that initiates differentiation of astrocytes. HL-1 supplement contains 29
ug/ m1 total protein with 15 u g/ ml insulin and contains no additional growth

factors or glutamate.

GLU Uptake Experiments

Neonatal astrocytes and astrocytoma cells used in these studies were
seeded at 6-12,000 cells/ ml into 3 ml of complete, serum-supplemented,
media at initial plating. The medium was changed the following day to

serum-free CDM. Three days later the CDM was replaced with fresh CDM of

146

the same type and NGF added. N GF was added at 500 ng/ ml or serum added
at 10% (300 ul/ 3 ml medium). Cells were incubated in NGF or serum for 48
hrs prior to the addition of GLU. Each GLU concentration that was studied
(2.5 and 25 nmol) contained 3 uCi of [3H]-GLU per 3 ml of culture medium per
petri dish. [3H]-GLU had a specific activity of (1 uCi/ 40pmol). GLU specific
activity was established at 1 uCi/2.5-25umol. The standard incubation period
for GLU uptake was set at 10 min at room temperature (RT) (20-25°C).

During the uptake incubation, cell culture plates were placed on a Lab-Line
rotary orbit shaker (60 rpm) to assure dispersion of GLU. The incubation was
stopped by removing the [3H]-GLU containing medium and rinsing the
cultures twice in a HEPES buffered saline solution, pH 7.4, which contained
the same concentration of unlabeled GLU. Each rinse was for a period of 10
min and was also carried out on a rotary shaker. Following this rinse
procedure the cells were extracted by placing 2 ml of a freshly prepared
solution of 1M perchloric acid into each well. Cells were extracted for 10 min,
again utilizing the rotary shaker. Cells were then given a post-extraction
rinse using 2 ml of 0.1M Tris buffer, pH 7.4. This rinse solution was
combined with the extract solution. Throughout these procedures cells
remained attached to the culture plate. GLU uptake data were recorded as
counts per minute (cpm) of extract and then normalized using the specific
activity of GLU and cpm/ protein concentration of extract and reported as
nmol GLU/ mg protein/ time (10 min). The experiment was designed to
compare uptake of GLU by neonatal astrocytes and astrocytoma cells cultured
in two different chemically defined media, HL1A and HL1B, and to determine
the effects of NGF, serum and hypothermia (Table 1). Three or more replicate

cultures were preformed for each study.

 

1 47
Scintillation Analysis
The procedure was carried out using a Packard 300 scintillation
counter. Uptake extract solution (0.8 ml) was added to Aquasol II scintillation
fluid (10 ml).

Protein Concentration

The Bio-Rad method of protein analysis was used. Extract uptake
solution (0.8 ml) was combined with 200 ul of reagent and absorbance checked
at 595 nm. This was compared to a standard curve using Bovine IgG and
quantitated accordingly. The average protein concentration for T-9 cultures

was 36.8 ug/ ml. and 38.8 ug/ ml for NR-l cultures.

Statistical Analysis

Determination of statistically significant differences between
experimental groups was performed utilizing the Students-t test. Differences
were considered to be significant when p values of _<_0.05 were obtained.
Standard error of adjusted means (SEM) was also determined and reported

where significance was indicated.

 

148

Results

Effects of Medium on net GLU Uptake

The net uptake of GLU (2.5 and 25 nmol) by NR-l and T-9 cells in HL1A
and HL1B CDM is compared. Cells were cultured in both media and
incubated with GLU for 10 min. There was a significantly lower uptake of
GLU (2.5 and 25umol) by NR-I cells as compared to T-9 cells cultured in both
HL1A and HL1B CDM (Table 1). In addition, there was a significant difference
in GLU uptake (2.5 and 25umol) between HL1A and HL1B media by both cell

types.

Effects of N GF on GLU Uptake

The effect of N GF (500 n g/ ml) on GLU uptake by NR-1 and T-9 cells
cultured in HL1A and HL1B CDM following a 10 min incubation is illustrated
in Table 1. NGF (500 ng/ ml) was not found to significantly change the net
uptake of GLU (2.5 and 25umol) by either NR-l nor T-9 cells.

Effects of Fetal Calf Serum on GLU Uptake

Table 1 illustrates the effects of serum (100 111/ ml medium) on GLU
uptake by N R-1 and T-9 cells cultured in HL1A CDM following a 10 min
incubation. The addition of fetal calf serum (FCS) in place of NGF was found
to significantly increase the net uptake of GLU (25umol) by N R-1 cells.
However, serum had no significant effect on GLU uptake by T-9 cells.

 

149

Effects of Hypothermia on GLU Uptake.

The study of GLU uptake by N R-1 and T-9 cells at temperatures <4°C
was performed to determine the presence or absence of an active uptake
system. Cell cultures were placed on ice for 90 min prior to and during a 10
min incubation with GLU (25umol). Hypothermia was found to cause a
significant decrease in the uptake of GLU (25umol) by T-9 cells cultured in
HL1A CDM while significantly increasing the GLU uptake by NR—l cells
under the same conditions. This study found that under these culture
conditions two different cell types can be shown to have different GLU uptake
systems. The results indicate that GLU uptake by T-9 astrocytoma cells is an

active uptake process.

 

 

150

TABLE 1. GLU uptake by neonatal astrocytes and anaplastic astrocytoma cells
cultured under varying conditions and treatments.

 

 

Culture GLU Uptake by Cell Type
Conditions and Treatment NR-1 T-9
(concentration, temp, medium, treatment) nmol GLU / mg protein/ 10 min)
25 nmol GLU, RT, HL1A 86.0 (2.4) 1726 (307)
2.5umol GLU, RT, HL1A 8.3 (0.9) 234 (30)
25 nmol GLU, RT, HLIB 55.0 (1.3) 1384 (47)
2.5umol GLU, RT, HL1B 6.3 (0.9) 156 (14)
25 nmol GLU, RT, HL1A, NGF 76.7 (12.3) 1581 (38)
2.5umol GLU, RT, HL1A, NGF 6.3 (0.9) 179 (2.6)
25 nmol GLU, RT, HLlB, NGF 63.0 (7.6) 1476 (77)
2.5umol GLU, RT, HL1B, NGF 5.0 (0.6) 175 (31)
25 nmol GLU, RT, HL1A, Serum 217.0 (13.6) 2218 (127)
25 nmol GLU, 4°C, HL1A 153.0 (15.7) 153 (6.1)

 

Statistically significant differences in GLU uptake are shown between
NR-1 and T-9 cells cultured in either HL1A or HL1B medium, between T-9
cells cultured in HL1A and HL1B medium as well as NR-I cells cultured in
different media, between serum treated NR-l cells and untreated cells
cultured in HL1A medium and in T-9 cells in response to cold temperature.
All comparisons were made using the Students-t test with p<0.05. Standard
error of the mean (SEM) is given in brackets.

151

Discussion

The purpose of this study was to compare GLU uptake by NR-l
astrocytes and T-9 astrocytoma cells under varying conditions and treatments,
with a special emphasis on the effect of NGF on GLU uptake. This was
generated by NGF's reported ability to initiate morphologic differentiation
(Greene and Tischler, 1976; Marushige et al., 1989) as well as its reported
ability to arrest proliferation (Marushige ct al., 1987) of different cell types.
These reports generated a hypothesis that NGF may be beneficial in the
treatment of tumors of the nervous system. Because of the importance of
how N GF might affect the uptake of GLU by tumor cells differently than it
does normal cells, this study draws a comparative picture of the differences in
cell type.

Medium preparation is a very important factor in any study where
morphology may be an intricate part of the expected outcome. Morphology
has been previously reported to affect amino acid neurotransmitter uptake
(Wilkin et al., 1983). Different medium preparations have also been shown to
initiate morphologic differentiation (Morrison and De Vellis, 1981). These
reports added to our interests of how different chemically defined media,
which influence morphologic differentiation, might affect GLU uptake.

There have been excellent studies on glutamate uptake by astrocytes
and astrocytoma cells previously reported in the literature, and some have
indicated that glutamate uptake was greater in astrocytes derived from brain
areas in which there was greater glutaminergic input( Drejer et al., 1982, 1983;

Nicklas and Browning, 1983; Hanson, 1986; Waniewski and Martin, 1986;

 

 

152

Schousboe et al., 1987; Flott and Seifert, 1991). However, fewer studies have
been reported showing differences in GLU uptake between astrocytes and
astrocytoma cells studied under identical in vitro conditions. One previous
study has shown that the glutamate carrier in astrocytic primary cultures
exhibits a substrate specificity which is somewhat different from that of
glioma cells (Balcar et al., 1987).

In this experimental model there was a significantly lower uptake of
GLU by neonatal astrocytes as compared to astrocytoma cells (Table 1). This
suggests that one of the factors present in the CDM medium may be
preferentially acting to decrease the uptake of GLU by neonatal astrocytes.
Upon examining the individual constituents of the CDM used in these
experiments, it appears that this action may be related to the presence of
insulin in the medium.

The insulin receptor is a tetrameric glycoprotein consisting of two 95
kD 6 -subunits and two 135 kD oc- subunits held together by a disulphide
bond. The B-subunit of the insulin receptor induces tyrosine kinase activity
(Kasuga et al., 1982). This receptor-mediated kinase activity results in
autophosphorylation as well as the phosphorylation of other appropriate
substrates. Cultured brain cells have been shown to express this type of
insulin receptor.

However, insulin's action on the uptake processes of the amino acid
neurotransmitters may be more involved than just activation of receptor-
mediated tyrosine kinase. Insulin has also been shown to mediate some of its
biologic effects through other membrane-bound cyclic AMP-dependent and
independent protein kinases (Marchmont and Housley, 1980; Walaas et al.,
1981; Heyworth et al., 1983). Therefore, the sequence of insulin's action starts

when insulin binding switches on receptor-mediated tyrosine kinase; this

153

activated receptor then becomes autophosphorylated. This action, in turn,
increases the activity of a kinase system to phosphorylate other substrate
target proteins, setting off a cascade of biological responses (Ebina ct al., 1985;
White et al., 1985; Riedel et al., 1986). In addition to this increase in protein
kinase activity, there are reports that insulin also inhibits [Ca++ Mg++]-ATPase
(McDonald et al., 1982). Therefore, the cellular changes that are brought about
as a result of insulin's interaction with the cell membrane include not simply
its well known'function in glucose transport but also its role in
phosphorylation, dephosphorylation processes as well as in calcium
mobilization.

Other consequences of the initial autOphosphorylation of the insulin-
receptor could be a cascade of cyclic AMP which could activate protein kinase
A leading to phosphorylation of serine and threonine residues or a
phosphoinositide cascade leading to increased levels of inositol trisphosphate
(IP3) with subsequent activation of protein kinase C following calcium
mobilization. Therefore, the dramatic difference in GLU uptake between
astrocytes and astrocytoma cells cultured in insulin-containing CDM may be
due to the phosphorylation and subsequent inhibition of the GLU receptor-
protein binding, brought about by insulin receptor-mediated activation of
tyrosine kinase. In addition, the inability of the T-9 rat astrocytoma cells to
respond to insulin, in a like manner, may be due to an insulin-dependent
depression in the number of insulin receptors, brought about by a receptor
down-regulation mechanism.

NGF treatment did not cause a significant change in the uptake of GLU
by either N R-1 neonatal astrocytes nor T-9 anaplastic astrocytoma cells (Table
1). However, NGF showed, except for one case, a consistent increase in GLU

uptake by cells cultured in HL1B (Table 1). This could be explained by a

 

 

154

synergistic effect being produced by N GF and basic-FGF, a constituent of HLIB
but not HL1A, which changes the affinity of the GLU receptor such that
uptake becomes concentration-dependent. NGF and basic-FGF may both act
through a receptor-linked kinase-mediated phosphorylation. But the
phosphorylation site would be expected to differ from any phosphorylation
mediated by insulin receptor-linked tyrosine kinase which might cause a
decrease in the affinity of the GLU receptor in NR-l astrocytes. NGF and
basic-FGF could mediate phosphorylation at this second site and cause the
GLU receptor-protein's tertiary structure to change there by affecting the
receptor's affinity for GLU binding. This phosphorylation produced by the
combined action of NGF and basic-FGF, dependent on 25umol or higher GLU
concentrations, could be responsible for the increase of GLU uptake by NR-l
cells seen in this study. NGF did not cause a significant change nor a
consistent increase in the uptake of GLU by NR-l astrocytes at 2.5umol GLU
concentration cultured in HL1B nor by astrocytes cultured in HL1A medium.
NGF also had this type of effect on GLU uptake by T-9 astrocytoma cells.
However, none of these changes noted were significant.

These speculations on the action of NGF and basic-FGF are supported
by the data showing a significant increase in the uptake of GLU by NR-1
astrocytes in the presence of serum (Table 1). Fetal calf serum is known to
contain growth factors whose receptors are associated with tyrosine kinase
activity. The availability of high energy phosphate could explain activation of
tyrosine kinase processes and could help explain an increased uptake of GLU
following incubation with serum for 48 hrs.

The hypothermia study was performed in order to block active uptake .
However, under conditions of hypothermia (temp <4°C) NR-l neonatal

astrocytes showed a significant increase, while T-9 anaplastic astrocytoma cells

 

155

showed a significant reduction in GLU uptake (Table 1). These data support
the fact that GLU uptake by T-9 astrocytoma cells is an active uptake process
dramatically different from that expressed by N R-l neonatal astrocytes. GLU
uptake by NR-1 astrocytes appears to be actively suppressed during culture in
these two CDM at room temperature,which is speculated to be caused by
insulin in the media, or NR-1 astrocytes might express an as yet unidentified
receptor, active at cold temperatures and unique to astrocytes.

In summary, GLU uptake by NR-l astrocytes is markedly lower as
compared to GLU uptake by T-9 astrocytoma cells. This response is suspected
to be caused by insulin-dependent mechanism, which may inhibit the uptake
of glutamate by neonatal astrocytes through an active phosphorylation of the
high affinity glutamate receptor. NGF caused no significant change in GLU
uptake by NR-l neonatal astrocytes or T-9 anaplastic astrocytoma cells. GLU
uptake by T-9 astrocytoma cells is an active uptake process that is not
significantly affected by NGF or serum.

The significance of this study is that in vitro glutamate uptake by
normal astrocytes is quite different than that of astrocytoma cells cultured
under the conditions presented here. NGF does not significantly affect the
net uptake of GLU by astrocytoma cells. However, NGF and basic-FGF may
act synergistically to increase GLU uptake by neonatal astrocytes and
astrocytoma cells. This information should be helpful in understanding
functional differences between normal astrocytes and astrocytoma cells as
well as in the development of appropriate therapeutic agents for the
treatment of brain tumors as well as the prevention of seizures and the

prevention of neuronal cell death.

 

156

References

Balcar, V. J., Schousboe, A., Spoerri, P. E. and Wolff, J. R. (1987) Differences
between substrate specificities of L-glutamate uptake by neurons and

glia, studied in cell lines and primary cultures. Neurochem. Int. 10,
213-217.

Bernasconi, R., Chapman, A. G., Martin, P., Sills, M., Williams, M. and
Meldrum, B. S. (1988) Effects of carbamazepine (CBZ) and 2-amino-7-
phosphono-heptanoic acid (2-APH) on insulin-induced convulsions
and increased aspartate levels. In, Frontiers in excitatory amino acid
research: Neurology and Neurobiology (E.A. Cavalheiro, J. Lehmann
and L. Turski, eds.), Vol. 46, pp. 275-278, Alan R. Liss, New York.

Bohlen, P., Esch, F., Baird, A., Jones, K. 1. and Gospodarowicz, D. (1985)
Human brain fibroblast growth factor: Isolation and partial
chemical characterization. FEBS Lett. 185, 177-181.

Choi, D. W. (1988) Glutamate neurotoxicity and diseases of the nervous
system. Neuron 1, 623-634.

Drejer, J., Larsson, 0. M. and Schousboe, A. (1982) Characterization of L-
glutamate uptake into and release from astrocytes and neurons
cultured from different brain regions. Exp. Brain Res. 47, 259-269.

Drejer, J., Meier, E. and Schousboe, A. (1983) Novel neuron-related
regulatory mechanisms for astrocytic glutamate and GABA high
affinity uptake. Neurosci. Lett. 37, 301-306.

Ebina, Y., Ellis, L., Jamagin, K., Edery, M., Graf, L., Clauser, E., Ou, J.,
Masiarz, F., Kah, Y. W., Goldfine, I. D., Roth, R. A. and Rutter, W. J.
(1985) The human insulin receptor cDNA: the structural basis for
hormone-activated transmembrane signaling. Cell 40, 747-758.

 

157

Faden, A. I. P., Demediuk, S., Panter, S. and Vink, R. (1989) The role of
excitatory amino acids and NMDA receptors in traumatic brain injury.
Science 244, 798-800.

Fagg, G. E. Foster, A. C. and Ganong, A. H. (1986) Excitatory amino acid
synaptic mechanisms and neurological function. Trends Pharmacol.
Sci. 7, 357-363.

Flott, B. and Seifert, W. (1991) Characterization of glutamate uptake
systems in astrocyte primary cultures from rat brain. Glia 4, 293-
304.

Fonnum, F. (1984) Glutamate: A neurotransmitter in mammalian brain.
J. Neurochem. 42, 1-11.

Greene, L. A. and Shooter, E. M. (1980) Nerve growth factor-biochemistry
and synthesis and mechanisms of action. Annu. Rev. Neurosci. 3, 353-
402.

Greene, L. A. and Tischler, A. S. (1976) Establishment of a noradrenergic
clonal line of rat pheochromocytoma cells which respond to nerve
growth factor. Proc. Natl. Acad. Sci. USA 73, 2424-2428.

Hannson, E. (1986) Co-cultivation of astroglial enriched cultures from
striatum and neuronal containing cultures from substantia nigra. Life
Sci. 39, 269-277.

Henn, F. A., Goldstein, M. N. and Hamberger, A. (1974) Uptake of the
neurotransmitter candidate glutamate by glia. Nature 249, 663-664.

Herschman, H. R. (1986) Polypeptide growth factors and the CNS. Trends
Neurosci. 9, 53-57.

Hertz, L., Schousboe, A., Boechler, A., Mukerji, S. and Fedoroff, S. (1978)
Kinetic characteristics of the glutamate uptake into normal astrocytes
in culture. Neurochem. Res. 3, 1-14.

158

Heyworth, C. M , Wallace, A. V. and Housley, M. D. (1983) Insulin and
glucagon regulate the activation of two distinct membrane bound cyclic
AMP phosphodiesterases in hepatocytes. Biochem. J. 214, 99-110.

Hooghe-Peters, E. L., Meda, P. and Orci, L. (1981) Co-culture of nerve cells
and pancreatic islets. Developmental Brain Research 1, 287-292.

Hosli, L., Anders, P. F. and Hosli, E. (1979) Action of amino acid transmitters
on cultivated cells of the mammalian peripheral and central nervous
system. J. Physiol (Paris) 75, 655-659.

Johnson, R. L. and Koerner, J. F. (1988) Excitatory amino acid
neurotransmission. J. Med. Chem. 31, 2057-2066.

Kasuga, M., Zick, Y., Blith, D., Karlson, F., Karing, H. and Kahn, C. (1982)
Insulin stimulation of phosphorylation of the 8 subunit of the insan
receptor. J. Biol. Chem. 257, 9891-9894.

Kettenman, H. and Schachner, M. (1985) Pharmacological properties of 7-
aminobutyric acid-, glutamate-, and aspartate-induced depolarization
in cultured astrocytes. J. Neurosci. 5, 3295-3301.

Leutz, A. and Schachner, M. (1981) Epidermal growth factor stimulates
DNA-synthesis of astrocytes in primary cerebellar cultures. Cell Tissue
Res. 220, 393-404.

Levi-Montalcini, R. and Angeletti, P. (1968) Nerve growth factor. Physiol.
Rev. 48, 534-569.

Levi-Montalcini, R. (1987) Nerve growth factor: thirty-five years later.
EMBO J. 6, 1145-1154.

Lobb, R. R., Alderman, E. M. and Fett, J. W. (1985) Induction of angiogenesis
by bovine brain derived class 1 heparin-binding growth factor.
Biochemistry 24, 4969-4973.

159

Marchrnont, R. and Housley, M. (1980) Insulin activates a rat liver
peripheral plasma-membrane cyclic-AMP phophodiesterase by a
phosphorylation mechanism. Biochem. Soc. Trans. 8, 537-538.

Marushige, Y., Marushige, K., Okazaki, D. L. and Koestner, A. (1989)
Cytoskeletal reorganization induced by nerve growth factor and glial

maturation factor in anaplastic glioma cells. Anticancer Res. 9,
1143-1148.

Marushige, Y., Raju, N. R., Marushige, K. and Koestner, A. (1987)
Modulation of growth and morphological characteristics in glioma
cells by nerve growth factor and glia maturation factor. Cancer Res. 47,
4109-4115.

McDonald, J., Chang, K-M., Goewert, R., Mooney, R. and Pershadsingh, H.
(1982) The [Ca2+ +Mg2+]-ATPase of adipocyte plasma membrane:
regulation by calmodulin and insulin. In, Transport ATPases (E.
Carafoli and A. Scarpa, eds.). Ann. N. Y. Acad. Sci.,Vol. 402, pp. 381-401,
N. Y. Acad. Sci., New York.

Morrison, R. S. and De Vellis, J. (1981) Growth of purified astrocytes in a
chemically defined medium. Proc. Natl. Acad. Sci. USA. 78, 7205-
7209.

Morrison, R. S., Keating, R. F. and Moskal, J. R. (1988) Basic fibroblast growth
factor and epidermal growth factor exert differential trophic effects on
CNS neurons. J. Neurosci. Res. 21, 71-79.

Morrison, R. 8., Sharma, A., De Vellis, J. and Bradshaw, R. A. (1986) Basic
fibroblast growth factor supports the survival of cerebral cortical
neurons in primary culture. Proc. Natl. Acad. Sci. USA. 83, 7537-
7541.

Nicklas, W. J. and Browning, E. T. (1983) Glutamate uptake and metabolism
in C-6 glioma cells: alterations by potassium ion and dibutyrl-CAMP. J.
Neurochem. 41, 179-187.

160

N icoletti, F., Iadarola, M. J., Wroblewski, J. T. and Costa, E. (1986) Excitatory
amino acid recognition sites coupled with inositol phospholipid
metabolism: developmental changes and interaction with o: 1-
adrenoreceptors. Proc. Natl. Acad. Sci. USA 83, 1931-1935.

N icoletti, F., Meek, J., Iadarola, M., Chuang, D., Roth, B. and Costa, E. (1986)
Coupling of inositol phospholipid metabolism with excitatory amino
acid recognition sites in rat hippocampus. J. Neurochem. 46, 40-46.

N icoletti, F., Wroblewski, J., Alho, H., Eva, C., Fadda, E. and Costa, E. (1987)
Lesions of putative glutaminergic pathways potentiate the increase of
inositol phospholipid hydrolysis elicited by excitatory amino acids.
Brain Research 436, 103-112.

Olney, J. (1969) Brain lesions, obesity, and other disturbances in mice treated
with monosodium glutamate. Science 164, 719-721.

Pearce, B. R., Albrecht, J., Morrow, C. and Murphy, S. (1987) Astrocyte
glutamate receptor activation promotes inositol phospholipid
turnover and calcium flux. Neurosci. Lett. 72, 335-340.

Riedel, H., Dull, D., Schlesinger, J. and Ullrich, A. (1986) A chimaeric
receptor allows insulin to stimulate tyrosine kinase activity of
epidermal growth factor receptor. Nature 324, 68-70.

Rothman, S. (1984) Synaptic release of excitatory amino acid
neurotransmitter mediates anoxic neuronal death. J. Neurosci. 4,
1884-1887.

Schmidek, H., Nielsen, S., Schiller, A. and Messer, J. (1971) Morphologic
studies of rat brain tumors induced by N-nitrosomethylurea. J.
Neurosurg. 34, 335-340.

161

Schousboe, A. and Hertz, L. (1981) Role of astroglial cells in glutamate
homeostasis. In, Glutamate as a neurotransmitter (G. Di Chiara and G.
L. Gessa, eds.). Adv Biochem Psychopharmacol,, vol 27, pp. 103-113,
Raven Press, New York.

Schousboe, A., Drejer, J. and Hertz, L. (1987) Uptake and release of glutamate
and glutamine in neurons and astrocytes in primary cultures. In,
Glutamine and Glutamate in Mammals (E. Kvamme, ed.), Vol. 2, pp.
21-39, CRC Press, Florida.

Schousboe, A., Svenneby, G. and Hertz, L. (1977) Uptake and metabolism of
glutamate in astrocytes cultured from dissociated mouse brain
hemispheres. J. Neurochem. 29, 999-1005.

 

Schousboe, A. (1981) Transport and metabolism of glutamate and GABA in
neurons and glial cells. Int. Rev. Neurobio. 50, 1-45.

Simon, R., Swan, J., Griffiths, T. and Meldrum, B. (1984) Blockade of N -
methyl-D-aspartate receptors may protect against ischemic damage in
the brain. Science 226, 850-852.

Sloviter, R. and Dempster, D. (1985) Epileptic brain damage is replicated
qualitatively in the rat hippocampus by central injection of glutamate
or aspartate but not by GABA or acetylcholine. Brain Res. Bull. 15, 39-
43.

Sloviter, R. (1986) On the role of seizure activity and endogenous excitatory
amino acids in mediating seizure-associated hippocampal damage. In,
Excitatory Amino Acids and Epilepsy (R. Schwarcz and Y. Ben-Ari,
eds.), vol. 203, pp. 659-671, Plenum, New York.

Springer, J. (1988) Nerve growth factor receptors in the central nervous
system. Exp. Neurol 102, 354-365.

 

162

Thomas, K., Rios-Candelore, M., Gimenez-Gallego, G., DiSalvo,
J.,Bennett,C.,Rodkey, J., and Fitzpatrick, S. (1985) Pure brain derived
acidic fibroblast growth factor is a potent angiogenic vascular
endothelial cell mitogen with sequence homology to interleukin 1.
Proc. Natl. Acad. Sci. USA. 82, 6409-6413.

Thonen, H., Brandtlow, C. and Heumann, R. (1987) The physiological
function of nerve growth factor in the central nervous system:
comparison with the periphery. Rev. Physiol. Biochem. Pharmacol.
109, 146-178.

Unsicker, K., Reichert-Preibsch, H., Schmidt, R., Pettmann, B., Labourdette, G.
and Sensenbrenner, M. (1987) Astroglial and fibroblast growth factors
have neurotrophic functions for cultured peripheral and central
nervous system neurons. Proc. Natl. Acad. Sci. USA. 84, 5459-5463.

Van Schravendijk, C., Hooghe-Peters, E., de Meyts, P. and Pipeleers, D. (1984)
Identification and characterization of insulin receptors on foetal-mouse
brain cortical cells. Biochem. J. 215, 165-172.

Walaas, 0., Horn, R., Lystad, E. and Adler, A. (1981) ADP-ribosylation of
sarcolema membrane proteins in the presence of cholera toxin and its

influence on insulin-stimulated membrane protein kinase activity.
FEBS Lett. 128, 133-136.

Walicke, P., Cowan, W., Ueno, N ., Baird, A. and Guillemin, R. (1986)
Fibroblast growth factor promotes survival of dissociated hippocampal

neurons and enhances neurite extension. Proc. Natl. Acad. Sci. USA.
83, 3012-3016.

Waniewski, R. and Martin, D. (1986) Exogenous glutamate is metabolized
to glutamine and exported by rat primary astrocyte cultures. J.
Neurochem. 47, 304-313.

 

163

White, M., Maron, R. and Kahn, C. (1985) Insulin rapidly stimulates
tyrosine phophorylation of a Mr-185,000 protein in intact cells. Nature
318, 183-186.

Whittemore, S. and Seiger, A. (1987) The expression, localization and
functional significance of beta-nerve growth factor in the central
nervous system. Brain Res. Rev. 12, 439-464.

Wiel-Malherbe, H. (1950) Significance of glutamic acid for the metabolism of
nervous tissue. Physiol. Rev. 30, 549-568.

Wieloch, T. (1985) Hypoglycemia-induced neuronal damage prevented by an
N-methyl-D-aspartate antagonist. Science 230, 681-683.

Wilkin, G., Levi, G., Johnstone, S. and Riddle, P. (1983) Cerebellar

astroglial cells in primary culture: expression of different
morphological appearances and different ability to take up [3H]D-

Aspartate and [3H]GABA. Developmental Brain Research, 10, 265-277.

Yankner, B. and Shooter, E. (1982) The biology and mechanisms of action
of nerve growth factor. Annu. Rev. Biochem. 51, 845-868.

 

 

 

Chapter 7

Summary and Conclusions

Oncogene Expression and Cell Proliferation

Carcinogenesis is an extremely complicated pathologic process. In the
brain, carcinogenesis is even more complicated because it can interfere with
and disrupt normal neurophysiology. Seizures, as well as other
neuropsychiatric symptoms could, therefore, be induced by brain tumors.
Alkylnitrosourea induced brain tumors in the rat provide an excellent
animal model for the study of brain tumors in humans and other animal
species. This model provides researchers the opportunity to study the
complete pathogenesis of brain tumors from initiation through progression.
In addition, the combination of in vivo and in vitro studies, using this
model, allows researchers to gain information on the molecular mechanisms
responsible for cell transformation and tumor progression, as well as to study
treatment and prevention.

It was not long ago that the discovery of tumor-causing genes called
oncogenes gave renewed hope that cancer might be prevented and controlled
by regulating the expression of certain genes. Later a set of related genes,
termed proto-oncogenes or cellular oncogenes (c-onc), were found to
normally exist in the genome of higher animals. Since then, these cellular
oncogenes have been found to code for two classes of protein. One class of
proteins acts at the nuclear level, while the other acts in the cytoplasm and
cytoplasmic membrane. However, the true function of many oncogenes

remains unknown. The oncogene products have been assumed to act on

164

165

specific nuclear and cytoplasmic cell targets because of their distribution and
localization within the cell. In brain tumors src, ras, sis, neu and crb
oncogenes have been found to express protein products which are related to
signal transduction. These proteins are present in the cytoplasm and
membrane compartments of cells (Rohrschneider and Gentry, 1984).
Oncogene products, which are components of the signal transduction
pathways, in turn play an important role in controlling normal growth and
differentiation. Other oncogenes have been found to express protein products
with a nuclear distribution. These proteins are assumed to have a nuclear
function (Eisenman et al, 1985). In brain tumors, the nuclear acting proteins
of the myc and fos oncogenes have been identified. Expression of these
oncogene proteins has been found to correlate to changes in cell growth
(Campisi et al, 1984; Kelly ct al, 1983; )

These reports just mentioned suggest that oncogenes may be involved
in the initiation of carcinogenesis. Therefore, this information led quickly to
the hypothesis that chemical agents which initiate cancer might act by
mutating proto-oncogenes to produce activated oncogenes. These activated
oncogenes in turn could be responsible for the abnormalities observed in
growth control and differentiation. This hypothesis has been confirmed, in
fact, in several types of tumors produced in rats and mice by chemical
carcinogens. This hypothesis especially proved true for those agents which
could cause base substitutions at specific sites in oncogenes. However, there
remains a considerable variation in experimental tumor studies. Cell
heterogeneity, between cells of tumors in the same tissues, different tissues or
different species, all show variation as to the type and extent of mutation and
oncogene involvement. In addition, there are other gene targets, which are

also exposed to mutations and could play as important a role as the

 
 

 

166

oncogenes. These targets are represented by the tumor suppressor genes and
transcriptional regulatory sequences (Weinstein, 1987). Therefore, it is
difficult to assume that mutations of proto-oncogenes are the major
mechanism of initiation. Given the number of signal transduction pathways
and the overall complexity of growth regulation and differentiation, it has
been suggested that more than 300 proto-oncogenes may exist in the
mammalian genome (Weinstein, 1988). As a result, oncogenes and oncogene
products may represent only a small fraction of a huge complex association of

mechanisms responsible for normal cell function, differentiation and growth.

 

In addition to changes in oncogene expression, many tumors also
exhibit a number of karyotypic variations, such as translocations, deletions
and amplifications (Barbacid, 1986). These genetic abnormalities have been
characterized in tumors for quite some time. The chromosomal changes
appear to influence carcinogenesis through increased gene dosage. Tumors
have long been reported to express a tremendous variety of aneuploidy and
chromosomal variation. Therefore, it appears that what is common among
tumor cells is definitely their heterogeneity.

One of the principal objectives of this study was to determine if nerve
growth factor (NGF) had an effect on proliferation. The results presented in
Chapter 2 indicated that N GF does affect the proliferation of astrocytoma cells.
NGF has been shown by others to induce a rapid but transient increase in the
expression of the fos oncogene product in PC12 cells (Kruijer et al, 1985).
Transcripts of fos mRNA can be detected 5 min after treatment with NGF (50
ng/ ml) and are maximally abundant at 30 min and then are found to decrease
(Kruijer et al, 1985). Other agents and conditions have also been reported to
induce the expression of the c-fos oncogene. These include cyclic-AMP and

epidermal growth factor, as well as K” depolarization. However, artificial

 

 

167

glucocorticoids, which also induce differentiation, do not induce c- as
oncogene expression (Kruijer et al, 1985). NGF induces PC12 cells to acquire
properties of sympathetic neurons (Greene and Tischler, 1976; Dichter et al,
1977). In contrast to NGF, artificial glucocorticoids induce PC12 cells to
acquire the characteristics of chromaffin cells (Schubert et al, 1980). Therefore,
future differentiation studies involving the effects of NGF on astrocytoma
cells should examine the expression of the c-fos oncogene to determine if
NGF concentration (5 ug/ ml and 50 ng/ ml) makes a difference in the extent
of c-fos oncogene expression. In addition, morphologic differentiation by
NGF at high levels and low levels of NGF concentration should be examined
to determine changes and differences in the induction process.

Astrocytoma cell proliferation was investigated by using acridine
orange (A0) flow cytometry. The technique used had the advantage of
characterizing cells in the cell cycle, while at the same time distinguishing
distinct cell populations by their expression of genes including oncogenes in
the cellular genome. Future studies should determine how these changes in
gene expression, especially changes in c-fos and c-myc, relate to changes in cell
cycle and proliferation potential.

N on-dividing, quiescent cell populations (G0) are characterized by
having a constant amount of DNA (2n) throughout their life span.
Proliferating cell populations are characterized by changes in their DNA
cycling and division into daughter cells. The content of DNA in proliferating
populations of cells increases during interphase from the diploid
chromosome complement (2n) to a tetraploid chromosome amount (4n).
DNA synthesis normally occurs only during the S phase of the cell cycle. This
S phase is preceded and followed by two separate stages of interphase which

are characterized by their differences in amount of chromosomal DNA.

 

 

 

168

These separate stages of interphase can be distinguished as subpopulations
and are referred to as G1 and G2, respectively. G2 cells have twice the
chromosomal DNA of G1 cells and are quite easily identified. However, G1
and Go cells have the same relative amount of chromosomal DNA and,
therefore, can not be distinguished. Quiescent (Go) cells in a proliferating
population of cells are actually dead cells. Cells which make up a true
quiescent (Q) subpopulation are, for the most part, Go cells. These cells are
derived from G1 cells but have withdrawn from the proliferating pool.
However, Go can not be distinguished from G1 by quanitation of DNA and,
therefore, quiescent cell subpopulations are identified as G0, G1. However,
quiescent cells differ in gene expression from cycling cells and therefore may
be distinguished by characterization of ss and dsDNA. The G2 phase
(subpopulation of interphase) of the cell cycle is followed by the mitotic (M)
phase and represents cell division. This M phase also contains cells with a 4n
DNA content. Therefore, cells of the G2 and M subpopulation also can not be
distinguished by DNA content. As a result of DNA characterization by
content and type (ssDNA, dsDNA), three distinct subpopulations of cells
should be distinguished. Proliferating cell populations are composed of a G0,
G1 subpopulation followed by a G2 + M subpopulation. S represents cells in
DNA synthesis and is found in the area between the two subpopulations. A
quiescent population of cells, which are non-dividing, are also represented by
G0, G1. However, this quiescent population is not followed by a G2 + M
subpopulation. In addition, cells in a quiescent population are characterized
by their greater stage of differentiation and degree of gene expression.
Therefore, quiescent cells should contain greater amounts of single-stranded

DNA.

 
 

169

Flow cytometry is a relatively new development in biophysics which
results in an efficient and accurate method for analyzing DNA in individual
cells (VanDilla et al, 1969,1975). Flow cytometry has the advantage of
measuring both single stranded and double stranded DNA (ssDNA, dsDNA)
by quantifying the intensity of fluorescence emitted by DNA-bound acridine
orange (A0) dye. A0 bound DNA, excited with blue light, fluoresces red
when bound to ssDNA and fluoresces green when bound to dsDNA. A
population of cells can, therefore, be characterized by their ssDNA to dsDNA
content. Each specific cell type is different because of a variation in gene
expression. This variation can be characterized by variation in the amount of
ssDN A by different regions of the genome and by the extent of gene
expression . This variation in ss and dsDNA, which in turn relates to gene
expression and differentiation can be detected by flow cytometry, while
carrying out cell cycle analysis based on DNA content. A proliferating cell
population, which is cycling, can be easily distinguished from a non-cycling,
quiescent cell population. In addition, closely related cell types, which
originate from a common precursor but which differ in proliferation
potential and gene expression, can also be distinguished from one another. .
Cell populations with greater proliferation potential and less differentiation
will have a greater ratio of dsDNA to ssDNA than a cell population with less
proliferation potential and greater differentiation. Cells characterized by
using these types of data analysis are illustrated by a scatter graph histogram.
These histograms clearly show the proliferation potential and degree of
differentiation between cell populations.

The technique of flow cytometry shown here works very well to
distinguish subpopulations of astrocytes and astrocytoma cells. Flow

cytometry could be utilized to an even greater extent in future studies by use

 

170

of a dual laser system. This would allow for the detection of specific proteins,
especially c-onc products, at the same time as cell cycle analysis.

The brain tumor model used here as an example, provides an excellent
model for continued research into cell cycle kinetics and the molecular

mechanism of action of reverse transforming agents like NGF.

Morphology and Intermediate Filaments

Alkylnitrosoureas, including MNU, impart their carcinogenic behavior
by producing alkylation of DNA bases . MNU exerts its effects by generating a
reactive intermediate released during its decomposition. The ultimate
electrophile produced by this process is the methyldiazonium ion (Kleihues
and Magee, 1973). The main product of MNU exposure has proven to be N7-
and 06-methylguanine. The N7-methyl alkylation of guanine is easily
removed from chromosomal DNA by non-enzymatic depurination while the
06-methyl alkylation is not (Kleihues ct al, 1982). The potentially mutagenic
product of MN U exposure has proven to be 06-methylguanine (Singer, 1975).
This DNA lesion leads to misincorporation of the uridine and adenine bases
during RNA co-polymer synthesis (Gerchman and Ludlum, 1973). Although
RNA polymerase normally makes a good copy of the DNA polynucleotide
template, based on the Watson-Crick base-pairing model, the presence of 0°-
methylguanine leads to extensive misincorporation. These nucleic acid base
pair substitutions are believed to cause the mutagenic and carcinogenic
events of this transformation. Such molecular events in the DNA may also,
therefore, be responsible for the aberrant synthesis of IFs.

The promoter sequence and transcriptional startpoint of the GFAP IF

gene (mouse) has been mapped out along with characterization of promoter

 

 

.., — -—~.n..:..e.—-~o -_-‘...:- -4. «m _ .

171

function (Miura et al, 1990). Three trans-acting binding sites of the promoter
region have been defined by DNase I footprinting and identified as GF I, GF II
and GF III. The GF III binding site has been found to be cyclic-AMP
responsive. Cyclic-AMP functions in this instance as an enhancer binding
protein for this specific promoter region. NGF is known to increase cellular
levels of cyclic-AMP (Schubert and Whitlock, 1977; Cremins ct al, 1986).
Therefore, NGF may enhance the promotion of GFAP IFs through its
generation of cyclic-AMP.

Mutations, specifically to the GF II binding site, have been found to
drastically reduce promoter activity, while base substitutions in GF I and GF
III were found to abolish the cell-specific expression of GFAP altogether
(Miura et al, 1990). Loss of cell-specific expression has since been shown to
often result in GFAP promoter expression even in non-GFAP producing cells
(Miura et al., 1990). The GFAP gene sequence has been analyzed and found to
be heavily concentrated with guanine bases. The GF II promoter region, for
example, is a 20 base pair sequence which contains a very high concentration
(50%) of guanine. Mutations to this region could severely affect the
expression of GFAP IFs. Therefore, exposure to MNU and the generation of
06-methylguanine could not only result in changes of the protein structure of
IFs, but could drastically reduce GFAP IF expression altogether. These
mutational changes could also be responsible for more subtle changes in
protein solubility characteristics as well as for changes in the expression of
GFAP epitopes. Changes in epitope specificity have been reported to be the
result of abnormal glycosylation (Hughes and Sharon, 1978). However,
epitope expression and concentration may be adversely affected by additional

mutations involving guanine.

 

172

Since their discovery, IFs have been reported to be constituents of the
cytoskeleton and to play a structural role in cells. However, more recent
reports have indicated that this may not be so. The microinjection of
antibodies directed specifically against the IF protein has been reported to
cause a collapse of the IF network without affecting the cytoskeleton and cell
morphology (Klymkowsky et al, 1983). As a result of this, many alternative
functions for IFs have been suggested. These functions assumed that IF
subunit proteins are responsible for targeted cellular functions and that the
formation of filaments only represented a storage and transportation form of
a protein messenger that could be released from the filamentous polypeptide
in response to intra- and extra-cellular signals (Traub, 1985).

Studies to determine IFs function have shown GFAP and vimentin to
both be nucleic acid binding proteins (Vorgias et al, 1983). In addition, the N-
terminal arginine residues of the IF protein subunits have been shown to
play an important part in filament assembly and disassembly. These subunit
proteins have been further shown to be processed for biological activity by
cleavage of this arginine-rich region by a Ca++-activated, neutral thiol
protease, which also inhibits them from re-assembling into filaments (Traub
and Vorgias, 1983,1984).

More detailed studies have reported an interaction of IFs with
ribosomes. IF protein subunits have been shown to bind with high affinity to
ribosomal RNA (rRNA). There is a preferential binding to 18s rRNA, which
suggests that these IF protein subunits may be single stranded (ss) nucleic
acid-binding proteins. The binding of these IF subunit proteins to rRNA was
also found to be influenced by nucleic acid base composition. The subunit
proteins were found to bind to DNA and RNA sequences with increasing

intensity as a function of their guanine content (Traub et al, 1983).

 

 

 

173

The results of GFAP and vimentin immunofluorescence studies,
presented in this report, illustrate and support these previous studies. The
binding of vimentin and GFAP IF proteins to rRNA may be illustrated by the
greater fluorescent intensity of the perinuclear and nucleolar regions of both
neonatal astrocytes and anaplastic astrocytoma cells.

Cells showing nucleolar immunocytochemical-staining have
morphologic characteristics suggesting them to be protoplasmic type
astrocytes. Astrocytes showing a stellate morphology are more characteristic
of fibrous astrocytes. Fibrous astrocytes did not show intense fluorescence of
nucleoli, however, they did show intense cellular immunofluorescence as
well as nuclear fluorescence. These results suggest that one IF protein may
share GFAP and vimentin epitopes in common and be expressed by
protoplasmic type astrocytes and anaplastic astrocytoma cells. In addition, the
GFAP epitope of this IF protein may be preserved by paraformaldehyde
fixation and degraded by alcohol fixation. This is further speculated to
involve aberrant changes in glycosylation, which imparts epitope specificity.

The expression of IF proteins by MNU-induced astrocytoma cells may
be further complicated by the transformation properties and molecular
involvement of guanine and guanine base substitution. Fibrous type
astrocytes may demonstrate an end-stage developmental process and
therefore, may express IF proteins with GFAP-specific epitopes which differ
from GFAP epitopes expressed by protoplasmic type astrocytes. This
expression of GFAP would again be independent of its processing and
glycosylation.

Besides the affinity of IF protein subunits for single-stranded rRNA
and ssDNA, these protein subunits exhibit a strong tendency to react with

histones (Traub, 1985). It has been reported that vimentin IF subunit proteins

 

 

174

induce a conformational change in histone-histone complexes rendering
them more sensitive to enzyme-induced proteolysis (Traub, 1985). This
finding also appears to be very significant and if truly valid, recognizes GFAP,
vimentin and other IF proteins as messengers of nuclear regulation.
Therefore, IF protein subunits may be capable of interacting with separate and
different constituents of chromatin, DNA and rRNA nucleic acids as well as
histone proteins. As a result, these interactions may involve two
functionally different binding sites, a nucleic acid-binding site and an acceptor
site for arginine rich polypeptides.

The significance of IF proteins having a nuclear function could be
acknowledged by future studies showing their involvement in regulation of
mitogenesis, gene expression or nuclear rRNA transport. Flow cytometry
techniques could be a useful tool in substantiating these speculations.

Previous studies in the department of pathology and elsewhere have
used high concentrations of NGF to induce morphologic differentiation in
glioma cells (Marushige at al., 1989). Normal concentrations of NGF in the
rat brain have been reported to be 0.5-2.5 ng/ g wet wt (Greene, 1977; Korsching
et al., 1985). In other studies involving NGF and epidermal growth factor
(EGF) a much lower concentration has been reported to be used (Leutz and
Schachner, 1981; Boonstra et al., 1983). It appears that NGF at 5000 ng/ ml may
be acting on rat anaplastic astrocytoma cells to induce uniform differentiation
characterized by the formation of intracellular bridges, activation of
cytoskeletal proteins and reduction of the nuclear-cytoplasmic ratio as a result
of an expanding cytoplasm with significant process development. However,
this response may represent an overexerted action by NGF to induce a
morphologically unique and uniform cell population but one that is still

quite different than what would be expected of a heterogeneous population

 

175

represented by rat fibrous and protoplasmic astrocytic cell types. These cells
are of neural ectoderm origin and, therefore, should be expected to express
NGF receptors and respond. However, use of NGF as a reverse transforming
agent should not unnecessarily include the morphologic differentiation of
anaplastic astrocytoma cells to different appearing but otherwise abnormal
cell types. Experimental studies involving NGF's action as a reverse
transforming agent, therefore, might better be carried out at a more
physiologic NGF concentration. The data presented here supports this
recommendation and strongly suggests that NGF may act silently to facilitate
the action of non-specific effectors on rat astrocytes and astrocytoma cells.
Whether NGF acts in addition to other effectors of membrane transduction,
acts by allowing certain specific biochemical reactions to take place, or by
simply acting to organize that which would takes place irregardless of its

presence is only for speculation at this stage of NGF research.

GLU and GABA Uptake Functions

Evidence has accumulated that indicates that the amino acid
neurotransmitters (GLU and GABA) and glutamine flow between two
metabolic compartments in the CNS. There is, in addition, a counter flow
mechanism, whereby these neurotransmitters flow in one direction and are
compensated by a flow of glutamine in the opposite direction. This concept of
metabolic compartmentation represents and distinguishes metabolic
differences between neurons and astrocytes (Waelsch ct al., 1964; Lajtha ct al.,
1959; Berl et al, 1961). Since introduction of the compartmentation concept, it
has been shown that the brain contains at least two different sub-

compartrnents or pools, a large pool which contains glutamate and a smaller

176

pool which also contains glutamate (Garfinkle, 1972; Van Den Berg et al.,
1974; Clarke et al., 1974). This smaller pool, however, contains a much
smaller fraction of total glutamate and gives rise to the rapid formation of
glutamine. However, glutamine from this pool does not, necessarily, contain
all of the cell's glutamine (Cremer et al., 1974).

There have been many reports indicating that numerous metabolites
enter into each of these glutamate pools. Glucose and pyruvate, for example,
have been found to enter both large and small pools where they can be
metabolized to either glutamate or glutamine. On the other hand butyrate,
citrate, bicarbonate, ammonia, succinate and others, as well as synaptic GLU
and GABA, predominantly enter the small pool (Van Den Berg, 1972; Blazas
et al., 1972; Mohler et al., 1974; Cheng and Bruenner, 1974; Clarke et al., 1974).

GABA, transferred from one compartment to another, is compensated
for by intermediates of the Krebs cycle. This metabolic interaction has been
referred to as the GABA-glutamine shunt. In addition, the fact that GABA
and glutamine function as links between two compartments has directed
special attention to this shunt mechanism and its regulation. In mouse brain,
this shunt has been shown to account for 8-10% of the total flux through the
Krebs cycle. Glutamate has also been found to cycle. GLU is released by '
neurons and taken up by astrocytes. It is then metabolized to glutamine and
cycled back to the neurons as glutamine, where it is hydrolyzed to regenerate
GLU or GABA and ammonia. This cycling of GLU has often been referred to
as the glutamate-glutamine shunt.

The importance of including a study on GLU and GABA uptake here is
due to the possibility that neoplastic transformation interferes with this
normal uptake process and that NGF treatment may normalize such an

uptake process. Many would agree that differences in uptake would be

 

177

expected because these are in fact different cells. Molecular anatomy as well as
physical and chemical characteristics of receptors and uptake processes are
certainly as great a target for mutational events of alkylnitrosourea as
intermediate filaments or components of the nucleus which regulate
proliferation. There are key enzymes in this uptake process and metabolic
system that are equally important. Glutaminase, glutamine synthetase and
glutamate decarboxylase are key enzymes for GLU, glutamine and GABA
respectively. These enzymes could constitute yet additional targets for
mutational events that could be expected to drastically affect uptake.

Regional differences in uptake function of astrocytes have been well
established by other studies (Monaghan et al, 1983). These differences in
uptake are in turn brought about by differences in the extracellular
environment and the influence of specific neurons. Therefore, studies on
uptake function are best done in a controlled environment, where variation
can be minimized. As a result, functional comparisons of GLU and GABA
uptake undertaken in this study between astrocytes and astrocytoma cells
should yield significant results pertaining to specific cell types and the effects
of NGF on uptake processes. Additional information may also be gained in
future studies relative to the causation of seizures and other types of impaired
neurologic function which may accompany specific types of cellular induced
pathology.

In the dog as well as the human the most prominent clinical sign at the
time of initial presentation for astrocytomas is a seizure. However, seizures
especially appear to be influenced by tumor location. Tumors occupying the
temporal, parietal and occipital lobes of the brain have a high correlation to
seizure induction. Tumors occurring in other areas of the brain have a high

correlation to other specific behavioral symptoms. If GABA, the major

 

178

inhibitory amino acid neurotransmitter, was to decrease in synaptic
concentration, due to an increase in astrocyte or astrocytoma cell uptake, it
could be sufficient to induce seizures. GABA concentrations in the rat
cerebral cortex are reported to be 0.5 umol/ g wet wt. and in the human 0.8-
2.3 umol/ g wet wt.. In human glioma tissues GABA concentrations have
been further reported to decrease to 0.5 umol/ g wet wt.. This suggests that a
significant change in GABA uptake and metabolism may exist.

GLU is the major stimulatory amino acid neurotransmitter. If synaptic
concentrations were to increase due to a decrease in astrocyte or astrocytoma
cell uptake, it also might be sufficient to induce seizures and result in brain
damage (Sloviter and Dempster, 1985). However, GLU appears to have an
even greater influence on mental function. Abnormal increases in synaptic
concentrations of GLU have also been correlated to spreading depression and
abnormalities in memory formation (Bures et al., 1974; VanHarreveld and
Fifkova, 1974; Cherkin and VanHarreveld, 1978; Hertz, 1979). The later in
turn may suggest that abnormalities in astrocyte function may be directly
related to symptoms of Alzheimer's disease and loss of immediate memory
capacity. GLU concentrations in the rat cerebral cortex are reported to be
2.1 umol/ g wet wt. and in the human 7.8-12.5umol/ g wet wt. In human
glioma tissues GLU concentration has been reported to be 4.2 umol/ g wet wt.

Decreases in GLU and GABA concentrations may result in important
changes to compartmentation and in the availability of metabolites from the
glutamate pools. If the small glutamate pool, for the immediate conversion
to glutamine, is disrupted, the supply of amino acid precursors to neurons is
also depleted. This suggests that glucose-derived glutamate would become
the major resource for GLU and GABA generation. This also suggests that

the large glutamate pool would increase and could become a source of

 

 

179

intermediary energy metabolism for these tumor cells. If nothing more, these
speculations suggest that further experimentation is necessary in the area of
GLU and GABA uptake.

The uptake studies presented here determined the uptake of GLU and
GABA at 2.5umol and 25umol concentrations of GLU or GABA. The lower
concentration was used to establish a normal appearing, resting concentration
and to observe the net effects of NGF on high affinity uptake mechanisms.
The higher concentration was used to determine uptake function at a
stimulatory or inhibitory concentration and to observe NGF's effects on low
affinity uptake mechanisms. It has been well established that high GLU
concentrations at 50-70umol are toxic and lethal to neuronal cells. Therefore,
these experiments were expected to generate the most useful preliminary
information in the shortest amount of time and at minimal costs.

These studies differ markedly from previous studies reported in the
literature. Differences include transmitter concentration, neonatal astrocyte
isolation and uptake methodology. A proliferating population of astrocytes
was generated from 4 day old rat pups. Previous studies had used 1 day old
rat pups and many times mixed cultures. The astrocytes used in this study
were at their 3rd passage, 75 days or longer in culture. They were
characterized as astrocytes by their morphology and GFAP
immunofluorescence. Over 95% of these cells were observed to be positive
for the astrocyte-specific intermediate filament GFAP. In addition,
approximately 95% of these cells exhibited epithelioid morphology
characteristic of protoplasmic astrocytes. These astrocytes were also positive
for GABA uptake, an additional defining characteristic for astrocytes. Other
studies on GLU uptake by neonatal astrocytes have reported uptake at 10 min

to range from 10-100nmol/ mg protein. This study in comparison reports 8-86

 

1 80
nmol/ mg protein for the GLU study and 15-483 nmol/ mg protein for GABA.

However, attempting to compare this study to previous studies for
quantifying of GLU and GABA uptake is not consistent with the intent and
purpose of this study. One would actually be trying to study and compare
apples to oranges. Previous studies on uptake have taken cultured cells and
perturbed them by initially culturing them in unrestricted medium and then
rinsing them in a restricted buffer solution once, twice or three times and
then measuring uptake in this buffer. This dissertation reports on an uptake
procedure for unperturbed cells cultured in a restricted chemically defined
medium; a system of analysis that, I personally feel, is more sensitive,
efficient and accurate in determining uptake. These data, therefore, must be
compared to other uptake studies with a great deal of caution. The best
comparisons to be made are between neonatal astrocytes and anaplastic
astrocytoma cells within this study.

Characteristics of GLU uptake in astrocytes are unique. Uptake is
reported to be greater in astrocytes than neurons. It is not electrogenic and can
be concentrated above extracellular concentrations. Astrocytes for the most
part, are considered to be a sink for GLU. Characteristics for GABA uptake by
astrocytes are also somewhat unique. Astrocytes express an active high
affinity uptake system. The sodium gradient is the driving force, making
GABA uptake electrogenic. GABA uptake by astrocytes has its limits. It has
been reported that GABA can not be concentrated above extracellular
concentrations. However, GABA uptake in unperturbed neonatal astrocytes
should be directed to the small glutamate pool for immediate conversion to
glutamine. Personal consultation with other neurochemists has indicated
that with a 10 min incubation metabolic influences should be negligible. The

area of amino acid neurotransmitter uptake and metabolism definitely needs

 

 

 

 

181

further research. This area has tremendous potential for research which
could result in a much better overall understanding of neuropsychiatric
symptoms that accompany many neurologic diseases.

The studies presented here were planned and intended to compare
neonatal astrocytes with astrocytoma cells and to determine the effects of NGF
on the uptake of GLU and GABA. This information, it was hoped, could
suggest whether or not NGF could influence amino acid neurotransmitter
uptake and hopefully data could be shown to support its potential use as a
reverse transforming agent for the treatment of astrocytomas. Fortunately or
unfortunately, the data leave more questions than answers. For example, the
effects of temperature on GLU uptake is completely opposite of any other
reports in the literature. These data are clear and crisp and significant, yet,
preliminary and premature to make profound statements. Nevertheless, the
results clearly suggest the discovery of another GLU receptor and mechanism;
one that would be the first to be unique to astrocytes. This is an exciting
discovery. However, there are so many new variables that were established
with this new protocol, that it simply necessitates further studies and

documentation in order to establish its reality.

Conclusions

This dissertation presents a new method for the isolation and long-
term culture of neonatal astrocytes. It also presents a new method of cell cycle
analysis of neonatal astrocytes and anaplastic astrocytoma cells using acridine
orange for characterizing cells by 55 and dsDNA. In addition it presents a new
method to determine amino acid neurotransmitter uptake by culturing cells
in a restricted chemically defined medium and measuring the uptake by

unperturbed cells.

 

 

182

From the experimental data it may be concluded that NGF has an effect
on proliferation and morphology of anaplastic astrocytoma cells. NGF was
found to not affect the proliferation potential of neonatal astrocytes.
However, NGF was found to facilitate the generation of a quiescent pool of
non-cycling astrocytoma cells. This is the first time that data supporting
changes in the cell cycle,from a proliferating population to a quiescent non-
cycling population, have been illustrated and reported in tumor cell
populations. This is a very significant finding.

NGF was found to facilitate the induction of morphologic
differentiation in neonatal astrocytes and anaplastic astrocytoma cells by GLU.
This facilitation by NGF was at a low concentration that did not induce
differentiation by itself. In addition, GLU at the concentration used, following
NGF treatment, did not induce morphologic differentiation without NGF.
This suggests that NGF may facilitate the action of other differentiation
promoter agents, which act by triggering membrane-mediated events that
lead to membrane transduction, which in turn is followed by morphologic
changes characteristic of differentiation.

NGF was also found to induce GFAP expression in neonatal astrocytes
and anaplastic astrocytoma cells. The GFAP epitopes induced by NGF in
anaplastic astrocytoma cells were found to be fixation-sensitive.
Paraformaldehyde fixation in cacodylate buffer preserved GFAP epitopes,
while methanol fixation abolished GFAP epitopes in anaplastic astrocytoma
cells.

In addition, NGF treatment did not have a deleterious effect on the
uptake of the amino acid neurotransmitters GLU and GABA. N GF did not
significantly affect GLU or GABA uptake by anaplastic astrocytoma cells as

compared to neonatal astrocytes. However, NGF was found to significantly

 

 

 

1 83
decrease the uptake of GABA by neonatal astrocytes at 2.5 and 25 nmol GABA

concentrations. This may or may not be related to NGF's ability to stimulate
protein synthesis or its facilitation of morphologic differentiation, which is
speculated to be the result of cyclic AMP, phosphorylation and membrane
transduction events.

The data accumulated in the studies on GLU uptake suggest that a 6th
GLU receptor exists, which may be astrocyte specific. This receptor
mechanism has been found to be actively suppressed under normal
physiologic conditions and increased in response to temperatures <4°C. .

These studies have laid the ground work for future studies in normal
and tumor cell biology. In addition, the new procedures presented in this
dissertation should enhance our ability to study the neuropathologic and

toxicologic effects of specific agents on astrocytes.

 

184

References

Balazs, R. and Richter, D. (1972) Metabolic compartments in the brain: Their
properties and relation to morphologic structures. In, Metabolic
Compartmentation in the Brain (Balazs, R. and Cremer, J., eds.), pp.
167-184, MacMillan Press, London.

Balazs, R., Machiyama, Y. and Patel, A. (1972) Compartmentation and the
metabolism of g-aminobutyrate. In, Metabolic Compartmentation in
the Brain (Balazs, R. and Cremer, J., eds.), pp. 57-70, MacMillan Press,
London.

Barbacid, M. (1986) Oncogenes and human cancer: Cause or consequence?
Carcinogenesis 7, 1037-1042.

Berl, S., Lajtha, A. and Waelsch, H. (1961) Amino acid and protein
metabolism. VI. Cerebral compartments of glutamic acid metabolism.
J. Neurochem. 7, 186-197.

Boonstra, J., Moolenaar, W., Harrison, P., Moed, P., van der Sagg, P., and
de Latt, S. (1983) Ionic responses and growth stimulation induced by
nerve growth factor and epidermal growth factor in rat
pheochromocytoma (PC12) cells. J. Cell Biol. 97, 92-98.

Bures, J., Buresova, 0. and Krivanek, J. (1974) The Mechanism and
Application of Leao's Spreading Depression of Electroencephalographic
Activity. Academic Press, New York.

Campisi, J., Gray, H., Pardee, A., Dean, M. and Sonenshein, G. (1984) Cell
cycle control of c-myc but not c-ras expression is lost following chemical
transformation. Cell 36, 241-247.

 

 

185

Cheng, S.-C. and Brunner, E. (1974) Further support for the subdivision of
the small metabolic compartment: Effects of Halothane. In, Metabolic
Compartmentation and Neurotransmission (Berl, 5., Clarke, D. and
Schneider, D., eds.), pp. 479-485, Plenum Press, New York.

Cherkin, A. and Van Harreveld, A. (1978) L-Proline and related compounds:
Correlation of structure, amnesic potency and anti-spreading
depression potency. Brain Res. 156, 265-273.

Clarke, D., Ronan, E., Dicker, E. and Tirri, L. (1974) Ethanol and its relation to
amino acid metabolism in brain. In, Metabolic Compartmentation and
Neurotransmission (Berl, 5., Clarke, D. and Schneider, D., eds.), pp. 449-

' 460, Plenum Press, New York.

Cremer, J., Heath, D., Patel, A., Balazs, R. and Cavanagh, J. (1974) An
experimental model of CNS changes associated with chronic liver
disease: Portocaval anastomosis in the rat. In, Metabolic

Compartmentation and Neurotransmission (Berl, 8., Clarke, D. and
Schneider, D., eds.), pp. 461-478, Plenum Press, New York.

Cremins, J., Wagner, J. and Halegoua, S. (1986) Nerve growth factor action is
mediated by cyclic AMP and ca/phospholoipid dependent protein
kinases. J. Cell Biol. 103, 887-893.

Dichter, M., Tischler, A. and Greene, L. (1977) Nerve growth factor-induced
increase in electrical excitability and acetylcholine sensitivity of a rat
pheochromocytoma cell line. Nature 268, 501-504.

Eisenman, R., Tachibana, C., Abrams, H. and Hann, S. (1985) v-myc- and
c-myc-encoded proteins are associated with the nuclear matrix. Mol.
Cell. Biol. 5, 114-126.

Garfinkel, D. (1972) Possible correlations between morphological structures
in the brain and the compartmentations indicated by simulation. In,
Metabolic Compartmentation in the Brain (Balazas, R. and Cremer, J.,
eds.), pp. 129-136, MacMillan Press, London.

 

186

Gerchman, L. and Ludlum, D. (1973) The properties of O6-methylguanine in
templates for RNA polymerase. Biochim. Biophys. Acta 308, 310-316.

Greene, L. (1977) A quantitative bioassay for nerve growth factor (NGF)
activity employing a clonal pheochromocytoma cell line. Brain Res.
133, 350-353.

Greene, L. A. and Tischler, A. S. (1976) Establishment of a
noradrenergic clonal line of rat pheochromocytoma cells which
respond to nerve growth factor. Proc. Natl. Acad. Sci. USA 73, 2424-
2428.

Hertz, L. (1979) Functional interactions between neurons and astrocytes. I.
Turnover and metabolism of putative amino acid transmitters.
Prog. Neurobiol. 13, 277-323.

Hughes, R. and Sharon, N. (1978) Carbohydrates in recognition. Trends
Biochem. Sci. 3, 275-278.

Kelly, K., Cochran, B., Stiles, C. and Leder, P. (1983) Cell-specific regulation of
the c-myc gene by lymphocyte mitogens and platlet derived growth
factor. Cell 35, 603-610.

Kleihues, P. and Magee, P. (1973) Alkylation of rat brain nucleic acids by N -
methyl-N-nitrosourea and methyl methanesulpfonate. J. Neurochem.
20, 595-606.

Kleihues, P., Patzschke, K. and Doerjer, G. (1982) DNA modification and
repair in the experimental induction of nervous system tumors by
chemical carcinogens. Annls. N. Y. Acad. Sci. 381, 290-303.

Klymkowsky, M. W., Miller, R. H. and Lane, E. B. (1983) Morphology,
behavior and interaction of cultured epithelial cells after the antibody-
induced disruption of keratin filament organization. J Cell Biol 96,
494-509

 

187

Korsching, S., Auburger, G., Heumann, R., Scott, J. and Thoenen, H. (1985)
Levels of nerve growth factor and its mRNA in the central nervous
system of the rat correlate with cholinergic innervation. EMBO J. 4,
1389-1393.

Kruijer, W., Schubert, D. and Verma, I. (1985) Induction of the proto-
oncogene fos by nerve growth factor. Proc. Natl. Acad. Sci. USA 82,
7330-7334.

Lajtha, A., Berl, S. and Waelsch, H. (1959) Amino acid and protein
metabolism of the brain. IV. The metabolism of glutamic acid. J.
Neurochem. 3, 322-332.

Leutz, A. and Schachner, M. (1981) Epidermal growth factor stimulates
DNA-synthesis of astrocytes in primary cerebellar cultures. Cell Tissue
Res. 220, 393-404.

Marushige, Y., Marushige, K., Okazaki, D. and Koestner, A. (1989)
Cytoskeletal reorganization induced by nerve growth factor and glial
maturation factor in anaplastic glioma cells. Anticancer Res. 9, 1143-
1148.

Miura, M., Tamura, T.-a. and Mikoshiba, K. (1990) Cell-specific expression of
the mouse glial fibrillary acidic protein gene: Identification of the cis-

and trans-acting promoter elements for astrocyte-specific expression.
J. Neurochem. 55, 1180-1188.

Mohler, H., Patel, A. and Balazs, R. (1974) Effects of 1-hydroxo-3-
aminopyrolidone-2 and other CNS depressants on metabolic
compartmentation in the brain. In, Metabolic Compartmentation and
Neurotransmission (Berl, 8., Clarke, D. and Schneider, D., eds.), pp. 385-
395, Plenum Press, New York.

Monaghan, D., Holets, V., Toy, D. and Cotman, C. (1983) Anatomical
distribution of four pharmacologically distinct L-[3H]-glutamate
binding sites. Nature 306, 176-179.

 

188

Rohrschneider, L. and Gentry, L. (1984) Subcellular locations of retroviral
transforming proteins define multiple mechanisms of transformation.
In, Advances in Viral Oncology (Klein, G., ed.), pp. 296-306, vol. 4,
Raven Press, New York.

Schubert, D. and Whitlock, C. (1977) Alteration of cellular adhesion by nerve
growth factor. Proc. Natl. Acad. Sci. USA 74, 4055-4058.

Schubert, D., LaCorbiere, M., Klier, F. and Steinbach, J. (1980) The
modulation of neurotransmitter synthesis by steroid hormones and
insulin. Brain Res. 190, 67-69.

Singer, B. (1975) The chemical effects of nucleic acid alkylation and their
relationship to mutagenesis and carcinogenesis. Prog. Nucleic Acids
Res. Mol. Biol. 15, 219-284.

Traub, P. (1985) Are intermediate filament proteins involved in gene
expression? In: Wang, E., Fischman, D., Liem, R. K. H. and Sun, T.-T.
(eds) Intermediate Filaments, Ann. N. Y. Acad. Sci. 455, 68-78

Traub, P. and Vorgias, C. E. (1983) Involvement of the N-terminal peptide of
vimentin in the formation of intermediate filaments. J Cell Sci 63, 43-
67

Traub, P. and Vorgias, C. E. (1984) Differential effect of arginine modification
with 1,2-cyclohexanedione on the capacity of vimentin and desmin to
assemble into intermediate filaments and to bind to nucleic acids.

J Cell Sci 65, 1-20

Traub, P., Nelson, W., Kuhn, S. and Vorgias, C. (1983) The interaction in
vitro of the intermediate filament protein vimentin with naturally
occurring RNAs and DNAs. J. Biol. Chem. 258, 1456-1466.

 

 

1 89
Van Den Berg, C. (1972) A model of compartmentation in mouse brain based
on glucose and acetate metabolism. In, Metabolic Compartmentation
in the Brain (Balazs, R. and Cremer, J., eds.), pp. 137-166, MacMillan
Press, London.

Van Dilla, M., Steinmetz, L., David, D., Calvert, R. and Gray, J. (1974) High-
speed cell analysis and sorting with flow systems: Biological
applications and new approaches. Nucl. Sci. 21, 714-720.

Van Dilla, M., Trujillo, T., Mullaney, P. and Coulter, J. (1969) Cell
microfluorometry: A method for rapid fluorescence measurement.
Science 163, 1213-1214.

Van Harreveld, A. and Fifkova, E. (1974) Involvement of glutamate in
memory formation. Brain Res. 81, 455-467.

Van Den Berg, C., Matheson, D. and Ronda, G. (1974) A model of glutamate
metabolism in the brain: A biochemical analysis of a heterogenous
structure. In, Metabolic Compartmentation and Neurotransmission
(Berl, 5., Clarke, D. and Schneider, D., eds.), pp. 515-540, Plenum Press,
New York.

Vorgias, C. E. and Traub, P. (1983) Isolation of glial fibrillary acidic protein
from bovine white matter and its purification by affinity
chromatography on single-stranded DNA-cellulose. Biochem Biophys
Res Comm 115, 68-75.

Waelsch, H., Berl, S., Rossi, C., Clarke, D. and Purpura, D. (1964) Quanitative
aspects of CO2 fixation in mammalian brain in vivo. J. Neurochem.
28, 457-459.

Weinstein, I. (1987) Growth factors, oncogenes and multistage
carcinogenesis. J. Cell. Biochem. 33, 213-224.

 

190

Weinstein, I. (1988) The origins of human cancer: Molecular mechanisms of
carcinogenesis and their implications for cancer prevention and
treatment. Cancer Res. 48, 4135-4143.