~ ".54; _ 2

 

. ... To ... 3.3.7... ....v....... u... .o...o.. .. o.. u . . .. . . .

 

       

    

 

 

awn ‘ x n)", , , “

 

 

.

 
 

 

 

 

 

 

 

 

  

., ,...V . ...I. u.
‘ 3 n, .0 .v‘. .— I...
. , .2, f ..
. Ti
. .3 r i
. <
‘. .
c. .‘. ‘
.;..
.....
. . .‘
.,;. ,-
7... r
.
.7. ,

 

 

 

 

 

 

 

 

 

 

 

...,..........V..

 

 

  
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 
 

. . . . .. .Z......:..;.; 7...

 
 

 
 
 
 

 

 
 
 
 
 

 
 
 
 
 
 

 

LBI BISRARE

IlllllllllllllllllllllllllllllllllIllllllllllll |

312 293008825

ill

 

 

 

 

 

 

 

 

This is to certify that the

dissertation entitled

The Use of Nerve Growth Factor as a Reverse Transforming
Agent for the Treatment of Tumors of Neurogenic Origin

presented by

Michael J. Yaeger

has been accepted towards fulfillment
of the requirements for

Ph . D . degree in Pathology

mm

Major professor

DateM 30‘ ((37!
0 d

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

 

 

1.68:“ S’V
Michlgen Stags)
L Universaty

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

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

i

 

 

 

 

::l
————l[_—l

MSU Is An Affirmative Action/Equal Opportunity Institution
cMMunét—nt

 

 

 

 

 

 

 

 

'PHE USE OF NERVE GROWTH FACTOR AS A REVERSE TRANSFORMING
‘AGENT FOR THE TREATMENT OF TUMORS OF NEUROGENIC ORIGIN

BY

Michael J. Yaeger

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

DOCTOR OF PHILOSOPHY

Department of Pathology

1991

ABSTRACT

lmHE USE OF NERVE GROWTH FACTOR AS A REVERSE TRANSFORMING
.AGENT FOR THE TREATMENT OF TUMORS OF NEUROGENIC ORIGIN

BY

Michael J. Yaeger

The standard approach to cancer therapy generally
involves selective removal or destruction of neoplastic cells
by surgery, radiation or chemotherapy, or a combination
thereof. However, because primary CNS tumors are often not
amenable to surgical removal and generally tend to infiltrate
the surrounding neuropile, novel therapeutic approaches may
be necessary to augment traditional treatment strategies. One
such option relies on the management of cancer through the
control of differentiation and growth of neoplastic cells.
The concept of reverse transformation proposes that certain
agents may be able to activate pathways involved in
differentiation, and in the process, suppress pathways
essential for maintenance of the transformed phenotype.

Nerve growth factor (NGF), a compound important in the
development, maintenance and regeneration of the mammalian
sensory and sympathetic nervous systems, represents one such
potential reverse transforming agent. Because NGF is able to
stimulate the differentiation of a variety of normal and
neoplastic cell lines of neurogenic origin, it may prove
useful as a reverse transforming agent for the treatment of

tumors of neuroectodermal origin.

This thesis chronicles research undertaken to investigate
the reverse transforming potential of NGF. Two principal
areas were studied. In the initial phase, a variety of tumor
cell lines were evaluated for response to NSF mm. This
was followed by an evaluation of the response of subcutaneous
tumors implants to NGF in experimental animals. m
studies demonstrated the ability of NSF to induce neoplastic
cells of neurogenic origin to develop a more differentiated
phenotype, to diminish or arrest the growth rate of neoplastic
cells in culture and to induce changes that persist in the
absence of the compound. Subsequent in viyg tests
demonstrated the ability of NSF to decrease the growth rate
of tumor implants by 65% or more.

The ability of NGF to induce differentiation m and
diminish tumor growth rate Mm indicates that nerve growth
factor may have significant potential as a reverse

transforming agent for the treatment of neurogenic tumors.

Dedicated to my parents, William J. Yaeger and Mary Ann Brink,
and five wonderful sisters, Sharon, Susan, Teresa, Coleen and
Anne, thank you for a lifetime of support. I would also like
to dedicate this thesis to my wife, Susan, with love and

appreciation.

iv

ACKNOWLEDGEMENTS

I would like to extend a special thank you to Dr.
Adalbert Koestner, my major professor, for his guidance,
support and friendship. I am indebted to Dr. Koestner for his
assistance and insights. His advice was invaluable during my
program and shall continue to aid me in future endeavors.

I would also like to acknowledge the remaining members
of my committee; Dr. Kathyrn Lovell, Dr. Keiji Marushige, Dr.
Jon Patterson and Dr. Clifford Welsch. It would be difficult
to find a nicer, more knowledgeable group with which to work
and I truly appreciate their guidance and support.

I would like to thank Dorothy Okazaki, Dr. Yasuko
Marushige, Cori Oakley and Ralph Common. Their technical
support and assistance was invaluable.

I am grateful to all the members of the department of
Pathology and AHDL. I wish to acknowledge their part in
making my stay at Michigan State University a rewarding and
enjoyable experience.

Lastly, I would like to thank my wife, Susan, for her
understanding and ready assistance during my program. She was
at once a wife, unpaid laboratory technician and assistant,
and my major means of emotional support.

Thank you all. I shall always have fond memories of

Michigan State.

TABLE OF CONTENTS
Page
‘LIST OF TABLES .................. ........ ........ ....... viii
LIST OF FIGURES .... .......... . .......................... .ix
KEY TO ABBREVIATIONS .....................................xi
INTRODUCTION ........ ......... .............................1
LITERATURE REVIEW .................................. ....... 3

The discovery of nerve growth factor ........ ..... ....3
The nature of NGF ....................................4
Properties of NGF ....................................6
Neurotrophism ...................................6
Neurotropic effects .............................8
Maturation and differentiation ..................9
NGF-Induced down—regulation
of EGF receptors ...............................10
NGF in the CNS ......................................11
Nerve growth factor receptors .......................13
Reverse transformation ..............................24
In yitro studies ....................................30
In yivg studies .....................................38
Models of tumor induction ......................38
Tumor implantation studies .....................42
Importance of the NGF-NGFR signal transduction
pathway ........................................43
Summary .............................................45

CHAPTER 1: The Reverse Transforming Effects of Nerve
Growth Factor on Five Human Neurogenic Tumor Cell
Lines: 111v1tzoresu1t8 OOOOOOOOOOOOOOOOO0.00.00.000.47

Abstract .......................................47
Introduction ...................................48
Methods ........................................51
Results ........................................54
Discussion .....................................64

vi

Chapter 1 Appendix ....... . . . . . . . . . ........ . . ...... . . 69

Introduction ..................... ............ ..69
NGF-Induced morphologic changes ... ............. 69
Image analysis ................. ..... ...........88
The effect of NGF treatment on NGFR

density in the U251 cell line ......... ......... 94

The length of NGF-exposure required to
induce a maximal morphologic response
in the H8294 cell line ................ ...... ...98
Conclusions ...................................104

CHAPTER 2: The Use of Nerve Growth Factor as a
Reverse Transforming Agent for the Treatment
of Neurogenic Tumors: in viyg results ..............105

Abstract ................................ ...... 105
Introduction ..................................106
Methods ...... ..... . ...... .....................109
Results .......................................113
Discussion ....................................122

Chapter 2 Appendix .................................126

Introduction ..................................126
Experiment A: A preliminary study of the

behavior of intracerebral and subcutaneous

T9 anaplastic glioma implants .................128
Experiment B: The effect of NGF treatment on

the growth of subcutaneous and intracerebral

T9 anaplastic glioma implants .................132
Experiment D: An evaluation of the effects

of various NGF treatment regimes on the

growth rate of subcutaneous T9 implants .......137
Experiments E: Stereotactically guided
intracerebral implantation of T9 anaplastic
glioma cells followed by NGF treatment

directly at the implantation site .............143
Intracerebral implantation of clone 16 ........152
Experiments involving the implantation

of two human tumor cell lines into

athymic nude mice .............................155
Conclusions ...................................158

CONCLUSIONS AND FUTURE STUDIES ........... ........... ....159

LIST OF REFERENCES ....................... . .............. 164

vii

LIST OF TABLES

Table Rae

1.1 Relative NGFR densities of the five human
neurogenic tumor cell lines .......... 55

1.2 Time and dose rate parameters for the response of
the human tumor cell lines to NGF 1n_yi§zg ...........57

1.3 The effect of NGF treatment on the number of
processes per cell and the percentage of cells
exhibiting well.developed processes ..................92

1.4 The effect of NGF treatment on the expression of
NGFR on the U251 glioblastoma multiforme cell line ...96

1.5 Length of NGF-exposure required to induce a
maximal morphologic response in the H8294
malignant melanoma cell line . . . . . . . . . . . . . . . . . . . ..... 102

2.1 Comparison of PBS, alpha-z-macroglobulin, mouse
and rat serum in terms of their ability to inhibit
or diminish NGF-induced morphologic changes in the
D54 mixed.glioma cell line ..........................121

2.2 The percentage of rats free from palpable
subcutaneous nodules following NGF treatment ........139

2.3 The effect of NGF treatment on stereotactically
guided intracerebra1.T9 implants ........ ............ 150

viii

LIST OF FIGURES

Figure Page

1.1

Phase-contrast photomicrographs of the NGF-induced
morphologic changes in the D54 (mixed glioma) and
H3294 (malignant melanoma) cell lines ...............58

Phase-contrast photomicrographs of the NGF-induced
morphologic changes in the U251 glioblastoma
multiforme cell line COCO.........OOOOOOOOOOOOOO0.0.060

The effect of NGF on the cell counts of four
human neurogenic tumor cell lines on day 4
post-NGF treatment 0.0............OOOOOOOOOOOOOO ..... 62

The effect of NGF-Treatment on the growth rate
of the U251 glioblastoma multiforme cell line
mm

.........OOOOCOOO0............OOOOOOOOOOOOO.63

Phase-contrast photomicrographs of the spectrum
of morphologic changes induced by NGF in the
U118 glioblastoma multiforme cell line ......... ..... 71

Phase-contrast photomicrographs of the spectrum
of morphologic changes induced by NGF in the D54
mixed glioma cell line ..............................74

Phase-contrast photomicrographs of the spectrum
of morphologic changes induced by NGF in the
TE671 medulloblastoma cell line .....................77

Phase-contrast photomicrographs of the spectrum
of morphologic changes induced by NGF in the
H3294 malignant melanoma cell line ..................81

Phase-contrast photomicrographs of the spectrum
of morphologic changes induced by NGF in the
U251 glioblastoma multiforme cell line ..............85

The effect of NGF treatment on the maximal

cross-sectional area of the TE671, U251 and
H8294 cell lines 0.........OOOOOOOOOOOOOOOO 0000000000 90

ix

Figure Page

2.1

The effect of NGF treatment on the growth rate
of subcutaneous T9 anaplastic glioma implants ......116

The effect of NGF treatment on the growth rate
of subcutaneous clone 16 anaplastic neurinoma
implants ......OOOOOOOOOOOO......OOOOOOOOOOOOOOOOO0.117

The effect of a repeated series of NGF treatments
on the growth rate of clone 16 implants ............119

The effect of NGF treatment on the growth of
subcutaneous T9 implants ................... ...... ..135

The effect of various NGF doses and dosage
schedules on the growth rate of subcutaneous
T9 implants O...OOOOOOOOOOOOOOOOOOO0.0.00.00... 00000 141

The effect of NGF treatment on the growth rate
of subcutaneous U251 glioblastoma multiforme
implants OOOOCOOOOOIOOOO0.00.0...0.00.00.00.00...000156

The effect of NGF treatment on the growth rate
of subcutaneous D54 mixed glioma implants ..........157

KEY TO ABBREVIATIONS

BID.......Twice-a-day

CNS.......Central nervous system

D54.......D54 mixed glioma cell line.
DMEM......Dulbecco's modified eagle's medium
EGF.......Epidermal growth factor
ENU.......Ethylnitrosourea

FBS.......Fetal bovine serum

Hs294.....HsZ94 malignant melanoma cell line
IP........Intraperitoneal

MRI.......Magnetic resonance imagery
NGF.......Nerve growth factor

NGFR......Nerve growth factor receptor
PC12......PC12 pheochromocytoma cell line
PNS.......Peripheral nervous system
SID.......Once-a-day

TE671.....TE671 medulloblastoma cell line
0251......U251 Glioblastoma multiforme cell line
0118......U118 Glioblastoma multiforme cell line

xi

INTRODUCTION

INTRODUCTION

The standard approach to cancer therapy generally
involves selective removal or destruction of neoplastic cells
by surgery, radiation or chemotherapy, or a combination
thereof. However, novel therapeutic approaches may be able
to augment these traditional treatment strategies. One such
option is the management of cancer through control of
differentiation and growth of neoplastic cells. Malignant
transformation, or an integral event in malignant
transformation, may be associated with an aberrant or
incomplete response to normal effectors regulating
differentiation and growth. If this were true, then specific
differentiating agents might be able to stimulate neoplastic
cells to differentiate and reverse many of the transformed
properties of the neoplastic cells. Such a process is the
antithesis of neoplastic transformation and is termed reverse
transformation. It is based on the premise that certain
agents may be able to activate processes which stimulate
neoplastic cells to differentiate to a point where they no
longer exhibit neoplastic behavior.

Nerve Growth Factor (NGF) represents one such potential
reverse transforming agent. The capability of NGF to
stimulate the differentiation of a variety of normal and

neoPlastic cell lines of neurogenic origin suggests that it

1

2

may’ prove useful as a reverse transforming agent. In
particular, studies demonstrating the ability of NGF to retard
cell proliferation and induce morphologic changes in
anaplastic glioma cells (Marushige et al. 1987: Kumar et al.
1990) , PC12 pheochromocytoma cells (Greene and Tischler, 1976)
and neuroblastoma cells (Reynolds and Perez-Polo, 1975; Jensen
1987: Matsushima and Bogenmann, 1990) indicate that it is
capable of reversing some of the transformed properties of
neoplastic cells of neurogenic origin. In response to NGF,
these tumor cell lines develop a more differentiated
phenotype, and, in the process, lose the capacity for
uncontrolled growth.

The literature review will outline the properties of NGF
and the rationale behind its evaluation as a reverse
transforming agent. Susequent chapters will describe studies
designed to evaluate the reverse transforming potential of

nerve growth factor.

LITERATURE REVIEW:

sc v e t

The discovery of NGF was, like many in science, a
fortuitous gift from an unexpected source: a mouse sarcoma.
NGF was discovered as a result of the striking morphologic
effects that it caused. When fragments of mouse sarcoma 37
or 180 were grafted to the body wall of 3-day chick embryos,
sympathetic (and sensory' nerve fibers ‘were. stimulated. to
innervate the neoplastic tissue where they built an
extraordinary network.of fibers (Bueker 1948: Levi-Montalcini
1987). This extensive outgrowth. was accompanied. by an
increase in the volume of corresponding ganglia of up to 6
times (Levi-Montalcini 1951). As the transplantation of the
sarcomas to the chorioallantoic membrane elicited similar
effects, it was hypothesized that the agent responsible for
the induced changes was a soluble, diffusible product that was
released from the sarcomas (Levi-Montalcini 1952: Levi-
Montalcini and Hamburger, 1953). An in_y1§;g assay confirmed
this suspicion as sensory and sympathetic ganglia explanted
from 8-day chick embryos and grown in a semisolid medium in
proximity to fragments of mouse sarcoma 37 or 180 exhibited
an eruption of nerve fibers of maximal density on the side
facing the tumor (Levi-Montalcini et a1. 1954). The majority

3

4

of this nerve outgrowth. activity ‘was identified. in the
microsomal fractions of these tumors (Cohen et al. 1954).

In another fortuitous discovery, snake venom was
identified as possessing similar nerve growth promoting
activity (Cohen and Levi-Montalcini, 1956: Levi-Montalcini and
Cohen, 1956: Cohen 1959). This led.directly to the evaluation
of the murine homolog of the venom producing glands, the mouse
submandibular salivary gland, for NGF activity (Cohen 1960).
When added in. minute quantities to the culture medium,
homogenates of the mouse salivary gland elicited an even
denser and more compact fibrillar halo (Cohen 1960; Levi-
Montalcini and Booker, 1960b: Aloe et al. 1981) . Though
considerable confusion centered around why such.diverse, non-
neural sources should possess nerve growth promoting activity,
it was nonetheless believed that this factor might represent
an important clue to understanding the development and
maintenance of the nervous system. The identification of this
very rich source of NGF launched an extensive and systematic
search into the compound's chemical structure, mechanisms and

spectrum of action.

W
Murine NGF exists as a high molecular weight complex with
a sedimentation coefficient of 7S and a molecular weight of
approximately 130,000 (Varon et al. 1954) . It consists of two

alpha subunits, two gamma subunits, and a single, 2.58 beta

5

subunit (Smith et al. 1968: Greene et al. 1971). All of the
neurotrophic activity resides in the beta subunit (Greene et
a1. 1971) which consists of a dimer of two identical chains
of 118 amino acids (Angeletti and Bradshaw, 1971) held
together by noncovalent bonds (Bocchini and Angeletti, 1969).
The other subunits do not influence the activity of the beta,
and the 78 must come apart and liberate the beta subunit
before any activity can be expressed (Stack and Shooter,
1979). The dissociation of the high molecular weight complex
into active and inactive subunits can be accomplished during
the extraction procedure through the use of low pH and
separation via ion-exchanges columns (Varon et al. 1954) .

The active NGF molecule appears to be a highly conserved
protein. Murine NGF elicits a strong response in avian, rat,
hamster, monkey and human tissues. A comparison of mouse
beta-NGF nucleotide and amino acid sequences with
corresponding human sequences reveals high levels of homology
(Ullrich et a1. 1983). The nucleotide sequences are 86%
homologous overall (Ullrich et al. 1983). Not only is the NGF
molecule highly conserved, but extensive sequence homology
also exists for the nerve growth factor receptor (NGFR). The
NGFR has been cloned in the rat, chicken and human, and the
three receptors, especially in the cysteine-rich binding site,
share extensive sequence homology (Johnson et al. 1986a;

Radeke et al. 1987: Large et a1. 1989).

6
W

The regulation of neuronal development can be considered
in terms of control of cell death, differentiation and
orientation. NGF appears to play an important role in each
of these areas in the peripheral nervous system. NGF plays
a vital trophic role especially during early developmental
stages. NGF enhances differentiation processes, such as
neurite outgrowth, and guides the direction of growing or
regenerating neurites along its concentration gradient
(Gunderson and Barrett, 1980: Campenot, 1982a: Collins and

Dawson, 1983).

MW
Some of the most astonishing examples of the importance
of NGF have been demonstrated through the use of anti-NGF-
antibodies administered to populations of prenatal, neonatal
and adult rodents. Anti-NGF-antibodies present during the
neonatal period resulted in near total disappearance of
sympathetic para- and prevertebral ganglia (Levi-Montalcini
and Booker, 1960a: Levi-Montalcini and Booker, 1960b: Levi-
Montalcini 1964: Otten et al. 1979: Johnson et al. 1980:
Kessler and Black, 1980a) a process which was termed
immunosympathectomy (Levi-Montalcini and Angeletti, 1966) .
When administered in utero, anti-NGF-antibodies not only
resulted in the degeneration of the sympathetic nervous system

but also in the degeneration of sensory neurons and chromaffin

'7
cell precursors in. the adrenal medulla (Aloe and Levi-
Montalcini, 1979; Johnson et al. 1980). Even in adult
animals, immunization against NGF results in a 30-40% cell
loss in the superior cervical ganglion (Gorin and Johnson,
1980).

Additional information as to the importance of NGF was
gained through the administration of supplemental NGF.
Initial experiments in mice indicated that injection of crude
NGF preparations into newborn mice resulted in a 4-6 fold
increase in the size of the superior cervical ganglia (Levi-
Montalcini and Booker, 1960b) . Additional experiments further
clarified the nature of this response. During normal
development, there is a considerable decrease in the number
of sympathetic neurons in the superior cervical ganglion.
Administration of NGF to 6 day old rats prevented this
decrease via a more rapid rate of neuronal maturation and the
prevention of cell death (Hendry and Campbell, 1976: Hendry
1976: Zaimis 1971). This, combined with the fact that NGF
levels in various target organs are correlated with the
density of innervation (Holzbauer and Sharman, 1972: Campenot
1982b), suggests that the density of target organ innervation
may be controlled by the level of NGF produced by the target
organ. The ability of NGF to support the survival of
neurons (neurotrophism) has also been demonstrated in studies
involving lesion induced neuronal degeneration. Exogenous NGF

administration prevented nerve cell death following

8
transection of NGF-dependent postganglionic axons of the
superior cervical ganglia (Levi-Montalcini et a1. 1975: Hendry
1975: Johnson et a1. 1980: Johnson et al. 1986b).
Additionally, chronic intracerebroventricular infusion of NGF
in rats or primates with fimbria-fornix lesions prevents the
degeneration of basal forebrain cholinergic neurons (Hefti
1986: Williams et al. 1986: Kromer 1987: Gage et al. 1988:

Tuszynski et a1. 1990).

W

NGF can direct growing or regenerating axons of sensory
and sympathetic fibers along its concentration gradient.
These neurotropic effects have been established both mm
and in yivg (Levi-Montalcini 1976: Menesini-Chen et al. 1978:
Gunderson and Barrett, 1979: Gunderson and Barrett, 1980:
Campenot 1982a) . When NGF was injected into the brain of
neonatal rats for 6-10 days, nerve fibers originating from
sympathetic ganglia were observed to enter the spinal cord and
ascend the dorsal columns to the site of injection in the
brain stem (Levi-Montalcini 1976) . In vitro, growth cones of
chick sensory ganglia were observed to rapidly turn and grow
towards the highest concentration of NGF when exposed to local
concentrations of the growth factor (Charlwood et al. 1972:
Letorneau 1978: Gunderson and Barrett, 1979: Gunderson and
Barrett, 1980) . Furthermore, NGF adsorbed in specific

patterns onto a substratum can guide axonal growth along the

9
patterns, thereby exerting a haptotactic effect (Gunderson
1985). It is through such neurotropic effects that NGF may
assist in the precisely oriented growth necessary for the
formation of specific synaptic connections within the nervous

system (Yanker and Shooter, 1982).

W

Maturation and differentiation are the two properties of
NGF central to its reverse transforming potential. Much of
the information in this area has been generated m
utilizing various neoplastic cell lines and will be discussed
in detail in later sections. However, 11111333 and Mg
studies involving non-neoplastic tissue offer compelling
evidence that NGF is capable of stimulating maturation and/or
differentiation. Administration of NGF can accelerate the
maturation of sensory and sympathetic neuroblasts to
essentially postmitotic neurons (Mobely et al. 1977: Kessler
et a1. 1979). Exogenous NGF is also capable of influencing
the course of development of certain cell populations.
Prenatal injection of NGF was shown to produce massive
transformation of chromaffin cell precursors to sympathetic
neurons (Aloe and Levi-Montalcini, 1979) . This was confirmed
W as NGF caused a phenotypic shift in chromaffin cells
(Unsicker et al. 1978: Aloe and Levi-Montalcini, 1979)
resulting in their neuronal differentiation accompanied by a

number of chemical, ultrastructural and morphologic changes

10
characteristic of neuronal rather than a neuroendocrine

phenotype (Greene and Shooter, 1980).

t.'- 12- -. 2-11- :-L 2 7. o u“ o-- 5c - -- o ;

There is considerable evidence that NGF is capable of
stimulating neurogenic cell lines to undergo differentiation
(Waris et al. 1973: Reynolds and Perez-Polo, 1975: Greene and
Tischler, 1976: Unsicker et a1. 1978: Aloe and Levi-
Hontalcini, 1979: Greene (and Shooter, 1980: ‘Vinores and
Koestner, 1981: Reynolds and Perez-Polo, 1981: Jensen 1987:
Marushige et al. 1987: Marushige et al. 1989b: Kumar et a1.
1990: Matsushima and Bogenmenn, 1990: Yaeger et al. 1991).
In at least one cell line, this process is accompanied by
changes that cause these cells to become less susceptible to
mitogenic signals. Treatment of PC12 cells with NGF reduces
the amount of epidermal growth factor (EGF) bound.to the cells
by 80% or more (Huff and Guroff, 1979: Huff et al. 1981).
This decrease in binding is due to a reduction in the number
of cell surface receptors caused by decreased receptor
biosynthesis (Lazarovici et. a1. 1987). Such. a process
fulfills the criteria for heterologous down-regulation and may
.represent. a coordinated. change in. the potential of the
«Sifterentiated PC12 cell to interact with the hormonal
environment (Lazarovici et al. 1987). This decrease in the
availability of mitogenic receptors on the surface of the

cells during differentiation may restrict the ability of the

11
differentiated neurons to proliferate (Lazarovici et al.
1987). Thus, as part of its differentiating response, NGF
appears to down-regulate systems that have the potential to
stimulate proliferative pathways. This may be a part of the

mechanism whereby the differentiated phenotype is maintained.

H§E_§n§_§h§_QH§

The role of NGF in the development and maintenance of
peripheral sensory and sympathetic neurons has been well
characterized. Studies indicate that NGF may also play a role
in the development and maintenance of cholinergic neurons in
the mammalian forebrain (Williams et al. 1986).

Initial evidence that NGF may play an active role in the
CNS was demonstrated by the identification of both NGF and
nerve growth factor receptors (NGFR) in the brain. The CNS
has been shown to contain NGF mRNA, NGF antigen and NGFR
(Greene and Shooter, 1980: Scott et al. 1981: Ayer LeLievre
et a1. 1983: Taniuchi and Johnson, 1985: Richardson et a1.
1986). Nerve cells, especially those located in the
hippocampus and cortical areas manufacture large quantities
of NGF mRNA and NGF protein (Korsching et al. 1985: Shelton
and Reichardt, 1986: Whittemore et a1. 1986) . Small and large
neuronal populations located in different brain areas have
been shown to exhibit all the properties and responses typical
of sensory and sympathetic cells such as: 1) the presence of

Specific receptors (Szutowicz et al. 1976), 2) retrograde

12

transport of NGF (Seiler and Schwab, 1984), 3) increased
neurotransmitter synthesis in response to NGF (Gnahn et al.
1983: Hefti et al. 1984: Mobley et al. 1985) and 4) a trophic
response characterized by prevention of cell death by
exogenous NGF (Williams et al. 1986: Kromer 1987). Reports
indicate that the majority of NGFR in the adult rat and monkey
brain is found in the cholinergic neurons and that NGF has a
neurotrophic effect on these cholinergic neurons (Schwab et
al. 1979: Seiler and Schwab, 1984: Taniuchi and Johnson, 1985:
Kordower et al. 1988: Batchelor et al. 1989). Chronic
intracerebroventricular infusion of NGF in rats or primates
with fimbria-fornix lesions prevents the degeneration of basal
forebrain cholinergic neurons (Hefti 1986: Williams et al.
1986: Kromer 1987; Tuszynski et al. 1990).

One of the motivating factors behind research into the
effects of NGF in the CNS centers around the well
characterized neurotrophic properties of this compound in both
the CNS and PNS. It is theorized that the neurotrophic
properties of NGF may prove useful in preventing or limiting
the severity of a number of neurologic disorders characterized
by degeneration of selected neuronal populations such as
Alzheimer's disease, Parkinson's disease, amyotrophic lateral

sclerosis, and others. (Appel 1981).

13
W

In the previous sections mention was made of the nerve
growth factor receptor (NGFR) . The receptor is a central
issue in any discussion of the effects of NGF because the
growth factor' must first bind. to specific cell surface
receptors in order to effect a response. Though this point
is incontrovertible, the exact nature of the receptor for NGF
and the NGF-NGFR interaction is unclear. The nature of the
receptor and the character of this interaction has recently
been clarified to a greater degree (Buxser et a1. 1990:
Hempstead et al. 1991), however, the mechanisms by which this
complex triggers changes within the cell have not, as of yet,
been fully elucidated.

A wide variety of cell types possess NGFR. These include
sensory and sympathetic neurons (Verge et al. 1989),
cholinergic CNS neurons (Cohen-Cory et al. 1989), meningeal
cells (Rettig et al. 1987), glial cells (Rettig et al. 1987),
Schwann cells (Zimmerman and Sutter, 1983: Taniuchi et al.
1986: Rettig et al. 1987), adrenal chromaffin cells, and the
neoplastic counterparts of many of these cell types including
neuroblastomas (Rettig et al. 1987: Marchetti et al. 1987:
Chen et a1. 1990) , medulloblastoma, pheochromocytoma (Landreth
arm: Shooter, 1980), gliomas (Kumar et a1. 1990), neurinomas
and melanomas (Fabricant et al. 1977) . Cells of hematopoietic

lineage, mast cells and B-lymphocytes, also express NGF-R
(Thomson et a1. 1988) .

14

The nerve growth factor receptor is generally regarded
to be a 70-80 kDa phosphorylated glycoprotein (Hosang and
Shooter, 1985: Chao et al. 1986: Green and Greene, 1986: Ross
et al. 1987) . NGF binds to specific NGFR located on the nerve
endings of sympathetic and sensory fibers by a selective and
saturable mechanism, is internalized and transported
retrogradely to the corresponding perikarya where it evokes
its characteristic biochemical effects (Hendry et al. 1974:
Stockel et a1. 1974: Duman et al. 1979: Thoenen and Barde,
1980: Hamburger et al. 1981: Schwab et al. 1982: Hosang and
Shooter, 1985: Johnson et a1. 1987). NGF, when retrogradely
transported to the neuronal some will assure maintenance of
the differentiated state of the neuron (Hendry 1976: Thoenen
and Barde, 1980). The NGF-NGFR complex is internalized in
vesicles in a fashion similar to that observed for EGF and
insulin (Schlessinger et a1. 1978: Levi et al. 1980).
Following internalization and retrograde transport, it is the
association of NGF and its receptor with the nucleus that
closely parallels the effects of the hormone on neurite
outgrowth (Yanker and Shooter, 1982) . Interruption of
retrograde axonal transport by surgical or chemical procedures
has the same effect as the neutralization of endogenous NGF
by specific antibodies (Thoenen and Barde, 1980: Harper and _

Thoenen, 1981: Schwab et al. 1982).
As with most polypeptide hormones and growth factors, the

receptor plays an essential role in signal transduction. This

15

has been demonstrated by the fact that NGF injected directly
into the cell cytoplasm has no effect on responsive cells.
NGF injected into the cytoplasm of either naive PC12 cells or
cells pretreated with NGF, did not induce process outgrowth
compared with non-injected PC12 cells in the same culture
dishes, despite the obvious responsiveness of the non-inj ected
sister cultures to extracellularly applied NGF (Seeley et al.
1983). '

Reports indicate that the binding of NGF to its specific
surface receptor is inconsistent with a simple, single step
interaction of NGF with its receptor (Sutter et al. 1979:
Landreth and Shooter, 1980: Olender and Stach, 1980: Schechter
and Bothwell, 1981: Tait et a1. 1981: Bernd and Greene, 1984:
Godfrey and Shooter, 1986: Green et al. 1986: Woodruff and
Neet, 1986: Marchetti and Perez-Polo, 1987). This is
indicated by a deviation from linear of Scatchard plots and
the evaluation of association and dissociation kinetics.
These types of analyses, along with the evaluation of physical
properties, support the existence of at least two distinct
Classes of NGF-binding sites in neurons and PC12 cells (Sutter
et a1. 1979: Olender and Stach, 1980: Landreth and Shooter,
1980: Schechter and Bothwell, 1981: Tait et a1. 1981:

Sonnenfeld and Ishii, 198.2: Stach and Wagner, 1982: Block and
Bothwell, 1983: Vale and Shooter, 1983: Bernd and Greene,
1984: Godfrey and Shooter, 1986: Green et al. 1986: Green and

Greene, 1986: Woodruff and Neet, 1986: Marchetti and Perez-

16

Polo, 1987: Kumar'et al. 1990: Welcher et al. 1991): 1) a slow
dissociation, high—affinity, trypsin resistent form that has
a Kd of 0.2 nM in PC12 cells and a Kd of 10 pH in sensory and
sympathetic neurons and is resistent to extraction with low
concentrations of Triton X-100 and 2) a rapidly dissociating,
low-affinity, trypsin sensitive form that has a Kd of 5 nM in
PC12 cells and a Kd.of'1 nM in sensory and sympathetic neurons
and is extracted with low concentrations of Triton x-1oo. The
high affinity receptors appear to be responsible for NGF
internalization and for the differentiating response in PC12
cells (Green et al. 1986). The biologic response in neuronal
cells is also thought to correlate with occupancy of the high
affinity receptor.

Though reports describe two different.NGF-binding sites,
data indicates that a single protein is responsible for both
the high-affinity and low affinity forms of the NGFR (Buxser
et al. 1985: Green and Greene, 1986) with indirect evidence
suggesting that the high-affinity NGFR consists of the low-
affinity receptor complexed with an additional cytoplasmic or
membrane protein that is necessary for signal transduction
(Marchetti et al. 1987: Hosang and Shooter, 1987: Hempstead

et al. 1989) . Experiments in which receptors were labeled
With '“I-NGF, crosslinked and then immunoprecipitated with an
antireceptor antibody, demonstrate the existence of two
receptor proteins, an 80Kd species and a 200 Kd species.

There appears to be structural homology between the ZOO-and

l7
80-kDa receptor protein and it is suggested that the 200-kDa
NGF binding protein is an oligomer, possibly a dimer, of the
80-kDa NGF binding protein (Buxser et al. 1985). It has been
proposed that low affinity, rapidly dissociating binding of
NGF results when NGF associates with a single 80-kDa receptor
molecule and that the high affinity NGF binding results when
each monomer of NGF binds to a separate receptor molecule,
forming a dimeric receptor structure bridged.by dimeric beta-
NGF creating the 200-kDa complex (Buxser et al. 1985).
Removal of most of the intracellular portion of the low-
affinity NGF-R has no effect on NGF binding to the
extracellular domain of the receptor (Welcher et al. 1990).
Furthermore, within the extracellular domain, the four
cysteine-rich sequences when expressed alone are capable of
binding NGF (Welcher et al. 1990). It has also been shown
that the NGFR binds the homologous neurotrophic factor, brain-
derived neurotrophic factor (Rodriguez-Taber et a1. 1990).
At this point, it becomes necessary to examine some
puzzling concepts concerning the NGFR. It appears that a
single protein is responsible for both high and low affinity
binding and that these two forms exhibit interconversion (Vale
and Shooter, 1982: Block and Bothwell, 1983: Buxser et al.
1983a: Buxser et al. 1983b). Yet if this were true, it is
difficult to imagine how one form could be trypsin resistent
and the other trypsin sensitive, considering that both must

be present on the cell surface in order to interact with NGF.

18

There is also difficulty reconciling questions concerning the
function of the numerous low affinity receptors. It has been
suggested that because the low affinity sites are present in
numbers at least 10-fold larger than high-affinity sites
(Sutter et al. 1979: Claude et al. 1982), these low affinity
receptors may be significant in terms of the absolute numbers
of sites occupied. A spare receptor system (Stephenson 1956)
might be involved, in which case a maximum response is
attainable with occupation of only a fraction of the available
sites. Even more puzzling is that some cell populations
appear to respond in the apparent absence of high-affinity
receptors and that lysis of PC12 cells results in the rapid
loss of the high-affinity sites (Buxser et al. 1983b)
apparently due to their rapid conversion to low-affinity sites
after cell lysis (Buxser et al. 1983b).

Recent findings have helped to clarify some of these
puzzling aspects of the NGFR. Innovative experimental
approaches to the NGF-NGFR interaction have shed new light on
the understanding of this association. These results do not
contradict data obtained in previous studies, but rather offer
new insight into the interpretation of existing information.

A major complication in the interpretation of receptor

affinity data is the internalization of the NGF-NGFR complex
at physiological temperatures. Scatchard plots, which are the
most commonly used method to detect the two different binding

affinities, are affected by internalization (Buxser et al.

19
1990) . Therefore, it becomes essential to distinguish between
high-affinity receptors and internalization. Using new
approaches to try and separate these two events, Buxser et.
al. (1990) addressed some of the aforementioned problems. The
conclusions from their studies are rather insightful in light
of existing data.

It does not appear that the heterogeneity of NGF binding
arises from the presence of two different species of
receptors. Rather, the identification of receptors with
different binding affinities appears to be a consequence of
the pathway by which the NGF-receptor complex is processed in
the cell (Buxser et a1. 1990) . The thermodynamics of binding
of “I-NGF to PC12 cells are only consistent with the initial
formation of a single type of cell surface NGF-receptor
complex consistent with the previously described low-affinity
binding (Buxser et al. 1990) . A delay of approximately 60 sec
was observed after addition of ‘a’I-NGF before significant
high-affinity receptor binding was detected (Buxser et al.
1990). This second type of site can be explained by the
internalization of the ‘aI-NGF complexes (Buxser et al. 1990) .
The previously described high-affinity complexes are not
distinguishable from internalized NGF-receptor complexes and

the dissociation from slow receptors is not significantly
different from the return of internalized receptor to the cell
surface (Buxser et al. 1990) . These results provide a

relatively simple model to explain binding of NGF to its

20

receptor and subsequent internalization (Buxser et al. 1990).
The data are consistent with the presence of a single class
of binding sites on the surface of PC12 cells (Buxser et al.
1990). The biologically relevant sites have an affinity of
approximately 0.5 nM, can be internalized by the cells and
require only a low level of occupancy, albeit for a long
period of time, to initiate activation of the neurite
outgrowth process (Buxser et al. 1990). This behavior is
similar to that of the transferrin receptor (Chvatchko et al.
1984) and similar phenomena have been described for insulin
(Chvatchko et al. 1984). It would appear that high-affinity
binding has been confused with the internalization of the NGF-
NGFR complex (Buxser et al. 1990). These results are
consistent.with.much.of what.has been.described about the NGF-
NGFR interaction and provide a relatively straight forward,
unifying concept for the NGF-NGFR association.

This had been the most recent and insightful information
published on the subject. However, since the article by
Buxser et. al. (1990), an entirely new concept has been
unveiled“ IRecent studies (Hempstead et.al. 1991) suggest.that
high-affinity NGF binding requires coexpression and
COOperation between the low-affinity NGFR and the tyrosine
Icinase tzk gene product (Hempstead et al. 1991).

The :13 proto-oncogene encodes for a 140-Kd, membrane-

sPanning protein tyrosine kinase (p140-prototrk) . Although

tad: was discovered as a rearranged oncogene from a colonic

21
carcinoma (Martin-Zanca et al. 1986), normal :13 expression
.in_yiyg is limited to neural crest-derived sensory neurons
(Martin-Zanca et al. 1990a: Martin-Zanca et al. 1990b). The
p140-prototrk is intriguing because it is active in a select
subset of neural crest derived neurons and appears to be the
most exquisitely regulated mammalian tyrosine kinase that has
been described (Barinaga 1991).

Cross-linking studies indicated the existence of two
types of NGF receptors (Hosang and Shooter, 1985). Cross-
linking of 1a’Ir-NGF to its receptor with the photoactivatable
agentNehydroxylsuccinimidyl-4-azidobenzoate(HSAB),generated
crosslinked complexes of 100K and 158K (Hosang and Shooter,
1985). The 158K crosslinked complex was generated only in
cells that had high-affinity binding sites (Hempstead et al.
1991) and was considered to be the high-affinity receptor.
These receptors were distinct from the previously
characterized low affinity NGF receptors (p75-NGFR) (Chao et
al. 1986: Johnson et al. 1986a: Radeke et al. 1987), which
were represented.by the 100K band. The 158-kD species cross-
linked to NGF in PC12 cells and sympathetic neurons (Massague
1981: Hosang and Shooter, 1985: Buxer et al. 1985: Green and
Greene, 1986) was not recognized by antibodies to p75-NGFR
(Hempstead et al. 1990: Meakin and Shooter, 1991) . The p75-

BKEFR.did not appear to mediate a biologic response (Johnson
and Taniuchi, 1987: Hempstead et al. 1988) following ligand

binding and generated only low-affinity NGF-binding sites with

22
a Kd of 10‘ M (Chao et al. 1986). The generation of both
high- and low-affinity sites required an additional factor
besides the p75-NGFR.

The identification of this "additional factor" has long
eluded investigators. Two recent studies have lead
investigators to believe that the high affinity receptor was
a tyrosine kinase. Maher (1988) demonstrated that NGF induced
rapid phosphorylation of tyrosine residues when added to NGF-
responsive PC12 cells while Meakin and Shooter (1991)
demonstrated that the high affinity NGF receptors could be
immunoprecipitated with anti-phosphotyrosine antibodies.

This high-affinity receptor expressing tyrosine kinase
activity exhibited a molecular weight reminiscent of members
of the 31;); family of tyrosine kinases (Barbacid et al. 1991) .
HSAB crosslinking experiments firmly identified the 158K
species as the p140-prototrk crosslinked to NGF (Kaplan et al.
1991a) . Following cross-linking to mI-NGF, the 158-kD
species was immunoprecipitated with antibodies to NGF or p140-
prototrk (Kaplan et al. 1991b) . The immunoprecipitation of
the 158-kD species was blocked by addition of the peptide from
the p140-prototrk (Kaplan et al. 1991a) used to generate the
antibody and was not seen if excess unlabeled NGF was added
to the cells prior to cross-linking (Kaplan et al. 1991b) .

Additional evidence that p140-prototrk is a receptor for
NGF come from physiochemical studies. Picomolar

concentrations of NGF stimulate p140-prototrk

23
autophosphorylation (Kaplan et al. 1991a) . The stimulation
of p140-prototrk tyrosine phosphorylation in response to
addition of NGF to PC12 cells is rapid and specific, and
occurs in the presence of physiological amounts of NGF (Kaplan
et al. 1991a).

Membrane fusion studies clarified the contributions of
p140-prototrk and p75-NGFR in the formation of the high
affinity receptors. Both p140-prototrk and p75-NGFR exhibit
binding characteristics of low-affinity NGFR (Kaplan et al.
1991b) . PAElc cells express only p75-NGFR and rtrk-3T3 cells
express only the p140-prototrk. Scatchard analysis indicates
that both these cell lines exhibit receptors with a single
binding affinity consistent with low-affinity binding.
However, if the membranes from these two cell lines were
fused, they exhibited both high- and low-affinity binding
(Hempstead et al. 1991).

Thus it appears that coexpression of proto-oncogene p140-
prototrk and p75-NGFR are required for the formation of high-
affinity NGF-binding sites (Hempstead et al. 1991) .
Experiments (Hempstead et a1. 1991) confirm that two distinct
protein species can bind NGF, and support earlier biochemical
data describing the different proteolytic susceptibility, pH
sensitivity and Triton X-100 solubility of the two kinetic
forms of the receptor (Schechter and Bothwell, 1981: Vale and
Shooter, 1982: Block and Bothwell, 1983: Buxser et al. 1983a) .

Results (Hempstead et a1. 1991) provide strong support for the

24
hypothesis that the high affinity NGF binding site is composed
of at least two NGF-binding components: the tzk gene product
and p75-NGFR.

These findings do not conclusively demonstrate that $13
alone can trigger the specific biochemical pathways required
for neuron survival and differentiation. They do greatly
clarify the nature of the nerveigrowth factor receptor and.the
NGF-NGFR interaction. However, the specific biochemical

pathways regulated by NGF remain to be elucidated.

W

The previous sections provide background information on
the properties of NGF and the interaction with its receptor.
Upcoming sections will detail the rationale behind the
evaluation of NGF as a reverse transforming agent. However,
before evidence detailing the ability of NGF to induce reverse
transformation is presented, it would be beneficial to discuss
the conceptual basis for this type of therapy and the crucial
attributes of a reverse transforming agent.

Alterations or mutations in the genetic information are
considered to be crucial in the process of neoplastic
transformation. The therapeutic approach to eliminate these
transformed cells generally involves surgical removal of the
Offending cells or their selective destruction through the use
Of radiation or chemotherapy. Since reverse transforming

agents do not directly destroy neoplastic cells and can not

25

repair the genetic defect(s) present in these cells, how then
can such agents restrain the malignant behavior of neoplastic
cells? An insight into the answer to this question can be
obtained from knowledge of normal cellular physiology.
Terminally' differentiated cells and their’ proliferation-
competent precursors possess identical genetic make-up.
However, one of these populations continues to proliferate
while the other is restrained in a non-proliferating state.
These two populations differ, not in their genetic content,
but in the suppression or expression of genetic information
by epigenetic factors. Herein lies the key to the potential
of reverse transforming agents. Though reverse transforming
agents are not able to repair the genetic abnormalities in
transformed cells, they may be able to suppress pathways that
are essential for the maintenance of the transformed
phenotype. Just as epigenetic factors are able to maintain
a cell in a terminally differentiated state, epigenetic
mechanisms may also be able to persistently suppress the
eXpression of the transformed phenotype in neoplastic cells.

The ability of epigenetic factors to modulate the
expression of malignant characteristics has been demonstrated
.both in_yitzg and in_yiyg. When the 816 mouse melanoma cell
(line was implanted subcutaneously into susceptible animals,
tumors developed rapidly and consistently. However, when 816
mouse melanoma cells were implanted into embryonic mouse skin

at the time when pre-melanocytes migrate into the skin, a

It a
m. P'
bus 5
ng'l'

in"
“Vail

Stu
lmti

eta

26

significant decrease in tumor incidence was noted (Gerschenson
<et al. 1986). When conditioned medium from embryonic skin
cultures was incubated with these same cells in culture, the
growth of the melanoma cells ceased, the cells acquired a more
normal morphology, and these cells failed to proliferate when

placed in fresh growth medium (Gerschenson et al. 1986). In
a similar study, significantly fewer tumors developed from C-

1300-3 neuroblastoma cells that had been implanted into the

second somite of neurula stage embryos than developed from

untreated control cultures of the neuroblastoma line (Podesta

et al. 1984). Thus, it appears that factors normally present

in the microenvironment during development are able to

suppress the malignant characteristics of certain neoplastic

cell lines.

Factors in the microenvironment that may contribute to

the suppression of the malignant characteristics include: 1)

the properties of the extracellular matrix, 2) communication

with neighboring cells and 3) the presence of soluble hormones

or growth factors. Our research will concentrate on the

ability of nerve growth factor, a growth factor with

differentiating capabilities, to suppress neoplastic behavior.
A problem arises when striving to evaluate the response

*Of neoplastic cells to this type of therapy. The mechanisms
responsible for neoplastic transformation have not been
elucidated for the cell lines under study. As such, it

be<30231I|es difficult to demonstrate that the reverse transforming

27

agent has suppressed the expression of pathways responsible
for transformation. Nonetheless, a set of crucial attributes
essential for reverse transformation can be formulated which
provide guidelines by which potential reverse transforming
agents can be judged. These criteria are based on the goals
of cancer therapy and the general properties of transformed
cells. Once these criteria are defined, it becomes easier to
appreciate the rationale behind the use of NGF as a reverse
transforming agent and the research strategy utilized to
evaluate the therapeutic potential of this compound.

As stated in the introduction, reverse transformation
could be described as the antithesis of neoplastic
transformation and characterized as the process whereby
neoplastic cells are stimulated to differentiate to a point
where they no longer exhibit neoplastic behavior. There are
many different mechanisms by which a cell can undergo
neoplastic transformation. It is likely that there are also
a number of ways to partially or completely reverse this
process. Though the mechanisms may vary, criteria vital for
reverse transformation can be readily conceptualized.

One essential component of reverse transformation is
suppression or cessation of the growth of neoplastic cells.
A hallmark of neoplasia is growth that exceeds and is
uncoordinated with that of normal tissues. This is not the
result of neoplastic cells behaving as pathologic dynamos with
Greatly reduced doubling times and a large proportion of cells

I.”

28

in the growth fraction. In fact, the replication rate of
neoplastic cells is generally less than that of many normal
renewing tissues. Rather, tumors grow progressively because
the fine homeostasis between cell growth and cell loss is
disrupted. Therefore, as a characteristic of neoplastic cells
is growth that exceeds that of normal tissue, it is vital that
reverse transforming agents curb this unrestrained growth.

Since neoplastic cells are not directly destroyed during
the process of reverse transformation, another vital
characteristic of a reverse transforming agent is that the
induced effects must persist. If the impact of an agent were
transient and the neoplastic cells retained the ability for
uncontrolled growth once the effects of the compound abated,
then the patient would be no nearer a cure than before
therapy, and the therapy would be of little value. Therefore,
another essential feature of an agent that does not directly
eliminate neoplastic cells during the course of treatment is
persistence of the induced effects.

The third characteristic of a reverse transforming agent
is the development of a more differentiated phenotype.
Conceptually, this property does not appear to be as crucial
as the previous two. If therapy fulfills the first two
criteria, i.e., brings about a persistent arrest of tumor
growth, than a significant therapeutic benefit would be
gained, without the necessity for development of a more

differentiated phenotype. However, the development of a more

29
differentiated state is generally the means by which a
persistent suppression of tumor growth is attained. The vast
majority of transforming agents are mutagens. Conversely,
agents capable of inducing reverse transformation are
generally compounds that promote differentiation. Indeed,
differentiation therapy is essentially a synonym for reverse
transformation. The establishment of a more differentiated
state can result in reverse transformation either through
terminal differentiation, in which the development of a
differentiated phenotype is accompanied by cessation of
growth, or through differentiation to a point where the
neoplastic cells are once again brought under control of
normal homeostatic mechanisms.

In conclusion, the basis for this type of therapy relies
on the capacity of the reverse transforming agent to activate
epigenetic mechanisms that persistently suppress the
expression of the transformed phenotype . Three
characteristics are of primary concern in evaluating the
response of neoplastic cells to a reverse transforming agent:
1) the development of a more differentiated phenotype, 2)
growth suppression or arrest and 3) persistence of the induced
changes. An appreciation for these key principles will
facilitate the analysis of data from 1114113129 and 184119
exPeriments which provide the basis for evaluating NGF as a

reverse transforming agent.

3O
Weiss

Information presented in this section describes results
of m experiments that demonstrate the ability of NGF
to reverse several of the transformed properties of neoplastic
cell lines of neurogenic origin. These studies indicate that,
in a defined 1841121252 environment, NGF is able to induce
several of the changes described as essential for reverse

transformation.

Some of the most intriguing evidence for the reverse
transforming ability of NGF comes from studies involving the
Pc12 pheochromocytoma cell line. As has already been
mentioned, NGF is able to stimulate adrenal chromaffin cells
( the cell type from which pheochromocytomas are thought to

a1l=‘.:lse) to develop a number of chemical, morphologic and
u:L‘Izrastructural changes characteristic of a neuronal rather
than neuroendocrine phenotype (Unsicker et al. 1978: Aloe and
I'QVi-Montalcini, 1979: Greene and Shooter, 1980) . NGF is able
tQ stimulate similar changes in PC12 cells. PC12 adrenal
lug(iullary pheochromocytoma cells, when grown in the absence
Q :3 NGF share many properties of adrenal chromaffin cells
( gleene and Tischler, 1976: Greene and Rein, 1977) . Within
§‘\reral days after the cells are exposed to physiologic levels
Q t NGF, they begin to acquire a number of phenotypic
b:-~‘<>perties of sympathetic neurons. These include initiation
Q :6 neurite outgrowth, cessation of cell division and the

Q.
Q\relopment of electrical excitability (Greene and Tischler,

Q
H‘
N

v :
rte.

ex:

2M

31
1976; Greene and Rein, 1977: Dichter et al. 1977; Schubert et
al. 1977; Tischler and Greene, 1978; Rudy et al. 1982; Reed
and England, 1986). Within 2-3 weeks, PC12 cells in the
presence of NGF form an intricate network of electrically
excitable neurites that can then form functional synapses with
co-cultured muscle cells (Schubert and Whitlock, 1977) .

The effects of NGF on the growth of PC12 cells warrants
additional comment. The NGF-induced differentiation of PC12
cells involves the transition from a mitotic to a nonmitotic
state (Lazarovici et a1. 1987) . It has been reported that if
NGF was added to the culture medium immediately following

removal of serum, the cells underwent approximately one
doubling time before cessation of growth. (Greene and
Tischler, 1976; Mobley et al. 1977; Greene 1978; Unsicker et
a1 . 1978; Aloe and Levi-Montalcini, 1979; Greene and Shooter,
1980: Greene et a1. 1983) .
Additionally, reports (Huff et al. 1981; Lazarovici et
‘1 . 1987) have shown that exposure of PC12 cells to NGF caused
q‘Dwn-regulation of epidermal growth factor (EGF) receptors by
Q“‘Ibproximately 50% after 3-4 days (Rudkin et al. 1989) . This
I“§‘t'.erologous down-regulation would appear to be an efficient
hfians of desensitizing the cell to proliferative stimuli,
thereby amplifying the differentiative pathways. (Lazarovici
Qt a1. 1987) Experiments involving the PC12 cell line
Q‘Qtlnonstrate the ability of NGF to reverse some of the

thansformed properties of neoplastic cells of neurogenic

32
origin. The ability to transform a rapidly dividing
population of neoplastic neuroendocrine cells into a non-
dividing, functional population that resembles mature neurons
illustrates the reverse transforming potential of NGF.

The differentiating effects of NGF are by no means
restricted to PC12 cells. A variety of neurogenic cell lines
respond in a similar manner. NGF reverses several of the
transformed properties of a number of tumor cell lines of
91 ial origin. The F98 anaplastic glioma cell line develops
a more differentiated phenotype in response to NGF manifested

by restoration of surface inhibition, a decrease in nuclear-
c)r‘lmplasmic ratio and an increase in processes formation

( Vinores and Koestner, 1980: Vinores and Koestner, 1981) . NGF
treatment also diminishes the growth rate of F98 cells in
culture (Vinores and Koestner, 1980) . The C6 glioma cells
13% spond to NGF by becoming rounded and developing processes
that form an interconnecting lattice or network between cells
(Rmar et al. 1990) . Eventually, cell proliferation of the
Q6 glioma line becomes arrested (as measured by ’H-thymidine
llb‘lzake assays) in response to NGF (Kumar et a1. 1990) . NGF
theatment clearly retards cell growth of T9 anaplastic glioma
QQ Ills (Marushige et a1. 1987) . T9 cells treated with NGF also
QQ‘JeIOp a number of morphologic changes consistent with a more
Q1 Iferentiated phenotype. NGF-treated cells become flattened
‘hfi extend numerous cytoplasmic processes. (Marushige et al.

1
3 B7) The NGF—induced cytoskeletal changes in T9 cells

r._
r"
I.

Em

Mai

M
Ind

ieve
res;

1981

res;
[25:
ad
and
I011}
How;

,9“
“I

33
support the concept that NGF is capable of morphological
differentiation of anaplastic glioma cells endowed with NGF
receptors (Marushige et al. 1989a).

As NGF plays an important role in the development and

maintenance of the sensory and sympathetic nervous systems,

.it would seem reasonable that neoplastic cell lines derived

from these cell types, such as neuroblastomas, a tumor of
probable sympathetic origin, would be responsive to NGF.
Indeed, several neuroblastoma cell lines have been shown to
develop changes suggestive of neuronal differentiation in
response to NGF (Waris et al. 1973: Reynolds and Perez-Polo,
1981; Harushige et al. 1987).

A number of human and murine neuroblastoma cell lines
e3tpress NGFR, bind NGF, differentiate, and extend neurites in
reaponse to NGF (Perez-Polo et al. 1979; Otten et al. 1980;
I(eesler and Black, 1980b; Kessler and Black, 1981; Reynolds
and Perez-Polo, 1981: Burmeister and Lyser, 1982: Sonnenfeld

and Ishii, 1982). Cultured neuroblastoma cells tend to be
I‘and and do not produce processes (Waris et al. 1973).
I.I'b‘wever, when these cells were grown in a medium containing
“QB; they were stimulated to extend processes (Waris et a1.
13 ‘23) . SYSY neuroblastoma cells extended processes exceeding
19 Q um in length after 3 days of NGF exposure (Perez-Polo et
‘1 ~. 1979; Sonnenfeld and Ishii, 1982). By 5 weeks, SYSY
§u ltures grown in the presence of 7S NGF exhibit a high degree

Qt
morphologic differentiation (Jensen 1987). The NGF-

34

induced neurites in SH-SY5Y cells had an ultrastructure

similar to that of developing sympathetic ganglion cells and

ended in structures typical of neuronal growth cones

(Burmeister and Lyser, 1982).
The development of electrical excitability’ provides

additional evidence that NGF is able to stimulate cultured

neuroblastoma cells to develop properties indicative of

neuronal differentiation. SYSY and KA, human neuroblastoma

have the appearance of undifferentiated neuroblasts

1977).

clones,
and are electrically unexcitable (Kuramoto et al.

Treatment of these cells with NGF over a 5-day period results

in the cells assuming morphologic characteristics of

differentiated neurons (Perez-Polo et a1. 1982) and the

development of electrical excitability when monitored
ihtracellularly (Kuramoto et al. 1981) or pharmacologically

( I’erez-Polo et al. 1982) . These differentiating effects were
iaentical whether the NGF was human or murine in origin

( bareas-Polo et al. 1982) .

NGF is able to stimulate neuroblastoma cells to develop

more differentiated phenotype and electrical excitability.

ilfl

ese morphologic and functional changes are accompanied by

Virtual arrest in cell division of SYSY neuroblasts (Perez-

7W

Q:lo et al. 1979; Perez-Polo et al. 1982; Sonnenfeld and

tekl-aii, 1982).
NGF appears to be able to induce at least two of the

Q
hanges described as essential for reverse transformation.

w
u

hi

n:

35

Specifically, NGF is able to stimulate the development of a

more differentiated phenotype and cause suppression of cell

proliferation. These effects have been demonstrated in a wide

variety of neurogenic tumor cell lines. However, evidence
must be presented that demonstrates the ability of NGF to
fulfill the remaining essential characteristic of a reverse
transforming agent: the ability to induce effects that are
persistent. Since neoplastic cells are not directly destroyed
during the process of reverse transformation, it is vital that
the differentiating effects persist.

Mm experiments indicate that the morphologic
changes induced by NGF persist in the absence of the compound.
(Waris et al. 1973; Reynolds and Perez-Polo, 1981; Marushige
at al. 1987) This has been demonstrated with neuroblastoma

and anaplastic glioma cell lines (Waris et al. 1973; Reynolds

and Perez-Polo, 1981; Marushige et al. 1987) . T9 anaplastic
91 ioma cells treated with NGF for 2 days undergo further
mQ :rphologic changes during the succeeding 2 days in the
ab sence of NGF and exhibit a pattern of cell growth and cell
mQIphology similar to those grown in the continuous presence
Q ‘3 NGF (Marushige et a1. 1987) . Cells grown in NGF for 2 days
thlowed by 4 days without NGF, cells grown in NGF for 4 days
and then for 2 days without and cells grown in the continuous
E3:.~‘desence of NGF for 6 days all show essentially identical
“Q rphological characteristics and growth pattern (Marushige
§§ al. 1987) . Thus, it appears that the effects of NGF

per:
net:

I”-
‘3,-
who

ind'
m

ana

to

dif
ion
Per

{Re
CQn
0f

is,

36

persist in its absence. Experiments involving a
neuroblastoma cell line also demonstrate that the effects of
NGF treatment are quite persistent as the NGF-induced
morphologic changes continued to persist for as long as three
weeks after switching to fresh medium (Reynolds and Perez-

Polo, 1981) .
The previous paragraph emphasizes a distinctive asset of
NGF: the persistent nature of the response. The ability to
induce a more differentiated phenotype is not unique to this
growth factor. A variety of chemical agents, primarily
analogs of second messengers, also possess this capability
(Perez-Polo et a1. 1982: Harushige et al. 1989b). Second
messenger analogs are able to mimic not only the morphologic
clianges induced by NGF but are also capable of diminishing the
growth rate. However, a crucial difference exists between
tl‘lese agents and NGF in terms of reverse transforming
thential. Second messenger analogs, at least those tested
tQ date, do not induce persistent effects. The
Q ifferentiating effects of agents such as Bromo-cAMP and
iQnomycin are readily reversible, whereas the effects of NGF
bfirsist following its removal from the culture medium
( Reynolds and Perez-Polo, 1975; Marushige et al. 1987) . This
§anept is so fundamental to the reverse transforming ability
Q. IE NGF that it requires further emphasis. Because the
hfioplastic cells remain following reverse transformation, it

3‘ Q essential that the differentiating effects persist. The

ail

repl:
Ltd‘s:
the I

asa

Rem
indi
tram

demo

ofn

glio

Ufa

ing

that

37
aim is to permanently remove neoplastic cells from the
replicative-competent pool. Second messenger analogs do not
induce persistent effects. It is the persistent nature of
the NGF-induced response that makes it particularly attractive
as a reverse transforming agent.

Results from m studies (Greene and Tischler, 1976:
Reynolds and Perez-Polo, 1981: Sonnenfeld and Ishii, 1982)
indicate that NGF is capable of reversing some of the
transformed properties of susceptible tumor cells. The data
demonstrate that:

1) NGF is capable of initiating a response in a variety
of neurogenic tumor cell lines including: pheochromocytoma,
glioma and neuroblastoma cell lines:

2) NGF treatment consistently results in the development
Of a more differentiated phenotype in responsive cell lines:

3) a common consequence of NGF treatment is a reduction

in growth rate or complete growth arrest:

4) experiments involving a variety of cell lines indicate

that these effects are persistent.

The first attribute indicates that NGF may be effective

Q“ a variety of neuroectodermal tumor types. The last three
bIt‘operties were described as crucial attributes of a reverse
hbansforming agent. Together, these results demonstrate that

I Q1? has significant reverse transforming capabilities in

mg.

etr
car

the

38

With this background, let us now examine the results from

111.1119 studies .

MW

1111112119 experiments have contributed a great deal to the
wealth of evidence confirming the reverse transforming
capabilities of NGF. Such information is an invaluable first
step in the process of evaluating the reverse transforming
potential of NGF. However, many compounds that prove
effective in the petri dish are unable to duplicate this
initial promise in vivg. This may be due to a variety of
factors including an inability to traverse physical and
physiologic barriers or a failure to elicit a response in the
complex environs of the living organism. Another concern is
the potential for systemic toxicity. These considerations
I“lecessitate evaluation of the potential therapeutic agent in
m. Upcoming sections address this aspect and provide
Q\"idence that NGF also exhibits reverse transforming

Q§pabilities M19.

W

The first type of mm system to be discussed'will be
§ model of tumor induction involving the use of
Q‘hhylnitrosourea (ENU) . One of the means by which a treatment
§§n be evaluated is to first induce tumor formation with a

Q*iemical carcinogen and then follow with the administration

of
one
As 1

tuac

admi
resz
of ‘
l96é
al.

dew

the

Wit]
Pup:

Pro:

193.

eitl

afte

39
of the proposed treatment. Such an approach requires
consistent tumor development in a specific organ or tissue.
As ENU is capable of inducing a high incidence of neurogenic
tumors in susceptible animals, it has proven to be a valuable
model for the study of neurogenic tumors. 50 mg/kg ENU
.administered to pregnant rats on the 20th day of gestation
:results in the formation of neurogenic tumors in nearly 100%
of the offspring (Druckrey et al. 1966a; Ivankovic et al.
1966: Wechsler et al. 1969: Druckrey et al. 1972: Koestner et
a1. 1971; Swenberg et al. 1972). With such reliable tumor
development, this method is an indispensable tool in assessing
the effects of therapy on several neurogenic tumor types.
The effect of NGF on the development of ENU-induced
iruunors has been examined and it appears that NGF levels play
a major role in the sensitivity of the nervous system to
malignant transformation by ENU. This statement is supported
by several lines of evidence:
1) Initial studies showed that treatment of pregnant rate
With NGF prior to ENU exposure, or postnatal treatment of the
Pups, resulted in a significant reduction in early neoplastic
PrOliferations (ENPs) in the trigeminal nerve of the offspring
by 90 days of age (Vinores and Koestner, 1982: Camp et al.
1984).
2) In subsequent experiments, treatment of rats with NGF
either transplacentally before ENU exposure or postnatally

after ENU exposure resulted in a reduction in neoplastic

pro]

. n
K‘,
(I)
n -

earl

'01)

car:

Phy:
MU:
Sta:

in 1

inc]

Car,

40
proliferation in the trigeminal nerves (Vinores and Koestner,
1983).

3) Administration of anti-NGF antibodies resulted in the
earlier appearance of tumors suggesting that the level of NGF
modulates the carcinogenic response of the nervous system to
ENU (Stahn et al. 1975; Vinores and Perez-Polo, 1976) .

4) Treatment with NGF resulted not only in a decrease in
the number of ENU-induced tumors that developed but also in
loss of the neurotrophism that ENU normally exhibits in rats.
In other words, when ENU is preceded by NGF, a high incidence
of neural tumors is no longer produced by the resorptive
carcinogen.

5) Nice, which characteristically have higher
Physiological levels of NGF than rats, exhibit a
neurologically nonspecific response to ENU irrespective of the
Stage of development and ENU induces very few neural tumors
in mice.

6) However, the number of trigeminal nerve neurinomas
and central nervous system gliomas in mice is increased nearly
f°urfold by prior treatment with anti-NGF IgG (Vinores 1976) .

The earlier appearance of tumors in groups where anti-
NGF antibody treatment preceded ENU exposure suggests that a

reduction in the endogenous levels of NGF may prevent the
nOll'lllal NGF-induced maturation of the target cells, thereby
increasing the number of ENU-sensitive cells available to the

carcinogen. The reduction in neurogenic tumors following

A‘ I
VJ

a p«
Stu:
a1:
tree
rat:

to

NGF
dew
Per:

init

is

mt:
M
or t
rece
in t

was

 

41
perinatal NGF administration suggests that NGF may render
target cells in the central and peripheral nervous systems
more resistant to ENU carcinogenesis, possibly by promoting
their maturation.

It does not appear that NGF treatment simply delays the

onset of tumor formation. Rather, NGF treatment results in
.a persistent suppression of tumor development. In a one year
study in which the offspring of ENU-treated rats were
administered NGF therapy, treatment resulted in a significant
(P < 0.01) decrease in the number of neurinomas in NGF-
treated groups compared to the untreated controls (16 of 34
rats in the NGF treated group developed neurinomas compared
to 29 of 34 control animals) (Raju et al. 1989). As the
resorptive carcinogen was administered prior to NGF treatment,
NGF appears to restrain the promotional phase of tumor
development. These results suggest that NGF causes a
Persistent suppression of the tumor forming ability of ENU-
1nitiated cells.

The nature of the response of ENU-initiated cells to NGF
is further clarified by immunohistochemical studies. In
untreated animals, 7 of 29 neurinomas exhibited positive
inIllinltnohistochemical staining for NGF receptors whereas none
‘33 the tumors in the NGF-treated group were positive for NGF
re<‘Jeptors (Raju et al. 1989) . The counterpart to the tumors
in the untreated control group that exhibited elevated NGFR

“98 not apparent in the NGF-treated group. Thus, it appears

to l
ind:
(Vi:

stuc

42
that NGF treatment suppressed the development of tumors from
initiated cells possessing sufficient NGFR to exhibit positive
immunohistochemical staining for the receptor protein. These
results suggest that NGF treatment restrained the progression
of a subpopulation of ENU-initiated cells possessing elevated
NGFR.

Results from the W models of tumor induction
demonstrate three important characteristics of the response
to NGF: 1) NGF treatment results in a reduction in the ENU-
.induced neoplastic proliferation in the trigeminal nerves
(Vinores and Koestner, 1983): 2) results from a one year
study indicate that these effects appear to be persistent
(Raju et al. 1989): and 3) whether or not an initiated cell
Will respond to NGF treatment appears to be dependent on the

Presence of NGFR.

W
The second system used to evaluate the effects of NGF in
2.1qu utilized the implantation of the F98 anaplastic glioma
Cell line into syngeneic animals. These experiments offer
Pel‘haps the most convincing evidence for the M19 efficacy
°f NGF as they demonstrate that NGF therapy is capable of
diminishing the growth of tumor implants in the environs of
9 living organism. This evidence is supplied by a series of
°XP8riments. In the first, rats received intracerebral

illllfldlnts composed of a mixture of F98 anaplastic glioma cells

C6 C4
Md I
does

with

inocx
sdrv:

did 5

dimir
exam

lorph

“spat
deopl

Effec

43

and NGF-producing C6 cells. Rats receiving the NGF-producing

C6 cells survived longer than those receiving only F98 cells

and 30% of these rats were tumor-free at 90 days (Vinores and
Koestner, 1981) . Additionally, when F98 cells were treated
with NGF prior to inoculation, the survival of the recipient
was increased significantly (P< 0.05) (Vinores and Koestner,
1980) . Finally, if untreated F-98 glioma cells were
inoculated and the recipient treated with NGF, the animals
survived even longer (Vinores and Koestner, 1980) . Not only
did survival time increase significantly (P<0.00005) , but also
the tumor volume was reduced and the tumor growth rate
diminished (Vinores and Hoestner, 1980) . 0n histologic
examination these tumors exhibited a more differentiated

morphology (Vinores and Koestner, 1980) . ‘
11.2119 tests support ill—11L“! evidence that NGF is

capable of reversing some of the transformed properties of

l"Ooplastic cells of neurogenic origin and verify that NGF is

fiffective m.

up...” ; . ,e G- a; — . .. 11.?!‘1001 9;. 1w.

Before concluding this introductory chapter, it seems
aPpl’mprzlate to underscore the relevancy of NGFR expression and
the signal transduction pathway. Earlier in this chapter, it
“98 postulated that malignant transformation, or an integral
eVent in malignant transformation, may be associated with an

aberrant or incomplete response to normal effectors regulating

ii ft
post‘
invo
show.
em
line
path

th
mp
comp
subs
tran
heur
abno
al. ;
that
Rett
aPpe
tram

I819

expre

44

differentiation and growth. A case could be made for this
postulate when considering the results from experiments
involving several neuroblastoma cell lines. Research has
shown that there are at least three different defects in the
expression or function of the NGFR in neuroblastoma cell
lines, thereby disconnecting one of the normal differentiating
pathways in these cells. These defects in the NGF-NGFR
pathway may represent an early change in developing
sympathetic neuroblasts which leaves them in a proliferative-
competent and undifferentiated state in which they may develop
subsequent mutations which ultimately lead to malignant
transformation (Azar et al. 1990). A proportion of human
neuroblastomas have been reported to contain specific
abnormalities of chromosome 17 or chromosome 17q (Biedler et
a]. . 1980: Gilbert et al. 1984) , which is the chromosome region
that carries the NGFR gene in humans (Huebner et al. 1986:
Retitig et al. 1986). As malfunction of the NGFR system
1appears to be relevant in the discussion of neoplastic
transformation, activation of the NGF/NGFR system may be
relevant in the process of reverse transformation.

Host neuroblastoma cells lack functional NGF-NGFR signal
transduction cascade (Matsushima and Bogenmann, 1990) . The
“TI-A NB cell line does not express functional NGFR (Bogenmenn
9"- al. 1983: Matsushima and Bogenmann, 1990) . Transfection
0f the human NGFR cDNA into these cells results in the

exPression of high-affinity cell membrane-bound NGF receptors

 

 

45

(Hatsushima and Bogenmann, 1990) . These cells are then able

to respond to NGF. They exhibit transient c-fos activation

in response to the growth factor which was followed by neurite

outgrowth. NGF treatment resulted in the cessation of DNA

synthesis as indicated by studies in which 95% of
morphologically differentiated cells had not incorporated 3H-

thymidine while more than 80% of the non-responsive cells were

labeled (Matsushima and Bogenmann, 1990) . Differentiated

cells maintained their differentiated state upon removal of
NGF (Hatsushima and Bogenmann, 1990) . The authors concluded
that, following transfection with NGFR cDNA, NGF was able to
stimulate these cells to terminally differentiate (Matsushima
and Bogenmann, 1990).

This last section underscores several points that form
the basis for our research. It would appear that defects in
the NGF-response pathway may be important in neoplastic
transformation. Conversely, it is evident that NGF treatment
Can. in the presence of a functional response pathway, reverse
I"any of the transformed properties of neoplastic cells by

inducing a state resembling terminal differentiation.

W

Not long after the NGF receptor was discovered, a number
(’15 tumor cell lines were identified that possessed abundant
NGFR. This finding lead to speculation that aberrant

exPression of NGF and/or its receptor may have been important

'9
in.

net
i:

{8‘

he:

ne
KG
of
he

re

Ch

46

in the process of neoplastic transformation in these cells.
That theory has since been discarded. After reviewing the
information presented in this chapter, it is easy to
understand why this theory was abandoned. In direct contrast
to the majority of "growth factors," NGF acts, not as a
mitogenic stimulus, but rather, as a differentiating agent.
It stimulates both normal and neoplastic cells of
neuroectodermal origin to develop the characteristics of more
differentiated cells. In the process, it is capable of
reversing some of the transformed properties of these
neoplastic cells.

The information presented in this introductory chapter
leaves little doubt as to the validity and potential benefits
of this line of research. These promising results indicate
a need to further investigate the reverse transforming
Potential of NGF. Principal among these needs is the
necessity to evaluate additional cell lines for response to
“GP in order to develop a better understanding of the spectrum
of this compound's potential. As the 1111119 effects of NGF
have been evaluated in but a single cell line, another
research priority is to evaluate additional cell lines in
11.232. These considerations will be addressed in upcoming

chapters dealing with my thesis research.

CHAPTER 1

THE REVERSE TRANSFORMING EFFECTS OF NERVE GROWTH FACTOR ON

FIVE HUMAN NEUROGENIC TUMOR CELL LINES: W RESULTS

sped
eve;
dif
neo;
1180:
Ha

EVa

 

CHAPTER 1

The Reverse Transforming Effects of Nerve Growth Factor on

Five Human Neurogenic Tumor Cell Lines: 111115.19 Results.

MEL
The ability of nerve growth factor (NGF) to stimulate the
differentiation of a variety of normal and neoplastic cell
lines of neurogenic origin suggests that it has potential as
a "reverse transforming agent." The binding of NGF to
specific surface receptors may set in motion a series of
events whereby neoplastic cells are stimulated to
differentiate to a point where they no longer exhibit
neoplastic behavior. Such a process is the antithesis of
neoplastic transformation and is termed reverse
transformation. Five human neurogenic tumor cell lines were
evaluated for their response to NGF mm to examine
whether the induced changes were consistent with a reverse
transforming response. Crucial alterations essential for
reverse transformation include: 1) the development of a more
differentiated phenotype, 2) a decrease in the rate of
Proliferation, and 3) persistence of the induced changes.
Each cell line exhibited a morphologic response to NGF which

47

in:
fr:
the

USE

ins
by
th.
to
0p'
di
tr
tr
in
di
di
Ce
th.
he<

tr;

 

48
was generally consistent with the development of a more
differentiated.phenotypes NGF treatment decreased the growth
rate of each cell line and produced complete growth arrest in
a single line. The effects of NGF were persistent as the
induced morphologic changes remained following removal of NGF
from the culture medium. These in_yitrg experiments indicate
that NGF has reverse transforming potential and may prove

useful for treatment of neurogenic tumors.

W

The standard approach to cancer therapy generally
involves selective removal or destruction of neoplastic cells
by surgery, radiation or chemotherapy, or a combination
thereof. However, novel therapeutic approaches may be able
to augment these traditional treatment strategies. One such
option is the management of cancer through control of
differentiation and growth of neoplastic cells. malignant
transformation, or an integral event in malignant
transformation, may be associated with an aberrant or
incomplete response to normal effectors regulating
differentiation and growth. If this is true, then specific
differentiating agents might be able to stimulate neoplastic
cedls to differentiate and reverse many of the properties of
fame transformed cells. Such a process is the antithesis of
neOplastic transformation and is termed reverse

transformation. It is based on the premise that certain

 

{8"

st:

fl€<

Re}

Rej

Ha:

Bo:

18‘

ch

in

an

19

Ma

he

di

St

or

‘31

no

 

49
agents may be able to activate processes which stimulate
neoplastic cells to differentiate to a point where they no
longer exhibit neoplastic behavior.

Nerve growth factor (NGF) represents one such potential
reverse transforming agent. The capability of NGF to
stimulate the differentiation of a variety of normal and
neoplastic cell lines of neurogenic origin (Waris et al. 1973;
Reynolds and Perez-Polo, 1975: Greene and Tischler 1976:
Reynolds and Perez-Polo, 1981: Levi-Hontalcini and Aloe, 1985:
Harushige et al. 1987: Kumar et al. 1990: Hatsushima and
Bogenmann, 1990) suggests that it may prove useful as a
reverse transforming agent. In particular, studies indicating
that NGF retards cell proliferation and induces morphologic
changes in anaplastic glioma cells (Marushige et al. 1987 ) and
induces the differentiation of PC12 pheochromocytoma (Greene
and Tischler, 1976) and neuroblastoma cells (Waris et al.
1973; Reynolds and Perez-Polo, 1981: Perez-Polo et al. 1982;
Natsushima and Begenmann, 1990) demonstrate that NGF is
capable of reversing some of the transformed properties of
neurogenic tumor cells. It is proposed that the
differentiating abilities of NGF could be utilized to
Stimulate neoplastic cells of neurogenic (neuroepithelial)
origin to develop a more differentiated phenotype, thereby
eliminating the capacity for uncontrolled growth.

The ability to induce a more differentiated phenotype is

not unique to NGF. A variety of chemical agents, primarily

50

analogs of second messengers, also possess this capability
(Perez-Polo et al. 1982; Marushige et al. 1989b). However,
an important difference exists between these agents and NGF
in terms of reverse transforming potential. The
differentiating effects of agents such as Bromo-cAMP and
ionomycin are readily reversible, whereas the effects of NGF
persist following its removal from the culture medium
(Reynolds and Perez-Polo, 1975: Marushige et al. 1987). It
is the persistent nature of the NGF-induced response that
makes NGF particularly attractive as a reverse transforming
agent.

The aim of this study was to determine whether NGF was
capable of inducing changes indicative of reverse
transformation in well characterized cell lines derived from
malignant human brain tumors. This was accomplished by
evaluating five human neurogenic tumor cells lines for in
yitzg_response to NGF. Each cell line was evaluated for NGF-
induced changes in cell morphology. Observations were made
as to the capacity of NGF to a) induce morphologic changes
suggestive of'a more differentiated phenotype and b) decrease
the rate of proliferation. As persistence of the induced
changes is an essential feature of a reverse transforming
agent, an evaluation of how long the NGF induced morphologic
changes remained following removal of NGF from the culture

medium was also undertaken.

The success or failure of NGF as a treatment depends,

51

from the onset, on the presence of NGF receptors (NGFR). The
initial event in NGF-mediated processes is the binding of the
growth factor'to specific surface receptors (Levi et al. 1980:
Levi-Montalcini 1987: Hempstead et a1. 1989). Experimental
evidence indicates that the presence or absence of NGFR may
determine the outcome of NGF treatment (Raju et al, 1989).
Therefore, an essential component of our research was to
identify the NGF receptor density of each cell line evaluated.

Investigation. of the effects of’ NGF on. growth and
differentiation.parameters, combined with information on NGFR
density, should provide a better understanding of the reverse
transforming potential of this growth and maturation factor

on human neuroepithelial tumors.

WWW

Wile:

Four of the human cell lines used in these experiments,
the D54 mixed glioma, glioblastoma multiforme cell lines 0118
& U251 and the TE671 medulloblastoma cell line, were generous
gifts from Dr. Bigner at Duke University. A fifth cell line,
the Hs294 malignant melanoma line, was obtained from the
American Type Culture Collection. Stock cultures were
maintained in Dulbecco's modified Eagle's medium containing
20% fetal bovine serum (Sigma) (10% FBS for the H8294 cell

line) and 4 mn glutamine in a humidified chamber with 8% CO2

52

at 37 C. Cultures were seeded in 12 well cluster plates
(Corning) at a density of 3.0 x 103 to 2.5 x 10‘
cells/well (1.0 ml/well). Cells were seeded in DMEM + 20%
F88. The medium was replaced with HL-l serum-free medium
(Ventrex) after 24 hours. The following day, the medium was
again exchanged with HL-l to which varying concentrations
(0.25, 0.5, 1.0, 2.0 and 4.0 ug/ml) of 2.58 NGF (Boehringer
Mannheim) were added. Cultures were observed daily via phase-
contrast microscopy. For cells that became detached from the
culture dish in response to NGF, viability was determined by
the trypan blue exclusion test.

Cultures were treated comparably for the growth studies.
Paired control and NGF-treated cultures were trypsinized at

intervals and counted with either a hemocytometer or Coulter

Counter.

32211321191531.1114};

Cultures were treated as for the growth and morphologic
studies except seeded at lower densities (5 x 10" cells/well) .
Each 12-well cluster plate was divided into three groups. In
the first, the NGF-control group, the medium was refreshed
with 4 ug/ml NGF in HL-l every fourth day throughout the
experiment. In the second, the persistence group, cells were
cultured in the presence of NGF for 48 hrs. The NGF-
containing medium was then removed and replaced with HL-l.

In the third group, the untreated control group, cells were

 

 

 

53
grown in HL-l throughout the experiment. The culture medium
was refreshed with HL-l every 4th day in both the HL-l control
and NGF-persistence cultures. Cultures were observed daily

via phase-contrast microscopy.

WM

NGF receptor (NGFR) density was determined for each cell
line using a modification of the ELISA procedure described by
Doherty et. a1. (1988). Cells were seeded at 15,000
cells/well in a 96-well microtiter plate. Following two days
in culture, the cells were fixed with paraformaldehyde.
Receptor density was determined using the ME20. 4 anti-receptor
antibody (1:20, Amersham) and an anti-mouse horseradish
peroxidase conjugate (1:1000, Sigma). Optical density (OD)
was measured with the Dynatech.HR600 microtiter plate reader.
A standard curve was developed using a cell line with known
NGF-binding properties, the H8294 malignant melanoma cell line
(Fabricant et al. 1977). OD was determined at various Hs294
seeding densities. Protein concentrations were determined
from which cell numbers were calculated. A standard curve was
generated that compared OD to H3294 cell numbers. Unknowns

were then compared to the standard curve.

WWW
HE20.4 is a monoclonal antibody that binds to the human

NGFR and in so doing inhibits the binding of NGF to its

54
receptor (Ross et al. 1984). Preincubation of cultures with
ME20.4 should greatly diminish or prevent the development of
NGF-induced changes. Cell lines were seeded at 1.25 x 10‘
cells/well in a 12-well cluster plate. On the second day the
medium was replaced with HL-l. On the third day, the plate
was divided into 6 groups: an HL-l control group, two NGF-
treated groups (2.0 and 4.0 ug/ml NGF) , an ME20.4 control
group (1.8 ug/ml) and two groups which were pre-incubated with
HE20.4 (1.8 ug/ml) 60 min. prior'to the addition.of either 2.0
or 4.0 ug/ml NGF. Groups were compared to determine whether
pre-incubation with.HE20.4 could prevent or diminish the NGF-

induced morphologic changes.

Results

An ELISA procedure was developed to determine NGFR
density (Table 1.1). Results from these tests indicated that
the H8294 cell line had abundant NGFR. Tumor cell lines of
glial origin had very few NGFR. The TE671 medulloblastoma
cell line exhibited a slightly higher density of receptors
than the glial tumor lines. The relative densities obtained
were consistent with those reported for cell lines of similar
origin (Fabricant et. al. 1977). Receptor' numbers were
estimated using information from published reports (Fabricant
et al. 1977). These calculations indicate that the H8294 cell
line would have approximately 6.4 x 105 receptors per cell,

the glial cultures approximately 6.0 - 10.0 x 10’

55

Table 1.1. Relative NGFR densities of the
five human neurogenic tumor cell lines

 

Cell Line Binding of ME20.4
CArbitrary units i S.E.)

 

H8294 106.7 i 12.36
TE671 2.9 i 0.58
054 1.7 i 0.34
U251 1.4 i 0.17
U118 1.0 i 0.31

 

Table 1.1 illustrates the relative NGFR density of the
human tumor cell lines expressed in terms of the binding
of ME20.4 (antibody directed against the human NGFR)

as determined by standard ELISA.

' 1.0 arbitrary unit corresponds to 0.165 O.D. units per
2.5 x 10‘ cells.

56
receptors/cell and the TE671 cell line approximately 1.7 x 10‘
receptors/cell.

NGF treatment resulted in the morphologic transformation
of each of the five human neural crest tumor cell lines
evaluated. The type of response and dose at which it was
observed varied with the cell line (Table 1.2). In general,
a morphologic response was first identified at a dose between
0.25 and 0.5 ug/ml NGF and was maximal at concentrations of
2.0 to 4.0 ug/ml NGF. Two of the cell lines, D54 and U118,
responded in a similar manner. Within 8 hours following
exposure, cells became more compact, 3-dimensional and
produced an increased number of very fine, short processes.
Over the next 16 to 24 hours, the cells became increasingly
more compact. eventually attaining a spherical shape with
concomitant loss of their processes. After 24 to 48 hours,
most of the cells had detached from the surface of the culture
dish and died. Cells that remained attached tended to be
present in clusters exhibiting numerous short, fine peripheral
processes. (Figure 1.1 C-D)

The remaining three cell lines (H8291, U251 and TE671)
exhibited a similar initial response. Cells became more
compact, 3-dimensional and produced more highly developed
processes. However, rather than becoming spherical and
detaching from the culture dish, these cells developed a more
differentiated phenotype characterized by an increasingly more

compact and 3-dimensional morphology and the production of

57

.omufiOflHm we: encommmu Huawxsa m soars us Aaa\osc moz no sofiusuusmocoo may

one .mocmso owmoHOAQMOE manneuflusoow cm m>mflsoe ou oouesveu Aaa\msc cowuouucmosoo
moz Hmawswa on» «sowusasawum Hosfixea ocesoaaou oe>uomno was emcmso DeooHocmuoa

m guess as mafia undefined one meanness“ mans» one .mosfia Haoo Mossy ofismmouses
amass e>flu may no some you mueumasnss omsommeu mesn as» no eceauso as we ~.H manna

 

o.e o.e o.v o.~ o.¢ “oncommmu Heawxsa you omuasumu
AHa\osc emoo huz asses“:

m~.o m.o m~.o m~.o m.o ”mmsoaneu owmoaonmuofi Mom
omufisveu Aaa\osc mmoo moz Hades“:

mu: m was em mu: m an: «N an: em "moz Ha\es o.¢ um massage
odooHonaoa om>ummno unmeausm

 

QHHD Hbmmfi «ma wmmmm HmND ”OCMH HHOU

 

.mHMHMIaH moz ou mecHH dado Mossy
amass was no encommmu ms» no“ mueumfimuen mush mmoo one wafia um.a manna

58

Figure 1.1: Phase-contrast photomicrographs of _the NGF-
induced morphologic changes in the D54 (mixed glioma) and
H8294 (malignant melanoma) cell lines.

 

Phase-contrast photomicrographs of the following groups were
taken on day 4 post-NGF treatment: A) H8294, untreated
control; B) H8294, 4 ug/ml NGF; C) D54, untreated control; D)
D54, 2 ug/ml NGF. In response to NGF the H8294 cell line
develops a more compact, three-dimensional morphology and
numerous short processes. The D54 cell line responds to NGF
by becoming considerably more compact and by developing
multiple, long, slender processes. (Bar equals 10 um)

59

numerous, well developed processes. This change varied from
the production of numerous short processes (Figure 1.1 A-B)
to the development of an extensive network of long slender
branching processes that formed an interconnecting network
with neighboring cells. (Figure 1.2 A-B) These changes were
quite stable in the U251 cell line. However, following the
development of a maximal morphologic response, cells from both
the H8294 and TE671 lines'would.gradually'begin.to}detach from
the culture dish.

Two procedures were undertaken to try and determine if
the morphologic changes were the result of binding of NGF to
specific surface receptors. In the first, 2.0 ug/ml
cytochrome C, a compound similar to NGF in both size and
charge (Korshing and Thoenen, 1983), was added to the culture
medium to determine if the NGF-induced morphologic changes
were caused simply by the addition of a positively charged
molecule to the culture medium. Differences between the
cytochrome C-containing cultures and untreated control
cultures were not identified, indicating that the changes
observed following NGF treatment were apparently not due to
the addition of a charged molecule. Preincubation with
HE20.4, an antibody that binds specifically to the human NGFR
(Ross et al. 1984), greatly diminished the development of NGF
induced morphologic changes, suggesting that the morphologic
transformation induced by NGF was the result of the binding

of the growth factor to specific surface receptors.

60

Figure 1.2: Phase-contrast photomicrographs of the NGF-
induced morphologic changes in the 0251 glioblastoma
multiforme cell line.

 

Phase contrast photomicrographs of the following groups were
taken: A) U251, untreated control, day 11; B) U251, 4 ug/ml
NGF, day 11: C) U251-Persistence, NGF control, 4 ug/ml NGF
every fourth day, day 43; D) U251—Persistence, persistence
group, 4 ug/ml NGF day 1-2, HL-l media change every fourth
day, day 43. In response to NGF, the U251 cell line becomes
much more compact, three-dimensional and develops an extensive
network of long slender branching processes that tend to form
an interconnecting network with neighboring cells. These
changes are quite persistent as is evidenced by the retention
of the NGF-induced changes in morphology even after the cells
had been cultured in the absence of NGF for 41 days. (Bar
equals 10 um)

61

The morphologic changes induced by NGF were accompanied
by a decrease in growth rate in each cell line. On day 4
post-NGF treatment, this ranged from a minimal decrease in the
TE671 cell line, to a nearly 50% decrease in the D54 cell line
(Figure 1.3). Due to various technical problems (cells
becoming detached in response to NGF), the development of a
complete growth response curve was possible only in the U251
cell line. NGF treatment resulted in a reduction in the
growth rate of this cell line with complete cessation of
growth after day 10-12 (Figure 1.4).

An evaluation of how long the NGF-induced morphologic
changes persisted in the absence of the compound was also
undertaken. Persistence was documented in the D54, H8294 and
U251 cell lines. The evaluation was limited by the varying
behavior and survival of the cells in culture. In the D54
cell line, the NGF-induced morphologic changes persisted for
7 days following removal of NGF from the culture medium.
Changes persisted in the H8294 cell line for 14 days following
removal of NGF. These values were not a true reflection of
the persistent nature of the NGF-induced response in these
cell lines. Even when grown on polylysine coated plates,
cells from D54 cultures became detached in response to NGF
after 7 days. In the H8294 cell line, morphologic changes
became obscured once these cultures became confluent at
approximately day 14. At the termination of each experiment,

there was no evidence of reversal of the NGF-induced

62

Figure 1.3: The effect of NGF on the cell counts of four
human neurogenic tumor cell lines on day 4 post-NGF treatment.

120

 

U251 H8294 TE671
100

 

 

D
O

 

 

 

 

 

 

/%*
1
...
|_.

ooooooooooooooooooooooo

 

 

L 0 000 0
I .0 0000
00 90000000
D 0 900000000001
b00000. 00000000000
b00000. 00000 000001
b00000 900000000001
0000 0

h

I... o 000 o 01
00000000000 so
oooooooooooooooooo
eooooop oooooooooo 1
eeoooopooooeooeeoe
b00000! ..... 000001
b.0000! oooooooooo 1
b00000! eeeeeeeeeee
\\ 1 he. eopeopeeeeqoot 1 l

O 2 4 0 2 4 0 2

 

 

Cell Counts
(percent of control)
«n
c:

 

 

l0 0-
O O

 

 

 

 

 

 

 

 

 

 

 

 

 

 

..

NGF Concentration (ug/ml)

-. ____.... .—___-._ ___. . -. .l .,..__—_. _ .__—.-.——

 

Cell Counts, Day 4 post-NGF treatment. Values are expressed
as a percentage of the corresponding control (i S.E.) so that
the cell lines can be readily compared.

63

Figure 1.4: The effect of NGF-Treatment on the growth rate
of the 0251 Glioblastoma multiforme cell line in_yi§r_.

 

   
     

 

 

 

 

100_M

qA _,.
O

H ...
35

S 10

C:
:3
O
U
H
H
(D

U ...,

1 l l l J l l l 4 l
0 £2 4 (5 8 12 16

Days Post-NGF Treatment

 

 

 

 

 

 

Figure 1.4 illustrates the effect of NGF-treatment on the
growth curve of the 0251 glioblastoma multiforme cell line.
Untreated control cultures (.) rapidly proliferate and reach
confluency by day 8 whereas the 4.0 ug/ml NGF-treated (o)
cultures exhibit a gradual reduction in growth rate with
eventual growth arrest by day 10.

64
morphologic changes. Thus, it was unclear how much longer
the effects would have persisted had technical problems not
prohibited further evaluation. Persistence was best
demonstrated in the 0251 cell line (Figure 1.2 C-D) where the
NGF-induced morphologic changes persisted for 53 days
following removal of NGF from the culture medium. Cells which
were exposed to NGF for 2 days retained a morphology
essentially identical to that of cells in NGF control cultures
and showed no evidence of reverting to a morphology similar
to untreated controls even after they had been grown without
NGF for 53 days. The experiment was terminated at this point
due to the unhealthy condition of the cultures following

prolonged maintenance in HL-l.

Discussisn

Control of growth and differentiation of tumor cells is
a basis for cancer therapy and is the principle behind the
concept of reverse transformation, However, since neoplastic
cells are not directly eliminated through the use of reverse
transforming agents, very stringent criteria need to be met
when evaluating such compounds as potential therapeutic
agents. Treatment must render neoplastic cells susceptible
to the body's intrinsic control mechanisms or result in a
persistent suppression of growth if a therapeutic advantage
is to be gained.

Our data indicate that NGF is able to induce several

65
changes in neoplastic cell lines critical for reverse
transformation. It was capable of inducing a more
differentiated phenotype and of reducing or arresting tumor
growth Mm. These are both vital components of a reverse
transforming response. However, in order for a reverse
transforming agent to be therapeutically useful, an additional
characteristic is desired. An valuable feature of agents that
rely on differentiation as a means of producing a therapeutic
effect is persistence of the differentiated phenotype. NGF
induced changes were highly persistent in_yit;g. It is the
persistent nature of this response that separates NGF from
many other compounds capable of inducing differentiation and
suggests that NGF may prove particularly useful as a reverse
transforming agent. Furthermore, reports indicate that the
effects of NGF may also be persistent in viyg. Following a
one year study in which neurogenic tumors were induced by
transplacental exposure to ethylnitrosourea (ENU) ,
significantly fewer animals developed neurinomas in the NGF-
treated group (47% as compared to 85% for the untreated
controls; P < 0.01), indicating that NGF has the ability to
persistently suppress the tumor forming ability of ENU
initiated cells (Raju et a1. 1989). As the resorptive
carcinogen was administered prior to NGF treatment, NGF
appears to restrain the promotional phase of tumor

development.

It is generally believed that NGF-responsive cells

66

possess two types of NGFR, high affinity receptors (thought

to be necessary for biologic response) and low affinity
receptors (Landreth and Shooter, 1980; Sonnenfeld and Ishii,
1985: Vale and Shooter, 1985: Green and Greene, 1986; Kumar
et al. 1990: Welcher et al. 1991). However, a recent report
(Buxser et a1. 1990) provides strong evidence for the
existence of but a single type of NGFR. The confusion appears
to stem from the internalization of the NGF-NGFR, mimicking
a second class of receptors (Buxser et al. 1990). In our
experiments, an ELISA was used to determine the NGFR-densities
of the various cell lines under study (Table 1.1). The
relative densities obtained were consistent with information
on similar cell lines (Fabricant et a1. 1977). A correlation
between receptor density and response to NGF was not
identified, suggesting that additional factors are involved
in the responsiveness of cell lines to NGF besides NGFR
density. The Hs294 cell line, which possessed the highest
density of NGFR, exhibited a maximal response at the lowest
dose of NGF, however, the magnitude of this response was not
as great as was observed in several cell lines exhibiting
fewer receptors. The cell line of intermediate density, the
T3671 line, proved to be the least responsive cell line.
Three of the cell lines, 054, U251, U118, had low NGFR.1evels.
Despite low receptor numbers, a reverse transforming response
was elicited in each cell line, albeit at high doses of NGF.

These results indicate that low levels of NGFR, such as were

67

identified in the cell lines of glial origin, were not a
significant obstacle to reverse transformation if ample NGF
was delivered to the neoplastic cells. Stereotaxic treatment,
which enables the targeting of agents to specific sites, may
allow for the administration of a sufficient dose of NGF to
neoplastic foci to stimulate a reverse transforming response
in v1 9.

The mechanisms by which NGF affects normal and neoplastic
cells are currently not completely understood. However, if
NGF is to be useful as a reverse transforming agent, it must
have persistent epigenetic effects that reverse or block one
or more of the steps that lead to neoplastic transformation.
It has been shown that NGF treatment results in the
heterologous down-regulation of epidermal growth factor
receptors (Lazarovici et al. 1987). Thus, as part of its
differentiating response, NGF appears to diminish pathways
that favor cell proliferation. This may be a part of the
mechanism whereby the differentiated phenotype is maintained
(Lazarovici et a1. 1987).

In_yitzg tests indicate that NGF fulfills several key
criteria of a reverse transforming agent. NGF treatment
results in a reduction in growth rate with complete growth
arrest in some cell lines. It stimulates neoplastic cells to
develop a more differentiated phenotype and these effects
appear to be highly persistent. As such, it appears that NGF

possesses significant therapeutic potential. Work is

68
currently underway to characterize the reverse transforming

potential of NGF 1111119.

CHAPTER 1 APPENDIX

IDSIQQEQLiQn

The preceding paper provides information supporting the
reverse transforming potential of NGF in;yi§rg. The material
presented did not include all the information obtained from
in_git;g studies. The scope of the article was limited to
information necessary to demonstrate the reverse transforming
capabilities of NGF in a tissue culture system. Additional
W studies were performed. The results of these
experiments are reported here, in the appendix, because they
provide additional information. on 'the response of ‘these

neoplastic cell lines to NGF in tissue culture.

l!3E-° I 3 l 1 . l

The preceding article did not allow for sufficient space
to illustrate the spectrum of morphologic changes induced by
NGF in all five of the neoplastic cells lines under study.
This appendix provides the opportunity to illustrate these
changes and to furnish a visual record of the NGF—induced
changes in morphology. The "Materials and Methods" for this

sections are described in the preceding article.

69

'70
s o ' o

The U118 glioblastoma multiforme cell line exhibits an
orderly progression of changes in response to NGF. The
initial NGF-induced changes consist of the cell bodies
becoming shortened and more compact with loss of cytoplasmic
processes (Figure 1.5) . As these changes progress, cells
become even more compact, three-dimensional and eventually
develop a spherical shape. These spherical cells then detach
from the surface of the culture dish. This progression can
be observed either by examining a single treatment group over
a period of several days or by examining the various NGF
treatment groups from a single time period. The first changes
were noted in the 4 ug/ml group following 8 hrs' exposure to
NGF. Morphologic changes were observed at a dose as low as
0.25 ug/ml NGF and were maximal at 4.0 ug/ml NGF. The NGF-

induced morphologic changes reached their peak by day 4 post-
NGF treatment.

71

Figure 1.5: Phase-contrast photomicrographs of the spectrum
of morphologic changes induced by NGF in the U118 glioblastoma
multiforme cell line.

Photomicrographs were taken at phase 1. (Bar = 10 um)

A: Day 1 post-NGF treatment, HL-l control: Control
cultures consist of a population of large, fairly flat spindle
shaped cells that generally possess from 1 - 2 processes per
cella Occasional cells exhibit. an angular’ or stellate
morphology.

B: Day 1 post-NGF treatment, 4 ug/ml NGF: Approximately
30% of the cells in this group have become detached from the
surface of the culture dish. The majority of the cells that
remain are spherical in shape. The few non-spherical cells
present are greatly shrunken and are devoid of processes.

C: Day 3 post-NGF treatment, HL-l control: Cells in the
control cultures have retained a morphology similar to that
described for the day 1 control cultures.

D: Day 3 post-NGF treatment, 4 ug/ml NGF: The majority
of cells in this culture have become spherical and detached
from the culture surface. Compared to the control cultures,
there are approximately 70% fewer cells. 80-90% of the cells
that remain attached are spherical. A few of the cells that
remain are ‘markedly shrunken, rod shaped and. devoid of
processes.

72

Figure 1 . 5

 

73
Wm

Components of the NGF-induced morphologic changes
exhibited by the D54 mixed glioma cell line are reminiscent
of changes observed with the U118 cell line. This cell line
also tends to become spherical and detach from the surface of
the culture dish in response to NGF. However, before this
stage is reached, these cells undergo a more dramatic spectrum
of morphologic changes than is observed with 0118 cultures.
Whereas the U118 cell line tends to become progressively more
compact until it eventually reaches a spherical shape, the D54
cell line progresses through a stage where cells exhibit well
developed processes.

Untreated D54 cultures consist of a population of very
large, flat, two-dimensional, irregular to spindle shaped
cells. NGF exposure stimulates these cells to become more
compact and three-dimensional. Cells then become compressed
into long, slender, three-dimensional cells that often exhibit
well developed processes at one or both poles (Figure 1.6).
These cells progress to become considerably more compact and
develop a number of long slender processes. The processes
eventually condense and the cells become spherical and detach
from the culture surface. In response to NGF the D54 cell
line exhibits the most.dramatic decrease in size of any of the
five human cell lines evaluated. These cells progress from
very large flat cells to cells with minute cell bodies

possessing long, slender processes.

74

Figure 1.6: Phase-contrast photomicrographs of the spectrum
of morphologic changes induced by NGF in the D54 mixed glioma
cell line.

 

Photomicrographs were taken at phase 1 on day 3 post-
NGF treatment. (Bar = 10 um)

A: HL—l control: cells in the control cultures consist
of a population of very large, flat, fusiform to irregular
cells that occasionally exhibit short processes.

B: 1.0 ug/ml NGF: These cells differ from the controls
in that they have become much more compact and three-
dimensional, and are beginning to sprout an increased number
of short processes.

C: 2.0 ug/ml NGF: Compared to untreated control cultures,
the cell density in this group has become greatly decreased
as many of the cells have become spherical and have detached.
Those that remain are even more compact and exhibit long,
slender processes.

D: 4.0 ug/ml NGF: The cell density in this group has
decreased compared to the previous group as cells continue to
detach. Those that remain are many times smaller than the
untreated controls and exhibit well developed processes.

75

 

1.6E: Day 1, HL-l control and F: Day 1, 4.0 ug/ml NGF:
These two photomicrographs were taken at phase L (Bar = 20 um)
to better demonstrate the proportion of cells that become
spherical and detach in response to NGF and how NGF treatment
stimulates these cells to become significantly more compact.
Fewer cells are present in the 4.0 ug/ml NGF-treated group.
The cells that remain have not retained the flat, spindle—
shaped morphology of cells in the corresponding control
cultures. These cells have become significantly more compact.
They will eventually attain a spherical shape and detach.

1.6G: Day 9, HL—l control, and H: Day 9, 0.8 ug/ml NGF,
phase 1 (Bar = 10 um): A fairly low dose of NGF was utilized
so that the morphologic changes would become more developed
before the cells detached. These two figures illustrates how
compact and three-dimensional the cells become in response to
NGF and how NGF will induce these cells to develop long
slender processes that from an interconnecting network with
neighboring cells.

76
In the D54 cell line NGF-induced changes in morphology
were noted in the 4 ug/ml group following 6 hrs' exposure to
NGF. Morphologic changes were observed at a dose as low as
0.25 ug/ml NGF and were maximal at 4.0 ug/ml NGF. The NGF-
induced morphologic changes reached their peak by day 4 post-

NGF treatment.

W

The TE671 medulloblastoma cell line undergoes possibly
the most subtle range of morphologic changes in response to
NGF. Though these changes are not as dramatic as the other
four cell lines, they are readily apparent and are consistent
with the development of a more differentiated phenotype. This
cell line develops an orderly progression of morphologic
changes in response to NGF. This progression can be
illustrated by comparing the various NGF treatment groups on
day 3 post-NGF treatment (Figure 1.7) . Untreated TE671
cultures consist of a population of very large, flat, two-
dimensional, spindle-shaped cells that only occasionally
exhibit well developed.processesw Low levels of NGF (0.25 and
0.5 ug/ml) stimulate these cells to develop a more compact,
oval to angular, more three-dimensional morphology with an
increased number of cells exhibiting short processes. As the
dose of NGF increases, the cells become progressively more
compact and three-dimensional, a greater number of cells

produce well developed processes and these processes tend to

77

Figure 1.7: Phase-contrast photomicrographs of the spectrum
of morphologic changes induced by NGF in the TE671
Medulloblastoma cell line.

Photomicrographs were taken at phase 2 on day 3 post-NGF
treatment. (Bar = 5 um)

A: HL-l control: Control cultures consist of a population
of large, flat, two-dimensional, spindle shaped cells that
exhibited an occasional well developed polar process.

B: 0.25 ug/ml NGF: When compared to the control cultures,
cells in this group are more compact, slightly more three-
dimensional and have developed an oval to angular morphology.

C: 0.5 ug/ml NGF: Cells in this group have become even
more compact and three-dimensional than the previous group and
are beginning to exhibit an increased number of moderately
well developed processes.

D:1.0 ug/ml, E: 2.0 ug/ml and F: 4.0 ug/ml: As the dose
of NGF increases to 1.0 ug/ml and above, cells continue to
become progressively' more compact, three-dimensional and
produce and increased number of well developed processes that
tend to form an interconnecting network with neighboring
cells.

78

Figure 1.7

 

79
form an interconnecting network with neighboring cells.

The earliest identifiable morphologic changes were
apparent following 24 hrs exposure to 4.0 ug/ml NGF.
Morphologic changes were observed at doses as low as 0.25
ug/ml NGF and.were maximal at 4.0 ug/ml NGF. The NGF-induced
morphologic changes reached their peak by day 4 post-NGF

treatment.

W

The Hs294 malignant melanoma cell line undergoes an
interesting succession of morphologic changes. The
predominant cell type in untreated H8294 cultures is a very
primitive appearing large, flat, two-dimensional, amorphous
cell that is devoid of processes. In response to NGF these
amorphous cells begin to develop a more compact, angular,
three-dimensional morphology. Whereas the amorphous cells
could be considered the most undifferentiated cell type, these
angular, more three-dimensional cells are an intermediate
stage in the progression towards the development of a more
differentiated phenotype. As either the length of exposure
or the dose of NGF increases, these angular cells become more
compact, three-dimensional and begin to sprout processes.
These cells eventually develop a stellate morphology and
produce numerous, short, often branching processes in response
to NGF. These stellate cells represent the most

differentiated state that H8294 cells develop in response to

80

NGF. Though these NGF-induced changes may not be as dramatic
as those exhibited by the D54 and U251 cell lines, they are
readily apparent and the transition from a population of
primitive amorphous cells to highly differentiated three-
dimensional cells exhibiting numerous processes effectively
demonstrates the ability of NGF to induce Hs294 malignant
melanoma cells to develop a more differentiated phenotype.

The first NGF-induced morphologic changes were noted in
the 4 ug/ml group following 24 hrs' exposure to NGF.
Horphologic changes were observed at a dose as low as 0.25
ug/ml NGF and were maximal at 2.0 ug/ml NGF. The NGF-induced
morphologic changes reached their peak by day 5 post-NGF

treatment.

81

Figure 1.8: Phase-contrast photomicrographs of the spectrum
of morphologic changes induced by NGF in the H5294 Malignant
Melanoma cell line.

These photomicrographs were taken at phase 1 on day 2
post-NGF treatment. (Bar = 10 um)

A: HL—l control: The predominant cell type in the control
cultures was a population of very large, flat, two-
dimensional, amorphous cells that were essentially devoid of
processes. A few scattered spindle shaped cells, angular
cells and an occasional stellate cell are also apparent.

B: 0.25 ug/ml NGF: Compared to the control cultures,
cells in the 0.25 ug/ml NGF-treated group exhibited a
decreased number of amorphous cells with a corresponding
increased number of angular, more three-dimensional cells
exhibiting short processes.

C: 0.5 ug/ml NGF: The morphologic changes are more:
progressed in this group. The percentage of amorphous cells
has gradually decreased compared to the two previous groups
while the number of angular and stellate cells continues to
increase.

D: 1.0 ug/ml NGF: Classical amorphous cells are rarely
identified in this group or in either of the two groups where
higher doses of NGF were employed. Cells in this group
consistently exhibit a more three-dimensional morphology and
the majority have sprouted short processes.

E & F: 2.0 S: 4.0 ug/ml NGF: The morphologic changes
present in these two groups are similar to those described in
the previous group, however, changes continue to progress.
Cells in these two groups are even more compact, three-
dimensional and exhibit increased numbers of processes.

82

Figure l . 8

 

83
W

The U251 glioblastoma multiforme cell line exhibits
perhaps the most pronounced morphologic changes in response
to NGF, however, it takes significantly longer for these
morphologic changes to become fully developed. Whereas the
other four cell lines exhibit fully developed morphologic
changes following 3—4 days' exposure to NGF, it takes 11-12
days for morphologic changes to become fully developed in the
U251 cell line. These cells eventually develop striking
morphologic changes that, once again, readily demonstrate the
ability of NGF to stimulate neoplastic cell lines of
neurogenic origin to development morphologic characteristics
of a more differentiated phenotype.

The morphologic changes in the U251 cell line are best
followed by examining the changes in the 4.0 ug/ml NGF—
treated group over time. Initially, U251 control cultures
consist of a population of very large, flat, two-dimensional,
spindle-shaped to angular cells. Following two days exposure
to NGF, these cells become much more compact, three-
dimensional and a larger proportion of cells attain a spindle-
shaped appearance. As the cell density increases in the
control cultures, the cells become increasingly flattened,
angular’ and. generally' exhibit. a single process or lack
processes altogether. NGF-treated cultures exhibit a
dramatically different morphology. The spindle-shaped cells

in NGF-treated cultures become much more compact and three-

84
dimensional. As the length of exposure to NGF increases,
these cells develop processes which gradually lengthen and
branch until, by day 11-12, NGF-treated cultures consist of
a population of cells with very compact, three-dimensional
cell bodies that have developed numerous long, slender, often
branching processes that tend to form an interconnecting
network with neighboring cells.
The earliest observed morphologic changes were noted in
the 4 ug/ml group following 24 hrs' exposure to NGF.
Horphologic changes were observed at a dose as low as 0.5

ug/ml NGF and were maximal at 4.0 ug/ml NGF.

85

Figure 1.9: Phase—contrast photomicrographs of the spectrum
of morphologic changes induced by NGF in the U251 glioblastoma
multiforme cell line.

 

Photomicrographs were taken at phase 2. (Bar = 5 um)
Day 2 Post-NGF treatment:

A: HL—l control: Control cultures consist of a population
of very large, flat cells that exhibit an irregular to
spindle-shaped morphology.

B: 4.0 ug/ml NGF: Cells in the 4.0 ug/ml NGF-treated
cultures exhibit a significantly more compact, three-
dimensional morphology than cells in the untreated control
cultures.

Day 6 Post-NGF treatment:

C: HL—l control: Cells in the control cultures are very
flattened, angular and exhibit 0—1 processes per cell.

D: 4.0 ug/ml NGF:. Cells in the 4.0 ug/ml NGF-treated
cultures are much more compact, three-dimensional and tend
exhibit a spindle-shaped to stellate morphology.

86

 

Photomicrographs were taken at phase 1. (Bar = 10 um)
Day 9 Post-NGF treatment:

1.9E: HL-l control: Cells in the control cultures remain
flat and angular and exhibit few processes.

1.9F: 4.0 ug/ml NGF: Cells in the 4.0 ug/ml NGF-treated
cultures are more compact, three-dimensional and are beginning
to develop numerous processes.

Day 11 Post-NGF treatment:

1.9G: HL—l control: On day 11 post-NGF treatment, cells
in the control cultures continue to exhibit a flat, angular
morphology and exhibit few processes.

1.9H: 4.0 ug/ml NGF: 4.0 ug/ml NGF-treated cultures
consist of a population of cells that have very compact,
three-dimensional cell bodies and have developed numerous
long, slender, often branching processes that tend to form an
interconnecting network with neighboring cells.

87

Dimsisn

Descriptions of the NGF-induced. morphologic changes
contain a recurring theme. In response to NGF, each of the
tumor cell lines become more compact and three-dimensional,
and in four of the five cell lines NGF stimulates cells to
produce well developed processes. These types of changes are
significant because the induction of a more compact, three—
dimensional morphology and the production of well developed
processes are suggestive of the development of a more
differentiated phenotype. One of the goals of these in_yitzg
experiments was to demonstrate that NGF was capable of
inducing the development of a more differentiated phenotype.
Information presented in the preceding paper and in this
appendix provides ample evidence that NGF is able to induce
the five neurogenic tumor cell lines under study to develop
morphologic characteristics of a more differentiated
phenotype. These NGF-induced.morphologic changes are readily
visualized. Some of the information provided in these
photomicrographs is also amenable to image analysis. The
succeeding section on image analysis provides additional
support for the ability of NGF to induce dramatic changes in
cell morphology that are consistent with the development of
a more differentiated phenotype.

88
m s s

The preceding article states that the neurogenic tumor
cell lines evaluated in these studies typically respond to NGF
by becoming more compact, three-dimensional, and by producing
well developed processes. These types of changes are
indicative of the development of a more differentiated
phenotype. Though these changes are readily visualized, they
are difficult to quantify. In particular, precise measurement
of three-dimensional changes are difficult to obtain.
However, measurement of alterations in maximal cross-sectional
area, which accompany changes in three-dimensional morphology,
and changes in the number of well developed processes are
readily obtained. These two parameters were measured in order

to collect quantitative information on NGF-induced changes in

morphology.

W

The Hs294 malignant melanoma, T3671 medulloblastoma and
0251 glioblastoma multiforme cultures were evaluated in this
study. Phase contrast photomicrographs of cultures described
in the preceding paper under "Horphologic Studies" were
analyzed with the Sigma Scan (Jandel Scientific) program. The
dose of NGF and length of NGF exposure of the analyzed
cultures are described in the Results section. Analysis was
performed on Phase 2 photographs of the 0251 and TE671 cell

lines and on Phase 1 photographs of the H8294 cell line. The

89
criteria for counting a process varied with the cell type.
0251 and TE671 cells developed long, slender processes in
response to NGF. For these two cell lines, the process had
to be at least twice the length of the average cell body in
order to be counted. In contrast to the previous two cell
lines, the cell bodies of the Hs294 cell line were larger and
the processes considerably shorter. In order for a process
to be counted in this cell line, the process had to be at

least 1/2 the length of the average cell body.

8min

Image analysis was not.performed.on the D54 and 0118 cell
lines because they readily became detached in response to NGF.
Image analysis of the 0251, TE671 and H8294 cell lines was
performed. Figure 1.10 illustrates the dramatic effect that
increasing doses of NGF had on the maximal cross-sectional
area of these cells. Increasing concentrations of NGF
resulted in a dramatic decrease in cross-sectional area. At
4 ug/ml NGF, the maximum concentration of NGF utilized in
these studies, the average cross-sectional area of these
neoplastic cell lines was only 30.3 - 34.1% that of the
untreated control cultures. The 0251 cell line exhibited a
dramatic decrease in cross-sectional area on day 6 post-NGF
administration, the measurement period in which image analysis
was performed. However, morphologic changes in these cells

continued to develop after this time. The cells became even

90

Figure 1.10: The effect of NGF treatment on maximal cross-
sectional area of the TE671, 0251 and H5294 cell lines.

 

 

 

 

 

 

 

 

 

 

 

120
100'
A
d!!—
22
.<43 80
C
0°
00
JEu. so
LO
:3
w:-
PO 40
Po
0&-
“2.
V 20
lLLLILLlillLlllIlLlLLLllLIlLlLLllLLL'll
35.5 1.0 2.0 4.0

Concentration of NGF (ug/ml)

Maximal cross-sectional area measurements of the TE671
(medulloblastoma), 0251 (glioblastoma multiforme), and Hs294
(malignant melanoma) cell lines were generated from
photomicrographs taken of cultures on days 3, 6 and 13
respectively, post-NGF treatment» Measurements were taken at
five different NGF concentrations (0.25, 0.5, 1.0, 2.0, and
4.0 ug/ml NGF) for each cell line. Increasing doses of NGF

result in.a corresponding decrease in maximal cross-sectional
area.

(.) H5294
(+) TE671
(o) 0251

91

more compact, three-dimensional and developed numerous long,
slender often. branching' process that tended to form. an
interconnecting network *with neighboring cells. Though
analysis of 0251 cells from these later time periods would
have demonstrated an even more dramatic morphologic change,
these measurements were not undertaken because of difficulties
inherent in the analysis of these cell outlines.

Increasing concentrations of NGF not only caused cross-
sectional area to decrease, it also resulted in an increase
in the number of cells exhibiting well developed processes and
the number of processes per cell (Table 1.3) This is
illustrated by comparing the untreated control cultures with
cells in.the 4.0 ug/ml NGF treated groups. In untreated.HsZ94
cultures only 35% of the cells exhibited well developed
processes with an average of 1.4 processes per cell. In
contrast, in the 4.0 ug/ml NGF treated cultures 87.6% of the
cells exhibited well developed processes with an average of
4.5 processes/cell. Similar results were obtained with the

TE671 and 0251 cell lines.

92

 

Table 1.3: The effect of NGF treatment on the number of
processes per cell and the percentage of cells exhibiting
well developed processes.

 

 

Cell Days Dose NGF Processes Percentage
Line Post-NGF (ug/ml) per Cell of Cells
Treatment (i S.E.) with

Processes

TE671 3 0.24 i 0.10 20.0
0.76 i 0.19 48.0

1.04 i 0.18 68.0

1.04 i 0.21 60.0

1.28 i 0.17 80.0

1.44 1 0.20 84.0

H8294 6 1.4 i 0.38 35.0
1.9 i 0.36 53.7

2.4 i 0.48 56.9

2.8 i 0.41 68.6

3.2 i 0.43 74.3

4.5 i 0.78 87.6

0251 14 0.9 i 0.24 61.9
9.3 i 0.57 100.0

 

93
sc 3 o

The induction of a more compact, three-dimensional
morphology and the production of an increased number of well
developed processes is consistent with the evolution of a more
differentiated phenotype. These changes are readily
visualized and.were illustrated both in the preceding article
and in the section on morphology in this appendix. Some of
the parameters associated.with these morphologic changes were
amenable to quantification. Image analysis of these same cell
lines provided quantitative information on.parameters related
to the development of a more differentiated state. The
development of a more compact, three-dimensional morphology
is accompanied by a decrease in maximal cross-sectional area.
Figure 1.10 illustrates the ability of NGF to induce a
dramatic decrease in cross-sectional area of these neoplastic
cell lines. 4.0 ug/ml NGF decreased the cross-sectional area
of these neoplastic cells by approximately 67%. Increasing
doses of NGF also induce a dramatic increase in the number of
cells bearing processes and the number of processes per cell.
Treatment with 4.0 ug/ml NGF resulted in a 3.2 - 10.3 fold
increase in the number of processes per cell and a 1.6 - 4.2
fold increase in the number of cells exhibiting well developed
processes. An increased number of well developed processes
is another manifestation of the development of a more

differentiated phenotype.

Results from image analysis support conclusions drawn

94
from morphologic studies and provide quantitative information
demonstrating the ability of NGF to induce morphologic changes
indicative of the development of a more differentiated

phenotype in these neoplastic cell lines of neurogenic origin.

In,- : _- to G tettau‘! 01GB; .- .= 1 1- 5
iob st u 'f e c n

The binding of a ligand to its receptor may result in
changes in the level of expression (either up- or down-
regulation) of the corresponding receptor (Doherty et al.
1988) or associated receptors (Lazarovici et al. 1987). It
has been demonstrated in PC12 cells that NGF causes an up-
regulation in the number of NGFR present on these cells
(Doherty et al. 1988) and a down-regulation of EGFR in these
same cells (Lazarovici et al. 1987). Up-regulation of
receptors is a mechanism whereby the cells may become more
responsive to the ligand. If up-regulation of NGFR increased
the responsiveness of neoplastic cells to NGF, there is the
potential to design a treatment strategy to exploit this
factor and elicit. a more dramatic reverse transforming
response to NGF. The ELISA test, used previously to measure
receptor density, has the capability to provide information
as to the effect of NGF treatment on NGFR density in our cell
lines. This method was utilized to determine the effect of
NGF on expression of NGFR in the 0251 glioblastoma multiforme
cell line.

95

W

Cell counts versus protein concentration was determined
from untreated and NGF-treated 0251 cultures. The ELISA
procedure was run in 96-well microtiter plates (See the
preceding paper for details on the ELISA procedure). 0251
cells were seeded at a density of 12,500 cells/well (0.2 mls
at 6.25 x 10‘ cells/ml). The following day the medium was
exchanged with HL-l. On the third day the following groups
were established: HL-l control, 1.0 ug/ml NGF, 2.0 ug/ml NGF,
4.0 ug/ml NGF. Following two days' exposure to NGF, cells
from two of the eight seeded wells per group were dissolved
in 1.0 N NaOH for 30 min at 37°C. The protein content was
determined using the Coomassie blue technique (Pierce) to
obtain data on the cell density in these wells. The remaining
cells were fixed in 4% paraformaldehyde and receptor density
determined using the ELISA technique described in the

preceding paper.

325111.128

Results from this experiment indicate that NGF treatment
causes neither up- or down-regulation of NGFR in the 0251 cell
line. Following exposure to NGF for two days, significant
differences between the various treated and untreated groups
were not apparent (Table 1.4).

Though measurement of the effect of NGF treatment on

tumor growth rate was not a primary objective of this

96

 

Table 1.4: The effect of NGF treatment on the expression
of NGFR on the 0251 glioblastoma multiforme cell line.

 

Concentration Density of NGFR i S.E.

of NGF (0.1). units/2.5 x 10" cells)
0.0 ug/ml 0.201 1 0.22

1.0 ug/ml 0.195 1 0.17

2.0 ug/ml 0.184 i 0.26

4.0 ug/ml 0.192 1 0.14

 

97
experiment, results from this study support previous data
indicating that NGF has the ability to decrease the growth
rate of neoplastic cell lines in culture. When the various
treatment groups were compared using the untreated control as
the standard, it was evident that NGF treatment resulted in
an incremental decrease in tumor growth rate. Following two
days' exposure to NGF, the 1.0 ug/ml group contained 15.2%
fewer cells, the 2.0 ug/ml 18.6% fewer and the 4.0 ug/ml group

20.1% fewer cells than untreated control wells.

nimssien

The 0251 cell line develops a number of interesting
changes in response to NGF. NGF induces cells from this
glioblastoma multiforme cell line to develop a more
differentiated phenotype and to stop dividing. These effects
appear to be highly persistent. In addition, the 0251 cell
line grows readily in irradiated nude mice. These
characteristics make it a logical choice for further
investigations to assess the effects of NGF on
neurogenic tumor cell lines.

It has been reported that NGF treatment induces the up-
regulation of NGFR in PC12 cells (Doherty et al. 1988).
Results from this experiment indicate that NGFR are neither
up- or down-regulated in the 0251 cell line following two
days' exposure to NGF. In fact, a significant difference was

not identified between the control and any of the NGF-treated

98

groups. NGFR up-regulation does not appear to be an aspect
that can be exploited in the development of a treatment
strategy for the 0251 glioblastoma multiforme cell line.

0251 cells respond well to NGF therapy. This is readily
demonstrated by the morphologic studies and studies evaluating
growth rate. Though this experiment was not designed to
illustrate the effect of NGF on growth rate, measurements
taken during the course of the study once again demonstrate
that NGF is capable of diminishing the growth rate of

neoplastic cell lines of neurogenic origin.

9‘- =1°l1°. _ -‘.1°°=-,‘ 9’1- ‘0. 0 1911C: ’-. Liv. {12”

!.....p .. g-;..,;- , ,- ,3 -. ,. 7..., ,-A.,.". - 1e

During the course of the in_yit;9 evaluation of these
five human neurogenic tumor cell lines for response to NGF,
it was observed that neoplastic cells exposed to NGF would
continue to develop progressive morphologic changes even after
the NGF-containing medium had been removed from the culture.
This was observed during routine morphologic studies and in
experiments evaluating the persistent nature of the NGF-
induced changes. The progressive development of morphologic
changes following withdrawal of NGF was particularly apparent
in the 0251 cell line. In studies to evaluate the length of
time that NGF-induced morphologic changes would persist in its
absence, cultures were exposed to NGF for two days and then

the NGF-containing medium was removed. Though 0251 cultures

99
were exposed to NGF for only 48 hours, cells continued to
develop morphologic changes for an additional 10 days
following removal of the growth factor from the culture medium
and these changes were identical to those exhibited by cells
continuously exposed to NGF.

The purpose of this experiment was to determine the
length of NGF exposure required to induce a maximal response
in one of the neoplastic cell lines under study. The Hs294
cell line was chosen because it developed prominent
morphologic changes in response to NGF, it developed these
changes relatively quickly and the cells did not tend to

become detached from the culture surface in response to NGF.

MAW

General culture procedures were as described in the
"Materials and Methods" section of the preceding paper. The
H3294 cell line was utilized for these experiments. 12-well
cluster dishes were seeded with 1.0 x lo‘cells (1.0 ml/well).
The following day the medium was exchanged with HL-l serum
free medium. On the third day the medium was exchanged with
HL-l containing 2.0 ug/ml NGF in all wells except the HL-l
control well. The NGF containing medium was removed and
replaced with HL-l after: 1, 2, 4, 8, 12, 16, 24, 33, and 48
hours. Two additional groups were also present. A group in
which the NGF-containing medium was left in the cultures

throughout the entire four days in which morphologic changes

100
were evaluated and a group in which the NGF-containing medium
was replaced daily. Photographs of each group were taken
daily through day 4 post-NGF treatment. One hundred cells
from each treatment group were evaluated and placed into one
of three morphologic categories: (A) amorphous cells (round
to irregular, 2-dimensional, amorphous cells devoid of
processes) : (B) angular cells (sharply angular cells that were
more refractile and three dimensional than cells in group A);
and (C) stellate cells (small, stellate, three dimensional

cells exhibiting numerous short processes).

39.5.1113:

The Hs294 malignant melanoma cell line develops an
interesting spectrum of changes in response to NGF. Untreated
control cultures are predominated by a very primitive
population consisting of large, flat, amorphous cells. NGF
treatment stimulates many of these cells to develop a more
differentiated phenotype. This transformation is
characterized by the development of a more compact, three-
dimensional morphology and the production of an increased
number of well develop processes. In progressing to this
state, responsive Hs294 cells pass through an intermediate
stage characterized by the development of a more compact,
three—dimensional, sharply angular morphology.

In response to NGF, 90-95% of the cells in Hs294 cultures

develop a more differentiated morphology. 40-50% of the cells

101

attain the most differentiated state characterized by the
development of a stellate morphology. Table 1.5 indicates
that, from the 48-hour time period on, the spectrum of
morphologic changes. were essentially indistinguishable for
those cultures exposed to NGF for 16 hours or more. Cultures
exposed to NGF for 16 hours exhibited the same percentage of
cells in the least differentiated category (amorphous cells)
indicating that approximately the same proportion of cells
were stimulated to develop a more differentiated phenotype.
H3294 cultures grown in the presence of NGF for 16 hours also
exhibited the same number of cells in the most highly
differentiated category, the stellate cells. Thus, the data
indicates that at a dose of 2.0 ug/ml NGF a maximal response
can be attained following 16 hours exposure to NGF. Indeed,
an argument could be made that 8-12 hours exposure to NGF is
sufficient to induce a maximal morphologic response as these
cultures were generally within 5% of the NGF controls in each
category.

Though full expression of the morphologic changes took
slightly longer at. the lower’ exposure times, a maximal
response was attained after limited exposure. Exposure for
longer periods of time did not stimulate a higher percentage
of cells to develop a more differentiated phenotype or

additional morphologic changes.

-]a(48]1234h‘ H1124811234Lfl.\ I i

| Cb ...! a -

102

 

Table 1.5: Length of NGF exposure required to induce a
maximal morphologic response in the H8294 malignant
melanoma cell line.

 

NGF
Exposure Day 1: Day 2:
°Amorph. ‘Ang. 'Stell . Amorph. Ang . Stell
HL-l 75% 20% 5% 75% 10% 15%
'NGF 50% 30% 20% 5% 40% 55%
1 hr. 60% 30% 10% 75% 10% 15%
2 hr. 60% 30% 10% 60% 25% 15%
4 hr. 40% 30% 30% 20% 40% 40%
8 hr. 40% 30% 30% 15% 40% 45%
12 hr. 40% 30% 30% 15% 40% 45%
16 hr. 40% 30% 30% 10% 40% 50%
24 hr. 20% 45% 35% 10% 45% 45%
38 hr. 20% 45% 35% 10% 40% 50%
48 hr. 20% 45% 35% 10% 45% 45%
”NGF 20% 45% 35% 10% 55% 35%
Day 3: Day 4:
HL-l 55% 40% 5% 45% 45% 10%
'NGF 2% 49% 49% 2% 49% 49%
1 hr. 40% 40% 20% 40% 40% 20%
2 hr. 40% 30% 30% 33% 33% 33%
4 hr. 25% 55% 20% 20% 45% 35%
8 hr. 15% 45% 40% 10% 45% 45%
12 hr. 15% 35% 50% '10% 45% 45%
16 hr. 10% 50% 40% 5% 55% 40%
24 hr. 10% 45% 45% 5% 55% 40%
38 hr. 10% 45% 45% 5% 55% 40%
48 hr. 8% 45% 27% 5% 55% 40%
bNGF 5% 50% 45% 5% 55% 40%

 

2. 0 ug/ml NGF, medium was not changed during the
course of the experiment.

:NGF-containing medium was replaced daily.

:Amorphous cells (least differentiated morphology)

:Angular cells (intermediate stage of differentiation)

‘Stellate cells (most highly differentiated morphology)

 

103
Discussisn

Results from this experiment confirm previous
observations which had indicated that continuous exposure to
NGF is not necessary for neoplastic cell lines to develop a
maximal morphologic response to the growth factor. Indeed,
at 2 ug/ml NGF, H3294 cells required.only'8 hours NGF exposure
to develop and.maintain.morphologic changes nearly equivalent
to NGF-control cultures and only 16 hours to develop changes
identical to cultures that had been continuously exposed to
the growth factor for four daysw This type of response is not
unique to the H3294 cell line. Both the T9 (Marushige et al.
1987), 054 and 0251 cell lines respond in a similar manner,
i.e., limited exposure to NGF (48 hours) results in the
development of a maximal morphologic response.

Information such as this is vital to understanding the
response of neurogenic tumor cell lines to NGF. It is only
after data is obtained on the minimum dose required for
maximal response, the minimal dose required to elicit a
response, and the length of exposure needed to obtain a
maximal response, that experiments can be designed to test

specific aspects of the response of these cells in_yit;g.

104
o ' i
Information presented in this appendix expands on some
of the statements made in the preceding paper and describes
studies that provide additional insight into the response of
these five human tumor cell lines to NGF in_yit;g. .In;gitrg
studies demonstrate the ability of NGF to reverse some of the
transformed properties of neurogenic tumors cell lines. The
succeeding chapter will focus on the effect of NGF on the

growth rate of many of these same cell lines in vigg.

CHAPTER 2

THE USE OF NERVE GROWTH FACTOR AS A REVERSE TRANSFORMING AGENT

FOR THE TREATMENT OF NEUROGENIC TUMORS: IN VIVO RESULTS

CHAPTER 2

The Use of Nerve Growth Factor as a Reverse Transforming Agent

for the Treatment of Neurogenic Tumors: 10.1119 Results.

511.551.3232

The rationale behind the evaluation of natural differentiating
agents, such as nerve growth factor (NGF) , for reverse
transforming potential is based on the theory that such
compounds may represent a nontoxic means of controlling tumor
growth. Previous in vitro experiments have shown that NGF is
capable of retarding growth and of inducing persistent
differentiation of neurogenic tumor cell lines. 111.1119. NGF
is capable of causing a persistent reduction in the number of
ethylnitrosourea-induced neurinomas. In this study, the
reverse transforming potential of NGF was evaluated by
implanting anaplastic glioma and neurinoma cell lines into
Fischer rats and selected human glial tumor cell lines into
athymic nude mice. Results indicate that NGF is capable of
causing a significant decrease in the growth rate of
subcutaneous T9 (anaplastic glioma) and clone 16 (anaplastic
neurinoma) implants in syngeneic rats. Significantly, NGF
treatment was accompanied by adverse effects that were minimal

105

Cu

106
and transient. NGF treatment of human glial tumor cell lines
implanted into athymic nude mice was less effective.
Continued tumor growth (although greatly retarded) following
NGF treatment and the limited effectiveness of NGF when using
the nude mouse model are aspects that require further
investigation. These experiments demonstrated that NGF was
able to significantly restrain the growth of two neurogenic
tumor cell lines ill—£119- Results suggest that NGF may
eventually prove useful as part of a therapeutic regime for

the treatment of tumors of neurogenic origin.

mm
Evidence that nerve growth factor (NGF) is capable of

reversing some of the transformed properties of neoplastic
cells of neurogenic origin continues to accumulate.
Endorsement of this concept is all the more convincing because
it is supported by a variety of experimental approaches. The
initial indication that NGF might possess reverse transforming
capabilities comes from an understanding of the physiologic
effects of NGF and the nature of reverse transforming agents.
Compounds capable of inducing reverse transformation are
generally differentiating agents. As NGF functions as a'
differentiating agent for neural crest-derived cells during
development (Ievi-Montalcini and Aloe 1985), evaluation of
this compound as a reverse transforming agent for anaplastic

tumors of the nervous system is warranted.

of
at
51:

C3

ar.
Pe
Pe
Ma
19
tr
di

01'

Me
4]
Ya

is

107

Support for this concept has been generated from a number
of in_yit;g studies. These experiments indicated that NGF is
able to reverse many of the transformed properties of
susceptible tumor cells. Data demonstrates that: 1) NGF is
capable of reversing several of the transformed.properties of
a variety of neurogenic tumor cell lines, including
pheochromocytoma, g1 ioma, medulloblastoma, malignant melanoma
and neuroblastoma cell lines (Waris et al. 1973: Reynolds and
Perez-Polo, 1975: Greene and Tischler, 1976: Reynolds and
Perez-Polo, 1981: Jensen 1987: Marushige et al. 1987:
Marushige et al. 1989: Matsushima et al. 1990: Knmar et al.
1990: Vinores and Koestner, 1981: Yaeger et al. 1991): 2) NGF
treatment consistently results in the development of a more
differentiated phenotype in responsive cell lines (Reynolds
and Perez-Polo, 1975: Greene and Tischler, 1976: Reynolds and
Perez-Polo, 1981: Vinores and Koestner, 1982: Jensen 1987:
Marushige et al. 1987: Marushige et al. 1989; Matsushima et
al. 1990: Kumar et al. 1990: Vinores and Koestner, 1981:
Yaeger et al. 1991): 3) a common consequence of NGF treatment
is a reduction in growth rate or complete cessation of growth
(Reynolds and Perez-Polo, 1975: Greene and Tischler, 1976:
Vinores and Koestner, 1981: Jensen 1987: Kumar et al. 1990:
Matsushima and Bogenmann, 1990: Yaeger et al. 1991): and 4)
these effects are persistent in a number of tumor cell lines
of neurogenic origin (Reynolds and perez-Polo, 1975: Reynolds

and Perez-Polo, 1981: Marushige et al. 1987: Yaeger et al.

108
1991).

In_giyg_studies further support the reverse transforming
ability of NGF. Treatment of rats with NGF either
transplacentally before ethylnitrosourea (ENU) exposure or
postnatally after ENU exposure caused a reduction in the
number of ENU-induced neurinomas (Vinores and Koestner, 1982:
Raju et al. 1989). Data from a one-year study indicates that
NGF treatment does not simply delay the onset of tumor
formation, but results in a persistent suppression of tumor
development (Raju et al. 1989) . Another line of evidence
demonstrating that NGF is effective 10.1129 comes from studies
involving intracerebral implantation of F98 anaplastic glioma
cells into syngeneic Fischer rats. When F98 cells were
pretreated with NGF, treated with NGF following implantation
or implanted along with NGF-producing C6 cells, tumor growth
rate was diminished, survival time increased and fewer animals
developed tumors (Vinores and Koestner, 1980: Vinores and
Koestner, 1981) . On histologic examination NGF treated tumors
had developed a more differentiated morphology (Vinores and
Koestner, 1981).

These results offer convincing evidence that NGF is
capable of reversing some of the transformed properties of
neoplastic cells both 1% and 1131119. We have evaluated
a number of tumor cell lines in_gitzg and have found that a
diverse collection of neurogenic tumor cell lines exhibit

reverse transformation in response to NGF (Yaeger et al.

19

SL'

ir

IE

1i

hu

mu

99

1i

tr

Ma

of

Ba

CO.

Si]

.
ATL
E

109
1991). 1n_yit;g results and initial in_yiyg data strongly
support the need for further investigation in this area,
particularly in yivg. The aim of this study was to evaluate
the response of several neurogenic tumor cells lines,
including cell lines of both human and rodent origin, for

response to NGF in_yigg.

MW

§£QQK_QEIEB£§§

Four cell lines were evaluated for response to NGF in
£129: two rodent tumor cell lines (the T9 anaplastic glioma
line and clone 16, an anaplastic neurinoma clone) and two
human cell lines (the 054 mixed glioma and 0251 glioblastoma
multiforme cell lines). The two human tumor cell lines were
generous gifts from Dr. Bigner at Duke University. Each cell
line had previously been shown to exhibit a reverse
transforming response to NGF 13411trg (Marushige et a1. 1987:
Marushige et al. 1989: Yaeger et al. 1991). Stock cultures
of the human cell lines were maintained in Dulbecco's modified
Eagle's medium (Gibco) containing 20% fetal bovine serum
(Sigma), and 4 mM glutamine in a humidified chamber with 8%
CO2 at 37 °C. The rodent cell lines were maintained under

similar conditions except in DMEM containing 10% FBS.

WW

Cultures were seeded in 12-well cluster plates (Corning)

Va

ho

Vi

Ho

110
at a density of 3.0 x 10’ to 2.5 x 10‘ cells/well (1.0
ml/well). Cells were seeded in DMEM + 20% FBS. After 24
hours, the medium was replaced with either HL-l serum-free
medium (Ventrex) or HL-l containing 10.0%, 1.0% or 0.1% FBS.
The following day the medium was again exchanged with either
the serum containing medium or serum containing medium pre-
incubated with 2 ug/ml 2.58 NGF (Boehringer Mannheim) for one
hour. Cultures were observed daily via phase—contrast
microscopy for evidence of NGF-induced morphologic changes.

This experiment was duplicated with alpha-2-macroglobul in
replacing the FBS. Concentrations of alpha-2-macroglobulin
(bovine, Boehringer Mannheim) equivalent to that found in
10.0%, 1.0% and 0.1% serum were utilized. The corresponding
concentrations of alpha-2-macroglobulin were 0.9 mg/ml, 0.09
mg/ml, and 0.009 mg/ml respectively.

In addition, an 1114139 bioassay was employed to
identify whether there was a difference in the ability of
mouse versus rat serum (Sigma) to inhibit NGF-induced
morphologic changes. A similar protocol to the one outlined
above was used. Various concentrations of each serum (0.75,
0.5, 0.25, 0.1 and 0.075%) were added to HL-l. 2.0 ug/ml NGF
was pre-incubated with the serum containing medium for one
lhour before the final medium change. Corresponding controls,
without NGF, were employed at each serum concentration.

Cultures were observed daily for NGF-induced changes in

morphology.

111
ts v e n e e

Cultures of T9 cells one day from confluency were
trypsinized, washed, and resuspended in DMEM at 8 x 10‘
cells/ml. A Hamilton syringe with 22-gauge needle was used
to implant 25 ul (2 x 10‘ cells) of this cell suspension into
the subcutaneous tissue of the left flank of 90-110g Fischer
344 rats (Charles River). Tumors were allowed 10 days to
become established before treatment was initiated. The
various NGF (2.58 Boehringer Mannheim) treatment regimes
employed are described in the Results section. Animals were
weighed once'a week. Tumors were palpated and measured twice
weekly with vernier calipers. Animals were euthanized with
CO2 once the tumors had reached a maximum of 1.0-1.5 cm in any
one dimension. Samples of each tumor were fresh frozen, fixed
in 4% paraformaldehyde for'16 hours and.then frozen, and fixed
in 10% neutral buffered formalin. Routine histologic sections
were examined from each tumor.

Procedures involving clone 16 implants were performed in
a similar manner. The implant dose was decreased slightly
(3.5 x 10° cells/implant) for these experiments and the tumors

were generally allowed 7 days to become established before

treatment was initiated.

 

The implantation procedure was modified slightly for the

human tumor lines. Four-week-old athymic nude mice (Harlan

112

Sprague Dawley) were irradiated (gamma irradiator) with 322R
on the day of implantation. Cultures of 0251 cells one day
from confluency were trypsinized, washed and resuspended in
DMEM. 0.25 mls (7.9 x 10' cells) of this cell suspension was
implanted into the subcutaneous tissue of the dorsal midline
between the shoulder blades. Tumors were allowed 7 days to
become established before treatment was begun.

Tumors from. 054 implants developed. slowly and
inconsistently. As such, the 054 cell line was passaged three
times in irradiated athymic nude mice to obtain an animal-
adapted population of cells. These cells, when re-established
in culture, continued to exhibit a reverse transforming
response to NGF. Following the third passage, tumors were
aseptically removed from 4 animals, minced, passed through 40-

and loo-mesh cytosieves, washed and then a standard volume
(0.25 mls) was implanted into each experimental animal.

Tumors were allowed 7 days to become established before

treatment was begun.

W

NGF receptor (NGFR) density was determined for each cell
line using a modification of the ELISA.procedure described by
Doherty et. a1. (1988). Cells were seeded at 15,000
cells/well in a 96-well microtiter plate. Following two days
in culture, the cells were fixed with paraformaldehyde.

Receptor density was determined using ME20.4, a human anti-

113
NGF-receptor antibody ( 1:20, Amersham) or Mab 192, a rat anti-
NGF-receptor antibody (1:200, Oncogene Science) and an anti-
mouse horseradish peroxidase conjugate (1:1000, Sigma).
Optical density (00) was measured with the Dynatech MR600
microtiter plate reader. A standard curve was developed using
two cell lines with known NGF-binding properties, the human
H3294 malignant melanoma cell line (Fabricant et al. 1977) and
the rodent PC12 pheochromocytoma cell line (Sonnenfeld and
Ishii, 1985: Azar et al. 1990) . 00 was determined at various
seeding densities for each standard. Protein concentrations
were determined from which cell numbers were calculated. A
standard curve was generated that compared 00 to H3294 or PC12
cell numbers. Unknowns were then compared to the standard

curve .

355111.125

As a preliminary screen, the NGFR density of each tumor
cell line was estimated using an ELISA. Results indicated
that all four tumor lines used in the 111119 experiments (T9,
clone 16, 054 and 0251) had very few NGFR compared to the two
control cell lines (H3294 and PC12) . Cell lines were compared
on an arbitrary scale (1.0 arbitrary unit corresponds to 0.165
0.0. units per 2.5 x 10‘ cells.). The positive control for
the human tumor lines, the H3294 malignant melanoma line,
exhibited a receptor density (106.7 + 12.36) that was nearly

two orders of magnitude higher than either the 054 (1.7 i

114

0.34) or 0251 (1.4 i 0.17) cell lines. The PC12 cell line
(85.9 3 10.01), the positive control for the rodent tumor cell
lines, also exhibited a receptor density that was
approximately two orders of magnitude higher than either clone
16 (0.27 i 0.13) or the T9 (0.91 1 0.33) cell line. Despite
low receptor numbers, each cell line responded well to NGF in
yitrg (Marushige et al. 1987: Marushige et al. 1989: Yaeger
et al. 1991), albeit at high concentrations of NGF.

Initial experiments, in which NGF was administered
subcutaneously at a site distant from the T9 anaplastic glioma
implants, demonstrated an encouraging trend. Nerve growth
factor treatment resulted in a statistically significant (P
< 0.05) decrease in the average volume of subcutaneous tumor
implants in the high dose NGF-treatment group (40 ug NGF SID
for 5 days) compared to saline-treated controls. At the time
of necropsy the average tumor volume in the high dose NGF-
treated group (12.62 i 2.02 mm’/day) was less than half that
of the saline-treated controls (27.72 1; 4.52 mma/day). The
average tumor volume in the low dose NGF-treated group (10 ug
SID for 5 days) was decreased slightly (24.14 3; 6.30 mma/day)
compared to the controls, but this difference was not
statistically significant.

Following these favorable preliminary results, an attempt
was made to develop a more effective treatment protocol.

Various doses and dosage schedules were evaluated. Of these

regimes, 35 ug NGF administered directly at the implantation

115
site once a day on 6 consecutive days (210 ug total dose)
yielded the most favorable results.

A full scale experiment was performed using this
treatment protocol. Results are shown in figure 2.1.
Treatment with this schedule resulted in a statistically
significant (P < 0.05) decrease in the. growth rate of
subcutaneous T9 implants in the NGF-treated group compared to
the untreated controls. A significant difference was
identified at each measurement period once tumors could be
reliably palpated. .At the time of necropsy, the average
tumor volume in the.NGF-treated.group was only 37% that of the
untreated controls.

Similar experiments were conducted with clone 16. Data
from clone 16 experiments demonstrated that NGF (40ug once-
a-day (SID) for'6 days) was even more effective at diminishing
the growth rate of subcutaneous neurinoma implants (figure
2.2). As with the T9 implants, a significant difference was
identified at each measurement period once tumors could be
reliably palpated. At the time of necropsy, the average
tumor volume in the NGF-treated group was only 27% that of the
untreated controls.

‘With. both. T9 and clone 16 implants, NGF treatment
resulted in a significant decrease in tumor growth rate.
However, tumors in the NGF-treated groups continued to grow
despite therapy. These findings prompted an additional

experiment. A second series of NGF injections (40 ug NGF SID,

116

Figure 2.1: The effect of NGF treatment on the growth rate
of subcutaneous T9 anaplastic glioma implants.

 

 

 

 

 

 

     

 

 

 

Thousands
10
A
"a
v 8 1
Q)
B
.3
o 6
>.
H
g T
5 4
E4
&
g 2
3
‘¢
0 1 11.1.1(1.11.11.11.11.
19 29 33 36 4O 43

Days Post-Tumor Implantation

 

35 ug NGF was administered directly at the implantation site
SID on days 10-15 post-implantation (210 ug total dose).
Results indicate that the average tumor volume in the NGF-
treated group was decreased significantly compared to saline-
treated controls from day 33 on. Each group contained ten
experimental subjects.

(.) NGF-treated group

(.) saline-treated controls

117

Figure 2.2: The effect of NGF treatment on the growth rate
of subcutaneous clone 16 anaplastic neurinoma implants.

 

1200

 

150

 

 

100

 

Average Tumor Volume (m3)
0:
o
.1

    

 

 

..I.J.l.l.'-l.l.l.l.l.
1 13 18 21 25 28 32
Days Post-Tumor Implantation

'1

 

 

 

 

40 ug NGF was administered directly at the implantation site
SID on days 10-15 post implantation (240 ug total dose).
Results indicate that the average tumor volume in the NGF-
treated group was decreased significantly compared to saline-

treated controls from day 18 on. Each group contained seven
experimental subjects.

(0) NGF-treated group
(.) saline-treated controls

118
days 8-12 & 22-26) was administered to determine if repeated
series of treatments would result in an additional decrease
in tumor growth rate. This second series of NGF-treatments
had no additional effect on tumor growth (Figure 2.3), even
though the tumors were still responsive at the time of the
second injection period (data not shown).

Experiments to determine the effect of NGF treatment on
two human tumor lines, ' v'vo, were also undertaken. In
contrast to the rodent tumor cell lines, only a minimal
response to NGF (35 ug NGF SID for 6 days) was observed
following treatment of these tumors in nude mice. In each
case, NGF treatment resulted in a slight decrease in growth
rate (data not shown). However, the decrease was not nearly
as dramatic as with the rodent tumor lines and was not
statistically significant.

Reports indicate that maturation factors are capable of
inducing a phenotypic change 111—11513 (Waris et al. 1973;
Reynolds and Perez-Polo, 1975: Greene and Tischler, 1976:
Reynolds and Perez-Polo, 1981: Jensen 1987: Marushige et al.
1987: Marushige et al. 1989: Matsushima et al. 1990: Kumar et
al. 1990: Vinores and Koestner, 1981: Yaeger et al. 1991) as
well as in_yiyg (Vinores and Koestner, 1981). All tumors in
this study were examined histologically. Differences in
morphology were not identified between NGF-treated and

untreated tumors.

NGF-treated animals exhibited.a transient decreased rate

119

Figure 2.3: The effect of'a repeated series of NGF treatments
on the growth rate of clone 16 implants.

 

 

 

 

 

 

 

 

 

 

 

120
"g 100 ~-
3 ....W... -..
a 80
....
:3
:5 80
5
e: 40
0
53"
E 20
1<

 

 

1 13 17 20 24 27 31 34
Days Post-Tumor Implantation

Results demonstrate that the average tumor volume in both NGF-
treated groups was decreased significantly compared to saline-
treated controls. However, the growth rate of the group
receiving a repeated series of NGF treatments was not
significantly different from the group that received a single
series of NGF-injections. Each group contained seven
experimental subjects.

(.) saline-treated control group:

(+) single series of NGF injections (40 ug NGF SID days 8-
12):

(.) repeated series on NGF treatments (40 ug NGF SID days 8-
12 a 22-26).

120

of weight gain compared to saline treated controls. However,
the animals recovered rapidly, and, within one week following
cessation of NGF injections, there were no differences in the
weights between the various groups. This was the only
evidence of adverse consequences of NGF therapy. Histologic
evidence of organ damage was not identified. Outgrowths of
neurons were not apparent at the treatment site and
intracerebral inoculation was without noticeable effect on
cerebral architecture.

The results from several in_xitrg studies are reported
here because they provide an insight into potential causes of
the discrepancy between implantation studies involving rodent
and human tumor cell lines. These data illustrate the effects
of serum on the response of tumor cell lines to NGF in tissue
culture. Tumor cell lines fail to exhibit a morphologic
response to NGF in the presence of 10% and 1.0% FBS but will
respond at concentrations 5 0.1% (Table 2.1). However, even
in the presence of only 0.1% serum, the NGF-induced
morphologic changes were diminished compared to corresponding
controls. When levels of alpha-2-macroglobulin, the major
NGF-binding protein in serum (Ronne et al. 1979), equivalent
to those present at the various serum concentrations were
added to the cultures, similar results were obtained (Table
2.1). In addition, both rat and mouse sera were evaluated for
their ability to prevent NGF-induced morphologic changes.

These experiments indicate that mouse serum and F88 are

121

 

Table 2.1: Comparison of FBS, alpha-2-macroglobulin, mouse
and.rat serum in terms of their ability to inhibit or diminish
NGF-induced morphologic changes in the 054 mixed glioma cell
line.

 

Days 1 2 3 4 5 6 7 8 9 10
Post-NGF

administration

i_£B§

10.0% 0 0 0 0 0 O 0 0 0 0
1.0% 0 0 0 0 0 0 0 0 0 0
0.1% 0 0 0 o + + + + + +

0.9 0 0 0 0 0 0 0 0 0 0

0.09 0 0 0 0 0 0 0 0 0 0

0.009 0 0 O 0 + + + + + +

i_meuss_ssrum

0.75% 0 0 0 0 0 0 0 0 0 0

0.5% 0 0 O 0 0 0 0 0 0 0

0.25% 0 0 0 0 0 0 0 0 0 0

0.1% 0 0 o o + + + + + +

0.075% 0 0 0 + + 2+ 2+ 2+ 2+ 2+
1.231.33193

0.75% 0 0 0 0 0 O 0 0 0 0

0.5% 0 0 0 o + + + + + +

0.25% 0 0 o + + 2+ 2+ 2+ 2+ 2+
0.1% 0 0 + + 2+ 2+ 2+ 3+ 3+ 3+
0.075% 0 + + 2+ 2+ 2+ 3+ 3+ 4+ 4+

 

2.0 ug/ml NGF was pre-incubated with each concentration of
serum or alpha-2-macroglobulin for one hour prior to exchange
with the culture medium. The degree of change was graded from
0 (no difference from untreated controls) to 4+ (majority of
cells exhibiting NGF-induced changes in morphology).

 

122
approximately equivalent in their ability to diminish NGF-
induced changes in morphology. In contrast, rat serum appears
to be only one-fifth as potent in its ability to prevent or

diminish NGF-induced morphologic changes (Table 2.1).

D' .

W experiments utilizing the two cell lines of
rodent origin clearly demonstrate that NGF treatment is
capable of causing a dramatic decrease in the growth rate of
subcutaneous T9 (anaplastic glioma) and clone 16 (anaplastic
neurinoma) implants. At the various measurement periods, the
average tumor volume in the NGF-treated group was only 8-27%
that of saline-treated.controls for clone 16 implants and.only
29-37% that of saline-treated controls for T9 implants. This
confirms the results from in;gitzg_studies which demonstrated
that NGF was able to reverse several of the transformed
properties of these tumor cell lines (Marushige et al. 1987:
Marushige et al. 1989). These results are all the more
encouraging because adverse effects associated with NGF
administration in experimental animals were minimal and
transient.

Based on the promising results obtained with the rodent
tumor lines in vivg and the dramatic response of the human
tumor cell lines to NGF in_yitrg (Yaeger et al. 1991), both
prior to implantation and once re-established from the

implants, it was expected that a similar decrease in tumor

123

growth rate would be observed with the human tumor lines
following NGF treatment 1334119. Implants of the human cell
lines displayed a slight decrease in growth rate in response
to NGF. However, the magnitude of response was not nearly as
great as with the rodent tumor lines and this decrease was not
statistically significant. A complete explanation for the
difference between the two model systems was not established
although several factors were identified that may have
contributed to this disparity. The presence of serum or
alpha-2-macroglobulin in the culture medium is capable of
diminishing or preventing NGF-induced morphologic changes in
11m (Table 2.1) . Additionally, data indicate that mouse
serum is more potent than rat serum in its ability to diminish
the NGF-induced response (Table 2.1). These factors become
significant when combined with the observation that slight
hemorrhage and/or serum transudation was not uncommon
following injections in the athymic nude mice due to the
delicate nature of their subcutaneous tissues. Rats, in which
these experiments were highly successful, did not exhibit such
a tendency. It is plausible that serum present at the time
of NGF-administration inhibited the NGF-induced response in
the athymic nude mice and may have played a key role in the
disparity between the two systems.

The inherent difficulties encountered with the athymic
nude mice were not a factor in studies involving the

implantation of rodent tumor cell lines into syngeneic Fischer

124

rats. These experiments clearly demonstrate the ability of
NGF to reduce thelgrowth rate of subcutaneous tumor implants.

However, the matter of continued tumor growth in NGF-treated
groups requires further analysis before definitive conclusions
can be made as to the potential of this type of therapy. It
does not appear that continued tumor growth is simply the
result of a transient response to NGF. If the effects of NGF
were transient, the second series of NGF-treatments should
have elicited an additional response. The fact that this was
not observed argues against such an interpretation. A
plausible hypothesis, and one that is supported by several
lines of experimental evidence, is that NGF causes persistent
growth suppression in.a NGF-responsive subpopulation of cells
and that an NGF-insensitive subpopulation of cells accounts
for continued growth of the tumors. The therapeutic potential
of NGF may hinge on the basis of these findings. If a
transient suppression of growth is all that can be attained,
then the therapeutic potential of NGF would be rather limited.
However, if NGF is capable of causing persistent growth
suppression in responsive cells, it may prove useful as a
reverse transforming agent.

The rationale behind evaluating natural differentiation
agents for reverse transforming potential is based on the
theory that these compounds may represent a nontoxic means of
controlling tumor growth. Our data indicate that NGF is

capable of causing a dramatic decrease in the growth rate of

125

subcutaneous T9 (anaplastic glioma) and clone 16 (anaplastic
neurinoma) implants. Significantly, this decreased tumor

growth rate is accompanied by adverse effects that are minimal
and transient. Further study is needed to answer questions
concerning the mechanism responsible for this growth
suppression and whether or not these effects are persistent
in vivg. None-the-less, it is clear that NGF is able to
suppress the growth of rapidly dividing neoplastic cell lines
of neurogenic origin. We are optimistic that this compound
may prove useful, alone or in combination.with other types of

therapy, for the treatment of tumors of neurogenic origin.

CHAPTER 2 APPENDIX

IDEIQQBEEIQD

The preceding paper provides information supporting the
ability of NGF to reverse some of the transformed properties
of implanted neurogenic tumor cell lines. As with the in
yitzg studies, the material presented in the article did not
include all of the information obtained from 1n_yiyg studies.
The scope of the paper was limited to information necessary
to demonstrate the reverse transforming capabilities of NGF
1n_yigg and to discuss some of the pitfalls associated with
these types of experiments. Additional in vivo studies were
performed. The results of these experiments are reported
here, in the appendix, because they provide additional
information on techniques involved in tumor implantation
studies and on some of the difficulties encountered in the
course of this line of experimentation.

Two approaches were utilized for the in viyg studies.
The first approach was to implant tumors subcutaneously. The
advantage of this method is that the tumors are readily
palpable and can be measured at regular intervals. This
provides information on tumor growth rate. The disadvantage
of this method is that the treatment is not evaluated in the

126

127
environment in which it will eventually be used. The second
approach was to implant tumors intracerebrally. The advantage
of this method is that treatment is evaluated in the
environment in which the tumors naturally occur. The drawback
to the intracerebral studies is that measurements of tumor
size lbetween the time of implantation and. necropsy are
difficult unless animals are sacrificed periodically
throughout the course of the experiment or expensive methods,
such as Magnetic Resonance Imagery (MRI), are employed.
Initial experiments concentrated on subcutaneous studies
because they allowed tumor growth rate to be evaluated and
these studies were complicated by fewer technical
difficulties. These experiments were designed to: 1)
determine whether the tumor cell lines would respond to NGF
1n_yiyg, 2) develop an optimum treatment protocol and 3) to
obtain measurements on the effect of NGF on tumor growth rate.
The results of the subcutaneous experiments are detailed in
the preceding paper. The subcutaneous experiments were
followed by intracerebral studies to evaluate NGF treatment
in the environment in which the tumors would normally be
found. Results from the intracerebral studies and additional
information on some of the subcutaneous experiments are

outlined in this appendix.

128

f1 1 - :o a :12 :-- u a -o-: ° and! :3 7c ol'oua m-_.1 s

Experiment.A was a preliminary study designed to develop
effective subcutaneous and intracerebral implantation
procedures and to obtain baseline information on the growth
of T9 anaplastic glioma implants in syngeneic Fischer 344

rats.

naisrisls_safi_us£nsfis

Ten, 90-110g Fischer 344 rats were anesthetized.with 0.9
mls of a 3:1 mixture of PromAce (Fort Dodge 10mg/ml) and
Ketaset (Bristol lOOmg/ml) administered intraperitoneal (IP).
The dorsum of the skull was shaved and swabbed with alcohol.
A hole was drilled through the skull with a 20 gauge needle
at the intersection of a line drawn directly caudal from the
medial canthus of the left eye and a line drawn between the
openings of the external ear canals. A 50 ul Hamilton syringe
equipped with a 22 gauge needle was used to implant the T9
suspensions. The needle was inserted to a depth of 4mm below
the surface of the skull. These measurements were developed
with the aid of cadaver specimens and were designed to implant
the cell suspensions into the left parietal lobe. 10" T9
cells in 25 ul DMEM-0 was injected into this location. An
equal ‘volume of this suspension. was implanted. into the
subcutaneous tissue of the right lateral thorax. Animals were

observed twice a day and were euthanized once they began to

129
exhibit progressive neurologic signs. At necropsy, the
subcutaneous tumors were measured with vernier calipers.
Sections of both the subcutaneous and intracerebral tumors
were fresh frozen, fixed in 4% paraformaldehyde for 16 hours
and then frozen, and fixed in 10% neutral buffered formalin.
Routine histologic sections were evaluated from the
subcutaneous and intracerebral tumors. The brain was
sectioned at the level of the implantation site, and 4.0 mm
cranial and caudal to the implantation site. The brain stem
and cerebellum were sectioned sagittally 2 mm lateral to the

median plane.

83311153

Rats receiving intracerebral implants generally exhibited
the following progression of clinical signs. The animals
would first develop a reddish oculonasal discharge
(chromodactoria). This was followed by depression and weight
loss. Eventually, rats would develop a variety of neurologic
signs including progressive ataxia, circling, and posterior
paresis. Animals were euthanized once progressive neurologic
signs became apparent. 90% (9 of 10) of the rats developed
intracerebral tumors and 80% (8 of 10) developed subcutaneous
tumors following implantation of T9 suspensions. The average
survival time of animals that developed intracerebral tumors
was 31.4 i 1.4 days. The average volume of the subcutaneous

implants at the time of necropsy was 1439 mmf.

130

On histopathologic examination, the tumors consisted of
an invasive population of large spindle-shaped cells arranged
in intertwining streams and bundles with minimal supporting
stroma. These spindle cells had moderate to abundant pale
eosinophilic cytoplasm.and indistinct cytoplasmic boundaries.
Marked variation in nuclear size and morphology was evident.
The large oval nuclei generally had coarsely clumped chromatin
with 1-3 variably sized, often large, irregular'nucleolin Two
to four, at times bizarre, mitotic figures were observed per
HPF. The invasive tendencies, marked pleomorphism, relatively
high mitotic index and rapid growth rate suggest that tumors
derived from T9 implants represent a highly anaplastic
neoplasm.

Though 9 of 10 rats developed intracerebral tumors
following implantation of T9 cell suspensions, only 4 of these
animals had tumors present at the implantation site in the
left parietal lobe. In all 9 cases, tumor was apparent in the
cerebellum. ‘Variably sized ribbons and nodules of neoplastic
cells were present between the cerebellar folia. The
neoplastic proliferation was invasive and frequently extended
into the cerebellar white matter. Isolated vessels within the
molecular layer were surrounded by cuffs of tumor cells.
Nodules of neoplastic cells were frequently identified at the
junction of the cerebellum and brainstem and involving the
area between the dorsal cerebellum and cerebral hemispheres.

In 8 of 9 rats the neoplastic proliferation was also apparent

131
along the ventral portions of the brainstem.
Discussion

This preliminary experiment demonstrated that both
subcutaneous and intracerebral tumors could be consistently
established in Fischer 344 rats following the implantation of
suspensions of T9 anaplastic glioma cells. This study also
provided information on the time frame for tumor development
and the progression of clinical signs. It also illustrated
several pitfalls. The two major problems identified were
inconsistent tumor development at the intracerebral
implantation site and widespread submeningeal dissemination
of neoplastic cells throughout the cranial vault. Neoplastic
cells would leak from the implantation site and settle in
locations where spacial factors favored growth or along
ventral portions of the brain. This resulted in tumors
forming around the cerebellum and along the ventral portions
of the brainstem.

We were encouraged by the high rate at which tumors
developed following implantation both intracerebrally and
subcutaneously, but were disappointed with the lack of
confinement of the intracerebral implants to the implantation
site. In order for meaningful information to be gained from
future experiments, methods to achieve more consistent
development of tumors at the intracerebral implantation site
and to minimize tumor dissemination throughout the cranial

vault would have to be developed.

132
419‘ '13! ° 9' ‘ ‘ ° , __3-11|.31 O 1‘ ’_'.L! ',
;__...,;.:.,.‘ 1 .‘-ve... -.._. ._;' . 0.111110 ;_,;

The focus of this experiment was to try and refine the
implantation procedure used in the previous experiment and to
evaluate the response of intracerebral and subcutaneous T9
implants to NGF.

Several modifications were made to our original
implantation procedure to try and alleviate some of the
problems identified in Experiment A. A guide was developed
to better control the depth at which the implants were seeded.
The implants were seeded slightly deeper into the cerebral
parenchyma to try and confine the tumors to the implantation
site. Also, the needle was left in position for 30 seconds
following implantation in an attempt to diminish the leakage

of neoplastic cells from the implantation site.

03211313403131.0213

A protocol similar to the one described for Experiment
A was used for this study, with the following modifications.
A depth gauge was developed to increase implant placement
accuracy. .An 18 gauge needle was cut at a level which allowed
the 22 gauge needle, when passed through the guide, to extend
5 mm below the end of the guide when it was positioned on the
dorsum of the skull. The tumors were implanted to a depth of
5 mm, compared to 4 mm in the previous study. The

implantation depth.was increased in an attempt to confine the

133
tumors to the implantation site. The needle was then left in
place for 30 seconds to try and diminish pressure induced
escape of cells along the needle tract.

The implants were allowed 10 days to become established
before NGF treatment was begun. Rats were divided into three
groups containing 10 animals each. All treatments were
administered subcutaneously (left flank) at a site distant
from.either'implant~ The control group received 0.4 ml of the
final dialysate solution (lomM sodium acetate + 15 mM sodium
chloride). The low dose treatment group received 10ug NGF
(0.1 ml) dissolved in 0.9% saline and administered once a day
(SID) for five days. The high dose treatment group received
40 ug NGF (0.4 mls) SID on five consecutive days. Animals
were observed twice a day and were euthanized once they began
to exhibit progressive neurologic signs. At necropsy, both
the subcutaneous and intracerebral tumors were measured with
vernier calipers. Sections of the subcutaneous and
intracerebral tumors were fresh frozen, fixed in 4%
paraformaldehyde for'16 hours and.then frozen.and fixed in 10%
neutral buffered formalin. Routine histologic sections from
the subcutaneous and intracerebral tumors were evaluated as

per Experiment A.

8332113
The average survival time for animals in the high dose

NGF treatment group (200 ug NGF total dose) was 32.4 i 0.8

134
days, for the low dose NGF treatment group (50 ug NGF total
dose) was 28.6 i 0.8 days and for the untreated control was
29.3 i 1.0 days. The value of the data on survival time is
questionable due to the variability in tumor development as
a result of widespread submeningeal dissemination of the
neoplastic cells.

Compared to the effect on survival time, the effect of
NGF treatment on the growth of subcutaneous tumor implants was
striking. The average tumor volume (average tumor volume at
the time of necropsy/survival time) in the high dose NGF-
treated group was less than half that of the average tumor
volume of the untreated control (Figure 2.4). Interestingly,
only seven (7 of 10) animals in the high dose treatment group
developed subcutaneous tumors compared to 9 (9 of 10) for the
low dose NGF treatment group and all 10 in the untreated
control group.

The changes made in the implantation procedure greatly
increased the number of tumors present at the implantation
site. 87.9% (29/33) of the rats developed tumors at the
implantation site compared to only 44.4% (4/9) in Experiment
A. However, these tumors were not well confined to the
implantation site. Neoplastic cells had leaked from the
implantation site and seeded other areas of the brain
including the cerebellum and brainstem. 81.8% (27/33) of the
animals exhibited cerebellar tumors and 75.8% (25/33) of the

animals exhibited brainstem involvement. In addition, T9

135

Figure 2.4: The effect of NGF treatment on the growth of
subcutaneous T9 implants.

 

 

 

 

 

 

 

  

 

 

 

 

 

‘ ‘ ( :
Control Low Dose NGF-Rx High Dose NGF-Rx

cu

E

1—

'; as

3 "r—

i 30

3

Q I

A 25 ~-
... 1

E \

OJ

5

5 ‘5 n-

> . _ . .
g 10 —-— ——-—-- ~ ' ' —
o 3 ,1
g _ , _ _-
§ 0 h ;. ,

S.

a:

>

<

Results are reported as average subcutaneous tumor volume at
the time of necropsy divided. by survival time (mmf/day).
Results are described in these units because of variations in
survival time between individual animals. The high dose NGF
treatment group (200 ug NGF) exhibited a significant decrease
(P < 0.05) in the average tumor volume at the time of necropsy
(12.6 i 2.02 nmfi/day) compared to the untreated control group
(27.7 i 4.52 mm/day). The low dose NGF treatment group (50
ug NGF) exhibited a slight decrease in tumor growth (24.1 i
6.30 mm3/day), however, this difference was not statistically
significant.

136
implants exhibited a tendency to form cap-like masses covering

the parietal lobe above the implantation site (42.4%, 14/33) .

Discussim

The results from these experiments, particularly those
from the subcutaneous implants, were very promising. The data
demonstrates that NGF treatment is able to decrease the growth
of subcutaneous tumor implants and increase survival time
following intracerebral implantation. However, problems
continued to plague the intracerebral studies. Though the
modifications resulted in more consistent tumor development
at the implantation site in the cerebrum, the presence of
additional tumor masses scattered throughout the cranial vault
continued to plague the procedure. The variability in tumor
development undermined the value of the intracerebral studies.
Differences in survival time of individual animals was not
exclusively a reflection of treatment, it was also dependent
on the variability in tumor development.

This experiment was the initial study in which the effect
of NGF treatment was evaluated in 21 Q. The dramatic decrease
in the growth of subcutaneous implants following NGF treatment

indicates that this form of therapy may offer significant
therapeutic potential. However, several aspects of the
exPerimental procedure need to be improved. If intracerebral
experiments are to be continued, a better method of localizing

the tumors to the implantation site needs to be developed.

137

This problem is addressed in future studies (Experiment E).

0 0
.19: 21:; ,D' 4.! 6V __._ .-01 .- 11‘ s 0. -. 9-: s

The previous experiment, Experiment B, demonstrated.that
NGF treatment was able to significantly reduce the growth rate
of subcutaneous T9 anaplastic glioma implants. It was also
apparent from. the results that there 'was a significant
difference between the efficacy of the two treatment regimes.
200 ug NGF total dose was far more effective at suppressing
tumor growth than the 50 ug total dose. Therefore, before
another large scale experiment evaluating the therapeutic
potential of NGF was initiated, it was deemed prudent to try
and optimize the ‘treatment. protocol. Experiment. 0 ‘was
designed to evaluate the effectiveness of various NGF
treatment doses and dosage schedules when injected directly

into the site of the subcutaneous T9 tumor implant.

s h s

A suspension containing 10‘ T9 cells in 25ul (4 x 105
czells/ml) was injected into the subcutaneous tissue of the
Iright caudolateral thorax of 90-1109 Fischer 344 rats. Rats
wmere palpated eight days following implantation at which time
88% had palpable nodules at the implantation site. Treatment

‘fllzs begun ten days following implantation. Rats were

138
restrained in lateral recumbency, the tumor palpated and the
NGF injected directly into the region of the tumor. 42 rats
were divided into 7 groups containing 6 animals each. The
following treatments were administered. Time was measured
from the date of implantation.
Group #1) (control 1): 0.1 ml of the dialysate, day 10.
Group #2) 35ug NGF (35ug/0.1 ml), day 10, 35ug total dose.

Group #3) 35ug NGF (35ug/0.1 ml), days 10-13, 105ug total
dose.

Group #4) 35ug NGF (35ug/0.1 ml) administered twice a day at
a 1/2 hour interval, days 10-13, 210ug total dose.

Group #5) 35ug NGF (35ug/0.1 ml), days 10-16, 210ug total
dose.

Group #6) 70ug NGF (70ug/0.2 ml), days 10-16, 420ug total
Group #7) ?2:§trol 2): 0.1 ml of the dialysate, days 10-16.

The subcutaneous tumors were measured weekly with vernier
calipers and tumor volume was calculated. Animals were
euthanized with CO2 once the tumors had reached a maximum of
1.0-1.5 cm in any one dimension. Samples of each tumor were
fresh frozen, fixed in 4% paraformaldehyde for 16 hours and

then frozen, and fixed in 10% neutral buffered formalin.

Routine histologic sections were examined from each tumor.

833111.113

Though the results from this experiment were not clear-
cut, useful information was gained from this study. It is
apparent from table 2.2 that fewer animals in the NGF treated
groups developed tumors. Because there were so few animals
in each group, comparing individual groups was not very

rewarding. However, when the percentage of rats developing

139

 

Table 2.2: The percentage of rats free from palpable
subcutaneous nodules following NGF treatment

 

 

Days Post- day 29 day 36 day 43 day 48
Implantation (necropsy)
Untreated 13.33% 33.0% 33.0% 26.7%
NGF treated 26.7% 50.0% 43.4% 40.0%

 

This table illustrates the percent of rats free from.palpable
tumor nodules following subcutaneous T9 anaplastic glioma
implants. These values are the product of combining the two

control groups and are compared to the combined NGF-treated
groups.

140
tumors in the combined NGF treated groups was compared with
that of the combined control groups, it was apparent that
fewer tumors developed in animals treated with NGF.

It is evident from figure 2.5 that NGF treatment
suppressed the growth rate of subcutaneous T9 implants. Four
of the five NGF-treated groups exhibited a decrease in growth
rate of subcutaneous T9 implants compared to the untreated
controls. This experiment once again demonstrates the ability
of NGF to diminish the growth rate of neurogenic tumors in
111g. However, a treatment protocol that was definitively
superior’ was not established. Despite the lack. of an
unambiguously optimum protocol, the results demonstrate that
210 ug NGF divided over a 6 day period is an effective
therapeutic regime.

On histologic examination morphologic differences between

NGF-treated and untreated tumors were not apparent.

121.321.135.128

Results from this experiment were both encouraging and
puzzling. It is difficult to explain the results from Group
4. Animals in this group received.a total dose of 210 ug NGF.
This was the same total dose administered to rats in Group 5
given on a schedule similar to that of Group 3. Both these
groups exhibited a decrease in tumor growth rate in response
to treatment, yet the animals in Group 4 had an average tumor

volume that was greater than either control group. In light

141
Figure 2. 5: The effect of various NGF doses and dosage

schedules on the growth rate of subcutaneous
T9 implants.

Thousands

 

.5
s

/ +-----< Group 4
4-----< Group‘T.

4m- Group—l———-
//// '7

+-----< Group 3

 

§§

..e
O

 

 

/5-—('-‘“-(-GF‘OUP—6——*
.v_%/{/ 6-----< Group 5

MM <------< Group 2

t14lillll l LLI
22 29 so 48
Days Post-Implantation

 

 

M

 

 

Average Tumor Volume (m3)
& 0 O

O

This graph illustrates the effects of various NGF treatment
regimes on the growth rate of subcutaneous T9 implants. The
following treatments were evaluated:

Group #1) (control 1): 0.1 ml of the dialysate, day 10.

Group #2) 35ug NGF (35ug/0.1 ml), day 10, 35ug total dose.

Group #3) 35ug NGF (35ug/0.1 ml), days 10-12, 105ug total
dose.

Group #4) 35ug NGF (35ug/0.1 ml) administered twice a day at
a 1/2 hour interval, days 10-12, 210ug total dose.

Group #5) 35ug NGF (35ug/0.1 ml), days 10-15, 210ug total
dose.

Group #6) 70ug NGF (70ug/0.2 ml), days 10-15, 420ug total

‘ dose.
Group #7) (control 2): 0.1 ml of the dialysate, days 10-15.

Four of the five NGF-treated groups exhibited a decrease in
growth rate compared to the controls. The regimes used in

groups 1 and 5 appear to be the most effective therapeutic
protocols.

142

of the available data, there does not seem to be a reasonable
explanation for these findings. Experiments involving
subcutaneous tumor implants were plagued by large variations
in tumor volume within individual groups. It is possible,
considering the low number of rats that were used in each
group and the large variations that occur within individual
groups, that this disparity was the result of random
experimental variation.

An equally curious finding was that the group exhibiting
the most dramatic decrease in tumor growth rate was Group 2,
the group that received the lowest dose. Again, this was
puzzling. In light of previous experimental data (Experiment
B), the significance of this finding was questionable. None-
the-less, this regime, along with the regime from Group 5 was
chosen for future experiments (Experiment E.).

The remaining groups provide more reasonable data. The
two control groups exhibited very similar growth curves.
Group 3, which received an intermediate dose of NGF, exhibited
an intermediate response. Groups 5 & 6 exhibited a dramatic
decrease in tumor growth rate reminiscent of that identified
in Experiment B. However, it did not appear that
administering twice the dose (Group 6) over the same time
frame (Group 5) provided added therapeutic benefit. As
comparable results were obtained in Group 5 while using
significantly less NGF, this schedule was adopted for future

experiments.

143

In conclusion, a clearly superior treatment protocol was I

not identified. However, results indicate that: 1) NGF
treatment is able to cause a decrease in the growth rate of
subcutaneous T9 implants, 2) fewer tumors developed in the NGF
treated groups compared to the untreated controls and 3)
protocols 5 & 2 were appeared to be the most effective and

should be considered for future experiments.

Experimsnts E: stersqtastisallwuisesw

 

._._. ._, t ., . ° ..., ._; ' . H”. -7 ; . . -. 9

e t c t 'o 3

Following the success of the subcutaneous experiments
involving T9 anaplastic glioma implants, Experiment E was
undertaken to test the ability of NGF to diminish the growth
rate of T9 implants in the brain. A number of obstacles
needed to be overcome in order to successfully evaluate the
response of an intracerebral tumor to NGF treatment. The
principal problems that needed to be addressed were: 1)
confinement of the implant to a localized area of the brain
and 2) the development of a method whereby NGF treatment could
be repeatedly administered directly at the intracerebral
implantation site.

It was of prime importance to confine the tumor to a
specific area of the brain so that differences in survival
time would be a reflection of treatment and not of variations

in tumor development. Several measures were taken to try and

144

accomplish this aim. The first was to place the implant
deeper into a solid mass of parenchyma. Once this resolution
was made, it became obvious that our previous implantation
site, the parietal cortex, was inadequate. The parietal
cortex was not an ideal place to implant the tumor because it
did not offer sufficient parenchymal mass to entrap the tumor.
In this region of the rat brain, there is a limited amount of
cortex overlying the lateral ventricles. Furthermore, the
implant cannot be placed deep into this region because it
would then become established in the ventricular system or
damage the underlying thalamic region. A large, solid area
of cerebral parenchyma was needed that would tolerate an
expansile mass. One of the few regions to fit this
description is the striatum. The tumor could be seeded deep
within the striatum to confine and contain the tumor to this
location and lessen the likelihood of submeningeal
dissemination.

Additional measures were taken to try and confine the
tumor to a specific location, besides implanting the tumor
deep within a solid mass of parenchyma. The volume of the
implant was greatly diminished to try and prevent leakage from
the needle tract. In addition, a very fine needle was used
to decrease the size of the tract through which cells could
escape. These measures were used in combination with a

stereotaxic apparatus to insure precise implant placement in

each animal.

145

With measures taken to confine the tumor to a specific
location, the second problem could be addressed. A system was
needed whereby repeated treatments could be delivered to the
intracerebral implantation site. This would avoid systemic
dilution of the administered dose of NGF and by-pass the
necessity to traverse the blood-brain—barrier. The
development of such a system represented a significant
obstacle. To repeatedly place the animal in a stereotaxic
apparatus in order to administer the treatment directly at the
implantation site would have been far too traumatic. Such a
procedure would require repeated anesthetic episodes and
repeated placement of the animal in the stereotaxic apparatus.
To restrain a rat in the stereotaxic apparatus is not a benign
procedure. The tympanic membrane is ruptured when the animal
is placed in the apparatus and repeated procedures could
result in middle and inner ear infections. More importantly,
in order to accurately locate a specific area of the brain
with a stereotaxic apparatus, it must be calibrated from a
landmark, either lambda or bregma, present on the surface of
the animals skull. This would necessitate repeated incisions
through the skin to locate one of these landmarks. This
process was considered too traumatic. An alternative would
be to place a guide through the skull to provide a access
route for the treatments. Such a procedure would offer the
following advantages: 1) it would eliminate the repeated

surgery and associated trauma necessary with stereotaxic

146
techniques 2) animals would not need to be completely
anesthetized to perform the treatments and 3) the needle guide
would provide a reliable access route to the implantation site
which would insure that the treatment was administered at the
appropriate location.

The implantation of a needle guide was determined to be
the method that most adequately fulfilled our needs. This
procedure did have one significant drawback. Several
experiments had demonstrated that 35 ug NGF administered once
a day for five consecutive days was an effective protocol.
However, this protocol seemed too traumatic to be used with
this system. Animals would need to be sedated daily for five
consecutive days and submit to a needle being passed through
their brain. For the welfare of the animal, it was decided

to spread the treatments out over a longer period of time.

M3§§I131§_§DQ_H§EDQQ§

90-110g Fischer 344 rats were anesthetized with a
Rompun/Ketamine mixture administered IP (55 parts Ketamine
100mg/ml + 45 parts Rompun 20 mg/ml) at a dose of 0.1
mls/100g. The animal was positioned in a stereotaxic
apparatus (Kopf Model 900 Small Animal Stereotaxic apparatus).
The animals head was shaved and swabbed with alcohol. A 2 cm
midline incision was made starting 0.5 cm posterior to the
orbit and extending slightly beyond the nuchal crest. Once

the skin had been reflected, the scalpel blade was again

147

passed over the midline surface of the skull to free any
remaining subcutaneous tissue which was then reflected. A
mark was made on the surface of the skull at the coordinates
of bregma and 3.5 mm to the left of the mid-sagittal suture.
A 0.8 mm hole was drilled at this point. Three addition holes
were drilled through the skull, one 5 mm cranial and slightly
medial to the first and two 5 mm caudal. These three holes
formed a triangle centered around the first hole. Three #00
x 1/8" self-taping stainless steel screws (TX00-2 Small Parts
Inc, Miami FL) were placed in the peripheral holes leaving 3/4
of the screw exposed, A.10 mm section of 23 gauge thin walled
hypodermic tubing was stereotaxically positioned in the
central hole to a depth of 1mm below the surface of the skull.
This stilet was cemented in place with dental acrylic. The
dental acrylic formed a cap on the surface of the skull with
the three peripheral screws acting as anchors. Once the
dental acrylic was dry, a 27 gauge needle was passed through
the stilet and 10‘ T9 cells (2 x 10' cells/ml) in 5 ul DMEM-
O was implanted 4.5 mm deep into the left striatum. The
stilet was then capped and the caudal portion of the incision
closed with 4-0 Ethylon suture. Animals were housed in
individual cages.

In order to administer the treatments, rats were sedated
with 0.05 mls of an acepromazine/ketamine mixture IP (2 parts
ketamine (Ketaset, Bristol 100mg/ml) + 8 parts acepromazine

(PromAce, Fort Dodge 10mg/ml}). Treatments were administered

148

through the needle guide so that the solutions could be
injected directly at the implantation site. The nine control
animals received.20 ul of the final dialysate solution.on‘days
8, 11, 14 and 17 post-implantation. The eight animals in the
single dose NGF treatment group received a single injection
of 35 ug NGF in 20 ul 0.9% saline on day 8. The eight animals
in the multiple dose NGF treatment group received 35 ug NGF
on days 8, 11, 14 and 17 post-implantation for a total dose
of 140 ug NGF. NGF activity was assessed prior to the
experiments and once treatment had been completed using an 19
219:9 bioassay. The sutures were removed at the time of the
first treatments.

Animals were observed three times per day and euthanized
with CO, once they began to exhibit progressive neurologic
signs. The experimental subjects were evaluated in terms of
survival time, and tumor volume at the time of euthanasia
divided by survival time. Routine histologic sections of the
intracerebral tumors were also evaluated. The brain was
sectioned at the level of the implantation site, and 4.0 mm
cranial and caudal to the implantation site. The brain stem

and cerebellum were sectioned sagittally 2 mm from the median

plane.

Bfiéfllié

In one respect, these experiments were quite successful.

Each animal developed a tumor at the implantation site and,

149

in all but one animal, the neoplastic proliferation was
confined solely to the implantation site in the striatum.
Therefore, it would appear that measures taken to confine
these tumors to the implantation site were very effective.
The method whereby NGF treatment was repeatedly administered
to the implantation site also appeared to be very successful.
The rats tolerated the procedure quite well. Complications
were not encountered . Rats did not develop
meningoencephalitis or surface infections associated with the
presence of the needle guide. Self-inflicted trauma
associated with the presence of the needle guide was not a
problem.

However, as table 2.3 illustrates, NGF treatment did not
result in a statistically significant difference in either
average survival time or tumor size at euthanasia divided by
survival time. There was essentially no difference between
the various groups in terms of survival time or in the measure

of tumor growth rate.

Discussien

Stereotactic implantation of a needle guide proved to be
a very useful technique. The procedure allowed for the
precise placement of the implant into the striatum followed
by direct treatment of this area. Remarkably, complications
associated with the implantation and maintenance of the needle

guide were nonexistent.

150

 

Table 2.3: The effect of NGF treatment on stereotactically

guided intracerebral T9 implants.

‘Control:

bSingle Dose NGF:
(35 ug total dose)

bMultiple doses NGF:
(140 ug total dose)

Average
Survival Time

27.37 days

27.0 days

27.37 days

Tumor volume/
survival time

6.84 mm"/day

7 . 30 mm3/day

6 . 38 mm3/day

 

'N=9
”Na-8

151

However, neither the single dose- or multiple dose-NGF
treatments resulted in a decrease in tumor growth rate or an
increase in survival time. This was very puzzling considering
the success of experiments involving subcutaneous implants.
A number of factors may have contributed to this failure. It
is quite possible that the change in therapeutic regime may
have diminished the efficacy of the treatment. It was
considered too traumatic to treat these animals daily at an
intracerebral location and we were limited by the amount of
NGF that could be administered during each intracerebral
treatment. As such, treatments were spread out over a longer
period of time and a lower total dose was administered. It
is quite possible that changes in the total dose and dosage
schedule may have decreased efficacy. There is also a
distinct possibility that serum leakage fellowing each NGF
treatment may have diminished the effectiveness of the NGF.

The preceding paper describes the adverse effects
associated with the present of serum when administering NGF
treatment. In the rats, hemorrhage associated with
subcutaneous NGF treatments was rare and only occurred when
the animal moved violently while the needle was positioned in
the subcutis. However, a completely different set of
circumstances was encountered in the brain. The tumor was
implanted deep into the striatum. As such, the meninges and
several millimeters of cerebral parenchyma had to be traversed

in order to reach the implantation site. This greatly

152

increased the likelihood of serum leakage and/or hemorrhage
at the treatment site which would diminish the amount of NGF
available to induce reverse transformation of the neoplastic
cells. It should be noted that on a number of occasions
following treatment a serosanguinous fluid welled up through
the needle guide indicating that significant hemorrhage had
occurred. Lastly, we may have been too efficient in confining
the tumors to a localized area in the brain. Even though
fewer cells were implanted in this experiment, survival time
was decreased compared to Experiment B. Tumors produced by
the implantation of T9 anaplastic glioma cells are aggressive
and grow very rapidly. Because the growth of these T9
anaplastic glioma implants is so rapid, this model may not
truly reflect the ability of NGF to diminish the growth rate
of slowly expanding naturally occurring CNS tumors.

These technically difficult and unrewarding studies were
abandoned in favor of subcutaneous experiments. The
subcutaneous studies allowed for better control of the
treatment procedures and the ability to continuously monitor
tumor growth. Results from subcutaneous experiments provide

the data on which the preceding paper was based.

a o on
Despite the failure of the previous experiment, thoughts
of additional intracerebral studies were not completely

abandoned. Following the success of subcutaneous experiments

153
involving the neurinoma clone, clone 16, intracerebral
experiments utilizing this cell lines were contemplated.
However, before intracerebral studies could be considered,
information on the consistency of tumor development and the
growth rate of intracerebral clone 16 implants needed to be

obtained.

W

A needle guide was implanted for this study using a
procedure similar to that described for the previous
experiment, Experiment E. Four animals received implants.
A suspension of ld‘clone 16 cells (passage 93) in 5 ul DMEM-
0 (2 x 10" cells/ml) was implanted into the left striatum via
the needle guide. Animals were euthanized at 30, 60, 97, and
128 days post-implantation. The brain was sectioned at the
level of the needle guide and both faces of this section were
examined for evidence of tumor. In addition, sections 4.0 mm
cranial and caudal to the implantation site, and sagittal
brain stem and cerebellum sections cut 2 mm from the median

plane were evaluated.

39.81.11.125

None of the experimental subjects developed progressive
neurologic signs during the course of this study; Tumors were
not apparent on gross inspection of brain sections at the time

of necropsy. On histologic examination the needle track was

154
successfully identified in each animal indicating that
sections had been taken at the appropriate level. However,
a neoplastic proliferation was not identified in any of the

sections.

Disassism

Though clone 16 implants grew readily in the subcutaneous
tissue of Fischer rats, none of the rats receiving
intracerebral implants developed tumors. At necropsy, though
the needle tract was successfully identified histologically,
these was no evidence of tumor formation either at the
implantation site or elsewhere in the brain. Tumors were not
apparent at any time period. There was no evidence that
neoplastic cells had survived or proliferated at the
implantation site in the brains of these animals.

It has been observed that neurinomas, tumors of probable
Schwann cell origin, may grow directly up to the brain but
will not infiltrate the CNS tissue. Apparently, the
environment of the brain is not conducive to the survival and
growth of tumors of Schwann cell origin. This appears to be
true with clone 16, a tumor of probable Schwann cell origin.
Though. these tumors. grew 'vigorously in ‘the subcutaneous
tissue, they did not grow in the environs of the brain.
Therefore, further consideration of the use of clone 16

implants for intracerebral studies was abandoned.

The preceding article mentions results from experiments
in which human tumor cell lines were implanted into irradiated
nude mice. 'The article states that "In contrast to the rodent
tumor cell lines, only a minimal response to NGF (35 ug NGF
SID for 6 days) was observed following treatment of these
tumors (0251 and 054) in nude mice. In each case, NGF
treatment resulted in a slight decrease in growth rate (data
not shown). However, the decrease was not nearly as dramatic
as with the rodent tumor lines and was not statistically
significant." The results of these experiments are
illustrated in this appendix (Figures 2.6 and 2.7). The
"Materials and Methods" and "Discussion" for these results are
contained in the preceding article. These graphs were
included to clarify the statement that "NGF treatment resulted

in a slight decrease in growth rate."

156

Figure 2.6: The effect of NGF treatment on the growth rate of
subcutaneous 0251 glioblastoma multiforme implants

 

300

3
)
M
m
o
I
l

M

O

O
l
l

4
0|
0

l

l

r

l

9

.6
O
o
I
l

 

0|
0
l
l
l
l

 

Averaqe Tumor Volume (mm

 

0111lJ_l_lLLlLllLLllllllLLllllllllllllLLllllllLLllllllll

14 17 21 24 28 as 42 49 56 83
Days Post-Tumor Implantation

Figure 2.6 illustrates the effect of NGF treatment on the
growth rate of 0251 implants in irradiated nude mice. The
growth rate of 0251 implants was decreased slightly in NGF-
treated animals (4) (35 ug SID for 6 days) when compared to
the tumor growth rate in untreated control mice (.), This
difference was not statistically significant.

157

Figure 2.7: The effect of NGF treatment on the growth rate of
subcutaneous 054 mixed glioma implants

2500

 

3)

 

N
O
O
O

.6
0|
0
O

..A
C
O
O

 

 

500

Average Tumor Volume (mm

 

 

 

Days Post-Tumor Implantation

Figure 2.7 illustrates the effect of NGF treatment on the
growth rate of 054 implants in irradiated nude mice. When the
growth rate of 054 implants in NGF-treated (o) and untreated
(.) mice are compared, it is apparent that the difference
between the two groups is minimal and not statistically
significant. Though it may be difficult to appreciate from
the graph, during the initial measurement period, tumors in
the NGF treatment group were actually significantly larger (P
< 0.05) than tumors from the untreated control. On day 21
post-implantation, tumors in the NGF-treated group were twice
the size of tumors in the untreated control group. As the
experiment progressed this trend gradually reversed, until at
the time of necropsy, tumors in the NGF-treated group were
only 86% the size of tumors in the untreated control group.

158

This appendix provides addition information on some of
the 1n_y1y9 techniques not described in the preceding paper.
It also discusses some of the pitfalls associated with these
types of experiments. Though a number of problems were
encountered during the course of the 1134129 studies, the
dramatic decrease in the growth rate of both T9 anaplastic
glioma and clone 16 neurinoma implants in response to NGF
indicates that the growth factor possesses significant
therapeutic potential. Results from chapters 1 and 2 provide
persuasive evidence that NGF may indeed prove useful for the

treatment of tumors of neurogenic origin.

159

CONCLUSIONS AND FUTURE STUDIES

A pertinent question to ask at this point would be, "what
do I perceive is the potential of this type of therapy and in
what direction should research in this area proceed?"

There are a number of valid approaches to cancer therapy.
Surgical removal, when an option, is without question the most
effective. Destruction of neoplastic cells through the use
of radiation or chemotherapy has also proven effective.
However, radiation and chemotherapy are generally plagued by
fairly severe adverse side effects, and the outcome of therapy
is generally remission and not complete cure. Therefore,
there is a definite need for the development of alternative
modes of therapy, especially in the area of brain tumor
therapy where surgical removal is often not an option.

The aforementioned forms of therapy rely on the removal
or destruction of neoplastic cells. These approaches, though
certainly effective, do not address the fundamental processes
involved in neoplastic transformation. They rely on
annihilation of the neoplastic cell. The proximate cause of
cancer is considered to be altered or aberrant expression of
normal cellular genes important in growth and development.
An equally valid approach to cancer therapy would be to
confront the disease directly at the pathways responsible for
neoplastic transformation. If pathways essential for the

maintenance of the transformed phenotype could be persistently

160
suppressed, the neoplastic behavior of transformed cells would
be effectively controlled. This is the principle upon which
reverse transformation is based.

Reverse transforming agents stimulate neoplastic cells
to differentiate to the point where they no longer exhibit
neoplastic behavior. In this process the expression of
pathways essential for maintenance of the transformed
phenotype become suppressed. As we learn more about the
mechanisms involved in neoplastic transformation and about the
processes involved in differentiation, I believe we will
identify points at which these two processes intersect. It
is at these points that reverse transforming agents may be
able to intervene in a therapeutically beneficial manner.

Reverse transformation is a conceptually sound approach
that has two additional advantages. Because natural
differentiating agents are employed, there is the potential
for diminished systemic toxicity, especially if the agent can
be administered locally and confined largely to the region of
the tumor. These agents also have the potential for increased
specificity over many types of chemotherapeutic agents because
only those cells possessing receptors and functional response
pathways will be affected.

I believe the concept of reverse transformation is valid.
Indeed, rather than annihilating the neoplastic cells because
we are unable to regulate neoplastic behavior, reverse

transformation attempts to control neoplastic behavior by

161
suppressing pathways responsible for maintenance of the
transformed phenotype. The more I contemplate this approach,
the more I am convinced that this is a valid strategy. I also
believe that NGF possess significant reverse transforming
potential. The preceding chapters chronicle the basis for
this belief.

If I were to continue to pursue this line of research,
there are three areas that I would be particularly interested
in exploring. First, there is a need to further characterize
the persistent nature of the NGF-induced response. Because
reverse transformation does not result in the direct
elimination of neoplastic cells, the effects must be highly
persistent. Data presented in Chapter 2 demonstrates the
persistent nature of‘ the NGF induced effects I1n_9z1t;9.
However, these experiments were conducted over a relatively
short period of time, considering the goals of cancer therapy.
Because persistence is such a vital component of reverse
transformation, the fate of neoplastic cells treated.with NGF
needs to be critically evaluated. Studies must demonstrate
that, at least in a subpopulation of neoplastic cells, the
effects of NGF are Ihighly’ persistent if not. permanent.
Because tumors are composed of a heterogeneous population of
cells, the various subpopulations will have to be separated
and evaluated individually for persistence. If it can be
demonstrated that the reverse transforming agent can

persistently or permanently suppress the growth of a

162
subpopulation of cells within the neoplasm, then a therapeutic
advantage will have been gained.

Second, it would be beneficial to conclusively establish
that the reverse transforming agent has suppressed pathways
essential for maintenance of the transformed phenotype.
Evidence presented in the preceding chapters clearly
demonstrates that NGF is capable of suppressing the growth of
neoplastic cells both 1n_y1tr9 and 1n_y1y9. The next step
would be to identify the mechanisms whereby this is achieved.
This could be accomplished by using cell lines in which the
defects leading to neoplastic transformation have been well
characterized or by transfecting cells with known oncogenes.
Demonstration that treatment has suppressed known pathways
essential for transformation would provide conclusive support
for this approach.

Finally, it would be interesting to see if a synergistic
effect could be achieved by combining NGF with other known
differentiating agents, such as TGF-beta and beta-carotene.
Just as combinations of chemotherapeutic agents are
significantly more effective then any single agent alone,
combinations of reverse transforming agents may also prove
more effective.

I feel strongly' that the concept of reverse
transformation is a valid approach to cancer therapy and that
NGF possesses significant reverse transforming potential.

This area of research.has been interesting and rewarding; 'The

163
ability of NGF to drastically diminish the growth rate of two
of the neurogenic ‘tumor cell lines ‘when. implanted into
syngeneic animals demonstrates potential therapeutic promise

that requires further investigation.

LIST OF REFERENCES

Aloe L., Cozzari C., and. Levi-Montalcini R. (1981) The
submaxillary glands of the African rodent 2199919

(mastgmxs) 9939199919 as the richest availible source of
nerve growth factor. Exp. Cell. Res. 133, 475-480.

Aloe L. and Levi-Montalcini R. (1979) Nerve growth factor
induced transformation of immature chromaffin cells 19
2129 into sympathetic neurons. Proc. Natl. Acad. Sci.
USA 76, 1246-1250.

Angeletti R.H. and Bradshaw R.A. (1971) Nerve growth factor
from mouse submaxillary gland: amino acid sequence.
Proc. Natl. Acad. Sci. USA 68, 2417-2420.

Appel S.H. (1981) A unifying hypothesis for the cause of
amyotrophic lateral sclerosis, parkinsonism, and
Alzheimer disease. Ann. Neurol. 10, 299-505.

Ayer LeLievre C.S., Abendal T., Olsen L., and Seiger'A. (1983)
Localization.of NGF-like immunoreactivity in rat nervous
tissue. Med. Biol. 61, 296-304.

Azar C.G., Scavarda N.J., Reynolds C.P., and Brodeur GM.
(1990) Multiple defects of the nerve growth factor
receptor in human neuroblastomas. Cell Growth and Diff.
1, 421-428.

Barbacid M., Lamballe F., Pulido 0., and Klein R. (1991) The
tgk family of tyrosine protein kinase receptors.
Biochim. Biophys. Acta. Rev. Cancer. (in press).

Barinage M. (1991) On the right track to the NGF receptor.
Science 252, 505-506.

Batchelor P.E., Armstrong D.M., Blaker S.H., and Gage F.H.
(1989) Nerve growth factor receptor and choline
acetyltransferase colocalization in neurons within the
rat forebrain: response to fimbria-fornix transection.
J. Comp. Neurol. 284, 187-204.

164

165

Bernd P. and Greene L.A. (1984) Association of 125-I-nerve
growth factor with PC12 pheochromocytoma cells. Evidence
for internalization via high-affinity receptors only and
long-term regulation by nerve growth factor of both high-
and low-affinity receptors. J. Biol. Chem. 259, 15509-
15516.

Biedler J.L., Ross R.A., Shanske A., and Spengle B.A. (1980)
Human neuroblastoma cytogenetics : search for
homogeneously staining regions and double minute
chromosomes. In "Advances in Neuroblastoma Research."
(Evans A.E., ed.) New York: Raven Press, p. 81.

Block T. and Bothwell M.A. (1983) The nerve growth factor
receptor on PC12 cells: interconversion between two forms
with different binding properties. J. Neurochem. 40,
1654-1663.

Bogenmann E., Mark C., Isaacs H., Neustein H.B., DeClerck
Y.A., Laug W.E., and Jones P.A. (1983) Invasive

properties of primary pediatric neoplasms 1pm.
Cancer Res. 43, 1176-1186.

Bocchini V. and Angeletti P.0. (1969) The nerve growth factor:
Purification as a 30,000 molecular weight protein. Proc.
Natl. Acad. Sci. USA 64, 787-794.

Bueker E.D. (1948) Implantation of tumors in the hind limb
field of the enbryonic chick and the developmental

response of the lumbosacral nervous system. Anat. Record
102, 369-390.

Burmeister 0.W. and Lyser K. (1982) Process formation in the

human neuroblastoma clone SK-N-SH-SYSY 1n_y1t;_. Diss.
Abstr. Int. 43, 1334-3.

Buxser S.E., Watson L., and Johnson G.L. (1983a) A comparison
of binding properties and structure of NGF receptor on

PC12 pheochromocytoma cells and A875 melanoma cells. J.
Cell. Biochem. 22, 219-233.

Buxser S.E., Kelleher D.J., Watson L., Puma P., and Johnson

G.L. (1983b) Change in state of nerve growth factor
receptor. J. Biol. Chem. 258, 3741-3749.

Buxser S., Puma P., and Johnson G.L. (1985) Properties of the
nerve growth factor receptor: Relationship between

receptor structure and affinity. J. Biol. Chem. 260,
1917-1926.

166

Buxser S. , Decker D. , and Ruppel P. (1990) Relationship amoung
types of nerve growth factor receptors on PC12 cells.
J. Biol. Chem. 265, 12701-12710.

Camp R.C., Koestner A., Vinores S.A., and Capen C.C. (1984)
The effect of nerve growth factor on ethylnitrosourea-
induced neoplastic proliferation in rat trigeminal
nerves. Vet. Path. 21, 67-73.

Campenot R.B. (1982a) Development of sympathetic neurons in
compartmentalized cultures. 11. Local control of neurite
growth by nerve growth factor. Dev. Biol. 93, 1-12.

Campenot R.B. (1982b) Development of sympathetic neurons in
compartmentalized cultures. II. Local control of neurite
survival by nerve growth factor. Dev. Biol. 93, 13-21.

Chao M.V., Bothwell M.A., Ross A.H., Koprowski H., Lanahan
A.A., Buck C.R. and Sehgal A. (1986) Gene transfer and
molecular cloning of the human NGF receptor. Science
232, 518-521.

Charlwood.K.A., Lamont D.M., and Banks B.E.C. (1972) In.”Nerve
Growth Factor and Its Antiserum", (Zaimis E., Knight J.,
ed.) London: Athlone, pp. 102-107.

Chen J., Chattopadhyay B., Venkatakrishnan G., and Ross A.H.
(1990) Nerve growth factor-induced differentiation of
human neuroblastoma and neuroepithelioma cell lines. Cell
Growth and Diff. 1, 79-85.

Chvatchko Y. , Van Obberghen M. , and Fehlmann M. (1984)
Internalization and recycling of insulin receptors in
hepatoma cells. Biochem. J. 222, 111-117.

Ciechanover A., Schwartz A.L., Dautry-Varsat A., and Lodish
H.F. (1983) Kinetics of internalization and recycling of
transferrin and the transferrin receptor in a human
hepatoma cell line. J. Biol. Chem. 258, 9681-9689.

Claude P., Hawrot. E., Dunis D.E., Campenot. R.B. (1982)
Binding, internalization, and retrograde transport of
125-I-nerve growth factor in cultured rat sympathetic
neurons. J. Neurosci. 2, 431-442.

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

Cohen S., and Levi-Montalcini R. (1956) A nerve growth-
stimulating factor isolated from snake venom. Proc.
Natl. Acad. Sci. USA 42, 571-574.

167

Cohen S. (1959) Purification and metabolic effects of a nerve
growth-promoting protein from snake venom. J. Biol.
Chem. 234, 1129-1137.

Cohen S. (1960) Purification of a nerve-growth promoting
protein from the mouse salivary gland and its
neurocytotoxic antiserum. Proc. Natl. Acad. Sci. USA
46, 302-311.

Cohen-Cory S., Dreyfus C.F., and Black 1.3. (1989) Expression
of high- and low-affinity nerve growth factor receptors
by Purkinje cells in the developing rat cerebellum. Exp.
Neurol. 105, 104-109.

Collins F. and Dawson A. (1983) An effect of nerve growth
factor on parasympathetic neurite outgrowth. Proc. Natl.
Acad. Sci. USA 80, 2091-2094.

Dichter M.A., Tischler A.S., and Greene L.A. (1977) Nerve
growth factor induced increase in electrical excitability
and acetylcholine sensitivity of a rat pheochromocytoma
cell line. Nature 268, 501-504.

Doherty P., Seaton P., Flanigan T.P., and Walsh F.S. (1988)
Factors controlling the expression of the nerve growth
factor receptor in PC12 cells. Neurosci. Lett. 92, 222-
227.

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

Druckrey H. , Ivankovic S. , Preussmann R. , Zulch K.J. , and
Manuel H.D. (1966b) In "Experimental Biology of Brain
Tumors.” (Kirsch W.M., Paoletti E.G., Paoletti P., eds.)
Springfield, Ill: Thomas, pp. 85-147.

Dumas M. , Schwab M.E. , and Thoenen H. (1979) Retrograde axonal
transport of specific macromolecules as a tool for

characterizing nerve terminal membranes. J. Neurobiol.
10, 179-197.

Fabricant R.N., De Larco J.E., and Todaro G.J. (1977) Nerve
growth factor receptors on human melanoma cells in
culture. Proc. Natl. Acad. Sci. USA 74, 565-569.

Gage F.H., Armstrong D.M., Williams L.R., and Varon S. (1988)
Mophologic response of axotomized septal neurons to nerve
growth factor. J. Comp. Neurol. 269, 147-155.

168

Garin Chesa P. , Rettig W.J. , Thomson T.M. , Old L.J. , and
Melamed M.R. (1988) Immunohistochemical analysis of nerve
growth factor receptor expression in normal and malignant
human tissues. J. Histochem. Cytochem. 36, 383-389.

Gerschenson M. , Graves K. , Carson S.D. , Wells R.S. , and Pierce
G.B. (1986) Regulation of melanoma by the embryonic skin.
Proc. Natl. Acad. Sci. USA 83, 7307-7310.

Gilbert F., Feder M., Balaban G., Brangman 0., Lurie D.K.,
Podolsky R., Rinaldt V., Vinikoor N., and Weisband J.
(1984) Human neuroblastomas and abnormalities of
chromosomes 1 and 17. Cancer Res. 44, 5444.

Gnahn H., Hefti F., Heumann R., Schwab M.E., and Thoenen H.
(1983) NGF-mediated increase of choline acetyltransferase
(ChAT) in the neonatal rat forebrain: Evidence for a
physiological role of NGF in the brain? Dev. Brain Res.
9, 45-52.

Godfrey E.W. and Shooter E.M. (1986) Nerve growth factor
receptor on chick embryo sympathetic ganglion cells:
binding characteristics and development. J. Neurosci.
6, 2543-2550.

Gorin P.D. and.Johnson E.M. (1980) Effects of long-term.nerve
growth factor deprivation on the nervous system of the
adult rat: An experimental autoimmune approach. Brain
Res. 198, 27-42.

Green S.H. and Greene L.A. (1986) A single Mr = 103,000 I-
125-B-nerve growth factor-affinity-labeled species
represents both the low and high affinity forms of the
nerve growth factor receptor. J. Biol. Chem. 261, 15316-
15326.

Green S.H., Rydel R.B., Connolly J.L., and Greene L.A. (1986)
PC12 mutants that possess low- but not high-affinity
nerve growth factor receptors neither respond nor
internalize nerve growth factor. J. Cell Biol. 102, 830-
843.

Greene L.A. , Varon S. , Piltch A. , and Shooter E.M. (1971)
Substructure of the beta subunit of mouse 78 nerve growth
factor. Neurobiol 1, 37-48.

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

169

Greene L.A. and Rein G. (1977) Release, storage and uptake of
catecholamines by a clonal cell line of NGF responsive
pheochromocytoma cells. Brain Res. 129, 247-263.

Greene L.A. (1978) Nerve growth factor prevents the death and
stimulates the neuronal differentiation of clonal PC12
pheochromocytoma cells in serum-free medium. J. Cell
Biol. 78, 747-755.

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

Greene L.A. , Liem R.K.H. , and Shelanski M.L. (1983) Regulation
of high molecular weight microtubule-associated protein
in PC12 cells by nerve growth factor. J. Cell Biol. 96,
76-83e

Gunderson R.W. and Barrett J.N. (1979) Neuronal chemotaxis:
chick dorsal root axons turn toward high concentrations
of nerve growth factor. Science 206, 1079-1080.

Gunderson R.W. and Barrett J.N. (1980) Characterization of the
turning response of dorsal root neurites toward nerve
growth factor. J. Cell Biol. 87, 546-554.

Gunderson R.W. (1985) Sensory neurite growth cone guidance by
substrate adsorbed nerve growth factor. J. Neurosci.
Res. 13, 199-212.

Hamburger V., Brunso-Bechtold V.J.I(., and Yip J.W. (1981)
Neuronal death in the spinal ganglia of the chick embryo
and its reduction by nerve growth factor. J. Neurosci.
1, 60-71.

Harper G.P. and Thoenen H. (1981) Target cells, biologic
effects, and mechanism of action of nerve growth factor
and its antibodies. Annu. Rev. Pharmacol. Toxicol. 21,
205-229.

Hempstead B.L., Patil N., Olson K., and Chao M. (1988)
Molecular analysis if the nerve growth factor receptor.
Cold Spring Harb. Symp. Quant. Biol. 53, 477-485.

Hempstead B.L., Schleifer L.S., and Chao M.V. (1989)
Expression of functional nerve growth factor receptors
after gene transfer. Science 243, 373-375.

Hempstead B.L., Patil N., Thiel B., and Chao M.V. (1990)
Deletion of cytoplasmic sequences on the nerve growth
factor receptor leads to loss of high affinity ligand
binding. J. Biol. Chem. 265, 9595.

170

Hempstead.B.Iu, Martin-Zanca.0., Kaplan.D.R., Parada.L.F., and
Chao M.V. (1991) High-affinity NGF binding requires
coexpression of the 1,113 proto-oncogene and the low
affinity NGF receptor. Nature 350, 678-683.

Hefti F., Dravid A., and Hartikka J. (1984) Chronic
intraventricular injections of nerve growth factor
elevate hippocampal choline acetyltransferase activity
in adult rats with partial septo-hippocampal lesions.
Brain Res. 293, 305-311.

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

Hendry IA. (1976) Control in the development of the vertebrate
sympathetic nervous system. Rev. Neurosci. 2, 149-194.

Hendry I.A., Stockel K., Thoenen H., and Iverson L.L. (1974)
The retrograde axonal transport of nerve growth factor.
Brain Res. 68, 103-121.

Hendry I.A. (1975) The response of adrenergic neurones to
axotomy and nerve growth factor. Brain Res. 94, 87-97.

Hendry I.A. and Campbell J. (1976) Morphometric analysis of
rat superior cervical ganglia after axotomy and nerve
growth factor treatment. J. Neurocytol. 5, 351-360.

Holzbauer M. and Sharman D.F. (1972) In ”Handbook of
Experimental Pharmacology." (Blaschke H. and Muscholl
E. eds.) Berlin: Springer, Vol. 33, pp. 110-185.

Hosang M. and Shooter E.M. (1985) Molecular characteristics
of the nerve growth factor receptors on PC12 cells. J.
Biol. Chem. 260, 655-662.

Hosang M. and Shooter E.M. (1987) The internalization of nerve
growth factor by high-affinity receptors on PC12
pheochromocytoma cells. EMBO J. 6, 1197-1202.

Huebner K., Isobe M., Chao M., Bothwell M., Ross A.H., Finan
J., Hoxie J.A., Sehgel A., Buck C.R., Lanahan A., Nowell
P.C. , Koprowsky H. , and Croce C.M. (1986) The nerve
growth factor receptor gene is at human chromosome region
17q12-17q22, distal to the chromosome 17 breakpoint in
acute leukemias. Proc. Natl. Acad. Sci. USA 83, 1403.

171

Huff K. and Guroff G. (1979) Nerve growth factor-induced
reduction of epidermal growth factor responsiveness and
epidermal growth factor receptors in PC12 cells: an
aspect of cell differentiation. Biochem. Biophys. Res.
Commun. 89, 175-180.

Huff K. and Guroff G. (1981) Nerve growth factor-induced
alteration in the reponse of PC12 pheochromocytoma cells
to epidermal growth factor. J. Cell Biol. 88, 189-198.

Huff K., End 0., and Guroff G. (1981) Nerve growth factor-
induced alteration in the response of PC12
pheochromocytoma cells to epidermal growth factor. J.
Cell Biol. 88, 189-198.

Ivankovic S., Druckrey H., and Preussmann R. (1966) Erzengung
neurogener Tumoren dei den Nachkommen nach einmaliger
Injektion von Aethylnitrosoharnstoff an schwangere
Ratten. Naturwissenshaften 53, 410.

Jensen L.M. (1987) Phenotypic differentiation of aphidicolin-
selected human neuroblastoma ciltures after long-term
exposure to nerve growth factor. Dev. Biol. 120, 56-
64.

Johnson 0., Lanahan A., Buck C.R., Sehgal A., Morgan C.,
Mercer E., Bothwell M., and Choa M.V. (1986a) Expression
and structure of the human NGF receptor. Cell 47, 545-
554.

Johnson E.M., Gorin.P.D., Brandeis L.0., and.Pearson.J. (1980)
Dorsal root ganglion neurons are destroyed by exposure
in utero to maternal antibody to nerve growth factor.
Science 210, 916-918.

Johnson E.M., Rich K.M., and Yip H.K. (1986b) The role of NGF
in sensory neurons in 21 9. Trends Neurosci. 9, 33-37.

Johnson E.M., Taniuchi M., Clark H.B., Springer J.E., Koh
S.Y., Tayrien M.W., and Loy R. (1987) Demonstration of
retrograde transport of nerve growth factor receptor in
peripheral and central nervous system. J. Neurosci. 7,
923-929.

Johnson E.M. and Taniuchi M. (1987) Nerve growth factor (NGF)
receptor in the central nervous system. Biochem. Parmac.
36, 4189-4195.

Kaplan D.R., Martin-Zanca 0., and.Parada.L.F. (1991a) Tyrosine
phosphorylation and tyrosine kinase activity of the trk
proto-oncogene product induced by NGF. Nature 350, 158-
160.

172

Kaplan D.R., Hempstead B.L., Martin-Zanca 0., Chao M.V., and
Parada IuF. (1991b) The trk Proto-Oncogene product: A
signal transducing receptor for nerve growth factor.
Science 252, 554-558.

Kessler J.A., Cochard P., and Black I.B. (1979) Nerve growth
factor alters the fate of embryonic neuroblasts. Nature
280, 141-142.

Kessler J.A. and Black I.B. (1980a) The effects of nerve
growth factor (NGF) and antiserum to NGF on the
development of embryonic sympathetic neurons in vivo.
Brain Res. 189, 157-168.

Kessler J.A. and Black I.B. (1980b) Nerve growth factor
stimulates the development of substance P in sensory
ganglia. Proc. Natl. Acad. Sci. USA 77, 649-652.

Kessler’ J.A. and. Black I.B. (1981) Nerve lgrowth factor
stimulates development of substance P in the embryonic
spinal cord. Brain Res. 208, 135-145.

Klein R., Jing S., Nanduri V., O'Rourke., and Barbacid M.
(1991) The 1:111; proto-oncogene encodes a receptor for
nerve growth factor. Cell 65, 189-197.

Koestner A., Swenberg J.A., and Weschler W. (1971)
Transplacental production with ethylnitrosourea of
neoplasms of the nervous system in Sprague-Dawley rats.
Am. J. Pathol. 63, 37-56.

Kordower J.H., Bartus R.T., Bothwell M., Schatteman G., and
Gash D.M. (1988) Nerve growth factor receptor
immunoreactivity in the non-human primate (99999 999119) :
distridution, morphology, and colocalization with
cholinergic enzymes. J. Comp. Neurol. 277, 465-486.

Korshing S. and Thoenen H. (1983) Nerve growth factor in
sympathetic ganglia and corresponding target organs of
the rat: correlation with density of sympathetic
innervation. Proc. Natl. Acad. Sci. USA 80, 3513-3516.

Korshing 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.

Kromer L.F. (1987) Nerve growth factor treatment after brain
injury prevents neuronal death. Science 235, 214-216.

173

Kumar S., Huber J., Pena L.A., Perez-Polo J.R., Werrbach-
Perez K., and de vellis J. (1990) Characterization of
functional nerve growth factor-receptors in a CNS glia
cell line: Monoclonal antibody 217c recognizes the nerve
growth factor-receptor on C6 glioma cells. J. Neurosci.
Res. 27, 408-417.

Kuramoto T., Perez-Polo J.R., and Haber B. (1977) Membrane
properties of a human neuroblastoma. Neurosci. Lett. 4,
151-159.

Kuramoto T., Werrbach-Perez K., Perez-PololJ.R., and Haber B.
(1981) Membrane properties of a human neuroblastoma II.
Effects of differentiation. J. Neurosci. Res. 6, 441-
449.

Landreteh G.E. and Shooter E.M. (1980) Nerve growth factor
receptors on PC12 cells: ligand-induced conversion from
low- to high-affinity state. Proc. Natl. Acad. Sci. USA
77, 4751-4755.

Large T.H., Weskamp G., Helder J.C., Radeke M.J., Misko T.P.,
Shooter E.M., and Reichardt IuF. (1989) Structure and
developmental expression of the nerve growth factor
receptor in the chicken central nervous system. Neuron
2, 1123-1134.

Lazarovici P., Dickens G., Kuzuya H., and Guroff G. (1987)
Long-term heterologous down-regulation of the epidermal
growth factor receptor in PC12 cells by nerve growth
factor. J. Cell Biol. 104, 1611-1621.

Letourneau P.C. (1978) Chemotactic response of nerve fiber

elongation to nerve growth factor. Dev. Biol. 66, 183-
196.

Levi A. , Shechter Y. , Neufeld E.J. , and Schlessinger J. (1980)
Mobility, clustering, and transport of nerve growth
factor in embryonal sensory cells and in a sympathetic

neuronal cell line. IProc. Natl. Acad. Sci. USA.77, 3469-
3473.

Levi-Montalcini R. and Hamburger V. (1951) Selective growth
stimulating effects of mouse sarcoma on the sensory and

sympathetic ganglia system of the chick embryo. J.
Exptl. Zool. 116, 321-362.

Levi-Montalcini R. (1952) Effects of mouse tumor
transplantation on the nervous system. Ann. N.Y. Acad.
Sci. 55, 330-343.

174

Levi-Montalcini R. and Hamburger V. (1953) A diffusible agent
of mouse sarcoma, producing hyperplasia of sympathetic
ganglia and hyperneurotization of viscera in the chick
embryo. J. Exp. Zool. 123, 233-288.

Levi-Montalcini R. , Meyer H. , and Hamburger V. (1954) m
experiments on the effects of mouse sarcoma 180 and 37
on the spinal and sympathetic ganglia of the chick
embryo. Cancer Res. 14, 49-57.

Levi-Montalcini R. and Cohen S. (1956) ln_yitzg and in_yiyg
effects of a nerve growth-stimulating agent isolated from
snake venom. Proc. Natl. Acad. Sci. USA 42, 695.

Levi-Montalcini R. and Booker B. (1960a) Destruction of the
sympathetic ganglia in mammals by an antiserum to the
nerve-growth promoting factor. Proc. Natl. Acad. Sci.
USA 46. 384-391.

Levi-Montalcini R. and Booker B. (1960b) Excessive growth of
the sympathetic ganglia evoked by a protein isolated from
mouse salivary glands. Proc. Natl. Acad. Sci. USA 46,
373-384.

Levi-Montalcini R. (1964) Growth control of nerve cells by a
protein factor and its antiserum. Science 143, 105.

Levi-Montalcini R. and Angeletti P.U. (1966)
Immunosympathectomy. Pharmacol. Rev. 18, 619-628.

Levi-Montalcini R., Aloe L., Mugniani E., Oesch F., and
Thoenen H. (1975) Nerve growth factor induces volume
increase and enhances tyrosine hydroxylase synthesis in
chemically axotomized sympathetic ganglia in newborn
rats. Proc. Natl. Acad. Sci. USA 72, 595-599.

Levi-Montalcini R. (1976) The nerve growth factor: Its role
in growth, differentiation and function of the

sympathetic adrenergic neuron. Prog. Brain Res. 45, 235-
256.

Levi-Montalcini R. and Aloe L. (1985) Differentiating effects
of murine nerve growth factor in the peripheral and

central nervous systems of W tadpoles.
Proc. Natl. Acad. Sci. USA 82, 7111-7115.

Levi-Montalcini R. (1987) The nerve growth factor 35 years
later. Science 237, 1154-1162.

Maher P.A. (1988) Nerve growth factor induces protein-tyrosine

phosphorylation. Proc. Natl. Acad. Sci. USA 85, 6788-
6791.

175

Marchetti D. and Perez-Polo J.R. (1987a) Nerve growth factor
receptors in human neuroblastoma cells. J. Neurochem.
49, 475-486.

Marchetti D., Stach R.W., Saneto R., de Vellis J., and Perez-
Polo J .R. (1987b) Binding constants of soluble NGF-
receptors in rat oligodendrocytes and astorcytes in
culture. Biochem. Biophys. Res. Commun. 147, 422-427.

Martinez H.J., Dreyfus C.F., Jonakait G.M., and Black I.B.
(1985) Nerve growth factor promotes cholinergic
development in brain striatal cultures. Proc. Natl.
Acad. Sci. USA 82, 7777-7781.

Martin-Zanca D., Hughes S.H., and Brabacid M. (1986) A human
oncogene formed by the fusion of truncated tropomyosin
and protein tyrosine kinase sequences. Nature 319, 743-
748.

Martin-Zanca D., Brabacid M., and Parada L.F. (1990a)
Expression of the trk proto-oncogene is restricted to the
sensory cranial and spinal ganglia of neural crest origin
in mouse development. Genes Dev. 4, 683.

Martin-Zanca D., Klein R., Brabacid M., and Parada L.F.
(1990b) In "The avian model in developmental Biology:
From organism to genes.” (Le Douarin N., Dieterlen-Livre
F., and Smith J., eds.) Paris: CNRS, pp. 291-302.

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

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

Marushige Y., Marushige K., and Koestner A. (1989b) Chemical
control of growth and morphological characteristics of
anaplastic glioma cells. Anticanc. Res. 9, 1729-1736.

Massague J. (1981) Identification of a nerve growth factor
receptor protein in sympathetic ganglia membranes by
affinity labeling. J. Biol. Chem. 256, 9419-1924.

Matsushima H. and Bogenmann E. (1990) Nerve growth factor
induces neuronal differentiation in neuroblastoma cells
transfected with NGF receptor cDNA. Mol. Cell. Biol. 10,
5015-5020.

176

Meakin 8.0. and Shooter E.M. (1991) Molecular investigations
on the high-affinity nerve growth factor receptor.
Neuron 6, 153-163.

Menesini-Chen M.L., Chen J.S., and Levi-Montalcini R. (1978)
Sympathetic nerve fiber ingrowth in the central nervous
system of neonatal rodent upon intracerebral NGF
injections. Arch. Ital. Biol. 116, 53-84.

Milbrandt J. (1986) Nerve growth factor rapidly induces c-
fos mRNA in PC12 rat pheochromocytoma cells. Proc.
Natl. Acad. Sci. USA 83, 4789-4793.

Mintz B. and Illmensee. (1975) Normal genetically mosaic mice
produced from malignant teratocarcinoma cells. Proc.
Natl. Acad. Sci. 72, 3585-3589.

Mobley W.C., Server A.C., Ishii D.N., Riopelle R.J., and
Shooter E.M. (1977) Nerve growth factor. New Eng. J.
Med. 297, 1096-1104, 1149-1158, 1211-1218.

Mobley W.C. , Rutkowski J.L. , Tennekoon G.I. , Buchanan K. , and
Johnston M.V. (1985) Choline acetyltransferase activity
in striatum of neonatal rat increased by nerve growth
factor. Science 229, 284-287.

Olender E.J. and Stack R.W. (1980) Sequestration of 125-I-
labeled.beta-nerve.growth factor by sympathetic neurons.
J. Biol. Chem. 255, 9338-9343.

Otten 'U., Goedert. M., Mayer N., and Lembeck. F. (1980)
Requirement of nerve growth factor for development of
substance P-containing sensory neurones. Nature 287,
158-159.

Otten U., Goedert M., Schwab M., and Thibault J. (1979)
Immunization of adult rats against 2.58 NGF: effects on
the peripheral sympathetic nervous system. Brain Res.
176, 79-90.

Perez-Polo J.R., Werrbach-Perez K., and Tiffany-Castiglioni
E. (1979) A human clonal cell line model of
differentiating neurons. Dev. Biol. 71, 341-355.

Perez-Polo J.R., Reynolds C.P., Tiffany-Castiglioni E.,
Ziegler M., Schulze I., and Werrback-Perez K. (1982) NGF
effects on human neuroblastoma lines: A model system.
In "Proteins in the nervous system: Structure and
function.” New York: Alan R. Liss Inc., pp. 285-299.

177

Podesta A.H., Mullins J., Pierce G.B., and Wells R.S. (1984)
The neurula stage mouse embryo in control of
neuroblastoma. Proc. Natl. Acad. Sci. USA 81, 7608-
7611.

Radeke M.J., Misko T.P., Hsu C., Herzenberg L.A., and Shooter
E.M. (1987) Gene transfer and molecular cloning of the
rat nerve growth factor receptor. Nature 325, 593-597.

Raju N.R. , Koestner A. , Marushige K. , Lovell K.L. , and Okazaki
D. (1989) Effect of nerve growth factor on the

transplacental induction of neurinomas by
ethylnitrosourea in Sprague-Dawley rats. Cancer Res. 49,
7120-7123.

Reed J.R. and England D. (1986) The effect of nerve growth
factor on the development of sodium channels in PC12
cells. Biochem. Cell. Biol. 64, 1153-1159.

Rettig W.J., Spengler B.A., Garin Chesa P., 01d L.J., and
Biedler' J.L. (1987) Coordinate changes in neuronal
phenotype and surface antigen expression in human
neuroblastoma cell variants. Cancer Res. 47, 1383.

Rettig W.J., Thomson T.M., Spengler B.A., Biedler J.L., and
01d L.J. (1986) Assignment of human nerve growth factor
receptor gene to chromosome 17 and regulation of receptor
expression in somatic cell hybrids. Somat. Cell Mol.
Genet. 12, 441.

Reynolds C.P. and Perez-PolotJ.R. (1975) Human neuroblastoma:

glial induced morphological differentiation. Neurosci.
Lett. 1, 91-97.

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. J. Neurosci. Res. 6, 319-325.

Richardson P.M., Verge Isse V.M.K., and Riopelle R.J. (1986)
Distribution of neuronal receptors for nerve growth
factor in the rat. J. Neurosci. 6, 2313-2321.

Rodriguez-Tebar A., Dechant G., and Barde Y.A. (1990) Binding
of brain-derived neurotrophic factor to the nerve growth
factor receptor. Neuron 4, 487-492.

Ronne H., Anundi H., Rask L., and Peterson P.A. (1979) Nerve
growth factor’ binds to serum. alpha-z-macroglobulin.
Biochem. Biophys. Res. Comm. 87, 330-336.

178

Ross A.H., Grob P., Bothwell M., Elder D.E., Ernst C.S.,
Marano N. , Ghrist B.F.D. , Slemp C.C. , Herlyn M. , Aikinson
B., and Koprowski H. (1984) Characterization of nerve
growth factor receptor in neural crest tumors using
monoclonal antibodies. Proc. Natl. Acad. Sci. USA 81,
6681-6685.

Ross A.H., Bothwell M., Chao M.V., Dietzschold B., and
Koprowski H. (1987) The nerve growth factor receptor
using monoclonal antibodies. In "The molecular biology
of receptors: Techniques and applications of receptor
research." (Storsberg A.D., ed.) New York: VCH, pp. 72-
91.

Rudkin B.B., Lazarovici P., Levi B.Z., Abe Y., Fujita K., and
Guroff G. (1989) Cell cycle-specific action of nerve
growth factor in PC12 cells: differentiation without
proliferation. EMBO J. 8, 3319-3325.

Rudy B., Kirschenbaum B., and Greene L.A. (1982) Nerve growth
factor-induced increase in saxitoxin.binding to rat PC12
pheochromocytome cells. J. Neurosci. 2, 1405-1411.

Schechter A.L. and Bothwell M.A. (1981) Nerve growth factor
receptors on. PC12 cells: Evidence for 'two receptor

classes with differing cytoskeletal association. Cell
24, 867-874.

Schlessinger J., Shechter Y., Willingham M.C., and Pastan I.
(1978) Direct visualization of binding, aggregation, and
internalization of insulin and epidermal growth factor
on living fibroblastic cells. Proc. Natl. Acad. Sci. USA
65, 2659-2663.

Schubert.D., Hienenmann.S., and Kidakoro'Y. (1977) Cholinergic
metabolism and synapse formation by a rat nerve cell
line. Proc. Natl. Acad. Sci. USA 74, 2579-2583.

Schubert D. and Whitlock C. (1977) Alteration of cellular
adhesion by nerve growth factor. Proc. Natl. Acad. Sci.
USA 74, 4055-4058.

Schwab M.E., Otten U., Agid Y., and Thoenen H. (1979) Nerve
growth factor (NGF) in the rat CNS: absence of specific
retrograde axonal transport and tyrosin hydroxylase
induction in the locus coeruleus and substantia nigra.
Brain Res. 168, 473-483.

Schwab M.E., Heumann R., and Thoenen H. (1982) Communication
between target organs and nerve cells: retrograde axonal
transport and site of action of nerve growth factors.
Cold Spring Harb. Symp. Quant. Biol. 46, 125-134.

179

Scott S.M., Tarris R., Eveleth D., Mansfield H., Weichsel
M.E., and Fisher D.A. (1981) Bioassay detection of mouse
nerve growth factor (mNGF) in the brain of adult mice.
J. Neurosci. Res. 6, 653-658.

Seeley P.J. , Keith C.H. , Shelanski M.L. , and Greene L.A.
(1983) Pressure microinj action of nerve growth factor and
anti-nerve growth factor into the nucleus and cytoplasm:
Lack of effects on neurite outgrowth from
pheochromocytoma cells. J. Neurosci. 3, 1488-1494.

Seiler M. and Schaeb M.E. (1984) Specific retrograde transport
of nerve growth factor (NGF) from cortex to nucleus
basalis in the rat. Brain Res. 300, 33-39.

Shelton D.L. and Reichardt L.F. (1986) Studies on the
expression of the beta nerve growth factor (NGF) gene in
the central nervous system: level and regional
distribution of NGF mRNA suggest that NGF functions as
a trophic factor for several distinct populations of
neurons. Proc. Natl. Acad. Sci. USA 83, 2714-2718.

Smith A.P., Varon S., and Shooter E.M. (1968) Multiple forms
of nerve growth factor protein and its subunits.
Biochem. 7, 3259-3268.

Sonnenfeld 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.

Sonnenfeld K.H. and Ishii D.N. (1985) Fast and slow nerve
growth factor binding sites in human neuroblastoma and
rat pheochromocytoma cell lines: Relationship of sites
to each other and to neurite formation. J. Neurosci. 5,
1717-1728.

Stack R.W. and Wagner B.J. (1982) Decrease in the number of
lower affinity (type II) nerve growth factor receptors
on. embryonic sensory' neurons. does? not affect fiber
outgrowth. J. Neurosci. Res. 7, 103-110.

Stahn R., Rose S., Sanborn S., West G., and Herschmenn H.
(1975) Effects of nerve growth factor administration on
N-ethyl-N-nitrosourea carcinogenesis. Brain Res. 96,
287-298.

Stephenson R.P. (1956) A modification of receptor theory. Br.
J. Pharmacol. 11, 379-393.

Stockel K., Paravicini U., and Thoenen H. (1974) Specificity
of the retrograde axonal transport for nerve growth
factor. Brain Res. 76, 413-421.

180

Sutter A. , Riopelle R.J. , Harris-Warrick R.M. , and Shooter
E . M . ( 1979 ) Nerve growth factor receptors :
Characterization of two distinct classes of binding sites
on chick embryo sensory ganglia cells. J. Biol. Chem.
254, 5972-5982.

Swenberg J .A., Koestner A. , Weschler W. , and Denlinger R.M.
(1972) Quantitative aspects of transplacental tumor
induction with ethylnitrosourea in rats. Cancer Res. 32,
2656-2660.

Szutowicz A. , Frazier A. , and Bradshaw R.A. (1976) Subcellular
localization of nerve growth factor receptors. Thirteen-
day chick embryo brain. J. Biol. Chem. 251, 1516-1523.

Taniuchi M. and Johnson E.M. (1985) Characterization of the
binding properties and retrograde axomal transport of
monoclonal antibody directed against the rat nerve growth
factor receptor. J. Cell Biol. 101, 1100-1106.

Taniuchi M. , Clark H.B. , and Johnson E.M. (1986) Induction of
nerve growth factor receptors in Schwann cells after
axotomy. Proc. Natl. Acad. Sci. USA 83, 4094-4098.

Tait J.F., Weinman S.A., and Bradshaw R.A. (1981) Dissociation
kinetics of 1"“‘I-nerve growth factor from cell surface
receptors. Acceleration by unlabeled ligand and its
relationship to negative cooperativity. J. Biol. Chem.
256, 11086-11092.

Thoenen H. and Barde Y.A. (1980) Physiology of nerve growth
factor. Physiol. Rev. 60, 1284-1335.

Thomson T.M. , Rettig W.J. , Garin Chesa P. , Green S.H. , Carrato
Mena A., and Old L.J. (1988) Expression of human nerve
growth factor receptor on cells derived from all three
germ layers. Exp. Cell Res. 174, 533-539.

Tischler A.S. and Greene L.A. (1978) Morphologic and
cytochemical properties of clonal rat pheochromocytoma

cells which respond to nerve growth factor. Lab. Invest.
39, 77-89.

Turkington R.W. and Riddle M. (1970) Expression of
differentiated function by mammary carcinoma cells in
yitrg. Cancer Res. 30, 127-132.

Tuszynski M.H. , Hoi Sang U. , Amaral D.G. , and Gage F.H. (1990)
Nerve growth factor infusion in the primate brain reduces
lesion-induced cholinergic neuronal degeneration. J.
Neurosci. 10, 3604-3614.

181

Ullrich A., Gray A., Berman C., and Dull T.J. (1983) Human B-
nerve growth factor gene sequence highly homologous to
that of mouse. Nature 303, 821-825.

Unsicker K., Krisch B., 0tten.U., and Thoenen H. (1978) Nerve
growth factor induced outgrowth from isolated rat adrenal
chromaffin cells: impairment by glucocorticoids. Proc.
Natl. Acad. Sci. USA 75, 3498-3502.

Vale R.D. and Shooter E.M. (1982) Alteration of binding
properties and cytoskeletal attachment of nerve growth
factor receptors in PC12 cells by wheat germ agglutinin.
J. Cell Biol. 94, 710-717.

Vale R.D. and Shooter E.M. (1983) Conversion of nerve growth
factor-receptor complexes to a slowly dissociating,
Triton X-100 insoluble state by anti-nerve growth factor
antibodies. Biochem. 22, 5022-5028.

Vale R.D. and Shooter E.M. (1985) Assaying binding of nerve
growth factor to cell surface receptors. Methods
Enzymol. 109, 21-39.

Varon S., Nomura J., and Shooter E.M. (1954) The isolation of
the mouse nerve growth factor protein in a high molecular
weight form. Biochem. 6, 2202-2209.

Verge V.M., Riopelle R.J., and Richardson P.M. (1989) Nerve
growth factor receptors on normal and injured sensory
neurons. J. Neurosci. 9, 914-922.

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

Vinores S.A. and Koestner A. (1981) Effect of nerve growth
factor producing cells on anaplastic glioma and
pheochromocytoma clones: involvement of other factors.
J. Neurosci. Res. 6, 389-401.

Vinores S.A. and. Koestner .A. (1982) Reduction of
ethylnitrosourea-induced neoplastic proliferation in rat

trigeminal nerves by nerve growth factor. Canc. Res. 42,
1038-1040.

Waris T., Rechardt L., and Waris P. (1973) Differentiation of
neuroblastoma cells induced by nerve growth factor in
yi;;_. Experientia 29, 1128-1129.

182

Whittemore S.R. , Ebendal T. , Larkfors L. , Olson L. , Seiger A. ,
Stromberg I. , and Persson H. (1986) Development and
regional expression of beta nerve growth factor mRNA and
protein in the rat central nervous system. Proc. Natl.
Acad. Sci. USA 83, 817-821.

Wechsler‘W., Kleihues P., Matsumoto S., Zulch K.J., Ivankovic
S., Preussmann R., and Druckrey H. (1969) Pathology of
experimental neurogenic tumors chemically induced.during
prenatal and postnatal life. Ann. N.Y. Acad. Sci. 159,
360-408.

Welcher A.A., Bitler C.M., Radeke M.J., and Shooter E.M.
(1991) Nerve growth factor binding domain of the nerve
growth factor receptor. Proc. Natl. Acad. Sci. USA 88,
159-163.

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

Woodruff N.R. and Neet K.E. (1986) Beta nerve growth factor
binding to PC12 cells. Association kinetics and
cooperative interactions. Biochem. 25, 7956-7966.

Yaeger M.J., Koestner A., Marushige K., and Marushige Y.
(1991) The reverse transforming effects of nerve growth
factor on five human neurogenic tumor cell lines: I)!
21:39 results. Acta Neuropath.

Yanker B.A. and Shooter E.M. (1982) The biology and mechanism

of action of nerve growth factor; .Ann. Rev. Biochem. 51,
845-868.

Zaimis E. (1971) Nerve growth factor (NGF) and multipotential
cells. J. Physiol. 216, 65-66.

Zimmerman A. and Sutter A. (1983) Beta nerve growth factor
(NGF) receptors on glial cells. Cell-cell interaction
between neurons and Schwann cells in cultures of chick
sensory ganglia. EMBO J. 2, 879-885.

VITA

The author was born in Minneapolis, Minnesota on June 17,
1958. He received the<degree of Doctor'of‘Veterinary’Medicine
from the University of Minnesota in 1984.“ He author spent
two years in private practice, principally as a small animal
practitioner in Eagle River, Alaska. In July, 1986 he
accepted a position as a resident/instructor in anatomic
pathology at Michigan State University in the Department of
Pathology. After completion of the residency program, the
author was admitted into a Ph.D. program as a postdoctoral
fellow in the Department of Pathology.

The author will join the Veterinary Science Department
at South Dakota State University where he will participate in
teaching, research and diagnostics at the South Dakota Animal

Disease Research and Diagnostic Laboratory.