CURRENT CONCEPTS OF MALIGNANT GROWTH

drivercutInternet και Εφαρμογές Web

13 Νοε 2013 (πριν από 3 χρόνια και 9 μήνες)

564 εμφανίσεις

CURRENT CONCEPTS OF
MALIGNANT GROWTH

Part A. From a Normal Cell to Cancer

JANlS 0. ERENPREISS Latvian Institute of Experimental & Clinical
Medicine, Riga, LATVIA

Edited by
GUNTIS
BRUMEUS

and
MARUTA DZERVE


1993

ZVAIGZNE PUBLISHERS RIGA, LATVIA


© Janis
Erenpreiss, 1993

All rights reserved. No part of this book may
be

reproduced, stored in a retrieval system or
transmitted in any form or by any means, electronic, mechanical, photocopying or otherwise without
the prior written permission of the Author



IS
BN

5

405

01282

3

Published by


Zvaigzne Publisher



Kr. Vademara iela 105,


Riga, LV 1013


LATVIA


Licence N 000304186.

Cover design by
Arnolds Kr
eslinsh

Layout design by
Elizabete
G
urska

Desktop
pub
lishing by
M
ari
s
Polkmanis

Printed in the Republic of Latvia

Preface


5

Abbreviations

................................



6

Chapter I. DEFINITIONS OF TERMS


7

Chapter 2. CHEMICAL CANCEROGENESIS:

DISCOVERY OF THE BASIC
PRINCIPLES

OF CANCEROGENESIS

...........................


13

2.1.

General

................................
................................
.



13

2.2.

Initiation
...........................



14

2.3.

Promotion


16

2.4.

The latency
period

of tumour development

..............


.


................................
.....


21

2.5.

The role of oncogenes

in chemical cancerogenesis

.................


............................


22

Contents


Chapter 3. THE ONCOGENE CONCEPT:

A HISTORICAL OVERVIEW

............................


23

3.1.

Discovery of oncogenes. The molecular mechanism of
cancero
genesis seems to be simple and clear

........


.

23

3.2.

The oncogenes are numerous and diverse


25

3.2.

Cl
assification

................................
......




27

3.2.2.

Distribution

................................
............

in living
organisms


................................
................................
...

28

3.2.3.

Location

................................
.........................

on
human chromosomes

................................
.........................


32

3.2.4.

Mechanisms of activation

................................
.........


33

3.3.

The
sis

oncogene: the last effort

for an unified concept

................................
.......................


39
Chapter 4. THE ONCOGENE CONCEPT:

CURRENT INTERPRETATION

................


43

4.1.

Complementary oncogenes

................................
.................


43

4.2.

Oncogene expression in tumours

................................
......


40

4.3.

The key oncogenes,
myc

and
ras


................................
.......


4
g

4.3.1.

myc


................................
................................
.................


48

4.3.2.

ras

..........................


.


. .

...


................................
..............


54

4.3.2.1.

General

................................
..............................


54

4.3.2.2.

p2I
ras


................................


................................
..


59

4.3.2.3.

Mechanisms of
ras

oncogene activation

.....


61

4.4.

Protein
-
kinase and related oncogenes

..............................


67

4.4.1.

The
abl

oncogene .. . /

................................
...................


67

4.4.2.

src

and related oncogenes

................................
..........


69

4.5.

A classification of oncogenes according to their function

in cancerogenesis

................................
................................
..


71

4.6.

Some other oncogenes

................................
.........................


73

Chapter 5.
THE EMBRYOLOGICAL THEORY

OF CANCER: AN ATTEMPT TO COMPREHEND THE
MULTITUDE

OF

DATA


79

5.1.

From Julius Cohnheim to the oncogene

..........................


79,

5.2.

Gametogenesis: the role of oncogenes

..............................


85

5.2.1.

Origin of germ cells

................................
.....................


85

5.2.2.

Female germ cell development

................................
..


88

5.2.3.

Regulatory mechanisms of oocyte meiosis

.......

-
...

89

5.2.4.

Male germ cell development

................................
......


97

5.2.5.

Gene expression during spermatogenesis

.............


98

Chapter 6. SYNOPSIS: CONC
LUSIONS,

SPECULATIONS, PREDICTIONS

.............

103

References

................................
................................
.........................


113
Preface

Theory provides a conceptual
coherence

for a certain field of knowledge. Factual
data of this field serve as material for the construction of a theory.

Any attempt to develop a theory of cancerogenesis encounters a major
impediment, which is not lack but plethora of data. That is why the first

task is to
select the appropriate data from this pool of oncological phenomenology.

During its century
-
long history, experimental oncology has arrived at certain
statements which permit outlining of a general framework for the theory. Research
in chemical

cancerogenesis disclosed the most general rules of the conversion of a
cell from normal into tumorous. The discovery of oncogenes, which followed and is
continuing, revealed the molecular mechanisms of cancerogenesis.

However, extension of research, under
taken with great enthusiasm, yielded such
a variety and diversity of cancerogens and oncogenes that the statements which
were initially clear became blurred.

It is obvious that a guiding idea, a kind of working hypothesis that could serve
as a criterion fo
r selection of the basic data, is needed for outlining a general theory
of cancerogenesis.

The guiding idea proposed in this book is as follows.

A normal biological process may provide a basic pathway for cancerogenesis.
Thus, the key to understanding canc
erogenesis is to be searched for in both
identification and investigation of these normal pathways as tentative analogues of
cancerogenesis. In my view, a molecular analogy between embryonal and cancer
cells, i.e., the ability of any tumour cell to synthes
ize proteins of embryonal type,
gives sound reasons for searching for normal analogues of cancerogenesis in the
Definitions

5


pathways of early ontogenesis.

These considerations gave an impetus for an attempt in this book to analyse
contemporary data of experimental onc
ology from the viewpoint of the oldest concept
of tumour growth, the embryological theory of cancer.

I should like to express my gratitude for the help and advice that several persons
gave me during the making of this monograph. So, I should like to thank
my
colleages, Mrs. Taiga Gulite, Mrs. Ruta Zirne, and Mr. Maris Brazma, who helped
me with the bibliographic reference system. Also, I note with appreciation the work
done by Mr. Brazma in keyboarding the text and amendments.

J.O.Erenpreiss Riga, Latvia,
March
1993.

2
-

AAP



2
-
acetylaminophenanthrene

#

ASV



avian sarcoma virus

B(a)P, BP



benzo(a)pyrene

DMBA



7,12
-
dimethylbenz(a)anthracene

EBV



Epstein
-
Barr virus

ENU



ethylnitrosourea

GF



growth factor

EGF



epidermal growth factor

FGF



fibroblast
growth factor

PDGF



platelet
-
derived growth factor

TGF



transforming growth factor

IP
3



inositol triphosphate

3
-

MCA



3
-
methylcholanthrene MNU



N
-
methyl
-
N
-
nitrosourea MSV



murine sarcoma virus

PAH



polycyclic aromatic hydrocarbon

PK
-
C



calcium
activated, phospholipid
-
dependent

protein
-
kinase, protein
-
kinase C

RSV



Rous sarcoma virus

SV40



Simian virus 40

TPA



12
-
0
-
tetradecanoylphorbol
-
13
-
acetate (also

called myristate acetate)

DAG



diacylglycerol

C
HAPTER

'

Definitions

I

6

Definitions


of terms

A complete
and widely recognized theory of cancerogenesis is lacking at present.
Therefore, the interpretation of cancerogenesis terms is variable. It is extremely
important to define the terms used here, at the beginning of this work. The
definitions given below may

or may not correspond to those given by other
authors. It is most important that the reader clearly comprehends what the
author wishes to express when using each term.

Cancerogenesis terminology also differs depending on the level being
analysed: organism
, cellular or molecular. The cellular level is given priority in
this book when describing any events. However, an attempt is constantly made to
describe any related molecular mechanisms. Most of the concise definitions given
in this section will be expand
ed later with appropriate references. Only the
literature which deals directly with the terminology of cancerogenesis is cited in
this chapter.

Cancerogenesis

is the standard term to describe generation of neoplasia in the
broadest sense [28, 405]. In Engl
ish, the term "carcinogenesis" is most often used
[404, 461, 503]. This term in its strict sense signifies the development of an
epithelial tumour only (a carcinoma). For this reason a group of experts [28]
propose to replace the term "carcinogenesis" by "
cancerogenesis".

Cancerogenesis is a specific biological process which is the same regardless of
etiological agent, animal species, and reacting tissue.

For cells growing
in vitro,

the term "(malignant) transformation" is commonly
used. This means that a c
ell line is capable of growth as a tumour when implanted
in a suitable host [643].

Induction

implies the generation of neoplasia that would not have occurred in
the absence of an inducing agent [302].

It is necessary to outline the boundaries of cancerogenesis: it begins with the
appearance of specific cellular changes and ends with the emergence of the first
tumour cell. Cancerogenesis does not include normal development or pathology
which precedes or
follows
malignization, such as progression, metastatic spread,
differentiation, etc. Since cancerogenesis is treated as a cellular process, the
metabolism of the cancerogen, for example, its activation, cannot be attributed to
cancerogenesis, as it is some
times done [289J.

No satisfactory definition for cancer (cell) has hitherto been
suggested. Instead, multiple characteristics are proposed, which are for
the most part optional (not indispensable). The most fundamental
features of tumour cells are those de
signated by the term
autonomy
.
This means that tumour cells do not behave as integral elements of the
body, irrespective of whether they proliferate more or less rapidly than
the normal cells, remain undifferentiated, or undergo terminal
differentiation.
"Autonomy’’ is sometimes equated with "independence",
which is untenable both linguistically and oncologically (Greek "autos”
-

self, "nomos"
-

law, rule). The tumour cell autonomy is closely related to
Definitions

7


(yet not identical with) their autodynamic properties
, viz. fixation of
malignant properties, their persistence after removal of etiological
agents, and transmission of these properties to subsequent cell
generations. Fixation of malignant properties distinguishes tumorous
from phenotypicalIy transformed cel
ls. Phenotypically transformed
cells may display the same features as tumour cells, but they return to
the normal state as soon as the agent inducing these properties is
removed.

The establishment of cell autonomy indicates that cancerogenesis is
complete
and a tumour has emerged.

Cancerogen

is a chemical, physical or biological agent which induces
cellular changes that are both specific and necessary for
cancerogenesis.

Cancerogenesis has two stages, initiation and promotion.
Accordingly, cancerogens fall
into two groups, complete and incomplete
cancerogens
[
461
]
. A c
omp
lete
ca
nc
erog
en (synonym: solitary
cancerogen) [28] is an agent that has the ability to induce the entire
process of cancerogenesis
[28, 155, 461]
. An i
ncom
plete
cancerogen

is
able to induce

only one of the two stages. Hence, the incomplete
cancerogens are subdivided, into initiators and promoters. Every
division (group) includes chemical, physical, and biological agents.
Practically all chemical and physical initiators act as complete
cancer
ogens when applied at sufficient dosage, whereas biological
initiators fail to do so. Under certain experimental conditions, some
promoters behave as complete cancerogens of low potency.

Continuous application of a cancerogen causes tumours only after a
ce
rtain minimum latent period. The formula that takes this principle
into account is called the Weybull distribution
[
802
]
.

The latency period

of tumour development, or tumour latency [608],
or induction period
[
302
]

is the time between the first application of a
cancerogen and the emergence of a tumour
[
29
]
. The duration of
tumour latency is determined

by the animal species, the cancerogen type,
the experimental regime, and the cellular and molecular mechanisms of
cancerogenesis.
Shortest latency

is determined exclusively by the
cancerogenesis mechanism, and it is an inherited species
-
specific trait. Its
length cannot be reduced by elevating the cancerogen dosage [928]. A
reduced length of the shortest latency is possible when the cell omits some
event(s) of cancerogenesis. This is possible if: (I) at the onset of experiment,
the cell already possesses certain tumorous properties; (
2
) the cell genome
becomes integrated with an activated oncogene; or (3) only phenotypic
transformation but not actual

cancerogenesis occurs.

Phenotypic transformation

is a labile manifestation of certain tumorous
cellular features that occur in response to extrinsic factors and disappear
when the action of the agent is cancelled. It can be induced, e.g., by the
transform
ing growth factor

(TGF) that causes anchorage
-
independent cell
growth [461].

Initiation

is the first stage of cancerogenesis (both induced and
spontaneous) [461].
Initiators

are agents capable of inducing the first stage
of cancerogenesis, the process of i
nitiation [814]. The term "initiator" is
sometimes misused as a synonym of "complete cancerogen" [62, 481].
8

Definitions


Correspondingly, initiated cells are defined as cells capable of forming
carcinomas when grafted into an animal [798]. Initiated cells do not exhibi
t
tumorous phenotype.

Promotion

is the second and final stage of cancerogenesis. A (tumour)
promoter

is an agent capable of inducing the second stage of cancerogenesis
where a cell is converted from initiated to neoplastic [461]. Besides the
above
-
mentione
d (oncological) definition, the term "promoter" is used in
biology to denote the site of transcription initiation on the DNA molecule.

Until 1976, promoters (croton oil, TPA) were commonly known as
"cocancerogens" and the term "promotion" described the pro
cedure of
cocancerogen application [404]. Soon after, Berenblum substantially
expanded the meaning of "cocancerogen" [77]. As a result, this term was in
the latest terminology recommendations [28] deprived of semantic value
and was hence abandoned from use

in this book.

An
oncogene

is a gene, whose activity causes one of the stages of
cancerogenesis. The terms, and corresponding symbols, used for designating
oncogenes were proposed by J.M.Coffin et al. [173]: viral oncogene=v
-
onc;
cellular oncogene=c
-
onc. T
he frequently used term, "protooncogene", denotes
"the normal cellular counterpart of a gene identified as causing a tumour"
[
461]. However,
ras

seems to be the only gene for which a clear distinction
can drawn between its activities as an oncogene and as
a protooncogene.
Therefore, it is not surprising that the two terms were freely interchanged
when
describing in the same paper both "the role of oncogenes in
normal development" and "the expression of protooncogenes in
tumour cells” [98]. I share R.A.Weinb
erg's view that
"protooncogene" is merely an ill
-
made synonym of c
-
one

[1113],
and hence I shall not use it in this book.

Oncogene activity

implies three different meanings: (i) oncogene
function, which like the function of any other structural gene is
man
ifested through transcription and translation (expression). However,
expression does not necessarily signify that the gene would function as
oncogene in a given situation, since most, if not all, oncogenes fu
l
fil
physiological functions under certain condi
tions; (ii) the transforming
activity of oncogenes, which can be detected in transfection assay.
However, only some oncogenes possess this activity. The
myc

oncogene,
for example, cannot be assayed [8]. Consequently, a negative result of
transformation tes
t does not exclude the possibility that the tested gene is
an oncogene, (iii) for the
ras

oncogene, the mutant allele is considered to
be active.

Complementary oncogenes

are pairs of oncogenes that jointly realize
the entire process of cancerogenesis. Synonymous to the term
"complementary oncogenes" is "cooperating oncogenes" [452, 548].

The protein encoded by an oncogene is designated "p" if it is
exclusively the product

of an oncogene, and "P", if it is a fused protein, i.e.
the product of both an oncogene and an adjacent viral gene
(gag,
pol

or
env).

Phosphorylated and glycosylated oncoproteins are abbreviated as
"pp" and "gp", respectively.

Definitions

9


Reference is made to "early
event" and "late event" in speculations on
the role played by oncogenes at any stage of cancerogenesis, especially
when the time of activation of the
ras

oncogene is discussed. The "early"
and "late" events, used in this context, are not attributed to any
clearly
-
cut stages of cancerogenesis. This confusion in terminology is responsible
for a good deal of controversy as to the role of the
ras
oncogene in the
stages of cancerogenesis (for details, see Chapter 4.3.2.3.).

Much confusion has also been introduce
d by the use of the term
"progression". Each author tends to attribute a personal meaning to this
word, as, f.i., the following: the entire process of cancerogenesis [820]; a
particular stage thereof, e.g., promotion [90]; a special case of
cancerogenesis,

e.g., neoplastic transformation
in vitro

[360]; or a third
(after initiation and promotion) stage of cancerogenesis [295, 813]. The
term "progression" was introduced by Foulds, who gave an unequivocal
description of the process and its definition [302]. A
ccording to Foulds,
progression

means phenotypic and genotypic
changes of

tumour cells
,
transition from benign to malignant tumours and a rise in malignant
properties [28, 302]. Since the emergence of tumour cells
indicates the
completion of cancerogenesis
, it is clear that progression, which represents
further multiple changes of the tumour that has already emerged, cannot be
attributed to any period of cancerogenesis. Relations between
cancerogenesis and progression are schematically depicted in Fig. I.


Besides the oncological terminology discussed above, some
other terms
that are often used in relation to cancerogenesis should be explained.

Gene
regulation

and also the regulation of oncogenes are governed by
various mechanisms. Two types of interactions of genes within the same
genome can be distinguished,
cis
-

and trans
-
regulation [232].

Regulatory factors are called "trans
-
acting", if the genes that encode
them do not reside on chromosomes containing the genes they control. The
cis

acting factors interact with "cis" recognition sequences, which are
defined as
the transcriptional apparatus elements residing beside the genes
they regulate.


Fig. 1.

The stages of cancerogenesis.

10

Definitions



Molecular communication between cells occurs by autocrine, paracrine
and endocrine regulation [461, 993,

1055] (Fig. 2).
Autocrine

r
egulation

is
self
-
stimulation of a cell by production of both a stimulating

Fig. 2.

Three types of intercellular regulation
(stimulation): a = autocrine, b = paracrine, c =
endocrine.

11




Definitions


factor and its specific receptor.
Paracrine regulation

involves stimulation
of a cell via the action of a substance produced by a neighbouring cell.
Endocrine regulation

denotes s
timulation by a factor that is produced by a
specific cell in a gland, and acts at a distance from this cell.

Cells belonging to the
germ (cell) line

give rise to spermatozoon and egg
(ovum), and transmit genes from generation to generation and are therefo
re
potentially
immortal
. All other cells are
somatic
.

A cell strain

is a population of cells subcultured more than once
in vitro,

and it lacks the properties of indefinite serial passage.
A cell

line

is a
population of cells grown for an indefinite period of time
in vitro

by serial
subcultures. This period of time presumes potential "immortality" of the
cells when serially cultured
in vitro

[401],
CHAPTER
Chemical Cancerogenesis
:


Discovery of the Basic

Principles of Cancerogenesis

2.1.

GENERAL

Most of the present research in tumour development is directed to the
study of oncogenes. However, the basic laws of cancerogenesis have been
established by research of chemical cancerogenesis. These laws are as
follows: stage
-
wise development; irreversibility; specific laws of tumour
latency.

The two stages of cancerogenesis were discovered by Deelman [229,
230] who found that scarification of
mouse skin pre
-
treated with coal tar
stimulated the eme
r
gence of tumours that arose on the edges of the healing
wound. A decade later, Twort and Twort [1079] induced tumours with oleic
acid in mouse skin pre
-
treated with BP. Oleic acid was later substitute
d by
croton oil [75].

The term "cocancerogen" was initially proposed to describe a secondary
agent capable of inducing tumours only after application of a sub
-
threshold
dosage of a cancerogen [404]. This term has now been replaced by
"promoter". The notion
s of two
-
stage cancerogenesis were completely
formulated by Berenblum and co
-
authors [76, 78, 80, 318, 461, and 928].
They argue that cancerogenesis in mouse skin consists of two stages,
2

Chemical Cancerogenesis


initiation and promotion. A complete cancerogen (for example, B(a)P)
acts
both as initiator and promoter when applied at sufficient dosage. However,
only the initiating potencies are manifested when applied at sub
-
cancerogenic doses. Initiated tissues can develop tumours after application
of a promoting agent. The promoting

agents do not exhibit tumorigenic
potency in no

initiated tissue.

The initiated state of a cell persists throughout the life of affected
animals [250]. Further, after exposure of a pregnant P generation to an
initiating dosage of a cancerogen, the cells o
f the Fl, and even F2,
generations are promotable with TPA [679]. The laws of two
-
stage
cancerogenesis, revealed by using B(a)P, extend to other cancerogens from
the PAH
-
group, and other groups of chemicals [28]. Physical
cancerogens
also
possess the initi
ation activity
[308, 466, 683,
718J. Incomplete
physical and chemical cancerogens are mutually
interchangeable. For example, initiation can be induced by
radiation, and subsequent promotion by TPA [685J. Two
-
stage
cancerogenesis, which has been demonstrate
d primarily on
mouse skin, can also be induced in other species, organs, and
tissues
[928].
The processes of initiation and promotion can also
be reproduced
in vitro

[166,
414;
9641. Cells exposed to ENU
transplacentally
in vivo

are promotable by TPA
in
vitro

[520].

The universality of initiation and promotion is considered to be
integral to the general character of the entire mechanism of
cancerogenesis, independent of the animal species, reacting tissue, and
etiological agent. According to Druckrey [266
], the universality of the
principles of the latency period also argues in favour of the general
character of the cancerogenesis mechanism.

Investigation of the mechanisms of initiation and promotion has
continued until today. Proliferous new evidence has
considerably
entangled the previously logical interpretations. The terminology used is
also very ambiguous (see Chapt.l). Sometimes, even the two
-
step
mechanism of cancerogenesis was rejected [45, 870]. Thus, before
continuing with the analysis of contempo
rary data, the general
principles should be outlined.

Initiation and promotion are essential and separate stages of
cancerogenesis. Each of these stages possesses its own intrinsic
mechanism. AU complete cancerogens exhibit initiator capacity. For this
rea
son, the term "initiator" is sometimes used as a synonym of
"cancerogen" (see, f.ex., [62 and 787]). This is not correct, as initiation is
only one step of cancerogenesis.

2.2.

INITIATION

Initiated cells are characterized by three distin
ctive features: 1)

immortality (cancelling of the Hayflick limit), 2) the blockage o
f

terminal
differentiation, and 3) promotability

(under the influence of promoters the
initiated cells can complete the process of cancerogenesis) [29, 344, 504,
507, and 9
27]. The initiated cell does not exhibit phenotypic
characteristics of the tumour cell. The three basic properties are
obviously interrelated, since data about their dissociation are lacking.
Therefore, verification of one of these properties is sufficient

to suggest
the presence of the other two and to regard the cell as initiated. For
example, epidermal cells of SENCAR mice display constitutive blockage
of terminal differentiation and are promotable without treatment of the
initiator [491, 790, and 1012].
Initiation

14


Immortalization likely represents a key feature of initiation, leading
to both the blockage of differentiation and promotability. Immortalization
initiates cancerogenesis, whether induced by a chemical cancerogen
[58,101, 727], irradiation [718, 719], or o
ncogenes [344]. Immortalization
is the earliest event in spontaneous transformation
in vitro

[525] (see
also Chapt. 4.1.). Two genes that determine promotability (termed
pro
-
1
and pro
-
2
), have been identified [180]. These genes in
it
i
a
te life span
extension

of the cells.

Besides these three basic features of the initiated cell, some others
have been described. Initiation represents an irreversible cell state,
preserved through the life
-
time of an animal [28]. However, it is possible
to experimentally rever
se

this state. F
.

ex., protease inhibitor treatment
of radiation initiated cells has been shown to rever
se

the cells to their
"uninitiated" condition [495]. The initiation state of the cell disappears
during promotion. Initiation is assumed to take place ins
tantaneously
[28], since a single application of a cancerogen is sufficient for induction
of this process. However, a clear rationale is lacking here. A single
contact of a cell with an initiator cannot imply instantaneous action,
since the DNA
-
adducts of
the cancerogen conserve for a long time.

The majority of chemical cancerogens act by yielding highly reactive
intermediates that bind covalently to DNA. It is assumed that this bond
with DNA causes the initiation effect [469, 529, 594, 760, 838, and 1117];

the degree of tumour
-
initiating activity of various PAH correlates with
their ability to bind to DNA [122, 246]. During the first days after
application of a cancerogen, the binding ability with DNA is massive,
reaching a maximum in approximately 24 h [21
9
, 716]. Later, a portion
of the
adducts disappear, but a substantial amount remains bound to
DNA

for a few weeks [37, 716, 815]. For example, 42 days after the last
treatment of rats with 2
-
acetylaminophenanthrene (2
-
AAP), the adduct
concentration was sti
ll 10
-
40% of the peak adduct value [368]. The
persistence of DNA adducts is species
-
specific. Adducts were observed to
preserve in mice tissue better than in rats [874]. It has been proposed
that differences in the kinetics of preservation of adducts expla
in the
differences between the resistance of tissue and species to chemical
cancerogens. Transplacental induction of cancerogenesis by

3
-

MCA in mice is partly determined by the Ah locus which controls
the DNA adduct level [586]. Some estrogens induce cell im
mortality
in
vitro

[766] and cancerogenesis
in vivo

[349, 883]. Estrogens, capable of
inducing tumours (for example, diethylstilbestrol), exhibit a target organ
-
specific covalent binding to DNA [574, 841].

However, the degree of overall binding of any activated cancerogen to
DNA does not always relate to the initiating efficiency [407, 929, and
969]. The distribution of 2
-
AAP adducts, the extent of binding and
adduct
persistence are similar between target an
d non

target tissues (368J. The
yield of induced tumours is independent of the timing of promotion after
initiation. Induced tumour yield was similar when promotion occurred 7
Promoti
on

15


and
21

days after initiation. However, the amounts of DNA adducts at these
point
s of time differ considerably.

Mutation is traditionally proposed to be the mechanism of initiation,
whether induced by a chemical cancerogen or by radiation. This suggestion
is supported by the following rationales: I) the cancerogens possessing
initiator

capacity bind to DNA, 2) initiation occurs instantly, 3) initiation is
an irreversible event, and 4) the cancerogens are mutagens. The first three
reasons are clearly indirect and invalid. The mutagenicity of cancerogens
was investigated very extensively.

However, the relationship between
mutagenicity and the cancerogenic potency of chemical compounds may be,
depending on the mode of analysis, both confirmed [474, 877] and rejected
[256, 821, 1117]. The correlation is absent for various groups of chemical
cancerogens: N
-
nitroso

compounds [575], dibenzo[a,e]fluoranthrene
proximate metabolites [1175], aromatic amines [290]. The observation that
weak mutagens are also poor initiators is the only real relationship found
between mutagenicity and cancerogenic pot
ency [928]. However, this
relationship was discredited by the fact that the transformation frequency
was compared with the mutation frequency of genes not related to
cancerogenesis in these studies [57].

The metabolic activation pathways of cancerogenic co
mpounds are
different for mutagenesis and cancerogenesis, indicating that the mutagenic
properties of cancerogens might be coincidental rather than causal [576]. In
both
in vitro

and
in vivo

experiments, the process of transformation can be
clearly dissoci
ated from the mutation frequency [ 1147]. The
in vitro

transformation of rodent cells by chemical cancerogens and radiation can
occur with a much higher efficiency than by random mutation. For example,
the efficiency of transformation induction by 3
-
MCA
in

vitro

can reach
100% while a specific genetic mutation may occur at a rate of
10

6
/cell
division, or even less [682]. This argues strongly against a mutational
origin of initiation.

Initiation represents only the first stage of cancerogenesis, and the
ini
tiated cell does not possess the tumorous phenotype. This phenotype
becomes established only in promotion, the second and final stage of
cancerogenesis.

2.3.

PROMOTION

Promotion should not be regarded as an augmentation of properties
acquired by the cell during

the course of initiation, or as acquisition of
some complementary properties. It is a qualitatively new stage of
cancerogenesis, where many cellular events take an opposite course.

Each of the two stages of cancerogenesis differs by the agents which
elicit
or inhibit them, the mechanism of action and biological characteristics.

Promoters cause a wide range of cellular changes, most of which are
likely not directly associated with cancerogenesis [250]. The specific
action of promoters is generally mani
fested by a complex of phenotypic
features designated as mimicry of transformation [1116, 1117]. This
16

Chemical Cancerogenesis


complex includes, for ex., fetalism [970]. Promoters induce the
transformed phenotype in both normal and initiated cells. However, this
state becomes stab
ilized in the initiated cells only after a period of action
that has lasted at least 1
-
2 weeks [251]. In non
-
initiated (intact) cells, the
phenotype reverses to the normal state as soon as the action of the
promoter has been ceased. The molecular basis of
this difference between
initiated and intact cells is poorly understood [1092, 1132].


Modulation of differentiation is one of the most regular effects of
promoter treatment [11.592]; both stimulation and suppression of
differentiation have been reported [
705, 1087, and 1160]. In each case,
the particular effect, or its magnitude, may depend on the cell type, the
tissue
-
specific and species
-
specific origin, and the level of differentiation
[9, 250, 378, 1129, and 1132]. However, the promoters most often
enh
ance differentiation. The use of the enhancement of differentiation
was suggested as a test assay of the promoting potency of a chemical
[474]. The ability to induce differentiation is one of the main properties
that distinguish promoters from initiators w
hich cause the blockage of
differentiation. This same property may be used for classification of
oncogenes (see Chapter 4).

The main effects of promoters are mediated by their action on the cell
membrane [251, 813, 916, 1087, 1159, and 1161]. The phorbol e
ster
receptor has been identified as calcium
-
activated phospholipid
-

dependent proteinkinase C (PK
-
C) [146, 1040]. The activity of PK
-
C
increases sharply after TPA binding [21, 221, 250, and 1126].
Simultaneously, the enzyme becomes redistributed between c
ell
compartments [161, 168, 674, 784, 954, and 1126]. Both phorbol and non
-

phorbol ester tumour promoters act as PK
-
C activators [156, 413, 470,
672]; the same applies to ultraviolet radiation when applied at a
promoting dose [626]. The different response
s of various cell types to
phorbol esters correlate with the differential expression of PK
-
C [436].
The inhibition of PK
-
C activity prevents the promoting effect of TPA
[413], All of these data lead to the conclusion that the activation of PK
-

C
is a key m
echanism of the action of tumour promoters [161, 470, 731, 744,
1148].



The other effects of phorbol esters, such as induction of
differentiation, are also mediated by protein kinase C activation [300,
342,408, 413, 627, 653, 733, and 908] and translocati
on of the enzyme [343,
1110]. In this respect, tumour promoters likely mimic events related to
normal differentiation. PK
-
C is involved in the endogenous signals that
induce differentiation of competent
Xenopus

ectoderm to the neural tube
[762]. PK
-
C media
ted induction of terminal differentiation is required for
tumour promotion, at least in skin [413]. Based on the data mentioned
above, a definition may be proposed:
a tumour promoter is

a
chemical or
any other agent that activates PK
-
C.

This definition may

lack precision,
but is useful for both focussed investigations of the molecular mechanism of
promotion and preliminary screening of agents of unknown nature [473,
Promoti
on

17


686
].

Tumour promoting phorbol esters activate PK
-
C by direct coupling with
this protein kin
ase [35, 406]. Hydrolysis of phosphatidylinositol produces
diacylglycerol (DAG), a natural endogenous activator of PK
-

C. The
biochemical effects induced by DAG are similar to those of TPA [974, 1091].
PK
-
C can also be activated by protein products encoded

by proteinkinase
oncogenes. For example, pp60
SfC

and p
68
ro
* act by phosphorylation of
phosphatidylinositol [160].

Wounding has been repeatedly shown to act as a promoter. Partial
hepatectomy can act as a promoter in rat liver cancerogenesis [461], and
wounding has a promoting effect in viral cancerogenesis [918, 959]. Lesion
of cells and tissues activates growth factors, such as epidermal growth
factor (EGF), transforming growth factor (TGF), platelet derived growth
factor (PDGF) [27], etc. TGFs are gen
erally considered to be "wound
hormones" [
1100
], which are released from platelets and are produced both
upon wounding [36] and during promoter treatment [10]. Tumour
promoting phorbol esters and growth factors similarly affect various
cellular events such

as proliferation and angiogenesis [487, 694, and 1036].
Since all of the effects are realized through activation of PK
-
C, the causal
agents
-

tumour promoters and GF
-

can be mutually substituted [1073,
1124]. Also, trombocytes contain their own PK
-
C [112
5].

The phorbol ester receptors (PK
-
C) have been detected in various
normal cells and tissues, but not in mature red blood cells of mammals [34].
The highest number of receptors was observed in the brain. These
receptors exhibit high evolutionary conservat
ism and occur in mammals,
hydra, sea urchin, and
Drosophila melanogaster.

A certain gene, which
encodes a PK
-
C homologue in the budding yeast
Saccharomyces
c
erevisiae,

is essential for cell growth and G2

> M transit [571],

PK
-
C is encoded by a multigene f
amily that includes at least six
members that code for closely related, yet unique enzymes [97, 439,
528,
731, 734,744,675, 762, and 1123]. The agents that influence PK
-
C not
only cause activation of the enzyme, but also induce directed translocation
of sep
arate isozymes. Tumour promoting phorbol esters, growth factors and
the proteins encoded by protein
-
kinase oncogenes have similar effects on
the activity of PK
-
C and its intracellular redistribution [23, 235].

-

The diverse and sometimes opposite effects of PK
-
C may be explained
by differences in the translocation sites of individual isozymes, and
therefore, the phosphorylation of different substrates
[389, 675].
The
redistribution of PK
-
C isozymes may be crucial
for promotion
[456].
One of
the PK
-
C isozymes is translocated to cytoskeleton elements on activation
[675], while another moves to the nuclear envelope resulting in
phosphorylation of several nuclear envelope polypeptides
[296, 468,
and
555]. In this way,
the effects caused by tumour promoters can reach the cell
nucleus and affect transcription of a set of genes, named phorbol
-
ester
-
responsive genes [393, 798] or TPA
-
induced sequences (TIS) [1062].

At least three of the oncogenes that encode cytoplasmic Ser
/Thr kinases
18

Chemical Cancerogenesis


(
mos
,
raf,

and

pi
m
)

are phorbol ester
-
responsible
[11, 97, 464,
and 555]. A
partial clone of the
fos

oncogene also is TIS
[1062].
The genes induced by
TPA in mouse osteoblastic cells were called OTS
[740].
The constitutive
expression of one of

these genes was found in
cis
-

transformed cells.
Modulation of differentiation by promoters may also be realized either in
cooperation with oncogenes or by their activation [179], a short
-
lived
nuclear phosphoprotein encoded by the
ets
-
2
oncogene is stabi
lized by
activation of PK
-
C under the influence of tumour promoters. This
oncoprotein is one of the genome regulating elements [313]. Another target
for TPA is the "A
-
element", an enhancer binding protein that stimulates
DNA replication
[705].
Tumour promo
ters activate the transcription factor
AP
-
1 which is considered to play a key role in switching on gene
expression that ultimately leads to DNA replication and cell division
[25],

Curcumin, a potent inhibitor of tumour promotion, suppresses AP
-
I activity
[445].
The influence of tumour promoters on the cell genome can also be
mediated by topoisomerase II (Topo II), which is activated by PK
-
C
[142,
1185].
Topo II is a nonhistone protein of the nuclear matrix. This enzyme
participates in the regulation of chr
omatin structure and function
[324].
Phorbol ester treatment induces changes in the chromatin structure which
are similar to the Topo II
-
mediated DNA cleavage
[1185].
Free
-
radical
reactions can mediate the action of tumour promoters of DNA and
chromosomes
[148,
270,
418,
509, and 613]. Tumour promoters are reported
to be able to induce formation of the extrachromosomal circular DNA, sister
chromatid exchanges and other genetic effects [250].

Recently promotion was discovered to consist of two distinctive
processes, conversion and propagation [510, 608, and 969]. As a result,

two
-
stage cancerogene
sis with a strictly fixed sequence of
initiation and promotion was transformed into a three
-
stage
process: initiation I conversion
-

propagation. The terms "Stage
I
tumour promotion" and "Stage II tumour promotion" are also
used. Converting agents (Stage I promoters)


f
.

ex., TPA
-

are full
promoters, since they act as propagators as well. Agents such as
12
-
O
-
retinol
-
phorbol
-
B
-
acetate (RPA) and mezerein are
propagati
ng agents only (Stage
II
promoters) [318, 407, 608, 970].

DAG activates PK
-
C and acts as a propagating agent but lacks
converting activity [1091]. This indicates that PK
-
C participates in Stage
II tumour promotion only. However, the problem was not fully
clarified.
The synthetic lipid second messenger /h
-
l,2
-
didecanoyl
-
glycerol
stimulates PK
-
C activity in DMBA
-
initiated skin and acts as a complete
tumour promoter. Prostaglandin F may participate in both stages of
promotion [317],

A single converting treatm
ent is sufficient to induce conversion in
initiated mouse skin [608]. The conversion stage induced by a single TPA
application has a half
-
life of 10 to 12 weeks. When the time between
initiation and the start of Stage I promotion is prolonged up to several

weeks, a conversion treatment becomes more and more dispensable,
indicating that conversion occurs spontaneously at a slow rate in
initiated skin [608]. This has also been observed
in vitro

[725]. It was
hypothesized that TPA, though directly affecting th
e initiated cell, only
accelerates the (conversion) process that occurs spontaneously as a
consequence of initiation [535].

The non
-
inversibility of the initiation
-
promotion sequence was
considered to be one of the fundamental characteristics of the classi
cal
two
-
stage experiment [79]. In contrast, the converting treatment can be
carried out several days to weeks prior to initiation [318, 511, and 758].

The only known response that correlates with converting efficiency is
chromosomal damage (clastogenesis),

which was observed for converting
agents (full promoters) and not for mezerein as a propagating agent [319,
608, 804]. Clastogenesis is unlikely to be the mechanism of conversion.
Rather, it suggests the presence of a genetical mechanism of
conversion.

Th
e majority of data suggests that in two
-
stage skin cancerogenesis
strain differences are inherited as an incomplete dominant trait related
to the promotion phase [412, 712]. A common genetic pathway controls
sensitivity to a variety of skin tumour promotin
g agents [247]. Possibly,
the genes pro
-
1 and
pro
-
2 are involved [180]. This mechanism does not
imply differences in PK
-
C activity or isozyme redistribution [247], which
means that the inheritance of promotability and the entire
cancerogenesis is governed
by the conversion stage.

Tumour Latency

21

20

Chemical
Cancerogenesis


2.4.

THE LATENCY PERIOD OF TUMOUR
DEVELOPMENT

One of the characteristic features of cancerogenesis, whether induced by
chemical, physical, or biological agents, is the extended period of tumour
latency (TL), the time b
etween the first application of a cancerogen and the
development of a tumour [29,608]. Tumour latency considerably exceeds the
time necessary for elementary cell processes, such as mitosis, protein
synthesis, mutation, etc. The latent period positively cor
relates with the
species
-
specific life span of the animal. All of the authors who have worked
on this problem view that
TL
shares a certain part of the species
-
specific
life span

(SSL) [197,554, 722, and 1077]. In other words,

TL/
SSL = constant (1).

However, the values given for this constant vary between 1/3 and I/ 10.
The true value can only be determined if both TL and SSL are known.

Variation in the values of SSL given in literature is due to differences in
the methods of determination [204, 329,
385, 618, 777, 873, and 922]. The
simplest practical method is to assume that the maximum life span equals
the SSL. This approach can be objected to, but the differences between
maximum and species
-
specific life spans are minimal. The SSL is 2 years
for mo
use, 3.5 years for rat, 100 years for human (more accurately, 98±5
years). The average life span cannot be used since it varies with life
conditions, and is approximately only half of the SSL [539, 818].

The determination of tumour latency is more complica
ted. The average
and the mean latencies are highly variable values depending on the
cancerogen used and the experimental schedule. The only value that is
fixed and can be estimated is the shortest latency (ShL). ShL is the time
necessary to induce a tumour

by using the most active cancerogen in an
optimum schedule. Data on the shortest latency of three most explored
species are as follows: 40 days for mouse, 70 days for rat, and 5 years for
human [197, 265, and 536]. A comparison of shortest


These results are unexpectedly consistent and provide the constant for
formula (I):

ShL/SSL

=
5%

(2).
By studying rat liver tumour induction with cancerogens under various
schedules of treatment, Druckrey found that the yield of tumours is
dependent solely on the summary dose of the cancerogen
applied during the
whole experiment [265,266]. The increase of cancer with advancing age is
well known [254]. If age only provides exposure to environmental
latency with SS
L for these species follows.



SSL

ShL

ShL/SSL

mouse

730 days

40 days

5%

rat

1280 days

70 days

5%

human

100

yrs

5 yrs

5%


cancerogens, and if cancerogenesis is a stochastic process [1077], then the
Druckrey principle pred
icts a positive correlation between spontaneous
tumour incidence and SSL [26, 254]. However, this is not the case. The
incidence of spontaneous tumours is the same in outbre
e
d mice, rats,
hamsters and humans, although their SSL differ significantly [254, 3
90].

Such a result is quite expected from the formula (I) which shows that an
increase in SSL is coupled with a proportional increase in TL. The Druckrey
formula is true exclusively within the limits of a single species.

2.5.

THE ROLE OF ONCOGENES

IN CHEMICAL
CANCEROGENESIS

After discovery of oncogenes, the study of chemical cancerogenesis acquired a
new character. In accordance with the oncogene concept, the essence of
cancerogenesis is activation of oncogenes. Indeed, chemical and physical
cancerogens are kno
wn to induce all of the effects that lead to activation of
oncogenes: DNA demethylation [430], gene amplification [434], point
mutations and chromosomal aberrations [474, 821], etc. Tumour promoters
and growth factors enhance expression of many oncogenes [
114,196, 533,
703,835, 967]. Expression of c
-
ras

[19, 366, 392, 575, 1114, 1167] and c
-
myc

[906, 1166, 1167] is elevated in tumours induced by chemical and physical
agents. Cell cycle control of c
-
myc

(but not c
-
ras)

expression is lost during
chemical canc
erogenesis [138]. In some cases, deregulated c
-
myc

expression
is a direct consequence of the cancerogen action [117]. Carcinomas induced
in mice [40] and rats [1022] by chemical cancerogens carry the mutationally
activated c
-
H
-
ras oncogene.

Activated
ras

a
nd
src

oncogenes, like tumour promoters, may exert their
effects through both activation of PK
-
C [326, 460] and stabile intracellular
redistribution of this enzyme [252].

Activation of the oncogenes
myc

and
ras

is very likely the most common
event in both
chemical [49] and physical [131] cancerogenesis. Fetalism, one
of the most characteristic features of tumour cells, is also determined by
oncogene activity [1167].
22

Chemical
Cancerogenesis


3

CHAPTER


The Oncogene Concept: A
Historical Overview

3.1.

DISCOVERY OF ONCOGENES. THE
MOLECULAR
MECHANISM OF CANCERO
GENESIS SEEMS TO BE
SIMPLE AND CLEAR

The existence of genetic information within the genome that can manifest in
tumorous transformation was proposed long before the discovery of
oncogenes. Hypothetical schemes considering th
e origin, nature and function
of this type of genetic information were proposed [1051, 1052, and 1053]. The
term "oncogene" was introduced to describe this transforming genetic
information [447],

The genome of Rous sarcoma virus (RSV) was found to contain
four genes
-

gag, pol, env,

and
src.

In 1970, it was first observed that only one of these
four genes,
src,

is responsible for RSV
-
mediated cell transformation [348,
998]. The study of deletion mutants that were competent for replication but
defective in t
ransformation resulted in the localization of the
src

gene in the
3' region of the viral genome.

Tumours were induced in young chicken with cloned DNA encoding only
v
-
src

[316]. The oncogene was always found included in the genome of the
transformed cell.
Oncogenes were also found in the genomes of other
retroviruses, thus confirming the universality of this discovery.

Further, the active oncogenes were revealed in the genome of tumour
cells of non
-
viral origin [288]. Hybridization exp
eriments using oncogene
sequences showed that not only tumorous but also normal cells possess genes
homologous to
src

[988, 1000]. This discovery, in concert with subsequent
experiments, provided strong evidence on the origin of virus transforming
sequence
s from a homologous cellular gene acquired by the retrovirus via
recombination [348]. The terms v
-

src

and c
-
src

were recommended to
indicate whether a gene is located in the viral or cellular genome, and
w
-
onc

and c
-
one

were proposed as general terms [173
]. The c
-
src

gene was shown to
be widespread among vertebrates and expressed in normal cells [989]. The c
-
src

gene
The Oncogenes Are Diverse

23


can replace the v
-
src gene in an RSV. Normal c
-
src

sequences are
tumorigenic if placed under the control of the promoter of RSV [1153],

Both

the viral and cellular
src

genes code for phosphoproteins of
molecular weight 60 kDa, called pp60
v
-
s
rc

and pp
60
c
-
srs
, respectively. This
oncoprotein possesses an intrinsic protein
-
kinase activity with a strict
specificity for tyrosine residues [126, 183].

pp60
sre

is localized on the
cytoplasmic side of the plasma membrane and is abundant at sites of
adhesion plaques and cell
-
cell contacts of transformed cells. The pp60
sre

substrates, vinculin and talin, are found at the same locations and are
necessary for

cell
-
to
-
cell adhesion [348].

The degree of vinculin phosphorylation correlates with the amount of
anchorage
-
independent growth in transformed cells. The products of other
oncogenes were also found to associate with the plasma membrane and
possess an enzym
atic activity similar to that of
src.

The protein encoded by
v
-
abl

is a tyrosine
-
specific proteinkinase. However, c
-
abl

lacks this activity.
Translocation of c
-
abl

in chronic myeloleukaemia, followed by oncogene
amplification, leads to the appearance of th
e protein product with the same
protein
-
kinase activity
as v
-
abl

[222]. Modification of pp60
src
, which
preserves tyrosine
-

kinase activity of the protein, but deprives it of the cell
membrane
-

binding ability, eliminates the transforming potency [483].

All

of these observations formed the foundation of a simple and
understandable scheme which described the transforming action of the
oncoproteins. The oncogene products were viewed to change the properties of
the cellular surface, causing the inability of nor
mal cell communication. A
cell which has lost normal functioning of intercellular contacts becomes
tumorous [190, 1013].

Simultaneously, the old dispute of whether a cell surface or a genome
changed in the tumour cell was resolved, since the changes in the

cell
surface are determined by the genome. Observations on the changes induced
by pp60
s
rc

in the proteins of cytoskeleton [52, 493] did not conflict with the
scheme, but complemented it.

On the whole there is a good correlation between the activity of
onc
ogenes and the tumorous cell phenotype (though with some exceptions)
[879, 1115]. Tumorous transformation is preceeded by oncogene
transcription and synthesis of the encoded protein. Neutralization of the
oncoprotein, for example by means of a monoclonal a
ntibody, results in a
phenotypic reversion of transformed cells [264]. This led to the conclusion
that the activation of the oncogene is always the mechanism of
cancerogenesis, and that tumour growth arises from the activity of
oncoproteins.

As a result of all these discoveries, the oncogene concept, which
originated by studying RSV, grew into a theory of tumour
growth. This
theory united all tumours by the mutual pathogenetic trait, regardless
of the etiology and histogenesis, for all animal species including humans. The
conception of the oncogene was closely involved with the evolution of a theory
24


The Oncogene Concept


on the mechani
sm of cancerogenesis and oncogene activation, as well as the
action of oncoproteins. The definition of the oncogene clearly reflected the
current views of the 1980's:
the oncogene is a gene whose activity is
indispensable and sufficient for the tumour grow
th. At the same time a
serious conflict existed between the oncogene concept and the multi
-
stage
nature of cancerogenesis

[46].

3.2.

THE ONCOGENES ARE NUMEROUS AND
DIVERSE

After their discovery, oncogene research quickly diversified. The profuse
amount of data
obtained altered the initially elegant scheme and staggered all
of the conclusions that at the time seemed clear.

An incomplete list of the virus oncogenes with their encoded proteins is
given in Table I. In cases where the
v
-
onc

w
as

released from the vari
ous
retroviral isolates, the molecular mass of the oncoprotein and some other
characteristics are often variable. The variables which do not essentially
deviate are omitted in the table.

Table I

RETROVIRAL ONCOGENES [89, 173, 461, 610, 1174].


v
-
onc

Virus


Protein product

symbol

strain

animal

origin

name

cellular

localization

src


Rous sarcoma virus

chicken

pp60
s
rc

plasma membrane

fps

Fujinami sarcoma
virus






P130
gag
-
fps







fes

Snyder
-

Theilen
feline
sarcoma virus

cat

P85
gag
-
f
e
s

H Il

yes


Yamaguchi ASV

chicken

P
9
O

gag
-
ye
s

1» Il

fgr

Gardner
-

Rasheed
feline SV

cat


P
7
O
gag
-
actin
-
fgr

Il Il

ros


Rochester
-
2 ASV

chicken

P
68

gag
-
ros

Il Il

Continued

The Oncogenes Are Diverse

25




Table I continued

v
-
onc

Virus


Protein product

symbol

strain

animal

origin

name

cellular

localization

abl

Abelson murine
leukemia virus

mouse

P90
-


P1
60

gag
-
abl


plasma membrane

ski

Sloan Kettering ASV

chicken

P
125

gag
-
erbA

nucleus

erbA

Avian
erythroblastosis virus





P
7
5

gag
-
ski
-
pol




erbB






gp65
er
b
B


plasma membrane

fms

Snyder I McDonough
feline sarcoma virus

c
at

gP
18
O
gag
-
fms

endoplasmatic

membranes

raf

MSV
-
3611

mouse


gP9
0
gag
-
raf
,


P
75
gag
-
raf

plasma membrane

mil
-

myc

Avian myelocyto
-

matosis and
carcinoma virus


chicken


P
100

gag
-
mil
-
myc

cytoplasm

myc

Avian myelocyto
-

matosis virus MC29






P
110

gag
-

myc

nucleus


Avian myelocyto
-

matosis virus MH2





p58
myc




myb

Avian myeloblas
tosis
virus (AMV)





p45
my
b




myb
-

ets


AMV
-
26





P
135
gag
-

myb
-
ets





fos

Finkel
-

Biskis
-

Jenkins MSV

mouse


pp55
fos




mos


Moloney MSV




p37
mo
s

cytoplasm

H
-
ras

Harvey MSV

rat

p
21
ras

plasma membrane

K
-
ras

Kirsten MSV

it

p
21
ras

“ “

sis

Simian sarcoma virus
(SSV)

monkey


P28
env
-
sis

secreted

rel

Reticuloendotheliosis
virus

turkey


p64
rel


cytoplasm 189],


? [1050]

kit

Hardy
-

Zuckerman
feline sarcoma virus

cat


P80
gag
-
kit

plasma membrane

jun

ASV
-
17

chicken

P65
ga
g
-
j
un

nucleus

26


The Oncogene Concept


The above table is to some extent selective and obviously
incomplete (this
equally applies for Table 2).

Avian myelocytomatosis virus Mill Hill 2 (MH2) carries two oncogenes,
v
-
myc

and
v
-
mil

(the latter also called
v
-
mht).

The complex protein product
P
100

gag
-
mil
-
myc

is localised in
the cytoplasm, while the produ
ct of MC29 (
P
110

gag
-

myc
) is
found in the nucleus [1050, 1174]. Obviously, the
mil

moiety of the
v
-
myc

complex fusion protein determines the cytoplasmic location of the protein
product; yet, some data suggest that the product of MN2 (
p5
7
myc
) also has a
cy
toplasmic localiz
a
tion where it binds to single
-
stranded RNA [1174].

Recently, oncogenes, their products and functions have been reviewed [67,
189,283,680,681,752
-
755,828]. Descriptions of particular oncogenes selected for
theoretical speculation are given

in Chapter 4. Here, however, attention is
focussed on more general, sweeping data to give an overview.

3.2.1.

CLASSIFICATION

A rather large group of oncogenes codes for protein
-
kinases with tyrosine
-
kinase activity, which are similar to pp60
src
. Another class of

oncoproteins
possesses protein
-
kinase activity of serine/threonine
-
specific character. There
are oncoproteins lacking any enzymatic activity. Similar high diversity is seen
in the localization of oncoproteins. This wide diversity led to the various
classi
fications of the oncogenes.

A
ll

of the classifications of oncogenes consider the properties of the encoded
oncoprotein. The following characteristics are taken into account: I) biochemical
properties of the oncoprotein,
2
) functional characteristics and 3)

intracellular
localization. Although the oncogenes are most often grouped using various
criteria, they will be classified here by individual characters for clarity.

Oncoproteins may be grouped by biochemical properties as follows: I)
tyrosine
-
specific pro
tein kinases,
2
) serine/threonine
-
specific protein kinases, 3)
GTP/GDP
-
binding proteins possessing GTPase activity, and 4) DNA
-
binding
proteins. Some oncogenes, f.ex. those coding the growth factors are not included
in this scheme.

Classification by functi
onal characteristics leads to the following groups
[141]: I) growth factors, 2) growth factor receptors, 3) transducers of growth
factor responses, and 4) transcription factors.

Using the site of intracellular localization, the oncogenes (their
oncoprotein
s) may be grouped as follows: I) secreted proteins,
2
) cell surface
receptors and other membrane
-
associated proteins, 3) cyto
plasmic proteins, and
4) nuclear proteins.

To obtain a more complete classification, the criteria are combined and
superimposed. In such cases, the number of groups reached seven and more.
However, the "unclassified'' group is unavoidable f452J. The oncogenes of the
ras

family are usually distingui
shed as a separate group, also called G
-
protein
-
like oncogenes, membrane
-
associated G
-
proteins, and GTP/GDP
-
binding proteins (217, 452J.

,

The Oncogenes Are Diverse

27


The observed functional diversity of oncogenes caused the resurfacing of
the cancerogenesis mechanism problem,
which

had already seemed resolved.

Since some of the oncogenes encode tyrosine
-
specific protein
-

kinases,
similar to
src
,

all c
-
onc

were initially postulated to arise from a single
precursor [432] and to have the same action mechanism as established for
the onco
gene
src
.

However, the data presented in Tables I and 2 refute this
explanation, and the universal molecular mechanism of cancerogenesis is
not recognized. Some speculative attempts have been made to unite the data
obtained. For ex., the various oncogenes
are perceived to encode proteins,
which results in a cascade of growth signalling pathways transmitting a
stimulus from the cell surface to the nucleus. Damage at any step of this
chain (its interruption), as well as excessive exaggerated stimulation, will

result in uncontrolled cell division, or, in other words, malignant growth
[141,1014]
.

However, this scheme lacks validity.
The oncogenes are not functionally
interconnected in a mutual chain of signal conductors which provide the
normal regulation of cel
l division, or any other chain.

This may be confirmed
by classifying oncogenes by their activity: I) oncogenes, whose expression is
necessary for a normal mitotic cycle of any cell; 2) oncogenes, whose activity
under normal conditions takes place only in p
articular categories of cells,
and 3) oncogenes that are active only during certain stages of ontogenesis.

3.2.2.

DISTRIBUTION IN LIVING ORGANISMS

The DNA loci homologous to oncogenes have been found in the genomes of
all animals studied. Initial
comparative
-
evolutionary research was
conducted on vertebrate species from reptilia to human [88]. The genome of
any of the species investigated was found to carry at least 10 oncogene
homologues. By enlarging the scope of investigated organisms and oncoge
nes
tested, this list is being continuously appended [687].

The genome of
Drosophila melanogaster

contains at least six homologues
of different oncogenes [490, 724]. The protein encoded by
c
-
src

of

Drosophila

possesses proteinkinase

activity similar to vertebrate pp60
src
[432]. Also,
pp60
src

related protein is expressed in nerve cells
The Oncogenes Are Diverse

28


of early metazoan hydra
[913].
Saccharomyces
c
erevisiae

contains the loci
RASl and RAS2, that are homologous to the oncogenes H
-
ras
-
1
and H
-
ras
-
2
of
mammals, as well as the homologue of the oncogene
ets
[231, 488]. The
protein that is encoded by the gene RAS of yeast has
a
structure similar to
that of ver
te
brates [742, 823].
Missing

of RAS
-

1 and RAS
-
2 causes a lethal
effect in yeasts, which may be
re
covered

by
the
human H
-
ras gene
. In turn,
yeast RAS may induce morphological transformation of mammal cells [1043].

Computer analysis showed that the primary structure of the gene
CDC28, that controls division in yeast cells, was homologous with the fami
ly
of protein
-
kinase
oncogenes
(mos, src, fes, yes, raf).

The genomes of some
fungi possess sequences homologous to oncogenes
ras, mos,
and
abl

[824].

Substantial evidence indicates that c
-
one represents a gro
up of
evolutionary conservative
genes
, which

appeared at the very early stages of
evolution and did not undergo any major change during a period of about 800
million years. This fact suggests that
the main role of cellular oncogenes
i
s
the accomplishment of some substantial, ancient and fundamental
functions
of the cell and not just induction of cancerogenesis

[
199, 1113].

In view of these ideas, the

subsequent discovery of cellular oncogenes
that are not virus homologues was not surprising [461]. A partial list of these
c
-
onc

is given in Table 2. Ho
wever, the determination of criteria by which a
gene may be attributed to the oncogenes become
s

a new problem. A possible
methodical definition is that "oncogenes are genes identified by their
biological activity in transfection assay and/ or by homology t
o the
transforming genes of retroviruses


[190].

Homology to v
-
o
nc

is evidently unreliable. The biological activity in
transfection is also invalid, since many of the oncogenes are not able to
induce transformation [190, 1109]. Thus, the category of cellul
ar oncogenes
that lack their virus counterparts is defined very uncertainly. Some
representatives of this class are clearly oncogenes. For example, L
-
myc or
N
-
ras

are related to the families of myc and
ras
,

respectively, by all general
criteria. The situation appears more complicated for the oncogene
trc,

involved in the incidence of human colon carcinoma and thyroid papillary
carcinoma [48, 612]. The
trc

gene group involves about 40 somewhat different
oncoge
nes belonging to the tyrosine
-
kinase superfamily (
src, ret, ros
,

and
others). Tyrosine
-
kinase genes also encode for the receptor of insulin,
epidermal growth factor (EGF), platelet
-
derived growth factor (PDGF)
(chains
A
and
B)
and fibroblast growth factor
(FGF). The
trc

gene codes for
the receptors of neurotropic factors [48]. Somatic rearrangement of the
trc

gene leading to replacement of the ligand
-
binding domain of gpl40
trc

by
tropomyosin results in the generation of a highly transforming gene.


CELLULAR

ONCOGENES
[108,
253,
372, 1050]

The Oncogenes Are Diverse

29



c
-
onc

Protein product

symbol

cellular localization

biochemical properties

c
-
onc that are
the
homologues

of

viral oncogene

src

plasma

membrane tyrosine
-
specific
PK

fps

it


fes

it


yes

ti


fgr

II


ros

II


abl

ti


ski

nucleus

DNA
-
binding protein

erbA

it

DNA
-
binding thyreoid
-

hormone receptor

erbB

(
neu
)

plasma membrane

EGF receptor

fms

cytoplasm

tyrosine
-
specific PK

raf

plasma membrane

serine/threonine PK

mil

II

II

myc

nucleus

DNA
-
binding protein

myb

II

II

fos

it

binds
the
jun

protein

ets

it

DNA
-
binding protein

mos

cytoplasm

serine/threonine PK

H
-
ras

plasma membrane

GTP/GDP
-
binding GTPase

K
-
ras

II

II

sis

secreted

PDGF
-

B
-
chain

rel

cytoplasm

?

kit

plasma membrane

tyrosine
-
specific PK

jun

nucleus

complexes with the
fos

protein

The Oncogene Concept

Table 2



int
-
1

and
int
-
2

represent the integration locus for mouse mammary
tumour virus [253, 1050]. No oncogenic potential has yet been
demonstrated for the activated form of these genes
per se.

The genes
bcr

and
bcl

are the breakpoint regions of chromosomes in ce
lls of some
leukemias, and lack any oncogenic potential in transfection assay [1050].
A hypothetical oncogene mapping in the
bcl

locus is
prad
I [876]. Some
authors include one of the homeobox genes, /fox
-
2.4, as well as the gene
coding for interleukin 3 (a

multipotential colony stimulating factor), in
the oncogene group [141]. Coexpression of
Hox
-

2.4 (which apparently
blocks myeloid differentiation) and IL
-
3 (which provides the necessary
autocrine growth stimulus for myeloid cells) generates murine myeloid

leukemias [523, 795].

The rat insulinoma gene
(rig)

was found to code for a highly
conserved nucleoprotein in various tumours, but the possible designation
of the
rig
-
gene as an oncogene was not discussed [951].
Continued

c
-

one

Protein product

symbol

cellular localization

biochemical properties

c
-
onc

lacking a corresponding v
-
onc

N
-
ras

as H
-

and K
-
ras


L
-
myc

as c
-
myc


N
-
myc

as c
-
myc


met

cytoplasm

tyrosine
-
specific PK or
hepatocyte GF

hst

?

FGF
-
related protein

ret

?

?

mas

plasma membrane

angiotensin receptor (lacking
PK activity)

B
-
Iym

?

?

Ick

plasma membrane

tyrosine
-
specific PK

bcl

plasma membrane

tyrosine
-
specific PK G
-
protein

trc

cytoplasm

II

pim
-
1

•I

serine/threonine PK

int
-
1

secreted

glycoprotein

int
-
2

?

FGF
-
related protein

31

The Oncogene Concept


3.2.3.

LOCATION ON HUMAN
CHROMOSOMES


The position on human chromosomes, as well as on murine (325]
and
rat [1023] chromosomes, has been established for nearly all the known
oncogenes (Table 3).

Table 3

ONCOGENE LOCATION ON HUMAN CHROMOSOMES
[513, 882,
965, 1070].

Chromosome

Oncogen
e

Chromosome

Oncogene

Number

Segment

Numbe
r

Segment

I.

pl3

N
-
ras

10
.

qll
.2

ret


p31
-
p32

jun

11
.

pl5.5

H
-
ras
-
1


p32

B
-
Iym


pl3.3

hst


p32

L
-
myc


ql3

int
-
2


p32
-
p35

Ick


ql3

sea


p34

src
-
2


ql3

bcl
-
1


p36.2
-
p36.1

fgr


q23
-
q24

ets
-
1


q21.1
-
q24

ski

12
.

P12.1

K
-
ras
-
2


q32
-
q41

trc


ql3

int
-
1

2
.

p24

N
-
myc

13.


-



pl3
-
cen

rel

14.

q24.3

fos

3.

p21
-
p25

erbA
-
2

15.

q24
-
q25

fps


p25

raf
-
1


q26

fes

4.

P16.1

raf
-

2

16.

-



qll
-
q
22

kit

17.

qll
-
ql
2

erbB
-
2

5.

q34

fms


q
21
-
q
22

erb

A
-
I

6
.

p23
-
cl2

K
-
ras
-
I

18.

q21.3

yes


p
21

pim


q21.3

bcl
-
2


q
21
-
q
22

ros
-
1

19.

?

bcl
-
3


q22
-
q24

myb

20
.

qll
-
ql
2

Iok


q24
-
q27

mas


qll
-
ql
2

hck

7.

pl4
-
p21

A
-
raf
-
2


ql2
-
ql3

src
-
1


pter
-
q
22

erbB
-
l

21
.

q22.3

ets
-
2(erg)


q31
-
q32

met

22
.

qll
-
ql
2

yes

8
.

ql3
-
qter

Iyn


ql2.3
-
ql3.1

sis


q21
-
q23

mos


?

bcr

9.

q24.13

myc

X

pter
-
q26

H
-
ras
-
2

q34

abl


pll.4
-
pll.2

Araf
-
I

32


The Oncogene Concept


Table 3 indicates that the oncogenes are scattered throughout the
entire genome and located in both autosomes and the X chromosome.
As of yet, oncogenes have not

been found in only two human chromo
-
somes (13 and 16). Some of the oncogenes have two locuses in different
chromosomes. In such cases they are designated as onc
-
1 and
onc
-
2.

On the basis of analysis of the oncogene distribution within the
genome, Lima
-
de
Faria and Mitelman [577] state the preferential
location of most of the human oncogenes in the telomere region and
classify them as telons

(with the exception of

er
b
-
B and N
-
ras).

However, the material considered in this analysis includes only 20
oncogenes

located in 15 human chromosomes. Table 3 lists other
oncogenes, such as
kit, ret, hck, Araf
-
1,
yes,

which cannot be
designated as telons.

3.2.4.

MECHANISMS OF ACTIVATION

During the early period of the study of oncogenes, cancerogenesis was
considered to be a result of insertion of a retroviral oncogene into the
cell. After it was established that every cell contains the set of the
oncogenes,
the mechanism of cancerogenesis

was regarded as oncogene
activation.

Due to the variety in the mechanisms of activation,
revealed

and
suggested, these mechanisms have been organized into groups based
on this character (301, 451,849]. The term "activity of the oncogene


differs in meanin
g (see Chapter I), but, most often, refers to the ability
to cause cell transformation by the gene. Generally, all the methods of
oncogene activation may be defined as either quantitative or
qualitative.

Quantitative activation occurs by the increase in th
e concentration
of the relevant gene product without alteration of the coding sequences
of the gene. Qualitative activation is the alteration of the oncogene by
a mutation in the coding sequence. Various mechanisms act in these
processes, for each of these

two categories [736,834,850]. In some