Oncogenes and Retroviruses: Selected Reviews

138 Pages·1989·6.041 MB·English

by  David Kabat (auth.), Peter K. Vogt Ph. D. (eds.)

Current Topics in
Microbiology
148 and Immunology
Editors
R W. Compans, Birmingham/Alabama . M. Cooper,
Birmingham/Alabama . H. Koprowski, Philadelphia
I. McConnell, Edinburgh . F. Melchers, Basel
Y.Nussenzweig, NewY ork . M.Oldstone,
La Jolla/California . S. OIsnes, Oslo . H. Saedler,
Cologne . P. K. Vogt, Los Angeles· H. Wagner, Ulm
I. Wilson, La lolla/California
Oncogenes and
Retroviruses
Selected Reviews
Edited by P. K. Vogt
With 7 Figures
Springer-Verlag
Berlin Heidelberg NewY ork
London Paris Tokyo Hong Kong
PETER K. VOGT, Ph. D.
Department of Microbiology,
University of Southern California,
School of Medicine
2011 Zonal Avenue HMR-40l
Los Angeles, CA 90033-1054
USA
ISBN-13: 978-3-642-74702-1 e-ISBN-13: 978-3-642-74700-7
DOl: 10.1007/978-3-642-74700-7
This work is subject to copyright. All rights are reserved, whether the whole or part of
the material is concerned, specifically the rights of translation, reprinting, re-use of
illustrations, recitation, broadcasting, reproduction on'microfilms or in other ways,
and storage in data banks. Duplication of this pUblication or parts thereof is only per
mitted under the provisions of the German Copyright Law of September 9, 1965, in
its version of June 24, 1985, and a copyright fee must always be paid. Violations fall
under the prosecution act of the German Copyright Law.
© Springer-Verlag Berlin Heidelberg 1989
Softcover reprint of the hardcover 1st edition 1989
Library of Congress Catalog Card Number 15-12910
The use of registered names, trademarks, etc. in this publication does not imply,
even in the absence of a specific statement, that such names are exempt from the rele
vant protective laws and regulations and therefore free for general use.
Product Liability; The publishers can give nO guarantee for information about drug
dosage and application thereof contained in this book. In every individual case the
respective user must check its accuracy by consulting other pharmaceutical
literature.
2123/3020-543210 - Printed on acid-free paper
Table of Contents
D. KABAT: Molecular Biology of Friend Viral
Erythroleukemia. . . . . . . . . . . .
S. PALMIERI: Oncogene Requirements for Tumorigenicity:
Co-operative Effects Between Retroviral Oncogenes. . 43
S. SUKUMAR: ras Oncogenes in Chemical Carcinogenesis 93
E. LARSSON, N. KATO, and M. COHEN: Human Endogenous
Pro-viruses 115
Subject Index 133
Indexed in Current Contents
List of Contributors
You will find their addresses at the beginning of the respective
contributions
M. COHEN E. LARSSON
K. KABAT S. PALMIERI
N. KATO S. SUKUMAR
Molecular Biology of Friend Viral Erythroleukemia
DAVID KABAT
1 Introduction 1
2 Normal Hematopoiesis, and Its Control by Host Genes That also Modulate Susceptibility to
Friend Virus 3
3 Early Description of Friend Virus Disease 5
4 A Component of the Friend Virus Complex: F-MuLV 7
4.1 Host-Range Classes of MuLV, and Their Corresponding gp70s and Cell Surface Receptors 7
4.2 How Do Retroviral env Glycoproteins Mediate Viral Entry into Cells?
4.3 Diseases Caused by Different Host-Range Classes of MuLVs 11
4.4 Pathogenic Roles ofF-MuLVand F-MCF, and Structures of Their env Glycoproteins 13
5 SFFV 13
5.1 Sequence Comparisons of SFFVs and of Their env Glycoproteins 13
5.2 gp55 as an Erythroblast Mitogen 17
5.3 Outline of gp55 Structure and Processing 19
5.4 Inefficient Processing and Heterogeneous Disulfide Bonding of gp55 21
5.5 Possible Mechanisms of gp55 Mitogenic Function 23
6 Studies of the Viral LTR-U3 Regions 24
7 Studies of Helper-Free SFFV 27
8 Common Sites for Proviral Integration and Other Aspects of Leukemic Progression 29
9 Summary and Conclusions 30
References 31
1 Introduction
Friend virus induces rapid progressive erythroleukemia in susceptible mice. To
oncologists and to cell biologists, this disease has provided a fascinating model for
analyzing neoplastic progression, the role of host genes in controlling susceptibility
to cancer, and the differentiation of erythroid cells in culture. However, to molecular
biologists, Friend virus (and the closely related Rauscher and Cas virus complexes)
has been generally viewed as a relatively complex anomaly. The virus lacks a classical
oncogene of the sort typified by the sre gene of Rous sarcoma virus. Such classical
viral oncogenes (v-ones) are modified versions of normal cellular genes (proto
oncogenes or c-ones) and they are present in all of the other known retroviruses
that cause rapidly developing neoplasms (BISHOP 1983, 1985; WEINBERG 1985).
The c-ones have been highly conserved throughout evolution, and there is evidence
Department of Biochemistry, School of Medicine, Oregon Health Sciences University, Portland,
OR 97201, USA
Current Topics in Microbiology and Immunology, Vol. 148
© Springer-Verlag Berlin' Heidelberg 1989
2 D. Kabat
that they perform important cellular functions. Because Friend virus lacks such a modi
fied cellular gene and contains only retroviral-specific nucleic acid sequences, and
because it induces a progressively developing neoplasm rather than an immediate
cancer, it has been widely assumed that it is relatively "different" and "complex,"
perhaps too different to provide general insights and too complex for molecular
biological analysis.
In the last few years these distinctions have eroded. First, it has been learned
that neoplasms formed by oncogene-containing retroviruses may also be progressive
(LAND et al. 1983; WHITLOCK and WITTE 1985). Indeed, cancer seems to require
more than one genetic change (FOULDS 1975; LAND et al. 1983; WEINBERG 1985).
It has been proposed that one essential change may enable the cell to grow
autonomously of normal regulatory signals, and that a second change may lead to
immortality, an ability of the cell to grow forever (LAND et al. 1983). However,
neither of these putative changes has been unambiguously defined or is understood
at the molecular level. Several retroviruses even contain two v-ones that collaborate
to cause full transformation of the infected cell (OSTERTAG et al. 1987; BISHOP 1983,
1985; LAND et al. 1983). Therefore, viruses with v-ones can no longer be assumed
to cause simpler diseases. Secondly, direct studies of Friend disease have shown
it to be relevant to leukemia in general and to many of the horizontally trans
mitted retroviral diseases in particular. For example, Friend virus encodes an
envelope-related membrane glycoprotein (gp55) that causes erythroblast mitosis
when it is expressed on the surfaces of infected cells (LI et al. 1987 a). Recent
evidence suggests that the related envelope (env) genes of other retroviruses [e.g.,
human immunodeficiency virus (HIV), feline leukemia virus (FeLV) subgroups
Band C, FeLV-FAIDS, avian leukosis virus subgroup F, polytropic murine
leukemia viruses (MuLVs)] can cause immunosuppression, aplastic anemia, prolifera
tive diseases including lymphomas and angiosarcomas, and central nervous system
degeneration (HEBEBRAND et al. 1977; MATHES et al. 1978; FAMULARI 1983;
SIMON et al. 1984, 1987; SNYDERMAN and CIANCIOLO 1984; WEISS et al. 1985;
LIFSON et al. 1986; SITBON et al. 1986; SODROSKI et al. 1986; KLEINERMAN et al. 1987;
MITANI et al. 1987; SCHMIDT et al. 1987; KLASE et al. 1988; OVERBAUGH et al.
1988a, b; RIEDEL et al. 1988; SZUREK et al. 1988). Consequently, env gene-mediated
pathogenesis can no longer be viewed as an anomaly.
In addition, new evidence has indicated that Friend virus may actually cause
leukemia by a simple two-step mechanism. In the first step, gp55 causes mitogenic
activation of infected erythroblasts without abrogating the cells' commitment to
terminally differentiate. The consequence is a polyclonal proliferation of infected
erythrob1asts that have only a limited self-renewal capability. At this first stage,
the disease can be maintained by continuous viral replication and infection of new
erythroblasts. The second stage is caused by rare single proviral integrations that
cause the infected cells to become immortal. This immortality appears to be associated
with abrogation of the cell's commitment to differentiate (SPIRO et al. 1989).
Thus, for Friend disease the mitogenic and immortalizing stages of cancer are
conveniently separated into two discrete genetic changes that can now be readily
analyzed.
The gp55 glycoprotein is also processed only inefficiently from the rough endo
plasmic reticulum (RER) to the plasma membranes. Recent evidence has implied
Molecular Biology of Friend Viral Erythroleukemia 3
that newly made gp55 folds heterogeneously into different disulfide-bonded forms,
and that only one homodimeric component is competent for export from the RER
(GLINIAK and KABAT 1989). This Friend viral glycoprotein provides an excellent
model for analyzing the factors that control the sorting of proteins in the
RER and the formation and isomerization of disulfide bonds.
My purpose here is to review recent work on the molecular biology of Friend
virus, and (hopefully) to explain its excitement and relevance to workers in other
fields. In addition, I will briefly review earlier evidence that provides a necessary
background to the current work. Excellent previous reviews should be consulted for
detailed descriptions of Friend disease (TEICH et al. 1982; RUSCETTI and WOLFF
1984; FRIEND and POGO 1985; OSTERTAG et al. 1987), the gp55 glycoprotein
(RUSCETTI and WOLFF 1984; PINTER 1988), murine retroviral env genes (FAMULARI
1983; PINTER 1988), normal hematopoiesis (e.g., TILL and MCCULLOCH 1980;
BURGESS and NICOLA 1983; METCALF 1984; CLARK and KAMEN 1987; OSTERTAG
et al. 1987), and mouse genes that control susceptibility to leukemia (STEEVES
and LILLY 1977; RUSSELL 1979; TEICH et al. 1982).
2 Normal Hematopoiesis, and Its Control by Host Genes
That Also Modulate Susceptibility to Friend Virus
Normal hematopoiesis involves a hierarchy of stem cell pools that derive from
each other and that have limited self-renewal capabilities. Although more primitive
stem cells are multipotential and can form progeny that differentiate along more
than one pathway, more mature stem cells are committed to differentiate along
<
single pathways (see OSTERTAG et al. 1987; TEICH et al. 1982; CLARK and KAMEN
1987). The proliferation and differentiation of the hematopoietic stem cells are
regulated by a complex network of both humoral and microenvironmental factors
(BURGESS and NICOLA 1983; METCALF 1984; CLARK and KAMEN 1987; OSTERTAG
et al. 1987). In general, the humoral factors bind to receptors on surfaces of
specific stem cells, and this reception leads to proliferation of the cells and to
alterations in their responsiveness to other factors (i.e., to induction of certain
receptors and to downregulation of others) (WALKER et al. 1985). In addition, the
proliferation can continue only while the factor remains present and only until the
self-renewal capability of the stem cells becomes exhausted, at which time the
cells enter the next pool in the pathway. The lineage that produces erythroid
cells occurs in the sequence: pluripotent progenitor (CFU-S or CFU-MIX) ---> burst
forming unit-erythroid (BFU-E) ---> colony-forming unit-erythroid (CFU-E) ---> termi
nal differentiation via the erythrocyte series to form erythrocytes.
When normal bone marrow cells are cultured in the presence of erythropoietin
(Epo, the major regulator of the erythroid lineage), small colonies grow for 2-3
days before terminally differentiating to form erythrocytes (STEPHENSON et al. 1971;
EAVES et al. 1979; NICOLA et al. 1981). Such small colonies (containing usually
20-60 cells) are seeded by CFU-E that can proliferate only briefly before they
express their commitment to differentiate. Epo is required not only to induce
CFU-E proliferation but also throughout the process of terminal differentiation
(HANKINS and TROXLER 1980). The differentiated colonies contain no remaining stem
4 D. Kabat
cells; and, because they lack self-renewal capability, they disappear from the
cultures within several days of their formation. However, between 5-8 days of
culturing with Epo, large clusters of hemoglobinized colonies appear (AxELROD
et al. 1974; GREGORY 1976; EAVES et al. 1979). These clusters are called "bursts"
and it is believed that they are seeded by BFU-E that proliferate before differentiating
into CFU-E. Similarly, these clusters of CFU-E then proliferate briefly before they
differentiate to form bursts of hem oglobiniz ed descendent colonies. Like the colonies
founded by CFU-E, the bursts derived from BFU-E also lack cells capable of
self-renewal. Thus, at least for the committed erythroid progenitors, each stem
cell proliferates until its self-renewal capability becomes exhausted, at which time
the colony synchronously expresses its commitment to move down the pathway.
Accordingly, the colonies founded by BFU-E and CFU-E have characteristic
sizes and times of hemoglobinization. Moreover, because BFU-E recovered from
normal bone marrows are heterogeneous in the proportion of their self-renewal
capacity that was used in vivo, the bursts that form on day 5 in culture with
Epo are believed to derive from more mature "late BFU-E," whereas those that
appear later are larger and derived from "early BFU -E" (HEALTH et al. 1976; GREGORY
and EAVES 1978; JOHNSON and METCALF 1978; EAVES et al. 1979; NICOLA et al.
1981). .
Studies using serum-free medium have established that Epo alone is insufficient
for causing proliferation of BFU-E (ISCOVE 1978). A "burst promoting activity"
(BPA) is also required. Interleukin 3 (IL-3) and granulocyte-macrophage colony
stimulating factor (GM-CSF) appear to have BPA (ISCOVE 1978; METCALF et al.
1980; METCALF and NICOLA 1983). Both of these factors can also stimulate
other lineages of early hematopoietic progenitors. Moreover, hemoglobinized bursts
that form in cultures in the presence of both BPA and Epo continue to appear
for up to 21 days, substantially longer than in Epo alone (JOHNSON and METCALF
1978; OSTERTAG et al. 1987). As expected, these colonies are also larger (up
to 20000 cells). These results have implied that some of the earliest BFU-E
are responsive to BPA but not to Epo. These Epo-nonresponsive, committed
erythroid BFU-E also have different sizes and membrane properties than the late
BFU-E (HE~TH et al. 1976; JOHNSON and METCALF 1978; EAVES et al. 1979;
KOST et al. 1981; NICOLA et al. 1981). Therefore, the BFU-E are a dynamic and
heterogeneous population of developing committed cells.
The role of the microenvironment in hematopoiesis has been established using
specific mutant mice (STEEVES and LILLY 1977; TEICH et al. 1982). For example,
steel (SI/SId) mice have an inherited anemia that seems to be caused by a micro
environmental deficiency. This anemia cannot be rapaired by transplantation of
bone marrow cells from normal isogenic donors (MCCULLOCH et al. 1956;
RUSSELL 1979). However, transplants from SI/Sld mice can reconstitute normal
hematopoiesis in X -irradiated normal recipients. Other genes that control hematopoie
sis and immune responsiveness also often affect susceptibility of mice to Friend
viral erythroleukemia (STEEVES and LILLY 1977; TEICH et al. 1982; GEISSLER et al.
1988; CHABOT et al. 1988). These host susceptibility genes for Friend virus provide
a fruitful approach not only for characterizing the primary events in leukemic trans
formation, but also for understanding the interactions of normal and malignant
cells with their regulatory microenvironment.
Molecular Biology of Friend Viral Eryhtroleukemia 5
3 Early Description of Friend Virus Disease
In 1957, Charlotte FRIEND described a C-type retrovirus that caused susceptible
mice to develop a rapid-onset leukemia of hyperbasophilic blast-like cells that was
associated with splenomegaly, hepatomegaly, and anemia. Because it causes anemia,
this original Friend virus isolate is called FV-A. Subsequently, variant viruses were
isolated in different laboratories that were somewhat more pathogenic and that
caused polycythemia (erythrocytosis) rather than anemia (see TEICH et al. 1982;
RUSCETTI and WOLFF 1984; OSTERTAG et al. 1987). Because of this polycythemia,
these variant strains are called FV-P. Subsequently, other viruses have been isolated
that are similar in their molecular and pathogenic properties to FV-A (RAUSCHER
1962; BARBIERI-WEILL et al. 1983; LANGDON et al. 1983a, b). These include the
Rauscher erythroleukemia and Cas virus complexes.
In the years following these discoveries, the basic pathology of Friend disease
was described (see LEVY et al. 1979; TAMBOURIN et al. 1981; TEICH et al. 1982; Rus
CETTI and WOLF 1984; OSTERTAG et al. 1987). Briefly, the primary target cells for
pathogenesis by Friend virus are committed erythroblasts consisting of CFU-E
and late BFU-E (KOST et al. 1979, 1981; HANKINS and TROXLER 1980; PESCHLE
et al. 1980). It .is noteworthy that these are the Epo-responsive erythroblasts.
These cells, but not the earlier BFU-Es, are also engaged in rapid mitotic cycling in
vivo (GREGORY et al. 1973; EAVES et al. 1979), a condition conducive to successful
retroviral infection (TEMIN 1967; HOBOM-SCHNEGG et al. 1970). Accordingly, the
target cell population is increased in mice made anemic by bleeding or by
hemolytic chemicals (e.g., phenylhydrazine), whereas it is decreased in mice that have
a plethora of red cells due to chronic transfusions (KqST et al. 1979, 1981).
Within 1-2 days of injection with Friend virus, these infected target erythroblasts
begin to proliferate, to migrate from the bone marrow, and to sequester in the spleen
(TAMBOURIN et al. 1981). By 9-10 days postinfection, macroscopic foci of proliferating
erythroid cells can be easily seen on the surfaces of spleens fixed in Bouin's
fixative (AXELROD and STEEVES 1964). The number of foci is a measure of virus titer.
By 14-21 days, the spleens become extremely enlarged (2-2.5 g compared with 0.13 g
for normal) (ROWE and BRODSKY 1959). Often, the mice die of splenic rupture
during this time. However, more prolonged survival may occur if the virus is injected
in a low dose, if the FV-A strain is used, or if the mice are older than several
months at the time of injection. Recovery is rare. However, there is one virus
variant from which mice recover after 21 days (DIETZ et al. 1977; MARCELLETI and
FURMANSKI 1979). Mice with certain genotypes can also partially recover (TEICH
et al. 1982; MORRISON et al. 1986).
Because the proliferating erythroid cells continue to release virus, their trans
plant ability has been difficult to analyze. This difficult occurs because the virus from
the transplant can cause erythroblastosis of host origin. Consequently, such studies
have generally involved use of host mice that are genetically resistant to Friend virus
(TEICH et al. 1982), or that have been X-irradiated to reduce their target cell
population (MAGER et al. 1980; EENDLING et al. 1981). Although complex, such
studies have suggested that Friend disease is a progressive leukemia. Within the first
4-6 weeks of infection almost all of the proliferating cells are non transplantable,
suggesting that they have a limited self-renewal capability (WENDLING et al. 1981).