[ LIBRARV
Michigan 92L
University

V__.__

 

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

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

LT—

MSU Is An Affirmative Action/Equal Opportunity Institution
czbmmpma-pn

 

 

 

 

 

\a

wrigg-

I’|--

‘vfivo

THE ROLE OF MIDDLE T ANTIGEN MEDIATED
ENHANCER ACTIVITY IN POLYOMA VIRAL

DNA REPLICATION AND NEOPLASTIC TRANSFORMATION

BY

Ming Chu Chen

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Microbiology and Public Health

1991

(it

ABSTRACT

THE ROLE OF MIDDLE T ANTIGEN MEDIATED
ENHANCER ACTIVITY IN POLYOMA VIRAL

DNA REPLICATION AND NEOPLASTIC TRANSFORMATION

BY

Ming Chu Chen

Effects of polyomavirus enhancer on viral DNA synthesis
and neoplastic transformation were investigated.
Competition assays indicate that NIH-3T3 mouse cells contain
limiting factors for polyoma viral DNA replication and that
target sequences for these factors are located within the A
enhancer element. In the absence of middle T antigen, virus
with a duplicated A enhancer element displayed growth
advantage over the wild-type A2 enhancer. Enhancer
competition was alleviated by middle T antigen, and
partially by serum growth factors and/or TPA. This suggests
that middle T antigen affect enhancer activity through a
signal transduction or a transformation pathway. Given that
all natural hr-t isolates have evolved with rearrangements

in the same region of the enhancer, we propose it as a

compensatory event for the loss of middle T and/or small T
antigens.

Enhancer mediated competition was also observed in
nonpermissive FR-3T3 cells, suggesting that rat cells
contain limiting enhancer factors. Wild-type virus induced
stable transformation was interfered more pronounced than
viral DNA replication and abortive transformation. Taken
with the observation of low probability for double
integration, we propose that integration is a rate limiting
step during the process of polyomavirus mediated neoplastic
transformation. It can be either due to the existence of
limiting amounts of one or several factors required for
integration step, or due to the limitation of integration
events in the host chromosome. Since both viruses produce
normal large T antigen, which is required for integration,
it is likely that integration of one viral genome prevents a

secondary event.

To my parents

ii

ACKNOWLEDGEMENTS

I would express my deep appreciation and gratitude to
Dr. Michele Fluck for her encouragement, invaluable
discussions, and financial support throughout my graduate
program.

I thank the members of my guidance committee for their
advice and interest: Drs. Susan Conrad, Larry Snyder, and
John Wang.

I would like to acknowledge Drs. Karen Friderici and
Andrea Amalfitano who initiated studying enhancer sequences
of hr-t mutants, Makda Fisseha for assistance with data
presented in Chapters 2 and 3. I also thank the people in
Dr. Fluck's laboratory for their help and shared ideas.

Finally, my special thanks go to my family for their
love and constant support, my fiance' Dr. Perng Kuang Chang
for his friendship with great understanding and patience.

iii

TABLE OF CONTENTS

List of Tables . . . . . . . . . . . . . . . . . . . . vi

List of Figures . . . . . . . . . . . . . . . . . . . vii
Chapter 1: Literature Review
Introduction . . . . . . . . . . . . 1
Structural Organization of the Polyomavirus
Genome . . . . . . . . . . . . . . . . . . 4
Structure and Function of the Polyomavirus
Enhancer . . . . . . 11

Factors which Interact with the Polyoma
Enhancer . . . . . . . . . , . . . . . . . 13
Altered Enhancer Regions in Polyomavirus
Host-range Mutants . . . . . .'. . 17
Gig-acting Sequences which Control the Replication
of Polyomavirus DNA . . . . . . . . . . . . . 21

Mechanisms of Cell Transformation by Polyomavirus 3O
sman I O I O O O O O O O O O O O O O O O O O O 4 5
References . . . . . . . . . . . . . . . . . . . . 48

Chapter 2: DNA Replication of Polyoma Variants with
- Altered Enhancers

Abstract . . . . . . . . . . . . . . . . . . . . . 60
Introduction . . . . . . . . . . . . . . . . . 61
Materials and Methods . . . . . . . . . . . . . . 63
Results . . . . . . . . . . . . . . . . . . . . . 67
Discussion . . . . . . . . . . . . . . . . . . . . 82
References . . . . . . . . . . . . . . . . . . . . 89

Chapter 3: Functional Interdependence between Middle
T Antigen and the Polyoma A Enhancer Region
Abstract 0 O O O O O O O O O O O O O 9 4

Introduction . . . -. . . . . . . . . . . . . . 96
Materials and Methods . . . . . . . . . . . . . . 99
Results . . . . . . . . . . . . . . . . . . . . . 104
Discussion . . . . . . . . . . . . . . . . . . . . 122
References . . . . . . . . . . . . . . . . . . . . 129

iv

Chapter 4: The Role of the Polyomavirus Enhancer in
Neoplastic Transformation
Abstract . . . . .

References .

O O O O O O O 135
O O O O O O O O O O 155

Summary and Conclusions .

O O 0 O 159

LIST OF TABLES

Chapter 2: Table 1. Yield of infectious virus from
mixed infections . . . . . . . . . . . . 81

Chapter 4. Table 1. Integration event in

polyomavirus transformed FR-3T3 cell
lines derived from mixed infections . 146

vi

Chapter 1:

Chapter 2.

Chapter 3.

LIST OF FIGURES

Figure 1. Physical map of polyomavirus
genome . . . . . . . . . . . . . . . .
Figure 2. Physical map of the noncoding
region of polyomavirus . . . . . .
Figure 3. Cellular factors which bind
to the enhancer region of polyomavirus
Figure 4. Schematic representation of
phosphorylation sites and functional
domains in the polyomavirus large T
antigen . . . . . . . . . . . .
Figure 5. Model for the transforming
action of polyomavirus middle T
antigen . . . . . . . . . . . . . . . . .

Figure 1. Replication of A2 virus in
competition with deletion mutant 18-5 .
Figure 2. Competition assay in the mixed
infections with various wild-types and
deletion mutant 82 . . . . . . . .
Figure 3. Effect of hr-t mutants on A2
replication . . . . . . . .
Figure 4. Competition for DNA synthesis
between middle T antigen-deficient
mutants . . . . . . . . . . .
Figure 5. Kinetics of viral DNA
synthesis . . . . . . .
Figure 6. Accumulation of viral DNA in
virions and whole lysate . . . . . . .

Figure 1. Effect of hr-t mutant on A2
viral DNA synthesis . . . . . . . . .
Figure 2. Comparison of enhancer-origin
sequence of hr-t mutants . . . .
Figure 3. Effect of reiterated A enhancer
element on viral DNA synthesis . . .
Figure 4. Effect of middle T antigen on
viral DNA synthesis . . . . . . .
Figure 5. TPA responsiveness of viral DNA
replication . . . . . . . . . .
Figure 6. Effect of TPA, serum or both. on
the competition assays . . . . . . . .

vii

12

27

38

69

71

73

75
77

80

105
108
112
115
117

119

Chapter 4.

 

 

Figure 7. Viral DNA synthesis in BMK

cells . . . . . . . . . . . . . . . .
Figure 8. Mixed infections of BMK cells

Figure 1. Viral DNA replication in mixed
infection of FR-3T3 cells . . . . . .

Figure 2. Transformation in mixed
infections of FR-3T3 cells . . . . .

Figure 3. DNA analysis of transformants
derived from mixed infection with two
transforming strains . . . . . . . .

viii

 

121
123

140

142

147

Chapter 1.

Literature Review

Introduction

The mouse polyomavirus is a member of the papovavirus
family (1). Its genome is a double-stranded, covalently-
closed circular DNA molecule consisting of 5297 base pairs
(2). Viral DNA exists in virion as a minichromosome
complexed with cellular histones surrounded by capsid
proteins. The structure and histone composition of the
viral nucleosomes are indistinguishable from those of the
host (3). The polyoma virion is approximately 45 nM in
diameter and is composed of three virus-coded proteins,-VP1,
VP2, and VP3, forming a icosahedral structure. VP1 is the
major component of the viral capsid, making up about 75% of
the total virion protein (1).

Polyomavirus was first discovered in 1953, and was
later shown to cause a wide variety of solid tumors in mice
and other rodents (4,5). Although the virus is prevalent,
the incidence of tumors in the natural population is quite
low. This has been attributed to the immune protection of
the host acquired during gestation, low doses of infecting
virus, and to the existence of an anti-tumor response. The
mechanism for such immunological resistance to tumor

formation provides an interesting field to be explored.

2

In tissue culture systems, a productive infection, also
defined as lytic infection, takes place in mouse fibroblast
cells. The vast majority of these cells, designated as
permissive cells, support viral DNA replication, express all
viral gene products, and eventually are lysed and release
infectious progeny virus particles. To find a tumor
induction process in tissue culture equivalent to that in
mouse, other hosts were tested. Hamster and rat cells are
semi- and non-permissive, respectively, meaning that viral
infection produces much less or no viral particles. Such
permissivity seems to be determined by one or more trans-
dominant factors of the host, since hybrids from permissive
and nonpermissive cells are productively infected (6).
Infection of nonpermissive cells results in cellular
immortalization and/or neoplastic transformation.

0n infection of permissive cells, the virion adsorbs to
the cell membrane, penetrates into the nucleus, and then is
uncoated (7). The interaction between the virion and its
membrane receptor induces the expression of two cellular
oncogenes, own and c-fos, which are important for both
viral DNA synthesis and gene expression (8). Three early
genes coding for proteins referred to as T antigens are
' expressed. These proteins are involved in stimulating
expression of cellular genes and synthesis of cellular DNA,

and in facilitating viral growth. The large T antigen is

3
required for the initiation of viral DNA synthesis, and
regulates the expression of viral early and late genes.
Following the amplification of the viral genome, three late
genes encoding the capsid proteins are expressed. Virions
are then assembled in the nucleus and released after cell
lysis (9). An in 31:19 system that replicates polyoma viral
DNA in cell-free condition has been established. It
requires extracts of mouse cells, circular DNA containing
the viral replication origin, and purified polyoma large T
antigen (10).

Growth of polyomavirus in nonpermissive cells has
provided a useful model system for studying virally induced
neoplastic transformation. Host cells respond to viral
infection by undergoing DNA synthesis and cell division. A
number of properties associated with transformed cells are
transiently expressed. However, most of the infected cells
soon lose the viral DNA, and revert to their uninfected
state. This process during which the viral early genes are
expressed is termed "abortive transformation" (11).
Eventually, a small portion (<1%) of the infected cells
become stably transformed. Integration of viral DNA into
the host chromosome and continuous expression of middle T
antigen are required for "stable transformation".
Transformed cells display reduced requirement for serum,
altered morphology with a disrupted cytoskeletal array, loss
of contact inhibition, ability to grow in semisolid medium

4
without surface attachment, and ability to form tumors in
syngeneic animals (1,12).

This thesis will address certain specific issues on
middle T antigen mediated polyoma enhancer activities, and
on stimulation of viral DNA replication and neoplastic
transformation by these activities. Therefore, I will
emphasize this literature review on the following topics:
structural organization of the viral genome, functional
analysis of the noncoding regulatory region, functions of
the early viral proteins on viral DNA replication, and the

mechanisms of cell transformation by polyomavirus.

W

The polyomavirus genome, though having a small size, is
organized in a way to obtain maximum utility by overlapping
genes which are expressed through differential processing of
RNA transcripts (Fig.1) (13,14). The viral genome is
divided into an early region and a late region, separated by
approximately 350 base pairs of‘a noncoding regulatory
region (15). The primary transcripts of early and late RNA
are synthesized in opposite direction using each own
initiation signals. The early promoter possesses several
canonical TATA and CAAT boxes, and the major start sites of
early transcripts are at nucleotides 147 and 152 as
determined by primer extension and $1 mapping (Fig.2)

(16,17), whereas, the late promoter lacks a TATA box, CAAT

 

Figure 1. Physical map of polyomavirus genome.

The inner circle represents the HpaII/HspI restriction

map of polyoma genome with the numbering system of Soeda g;

al
The noncoding region flanks the origin of DNA replication.

(15) beginning at the junction of fragments 3 and 5.

The early and late precursor RNAs are spliced as indicated
(jagged lines represent introns) to generate the mRNAs for
the T antigens or the capsid proteins, respectively. The
boxed regions represent the protein coding sequences. The
nucleotides of the splicing junctions and of the termini of

the coding regions are also shown. The figure is taken from

Soeda g; :1. (15).

 

ECORI
0559)

8911 (87)

BamH1(4631)

 

7
box, or G+C-rich region, which are sequences often
associated with the regulation of transcription initiation.
Several preferred sites for late transcription initiation
span more than 100 nucleotides which results in late RNA
with extremely heterogeneous sizes (Fig.2) (18).
Termination signals for transcription are located at the 3'
end near poly A residues. However, it has been suggested
that termination signals are inefficient since small amount
of "run through" transcripts larger than late RNA are
observed (19). Alternative splicing of the primary early
transcript generates three transcripts coding for the early
proteins: small, middle and large T antigens (16,20). The
late primary transcript is also differentially spliced to
generate three transcripts which code for the viral capsid
proteins: VP1, VP2 and VP3.

Studies of heteroduplex formation by electron
amicroscopy (21), DNA hybridization (22), and immunological
cross-reactivity (23,24) reveal extensive homology between
polyomavirus and simian virus 40 (SV40). The primary
homologous regions are the major viral capsid proteins, the
VP1, and large T antigens. These data suggest that two
viruses may have evolved from a common ancestor (25).
However, three major regions of non-homology are present,
including the noncoding regulatory regions, the region
coding for the unique portion of middle T antigen in polyoma

8

genome, and the region encompassing the C-termini of both
the large T antigen and the major viral capsid proteins.

All these regions are of unique importance, the first
with its obvious regulatory effect on gene expression and
replication, and two of the others in connection with
cellular transformation by the individual virus. Since the
two viruses interact with different permissive hosts, and
since the activities of the regulatory region are dependent
upon viral-cellular interactions, it is not surprising that
this region contains different primary sequences. The major
differences between polyomavirus and SV40 in the control
region are in sequences defined as "enhancer" which have
been demonstrated to be important for both viral DNA
replication and gene expression in a host-specific manner
(26,27). Distinct transformation mechanisms have been
invoked by polyomavirus and SV40. Middle T antigen is the
transforming protein of polyomavirus, which is located in
the cell membrane and interacts with cellular kinases
(28,29), and phosphatase 2A (30), leading to transformation
of the host. By contrast, the transformation by SV4O is
presumed to be due to the carboxy-terminus of its large T
antigen binding to a cellular protein p53 (31). It has been
suggested that both viruses may have diverged from a common
ancestor and have acquired distinct properties to

accommodate to their respective hosts during evolution (25).

Figure 2. Physical map of the noncoding region of the
polyomavirus.

The center of the major palindrome in the origin of
replication is (ori) defined as "O" and the nucleotide
numbers extending from the late-gene side on the left to the
early-gene side on the right are shown. The landmarks
include the enhancer region, a 15-bp A/T sequence (open
box), a 34-bp palindrome (pinwheel), a 17-bp inverted repeat
(4+), a TATA box (solid box) and early and late mRNA start
sites. DNA binding sites for large T antigen are shaded
according to binding affinity (B,C > A > 1,2,3). ctstcting
sequences that function as the polyoma origin of replication
(a,B and Core) are shown (shaded areas designate the minimum
required sequence for these elements). The transition
points from discontinuous to continuous DNA synthesis that
define the origin of bi-directional DNA synthesis (OBR) are

indicated by arrows protruding from ori-core. The four

replicational enhancer domains (A-D) are shown in boxes.

The DNase hypersensitive regions (DHSR) are shown in
ellipses. The figure is summarized from Veldman g; :1. (34)
and Hendrickson, g5 g1. (91).

10

 

 

f I H irf l

 

Landmarks
mRNA; TATA box
T-ag Binding

Replication
Origin

DHSR

 

 

11
W

The noncoding region of the polyomavirus genome
contains regulatory elements, including promoters for early
and late gene transcription, origin sequences (ori) for
initiation of DNA replication, and enhancer region (Fig.2)
(2). "Enhancer elements" have been found in a variety of
viral genomes and cellular genes (32). Generally, they act
by stimulating transcription from linked polymerase II
promoters in an orientation- and distance- independent
fashion (33). The polyomavirus enhancer is required for
both viral DNA replication, early and late gene
transcription (34).

The polyoma enhancer is composed of two distinct non-
overlapping enhancer elements, which are defined as the
restriction fragments BcII-PvuII (A or a enhancer, from
nucleotides 5021 to 5132), and PvuII-PvuII (B or B enhancer,
from nucleotides 5133 to 5269) (35,36) (Fig.3). Although
lacking appreciable homology, these two elements can
substitute for each other in mouse fibroblasts to activate
both early transcription and DNA replication (34,35).
Extensive deletion mutagenesis and linker scanning analyses
have further dissected the entire enhancer region into
several structurally distinct domains that are functionally
redundant. A and D domains are located in the restriction
fragment BcII-PvuII, B and C domains are in the fragment

PvuII-PvuII (34) . Each domain contains sequence motifs which

12

 

 

 

 

 

 

5021 5132 5269
i-——A (0) element 2: B (B) eiemem‘———4
L D 1 LA] i C i B J
8011' Pqu FvuII
Y Y Y
lflfllflfi REE EEE an In
Consensus ORE AdEia c-Fos SV48 BPv
SGQUenCeS E
19M
ea) .60 €52 o
F T H .63.?sz Er-C ear-20
ac ors ® . an .
PEBl 2 P-i 1

Figure 3. Cellular factors which bind to the enhancer
region of polyomavirus.

The enhancer A (a) and B (5) elements, and the enhancer
subdomains A, B, C, and D are indicated according to the
.restriction map of the polyoma enhancer. Regions of
homology to sequences found in bovine papillomavirus (BPV),
'SV40, mouse immunoglobulin M (IgM), c-fos, adenovirus Ela
enhancers, and glucocorticoid response element (GRE) are
shown in boxes. Positions of cellular factor binding sites

are depicted by ovals.

13
are homologous to several enhancer consensus sequences found

in numerous other unrelated enhancers (Fig.3) (34).

WW

Cellular proteins have been shown to bind to enhancers
in order to regulate transcription in a positive or a
negative manner (37,38). Utilizing gel shift, DNase I
footprinting, and methylation interference-protection
assays, several of these transcription factors have been
shown to bind to the polyoma enhancer (Fig.3). The
interaction between the host protein and the enhancer
sequence is tissue- and growth-state specific.

PEA1, PEA2 and PEA3 are three proteins from mouse
nuclear extracts found to bind to the sequences within the A
domain (nucleotides 5110-5132) (Fig.3) (39). These three
nuclear factors recognize distinct neighboring sequences.
PEA1 belongs to a closely-related family of transcription
factors which includes human APl, the product of the avian
sarcoma virus V‘jun oncogene and yeast GCN4 (40-42).
Recently, products of the proto-oncogenes ets-l and -2 have

been shown to activate transcription through binding to the
PEA3 motif (43). Cooperation of three oncoproteins c-Jun,

c-Fos and p68""""1

, therefore may activate from A domain of
polyoma enhancer to the highest transcription efficiency.

Both PEA1 and PEA3 but not PEA2 are able to mediate

14
transcriptional activation by serum growth factors, tumor
promoter TPA (12-O-tetradecanoylphorbol-13—acetate) and
several transforming oncogenes (44-46). PEA2 has been shown
to act as a negative factor to repress the PEA1 activity in
undifferentiated F9 EC cells, leading to a low activity of
the A domain in these cells (47). In fibroblast cells and
retinoic acid treated F9 cells, in contrast, a decrease in
repression by PEA2 as well as an increase in the activity of
a positive factor result in a large increase of A domain
activity (47). The complex interactions of three
transcription factors with the 23 nucleotides of the A
domain suggest that a comparable balance must be involved in
the cell response to growth stimuli.

The PvuII-PvuII fragment contains two functional
enhancer domains, B and C, which interact with numerous
cellular factors (Fig.3). PEBl is a nuclear protein from
murine cellular extracts interacting with the polyoma
enhancer B domain. Its major binding sequences map to
nucleotides 5184-5196, which encompass a region homologous
to the early proximal portion of the GC-rich palindrome and
the core sequences of SV40 enhancer (54-56). A nuclear
[factor isolated from rat liver, EBP20, strongly interact
with sequences along nucleotides 5189-5199 within the 8
domain (57). The binding motifs of EBP20 and PEBl overlap

the SV40 core enhancer homology sequences. A minor PEBl

15
binding region (nucleotides 5146-5162) is located in the C
domain, which partially overlaps the enhancer consensus
sequences of c-fos. Nucleotides 5159-5175 within the C
enhancer domain interacts with a nuclear factor, EF-C. The
EF-C factor was initially identified in F9 cells but is also
present in various other differentiated cell lines (58,59).

D domain is less well defined than the other three
enhancer domains. Recently, a nuclear factor (NF-D) was
found to bind to the nucleotides 5038-5045 within the
polyoma enhancer D domain (Fig.3). A specific effect from
the sequences comprising the binding site for NF-D has yet
to be demonstrated, however, this factor is ubiquitous and
comparably abundant in several murine cell types (61). Both
the A and D domains have been shown to provide basal late
promoter activity, and alteration of the NF-D binding site
can decrease the late promoter activity (62).

Negatively acting transcription factors which interact
with the enhancer sequences have also been described
(47,54,56,65,66). The negative activity of the adenovirus 2
(Ad2) ElA proteins has been examined extensively. The ElA
proteins are able to repress the transcription of a reporter
gene which is linked to the polyoma enhancer (67,68). This
inhibitory effect of the EIA proteins is also observed with
API factor-binding motifs from the SV40 and human

metallothionein enhancers, and the MHC class I gene H-2Kb

16
enhancer, which binds the KBFl/HZTFl/TC-IIB protein (66).
It is possible that the EIA proteins affect the activation
of polyoma enhancer by inhibiting the activity of PEA1.
Nuclear receptors (eg. the estrogen and glucocorticoid
receptors), yeast enhancer factor GAL4 expressed in HeLa
cells and chimeric trans-activators (such as GAL-VP16) are
also similarly inhibited by the EIA gene products. It is
therefore conceivable that the E1A proteins play a crucial
role in repression by interfering in the transcriptional
activation process at a step common to all tnuw-acting
enhancer factors (69).

The polyoma enhancer is likely to have evolved in the
context of the intricate cellular control mechanisms
responsible for modulating gene expression. The presence of
negative enhancer sequences seems contradictory to the
presumed positive evolutionary selection. It could be
because that the same enhancer sequences interact with
either activators or repressors depending on different
stages of cell differentiation. Alternatively, enhancer
sequences may interact with a single cellular factor, whose
activity is determined by the stage of cell differentiation
(i.e., PEA2). Villarreal and colleagues have shown that
individual polyoma enhancer elements display their own
tissue specificity which is different from the wild-type

pattern (86). Therefore, coexistence of several motifs

17
recognized by various cellular factors in the polyoma
enhancer may provide a survival advantage through stages of
cell differentiation and host development, and wide tropism

in the mouse.

,, ; ;. r..., ; --. .,. , -. o“. g,w~ ,.7 - .,.- !‘ =,ts
Reiterations in the polyoma enhancer have been found
frequently among natural viral strains (TOR, P16, and CSP)
(71). These sequence rearrangements appear to have a
selective advantage in replication, resulting in the
emergence of these viruses. Recently, much attention has
been focused on the enhancer region of polyoma mutants that
have a- broader host range than do the wild-type strains.
Many host-range mutants have been isolated by selection
for their ability to grow in undifferentiated mouse
embryonal carcinoma (EC) cell lines. It appears that
enhancer sequences in the B element are critical for growth
of polyoma EC mutants in undifferentiated EC cell lines.
Point mutations and tandem duplications of sequences around
this region are always observed in the mutants isolated from
F9 EC cells (48-51). By contrast, those isolated from
infection of PCC4 cells have deletions in the B element and
duplications in the A element (52,53). A variety of host-
range mutants capable of growth in normally nonpermissive
cell lines have been isolated. They include mutants which

grow in trophoblastic, neuroblastic, or Friend erythro-

18
leukemic cell lines (63,64,72-74). Each type of mutant
carries a distinct rearranged enhancer structure responsible
for the respective host range. Rearranged enhancer elements
can also strongly affect the ability of polyoma to replicate
in mouse lymphoid cell lines (75). Therefore, the spectra
of rearrangements in the enhancer region coincide with
alterations of the host range.

Mutations within either the EBP20 or EF-C binding
motifs has been shown to be detrimental to polyomavirus
enhancer activity in F9 cells. In addition, a critical
point mutation for productive infection of F9 cells is an AT
to CC transition at nucleotide 5235 of the polyomavirus A2
genome. This mutation, located within the enhancer B
.element of the F9 EC mutant F441, creates a new positive
acting sequence that can interact with a cellular factor in
undifferentiated F9 cells (60). An endogenous "E1a like"
activity has been proposed to exist in the undifferentiated
EC bells. It is based upon the observation that the EIA
products do not repress a point mutant (Py ECF9.1) of the
polyoma enhancer which is active in the undifferentiated F9
cells (67). Mbreover, a labile protein has been reported to
prevent the replication of wild-type polyomavirus in
undifferentiated PCC4 cells (70). The properties of the EC
mutants implicate that alteration of the enhancer sequences
responds to the qualitative or quantitative spectra of
regulatory factors that exist within a particular cell type.

 

41'

 

19

PyNBll/l is a polyoma mutant selected for its high
efficiency of growth in neuroblastoma cells (63). It has an
altered enhancer region which contains a 91-base-pair tandem
duplication spanning nucleotides 5047-5137 involving the A
domain and part of the D domain (64). A novel site at the
A/D junction of its enhancer region has been found to bind
to a nuclear factor that belongs to the NF-l family of
transcription factors (61). It is thus suggested that new
protein recognition sites in the polyomavirus enhancer may
be generated by using modules which may pre-exist.
Consequently, the new protein binding sites allow the host
range flexibility in viral growth.

A cooperative action of enhancer components has also
been implicated in tissue-specific replication. Replacement
of the B domain with the Moloney murine leukemia virus
enhancer drastically alters in 112g mouse organ infection
patterns. This mutation redirects virus replication to the
pancreas, which is normally nonpermissive to either of the
parental viruses (76). An F9 mutant strain nyF9 acquires
three copies of cellular sequences (box DNA) which are
inserted into the enhancer B domain. Box DNA functions as a
negative transcriptional element in undifferentiated F9
cells but not in differentiated L cells, and allows the
. persistence of an episomal state of the viral genome in F9
EC cells (65,77). Comparing the enhancer sequences of these

mutants, it is obvious that changes in specific components

 

20
of the enhancer region correlate with the ability of nyF9
to propagate in specific cell lines.

In the development of this thesis, I will discuss novel
alterations of enhancer discovered in host range
transformation defective (hr-t) mutants of polyomavirus.
These mutants were first isolated by Benjamin in 1970 using
polyoma-transformed 3T3 cells as a permissive host and
normal 3T3 cells as a nonpermissive host (78). The host
range and nontransforming mutations coincide in these
mutants, and are located in a viral protein coding region
common for both small and middle T antigens (79,80). These
mutants also contain differences in the noncoding regions
comparing to the wild-types A2 and A3 (80,81). In the past,
this has been assumed to reflect the organization of the
wild-type from which they were derived. However, sequence
analyses of the enhancer region of 7 hr-t strains show that
each strain bears a unique alteration in this region (81,
this study), which does not correspond to the wild-type. It
is more likely that each sequence rearrangement in
individual mutant has been evolved independently. A common
feature to all the rearrangements involves the reiteration
(1-3 times) of the A enhancer domain which contains middle T
antigen responsive sequences, PEA1 and PEA3 (44,46). The
theme which will be developed here is that the pathway of
transformation or middle T antigen induced signal

transduction is exquisitely coupled to the pathway for

21
polyoma enhancer activation. Thus, the duplication of the
PEA1/PEA3 binding sites may represent a compensatory
mutation for the absence of middle T and/or small T
antigens, and confer a.cis-advantage for viral growth during
the isolation/selection.

Sequence rearrangements in the enhancer region which
alter the spectra of factor binding, gene expression and
viral replication also have been described in SV40, human
polyoma virus (BK, JV). It is therefore suggested that
papovaviruses may utilize this variability to adapt to

various host cells and tissue types.

W
W

(Ms-acting sequences that function as origins of DNA
replication in polyomavirus genome contain two primary
components: a core component.(ow1-core region), dedicated to
DNA replication, that is required for replication under all
conditions, and an auxiliary component, containing enhancer
elements, that is involved in transcription as well as
replication and is dispensable under some conditions.

The ori-core contains a 17-base-pair inverted repeat on
the early gene side, multiple T-antigen binding sites within
a GC-rich 32-base-pair sequence of dyad symmetry adjacent to

a region in which 15 residues form A-T pairs (Fig.2). These

22
features are similar among polyomavirus, SV40 and BK virus,
suggesting that secondary structures or symmetric binding is
involved in the function of the origin (2). However, the
emu-cores of polyomavirus and SV40 differ in several
aspects. The major T-antigen binding sites in SV40 are
superimposed with the replication origin, while the major
polyoma T-antigen binding sites B and C lie outside of’ori-
core. The TATA box and cap-sites for SV40 early mRNA
synthesis are superimposed on ori-core, while these elements
in polyomavirus are outside of’orL-core (Fig.2) (2). The
fact that polyoma viral DNA replication requires a
transcriptional enhancer may be attributed to its different
genome structure from SV40's around the arr-core.

The enhancer region is required in 915 for efficient
synthesis of polyoma viral DNA (34,82). In mouse COP-5
cells, which constitutively express polyoma large T antigen,
SV40 or immunoglobulin enhancers can replace polyoma
enhancer to restore replication of a plasmid containing
.polyoma,or1 sequences (83). Further studies show that the
dimeric ordered combination of PEA3 and PEA1 factor-binding
sites represents the minimum sequences that are able to
stimulate viral DNA replication to wild-type level (84,85).
The perfect correlation between stimulation of DNA
replication and those binding sites for transcription

factors suggests that the enhancer function per se is

23

required for viral DNA replication and may not be specific
to the sequences utilized. Unlike the transcription
enhancement, the distance between the enhancer and the ori
core may be critical in some but not all cells for
replication activation (83,85,86). It is hypothesized that
the closeness of the enhancer to the AT-rich end of’ort-core
might facilitate interaction between.orL-core and proteins,
thereby unwinding DNA templates and initiating DNA
synthesis. These processes could be accomplished by either
direct interaction of the enhancer binding proteins with ori
binding proteins or modification of the chromatin structure
around the arr-core by enhancer binding proteins (87). It
has been shown that, in cell-free systems the polyoma ori-
core functions without an enhancer (88). Furthermore, the
polyoma art-core sequence does not require an associated
enhancer component to initiate replication until formation
of two-cell mouse embryos (89). Perhaps some negative
regulatory factors that interfere DNA replication are I
associated with the nuclear structure (i.e., chromatin
structure, matrix, scaffold). The enhancer elements, when
associated with its specific binding factors, release the
origin from the nuclear matrix.

Analyses of initiation of DNA synthesis using various
eukaryotic and prokaryotic model systems indicate that

similar strategies are shared, which involve complex

24

formation of the orl-core sequence with initiation proteins
-(90). The initiation site of polyoma DNA synthesis has been
mapped to a l6-base-pair position of the inverted repeat
(91). The first step is the ATP-dependent binding of large
T antigen to the specific sites, which leads to cascades of
DNA conformation change (untwisting, and unwinding) (92-95).
Production of underwound DNA appears to be a prerequisite
for initiation of DNA synthesis. Two diverging replication
forks are generated by the initiation complex. Large T
antigen acts as a helicase translocating along the parental
DNA strand allowing the entrance of other proteins (94,
100). DNA synthesis from the replication fork is semi-
discontinuous, using host enzymes for primer synthesis (96),
elongation (97,98), gap filling (99), ligation (2,9), etc.

The general features of polyoma viral DNA replication
resemble those observed for SV40 DNA replication. However,
it has been noticed that mouse cell lines support the
replication of polyomavirus but not of SV40, whereas monkey
and human cells replicate SV40 but not polyomavirus. The
established cell-free replication systems provide a way to
examine the cellular proteins involved in permissivity
(10,101,102). SV40 DNA is replicated in the presence of the
SV40 large T antigen and extracts from either human or
monkey cells (103). In contrast, polyoma viral DNA

replication requires the polyoma large T antigen and mouse

25

cell extracts (10). Addition of purified mouse DNA
pol a/primase complex to HeLa or monkey cell extracts allows
in yigrg synthesis of polyoma viral DNA, suggesting that the
DNA pol a/primase is a major factor in permissivity for
papovaviruses. Monoclonal antibodies to SV40 large T
antigen have been shown to co-precipitate large T antigen
and DNA pol a from SV40 infected HeLa cell extracts,
indicating that these proteins interact directly (104).

It is likely that cellular proteins other than
pol a/primase are responsible for the permissivity of
polyoma DNA replication. In rat cell lines, viral DNA
synthesis is higher at 33°C than at 37°C, and only 1-2% of
the polyomavirus transformed cells replicate the integrated
viral genome to high level (105). Treatment of polyoma
transformed rat embryo muscle cells (LPT line) with physical
and chemical carcinogens or agents that inhibit protein
synthesis can increase the percentage ($0.2%-90%) of cell
population producing large amounts of viral DNA (106). The
permissivity factors in those systems remain to be
determined. As discussed above, the varieties of enhancer
binding factors in different mouse cell types affect
permissivity of viral DNA replication: the hr-t mutants
replicate less efficiently in normal mouse cells than in
Py3T3 cells (78): wild-type polyoma strains fail to

replicate in undifferentiated EC cells (48-53).

26

Large T antigens of polyoma and SV40 are the only
virus-encoded proteins that are involved in viral DNA
replication in cell-free systems (10,101,102). The
function(s) of large T antigen in initiation have in fact
shown considerable similarity to those of E; 9911 gene
products, dnaA, dnaB, and dnaC, in initiating replication at
ori C (90). Requirement of large T antigen for initiation
of DNA synthesis was first shown by utilizing a viral mutant
(ts-a). The thermolabile large T antigen of the ts-a strain
fails to initiate new rounds of viral DNA replication at the
nonpermissive temperature (106). Binding of large T antigen
to the ori also results in negative autoregulation of early
gene expression and induction of late gene expression (108-
111). However, the overproduction of SV40 viral RNA in ts
mutant infections is temperature independent, suggesting
that the effects on RNA transcription and DNA replication
are not coordinate (112). Although large T antigen
stimulates transcription from cellular promoters (113,114),
the DNA binding domain of large T antigen is not required
for the activation (114).

Polyoma large T antigen is a nuclear phosphoprotein
that contains 785 amino acids (Fig.4) (1). At the
initiation step of polyoma viral DNA synthesis, large T
antigen functions as a sequence specific binding protein

(107-109), with ATPase (115), and helicase activities

27

 

 

N— (a a —C
I I I I I I I

mo 2% an «m 3M mm NM

Rb binding

immortalizaiion
B I nuclear localization

2 origin binding
mm ATP binding

 

 

Figure 4. Schematic representation of phosphorylation sites
and functional domains in the polyomavirus large T antigen.
Two main regions of phosphorylation sites are depicted
by circles inside the large T antigen box. Below the T
antigen are the approximate positions of domains related to

its function.

28

(94,100,117). The nucleotide binding domain of the large T
antigen has been mapped to a region between amino acids 290
and 310, whereas, the ATPase activity is located near the
carboxy-terminus between amino acids 565 and 675 (116,117)
(Fig.4). While acting as a DNA helicase, large T antigen
moves in the 3'-5' direction on the DNA strand to which it
is bound.

The phosphorylation state of large T antigen may play a
critical role in its ability to perform multiple functions
within the infected cells. Since large T antigen encoded by
ts-a mutants is underphosphorylated (118) and is not able to
initiate new round of viral DNA synthesis (106), the
phosphorylation state may be crucial to viral DNA synthesis.
McVey et al. have shown that cdc2 protein kinase
phosphorylates the large T antigen of SV40, which results in
an efficient binding of the large T antigen to the SV40 ori
(119). Another protein that regulates the phosphorylation
state of the SV40 large T antigen is protein phosphatase 2A
(PP2A). PP2A removes serine phosphates from large T
antigen, which reduces the lag time prior to new DNA chain
synthesis and stimulates the T antigen- and SSB-dependent
unwinding of DNA fragments containing the SV40 ori
(120,121). Since the activities of cdc2 protein kinase and
PP2A are both cell cycle regulated, the cell cycle stage of

infected cells may affect the level of large T antigen

29

phosphorylation. It is consistent with the observation that
the large T antigen of polyomavirus is more phosphorylated
in the growing cells early in infection than in the
quiescent cells (118). The cdc2 kinase and PP2A studies
indicate that the replication-related phosphorylation of
large T antigen may vary at different stages of the cell
cycle. Therefore, it suggests that actively cycling cells
rather than resting cells provide the optimal state to
modify large T antigen to initiate viral DNA replication.

The other two early proteins, middle T and small T
antigens play a major role in cell transformation as will be
discussed below. Their potential roles in viral DNA
replication have not been examined in detail. However, most
studies using mutants with altered middle T and/or small T
antigens suggest that these proteins are not directly
required for viral DNA replication. Hr-t mutants, which are
both middle and small T antigen defective, have been shown
to display the following phenotypes: absence of secondary
induction of c-myc and c-fos expression in host cells
(8,125): decreased output of viral DNA (123,124):
underacetylation of histone H3 and H4 in the viral
minichromosome (122); underphosphorylation of VP1: and
defects in the assembly of viral capsids into mature virions
(123).' Hr-t mutants stimulate only one round of cellular

DNA synthesis in contrast to the wild-type strains which

30
stimulate multiple rounds (126). As discussed further
below, middle T antigen has been well characterized for its
role in oncogenic transformation, correlated with its
interactions with cellular membrane proteins such as
pp60c””c and phosphatidylinositol (PI) kinase (28,29). In
the presence of small T antigen alone, NIH-3T3 cells reach a
higher saturation density than untreated monolayers (127).
These observations suggest that these two proteins act
pleiotropically to alter the physiological state of the host
in a manner which facilitates virus production in the lytic
infection. Interestingly, these two proteins have been
shown to associate with the regulatory subunits of PP2A

(30,128). How these interactions affect the activities of

large T antigen remains to be resolved.

 

The early viral gene products of polyomavirus, the so
called T antigens, were recognized originally by use of sera
from tumor-bearing animals (129). Each of the T antigen
species plays some role in the process of establishment or
maintaining cell transformation. Complementation studies
utilizing two groups of nontransforming mutants, hr-t and
ts-a strains, determine the viral proteins involved in
neoplastic transformation (12,130,131). .As mentioned above,

the major mutation of the hr-t strains has been mapped to

31
the intron of the large T antigen and affects both middle T
and small T antigens (79,80). The ts-a mutants carry
‘ mutations at the carboxy-terminus of large T antigen
(79,132). Therefore, the stable transformation caused by
polyomavirus requires cooperative functions of large T
antigen and either middle T antigen, small T antigen, or
both middle T and small T antigens.

Genetic studies have shown that large T antigen is
essential for the initiation of transformation (11). At the
nonpermissive temperature, ts-a mutants are capable of
inducing a transformed cell phenotype which lasts a couple
of generations (abortive transformation), but are defective
in stable transformation. However, ts-a mutants can
transform cells at the permissive temperature, and a delayed
shift to the nonpermissive temperature past the first two
days does not affect transformation. The process of
initiation is defined as events requiring transient
expression of large T antigen and leading to the integration
of viral genome into the cellular chromosome. Once the
viral genome is integrated, large T antigen is no longer
required.

Southern blot analyses of DNA from polyomavirus
transformed cell lines have provided extensive knowledge
regarding the arrangements of integrated viral genomes. The
viral genomes are usually integrated in a head-to-tail

tandem manner and inserted at multiple sites of the host

32
chromosome (133-135). Recombination between the host and
viral DNA is non-homologous, with only 2-5 base pairs of
partial homology at the viral host junction (136-139).
Integration of polyomavirus can result in deletions,
duplications, and rearrangement of host DNA at the
integration site (139,140). There are no preferred
sequences in the host genome found for polyomavirus
integration (134,135,141). The variations in patterns among
transformants imply that multiple independent integration
sites are available in the host chromosome. In contrast,
the observation of the double integration events from mixed
infections of FR-3T3 cells are rare, which suggests that
only a limited number of integration sites exist in the host
chromosome (142).

Two models have been proposed for the integration of
head-to-tail tandem viral genomes. Basilico and
collaborators have shown that formation of integrated
tandems requires a functional large T antigen (143) and
viral origin of replication (143,144). It has been
postulated that rolling-circle replication of the viral
genome produces a linear multimer which may serve as a
substrate for integration, consistent with a role for viral
DNA replication. High molecular weight concatamers of viral
genomes have been identified in infections of nonpermissive
cells with SV40 (145). These have been hypothesized to

represent precursors of the integrated tandem originating by

33
rolling-circle replication. However, a number of
observations suggest that this is not the normal mechanism
of how tandem genomes are generated. This mode of viral DNA
replication represents a rare pathway in polyomavirus-
infected mouse cells (146), and has not yet been documented
in rat cells. Mereover, increased levels of DNA replication
does not correlate with increased transformation frequency
(105). Further evidence against this model come from
"recombinant" tandem genomes. Recombinant tandems of viral
genomes were first described in complementation experiments
using ts-a and hr-t strains. Two parental genomes were
found cointegrated at a single site in the host chromosome,
suggesting that homologous recombination had generated a
concatemeric structure as a precursor for the integrated
tandem (147). In a system not selecting for recombination,
a very high level of interviral recombination accompanying
integration of the viral genome has been observed (148). At
present, the data does not conclusively support either the
replication or recombination model.

A recombinogenic activity for large T antigen has been
implicated from several lines both with polyomavirus and
SV40. Large T antigen is able to promote amplification and
excision of integrated viral sequences as long homologies
are provided (i.e., a repeated viral genome) (134,135,150).
Large T antigen is required for resolving a polyomavirus-

mouse hybrid replicon (149). A recent study by St-Onge et

34
al. shows that polyomavirus large T antigen promotes
homologous recombination between viral sequences very
efficiently, when the latter carry a functional origin of
viral DNA replication (151). Direct evidence for the exact
mechanism involved in the integration of polyomavirus genome
concomitantly with neoplastic transformation is still
missing. Possibly, it involves both recombination and
replication of viral DNA.

Extrachromosomal forms of viral DNA (10-60 copies per
transformed cell) are present in many transformed cells in
addition to the integrated copies (152). Most of these free
viral genomes are thought to result from excision of the
integrated viral DNA following in ELLE replication
(150,156). Excision requires homologous regions of polyoma
sequence (153,154), large T antigen (135), and a functional
origin of replication (155). It is likely that replication
and recombination are involved in generating free copies of
the viral genome. In sign amplification of integrated viral
genome requires the same parameters as for excision
(153,155). According to the "onionskin" model (156),
initiation of replication at a given proviral locus leads to
multiple rounds of DNA synthesis. The resulting structure
of a localized onionskin with amplified sequences can then
present a substrate for homologous recombination, leading to
excision or amplification. Moreover, Basilico and coworkers

have demonstrated that excision of integrated tandem repeats

35
can lead to the appearance of phenotypic revertants
(134,135).

Another property of large T antigen concerns its
ability to immortalize primary mouse and rat embryo cells
(157). Immortalized cells do not display phenotypic changes
associated with transformation, but usually a decrease in
serum requirement is observed. The immortalizing function
has been mapped to the amino-terminal portions of the large
T antigen (Fig.4) (157,158). The mechanism of the large T
antigen induced immortalization is not clear, but may be
related to its action as a regulator of,cellular gene
transcription.

Middle T antigen is the major transforming protein of
polyoma. This assertion stems back to the properties of
hr-t mutants. Hr-t mutants, though making normal large T
antigen and integrating into the cellular genome as a wild
type, do not induce either abortive or stable transformation
(159). Such findings support the argument that middle T
and/or small T antigens maintain the transformation state.
This was confirmed since plasmids which encode only middle T
antigen are able to transform established rat cell lines
(160,161) and are tumorigenic in newborn hamsters (162,163),
whereas, transformation of primary rat embryo fibroblaSts
requires all three T antigens (161). In normal established
rat fibroblasts, expression-of middle T antigen under the

control of a dexamethasone-inducible promoter elicits

36
various degrees of transformation correlating with levels of
middle T antigen (168). Anchorage-independent growth
(ability to grow in soft agar) and tumor formation require
higher levels of expression than do focus formation and
morphological changes (i.e., loss of actin cables and
decreased adhesion). Mutations which affect the middle T
antigen but not large T or small T antigen, can severely
impair the ability of transformation (171).

As mentioned above, either small T or large T antigen
has to be present to cooperate with middle T antigen to
cause tumor in primary rat embryo fibroblasts (161-163).
These observations support the theory that tumorigenesis is
a complex process that requires multiple cellular events
(164). Co-transfection of primary cells with proto-
oncogenes and polyoma early genes provide further support
for the theory. In the transformation of primary mouse

embryo fibroblasts either large T antigen or c-myc can
complement c-ras (165). Moreover, primary baby rat kidney
cells can be transformed by a combination of the Ela
proteins of adenovirus and either middle T antigen or c-ras.
The large T antigen, c-myc, as well as Ela genes are
identified as "immortalizing oncogenes" which are able to
convert cell lines of limited replicative potential into
ones which are capable of being passaged indefinitely. The

middle T antigen gene belongs to the “transforming oncogene

37
family” which confers anchorage-independent growth. The
collaboration of both types of oncogenes prompts malignant
transformation.

Middle T antigen is a membrane phosphoprotein of 421
amino acids, the amino-terminal 79 of which are identical to
the amino termini of small T and large T antigens (167).
Plasma-membrane associated middle T antigen is the major
active fraction responsible for cell transformation (167-
169). The protein is anchored in the membrane on the
cytoplasmic side via a stretch of 20 hydrophobic amino acids
near its carboxy terminus (170). Targeting the protein into
the membrane is presumed to be a post-translational event
since it lacks a signal peptide sequence at the amino
terminus.

Although the middle T antigen has no known enzyme
activities, at least three associated activities have been
described. The major effects of middle T antigen come from
its actions in regulating phosphorylation events in the
cells. One of these is from its association with cellular
tyrosine kinase of the src family (src, fyn, and yes, but
not lck) (Fig.5) (28, 172-178). These three tyrosine
kinases can be distinguished in that they are not identical
in the way they bind to middle T antigen, with p59fyn
showing a much lower affinity (28, 175-177). The

association results in the activation of the tyrosine kinase

38

Figure 5. Model for the transforming action of polyomavirus
middle T antigen.

Polyoma middle T, PyMT: Phosphoinositol kinase-3, PIK:
cellular src proto-oncogene, c-src: phosphatidylinositol,
PI; phosphatidyl inositol-bis-phosphate, PIPé: inositol-tri
phosphate, 1P3: diacyl glycerol, DG: phospholipase C, PLC:
protein kinase C, PKC: 12-0-tetradecanoylphorbol 13-acetate,
TPA: mezeriein, Mez: OAG, the diacylglycerol analogue 1-
oleoyl-2-acetyl-glycerol: protein phosphorylation,
PyMT/c-src-dependent tyrosine kinase activity and PKC-

idependent serine/threonine kinase activity.

39

cormsbamcm: . cofimmadxw memo
Lm_::mo mimwcacmm <20

mmcoammz oEmmotz

. A, s v
aomWwWmem 05%va V_E Q E
\|\\\\\\\\\

corn—3.65395
59.8.1

 

 

. 00 j w
" . _ LE
ova Aimee no: AIIIL WE

\ /

93 No: 5:

.6683?
5309.

40
activity. The complex formation between ppGOV‘" and middle
T antigen has been studied most extensively. The kinase
complex is associated with the plasma membrane (167,179,
180), and monoclonal antibodies to either middle T antigen
or pp605‘" co-precipitate the two proteins from cell
extracts of polyoma-infected cells (28). Only a small
fraction of middle T antigen in an infected cell is
associated with pp6or‘" (174), and it seems to be the level
of this complex which determine the transformation
phenotype. The membrane association of pp6051" depends on
an amino-terminal myristate group covalently linked to
glycine 1 of the mature protein (180,181). However, a
nonmyristylated pp60""”““ though failing to reside on the
membrane, becomes membrane associated upon complex formation
with middle T antigen with activation of the pp60"” kinase
activity (182). Therefore, the putative carboxy-terminus of
.middle T antigen is sufficient to hold the middle-T-pp6ov‘“
complex in the plasma membrane.

The phosphorylation state of middle T antigen varies
and influences the molecular weight of the protein. Middle
T antigen has a molecular weight of 56-58 Kd depending on
its state of phosphorylation (167,168). It is the 58 Rd
form of the middle T antigen that associates and activates
pp60w‘". A nontransforming hr-t mutant strain, N659,
produces the 56 Kd phosphorylated middle T antigen which is
able to associate with pp605‘" but is inactive in in 215:9

41
kinase assay (185,186). In wild-type polyoma infections,
TPA stimulates the phosphorylation of the 58 Rd form but not
the 56 Kd one (187). Therefore, the increase of tyrosine
kinase activity is likely determined by the phosphorylation
state of the middle T antigen and pp6ov‘“, which affects the
interactions between the two proteins (183,185). By using
site-directed mutagenesis and hr-t mutants, phosphorylation
of the tyrosine residues on the carboxy-terminal half of the
middle T antigen appears to affect the interaction between
middle T antigen with pp6or‘“ which results in the formation
of an active kinase complex (181,186,188). Genetic and
mutagenesis analyses reveal that the amino-terminus of
middle T antigen is also important for activation of the

-src

tyrosine kinase activity of pp60° and for transformation
(189). Underphosphorylation at tyrosine 527 of pp605‘“ is
critical for the tyrosine-specific kinase activity (183).
Phosphorylation at this position decreases the tyrosine-
specific kinase activity of the protein (184).

Similar to mutated src protein, middle T antigen
induces the signal transduction pathway mediated by
phosphatidyl inositol (PI) kinase. Interestingly, however,
the PI kinase involved in this pathway is a recently
discovered enzyme which phosphorylates position 3 of the
inositol ring to form phosphatidylinositol(3) phosphate

[PtdIns(3)P] rather than the position 4 seen in the

42

phospholipase C-dependent signal transduction pathway (195).
PI 3-kinase is a 81 Kd phosphoprotein which forms a stable
complex with middle T antigen and all three activated forms
of cellular tyrosine kinases (29,192,193), and is
phosphorylated on tyrosine in the complex (29,194).
Dephosphorylation of this protein either in yitzg or during
extraction has been correlated with the loss of its PI
kinase activity. The presence of this PI kinase is most
likely identical to the putative 85 Kd PI 3-kinase which
binds to the PDGF receptor. The physiological functions of
the PtdIns(3)P are not known, nevertheless, studies indicate
that it might be important in mitogenesis and transformation
(196-198). Genetic analysis has revealed a relatively close
association between this PI 3-kinase activity and
transformation. Its activity can be immune precipitated
with all transforming middle T antigens and certain
nontransforming middle T antigen mutants (29,192,199,200)
but not with other nontransforming middle T antigen mutants
which retain the ability to bind pp60”“" (190,191).
Association of PI 3-kinase activity with the middle T
antigen, therefore, appears necessary for transformation.
But whether association is sufficient for cell
transformation has not yet been resolved.

Although all middle T antigen mutants that lack
associated in yitrg tyrosine kinase activity are

transformation negative, mutants which are transformation

43
negative yet are fully active in in yitrg tyrosine kinase
assay have been isolated. Substitution of tyrosine by
phenylalanine at position 315 results in a mutant middle T
antigen that activates pp6ov‘“ but fails to serve as a
substrate for pp60°‘" and subsequently fails to bind to PI
3-kinase (170,190). This mutant middle T antigen transforms
cells in culture weakly, and is decreased in the efficiency
and spectrum of tumors in the newborn mouse. These data
indicate that phosphorylation state of middle T antigen by
ppGOV‘" is a critical event for binding of PI 3-kinase and
for expression of the full tumorigenic potential of the
virus. Furthermore, the observation of a similar elevated
level of inositol triphosphate (Inng) in both mutant and
wild-type infection suggests that association of middle T
antigen with pp6ow‘“ activates phospholipase C which is
unrelated to the binding with PI 3-kinase (191).

Two proteins with molecular weights of 36 Kd and 63 Rd
present in middle T antigen immunoprecipitates have been
recently determined to be the catalytic and regulatory
subunits of the serine- and threonine- specific protein
phosphatase 2A (PP2A) (30,128). Again, genetic analysis
suggests that the presence of these proteins is necessary
but not sufficient for transformation by middle T antigen
(201,202). However, this conclusion is based on data
regarding the association of these two proteins with various

middle T antigen mutants, rather than on measurements of

44
PP2A activity. PP2A is active in 21:19 against a number of
enzymes involved in metabolic pathways and is thought to
have an important role in the regulation of various
metabolic processes (203). The precise physiological role
for this enzyme in cellular transformation remains to be
examined.

The 36 and 63 Rd proteins also bind to the small T
antigens of polyomavirus and SV40 (30). These two proteins
are the only proteins found associated with small T antigens
in cell lysates. No mutant small T antigen, that lose
function while retaining the ability to bind to these two
proteins has been reported. It suggests that binding to
PP2A may be sufficient for the small T function. The small
T antigens cannot by themselves transform cells but are
thought to act synergistically with the other T antigens to
elicit various aspects of cell transformation (162,163). In
particular instances, the small T antigen has mitogenic and
cell-DNA-synthesis-stimulating activities for cells in
monolayer culture (127,204,205). Through the interaction
with PP2A, small T antigen may anus-activate certain
promoters by modifying the activity of certain transcription

factors (206).

45
Summer! .

From this literature review, it is obvious that a
wealth of information about the polyomavirus has been
accumulated over the past forty years. During the early
years, most studies have been focused on the observation of
biological interactions of the virus with permissive and
nonpermissive hosts. The later progress made in molecular
biology facilitates the elucidation of the intricate
arrangement of the polyomavirus genome and of the
biochemical changes which accompany viral infection. The
discoveries of the alternative splicing of viral RNA and the
replicational/ transcriptional enhancers shed light on
control mechanisms of eukaryotic gene expression.
Requirement of cellular enzymes for both viral DNA
replication and RNA transcription provides a valuable model
system for studying eukaryotic gene function. Each virally
encoded T antigen plays some role in the process of
establishing or maintaining cell transformation mainly by
altering cellular properties. The primary role of these
viral gene products are to promote virus growth. In the
meantime, it provides an intriguing system for oncogenesis
study. 4

In this study, effects of the polyomavirus enhancer on
viral DNA synthesis and neoplastic transformation were
investigated. Evidence for the existence of limiting

factors for viral DNA replication in normal mouse cells was

MING CHU CHEN
‘compensatory event for the loss of middle T and/or small T
antigens.

Enhancer mediated competition was also observed in
nonpermissive FR-3T3 cells, suggesting that rat cells
contain limiting enhancer factors. Wild-type virus induced
stable transformation was interfered more pronounced than
viral DNA replication and abortive transformation. Taken
with the observation of low probability for double
integration, we propose that integration is a rate limiting
step during the process of polyomavirus mediated neoplastic
transformation. It can be either due to the existence of
limiting amounts of one or several factors required for
integration step, or due to the limitation of integration
events in the host chromosome. Since both viruses produce
normal large T antigen, which is required for integration,
it is likely that integration of one viral genome prevents a

secondary event.

46
obtained. Competition assays revealed that the target
sequences for these factors are located within the A
enhancer element of the polyomavirus genome. In the absence
of middle T antigen, a simple duplicated A enhancer element
displayed growth advantage over the wild-type A2 enhancer.
This phenomenon was alleviated in the presence of middle T
antigen. Alleviation of enhancer competition was obtained
by adding TPA and serum growth factors, suggesting that the
middle T antigen affect enhancer activity on DNA replication
through a transformation or signal transduction pathway.
Given that all natural hr-t isolates have evolved with
sequence rearrangements in the same region of the enhancer
element, it is likely to be a compensatory event for the
loss of middle T and/or small T antigens.

Enhancer mediated competition was also observed in
nonpermissive rat cells, since transformation frequency of
the wild-type virus was decreased by the hr-t mutants in a
dosage- and enhancer context- dependent manner. The.
observation of low probability for double integration
suggests that integration is a limiting step during the
process of polyomavirus mediated neoplastic transformation.
It can be either due to the existence of a limited amount of
one or more of the factors required for integration step or
the limitation of integration events in the host chromosome.

Since both viruses produce normal large T antigen in the

47
infected cells, we propose that the competing virus prevents

the integration of wild type virus.

4.

5.

10.

ll.

12.

13.

14.

15.

48

References

Tooze, J. (ed.) 1981. In DNA tumor viruses. Cold Spring
Harbor Laboratory, Cold Spring Harbor, New York.

Salzman, N.P. (ed.) 1986. In The papovaviridae, Vol. 1:
The polyomaviruses. Plenum Publishing Corp., New York.
pp.99-246.

Fey, G., and Hirt, B. 1975. Cold Spring Harbor Symp.
Quant. Biol. 39:235.

Gross, L., 1953. Proc. Soc. Exp. Biol. Med. 83:414-421.

Stewart, S.E., Eddy, 3.3., and Borgese, N. 1958. J.
Natl. Cancer Inst. 20:1223-1243.

Basilica, C., Matsuya, Y., and Green, H. 1970. Virology
41:295-305. .

Mackay, R.L., and Consigli, R.A. 1976. J. Virol.
19:620-636.

Zullo, J., Stiles, C.D., and Garcea, R. 1987. Proc.
Natl. Acad. Sci. 84:1210-1214.

Fields, B.N. and Knipe, D.M. (ed.) 1990 In Virology,
2nd edition. Raven Press, Ltd.. New York. pp.1593-1607;
pp.3l7-367.

Murakami, Y., Eki, T., Yamada, M., Prives, C., and
Hurwitz, J. 1986. Proc. Natl. Acad. Sci. USA 83:6347-
6351.

Stoker, M., and Dulbecco, R., 1969. Nature (London)
223:397-398.

Fluck, M.M. and Benjamin, T.L. 1979. Virology 96:205-
228.

Smart. J.E. and Ito, Y. 1978. Cell 15:1427-1437.

Hunter, T., Hutchinson, M.A., and Eckhart, W. 1978.
Proc. Natl. Acad. Sci. USA 75:5917-5921.

Soeda, E., Arrand, J.R., Smolar, N., Walsh, J.E., and
Griffin, B.E. 1980. Nature (London) 283:445-453.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

49
Kamen, R.I., Favaloro, J.T., Parker, R.H., Treisman,
R.H., Lania, L., Fried, M., and Mellor, A. 1980. Cold
Spring Harbor Sympo. Quant. Biol. 44:63-75.

Kamen, R., Jat, R., Treisman, R., Favaloro, J., and
Folk, W.R. 1982. J. Mol. Biol. 159:189-224.

Cowie, A.C., Tyndall, C., and Kamen, R. 1981. Nucl.
Acids Res. 9:6305-6322.

Acheson, N.H.,1978. Proc. Natl. Acad. Sci. USA 75:4574.

Treisman, R., Cowie, A., Favaloro, J., and Kamen, R.
1981. J. Molec. Appl. Gen. 1:83-92.

Wu, M., Manor, H., and Davidson, N. 1979. J. Virol.
32:334-338.

Howley, P.M., Israel, M.A., Law, M.-F., and Martin,
M.A. 1979. J. Biol. Chem. 254:4876-4883.

Shah, K.V., Ozer, R.L., Ghazey, H.N., and Kelly,
T.J.Jr. 1977. J. Virol. 21:179-186.

McCormick, F., Lane, D.P., and Dilworth, S.M. 1982.
Virology 116:382-387.

Soeda, E., Maruyama, T., Arrand, J., and Griffin, B.E.
1980. Nature (London) 285:165-167.

Amati, P. 1985. Cell 43:561-562.

de Villiers, J., Olson, L., Tyndall, C., Schaffner, W.
1982. Nucl. acids Res. 10:7965-7976.

Courtneidge, S.A., and Smith, A.E. 1983. Nature
(London) 303:435-439.

Courtneidge, S.A., Hiber, A. Cell 1987:50:1031-1037.
Pallas, D.C., Shahrik, L.K., Martin, B.L., Jaspers, S.
Miller, T.B., Brautigan, D.L., Roberts, T.M. 1990. Cell
60:167-176.

Peden, K.W.C., Srinivassan, A., Farber, J.M., Pipas,
J.M. 1989. Virology 168:13-21.

Picard, D. 1985. pp.24-48. In Oxford Survey of
Eukaryotic genes Vol. 2 Maclean, N. (ed.), Oxford
University Press, Oxford, U.K.

McKnight, S., and Tjian, R. 1986. Cell 46:795-805.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

50

Veldman, G.M. Lupton, 8., and Kamen, R. 1985. Mol.
Cell. Biol. 5:649-658.

Herbomel, P., Bourachot, B., Yaniv, M. 1984. Cell
39:653-662.

Mueller, C.R., Muller, W.J., and Hassell, J.A. 1988. J.
Virol. 1667-1678.

Scholer, H.R., and Gruss, P. 1984. Cell 36:403-411.

Sleigh, M.J., Lockett, T.J., Kelly, J., Lewy, D. 1987.
Nucl. Acids Res. 15:4307-4324.

Martin, M.E., Piette, J., Yaniv, M., Tang, W.-J., and
Folk, W.R. 1988. Proc. Natl. Acad. Sci. USA 85:5839-
5843.

Piette, J., and Yaniv, M. 1987. EMBO J. 6:1331-1337.

Angel, P., Imagawa, M., Chiu, R., Stein, B., Imbra,
R.J., Rahmsdorf, H.J., Jonat, C., Herrlich, P., and
Karin, M. 1987. Cell 49:729-739.

Lee, W., Mitchell, P., and Tjian, R. 1987 Cell 49:741-
752.

Wasylyk, B., Wasylyk, C., Flores, P., Begue, A.,
Leprince, D., and Stehelin, D. 1990. Nature (London)
346:191-193.

Wasylyk, C., Imler, J.L., Wasylyk, B. 1988. EMBO J.
7:2475-2483.

Imler, J.L., Schatz, C., Wasylyk, C., Chatton, B.,
Wasylyk, B. 1988. Nature (London) 332:275-278.

Wasylyk, C., Flores, P., Gutman, A., Wasylyk, B. 1989.
EMBO J. 8:3371-3378.

Wasylyk, B., Imler, J.L., Chatton, B., Schatz, C.,
Wasylyk, C. 1988. Proc. Natl. Acad. Sci. USA 85:7952-
7956.

Fujimura, F.K., Deininger, P.L., Friedmann, T., and
Linney, E. 1981. Cell 23:809-814.

Sekikawa, D., and Levine, A.J. 1981. Proc. Natl. Acad.
Sci. USA 78:1100-1104.

Tseng, R.W., and Fujimura, F. K. 1988. J. Virol.
62:2890-2895.

51.

52.

53.

54.

55.

56.

57.

58.

59.

60.

61.

62.

63.

64.

65.

66.

51

Tseng, R.W., Williams, T., and Fujimura, F. K. 1988. J.
Virol. 62:2896-2902.

Vasseur, M., Kress, C., Montreau, N., and Blangy, D.

Melin, F., Pinon, H., Kress, C., Montreau, N., and
Blangy, D. 1980. Proc. Natl. Acad. Sci. USA 77:1068-
1072.

Piette, J., Kryszke, M.H., and Yaniv, M. 1985. EMBO J.
4:2675-2685.

Piette, J., and Yaniv, M. 1986. Nucl. Acids Res.
14:9595-9611.

Bohnlein, E., and Gruss, P. 1986. Mol. Cell. Biol.
6:1401-1411. -

Johnson, P.F., Landschulz, W.H.,Graves, B.J. and
McKnight, S.C. 1987. Genes Dev. 1:133-146.

Fujimura, F. 1986. Nucl. Acids Res. 14:2845-2861.

Ostapchuk, P., Diffley, J.F.K., Bruder, J.T., Stillman
B., Levine, A.J., and Hearing, P. 1986. Proc. Natl.
Acad. Sci. 83:8550-8554.

Kovesdi, I., Satake, M., Furukawa, K., Reichel, R.,
Ito, Y., and Nevins, J.R. 1987. Nature (london) 328:87-
89.

Caruso, M., Iacobini, C., Passananti, C., Felsani, A.,
and Amati, P. 1990. EMBO J. 9:(3).

Bourachot, B., Yaniv, M., and Herbomel, P. 1989. J.

Maione, R., Passananti, C., De Simone, V., Delli-Bovi
P., Augusti-Tocco, G., and Amati, P. 1985. EMBO J.
4:3215-3221.

Caruso, M., Felsani, a., and Amati, P. 1986. EMBO J.
5:3539-3546.

Ariizumi, K., Takahashi, H., Nakamura, M., Ariga, H.

Borrelli, E., Hen, R., and Chambon, P. 1984. Nature
(London) 312:608-612.

67.

68.

69.

70.

71.

72.

73.

74.

75.

76.

77.

78.

79.

80.

81.

82.

83.

84.

52

Hen, R., Borrelli, E., Fromental, P., and Chambon, P.
1986. Nature (London) 321:249-251.

Velcich, A., Kern, F.G., Basilico, C., and Ziff, E.B.
1986. Mol. Cell. Biol. 6:4019-4025.

Rochette-Egly, C., Fromental, C., and Chambon, P. 1990
Genes Dev. 4:137-150.

Cremisi, C., and Babinet, C. 1986. J. Virol. 59:761-
763.

Ruley, H.E., and Fried, M. 1983. J. Virol. 47:233-237.

Tanaka, K., Chowdhury, K., Chang, K.S.S. Israel, M.,
and Ito, Y. 1982. EMBO J. 1:1521-1527.

Delli-Bovi, P., De Simone, V., Giordano, R., and Amati,
P. 1984. J. Virol. 49:566-571.

De Simone, V., and Amati, P. 1987. J. Virol. 61:615-
620.

Campbell, B.A., and Villarreal, L.P. 1987. Virus Genes
1:23-24.

Rochford, R., Campbell, B.A., and Villarreal, L.P.
1987. Proc. Natl. Acad. Sci. USA 84:449-453.

Ariizumi, K., Takahashi, H., Nakamura, M., Ariga, H.
1989. Mol. Cell. Biol. 9:4026-4031.

Benjamin, T.L. 1970. Proc. Natl. Acad. Sci. USA 67:394-
399.

Feunteun, J., Sompayrac, L., Fluck, M., and Benjamin,
T. 1976. Proc. Natl. Acad. Sci. 73:4169-4173.

Hattori, J., Carmichael, G.G., and Benjamin, T.L. 1979.
Cell 16:505-513.

Amalfitano, A. 1989. In thesis.

Tyndall, C., La Mantia, G., Thacker, L.M., Favaloro,
J., and Kamen, R. 1981. Nucl. Acids Res. 9:6231-6251.

de Villiers, J., Schaffner, W., Tyndall, C., Luptons,
S., and Kamen, R. 1984. Nature (London) 312:242-246.

Wasylyk, C., Schneikirt, J., and Wasylyk, B. 1990.
Oncogene 5:1055-1058. '

85.
86.
87.
88.
89.

90.

91.
92.
93.
94.
95.
96.
97.
98.
99.

100.

53

Rochford, R., Davis, C.T., Yoshimoto, K.K., and
Villarreal, L.P. 1990. Mol. Cell. Biol. 10:4996-5001.

Campbell, B.A., and Villarreal, L.P. 1988. Mol. Cell.
Biol. 8:1993-2004.

DePamphilis, M.L. 1988. Cell 52:635-638.

Prives, C., Murakami, Y., Kern, F.G., Folk, W.,
Basilico, C., and Hurwitz, J. 1987. Mol. Cell. Biol.
7:3694-3704.

Martinez-Salas, E., Cupo, D.Y., and DePamphilis, M.L.
1988. Genes Dev. 2:115-126.

Bramhill, D., and Kornberg, A. 1988. Cell 54:915-918.

Hendrickson, E.A., Fritze, C.E., Folk, W.R., and
DePamphilis, M.L. 1987. EMBO J. 6:2011-2018.

Borowiec, J.A., and Hurwitz, J. 1988. Proc. Natl. Acad.
Sci. 85:64-68.

Borowiec, J.A., and Hurwitz, J. 1988. EMBO J. 7:3149-
3158.

Stahl, H., Drége, P., and Knippers, R. 1986. EMBO J.
5:1939-1944.

Wold, M., and Kelly, T. 1988. Proc. Natl. Acad. Sci.
85:2523-2527.

Borowiec, J.A., Dean, E.B., Bullock, P.A., and Hurwitz,
J. 1990. Cell 60:181-184

Hendrickson, E.A., Fritze, C.E., Folk, W.R., and
DePamphilis, M.L. 1987. Nucl. Acids Res. 15:6369-6385.

Tsurimoto, T., and Stillman, B. 1989. EMBO J. 8:3883-
3889.

KrOkan, H., Schaffner, P., and DePamphilis, M.L. 1979.
Biochemistry 18:4431.

Weaver, D., Krokan, H., and DePamphilis, M.L. 1980. J.
Supramol. Struct. Suppl. 4(Abstr.), 895.

101. Murakami, Y., Eki, Y., Yamada, M.-A., Prives, D., and

102.

Hurwitz, J. 1986. Proc. Natl. Acad. SCi. 83:6347-6351.

Li, J.J., and Killy, T.J. 1984. Proc. Natl. Acad. Sci.
81:6973-6977.

103.

104.

105.

106.

107.

108.

109.

110.

111.

112.

113.

114.

115.

116.

117.

118.

119.

54

Wobbe, C.R., Dean, F.B., Weissbach, L., and Hurwitz, J.
1985. Proc. Natl. Acad. Sci. 82:5710-5714.

Smale, S., and Tjian, R. 1986. Mol. Cell. Biol. 6:4077-
4087.

Hacker, D.L., and Fluck, M.M. 1989. Mol. Cell. Biol.
9:648-658.

a., and Manor, H.

Mendelsohn, E., Baran, N., Neer,

1982. J. Virol. 41:192-209.
Franks, B., Eckhart, W. 1973. Virology 55:127-135.

Gaudray, P., Tyndall, C., Kamen, R., and Cuzin, F.
1981. Nucl. Acids Res. 21:5697-5710.
1983.

B.J., Muller, C.R., and Hassell, J.A.

47:600-610.

Pomerantz,
J. Virol.

Dilworth, S.M., Cowie, A., Kamen, R.I., and Griffin,
B.E. 1984. Proc. Natl. Acad. Sci. 81:1941-1945.

Kern, F.G., Pellegrini, 8., Cowie, A., and Basilico, C.

Clark, R., Penden, K., Pipas, J.M., Nathans, D., Tjian,
R. 1983. Mol. Cell. Biol. 3:220-228.

Matter, J.M., Tiercy, J.M., and Weil, R. 1983. Nucl.
Acids Res. 11:6612-6629.

Kingston, R.E., Cowie, A., Morimoto, R.I., and Gwinn,
R.A. 1986. Mol Cell. Biol. 6:3180-3190.

Gaudray, P., Clertant, P., and Cuzin, F. 1980. Eur. J.
Biochem. 109:553-560.

Tang, W.-J., and Folk, W.R. 1989. J. Virol. 63:242-249.

Dean, F.B., Bullock, P., Murakami, Y., Wobbe, C.R.,
Weissbach, L., and Hurwitz, J. 1987. Proc. Natl. Acad.
Sci. 84:16-20.

Bockus, B.J., and Schaffhausen, 8.8. 1987. J. Virol.

-61:1147-1154.

McVey, D., Brizuela, L., Mohr, I., Marshak, D. R.,
Gluzman, Y., and Beach, D. 1989. Nature (London)
341:503-507.

120.

121.

122.

123.

124.

125.

126.

127.

128.

129.

130.

131.
132.

133.

134.

135.

136.
137.

138.

55

Virshup, D.M., and Kelly, T.J. 1989. Proc. Natl. Acad.

Sci. 86:3584-3588.

Virshup, D.M., Kauffman, M.G.,
EMBO J. 8:3891-3898.

and Kelly, T.J. 1989.
Schaffhausen, B.S., and Benjamin, T.L. 1976. Proc.
Natl. Acad. Sci. 73:1092-1096.

Garcea, R.L., and Benjamin, T.L. 1983. Proc. Natl.
Acad. Sci. 80:3613-3617.

Turler, H., and Solomon, C. 1985. J. Virol. 53:579-586.

Glenn, G. M., and Eckhart, W. 1990. J. Virol. 64:2193-

2201.
Schlegel, R., and Benjamin, T.L. 1978. Cell 14:587-599.

Noda, T., Satake, M., Robins, T., and Ito, Y. 1986. J.
Virol. 60:105-113.

Walter, G., Ruediger, R., Slaughter, C., and Mumby, M.
1990. Proc. Natl. Acad. Sci. 87:2521-2525.

Habel, K. 1965. Virology 25:55-61.

Pluck, M., Staneloni, R., and Benjamin, T. 1977.
Virology 77:610-624.

Eckhart, W. 1977. Virology 77:589-597.
Miller, L.K., and Fried, M. 1976. J. Virol. 18:824-832.
Lania, L., Gandini-Attardi. D., Friffiths, M., Cooke,

B., DeCicco, D., and Fried, M. 1980. Virology 101:217-
232.

Birg, F., Dulbecco, R., Fried, M., and Kamen, R. 1979.
J. Virol. 29:633-648.

Basilico, C.,,Gattoni, S., Zouzias, D., and Della
Valle, G. 1980. Cell 17:645-659.

Stringer, J.R. 1981. J. Virol. 38:671-679.

Stringer, J.R. 1982. Nature (London) 296:363-366.
Hayday, A., Ruley, H.E., and Fried, M. 1982. J. Virol.

44:67-77.

139.

140.

141.

'142.

143.

144.

145.

146.

147.

148.

149.

150.

151.

152.

153.

154.

155.

156.

56

Bullock, P., Forrester, W., and Botchan, M. 1984. J.

Williams, T.J., and Fried, M. 1986. Mol. Cell. Biol.
6:2179-2184.

Winberry, L.K., Stewart, C.J., Schaffhausen, B.S., and
Fluck, M.M. 1985. Virology 144:433-447.

Oh, S.Y., Amalfitano, A., Friderici, K., Chen, M-C.,

Della Valle, G., Fenton, R., and Basilico, C. 1981.
Cell 23:347-355.

Dailey, L., Pallegrini, S., and Basilico C. 1984. J.
Virol. 49:984-987.

Chia, Q., and Rigby, P.W.J. 1981. Proc. Natl. Acad.
Sci. 78:6638-6642.

Bjursell, G. 1978. J. Virol. 26:136-142.

Fluck, M.M., Shaikh, R., and Benjamin, T.L. 1983.
Virology 130:29-43.

Hacker, D.L., and Fluck, M.M. 1989. Mol. Cell. Biol.
9:995-1004.

Piche, A., and Bourgaux, P. 1987. J. Virol. 61:845-850.
Botchan, M., Topp, W., and Sambrook, J. 1979. Studies
on SV40 excision from cellular chromosomes. Cold Spring
Harbor Symp. Quant. Biol. 43:709-719.

St-Onge, L., Bouchard, L., Laurent, 8., and Bastin, M.
1990. J. Virol. 64:2958-2966.

Prasad, I., Zoizias, D., and Basilico, C. 1976. J.
Virol. 64:2958-2966.

Colantuoni, V., Dailey, L., Della Valle, G., and
Basilico, C. 1982. J. Virol. 43:617-628.

Lania, L., Boast, S., and Fried, M. 1982. Nature
(London) 295:349-350.

Pellegrini, S., Dailey, L. and Basilico, C. 1984. Cell
36:943-949.

Baran, N., Neer, A., and Manor, H. 1983. Proc. Natl.

157.

158.

159.

160.

161.

162.

163.

164.

165.

166.
167.

168.

169.

170.

171.

172.

173.

174.

57
Rassoulzadegan, M., Naghashfar, z., Cowie, A., Carr,
A., Grisoni, M., Kamen, R., and Cuzin, F. 1983. Proc.
Natl. Acad. Sci. 80:4354-4358.
Asselin, C. and Bastin, M. 1985. J. Virol. 56:958-968.

Lania, L., Griffiths, M., Cooke, B., Ito, Y., and
Fried, M. 1979. Cell 18:793-802.

Treisman, R., Novak, U., Favaloro, J., and Kamen, R.
1981. Nature (London) 292:595-600.

Rassoulzadegan, M., Cowie, A., Carr, A., Glaichenhaus,
N., Kamen, R., and Cuzin, F. 1982. Nature (London)
300:713-718.

Asselin, C., Gelinas, C., Branton, P.E., and Bastin, M.
1984. Mol. Cell. Biol. 4:755-760.

Asselin, C., Gelinas, C., and Bastin, M. 1983. Mol.
Cell. Biol. 3:1415-1459.

Cairns, J. 1975. Nature 255:197-200.

Land, H., Parada, L.L., and Weinberg, R.A. 1983. Nature
(London) 304:596-602.

Ruley, H.E., 1983. Nature (London) 304:602-606.

Schaffhausen, B/S. 1982. CRC Crit. Rev. Biochem.
13:215-286.

Raptis, L., Lamfrom, H., and Benjamin, T.L. 1985. Mol.
Cell. Biol. 5:2476-2485.

Schaffhausen, B.S., Dorai, H., Arakere, G., and
Benjamin, T.L. 1982. M01. Cell. Biol. 2:1187-1198.

Carmichael, G., Schaffhausen, B.S., Mandel, G., Liang,
T.J. and Benjamin, T.L. 1984. Proc. Natl. Acad. Sci.
81:679-683.

Benjamin, T.L. 1982. Biochim. Biophys. Acta 695:69-95.

Bolen, J.E., Thiele, C.J., Israel, M.A., Yonemoto, W.,
Lipsich, L.A., and Brugge, J.S. 1984. Cell 38:767-777.

Courtneidge, S.A. 1985. EMBO J. 4:1471-1477.

Courtneidge, S.A., and Smith, A.E. 1985. EMBO J. 3:585-
591.

175.

176.

177.

178.

179.

180.

181.

182.

183.

184.

185.

186.

187.

188.

189.

190.

58

Cheng, S.H., Harvey, R., Espino, P. Semba, K.,
Yamamoto, T., Toyoshima, K., and Smith, A.E. 1988. EMBO
1987.

Kornbluth, S., Sudol, M., and Hanafusa, H. Nature

(London) 325:171-173.

Kypta, R.M., Hemming, A., Courtneidge, S.A. 1988. EMBO

Louie, R.R., King, C.S., Macanley, A., Marth, J.D.,
Perlmutter, R.M., Eckhart, W., and Cooper, J.A. 1988.
J. Virol. 62:4673-4679.

Natl. Acad. Sci.

Segawa, K., and Ito, Y. 1982. Proc.

79:6812-6816.

Ballmer-Hofer, K., and Benjamin, T.L. 1985. EMBO J.
4:2321-2327.

Carmichael, G.G., and Benjamin, T.L. 1982. J. Biol.
Chem. 225:230-235.
wyss, A., Kaech, S., and Ballmer-Hofer, K. 1990. J.
Virol. 40:184-196.

Cartwright, C.A., Kaplan, P.L., Cooper, J.A., Hunter,
T., and Eckhart, W. 1986. Mol. Cell. Biol. 6:1562-1570.

Punica-Qorms, H.K., Saunders, B., Roberts, T.M., Smith,
A.E. and Chang, S.M. 1987. Cell 49:75-82.

Schaffhausen, B.S., and Benjamin, T.L. 1981. J. Virol.
40:184-196.

Cheng, S.H., Markland, W., Markham, A.F., and Smith,
A.E. 1986. EMBO J. 5:325-334.

Mathews, J.T., and Benjamin, T.L. 1986. J. Virol.
58:239-246.

Carmichael, G.G., Schaffhausen, B.S., Dorsky, D.I.,
Oliver, D.B., and Benjamin, T.L. 1982. Proc. Natl.
Acad. Sci. 79:3579-3583.

Cook, D.N., and Hassell, J.A. 1990. J. Virol. 64:1879-
1887.

Talmage, D.A., Freund, T., Young, A.T., Dahl, J., Dawe,
C.J., and Benjamin, T.L. 1989. Cell 59:55-65.

191.

192.

193.

194.

195.

196.

197.

198.

199.

200.

201.

202.

203.

204.

205.

206.

59

Gorga, F.R., Riney, C.E., and Benjamin, T.L. 1990. J.
Virol. 64:105-112.

Kaplan, D.R., Whitman, M., Schaffhausen, B., Raptis,
L., Garcea, R.L., Pallas, D., Roberts, T.M., and
Cantley, L. 1986. Proc. Natl. Acad. Sci. 83:3624-3628.

Whitman, M., Kaplan, D.R., Schaffhausen, B., Cantley,
L., and Roberts, T.M. 1985. Nature (London) 315:239-
242.

Kaplan, D.R., Whitman, M., Schaffhausen, B., Pallas,
D.C., White, M., Cantley, L., and Roberts, T.M. 1987.
Cell 50:1021-1029.

Whitman, M., Downes, C.P., Keeler, M., Keeler, T., and
Cantley, L. 1988. Nature (london) 332:644-646.

Auger, K.R., Serunian, L.A., Soltoff, S.P., Libbey, P.,
and Cantley, L.C. 1989. Cell 57:167-175.

Kypta, R.M., Goldberg, Y., Ulug, E.T., Courtneidge,
S.A. 1990. Cell 62:481-492.

Fukui, Y., Hanafusa, H. 1989. Mol. Cell. Biol. 9:1651-
1658.

Morgan, W.C., Kaplan, D.R., Pallas, D.C., and Roberts,
T.M. 1988. J. Virol. 62:3407-3414.

Druker, B.J., Ling, L.E., Cohen, B., Roberts, T.M., and
Schaffhausen, 8.8. 1990. J. Virol. 64:4454-4461.

Grussenmeyer, t., Scheidtmann, K.H., Hutchinson, M.A.,
Eckhart, W., and Walter, G. 1985. Proc. Natl. Acad.
Sci. 82:7952-7954.

Pallas, D.C., Cherington, V., Morgan, W., Deanda, J.,
Kaplan, D., Schaffhausen, B., and Roberts, T.M. 1988.
J. Virol. 62:3934-3940.

Cohen, P. 1989. Annu. Rev. Biochem. 58:453-508.
Cherington, V., Morgan, B., Spiegelman, B.N., and
Roberts, T.M. 1986. Proc. Natl. Acad. Sci. 83:4307-
4311.

Liang, T.J., Carmichael, G.G., and Benjamin, T.L. 1984.
Mol. Cell. Biol. 4:2774-2783.

Loken, M., Bikel, I., Livingston, D.M., and Brady, J.
1988. Cell 55:1171-1177.

Chapter 2.

DNA Replication of Polyoma Variants with Altered Enhancers1

ABSTRACT

The DNA replication capacity of polyomavirus was
studied in mixed infections of NIH-3T3 cells. The results
show that viral genomes compete for cellular factor(s)
present in limiting amounts. Studies of enhancer mutants
suggest that nuclear factor(s) binding to the A enhancer
element represent the major class of candidates. Kinetic
studies imply that.competition occurs and accumulates after
the initial round(s) of DNA replication and that the
limiting step is an initiation step. Furthermore, studies
with hr-t mutants (lacking both middle and small T antigens)
reveal that middle T and/or small T antigen play a major
role in DNA synthesis, presumably by inducing factors which

bind to the A enhancer.

 

1Ming Chu Chen and Michele M. Fluck, Department of
Microbiology, Interdepartmental Program in Molecular
Biology, Michigan State University, East Lansing, MI 48824-
1101.

60

61
Introduction

Polyomaviruses such as polyoma (Py) and SV40 DNA have
provided a useful model for the study of eukaryotic DNA
synthesis (7). Efficient replication of the Py genome
requires.cis-acting sequences, virally encoded large T
antigen, and cellular components from a permissive host
(24,28). The:cIs-acting sequences, encompassing the non-
coding region of the viral genome (35), include the origin-
core and enhancer elements (6,8). The origin-core contains
minimum sequences needed to replicate Py DNA in a cell free
system (29,45). Large T antigen binds to sequences within
and adjacent to the origin-core, and is required for the
initiation of viral DNA synthesis (11,14,28). Once bound to
the viral genome, large T antigen unwinds DNA around the
replication origin, thereby permitting subsequent entrance
of host cellular enzymes to proceed DNA replication.

The mechanism by which the enhancer sequences affect
transcription and replication of Py DNA is under
investigation. It has been shown that the enhancer plays
dual roles in directing Py DNA replication as well as
transcription in a cell-specific manner (6). Mutations in
this region simultaneously alter both DNA replication and
transcription in specific cell types, suggesting that the
enhancer activates both processes through a common pathway

(2,4,40). The minimal enhancer sequences required for full

62
levels of viral DNA replication varies from cell-type to
cell-type (30,40). However, the enhancer is dispensable for
DNA replication In yitng (29), and in the early development
of mouse embryos at the one-cell stage (23). Therefore, it
is possible that different cellular anus-acting factors,
chromatin and nuclear structures lead to the tissue-,
stage-, or cell-specificity.

Up to now, a number of enhancer binding proteins have
been identified in different cell lines (5,12,18,22,25,26,
27). Some of these proteins interact In 21:19 with specific
sequence motifs within the Py enhancer. In yiyg studies
demonstrate that binding of some of these anus-acting
factors activates replication of the Py genome cooperatively
(24,30). Mutation of certain factor binding sequences can
abolish the replication in some cell lines. More
interestingly, mutations may create novel factor recognizing
sequences, and allow the virus to grow in a previously
nonpermissive host (5,19,21,26,38). Natural variations in
enhancer sequences have been reported among Py strains of
similar as well as different host-range groups, but certain
common features are observed within the same group
(19,21,38,39). It appears that the rearrangements in the
enhancer somehow modify the interaction between the viral
genome and cellular factors, consequently leading to altered

tissue tropism with regard to the ability to replicate and

63
transcribe in different hosts (2).

Host range transformation deficient (hr-t) mutants were
selected as viable mutants of Py with altered host range and
transformation deficiency shown to be mediated by mutations
in middle T and small T antigens (9,37). As shown
elsewhere, these mutants evolved with a uniquely rearranged
enhancer compared to the wild-type strains A2 or A3 (1,16
and Chapter 3 in this thesis). In addition to other unique
variations in the enhancer structures, sequences between
nucleotides (nt) 5088 and 5146 are found reiterated two to
four times in each of seven viral genomes analyzed. The
region of overlap contains binding sites for 3 nuclear
factors, PEA1, PEA2 and PEA3, of mouse cells (22). PEA 1
and PEA 3 represent the mouse analogs of the oncogenes jm:

and.ecs, respectively (27,41). They also function

cooperatively to activate transcription and replication
(30,42,43). In this article, we studied the effect of
amplification of these oncogene-responsive transcription
motifs on Py DNA replication and their biological

significance in hr-t mutants.

W

QQII§_an§_yirn§gfiI Mouse NIH-3T3 cells were grown in
1x Dulbecco modified Eagle's medium (DMEM) supplemented with

10% heat-inactivated newborn calf serum and maintained in a

64
humidified incubator with 5% CO, at 37°C.

All viral strains were grown on baby mouse kidney (BMK)
cultures from plaque purified viruses (44). A2 strain is a
large-plaque virus, used as a reference wild-type in our
laboratory (15). Hr-t mutants were originally screened for
their ability to grow on polyoma-transformed 3T3 cells (Py-
3T3) and their failure to grow on normal mouse 3T3 cells
(3). They have lesions in both small T and middle T
antigens but none in large T antigen (33,34). The mutations
have been mapped to MspI restriction fragment #4 in the
proximal part of the early region (9,16). A9, B2, 384, and
6B5 were isolated from lysate of a Pasadena small plaque
strain mutagenized by the mutagen ICR-191 (37). Mutant II-5
was isolated from lysates of a Pasadena small plaque virus
after two high multiplicity passages on Py-3T3 cells (37).
18-5 is an hr-t mutant constructed by site directed
mutagenesis from wild-type A2, and contains an approximately
50 bp deletion in the MspI fragment #4 (20).

Enhangez_gxgnange§I Viral DNA from A2- or 82- infected
NIH-3T3 cells were prepared by the method of Hirt (17).
Viral DNA was digested with restriction enzymes.8c1I and
8311, and each BclI-BglI fragment was gel-purified. The 375
base-pair (bp) BcII-BglI fragment of mutant strain 82 was
ligated to the 4954 bp BclI-BgII fragment of wild-type A2. A

total of 50 ng of viral DNA fragments were ligated in 10 pl

65
of ligation buffer (66 mM Tris, pH 7.5; 10 mM MgC12: 1 mM
DTT: and 1 mM ATP) for 16 hours at 16°C using T4 DNA ligase
(New England Biolab). For transfection, 10 ng of the
ligation mixture was applied to 1.5x1015 NIH-3T3 cells/35 mm
dish using 500 ug/ml DEAE-dextran at 33°C for 1 hour. The
dishes were rinsed with 1X Tris-buffered saline and serum
free 1X DMEM, overlayed with DMEM containing 0.9% agar and
5% newborn calf serum, and incubated at 37”C. Plaques were
picked, and viral stocks were grown on BMK or NIH-3T3 cells.
The resulted recombinant virus is named as A2(e82), which is
transformation-competent as strain A2 and contains a 82
enhancer-origin region.

InfegtigngI NIH-3T3 cells were seeded at a density of
2x105 per 60 mm plate and were infected after‘attachment.
The relative ratio of two viral genomes in the infection
mixture was established from the titers of the viral stocks,
and confirmed from inputs by hybridization analysis of low
molecular weight DNA extracted from infected cells at 4
, hours post infection. Cells were fed with 1x DMEM
supplemented with 2% heat-inactivated newborn calf serum
after infection.

Aneg§I§_gf;yirnI_DNAI Low-molecular-weight DNA was
sampled by the method of Hirt (17) at the designated time
points post infection. After phenol/chloroform extraction
and 95% ethanol precipitation, the DNA was suspended in 0.05

ml of 10 mM Tris hydrochloride (pH 7.4)-1 mM EDTA (Tufih).

66

For analysis, unless otherwise indicated, 1/10 volume of
each sample was digested with.MspI, electrophoresed on a 2%
Borate-Tris agarose gel, and hybridized with a specific
probe after Southern transfer to nitrocellulose (36).

MW
Infected cell monolayers and supernatants were harvested at
5 days post infection when the cells showed signs of lysis.
Lysates were sonicated for 1.5 minutes to release the virus
from cell debris. Total lysate DNA was isolated by
digestion of 0.4 ml of lysate with 50 pg/ml proteinase K and
0.5% SDS at 37°C overnight. After phenol/chloroform
extraction, DNA was precipitated with 0.3 M sodium acetate
(pH 5.2) and 2 volumes of 95% ethanol. The DNA was
resuspended in T1,,E1 as described above. To obtain
encapsidated viral DNA, we removed free DNA from the lysates
by treatment with 50 ng/ml DNase I in the presence of 10 mM
magnesium chloride at 37°C for 2 hours. DNase I was
inactivated by adding 15 mM EDTA and 0.2% SDS. The resulted
mixture was then treated with proteinase K and SDS to purify
virion DNA. For analysis, one-tenth volume of total lysate
DNA and one-third of the virion DNA were treated as
described above.

WW 3’ZP-lalnelo‘aml pPy-l.
representing the complete polyomavirus genome in pBR322, or

» ”P-labeled pG4PyH4, representing the HspI-4 in pGem 4

67
plasmid (Promega), was used as probe in DNA hybridizations
to Southern blots at a specific activity of 1x109 to 6x109
cpm/ug. Hybridizations were carried out in 1x Denhardt's
solution-2x SSC (1x SSC is 0.15 M NaCl plus 0.015 M sodium
citrate) (0.1 ml/cmz) at 65°C for 24 hours with 0.5x10° cpm/ml
probe. Comparisons of the intensities between DNA bands in

the same blot were determined by densitometric scanning.

825311:

WWW
ggnngt1;19n_n§§gyI As reviewed in the Introduction, all the

factors and enzymes required for replication of Py DNA are
derived from the host cell, except for large T antigen (7).
To study the capacity of the host replication machinery, we
designed a competition assay. NIH-3T3 cells were infected
with a mixture of wild-type strain A2 and a viable deletion
mutant strain of the hr-t type (18-5), and the levels of
accumulated viral DNA were analyzed. In each infection
mixture, we varied the input ratio between parental strains
by keeping the multiplicity of A2 at 1 and varying the
multiplicity of 18-5 from 1 to 10. Input ratios were
verified by hybridization analysis of DNA extracted four
hours post infection (Data not shown). Low molecular weight
DNA was extracted at 60 hours post infection and analyzed by
rmpI digestion, which differentiated the two viruses by the

68
fragment #4 (Material and Methods). Fig.1 indicates that as
the multiplicity of 18-5 is increased from 0 to 10 in the
input, a 20% decrease in the yield of wild-type A2 DNA is
observed. This result implies that limiting amounts of
replication factors are present in NIH-3T3 cells. Since
both parental strains produce normal large T antigen
(33,34), it is likely that the limiting factor is of
cellular origin.

At the MOI ratio of 1 to 1, we found that deletion
mutant 18-5 replicated less efficiently than the wild-type
A2. We think this could have resulted from an effect of low
multiplicity. Since at the MOI of 1, approximately 30% of
the cell population received both parental viruses and most
infected cells contained either A2 or 18-5. Most signals in
the extracts, therefore, represent accumulated DNA from
singly infected cells. The result of underreplication of
18-5 in this infection is consistent with our observation
that this virus grows poorly (see Discussion).

‘; .‘ ,- ;,‘,4._, - 01'01- ._ ., ., ,; .9..- '01
IQ:_DNA_:§nIIg§;IQnI Given the precedent that binding of
nuclear factors to the enhancer A domain is required for Py
DNA replication and transcription (22), and reiterations in

the enhancer A domain provide a cIs-advantage for growing in

several types of mouse cells (21), we investigated the

69

A2/18-5

92%); 1/01/ 1/3 V10 bp,

.A2:> I'l'lllifli.ilil .<702
18—5> ~6. <652

 

 

Figure 1. Replication of A2 virus in competition with
deletion mutant 18-5.

NIH-3T3 cells were infected with viral mixtures A2 and
18-5 at the M01 ratios of 1:0, 1:1, 1:3, and 1:10. Low
molecular weight DNA was isolated at 60 hours post
infection. One tenth of the sample was digested with
restriction endonuclease.MspI, electrophoresed on a 2%
agarose gel, transferred to nitrocellulose, and hybridized
with a polyoma virus probe specific to MSpI restriction
fragment #4. The upper band of 702 bp represents the MspI
fragment #4 of A2, and the lower band of 652 bp corresponds

to the MspI fragment #4 of 18-5.

70

effect of the enhancer duplication on the competition for
viral DNA synthesis. For this purpose, we made use of an
hr-t strain 82, which contains a 42 bp tandem duplication
(nt 5103-5144) that encompasses the A domain when compared
to wild-type A2 (Chapter 3). We also constructed a
recombinant wild-type virus A2(e82), whose noncoding
regulatory region was replaced with that of 82 (Materials
and Methods). We carried out competition assays by mix
infecting NIH-3T3 cells with wild-type strain, A2 or
A2(e82), and a deletion strain 82. Accumulated viral DNA
was analyzed as described above. We found that when the
input MOI of the 82 strain was increased up to 10, DNA
replication of wild-type A2 and A2(e82) was decreased 83%
and 30%, compared to that obtained in the single infections
(Fig.2). However, the level of competition was less
dramatic in the mixed infections with A2 plus 18-5, and
A2(e82) plus 82 than in that with A2 plus 82. The dosage-
dependence of the competition supports our previous
conclusion that NIH-3T3 mouse cells contain limiting amounts
of replication factors. Moreover, the difference in the
extent of competition influenced by different enhancer
structures suggests that one of the target(s) of limiting
factors is present on the duplicated A enhancer sequences.

WWW
nixed_infegt19nsI To generalize the effect observed in the

experiment using strain 82 as a competitor, a variety of

71

A 2 I 82 A2(e82)/B2

 

 

MQI 1/0

raho

Ill 1/3 1/10 1/0 ill 1/3 1/10
wr> -Cc’ - .*.... <702

82> <462

   

(A) i (a)

Figure 2. Competition assay in the mixed infections with
various wild-types and deletion mutant 82.

NIH-3T3 cells were doubly infected with viral mixtures
A2 and 82 (A), A2(e82) and 82 (B), at the MOI ratios of 1:0,
1:1, 1:3, and 1:10. Viral DNA was analyzed as described in
Fig.1. The size of MspI fragment #4 for the wild-types is

702 bp, and for strain 82 is 462 bp.

72

other hr-t mutants were tested. As shown in Fig.3, using A2
and hr-t deletion mutants at a ratio of 1 to 10, all these
hr-t mutants suppressed the replication of A2 DNA. When
compared to the mixed infection with A2 derived hr-t strain
18-5 (lane 6, Fig.3), naturally isolated hr-t strains appear
to be stronger competitors than 18-5. Furthermore, the
magnitudes of reduction of A2 DNA varied amidst different
infections. Strain A9, which contains 3-4 copies of A
enhancer sequences (1, Chapter 3), induced the strongest
competition (lane 2, Fig.3).

WWW TheA
domain of the Py enhancer, which contains binding sites for
factors PEA1 and PEA3 (22), appears to be the target of the
competition. Since both PEA1 and PEA3 are presumed to be
activated by middle T antigen (42), it became of interest to
study the effect of middle T antigen on the competition. An
example is shown in Fig.4. In this experiment, NIH-3T3
cells were doubly infected with middle T antigen deficient
strains 82 and 18-5. The multiplicities and the ratios of
two viruses are indicated in Fig.4. Accumulated viral DNA
extracted at 60 hours post infection was analyzed withIMspI,
since the two parental strains contain different sizes of
deletions in fragment #4. Compared to experiments with
wild-type (Panel A, Fig.2), competition by 82 was more
intense in the absence of middle T antigen. For example,

comparing the output of 18-5 at the MOI ratio of 1 to 1,

73

0‘ v- us In I? :2
Mutant - < 39, g I 0'3 9, bp.
O
<702

 

8‘“

Figure 3. Effect of hr-t mutants on A2 replication.

NIH-3T3 cells were infected with a mixture of A2 and an
hr-t deletion mutant at the MOI ratio of 1:10. The strain
of hr-t mutant used is indicated above each lane. Viral DNA
was analyzed as described in Fig.1. Output of A2 viral DNA
in the mixed infection (lanes 2-7) is compared to that from
the single infection with A2 virus (lane 1). The band of
702 bp represents the fragment #4 of A2, and lower band in
each lane is the fragment #4 of the respective hr-t mutant

used in the mixed infection.

74
we found a 50% and 93% decrease at the ratio of 1 to 3 and 1
to 10, respectively (Compare lanes 1-3, Fig.4). In
contrast, 18-5 did not exert a strong competitive effect on
82 (lanes 4 and 5, Fig.4), suggesting the existence of cis-
advantage growth of 82 over 18-5.

In fact, a significant competition was clearly observed
at the relative ratio of 1:1, when the MOI was increased
from 1 to 10 (lane 6, Fig.4). Less discrepancy between the
two viruses at the MOI of 1 could be due to more of a
population than a competition effect, and to the replication
incompetence of 18-5. This is consistent with the previous
interpretation that 82 replicates more efficiently than does
18-5.

Wis. To examine
the kinetics of competition, the time course of viral DNA
replication in a mixed infection with A2 and 82 was
followed. Fig.5 shows the analyses of low molecular weight
DNA collected at 4, 24, 48, and 72 hours post infection. To
get an adequate picture of the results, we applied various
amounts of DNA sample (see Figure legend), and analyzed
samples of the two earlier time points (4 and 24 hours)
hours separately from those of the later time points (48 and
72 hours). We observed that the onset of viral DNA
synthesis occurred before 24 hours post infection (Compare

Fig.5A and 58). At 24 hours post infection, A2 DNA

75

MOI
mic I/I I/3 I/103/IIOIIIO/Io bp,

1
18-5>~ - ~ Ce" <652

82> -

   

Figure 4. Competition for DNA synthesis between middle T
.antigen-deficient mutants.

NIH-3T3 cells were doubly infected with strains 18-5
and 82. The MOI ratio of 18-5 to 82 is indicated above each
lane. Accumulated viral DNA was isolated and analyzed as
described in Fig.1. The bands of 652 and 462 bp are the

fragments #4 of strains 18-5 and 82, respectively.

76
replication at the MOI of 1 reached the same level in both
single and mixed infections (Fig.58), and the ratio between
A2 and 82 genomes maintained the same as that in the input.
Competition occurred between 24 and 48 hours post infection,
since less A2 DNA accumulated in the mixed infections than
in the single infections (Compare lanes 2 to 3, Fig.5C and
5D). Whereas, the level of 82 DNA replication reached the
same level in both single and mixed infections (Compare
lanes 3 to 4, Fig.5c and 5D). Competition resulted in lower
ratios of A2 to 82 than that in the input (Compare lanes 3
in Fig.5C and 5D to 5A and 58). The repression of A2 DNA
synthesis in the mixed infection suggests that not every A2
genome entered rounds of replication after 24 hours post
infection.

When comparing the level of DNA accumulation in those
single infections using A2 virus, we found that the yield
obtained from the MOI of 1 never reach the same level as
that from the MOI of 10 (at least till 72 hours post
infection) (Compare lanes 1 to 2 in Fig.5). The increased
capacity of DNA synthesis at a higher multiplicity can be
attributed to an increase in the number of infected cells
and mitogenic effect induced by capsid attachment (46).
Moreover, the ratios of accumulated DNA between the two
infections were kept constant throughout the time course.
This implies that only a fixed portion of DNA molecules are

in the replication pool.

77

Figure 5. Kinetics of viral DNA synthesis.

NIH-3T3 cells were infected with wild-type A2 alone at
the MOI of 10 (lanes 1), and 1 (lanes 2): A2 and 82 at the
MOI ratio of 1:10 (lanes 3), or with 82 alone at the MOI of
10 (lanes 4). Viral DNA isolated at 4 (A), 24 (B), 48 (C),
and 72 (D) hours post infection was analyzed as described in
Fig.1, except for various amounts of samples was used. The
relative volumes used were one-half, one-fourth, one-tenth
'from the sample of 4, 24, 48 and 72 hours post infection,

respectively.

 

32>

A2>

82>

78

 

 

(D)

79

WW We also
analyzed viral DNA accumulated in the lysates from mixed
infections. Competition was also observed: for instance,
less A2 DNA replicated in the presence of excess amounts of
B2 (Fig.6A). To analyze whether and how the replication
competition is represented in the viral population, we
quantitated the yield of infectious particles in lysates
from doubly infected NIH-3T3 cells harvested at five days
post infection. Concentration of infectious virus in the
lysate was determined by plaque assay on NIH-3T3 cells.
Since this cell line is nonpermissive for hr-t mutants (3),
the titer obtained from this assay would represent mainly
the concentration of wild-type A2 virus in the lysate.
Results shown in Table 1 indicate a dosage dependent
decrease of wild-type virus yield in the presence of mutant
virus. We also analyzed encapsidated viral DNA in the same
lysates by treating the lysate with DNaseI prior to
extracting virion DNA (Materials and Methods). Such DNA was
then digested with HspI to determine the ratio of the two
parents. Results showed a consistency with that from the
plaque assays, in which less encapsidated A2 DNA existed in
the lysate from mixed infections (Fig.68). Moreover, the
magnitude of the decrease in virus production corresponded
to that in encapsidated A2 DNA. The fact that the ratio of

A2 to 82 are similar in virions and in the whole lysate

80

Total Encapsidoted

 

 

NKDI I I ,
ratio 1/0 1/1 1/3 1/10 l/Oi/i 1/31/10

   

A2> “(a .-

B?:>.

 

_ - \ .w ' 1.5:" I I ll ‘

(A) (3)

Figure 6. Accumulation of viral DNA in virions and whole
lysate.

Viral lysates from A2 and 82 mixed infected NIH-3T3
cells were collected at 5 days post infection. Ratios of A2
to 82 in the infection mixture are given. Total (A) and
encapsidated (8) viral DNAs in each lysate were isolated as
described in Materials and Methods, and analyzed as

described in Fig.1.

81

Table 1. Yield of infectious virus from mixed infections.

 

 

 

Yiral_§train 891° Yield°
w'r‘ W10 5)
A2 32 1:0 135

1:1 102
1:3 15
1:10 1
12 13-5 1:1 109
1:3 80
1:10 6
A2(e82) 32 1:0 174
1:1 112
1:3 60
1:10 6

 

"bstrains of middle T-proficient‘I and middle T-deletion
mutantID used in the mixed infection.

°Ratio of multiplicities between the parental viruses
in the infection.

°Summary of viral yields in lysates collected from
mixed infections. Concentration of infectious virus was
determined by plaque assay on NIH-3T3 cells.

82
(Compare Fig. 6A to 68) implies that there is no preference
of packaging either A2 or 82 DNA in the double infection.

121531114121:

Polyomavirus relies mainly on the host's proteins and
factors for DNA replication (7). The activities of these
cellular components are believed to be well regulated during
cell growth in the cell cycle, cellular differentiation, and
cellular transformation. However, Py is able to bypass the
cell-cycle regulation and reinitiates several rounds of DNA
replication during the interval when host DNA has replicated
only once. The experiments reported above were designed to
investigate the ability of a Py genome to replicate in the
host cell. For this purpose, we measured the DNA
replication of a wild-type strain in NIH-3T3 cells in the
presence of various amounts of a viable deletion mutant
which was used as a competitor. Accumulated DNA of the two
viruses was analyzed by restriction endonuclease digestion
followed by electrophoresis which resolved the two parental
genomes. The overall patterns of wild-type DNA yield
derived from such infections indicate that NIH-3T3 cells
have a limited capacity for Py replication, since the wild-
type DNA was decreased by the mutant strain in a dosage
dependent manner.

We also found that different competitor strains

inhibited DNA replication of the same wild-type to various

83
degrees depending on the context of enhancer sequences of
each mutant. Compared to the wild-type, strong competitors
contain reiterations of the A domain (1, Chapter 3),
suggesting that one of the targets for the competing factors
is located in the A domain of Py enhancer. It also implies
that a Py strain containing multiple copies of the A
enhancer domain has ale-advantage for DNA replication.

All of the rearranged enhancer sequences tested in this
study are derived from naturally isolated hr-t mutants. A
previous report has described a "dominant lethal effect" for
this group of Py mutants, in which hr-t strains exert a
growth inhibition effect on wild-type strains (10). Our
current studies suggest that competition for the A enhancer
binding factors accounts for (at least part of) the
"dominant lethal effect". Hr-t mutants have been selected
for their transformation deficiency, and are able to grow in
Py-transformed mouse 3T3 cells but not in normal mouse
fibroblasts (3). A viral function expressed during
transformation is thus proposed to be a prerequisite for
productive infection of hr-t mutants. The defects of these
viruses arise from their lack of middle T and small T
antigens (9,16). Other variations in the regulatory regions
of hr-t mutants have been noticed before (16), and detailed
sequence analyses of seven of the hr-t strains demonstrate

reiterations in the A enhancer domain (1, Chapter 3).. The

84
commonality of reiterated A enhancer domain found among hr-t
mutants leads to a speculation that this has been a
compensation (co-evolution) event for viruses which lose
their transformation function.

Duplications in the A enhancer domain have been found
in several middle T antigen-proficient strains of Py
(21,32). When compared to a strain with single copy of the
A domain, these viruses display little cis-advantage for
growth in normal mouse fibroblasts (21, Chapter 3). In
contrast, we show that hr-t strain with sequence repeats has
strong cur-advantage for viral DNA replication. It suggests
that lack of middle T and/or small T antigens casts a
selection pressure against those hr-t strains with single
copy of the A domain. In fact, we have made efforts to grow
up a chimeric hr-t mutant 82(eA2) which contains coding
sequences from strain 82 and the noncoding region of wild-
type A2. The difficulties in raising this particular virus
and in preparing high-titer viral stock of strain 18-5
support this assumption. Such might be the case if middle T
or small T antigen activate viral DNA synthesis by inducing'
the cellular factors that specifically bind to sequences in
the A enhancer domain.

PEA1, PEA2 and PEA3 are three characterized, distinct
nuclear proteins that bind to the A enhancer (22). The fact

that middle T antigen induces Py replication by mediating

85
the activities of PEA1 and PEA3 (43) links the function of
this transforming protein to a role in viral DNA
replication. Furthermore, it has been suggested that hr-t
gene induces a permissive state for Py growth in normal
mouse cell, which can be bypassed in some primary cells or
cells transformed by other viruses (37). Whether the amount
of PEA1 and PEA3 in the host cells is one of the
determinants of permissivity (or host range) for Py DNA
replication in normal mouse cell lines will be further
discussed in Chapter 3.

Py regulatory DNA includes an ori-core for replication
and enhancer sequences. The organization of regulatory
sequences in the viral genome, and the interactions of these
sequences with virally encoded large T antigen and a variety
of cellular factors are responsible for initiating viral DNA
replication as well as for viral adaptability (2,8). The
mechanism by which the enhancer activates DNA replication
remains unclear, it is thought that binding of factors to
the enhancer may prevent nucleosome assembly and allow
formation of a initiation complex at the origin region of
the viral genome. Thus we speculate that, in the presence
of limiting amounts of nuclear factors, only those viral
progeny DNA capable of binding to enhancer factors
effectively proceed to subsequent rounds of DNA synthesis.

In competition experiments, competitor strains with

86
reiterated factor binding sites strongly interfered the
replication of wild-type DNA. This suggests that amplified
A enhancer sequences in the mutant genome sequester PEA1 and
PEA3 factors, thus reducing the number of available factor
for the wild-type enhancer and leading to failure to form a
replication complex at the origin. Radiolabelling
experiments have shown that the majority of Py progeny are
removed from the replicating pool following their initial
synthesis, suggesting the existence of a rate limiting step
for initiation (31). Consistent with this, our kinetic
studies of competition showed that a greater proportion of
replicated molecules reenter the replication pool at early
rather than at late time post infection. In both single and
mixed infections A2 DNA replicated to the same level till 24
hours post infection, whereas the level of A2 DNA was lower
in the mixed infections during later courses of the
infection. Possibly, as the number of total viral DNA
molecules exceeded the amounts of PEA1 and PEA3 factors, and
competition between viral genomes limited the accessibility
of enhancer factors for A2 genomes in later rounds of DNA
replication.

Results of competition for DNA replication were
reflected in the output of viral particles, since the yield
of infectious wild-type virus was depressed in the presence
of mutants. The ratio between two viral genomes in the

virions was the same as that in the whole lysate, which

87
suggests that the different enhancer structure did not
affect encapsidation. Furthermore, we found that in both
single and mixed infections only a portion (~30%) of viral
genomes were effectively encapsidated, which can be due to
the underphosphorylation of capsid proteins and inefficient
viral assembly. It has been reported that middle T antigen
plays an important role in maturation of the Py virion by
phosphorylating capsid protein VP1 (13). In the mixed
infection, the level of middle T antigen expression could be
affected by two events: (1) competition for PEA1 and PEA3
results in competition for transcription, and (2)
outcompeted wild-type DNA provides fewer templates for
transcription. In analyses of virion DNA and plaque assays,
a sharp decrease of A2 was readily observed at higher input
ratio (1 to'3 rather than 1 to 10), which might result from
the synergistic effects of the above two factors.

In conclusion, this report suggests that normal mouse
3T3 cells have a limited capacity for replication of viral
genomes. One of the limiting host factors appears to
interact with enhancer A domain of the Py genome, since
strong competition can be observed for viruses with
reiterations in this region. Middle T antigen may play an
essential role in viral DNA replication by inducing the
activities of those A enhancer binding factors, thus

expanding cellular capacity to facilitate viral growth. The

88
mechanism of how a factor-enhancer complex is involved in

the initiation of DNA synthesis remains to be understood.

10.

11.

89

Reference

Amalfitano, A. 1989. Polyoma mutants with altered
enhancer regions redefine subdomains required for
replication in tissue culture systems, pp.35-67. In
Inggig: The role of the polyoma enhancer region in

replication of the virus In 21919 and in 2119.

Amati, P. 1985. Polyoma regulatory region: A potential
probe for mouse cell differentiation. Cell 43:561-562.

Benjamin, T.L. 1970. Host range mutants of polyoma
virus. Proc. Natl. Acad. Sci. USA 67:394-399.

Campbell, B.A., and L.P. Villarreal. 1988. Functional
analysis of the individual enhancer core sequences of
polyomavirus: cell-specific uncoupling of DNA
replication from transcription. Mol. Cell. Biol.
8:1993-2004. .

Caruso, M., C. Iacobini, C. Passananti, A, Felsani, and
P. Amati. 1990. Protein recognition sites in
polyomavirus enhancer: formation of a novel site for
NF-l factor in an enhancer mutant and characterization
of a site in the enhancer D domain. EMBO J. 9:

de Villiers, J., W. Schaffner, C. Tyndall, S. Lupton,
and R. Kamen. 1984. Polyoma virus DNA replication
requires an enhancer. Nature (London) 312:242-246.

DePamphilis, M.L., and M.K. Bradley. 1986. Replication
of SV40 and polyoma virus chromosomes. pp.99-246. In N.
P. Salzman (ed.). The papovaviridae. Plenum Publishing
Corp.. New York.

DePamphilis, M.L. 1988. Transcriptional elements as
components of eukaryotic origins of DNA replication.
Cell 52:635-638.

Feunteun, J., L. Sompayrac, M. Fluck, and T. Benjamin.
1976. Localization of gene functions in polyoma virus
DNA. Proc. Natl. Acad. Sci. USA 73:4169-4173.

Fluck, M.M., Staneloni, R.J., and Benjamin, T.L. 1977.
Hr-t and ts-a: two early gene functions of polyoma
virus. Virology 77:610-624.

Franke, 8. and W. Eckhart. 1973. Polyoma gene function
required for viral DNA synthesis. Virology 55:127-135.

12.

13.

14.

15.

16.

17.

18.

19.

20.

2]"

22.

90

Fujimura, F. 1986. Nuclear activity from F9 embryonal
carcinoma cells binding specifically to the enhancers
of wild-type polyoma virus and PyEC mutant DNAs.
Nucleic. Acids Res. 14:2845-2861.

Garcea, R.L., and T.L. Benjamin. 1983. Host range
transforming gene of polyoma virus plays a role in
virus assembly. Proc. Natl. Acad. Sci. USA 80:3613-
3617.

Gaudray, P., C. Tyndall, R. Kamen, and F. Cuzin. 1981.
The high affinity binding site on polyoma virus DNA for
the viral large T protein. Nucleic Acids Res. 9:569?-
5710.

Griffin, B.E., M. Fried, and A. Cowie. 1974. Polyoma
DNA: a physical map. Proc. Natl. Acad. Sci. USA
71:2077-2081.

Hattori, J., G. Carmichael, and T.L. Benjamin. 1979.
DNA sequence alterations in hr-t deletion mutants of
polyoma virus. Cell 16:505-513.

J‘

Hirt, B. 1967. Selective extraction of polyoma DNA from
infected mouse cell cultures. J. Mol. Biol. 26:365-369.

Johnson, P., W. Landschulz, B. Graves, and S. McKnight.
1987. Identification of rat liver nuclear protein that
binds to the enhancer core element of three animal
viruses. Genes Dev. 1:133-146.

Katinka, M., M. Yaniv, M. Vasscur, and D. Blangy. 1980.
Expression of polyoma early functions in mouse
embryonal carcinoma cells depends on sequence
rearrangements in the beginning of the late region.
Cell 20:393-399.

Lania, L., M. Griffiths, M., B. Cooke, Y. Ito, and M.
Fried. 1979. Untransformed rat cells containing free
and integrated DNA of a polyoma nontransforming (hr-t)
mutant. Cell 18:793-802.

Maione, R., C. Passananti, V. De Simone, P. Delli-Bovi,
G. Augusti-Tocco, and P. Amati. 1985. Selection of
mouse neuroblastoma cell-specific polyoma virus mutants
with stage differentiative advantages of replication.
EMBO J. 4:3215-3221.

Martin, H.E., J. Piette, M. Yaniv, W. Tang, and W.R.
Folk. 1988. Activation of the polyomavirus enhancer by
a murine activator protein 1 (APl) homolog and two

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

91

contiguous proteins. Proc. Natl. Acad. Sci. USA
85:5839-5843.

Martinez-Salas, E., D.Y. Cupo, and M.L. DePamphilis.
1988. The need for enhancers is acquired during
mammalian development with formation of a diploid
nucleus Genes Dev. 2:115-126.

Muller, W.J., C.R. Mueller, A.M. Mes, and J. Hassell.
1983. Polyomavirus origin for DNA replication comprises
multiple genetic elements. J. Virol. 47:586-599.

Ostapchuk, P., J.F.X. Diffley, J.T. Bruder, B.
Stillman, A.J. Levine, and P. Hearing. 1986.
Interaction of a nuclear factor with the polyomavirus
enhancer region. Proc. Natl. Acad. Sci. USA 83:8550-
8554.

Piette, J.M., H. Kryszke, and M. Yaniv. 1985. Specific
interaction of cellular factors with the B enhancer of
polyomavirus. EMBO J. 4:2675-2685.

Piette, J., and M. Yaniv. 1987. Two different factors
bind to the alpha-domain of the polyoma virus enhancer,
one of which also interacts with the SV40 and c-fas
enhancers. EMBO J. 6:1331-1337.

Pomerantz, B.J., and J.A. Hassell. 1984. Polyomavirus
and simian virus 40 large T antigens bind to common DNA
sequences. J. Virol. 49:925-937.

Prives, C., Y. Murakami, F.G. Kern, W. Folk, C.
Basilico, and J. Hurwitz. 1987. DNA sequence
requirements for replication of polyomavirus DNA in
vivo and in vitro. Mol. Cell. Biol. 7:3694-3704.

Rochford, R., C.T. Davis, K.K. Yoshimoto, and L.P.

~Villarreal. 1990. Minimal subenhancer requirements for

high-level polyomavirus DNA replication: a cell-
specific synergy of PEA3 and PEA1 sites. Mol. Cell.
Biol. 10:4996-5001.

Roman, A. 1979. Kinetics of reentry of polyoma progeny
form I DNA into replication as a function of time
postinfection. Virology 96:660-663.

Ruley, H.E., and M. Fried. 1983. Sequence repeats in a
polyoma virus DNA region important for gene expression.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

92

Schaffhausen, B., J. Silver, and T. Benjamin. 1978.
Tumor antigens in cell productivity infected by wild-
type polyoma virus and mutant NG-18. Proc. Natl. Acad.
Sci. USA 75:79-83.

Silver, J., B. Schaffhausen, and T. Benjamin. 1978.
Tumor antigens induced by non-transforming mutants of
polyoma virus. Cell 15:485-496.

Soeda, E., J.R. arrand, N. Smolar, J.E. Walsh, and B.E.
Griffin. 1980. Cording potential and regulatory signals
of the polyoma virus genome. Nature (London) 283:445-
453.

Southern,E.M. 1975. Detection of specific sequences
among DNA fragments separated by gel electrophoresis.
J. Mol. Biol. 98:503-517.

Staneloni, R., M.M. Fluck, and T.L. Benjamin. 1977.
Host range selection of transformation defective hr-t
mutants of polyoma virus. Virology 77:598-609.

Tanaka,K., K. Chowdhury, K.S.S. Chang, M. Israel, and
Y. Ito. 1982. Isolation and characterization of polyoma
virus mutants which grow in murine embryonal carcinoma
and trophoblast cells. EMBO J. 1:1521-1527.

Tseng, R.W., and F.K. Fujimura. 1988. Multiple domains
in the polyomavirus B enhancer are required for
productive infection of F9 embryonal carcinoma cells.
J. Virol. 62:2890-2895.

Veldman, G.M., S. Lupton, and R. Kamen. 1985.
Polyomavirus enhancer contains multiple redundant
sequence elements that activate both DNA replication
and gene expression. Mol. Cell. Biol. 5:649-658.

Wasylyk, B., C. Wasylyk, P. Flores, A. Begue, D.
Leprince, and D. Stehelin. 1990. The c-ets proto-
oncogenes encode transcription factors that cooperate
with c-Fos and c-Jun for transcriptional activation.
Nature (London) 346:191-193.

Wasylyk, C., P. Flores, A. Gutman, and B. Wasylyk.
1989. PEA3 is a nuclear target for transcription
activation by non-nuclear oncogenes. EMBO J. 8:3371-
3378.

Wasylyk, C., J. Schneikert, and B. Wasylyk; 1990.
Oncogene v-jun modulates DNA replication. Oncogene
5:1055-1058.

44.

45.

46.

93

Winocour, E. 1963. Purification of polyomavirus.
Virology 19:158-168.

Wobbe, C.R., F. Dean, L. Weissbach, and J. Hurwitz.
1985. Intdtro replication of duplex circular DNA
containing the simian virus 40 DNA origin site. Proc.
Natl. Acad. Sci. USA 82:5710-5714.

Zullo, J., Stiles, C.D., and Garcea, R.L.1987.
Regulation of c-myc and c-fos m-RNA levels by
polyomavirus: distinct roles for the capsid protein VP-
1 and the viral proteins. Proc. Natl. Acad. Sci. USA
84:1210-1214.

Chapter 3.

Functional Interdependence between

Middle T Antigen and the Polyoma A Enhancer Region1

ABSTRACT

We show that hr-t middle T antigen deficient mutants
which arose naturally and were selected for transformation
dependent growth, have evolved with rearrangements in the
enhancer. The variety of the rearrangements strongly
suggests that each one of them represents an independent
event. The rearrangements have one common feature, i.e.,
the reiteration of the A enhancer domain. This imparts a
eta-advantage for replication to any strain (wild type or
middle T antigen deficient) which harbors them. In the
absence of the reiteration, middle T antigen deficient
mutants have a strong defect in viral DNA replication. The
previously reported "dominant lethal effect" of hr-t mutants
in mixed infections with wild type can thus be explained as
an enhancer-mediated competition effect. When two strains
only differ by a duplication of the A element, the strong

ale-advantage of the duplication carrying strain can be

relieved partially by TPA or serum, and totally by middle T

 

1Ming Chu Chen, Andrea Amalfitano, Makda Fisseha, and
Michele M. Fluck, Department of Microbiology,
Interdepartmental Program in Molecular Biology, Michigan
State University, East Lansing, MI 48824-1101.

94

95
antigen. This suggests that these three agents, whose
common targets are PEA1 and PEA3 motifs in the A enhancer
element, act via the same or similar mechanism(s). The
experiments suggest that the middle T antigen is much more
effective than TPA or serum in inducing viral DNA synthesis

via the activation of PEA1 and PEA3.

96

IDIIQQDQLIQB

Hr-t mutants are a group of host range transformation-
defective mutants of polyomavirus (Py), which grow well on
Py transformed mouse 3T3 (Py-3T3) cells but poorly on normal
NIH-3T3 cells (2). They were isolated based on the
assumption that integrated Py genes essential for viral
growth would complement mutant strains defective in the
homologous function. A variety of cell types were later
found to be permissive for the growth of hr-t mutant without
being transformed by Py. These include certain primary
mouse cell lines, and some retrovirus-infected or -
transformed 3T3 cells (15,42). Thus, it was hypothesized
that efficient growth is achieved via gene activation of a
permissive state in the host cell by the hr-t gene
product(s), and that permissivity related cellular genes can
be expressed constitutively or regulated independently of Py
in some cell types (16). Whether the permissive state in
non-Py transformed cells involves the same mechanism as that
stimulated by the hr-t gene product(s) remain undetermined.

Cells infected with hr-t mutants lack certain traits
induced by wild type infection, i.e., the ability to undergo
multiple rounds of cell DNA synthesis, lectin
agglutinability, and cellular transformation (16,38). The
hr-t mutation has been mapped to the early gene coding
region (11,18) and shown to alter the expression of small T

and middle T antigens (37,39). Small T antigen, whose

97
function is still not very clear, has been shown to be
involved in productive infection and anchorage-independent
growth of transformed cells (6,26,31,46). Middle T antigen
is the major transforming protein of Py (45), which
associates with tyrosine kinases of the src family and
phosphatidylinositol (PI) 3-kinase in the plasma membrane
(3,5,7,24,44). Signals induced at the membrane are further
translocated into the cytoplasm and the nucleus. In the
cytoplasm, middle T antigen induces the phosphorylation of
several proteins, including VP-1 (14) and the oncoprotein
C-Raf (28). VP-l phosphorylation appears to be essential
for efficient encapsidation and assembly of infectious
particles (13). Phosphorylation of C-Raf stimulates
serine/threonine kinase activity of c-Raf, which may
generate a cascade of signal transduction in the cell (28).
Middle T antigen also enhances gene expression by activating
cellular transcription factors PEA1 and PEA3 (49,50), which
are the mouse homologs of human AP1 and c-ets, respectively
(27,48). These two factors mediate transcription by binding
to specific sequences located in the A enhancer element of
Py (27). PEA1 and PEA3 mediated transcription is also
stimulated by other oncogene products, TPA, and serum
(22,53). Collectively, the activities of middle T (and
small T) antigen alter the cellular transcription capacity,
inducing the S phase of the cell cycle which is required for
viral growth.

98

Hr-t mutants have been known to exert an inhibitory
effect on virus yield in mixed infection with wild type
(12). We recently observed that hr-t mutants interfere with
wild type at the level of viral DNA synthesis (Chapter 2).
This dominant lethal effect conferred by hr-t mutant is due
to competition for enhancer binding factors present in
limiting amounts. It has been noted that hr-t mutants
differ from the prototypic strains A2 and A3 in the enhancer
region (18). The Py enhancer has been divided into two
domains, A and 8, either of which can activate viral DNA
transcription and replication 1n 919 (47). Each element has
distinct binding sequences for different cellular enum-
acting factors (29,47). Different cell types demonstrate
dependence on various enhancer subdomains to reach high
levels of viral DNA synthesis, suggesting that each binding
motif contributes significantly to DNA replication
(4,20,35). The A enhancer element has been shown to
activate transcription in response to TPA and six
transforming oncogene products (22,49,53). TPA also
stimulates the replication of DNA molecules containing the
A enhancer and the Py origin (30). Both TPA and expression
of middle T antigen affect protein kinase C activity and
alter the physiological state of the host. Therefore, it is
likely that middle T antigen enhances viral DNA replication

by activating PEA1 and PEA3.

99

In this report, we have sequenced the enhancer
alterations in hr-t mutants. We demonstrate a selective
advantage for replicating virus with a duplicated A enhancer
element. The selective pressure was accentuated in the
absence of middle T antigen, in that a middle T deficient
mutant containing the duplication strongly outcompeted a
middle T deficient strain with single copy of the A enhancer
element. The competition could be relieved by introducing
TPA, serum, or middle T antigen. The results suggest that
amplification of the A enhancer element complements the loss

of middle T antigen in hr-t mutants.

W

Q9IIg_nn9_11:99991 Mouse NIH-3T3 cells (23) were
maintained in Dulbecco's modified Eagle's medium (DMEM)

supplemented with 10% heat inactivated newborn calf serum
and grown in a humidified incubator with 5% CO, at 37°C.
Kidneys were removed from approximately lo-day old random-
bred Balb/C mice and used to prepare primary cultures of
baby mouse kidney (BMK) cells (52). BMK cells were grown in
DMEM supplemented with 10% horse serum.

All viral strains were grown from plaque purified
viruses on BMK cultures. Strain A2 is a large-plaque virus
which is used as a reference wild type in our laboratory
|(17). When restriction endonuclease patterns of hr-t

mutants were established (18), it became evident that these

 

100
mutants were derived from at least two parental strains, A
and 8. Both wild types A and B are of a small plaque
phenotype. Wild type A harbors an 11 base-pair (bp)
deletion of one of the large T antigen binding sites. Hr-t
mutants NG-18, NG-59, HA-33, 3A-3, 82, 3B-2, 3B-4, and 30'b
belong to this group. The other parental strain (8), which
bears three large T antigen binding sites, harbors an
additional HpaII site in the late coding region resulting in
a split.HpeII fragment 1. Hr-t strains NG-23, A2, A8, A9,
A11, 685, SD-15, and II-5 are representatives of this
pattern. Mutants NG-18 (Group A) and NG-23 (Group B) were
derived at the same time from treatment with
nitrosoguanidine, while strain HA-33 (Group A) was prepared
from virus treated with hydroxylamine (2). Mutants A2, A8,
A9, A11, and 68-5 (Group B), as well as 3A3, 82, 38-2, and
3B-4 (Group A) were derived in a later experiment from wild
types treated with ICR-191 (42). Furthermore, mutant II-5
(Group B) was isolated from a wild type lysate after two
high multiplicity passages on Py-3T3 cells (42). Mutants
SD-15 (Group B) and 30'b (Group A) were isolated from the
light side of a band of wild type virus in a cesium chloride
equilibrium density gradient (42). Middle T antigen
deficient strain 18-5 is derived from wild type strain A2 by

deleting approximately 50 hp at 83 map units in the MspI

fragment #4 (25).

101

Enn9n99r_9n9n9n9999 Viral DNA was prepared from
infected NIH-3T3 cells by the method of Hirt (21). Viral
DNA was digested with the restriction enzymes.Bc1I and.Bg1I.
This generates two fragments: a small (approximately 350 bp)
fragment containing the noncoding region including the
enhancer, and a large (approximately 4,950 bp) fragment
containing the coding region (See Fig.2). The two fragments
were exchanged between various pairs of parental viral
genomes. For this purpose, digested DNA was resolved by
agarose gel electrophoresis, and each BclI-BgII fragment was
excised. A total of 50 ng of viral DNA fragments were
ligated in 10 pl of ligation buffer (66 mM Tris, pH 7.5: 10
mM MgC12: 1 mM DTT: and 1 mM ATP) for 16 hours at 16°C using
T4 DNA ligase (New England Biolab). For transfection, 10 ng
of DNA was applied to 1.5::105 NIH-3T3 cells/35 mm dish using
500 ug/ml DEAE dextran at 33°C for 1 hour. The dishes were
washed and overlayed with DMEM containing 0.9% agar and 5%
newborn-calf serum. Plaques were picked after staining with
neutral red, and viral stocks were grown on NIH-3T3 cells.

The names of the resulting viruses consist of the name
of the parental virus contributing the fragment with coding
sequences followed in parentheses by the name of the virus
contributing the fragment with enhancer region. For
instance, A2(e82) is an A2 virus in which the enhancer-

origin region has been replaced with the corresponding

102
region from the 82 virus.

WW Hirt DNA
from each of the recombinant viruses was isolated from
infected NIH-3T3 mouse cells and digested withIBcII and.PscI
to release the enhancer containing 600-700 bp fragment.

This fragment was purified by low melting temperature
agarose electrophoresis and ligated into the polylinker
region of either the bacteriophage M13mp18 or M13mp19 RF DNA
digested with BamHI and PstI. Insert containing clones were
isolated, and single stranded DNA of each was subjected to
Sanger dideoxy sequencing from opposite directions using
either the M13 universal primer or a 17mer on the early side
of the polyoma 8311 site.

In199919n99 Cells plated at a density of 2x105
cells/60 mm dish were either singly or doubly infected with
viral strains at a multiplicity of infection (MOI) of 1 or
10 plaque forming units (PFU)/cell. The relative ratio of
the 2 parental genomes in the infection mix was originally
established fromthe titer of the 2 parental viral stocks,
and further confirmed by hybridization analysis of the viral
sequences extracted from the cells at 4 hours post
infection. Cells were fed with DMEM supplemented with
newborn calf or horse serum as described in the text. TPA
(Sigma Chemical Co.) was dissolved and diluted with ethanol

and 10-100 nM was added into the incubation medium of each

103
dish.

Anegg1§_9fi_y1nnI_nn59 Low molecular weight DNA was
extracted using the Hirt procedure (21), extracted with
phenol and chloroform, precipitated with isopropyl alcohol,
and resuspended in 50 ul of 10 mM Tris-HCl (pH 7.4), and 1
mM EDTA. The DNA was digested with restriction endonuclease
prI, which cuts polyoma DNA 8 times (18). The size
differences in the enhancer region between viral strains can
be distinguished by different mobility of the Meal fragment
3 (18). The restriction fragments were separated on 1.8-
2.0% agarose gels in 0.089 M Tris-borate, and 0.002 M EDTA
(pH 8.0) and transferred to nitrocellulose or hybond
membranes (Amersham Corp.) according to the Southern
procedure (41). The DNA was fixed onto the blots and probed
to a viral DNA fragment which specifically hybridized to the
MspI fragment 3. The polyoma HspI-3 restriction fragment was

isolated by HindIII and EcoRI digestions of a pGem plasmid

containing this subfragment (which was originally inserted
into the 0181 site in a pBR322 plasmid, and recloned into
the HindIII, EcoRI sites of the pGem vector), and resolved in
low melting point agarose (FMC Bioproducts). The probes
were labeled to a specific activity of 1-2 x 10" cpm/ug
[”pJ-dCTP (3,000 Ci/mmole) using a multiprime DNA labeling
kit (Amersham Corp.). Hybridization was carried out at 65%:

for 1 day in 2X SSC/lx Denhardt's or in 5x SSPE /5X

104
Denhardt's/0.5% SDS solution (0.1 ml/cmz with 1-2 x 105

cpm/ml of the labeled probes in hybridization solution).

BEEHLIE

WWW
1nfi99919n99 We have recently demonstrated that in mixed
infections of NIH-3T3 cells, replication of wild-type strain
A2 is inhibited by increasing the amount of an hr-t mutant
strain 82. The competitive effect of hr-t mutant 82 was
shown to be greater than that of a middle T antigen
defective mutant isolated by mutagenesis without the
biological selection used for hr-t mutants (Chapter 2).
This interference effect had been noted before (12) and
referred to as ”dominant lethal effect". To examine whether
the "dominant lethal effect" exerted by strain 82 on wild-
type viral DNA synthesis is a trait common to all hr-t
mutants, we carried out mixed infections using wild type A2
and 15 hr-t mutants. In these, A2 was added at a
multiplicity of 1, and the hr-t parent at a multiplicity of
10. Viral DNA extracted at 60 hours post infection was
analyzed by ”8131 digestion. Except for HA-33, all other
hr-t mutants we used contain deletions of variable sizes in
impI fragment 4 (18). In all cases, the level of A2 DNA was
decreased in the mixed infections compared to the single

infection with A2 alone (Fig.1). The output ratio

105

Figure 1. Effect of coinfection with hr-t mutants on A2
viral DNA synthesis.

NIH-3T3 cells were infected at a multiplicity of l
PFU/cell of A2 with (lanes 1-15) or without (lane 16) an
hr-t mutant which was used at a multiplicity of 10 PFU/cell.
At 60 hours post infection, low molecular weight DNA was
isolated. One tenth of the sample was digested with HspI,
electrophoresed on a 1.8% agarose gel, transferred to
nitrocellulose, and hybridized to a probe specific to the

polyoma MspI restriction fragment #4.

Sl-(JS
S-II
6V
8V

SZON
Z8
CCVH

808
$89

788
388

8V8

llV

8i9N

hr-i

106

 

10 H 12 13 I4 15 16

9

107
between A2 and mutant DNA varied with each mixed infection.
In most cases except with NG-18, the band corresponding to
wild type DNA cannot be detected, while bands corresponding
to mutant DNA are approximately of similar intensity in all
infections. This suggests that each hr-t mutant displays a
certain extent of dominant lethal effect for viral DNA
synthesis.

WW
nntnn999 Since the dominant lethal effect of hr-t mutant 82
is associated with the enhancer region of this virus, which
contains a duplication (Chapter 2), we set out to sequence
another five hr-t mutants between the BcLI and the BgII
sites. The results are illustrated schematically in Fig.2.
The recently derived sequence of 82 as well as the sequences
of NG-59 obtained by Ruley and Fried (36) are also shown for
comparison. The sequence of NG-59 and 82 differ from the
other five hr-t strains in that both contain an 11 bp
deletion encompassing one of the major large T antigen
binding sites (Data not shown). This variation has been
reported between the prototype strains A2 and A3 (40).

Compared to wild types A2 and A3, all seven hr-t
mutants display rearrangements in their enhancer regions.
Strains NG-59 and 82 contain simple duplications of 31 bp

and 42 bp, respectively. SD-15 and II-5 contain 44 bp

108

Figure 2. Comparison of enhancer-origin sequence of hr-t
mutants.

The sequence of each of the hr-t strains is compared to
a restriction map of the A2 strain of Salzman (8). The
sequence of NG-59 is taken from Ruley, et al (36). The
strand equivalent to the early mRNA is presented. Parallel
lines represent duplications and dashed lines indicate their
relation to the genome. Nucleotide substitutions are placed
directly above their locations. Duplicated sequences between
nucleotides 5088 and 5146 are shown at the bottom, leftward
(-) and rightward (*) boundaries of each reiteration in the
hr-t strain are indicated above the corresponding
nucleotide. Nuclear factors PEA3, PEA1, and PEA2 binding

regions within the A enhancer domain are shown.

109

NG‘23 \ G C)?- \ \

 

A3 a g. ‘\

E
A-9 ‘GCJ‘

,SD-15 \ s

32 "<‘——'f‘

.NG'359 her-i
' as

Bi“ Pun]: Firm 3 i1

. y 1 I
l A enhancer 8 enhancer (origin)

5038 ,/’//r i\\\ 5146

GAAAAIGTCAAGTCAéTTAAGCACGAAcicACTAACTcAccccaccrcccccTEchci
PEA3 PEA1 PEA2

 

 

 

 

 

 

110

duplications identical to those described previously for the
P16 and Toronto strains (36). The three remaining viruses
have complicated patterns of rearrangement, with multiple
and more extensive reiterations of the region between
nucleotides 5088 and 5141. Interestingly, many of the
rearrangements have an identical border. The rightward
border of the duplication at nucleotide 5141 is shared by
mutants NG-23, A8, A9, II-5, and SD-15. The leftward border
at nucleotide 5088 is shared by NG-23, A8, and A9. Strains
II-S and SD-15 have the same leftward border at nucleotide
5098. Moreover, in strains A8 and NG-23 this amplification
is coupled with a deletion of 123 bp between nucleotides
5141 and 5264. This deletion encompasses the B enhancer
element of the A2 genome. Only SD-15 and II-5 have the same
rearrangement (except for a point mutation in II-5). All
other five strains differ from each other and from that of
SD-15/II-5.

Though each virus contains a unique alteration, all
have a common feature involving reiteration of the A
enhancer domain. As described in the Materials and Methods,
the seven hr-t mutants characterized here have different
histories. Mutants NG-23, A8, A9, SD-15, and II-5 belong to
the Group 8. Though isolated at different time, they share
a duplication between nucleotides 5098 and 5141. Thus, it

appears that various and independent duplications of the A

111
enhancer domain arose in at least two different parental
strains and were retained in hr-t mutants.

3,: ;“- o‘ - ;._to_;;,,._,- --:._-,:o.
W The enhancers (BclI-BglI fragment) of six
hr-t mutants were introduced one at a time into the A2
genome (see Materials and Methods). These recombinants are
designated A2(eX) where x refers to the origin of the
enhancer fragment. To investigate whether the altered
enhancer sequences affect viral DNA replication, competition
experiments were performed in which pairs of viruses were
used at a multiplicity of 10 for each. These viruses can be
distinguished from each other by the size of their enhancer,
for example HspI fragment 3. To verify the equal dosage of
the two parents in the input, viral DNA was extracted at
four hours post infection, digested with.MspI,

electrophoresed, blotted and hybridized with.MspI fragment 3

as described in Materials and Methods (Data not shown). A
mild advantage in DNA replication was observed for the
enhancer B2 in competition with A2 (lane 1, Fig.3). Strain
A2(eA9) which contains a triplicated A element outcompetes
strain A2(e82) with a duplicated A element (lane 4, Fig.3),
and displays extensive growth advantage over the A2 strain
(lane 3, Fig.3). These results confirm our observation that
NIH-3T3 cells contain limiting factors for DNA replication
whose targets are located in the A enhancer element (Chapter

2).

112

eA9>

932> a
eA2>
9A8>

 

Figure 3. Effect of reiterated A enhancer element on viral
DNA synthesis.

Recombinant viruses were generated by replacing the
enhancer containing BclI-BglI fragment of the strain A2 with
that of an hr-t strain. NIH-3T3 cells were infected at a
multiplicity of 10 PFU/cell of each of the following
viruses: A2 and A2(e82) (lane 1), A2 and A2(eA8) (lane 2),
A2 and A2(eA9) (lane 3), A2(eA9) and A2(e82) (lane 4),
A2(eA9) and A2(eA8) (lane 5). Viral DNA was analyzed as
described in the legend to Fig.1, except that a probe

specific to the polyoma MspI-3 was used in the

hybridization.

113

Though containing a triplicated A enhancer element,
strain A2(eA8) displayed the lowest replication efficiency
among the four strains tested (lanes 2 and 5, Fig.3). This
implies that either a function of the 8 domain in viral DNA
synthesis cannot be compensated by reiterations of the A
domain, or that a site is created for a repressor in the A
rearrangement. Given the similarities between the A8, A9,
and NG-23 rearrangements, the latter is not very likely.
The activities of the A and 8 elements of the Py enhancer in
DNA replication can be redundant or indispensable, depending
on the host cell type (4,33,34,47). Mutation in the 8
element does not totally abolish viral DNA replication,
suggesting that enhancer A element plays the major part in
viral DNA replication in NIH-3T3 cells. Furthermore, our
results suggest that the B enhancer element may also

facilitate Py DNA replication in NIH-3T3 cells.

 

an9199n_9911919n;_9t;91n§1 To examine the effect of middle
' T antigen on viral DNA synthesis in NIH-3T3 cells, we

compared the replication of middle T deficient mutants, 18-5
and 82, to that of wild types, A2 and A2(e82). Low
molecular weight DNA was extracted from infected cells and
analyzed by restriction endonuclease HspI as described in
Materials and Methods. we observed that for those pairs of

viral strains with the same enhancer structure, middle T

114
deficient mutant replicated less efficiently than wild type
in NIH-3T3 cells (Compare lanes 1 to 2, and 3 to 4, in
Fig.4). Underreplication of hr-t mutant DNAs suggests that
middle T and/or small T antigens stimulate viral DNA
synthesis. We also compared the effect of duplicated A
enhancer elements on viral DNA synthesis. Duplication of
the A enhancer resulted in higher yield of DNA synthesis
(Compare lanes 1 to 3, and 2 to 4, in Fig.4). Therefore,
duplication in the A enhancer element exerts a.cis-advantage
for viral growth. At the level of viral DNA synthesis, the
enhancer effect is of the same order of magnitude as the
middle T antigen effect.

eIIe. go o: ‘ - e ,‘ II ee ; ., e:, .!e

y1;nI_pNA_9yn;n99199 As discussed in the introduction,
middle T antigen and TPA appear to affect cells through the

'i on

same pathway. Thus it was of interest to compare their
effect on viral DNA synthesis. For this purpose, infected
cells were incubated in medium containing 1% serum either
treated with 10 or 100 nM TPA, or mock-treated with ethanol.
Under these conditions, TPA treatment had little effect on
the cell division, but cell morphology was altered in that
the cells appeared transformed (data not shown). The
results show that TPA strongly induced DNA synthesis
(approximately five folds increase) of middle T antigen

deficient strains 18-5 and 82 (Fig.5). The effect was more

115

 

I
882 4:3

702 an

 

bp, 1 2 3 4

Figure 4. Effect of middle T antigen on viral DNA
synthesis.

NIH-3T3 cells were infected with a middle T antigen
proficient strain: A2 (lane 1), A2(e82) (lane 3), or with a
middle T deficient strain: 18-5 (lane 2), 82 (lane 4) at a
multiplicity of 10 PFU/cell for each virus. Viral DNA was
analyzed as described in Fig.1, and hybridized to a probe
specific to the whole polyoma genome. Middle T antigen
effect on viral DNA replication was compared by MsnI
fragments #1 between viruses with a single copy of (lanes 1

and 2) and duplicated (lanes 3 and 4) A enhancer element.

116

pronounced for the strain 18-5, which contains a single A
enhancer element (see Materials and Methods). The TPA
effect did not alter the effect showing the duplication of
the A enhancer, since more viral DNA replication was
observed with strain 82 than with 18-5 in the presence of
TPA.

In contrast, in the presence of middle T antigen, no
TPA effect could be observed. Since the full effects of TPA
were observed at the concentration of 10 nM for mutants, and
since increasing the concentration of TPA to 100 nM in wild
type infected cells had no effect on viral DNA synthesis, it
appears unlikely that the absence of stimulating wild-type
DNA synthesis was due to insufficient TPA concentration.
Clearly, the middle T antigen is a more potent inducer of
the viral DNA synthesis than TPA.

W
W To examine
whether TPA and middle T antigen activate Py DNA synthesis
through the same target, we compared the effect of TPA,
middle T antigen, and serum on the competition between
viruses with single and duplicated A enhancer elements.
NIH-3T3 cells were infected at equal multiplicities with
pairs of strains which differ in the enhancer. One pair of
viruses were middle T antigen proficient, whereas the other

pair were middle T antigen deficient. Infected cells were

117

Virus A2 18-5 A2(OB2) 82

TPA ~
("W 0 10100 0 10100 a 10100 0 10100

.0. --6..-..

Figure 5. TPA responsiveness of viral DNA replication.
NIH-3T3 cells were infected with each virus at the
multiplicity of 2 and incubated with or without TPA. Viral

DNA was digested with EcoRI, electrophoresed on a .7%
agarose gel. Hybridization was conducted with a probe for

the whole viral genome (pPy-l).

118
incubated in medium containing either 0.5 or 5% serum, and
with or without TPA (20 nM).

The activation of the cell's capacity for viral DNA
synthesis by middle T antigen and TPA which was already
described in Figs. 4 and 5, is demonstrated again. The
stimulatory effect of TPA is particularly visible in low
serum (Compare lanes 1 and 2, 5 and 6, in Fig.6). In
addition, serum also increases the capacity for viral DNA
synthesis (Compare lanes 1 and 3, 5 and 7, in Fig.6).
Comparing the difference between lanes 1 and 2 with that
between lanes 1 and 3 in Fig.6, it appears that the effect
of serum is similar in magnitude to the effect of TPA and
little synergy is observed with these two agents. Neither
agent is as potent as middle T antigen (Compare A and 8,
Fig.6). Interestingly, the capacity for viral DNA synthesis
is not fully utilized at the low multiplicity, since a big
increase is observed by raising the multiplicity. The
multiplicity effect is maximal in cells infected in low
serum and without TPA treatment (Compare lanes 1 and 5,
Fig.6). Only a small fraction (approximately 15%) of this
effect can be attributed to an increase in the number of
infected cells at the multiplicities used. The mitogenic
effect of capsid attachment at the early step of infection
(54) may contribute to increasing the capacity for DNA

synthesis, or the number of templates may be rate limiting

119

 

 

 

 

 

 

 

 

 

Figure 6. Effect of TPA, serum or both on competition
assays.

NIH-3T3 cells were infected with a mix of middle T
antigen proficient viruses (A), or middle T antigen
deficient viral strains (8). The MOI ratios and incubation
conditions are indicated in the bottom panel. The filters
containing fractionated MspI restriction fragments were
hybridized with a probe specific for the MspI fragment 3.
Panel B was exposed 3 times longer than panel A. MT, middle

T antigen.

120

at lower multiplicities. This suggests that factors other
than those which bind in the A enhancer are at play.

When the fate of the two parental viruses is compared,
the advantage of the template with a reiteration of the A
enhancer element is obvious. In these experiments, the
amount of T antigen available for replication is identical
for both viral strains. The strain containing a duplicated
A enhancer outcompetes a template without reiteration under
most conditions: low serum, absence of TPA or of middle T
antigen. The fact that all three agents: middle T antigen,
TPA, and serum relieve the competition support the
hypothesis that all three affect a factor required for DNA
replication whose target of action lies in the A enhancer
element. In the case of relieving competition, the middle T

antigen appears to be the most potent agent.

 

W Baby mouse kidney (BMK)

cells have been shown to be permissive for the replication
of hr-t mutants, suggesting that the activity of the middle
T antigen is not essential in this cell type (15). In BMK
cells, the contribution of the middle T antigen to the yield
of viral DNA could not be demonstrated: middle T deficient
mutants accumulated viral DNA to a similar extent as middle
T proficient strains (Compare lanes 1 and 2, 3 and 4, in
Fig.7). In contrast, the effect of the A enhancer

duplication could be detected both in the presence and in

121

 

Figure 7. Viral DNA synthesis in BMK cells.

BMK cells were infected with A2 (lane 1), 18-5 (lane
2), A2(e82) (lane 3), 82 (lane 4) at a multiplicity of 10
PFU. Low molecular weight DNA was analyzed as described in

the legend to Fig.1.

122
the absence of the middle T antigen (Compare lanes 1 and 3,
as well as lanes 2 and 4, in Fig.7). As described above for
NIH-3T3 cells, the competition in viral DNA synthesis
between mutant strains 18-5 and 82 were observed in BMK
cells (panel A, Fig.8). Thus, the permissive factors in the
BMK cells are unlike middle T antigen, which alleviates the
competition between viral genomes with different A enhancers
in NIH-3T3 cells. As will be discussed in detail below, the
likely candidates for enhancer duplication effect in NIH-3T3
cells are PEA1 and PEA3. These factors are known to be
activated by middle T antigen resulting in increased viral
transcription (49). As the A enhancer duplication extends
beyond PEA1 and PEA3 binding sites, it is possible that
other factors are involved in BMK cells.

Moreover, we observed that in the BMK cells, strains
with 82 enhancer outgrow strains with A2 enhancer (panel 8,
Fig.8). Therefore, in this typical cell type, the

duplicatibn of the A enhancer element exerts a strong cis-
‘acting advantage for viral DNA synthesis regardless the

presence of middle T antigen.

21min
In this study, we have shown that alterations of the

enhancer have coevolved with deletions of the middle/small T

antigen in natural isolates of polyoma mutants selected for

123

(WT — +

 

NMDI i 10 'l 10

eB2> EPEEFE~
i‘vé'

9A2> é sweet 34-

 

(A) (3)

Figure 8. Mixed infections of BMK cells.
BMK cells were doubly infected with either
transformation-competent (+MT) or -defective (-MT) viral

strains. The viral DNA was analyzed as described in Fig.6.

124
the growth dependence on transformation (hr-t mutants).
These rearrangements of the enhancer confer a strong cis-
advantage in viral DNA replication to any genome which bears
such a rearrangement. However, thezcis-advantage is
particularly noticeable for middle T antigen deficient
mutants. In fact, as shown in Fig.6, a middle T antigen
deficient mutant without rearrangement is not capable of DNA
replication in NIH-3T3 cells in low serum. These results
provide an adequate explanation for our repeated failure to
obtain such mutants by construction (Chen, Friderici and
Fluck, unpublished data). The sis-advantage is also
observable in the presence of middle T antigen.

When strains carrying rearrangements are used in mixed
infections with strains carrying the A2 wild type enhancer,
a strong competition is observed in which the genome with
the enhancer duplication outreplicates the other genome.
This effect again is much stronger in the absence of middle
T antigen (i.e. in infections with 2 middle T antigen
deficient mutants) or at low middle T antigen levels (i.e.
in infections with a high ratio of middle T antigen
deficient to middle T antigen competent genomes). This
effect had been noted previously and referred to as a
dominant lethal effect (12). The enhancer competition is
also easily observable in rat cells in which hr-t mutants

prevent wild-type from transforming (32, Chapter 4). Cell

125
specific sis-acting advantages of A enhancer duplications
have been reported previously for polyoma mutants isolated
from Friend erythroleukemic cells and neuroblastoma cells
(9,10).

Our experiments demonstrate clearly that the target for
the ”dominant lethal effect" of hr-t mutants is the
rearrangement of the enhancer. Mutants with multiple
reiterations of the A domain compete more strongly than
those with simple structure. The simplest mutant such as
82, which has been studied in detail by competition assays,
shows that the target for these is located in the A domain
between nucleotides 5097 and 5132. This domain is known to
bind the transcription factors PEA1, PEA2, and PEA3 (27).
PEA1 has been identified as the mouse homolog of AP1 (27)
and PEA3 as the homolog of c-ets (48). TPA has been shown
to activate transcription from the PEA1 as well as the PEA3
sites (22,53). Synergy is observed between the two sites
(35,48,50). The same results have been obtained with middle
T antigen (49). TPA stimulates DNA replication of a plasmid
containing the PEA1 and PEA3 binding sites (30). The
activation by middle T antigen of genes which are the target
of protein kinase C activation is understood to be a result
of the activation of the phosphatidylinositol signal

transduction pathway by middle T antigen (44).

126

Our overall model represents a tie, which brought many
separate observations that have been derived by using
plasmids into the life cycle of the virus (30,51). The
major new aspect is the important role for middle T antigen
in DNA synthesis via the activation of enhancer binding
cellular factors and thus, an important role for PEA1/PEA3
in DNA replication in the polyoma system. The experiments
presented here as well as a detailed study presented in
Chapter 2, suggest that factors for DNA replication are
present in limiting amounts in NIH-3T3 cells. Key factors
are those which bind in the A element i.e., at least
PEA1/AP-1 and PEA3/c-ets. These factors can be induced by
serum, TPA and middle T antigen (22,49,53), presumably via
the industion of protein kinase C. Our hypothesis is that
reiteration of the A element represent a way to increase the
probability of binding to these limiting factors. Thus
middle T antigen deficient mutants with reiterations of the
A enhancer have a strong advantage over those without
reiterations, in competing for the cellular factors. In the
presence of the viral inducer, middle T antigen, the
concentration of these factors must rise sufficiently to
relieve most of the advantage associated with the
duplications. Both serum and TPA can in part complement for
the absence of middle T antigen, however, middle T antigen

is a much more potent inducer than either of these agents.

127

As other polyomaviruses, polyoma enhancer contains
motifs for binding multiple cellular factors (29,47). This
may represent an adaptation for growth in multiple tissues
in the natural host (Amalfitano et al., in preparation). An
intriguing aspect of the enhancer is that it can be
rearranged at high frequencies in various ways to generate
mutants with new host range (1). Many of these
rearrangements involve duplications. Both single copy and
duplicated enhancers have been described in naturally
occurring polyoma strains (36). The generation of
duplications in the polyoma enhancer has been observed in
many systems. In fact, the commonality of such
rearrangements distracted from giving importance to the
observation that hr-t mutants contained alterations in the
fragments around the origin when these were first noted
(11,18). The frequent generation of such duplication calls
for an explanation. The position of the origin proximal to
the duplication and the fact that the duplications are
always situated on the late side of the origin suggest that
they may arise when replication forks progress
preferentially over the nucleosome free region (19) and
stall when encountering the first nucleosome both on the
early and on the late sides. Evidence that replication
progresses faster on the late side is available for SV-40
(43). A partially duplicated molecule would result from
this situation which might be at times resolved by a

128
recombination event generating the duplication.

In conclusion, our results suggest a strong functional
interdependence between middle T antigen and the A enhancer
element. This dependence is possibly mediated by the
induction of the PEA1/AP1 and the PEA3/c-ets factors by
middle T antigen, and by the requirement for at least these

two factors for DNA replication in NIH-3T3 cells.

129

References

Amati, P. 1985. Polyoma regulatory region: A potential
probe for mouse cell differentiation. Cell 43:561-562.

Benjamin, T.L. 1970. Host-range mutants of polyoma
virus. Proc. Natl. Acad. Sci. USA 67:394-399.

Bolen, J.B., Thiele, C.J. Israel, M.A., Yonemoto, W.,
Lipsich, L.A., and Brugge, J.S. 1984. Enhancement of
cellular src gene product associated tyrosyl kinase
activity following polyoma virus infection and
transformation. Cell 38:767-777.

Campbell, B.A., and Villarreal, L.P. 1986. Lymphoid and
other tissue-specific phenotypes of polyomavirus
enhancer recombinants: positive and negative
combinational effects on enhancer specificity and
activity. Mol. Cell. Biol. 6:2068-2079.

Cheng, S.H., Harvey, R., Espino, P., Semba, K.,
Yamamoto, T., Toyoshima, K., and Smith, A.E. 1988.
Peptide antibodies to the human c-fyn gene product
demonstrate pp59””m is capable of complex formation
with the middle-T antigen of polyomavirus. EMBO J.
7:3845-3855.

Cherington, V., Morgan, 8., Spiegelman, B.M., and
Roberts T.M. 1986. Recombinant retroviruses that
transduce individual polyoma tumor antigens: effects on
growth and differentiation. Proc. Natl. Acad. Sci. USA
83:4307-4311.

Courtneidge, S.A., and Smith, A.E. 1983. Polyoma virus
transforming protein associates with the product of the
c-src cellular gene. Nature (London) 303:435-439.

DePamphilis, M.L., and Bradley, M.K. 1986. Replication
of SV40 and polyoma virus chromosomes. pp. 99-246. In
The papovaviridae Volume I. Salzman, N.P. ed. Plenum
press, New York and London.

De Simone, V., La Mantia, G., Lania, L., and Amati, P.
1985. Polyomavirus mutation that confers a cell-
specific cis advantage for viral DNA replication. Mol.
Cell. Biol. 5:2142-2146.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

130

De Simone, V., and Amati, P. 1987. Replicative cis-
advantage of polyomavirus regulatory region mutants in
different murine cell lines. J. Virol. 61:1615-1620.

Feunteun, J., Sompayrac, L., Fluck, M., and Benjamin,
T.L. 1976. Localization of gene functions in polyoma
virus DNA. Proc. Natl. Acad. Sci. USA 73:4169-4173.

Fluck, M.M., Staneloni, R.J., and Benjamin, T.L. 1977.
Hr-t and ts-a: two early gene functions of polyoma
virus. Virology 77:610-624.

Garcea, R.L., and Benjamin, T.L. 1983. Host range
transforming gene of polyoma virus plays a role in
virus assembly. Proc. Natl. Acad. Sci. USA 80:3613-
3617.

Garcea, R.L., Talmage, D.A., Harmatz, A, Freund, R.,
and Benjamin, T.L. 1989. Separation of host range from
transformation functions of the hr-t gene of
polyomavirus. Virology 168:312-319.

Goldman, E., and Benjamin, T.L. 1975. Analysis of host
range of nontransforming polyoma virus mutants.
Virology 66:372-384.

Goldman, E., Hattori, J., and Benjamin, T. 1979.
Cellular and c-type viral factors in infections by
polyoma virus hr-t mutants. Virology 95: 373-384.

Griffin, B.E., Fried, M., and Cowie, A. 1974. Polyoma
DNA : a physical map. Proc. Natl. Acad. Sci. USA
71:2077-2081.

Hattori, J., Carmichael, G., and Benjamin, T.L. 1979.
DNA sequence alterations in hr-t deletion mutants of
polyoma virus. Cell 16:505-513.

Herbomel, P., Saragosti, S., Blangy, D., and Yaniv, M.
1981. Fine structure of the origin-proximal DNaseI-
hypersensitive region in wild-type and EC mutant
polyoma. Cell 25:651-658.

Herbomel, P., Bourachot, B., and Yaniv, M. 1984. Two
distinct enhancers with different cell specificities
coexist in the regulatory region of polyoma. Cell
39:653-662.

Hirt, B. 1967. Selective extraction of polyoma DNA from
infected mouse cell cultures. J. Mol. Biol. 26:365-369.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

131

Imler, J.L., Schatz, C., Chatton, B., and Wasylyk, B.
1988. A Harvey-res responsive transcription element is
also responsive to a tumor-promoter and to serum.
Nature (London) 332:275-278.

Jainchill, J.L., Aaronson, S.A., and Todaro, G.J. 1969.
Murine sarcoma and leukemia viruses: assay using clonal
lines of contact-inhibited mouse cells. J. Virol.
4:549-553.

Kornbluth, S., Sudol, M., and Hanafusa, H. 1987.
Association of the polyomavirus middle-T antigen with
c-yes protein. Nature (London) 325:171-173.

Lania, L., Griffiths, M., Cooke, 8., Ito, Y., and
Fried, M. 1979. Untransformed rat cells containing
free and integrated DNA of a polyoma nontransforming
(hr-t) mutant. Cell 18:793-802.

Liang, T.J., Carmichael, G.G., and Benjamin, T.L. 1984.
A polyoma mutant that encodes small T antigen but not
middle T antigen demonstrates uncoupling of cell
surface and cytoskeletal changes associated with cell
transformation. Mol. Cell. Biol. 4:2774-2783.

Martin, M.E., Piette, J., Yaniv, M., Tang, W., and
Folk, W.R. 1988. Activation of the polyomavirus
enhancer by a murine activator protein 1 (AP1) homolog
and two contiguous proteins. Proc. Natl. Acad. Sci. USA
85:5839-5843.

Morrison, D.R., Kaplan, D.R., Rapp, U., and Roberts,
T.M. 1988. Signal transduction from membrane to
cytoplasm: growth factors and membrane-bound oncogene
products increase c-raf phosphorylation and associated
protein kinase activity. Proc. Natl. Acad. Sci. USA
85:8855-8859.

Mueller, C., Muller, W., and Hassell, J.A. 1988. The
polyomavirus enhancer comprises multiple functional
elements. J. Virol. 62:1667-1678.

Murakami, Y., Asano, M., Satake, M., and Ito, Y. 1990.
A tumor promoting phorbol ester, TPA, enhances
polyomavirus DNA replication by activating the function
of the viral enhancer. Oncogene 5:5-13.

Neda, T., Satake, M., Robins, T., and Ito, Y. 1986.
Isolation and characterization of NIH-3T3 cells
expressing polyomavirus small T antigen. J. Virol.
60:105-113.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

132

Oh, S.Y., Amalfitano, A., Friderici, K., Chen, M.C.,
and Fluck, M.M. 1990. Low probability of double
integration in transformation of nonpermissive cells by
polyomavirus. J. Virol. 64:1304-1313.

Piette, J., Kryszke, M-H., and Yaniv, M. 1985. Specific
interaction of cellular factors with the B enhancer of
polyoma virus. EMBO J. 4:2675-2685.

Rochford, R., Campbell, B.A., Villarreal, L.P. 1990.
Genetic analysis of the enhancer requirements for
polyomavirus DNA replication in mice. J. Virol. 64:476-
485.

Rochford, R., Davis, C.T., Yoshimoto, K.K., and
Villarreal, L.P. 1990. Minimal subenhancer requirements
for high-level polyomavirus DNA replication: a cell-
specific synergy of PEA3 and PEA1 sites. Mol. Cell.
Biol. 10:4996-5001.

Ruley, H.E., and Fried, M. 1983. Sequence repeats in a
polyoma virus DNA region important for gene expression.
J. Virol. 47:233-237.

Schaffhausen, 8., Silver, J., and Benjamin, T. 1978.
Tumor antigens in cell productively infected by wild-
type polyoma virus and mutant NG-18. Proc. Natl. Acad.
Sci. USA 75:79-83.

Schlegel.R., and Benjamin, T.L. 1978. Cellular
alterations dependent upon the polyoma virus hr-t
function: separation of mitogenic from transformation
capacities. Cell 14:587-599.

Silver, J., Schaffhausen, B., and Benjamin, T. 1978.
Tumor antigens induced by non-transforming mutants of
polyoma virus. Cell 15:485-496.

Soeda,E., Arrand, J.R., Smolar, N., Walsh, J.E., and
Griffin, B.E. 1980. Coding potential and regulatory
signals of the polyoma virus genome. Nature (London)
283:445-453.

Southern, E.M. 1975. Detection of specific sequences
among DNA fragments separated by gel electrophoresis.
J. Mol. Biol. 98:503-517.

Staneloni, R., Fluck, M.M., and Benjamin, T.L. 1977.
Host range selection of transformation defective hr-t
mutants of polyoma virus. Virology 77:598-609.

43.

44.

45.

46.

47.

48.

49.

50.

51.

52.

53.

133

Tack, L.C., and Proctor, G.N. 1987. Two major
replicating simian virus 40 chromosome classes. J.
Biol. Chem. 262:6339-6349.

Talmage, D.A., Freund, R., Young, A.T., Dahl, J., Dawe,
C.J., and Benjamin, T.L. 1989..Phosphorylation of
middle T by pp6or‘": a switch for binding of
phosphatidylinositol 3-kinase and optimal
tumorigenesis. Cell 59:55-65.

Treisman, R.,Novak, U., Favaloro, J., and Kamen, R.
1981. Transformation of rat cells by an altered polyoma
virus genome expressing only the middle T protein.
Nature (London) 292:595-600.

Turler, H., and Solomon, C. 1985. Small and middle T
antigens contribute to lytic and abortive polyomavirus
infection. J. Virol. 53:579-586.

Veldman, G.M., Lupton, S., and Kamen, R. 1985.
Polyomavirus enhancer contains multiple redundant
sequence elements that activate both DNA replication
and gene expression. Mol. Cell. Biol. 5:649-658.

Wasylyk, B., Wasylyk, C., Flores, P., Begue, A.,
Leprince, D., and Stehelin, D. 1990. The c-ets proto-
oncogenes encode transcription factors that cooperate
with c-Fos and c-Jun for transcriptional activation.
Nature (London) 346:191-193.

Wasylyk, C., Imler, J.L., and Wasylyk, B. 1988.
Transforming but not immortalizing oncogenes activate
the transcription factor PEA1. EMBO J. 7:2475-2483.

Wasylyk, C., Flores, P., Gutman, A., and Wasylyk, B.
1989. PEA3 is a nuclear target for transcription
activation by non-nuclear oncogenes. EMBO J. 8:3371-
3378.

wasylyk, C., Schneikert, J., and Wasylyk, B. 1990.
Oncogene v-jun modulates DNA replication. Oncogene
5:1055-1058.

Winocour, E. 1963. Purification of polyomavirus.
Virology 19:158-168.

Yamaguchi, Y., Satake, M., and Ito, Y. 1989. Two
overlapping sequence motifs within the polyomavirus
enhancer are independently the targets of stimulation
by both the tumor promoter 12-tetradecanoylphorbol-13-
acetate and the Ha-ras oncogene. 1989. J. Virol.
63:1040-1048.

54.

134

Zullo, J., Stiles, C.D., and Garcea, R.L.1987.
Regulation of c-myc and c-fos m-RNA levels by
polyomavirus: distinct roles for the capsid protein VP-
1 and the viral proteins. Proc. Natl. Acad. Sci. USA
84:1210-1214.

Chapter 4.

The Role of the Polyomavirus Enhancer in

Neoplastic Transformation1

ABSTRACT

We analyzed the transformation frequencies induced by
polyomavirus from experiments in which FR-3T3 cells were
infected with a mixed viral population containing a
transforming wild-type plus a transformation-deficient
mutant. The results demonstrate a dosage- and enhancer-
dependent "interference effect", in that increasing the
number of mutant genomes and/or reiterations of the A
enhancer element in the mutant genome induced a progressive
inhibition of transformation induced by wild type.
Inhibition of abortive transformation and viral DNA
replication of wild-type were also observed. All three
effects appear to be mediated by the limiting amounts of
factors which bind in the A enhancer element. One possible
interpretation for the inhibition of stable transformation
is that the binding of factors to the enhancer is an

important step in the integration of the viral genome.

 

1Ming Chu Chen and Michele M. Fluck, Department of
Microbiology, Interdepartmental Program in Molecular Biology,
Michigan State University, East Lansing, MI 48824-1101.

135

136

During infection of nonpermissive cells by polyomavirus
(Py), stable neoplastic transformation relies on integration
of the viral genome into the host chromosome with
concomitant expression of the viral oncogene, middle T
antigen. This is deduced primarily from the studies of
different classes of Py mutants. Infection of rat or
hamster cells with host range transformation-deficient
(hr-t) mutants does not cause either abortive or stable
transformation of the hosts (14). The hr-t gene encodes
middle T and small T antigens as dual products (27). A
mutant that expresses only large T and small T antigens is
unable to transform cells (22), thus middle T antigen is the
transforming protein of Py. Further evidence supports the
above conclusion: for instance, expression of middle T
antigen from a cDNA in normal established rat fibroblasts
induces full transformation (32). In contrast, large T
antigen temperature sensitive mutants (ts-a) induce a normal
frequency of abortive transformation but a decreased
frequency of stable transformation at nonpermissive
temperature (17,29). However, cells can be stably
transformed by these mutants at low temperature, and further
temperature shifts do not abolish the transformed phenotype
(29). Moreover, cells stably transformed by Py often fail
to express an intact large T antigen. Py large T antigen,
therefore, is important for establishing but not for

maintaining stable transformation. Complementation studies

137
using large T and middle T antigen-mutants demonstrate that
two early proteins, large T and middle T antigens, are
required for Py mediated stable transformation (for reviews,
see 15). Large T antigen plays an important part in the
initial steps, presumably by facilitating integration of
viral genomes into host chromosome (17). Once virus
integrates, continuous expression of middle T antigen is
required to maintain a transformation state.

DNA analyses of cell lines transformed by wild-type Py
demonstrate that integrated viral genomes are usually
arranged in a head-to-tail tandem manner (2,3). In those
rare tsa-transformants obtained at nonpermissive temperature
less than a complete copy of the viral genome is usually
integrated (12). Moreover, viral replication origin is
required for Py transformation and for formation of tandem
insertions (10). This implies that viral DNA synthesis
plays a role in the process of integration. A previous
report from our laboratory supports the role of large T
antigen in integration, but argues against the existence of
close correlation between Py DNA synthesis and viral induced
neoplastic transformation. It suggests a function of large
T antigen other than amplifying the viral genome to enhance
the probability of integration. Multiple functions have
been associated with large T antigen: helicase activity

(28), DNA-unwinding activity (11), and recombinogenic

138
activity (30): however, the mechanism by which large T
antigen mediates viral integration is still unclear.

We have followed an interesting observation which shows
that hr-t mutants of Py inhibit the growth of wild-type and
ts-a mutants in a dosage dependent manner (14). Analyses of
viral DNA replication in mixed infections of permissive
mouse cells demonstrate that the "dominant lethal effect"
results from competition between the viral genomes for
limiting cellular factors that bind specifically to the
sequences in A element of the Py enhancer (Chapter 2).
Interestingly, seven of the natural hr-t isolates which have
been sequenced contain reiteration(s) of the A enhancer
element (Chapter 3). Since middle T antigen activates Py
DNA replication and gene expression through PEA1 and PEA3
motifs in the A element (34,35), we hypothesize that
enhancer duplications in hr-t genomes represent a
compensational event for the loss of middle T function.
Results supporting the hypothesis come from experiments
which show'cis-advantage growth for an hr-t genome with a
duplicated A enhancer (Chapters 2,3).

Utilizing the same assays, we tested whether hr-t
mutants affect wild-type induced transformation in
nonpermissive cells. We first determined whether "dominant
lethal effect" exists in rat cells as well as in NIH-3T3

mouse cells. For this purpose, we carried out mixed

139
infections of FR-3T3 cells with three viral combinations
using a wild-type strain and an hr-t deletion mutant, i.e.,
A2 with 82, A2 with 18-5, and A2(e82) with 82. Other than
different in the transformation capability, A2 and 82
strains contain differences in their enhancer-origin
regions: whereas in the latter two pairs, both parental
viruses carry the same enhancer-origin structure (21,
Chapter 3). Each infection contained a fixed concentration
of A2 or A2(e82) at a multiplicity of 1. The concentrations
of the 18-5 and 82 were varied by increasing the
multiplicities from 1 to 20. Strains 82 and 18-5 are
deletion mutants with 241 and 50 bp deletions in their
middle T and small T antigen coding regions, respectively,
which can be distinguished from the wild-types by the size
of MspI fragment #4. The relative ratios of the two
parental viruses were confirmed by analyzing viral DNA in
the input (data not shown). Fig.1 shows an example of the
yield of viral DNA at seven days post infection. We found
that DNA synthesis of A2 and A2(e82) were decreased in the
presence of 82 in a dosage dependent manner. In addition, a
higher MOI of 82 was required to outcompete A2(e82) than to
outcompete A2. The dosage- and enhancer-dependent
competition results in underreplication of the wild-type
DNA, suggesting that FR-3T3 cells contain limiting amounts

of A enhancer factors.

140

 
  

 

A2/‘92 ,
1/0 In 1/3 mo v20 'l/O 1" bp.
_ e 0‘
I
wr> ,5” .. <702
O
82> - <642
e
D
O

 

Figure 1. Viral DNA replication in mixed infection of
FR-3T3 cells.

FR-3T3 cells were infected with mixtures of A2 and 82
(A), or A2(e82) and 82 (B). The ratios of MOIs of wild-type
[A2 or A2(e82)] to mutant (82) are given. Total DNA was
collected from infected cells at seven days post infection.
One tenth of each sample was analyzed by MspI digestion,
electrophoresed on a 2% agarose gel, transferred to
nitrocellulose, and hybridized to a probe specific to MspI
restriction fragment #4. The upper bands represent MspI-4
of A2 or A2(e82), and the lower bands corresponds to that of

B2.

141

During the course of infections, we found that abortive
transformation frequency was also reduced by hr-t mutant.
Since occurrence of abortive transformation is correlated
with transient expression of middle T antigen at the early
stage of infection, it is possible that transcription of the
wild-type genome is interfered with the presence of excess
amount of hr-t mutants. A common mechanism has been
proposed for enhancer mediated viral DNA replication and
transcription since single mutations in this region affect
both functions in the same cell type. Therefore,
competition for the A enhancer binding factors results in
underreplication as well as under-expression of the wild-
type genome.

Another observation we made was alterations of stable
transformation frequencies in the mixed infections. Stable
transformation was determined by the ability of anchorage-
independent growth. Hr-t mutants are defective in
transforming rat cells because of lacking functional middle
T antigen, however, they integrate normally as the wild-type
does (21). Therefore, transformation frequency derived from
double infection represented the number of cells in the
population containing integrated viral genomes expressing
functional middle T antigen. The results are summarized and
depicted in Fig.2. Primarily, we found that in single
infections A2(e82) induced a higher incidence of
transformation than A2 did (Fig.2), suggesting that A2(e82)

142

Figure 2. Transformation in mixed infections of FR-3T3
cells.

FR-3T3 cells were plated at a density of 1 x 105 cells
per 60 mm dish. After the cells attached to the plates,
they were infected with mixtures of transforming wild-type
[A2 or A2(e82)] and nontransforming hr-t mutant (82 or 18-5)
strains of polyoma. The ratios of MOIs of wild-type (WT) to
hr-t are given. Transformation was scored by the appearance
of colonies over the soft agar. The transformation
frequency was determined by dividing the number of colonies
obtained with the number of cells used in the infection.

5)

log(tronsio. motion irequ'ency :10

143

 

 

 

 

o--- A2(82)/82
, rum/32
3- . CI--- A2) Is-s
H I
I
I
2-Ii
].
L4 ". .
#0 III 1/3 1110 1120

WI! hm input rotio

 

144

is a stronger transforming virus. Addition of hr-t mutant
in the infection led to a decrease of transformation
frequency when compared to'that in the single infection.
The reduction of transformation frequency was dosage-
dependent, in that interference became more obvious when
higher doses of hr-t mutants were applied. Transformation
frequencies were less affected when parental viruses contain
the same enhancer sequences, i.e., in the mixed infection of
A2 and 18-5 or of A2(e82) and 82. In contrast, A2 induced
transformation was greatly inhibited by 82 when compared to
the reduction of transformation frequencies in the other two
pairs of infections. It suggests that the structure of
enhancer element in viral genome might be involved in a
rate-limiting step in the process of Py transformation.

Integration of viral DNA into one host chromosome is
required for Py mediated neoplastic transformation, and has
been suggested to be a rate limiting step during this
process (29). To test whether reiterations of the A
enhancer element affect viral DNA integrations, we carried
out several mixed infections of FR-3T3 cells with two
transforming viruses, A2 and A2(eX). A2(eX) represents a
transformation-competent recombinant whose coding sequences
are derived from A2, and enhancer-origin region is from an
hr-t mutant, X (Chapter 2). Various enhancer regions of
different hr-t mutants were utilized, whose sequences have

been determined and shown to contain reiterations in the A

145

enhancer element (1, Chapter 3). A multiplicity of 10 was
used for each virus in the infection. We isolated
transformants from soft agar and analyzed integrated viral
DNA in high molecular weight DNA. Analyses of a total of 96
transformants from eight independent infections are
summarized in Table 1. Species of the integrated viral
genome was determined by the size of Meal fragment #3 that
contained Py enhancer sequences. Fig.3 shows a
representative result of DNA analysis from experiment #4 in
the Table 1. Each transformant contained typical head-to-
tail tandem arrays of the viral genome, since all the MspI
restriction fragments can be recovered. We found that only
three out of 96 transformants (3%) have both parental
genomes integrated. In agreement with our previous report
(24), we demonstrated again that the probability for more
than one integration event in Py transformation is low. In
other words, it suggests that integration of a viral genome
prevents a secondary integration event in the same host. In
addition, half of the transformants contain A2 DNA
indicating no preference for integrating a wild-type genome
with duplicated A enhancer element.

In this report, we have tested the role of the Py
enhancer in neoplastic transformation of nonpermissive
cells. We carried out double infections of FR-3T3 cells

with a fixed amount of wild-type and various amounts of hr-t

146
Table 1. Integration event in polyomavirus transformed

FR-3T3 cell lines derived from mixed infections.‘

 

 

Yiral_strain° intesrations°
2813. no. HT LL WT L WT 1 WT 2 DQQDIQ
1. 12 A2(e82) 9 3 o
2. 12 A2(eRA) 1 9 0
3. 12 A2(e82) 3 a 1
4. 12 A2(e82) 7 6 0
5a. 12 A2(eA8) 12 1 0
5b. 12 A2(eA8) s s 0
6a. 12 A2(eA9) 7 7 2
6b. 12 121919) 7 3 0

 

 

‘Summary of the transformants analyzed in this report.
FR-3T3 cells were coinfected with two transforming strains
at the multiplicity ratio of 10 to 10.

bWT, transforming wild types used in the infections.
WTI and WT2 are different in their enhancer-origin regions
which can be distinguished by’uspI restriction fragments #3.

cSummary of the number of transformants containing
either one or both parental genomes analyzed as determined
by Southern blotting analysis.

147

Figure 3. DNA analysis of transformants derived from mixed
infection with two transforming strains.

FR-3T3 cells were infected with A2 and A2(e82) at a
multiplicity of infection of 10 PFU/cell of each parent
(Table 1, experiment 4). A2(e82) contains a tandem
duplication in the A enhancer element, which can be
distinguished from A2 by different sizes of the MspI
fragment #3. Transformants were cloned by anchorage
independence growth in soft agar. A 10 ug portion of high-
molecular-weight DNA isolated from these clones was digested
with MspI and was fractionated on 1.8% agarose, gel. DNA was
blotted to nitrocellulose paper and hybridized with a probe

for the whole viral genome (pPy-l). bp, Base pairs.

148

Nd. v
fimovwlx v

m—

 

DDS.

.‘uos

. Q. as «so

as, a... nu:

. .D are...”

N ~ 65

149
mutants. Two parental viruses contain the same or different
sequences in the A enhancer element. Viral DNA synthesis
and abortive transformation rate were analyzed during the
early stage of infections. Since expression of middle T
antigen early post infection induce transient anchorage
independent growth, the abortive transformation rate was
used as an indicator for the level of transcription from the
wild-type genomes. The results showed that both DNA
replication and transcription from the wild-type virus were
inhibited in the presence of mutant. Similar to the NIH-3T3
mouse cells, dosage- and enhancer-dependent competitions
imply that FR-3T3 cells contain limiting amounts of A
enhancer factors. PEA1 and PEA3 are the putative limiting
factors, since their binding to the A enhancer element
activates both Py DNA replication and transcription nicis_
(34,35).

Interestingly, the presence of hr-t mutant in the
double infections decreases the stable transformation
frequency. We have yet to determine how hr-t mutant
interferes with wild-type induced stable transformation.
There are two models that we have raised: (1) the wild-type
genome is prevented from integrating into the host
chromosome, and (2) the expression of middle T antigen from
integrated virus is suppressed. Our observation of

diminished wild-type DNA replication and lower abortive

150
transformation rate regarding the enhancer competition favor
the latter hypothesis. Above all, hr-t mutant has to exist
in the same cell with wild-type to compete for viral DNA
synthesis and transcription. Based on the very low
efficiency of viral DNA replication in nonpermissive cells,
we postulate that integration of both viral genomes into
same host DNA is required for co-existence. Nevertheless,
the evidence of low probability of double integration in a
clonal transformant (24) argues against both viruses being
in the same cell after integration occurs. Consequently,
the decreased transformation frequency is unlikely to result
from the suppression of middle T antigen expression from the
wild-type genome by a coexisting hr-t mutant at late stage
of Py infection.

It has been suggested that integration of the viral
genome is a rate-limiting step in transformation on the
basis of one lower frequency of stable transformants
compared with that of abortive transformants (29). The
elevated transformation frequency induced by A2(e82) could
be due to more viral DNA synthesis leading to a higher
integration rate. The lower frequency of stable
transformation in the mixed infection could result from
underreplication of the wild-type genome at early stage of
infection, which provides less templates for integration.
The above points suggest that the Py enhancer affects
neoplastic transformation by mediating the capability of

151
viral DNA synthesis, and the number of DNA templates is
correlated with transformation frequency. However, it has
been shown that the level of viral DNA synthesis does not
fully account for the change of transformation frequency in
Py infection of nonpermissive cells (18). In addition, in
the same mixed infection we found that the decrease of
stable transformation rate was more profound than those of
DNA replication and transcription were, i.e., we observed
decrease of stable transformation early at the input ratios
of 1 to 1 and 1 to 3 (Fig.2), whereas an input ratio of 1 to
10 was needed to observe obvious competition in DNA
synthesis (Fig.1). We speculate that other mechanism(s) are
involved in the Py enhancer mediated viral DNA integration
during neoplastic transformation.

Alternatively, we hypothesize that the Py enhancer
plays a direct role as a.cis-acting signal for an activation
of the viral genome required for integration. However,
analyses of integrated viral genome in transformants derived
from mixed infections utilizing two transforming viruses did
not prove the above postulation. Little competition between
transforming wild-types has been observed in NIH-3T3 cells

(Chapter 3) , moreover, a strong ole-advantage of duplicated

A enhancer element was only observed in the absence of
middle T antigen (Chapters 2 and 3). Duplication of this

region in the Py enhancer may represent a compensation for

152
the mutation of middle T antigen. Possibly cells express
plenty of A enhancer factors in the presence of middle T
antigen, thereby competition between viral genomes is
alleviated. Since middle T antigen mediates enhancer
activity and alleviates competition, it is understandable
that little competition for integration occurs between
transformation-competent viruses which have different copy
number of A enhancer element.

Several explanations could be applied for enhancer
function in integration of the Py genome. First, the
existence of DNase I-hypersensitive sites in the regulatory
region of viral chromosome is highly correlated with
enhancer for transcription and replication (7,20). Enhancer
activities of Py are dispensable for DNA replication and
transcription when little chromatin structure is present,
for instance, in a cell-free condition (25), and in early
development of mouse embryos at the two-cell stage (23). It
is generally thought that binding of nuclear proteins to
enhancer sequences provides nucleosome-free areas in
chromatin, which triggers signals in DNA to both replication
and transcription complexes. Such a nuclease accessible
region in the Py genome may enhance the probability of its
recombination (integration) by altering DNA conformation.
Integration of viral genome involves recombination with host
DNA. Occurrence of recombination has been correlated with

transcriptionally active gene (4,31), possibly due to its

153

DNA accessibility that is also required for initiating
recombination. Second, binding of enhancer factors may
facilitate interaction between viral DNA and the large T
antigen, which is responsible for the typical Py integration
pattern. It has been shown that subpopulation of the large
T antigen are associated with the nuclear matrix (8).
Enhancer may function to localize DNA at cellular structure
such as the nuclear matrix. Mereover, Bourgaux et al. have
indicated that proximity to the Py replication origin being
a critical determining factor for recombination event (6).

Py integration occurs by nonhomologous recombination
with host DNA and usually is associated with rearrangements
or deletion of host DNA (3,19,36). .Multiple copies of the
viral genomes are commonly present in the transformed cells
(2,16). Due to variable integration patterns found in
different transformants and no specific viral-cellular
joints yet being discovered, it is assumed that the host
chromosome comprises numerous integration sites and each
integration event is independent. The evidence against any
specificity in SV40-transformed cells is well known (5).
Similar indications of no specific integration site in host
DNA have also been cited for Py (12,19,26). However,
difficulties of recovering specific integration sequences
might be resulted from the instability of integrated DNA in
Py-transformed cells (2,9,37). We have followed the fate of

Py genomes in stable transformants derived from infection of

154
rat cells with pairs of transforming viruses which contain
differences in the enhancer region. Similar to our previous
studies (24), we found very low incidence of double
integration. Thus, we assume that either limited number of
chromosomal sites are available for Py integration and/or
limited number of viral genomes become capable to integrate.
Since each viral genome has equal opportunities to integrate
in these mixed infections (~50%), low rate of recovering
second parental genome in a transformant suggests that the
integration eVent is restricted.

In conclusion, competition assays demonstrated that
immortalized FR-3T3 cells contain limiting amount of A
enhancer factors which affect viral DNA replication,
transcription, and transformation. Integration has been
proposed to be a rate limiting step in stable
transformation, and appears to be most affected by the
presence of competitors, thus enhancer might be involved in
the step of integration of Py genome into host chromosome.
Furthermore, low frequency of double integrations were found
in the mixed infections of transforming viruses, suggesting
that integration event in a host DNA is restricted. We
hypothesize that enhancer factors act as a.cis-acting
signal, by binding to the A enhancer element, which

activates viral genome to integrate into the host genome.

10.

155

Reference

Amalfitano, A. 1989. Polyoma mutants with altered
enhancer regions redefine subdomains required for
replication in tissue culture systems, pp.35-67. In
In9§1§: The role of the polyoma enhancer region in

replication of the virus 1n y1tr9 and 1n y1y9.

Basilico, C., S. Gattoni, D. Zouzias, and G. Della
Valle. 1979. Loss of integrated viral DNA sequences in
polyoma-transformed cells is associated with an active
viral A function. Cell 17:645-659.

Birg, F., R. Dulbecco, M. Fried, and R, Kamen. 1979.
State and organization of polyomavirus DNA sequences in
transformed rat cell lines. J. Virol. 29:633-648.

Blackwell, T.K., M.W. Moore, G.D. Yancopoulos, H, Suh,
S. Lutzker, E. Selsing, and F.W. Alt. 1986.
Recombination between immunoglobulin variable region
gene segments is enhanced by transcription. Nature
(London) 324:585-589.

Botchan, M., W. T0pp, and J. Sambrook. 1976. The
arrangement of simian virus 40 sequences in the DNA of
transformed cells. Cell 9:269-287.

Bourgaux, P., D. Gendron, and D Bourgaux-Ramoisy. 1990.
Preferred crossover sites on polyomavirus DNA. J.
Virol. 64:2327-2336.

Bryan, P.N., and W.R. Folk. 1986. Enhancer sequences
responsible for DNase I hypersensitivity in
polyomavirus chromatin. Mol. Cell. Biol. 6:2249-2252.

Buckler-White, A.J., G.W. Humphrey, and V. Pigiet.
1980. Association of polyoma T antigen and DNA with the
nuclear matrix from lytically infected 3T6 cells. Cell
22:37-46.

Colantuoni, V., L. Dailey, and C. Basilico. 1980.
Amplification of integrated viral DNA sequences in
polyoma virus-transformed cells. Proc. Natl. Acad. Sci.
USA 77:3850-3854.

Dailey, L., S. Pellegrini, and C. Basilico. 1984.
Deletion of the origin of replication impairs the
ability of polyomavirus DNA to transform cells and to
form tandem insertions. J. Virol. 49:984-987.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

156

Dean, F.B., P. Bullock, Y. Murakami, C.R. Wobbe, L.
Weissbach, and J. Hurwitz. 1987. Simian virus 40 (SV40)
replication: SV40 large T antigen unwinds DNA
containing the SV40 origin of replication. Proc. Natl.
Acad. Sci. USA 84:16-20.

Della Valle, G., R.G. Fenton, and C. Basilico. 1981.
Polyoma large T-antigen regulates the integration of
viral DNA sequences into the genome of transformed
cells. Cell 23:347-355.

De Simone , V., G. La Mantia, L. Lania, and P. Amati.
1985. Polyomavirus mutation that confers a cell-
specific cis advantage for viral DNA replication. Mol.
Cell. Biol. 5:2142-2146.

Fluck, M.M., R.J. Staneloni, and T.L. Benjamin. 1977.
Hr-t and ts-a: two early gene functions of polyoma
virus. Virology 77:610-624.

Fluck, M.M., and T.L. Benjamin. 1979. Comparisons of
two early gene functions essential for transformation
in polyoma virus and SV-40. Virology 96:205-228.

Fluck, M.M., R. Shaikh, and T.L. Benjamin. 1983. An
analysis of transformed clones obtained by coinfections
with Hr-t and Ts-a mutants of polyoma virus. Virology
130:29-43.

Fried, M. 1965. Cell transforming ability of a
temperature-sensitive mutant of polyoma virus. Proc.
Natl. Acad. Sci. USA 53:486-491.

Hacker, D.L., and M.M. Fluck. 1989. Viral DNA synthesis
in nonpermissive rat F-111 cells and its role in
neoplastic transformation by polyomavirus. Mol. Cell.
Biol. 9:648-659.

Hayday, A., H.E. Ruley, and M. Fried. 1982. Structural
and Biological analysis of integrated polyoma virus DNA
and its adjacent host sequences cloned from transformed
rat cells. J. Virol. 44:67-77.

Herbomel, P., S. Saragosti, D. Blangy, and M. Yaniv.
1981. Fine structure of the origin-proximal DNase I-
hypersensitive region in wild-type and EC mutant
polyoma. Cell 25:651-658.

Lania, L., M. Griffiths, B. Cooke, Y. Ito, and M.
Fried. 1979. Untransformed rat cells containing free

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

157

and integrated DNA of a polyoma nontransforming (Hr-t)
mutant. Cell 18:793-802.

Liang, T.J., Carmichael, G.G., and Benjamin, T.L. 1984.
A polyoma mutant that encodes small T antigen but not
middle T antigen demonstrates uncoupling of cell
surface and cytoskeletal changes associated with cell
transformation. Mol. Cell. Biol. 4:2774-2783.

Martinez-Salas, E., D.Y. Cupo, and M.L. DePamphilis.
1988. The need for enhancers is acquired during
mammalian development with formation of a diploid
nucleus. Genes Dev. 2:115-126.

Oh, S.Y., A. Amalfitano, K. Friderici, M.C. Chen, and
M.M. Fluck. 1990. Low probability of double integration
in transformation of nonpermissive cells by
polyomavirus. J. Virol. 64:1304-1313.

Prives, C., Y. Murakami, F.G. Kerns, W. Folk, C.
Basilico, and J. Hurwitz. 1987. DNA sequence
requirements for replication of polyomavirus DNA in
vivo and in vitro. Mol. Cell. Biol. 7:3694-3704.

Ruley, H.E., L. Lania, F. Chaudry, and M. Fried. 1982.
Use of cellular polyadenylation signal by viral
transcripts in polyoma virus transformed cells. Nucleic
Acids Res. 10:4515-4524.

Silver, J., B. Schaffhausen, and T. Benjamin. 1978.
Tumor antigens induced by non-transforming mutants of
polyoma virus. Cell 15:485-496.

Stahl, H., P. DrOge, and R. Knippers. 1986. DNA
helicase activity of SV40 large tumor antigen. EMBO J.
5:1939-1944.

Stoker, M., and R. Dulbecco. 1968. Abortive
transformation by the tea mutant of polyoma virus.
Nature (London) 223:397-398.

St-Onge, L., L. Bouchard, S. Laurent, and M. Bastin.
1990. Intrachromosomal recombination mediated by
papovavirus large T antigens. J. Virol. 64:2958-2966.

Thomas, B.J., and R. Rothstein. 1989. Elevated
recombination rates in transcriptionally active DNA.
Cell 56:619-630.

Treisman R., Novak, U., Favaloro, J., and Kamen R.
1981. Transformation of rat cells by an altered polyoma

33.

34.

35.

36.

37..

158

virus genome expressing only the middle T protein.
Nature (London) 292:595-600.

Veldman, G.M., S. Lupton, and R, Kamen. 1985.
Polyomavirus enhancer contains multiple redundant
sequence elements that activate both DNA replication
and gene expression. Mol. Cell. Biol. 5:649-658.

Wasylyk, c., J.L. Imler, and s. Wasylyk. 1988.
Transforming but not immortalizing oncogenes activate
the transcription factor PEA1. EMBO J. 7:2475-2483.

Wasylyk, C., P. Flores, A. Gutman, and B. Wasylyk.
1989. PEA3 is a nuclear target for transcription
activation by non-nuclear oncogenes. EMBO J. 8:3371-
3378.

Williams, T.J. and M. Fried. 1986. Inverted
duplication-transposition event in mammalian cells at
an illegitimate recombination joint. Mol. Cell. Biol.
6:2179-2184.

Zouias, D., I. Prasad, and C. Basilico. 1977. State of
the viral DNA in rat cells transformed by polyoma
virus. II. Identification of the cells containing
nonintegrated viral DNA and the effect of viral
mutations. J. Virol. 24:142-150.

159
' SUMMARY AND CONCLUSIONS

The role of polyoma enhancer region in viral DNA
replication and neoplastic transformation was investigated.
The experiments described in Chapter 2 used competition
assays to determine that limited amount of cellular factors
are present in NIH-3T3 mouse cells. Studies of enhancer
variants suggest that nuclear factor(s) binding to the A
enhancer element represent the major class of the
candidates. The presence of this enhancer element has been
shown toIcis-activate both polyomavirus DNA replication and
transcription, though the mechanism is unclear. Results
from the kinetic studies suggest that function of the A
enhancer element is involved in an initiation step of DNA
synthesis. Possibly interactions with enhancer factors
facilitate the formation of replication complex around the
origin.

Numerous evidence have demonstrated that rearrangements
in the polyoma enhancer affect viral growth profoundly.
Combining the observations that naturally isolated host
range transformation deficient (hr-t) mutants are strong
competitors in viral DNA replication and that they share
common sequence rearrangements in the A enhancer element, we
suggest that duplicated A enhancer play a positive role in
the growth of this group of mutants. The previously
reported "dominant lethal effect" of hr-t mutants in mixed

160
infections with wild-type can thus be explained as an
enhancer mediated competition effect. A enhancer
duplications have been discovered in several transforming
polyoma strains, which do not display strong cis-advantage
growth. By contrast, in the absence of the reiteration,
nontransforming mutants have a strong defect in viral DNA
synthesis. Experiments described in Chapter 3 demonstrated
that strong'cis-advantage of the duplication carrying strain
can be relieved partially by TPA or serum, and totally by
middle T antigen. Binding motifs in the A enhancer for PEA1
and PEA3, the mouse homologs of AP1 and c-ets, are the most
likely targets since they both respond to the activation of
the above three factors. Middle T antigen is postulated to
be most effective than in inducing viral DNA synthesis via
the activation of PEA1 and PEA3. Thus, the reiteration of
the A enhancer element has possibly evolved during the
natural selection for transformation dependent growth.

The experiments in Chapter 4 studied the fate of
polyoma genome in the mixed infections of nonpermissive rat
fibroblast cell line, FR-3T3. Results from competition
assays demonstrated that FR-3T3 cells contain limited number
lof A enhancer factors. These proteins appear to mediate
viral DNA replication, gene expression (as indicated by
abortive transformation), and stable transformation. The

interference effect of hr-t mutants on wild-type induced

161
transformation frequencies was more pronounced than on the
viral DNA replication and transcription. Taken with results
generated from our and other laboratories: the number of
viral genome is not highly correlated with the
transformation frequency, the rarity of double integration
in the polyoma transformant, and integration is a putative
rate-limiting step during the course of neoplastic
transformation, we propose that enhancer plays a.cis-acting
role in the integration of the viral genome.

The experiments using two transforming strains with
different copy number of the A enhancer elements pursued the
above hypothesis. The results showed few double integration
events, whereas no preference was shown to integrate the
viral genome with reiterations in the enhancer. The failure
to detect the differences can be explained as follows.
Middle T antigen mediates A enhancer activity by inducing
sufficient cellular factors PEA1 and PEA3, and alleviates
the competition between the viral genomes, thus either
parental genome has a similar opportunity to interact with A
enhancer factors. A system other than using middle T-
mediated transformation as a selection marker would be
helpful to test the hypothesis. The intriguing restriction
in integration of the viral genome to only one site in the
genome of individual transformants remains to be understood.

There can be one specific site common in every transformant

162
or multiple integration sites in the host DNA, and
integration of one viral genome prevents the occurrence of a
secondary integration event.

Several lines have implicated that specific binding of
host factor with the enhancer sequences alters chromatin
structure and provides nucleosome-free regions, which may
trigger the initiation steps of DNA replication and
transcription. Moreover, actively transcribed genes are
correlated with the occurrence of recombination. The
results presented here follow the above models, therefore,
study of interaction between the polyoma enhancer and host
machineries may be helpful in elucidating the mechanism by

which enhancer mediates DNA activity.