‘51-..-
.791:

V!!! .‘al
. .nl.1.!t.

:11 aflsvvaa x
lil‘ . I‘In.
Tilt a 1...;

.t... 15’ Allin.
93!.lntéli’l-1. '2. ,Illn’vrobucl 1!!!!- .
.55.}...35 . .itf‘x‘frpfiv I57 ‘rir!’
53.2: 2.....va .Iv into-(‘1.t.’ 19:. 32.1.3.0!!925-81231 :0; w I...» .
It‘ll-13.5.".3‘ lifll’lc nu!“< p! 3:33,:
titloai 5 IvlliturcO-Oiatvfo;

{El-0.1.... i!..l!"|l
"A.I| .v . {‘35. in“;
. V - :1

54-. L‘

1“.:A.§oa

’1‘ 2|. In.
: ‘

5.91 .3.
$3! . o 'ioilfla v. .7“.(.10v vv..(o’a
i210: u‘l...‘

Inf; 3!!! - vii-Irv"
.. Ilvtibolo.lv)‘. . vvflllrluob.lvn

 

ESlS

JIllillllflllllllfilllllllllllllfllfllllll h A...

1293 00910 1

 

 

 

This is to certify that the

dissertation entitled

Molecular characterization of a defective Marek's disease
virus: Identification of a functional MDV replication origin
and a gene encoding a potential nuclear DNA binding protein.

presented by

Heidi Suk Camp

has been accepted towards fulfillment
of the requirements for

Ph. D. degree in Animal Science

 

 

 

 

 

 

@w.cw~

Major professor

Date 5/3fl0' 3

MS U is an Affirmative Action/Equal Opportunity Instilull'on 0-12771

 

LIBRARY
Michigan State
. University

I
l

 

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

ll DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

;_
43
3E
F—i—‘l

 

 

 

MSU is An Affirmative Action/Equal Opportunity Institution

smarts-9.1

 

MOLECULAR CHARCTERIZATION OF A DEFECTIVE MAREK’S DISEASE
VIRUS: IDENTIFICATION OF A FUNCTIONAL MDV REPLICATION
ORIGIN AND A GENE ENCODING A POTENTIAL NUCLEAR DNA

BINDING PROTEIN

By

Heidi Suk Camp

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

DOCTOR OF PHILOSOPHY

Department of Animal Science

1993

ABSTRACT

MOLECULAR CHARACTERIZATION OF A DEFECTIVE MAREK’S DISEASE
VIRUS: IDENTIFICATION OF A FUNCTIONAL MDV REPLICATION
ORIGIN AND A GENE ENCODING A POTENTIAL NUCLEAR DNA

BINDING PROTEIN
BY
Heidi Suk Camp

Characterization of molecular events involved in Marek’s disease virus (MDV) .

DNA replication is essential to understand the entire scope of the MDV
replication cycle. Defective MDV genomes (MDV replicons) contain limited viral
sequences important for MDV DNA replication. Therefore, they can replicate as
separate entities from that of standard virus DNA, creating a useful model system
to study MDV DNA replication. Based on DNA sequencing and transient
plasmid amplification assays, a functional serotype 2 MDV replication origin was
identified within the MDV replicon DNA. The serotype 2 replication origin is
approximately 90-bp long, arranged in an imperfect palindrome centered around
an AT—rich region. Sequence analysis of the 90~bp region revealed a sequence and
structure strikingly similar to origins of alpha-herpesvirus replication. In
addition, the 90-bp MDV origin sequence contains a 9-bp sequence
(CGTTCGCAC) sharing 82% sequence similarity to an ll-bp motif recognized by

the herpes simplex virus (HSV) origin binding protein (OBP). Mobility shift and

competition assays were performed to investigate viral or cellular proteins
interacting with the potential OBP recognition site of the MDV replication origin.
A virus specific protein-DN A complex, complex B, was identified using synthetic
22-mer oligonucleotide probe containing the potential 031’ binding site of MDV.
The 5’-region of MDV replicon DNA revealed a significant sequence identity to
the HSV a sequence, responsible for cleavage/ packaging process. Several
potential open reading frames (ORF) were identified within the MDV replicon
DNA. ORF-A codes for a putative 204 amino acid protein that shares 21% and
36% amino acid sequence identity to nuclear DNA binding proteins such as a
nuclear antigen (EBNA-l) of Epstein-Barr virus (a gamma-herpesvirus) and '
galline, a chicken sperm histone protein, respectively. The MDV replicon,
therefore, appears to contain characteristics of both gamma- and alpha-

herpesviruses at the molecular level.

As alyways, to love of my life, Steve and Lindsay.

iv

ACKNOWLEDGEMENTS
I am grateful to my mentors, Drs. R. F. Silva and P. M. Coussens for their
continue support and encouragement. Special thanks to members of my
quidence committee, Drs. JJ. Irland, M. T. Yokoyama and L. F. Velicer for
their valuable time and suggestions. Finally, to my parents and parent-in- ‘

laws for their infinite love and support.

Table of contents

Chapter I: Literature review ................................. 1
1. Biology of Marek’s disease virus (MDV) ................... 1
2. Pathology of Marek’s disease (MD) ....................... 3
3. Molecular biology of MDV ............................ 5

A. MDV genome structure .......................... 5

B. Physical maps of MDV genomes .................... 9
C. DNA replication ................................ 11
i). Alpha-herpesvirus DNA replication ............. 13
ii). Beta-herpesvirus DNA replication ............... 23
iii). Gamma-herpesvirus DNA replication ............ 26
iv). MDV DNA replication ....................... 28
4. Defective herpesviruses ............................... 32
A. Generation of defective particles ..................... 34
B. Structure of defective genomes ...................... 36
C. Replication of defective particles ..................... 38

D. Cis-acting recognition sites within defective
virus genomes .................................. 40
E. Role(s) of defective particles in pathogenesis
and virus-host cell interactions ..................... 42
F. Defective particles as eucaryotic vectors ............... 45
Chapter II: Cloning, sequencing, and functional analysis

of a Marek’s disease virus origin of replication ......... 48
Abstract ............................................ 49
Introduction .......................................... 50
Materials and Methods ................................. 52
Results ............................................ 56
Discussion .......................................... 59

Chapter III: Defective Marek’s disease virus contains a gene encoding
a potential nuclear binding protein and a region of the HSV
a-like sequence ................................. 72

Abstract ............................................ 73

vi

Introduction .......................................... 74

Material and Methods ................................... 77
Results ............................................ 80
Discussion ............................................ 84

Chapter IV: Cellular and viral protein interactions with Marek’s disease

virus replication origin ............................ 109

Abstract ............................................ 1 10
Introduction .......................................... l 11
Materials and Methods .................................. 113
Results ............................................ 1 16
Discussion ............................................ 1 19
Summary and conclusions .................................. 129
References ............................................ 137

vii

Chapter I

Literature review
1 Biology of Marek’s disease virus (MDV)

In 1907, a Hungarian scientist, Joseph Marek first described the clinical signs
of Marek’s disease (MD) (Marek, 1907). MD is a readily transmissible,
lymphoproliferative disease of chickens, often accompanied by leg paralysis and
lymphoid tumors (Payne and Biggs, 1967). During the early 1960’s, it was
discovered that the etiologic agent of MD is a highly cell-associated avian
herpesvirus, termed Marek’s disease virus (MDV) (Churchill and Biggs, 1967;
N azerian and Burmester, 1968).

The structure of the MDV capsid is that of a typical herpesvirus containing
162 capsomeres with icosahedral symmetry (N azerian and Burmester, 1968).
N ucleocapsids containing intact viral DNA, approximately 85-100 nm in diameter,
are present in MDV infected cells. Only a few enveloped particles, approximately
130-170 nm in diameter, are observed budding from the infected cell inner nuclear
membranes (N azerian and Burmester, 1968; Hamdy et a1, 1974). The only source
of a large number of cytoplasmic enveloped particles (270—400 nm) is the infected
feather follicle epithelium.

MDV is classified into 3 serotypes based on agar gel precipitation and indirect
fluorescent antibody assays (Bulow and Biggs, 1975; Bulow and Biggs, 1975b).

Serotype 1 MDV includes all oncogenic viruses and their attenuated derivatives.

2

With regards to oncogenic potential, different pathotypes exist within serotype 1
MDV. For example, "classical forms" of MDV which include GA and JM strains
can cause high incidences of MD in genetically susceptible, unvaccinated chickens.
RB-IB and Md/ 11 are classified as very virulent strains of MDV (vaDV) and
are responsible for causing outbreaks of MD in vaccinated chickens (Witter, 1985).
Only genetically resistant lines of chickens can be protected from vaDV by
current vaccine programs.

N on-oncogenic, naturally occurring chicken herpesviruses are represented in
the serotype 2 group. Though tumors have never been observed in susceptible -
chickens infected with serotype 2 strains, certain strains of serotype 2 MDV, such
as SB-1, can cause a low level of cytolytic infection in lymphoid organs resulting
in immunosuppression in infected birds (Schat and Calneck, 1978). Serotype 3
MDV is the non-oncogenic herpesvirus of turkeys (HVT) which is antigenically
related to MDV. HVT does not occur naturally in chickens, but is now prevalent
in commercial flocks due to its widespread use as a vaccine against MDV-induced
lymphomas. Attenuated serotype 1 strains and bivalent vaccines containing HVT
and serotype 2 strains are also being used as vaccines to protect birds against MD.
MDV is of interest not only because of its immense economic importance to the
poultry industry, but also because of its role as a model system to study virus-
induced malignancy.

MDV can be readily cultured from viable lymphocytes and MDV tumor cells.
Cell-free MDV can be recovered only from the feather follicle epithelium of

infected birds. Upon initial isolation, MDV can be propagated in chicken kidney

3
cells (CKC), duck embryo fibroblast (DEF) cells, chicken embryo fibroblast (CEF)

cells or other primary avian cell cultures (Purchase, 1974). In tissue culture, MDV
establishes strictly cell-associated infections resulting in poor virus recovery from
the culture media.

Most studies on the biological characterization of MDV have been done with
serotype 1 and serotype 3. More attention was given to serotype 2 strains when
Bacon et a1. (1989) reported that serotype 2 MDV-based vaccines caused the
enhancement of lymphoid leukosis in certain strains of chickens co-infected with
exogenous avian leukosis virus (ALV). Further reports suggest that serotype 2
MDV may encode a trans-activation protein(s) which stimulates the retrovirus ‘
long terminal repeat promoter region (Tieber et al., 1990), leading to an increase
in ALV gene expression and virus production (Pulaski et al., 1992). Further
biological characterization of all three serotypes is needed to understand the
nature and relationships among all MDV serotypes at the molecular and antigenic

levels.

2. Pathologenesis of MDV.

The pathogenesis of MDV has been well defined in experimental settings
using mostly "classical forms" of MDV and genetically susceptible maternal
antibody-free chicks (Payne, 1985). MDV is readily transmitted horizontally by
direct contact with infected birds and through the air (Sevoian et al, 1963). MDV
infection usually occurs via the respiratory tract, followed by a cytolytic infection

in lymphoid organs at 3-6 days post-inoculation. Mononuclear infiltration of

4

peripheral nerves, gonad, iris, various visceral organs, muscle, and skin can be
observed at approximately 2 weeks post infection. Clinical signs and gross
lesions generally appear after about 3-4 weeks.

The most common clinical signs of MD are leg paralysis primarily due to
lymphocyte infiltration and nerve demeylination of spinal cords. Clinical signs
may vary, however, among infected birds depending upon the location of infected
nerves. For example, the head will be paralyzed if the nerves controlling neck
muscles are affected. Vagal nerve involvement can lead to paralysis of the cr0p.
In addition, a certain strain of MDV can cause depigmentation of the iris,
resulting in blindness. Infected visceral organs, particularly the gonads, liver, .
lungs, heart, muscle, and skin, may have diffuse or grayish-white lymphoid
tumors (Payne and Biggs, 1967). In addition, severe atrophy of the thymus and
bursa of Fabricius often occurs in infected chickens. However, several factors can
affect the outcome of MDV pathogenesis such as the age of host, strain and
species of host, and strain of viruses (Payne, 1988).

There are three types of MDV infections in viva; lytic infection, latent infection
and transformation. The lytic infection can be sub-divided into fully-productive
and semi-productive infections. Fully-productive infections are characterized by
the production of cell-free enveloped virus particles which occurs only in the
infected chicken feather follicle epithelium. Semi-productive infections result in
the production of predominantly nonenveloped nucleocapsids in the nuclear
membrane of B lymphocytes from infected spleen, bursa, thymus, and peripheral
blood (Calneck et al., 1970; Schat et al., 1978). The establishment of latent

5
infections by MDV is usually detected at the end of the early cytolytic infection

and results from development of host cell immune responses. Latent infections
are primarily restricted to T-lymphocytes and are characterized by the presence
of viral genomes without viral antigen expression or virus particle production.
T-lymphocytes are also the target for oncogenic transformation by MDV (Calneck
et al., 1970; Nazerian, 1973). As with most of the other DNA tumor viruses, the
oncogenic potential of MDV infections may be due to either the presence of viral

oncogenes, or the induction of host cell immune response.

3. Molecular biology of MDV
A. MDV genome structure.

The MDV genome consists of a linear double stranded DNA molecule of
approximately 165-180 kilo-basepairs (kbp) with nicks and gaps as in other
herpesviruses (Lee et al., 1971; Hirai et al., 1979; Cebrian et al., 1982; Wilson et al.,
1991). The density of MDV DNA in CsCl gradients is 1.705 g/cmB, close to that
of chicken cell DNA, presenting difficulties isolating pure virus DNA.

Originally, MDV was classified as a gamma-herpesvirus, based primarily on
its lymphotrophic nature, similar to Epstein-Barr virus (EBV). However, recently
MDV has been re-classified as an alpha-herpesvirus based on MDV genomic
structure and gene collinearity with those of other alpha-herpesviruses
(Buckmaster et al., 1988; Roizman, 1990; Roizman, 1992; Brunovski and Velicer,
1992; Velicer and Brunovski, 1992). In addition, random nucleotide sequencing

of MDV genomes strongly indicate that MDV has a greater sequence similarity

6

to the alpha-herpesviruses than to gamma-herpesviruses (Buckmaster et al., 1988).

The genome structure of MDV belongs to the Herpesvirdae group E genome
family along with HSV and Varicella-Zoster virus (VZV) (Cebrian et al., 1982).
The group E genome structure is composed of covalently linked long (L) and
short (S) regions, each consisting of unique sequences (UL or Us) flanked by
inverted repeats (Wadsworth et al., 1975) (Fig. 1A). The inverted repeat regions
flanking the L component are designated ab and b'a’, whereas the inverted
repeats of the 5 component are termed a’c’ and ca. The standard group E virus
genome can thus be represented by ab-UL-b’a’c’-Us-ca. The a sequence in HSV
genomes, approximately 250-500-bp long, is variable among different virus
isolates (Mocarski et al., 1985). A single unit of the HSV a sequence is composed
of two copies of a 20-bp terminal direct repeat (DR1), a unique 64»bp sequence
(Uh), 19-22 tandem repeats of a 12-bp sequence (DR2), 3 repeats of a 37-bp
sequence (DR4), and a unique 58-bp sequence (Uc) (Fig. 1B). Therefore, the HSV
a sequence can be represented by URI ~Ub-DR2mw-DR43-Uc-DR1. Numerous
reports suggest that the a sequence present in the L-S junction and both termini
of HSV genomes contains two cis-acting recognition signals commanding
inversion of L-S components relative to each other, as well as for proper
cleavage/ packaging of unit length viral DNA. A more extensive literature review
on the functional property of the a sequence will be presented in the QM
rgglication section (p. 11).

In addition to the standard herpesvirus group B genome structure, serotype

1 MDV contains several sets of direct repeats consisting of more than 100-bp

 

 

 

Fig.1(NTheg-wpfiherpeevlrusgermsmnctmm)Ashrgle
oopyoftheHSVaseqrence

 

 

 

(designated DRl to DR5) scattered throughout the genome. Direct repeat
elements of MDV DNA are reminiscent of direct repeats present in EBV DNA
(Hirai, 1988).

Based on gene colinearity between MDV and HSV, DRS sequences (located
in both termini of the MDV serotype 1 genome) are of interest in regards to a
potential a-like sequence. Southern blot analysis depict DRS regions as
heterogenous Egg} bands hybridizing to appropriate probes, suggesting the
existence of terminal heterogeneity (Hirai et al., 1988). However, DRS sequences
are not present within the L-S junction of the MDV genome, indicating that DRS
sequences may play different functional roles than the HSV a sequence. In
support of this notion, Kishi et a1. (1991) reported the identification of an a-like
sequence (269-bp) located within the L-S junction and both termini of the serotype

1 MDV genome. DNA sequencing analysis of the 1.6-kbp BamI-II to I-IindIII

8
subfragment of the serotype 1 MDV BamHI-A genomic clone revealed 215

nucleotides containing two copies of a 12-bp repeat (DR1), 17 tandem repeats of
a 6-bp sequence (DR2), two copies of a 13-bp sequence (DR4) and two unique
sequences within the region flanking the 12-bp repeats (Ub and Uc, respectively)
(Fig. 2). Therefore, the putative a-like region of serotype 1 MDV genome contains
a structural similarity to the HSV a sequence. The a-like region of serotype 1
MDV, however, did not contain any nucleotide sequence similarity to the HSV a
sequences. On the contrary, DR2 sequences (17 tandem repeats of GGGTI‘A) of
serotype 1 MDV were found in human herpesvirus type 6 DNA (Kishi et al.,1988).

A potential a-like sequence in serotype 3 MDV was reported by Reilly and '
Silva (1993). Based on southern hybridizations and exonuclease digestions of the
termini of HVT genomes, a 1.2-kbp fragment which is located in both termini and
the L-S junction region of HVT genome was identified. The 1.2-kbp fragment was
reported to share some structural similarity to the HSV a sequence. The number
of copies of this 1.2-kbp region were variable among different HVT isolates as in
HSV. Additionally, different HVT isolates exhibited intra-strain variations in the
number of copies of 1.2-kbp a-like sequence within L component termini and L-S
junctions. The S termini contained only a single copy of the 1.2-kbp fragment.
Interestingly, the 1.2-kbp a-like region in HVT is similar in size to the 950-bp a
sequence of cytomegalovirus, a beta-herpesvirus (Tamashiro et al., 1984).
Nucleotide sequences of the 1.2-kbp region, however, are not yet available for

direct comparison with a-like sequences reported for other herpesviruses.

 

 

 

TR
1. UL
la 6:
5' 3'
DR] - GGCCGCGAGAGC
5, 3, i——— 269bp ———4
(DR2)26 - GGCI'I' A DR] (DR2)26 (DROZ DR!

. U U
(DR4n.sccmcaccccrA3 -—_—i:l

Fig. 2 An a-like sequence in serotype 1 MDV

 

 

 

B. Physical maps of MDV

The construction of physical maps of serotype 1 and HVT DNA were
accomplished by Fukuchi et al.(1985) and Igarashi et a1. (1987), respectively (Fig.
3A and 3B). More recently, Ono et a1. (1992) reported the construction of
restriction endonuclease (RE) maps for serotype 2 DNA, strain I-IPR16 (Fig. 3C).
The availability of RE maps and genomic clones of MDV DNA have greatly
facilitated our understanding of the molecular nature of MDV.

Although all three serotypes of MDV are related antigenically, their genomic
RE patterns are very different, and these differences can be used to evaluate new
MDV isolates (Silva and Barnett, 1991). There have been conflicting results
regarding the overall degree of homology between serotype 1 MDV and HVT; 5%

to 70% homology was reported based on reassociation kinetics and low stringency

IO

 

 

>. manage. H ZU<

 

 

 

 

gr CF a? =flm Cm gm
Hr " _ 1+an
L .N o _ o n m 5; _ m / : . T” a //z_z:n= 5: > w.
. —— — L— p _ _ _ b p —_i b _—_ :Zhb— _ — .P bb _— _
——— q dd — q u d 4 — —— — «A: —:——u a fi _ —d —d —_ _

w. manoaaum N §U<

 

 

 

 

:3. 5. =3. Sm cm if».
H m L Til—U
e : 2.: n z is??? 22.5: n x 3.93.5 o ”3 _ we» > a» a
— _ LPP F P p _ p : i— L — _ p _ — hpp — —_ _— _
fi _ a d —J — — — fi dfi qifi _ _ — q a q —d — q— 4— _

0. mmeoaflm w ZU< Ag

 

 

 

.Ewr Cr Em :~r Cm dam
D u _ DIE
x
e2. 0 a
L m o a w z”: n // z x» _ o _ a o > o
F p: P P hik _ P _ p n — p _
a flu q I d u d d A q q — u u _

 

16.9 35.8. 328 o. E. 3323 . zo< 62:1. :52. Amy. 33on n Zo< “moor. .33 man A9. .8336 w §o< Ema... :53

 

 

11
southern blot analysis, respectively (Kaschka-Dierich et al., 1979; Gibbs et al.,

1984). A recent report by Ono et a1. (1992) suggested that serotype 2 MDV is
similar to serotypes 1 and 3 MDV, with regards to genome structure and gene
colinearity. Determination of complete nucleotide sequences for all three MDV

serotypes would enhance our understanding of MDV biology.

C. DNA replication.

Much of the interest in animal virus genome replication comes from a desire
to understand the events that occur during replication of eucaryotic chromosomes.
Viruses offer many advantages as model DNA replication systems. The relatively
simple genome structure of most viruses allows easy manipulation at the
molecular level. Studies of Simian virus 40 (SV40) and adenovirus DNA
replication, which relies mostly on host-cell replication machinery, have led to the
discovery of many eucaryotic proteins which are necessary for viral replication
processes. In contrast to SV40 and adenovirus, herpesvirus genomes are more
complex and encode many of the proteins required for replication of their
genomes. Thus herpesviruses represent a model system in which to study
interactions of virus-encoded and cellular proteins involved in DNA synthesis
(Olivo and Challberg, 1989). Genetic and molecular dissection of herpesvirus
replication events, will therefore increase our understanding of eucaryotic genome
replication.

Study of MDV DNA replication is important for several reasons. First, MDV

exhibits characteristics of both alpha-and gamma-herpesviruses; the MDV genome

12
structure and gene arrangements are similar to HSV and VZV (alpha-

herpesviruses), whereas the biological properties of MDV are similar to EBV (a
gamma-herpesvirus) (Buckmaster et al., 1988; Roizman, 1982; Roizman, 1992;
Brunovskis and Velicer, 1992). Based on ethidium bromide-CsCl equilibrium
centrifugation, several investigators have reported that MDV genomes exist as
closed circular DNA in lymphoblastoid cell lines established from MDV tumors
(Tanaka et al., 1978; Hirai et al., 1981). The closed circular structure of MDV
DNA in established MDV cell lines resembles that of EBV genomic structure in
latently infected cells. However, based on Gardella gel electrophoresis and in situ
hybridization techniques, Delecluse and Hammerschrnidt (1993) recently reported
that the majority of MDV genomes are integrated into the host cell chromosomes
in various MDV lymphoblastoid cell lines. Only a minor population of MDV
genomes were detected as linear or covalently closed circular DNA. As in the
MDV system, the integration of EBV DNA was observed in a number of Burkitt’s
lymphoma cell lines, together with circular forms of EBV DNA (Anvret et al.,
1984; Adams et al., 1987). Study of MDV DNA replication, therefore, will
enhance our understanding of MDV lineage in the herpesvirus family.
Secondly, MDV still causes a great economic loss in domestic poultry
industries. Although successful vaccines exist, the mechanisms of anti-tumor
immunity are virtually unknown. An emergence of very virulent MDV strains
adds further complications in controlling MD. Additionally, conditions
appropriate for the establishment of latency and transformation by MDV infection

are not clear at present. It is likely that multiple steps are involved in MDV

13

transformation rather than a single viral oncogene product. Restriction of viral
DNA replication has been implicated in DNA virus transformation. Integration
of viral genomes into host cell chromosomes can lead to activation or repression
of certain cellular proteins, leading an imbalance of host gene expression and,
thereby contributing to the transformation process. The mode of MDV DNA
replication during lytic and latent infections, and its association with host cell
replication, are poorly understood. One can ask a series of questions regarding
MDV DNA replication. 1). Does MDV contain origins of DNA replication? 2).
Does MDV contain signals specifying the cleavage-packaging process? 3). Are
there viral-specific proteins necessary for MDV DNA replication? 4). Are there
host factors involved in MDV DNA replication? 5). How does MDV DNA
replicate and maintain its genome in lytic and transformed/ latent infection
stages?

The current literature review will be focused on comparisons of DNA
replication from all three herpesvirus classes (alpha, beta, and gamma-
herpesviruses). Reviewing studies and results on replication of other
herpesviruses will help us gain some insights regarding future experimental

designs in MDV DNA replication research.

i). Alpha-herpesvirus DNA replication.
HSV is an alpha-herpesvirus, and is the most extensively characterized of all
herpesviruses. The importance of defective herpesvirus genomes in study of viral

replication is accentuated in the HSV system. Existence of three separate HSV

14
origins of DNA replication was first inferred from studies of two different types

of defective herpesviruses (amplicons) (Frenkel, 1981). A more extensive
literature review on the description of HSV amplicons will be addressed in a later
section entitled,"c1efective herpesviruses" (see p.32). All three HSV origin sequences
have been cloned from the standard virus genome and analyzed in some detail.

Origin function analyses are usually performed by transient plasmid
amplification assays. Standard plasmid amplification assay is performed by
transfecting plasmids containing origin sequences into HSV infected cells, or co-
transfecting cells with origin containing plasmids and standard virus DNA
(Mocarski and Roizman, 1982; Stow, 1982; Stow and McMonagle, 1983).
Subsequent studies revealed the identification of two copies of lytic origins within
the inverted repeats flanking the S component (oris). A third origin, oriL was
localized to the center of the L component (oril). A core oriL is located in a 425-
bp segment containing a perfect 144-bp AT-rich palindrome (Weller et al., 1985).
Or'i5 and oriL sequences share extensive nucleotide sequence similarity, except for
sequences extending to rightward from the AT-rich palindrome (Murchie and
McGeoch, 1982; Weller et al., 1985). To determine functional significance of the
existence of three separate HSV origins, several mutant viruses were created.
Mutant viruses lacking oriU or with one copy of oris replicated normally in vitro
(Longnecker and Roizman, 1986; Polvino-Bodnar et al., 1987). Attempts to
construct mutant viruses lacking both copies of ori,5 have not been successful,
implying that HSV DNA replication requires at least one copy of oris or,

alternatively, at least two origin sequences (oriS and any or two copies of oris).

15
Functional significance of three separate origins in HSV DNA replication is not

clear.

Deletion analysis of oris sequences revealed a core origin of 75-bp region
which included both arms of a 56-bp palindrome centered on alternating AT
residues, and an additional 30—bp leftward of the AT-rich palindrome (Stow and
McMonagle, 1983; Lockshon and Galloway, 1988). Substitution of the AT-rich
region with GC-rich sequences impaired oris replication activity (Lockshon and
Galloway, 1988). A large insertion of AT dinucleotide within the AT-rich motif
resulted in an oscillating effect on the replication activity; the insertion of 3 copies
of an AT dinucleotide caused a dramatic decrease in DNA synthesis, the insertion
of 5 or 6 copies of the AT dinucleotide returned replication efficiency to the
normal level. Insertion of 8 AT dinucleotides decreased replication efficiency.
These results suggest that spacing between the two palindromic arms in oriS is
critical for DNA replication.

Deb and Doelberg (1988) performed systematic base substitutions in a triple
tandem arrangement across the AT- rich region, and found that changing the first
3-bp of the left end of the AT-rich sequences caused a dramatic loss of replication
function. This suggests that the AT-rich sequences at the center of the palindrome
are essential and have an unknown functional role in replication which is not
clear at present.

Two origin-binding protein (OBP) recognition sites (designated sites I and H)
overlap the ends of each arm of the oris palindrome (Fig. 4) (Elias and Lehman,

1988; Elias et al., 1990). The HSV OBP is one of seven viral proteins that are

16
required for HSV DNA replication: a helicase and prirnase complex (ULS, UL8,

UL52), the OBP (UL9), the major single-stranded DNA-binding protein (ICP8),
DNA polymerase, and a polymerase accessory protein (UL42) (McGeoch et al.,
1988; Wu et al., 1988).

Based on site-directed mutagenesis and methylation interference, the OBP.
recognition site was localized to a sequence YGYTCGCACT; where Y represents
C or T (Koff and Tegtrneyer, 1988; Deb and Deb, 1989; Elias et al., 1990). The

deletion of OBP-binding sites I and II completely eliminates both the OBP-binding

 

 

HIE

 

 

 

 

 

 

 

WWRAWHMWAW
an. It'll ll'lll

“put-Madum

Fig. 4 HSV origin Binding protein (UL9) recognition sites

 

 

 

activity and HSV DNA replication in transient plasmid replication assays (Deb
and Deb, 1989; Weir and Stow, 1990). Therefore, the interaction of OBP with oris

is critical to optimal HSV DNA replication. As mentioned earlier, an oscillating

17

activity was observed with plasmids containing different copy numbers of AT
dinucleotide within the AT-rich region (Lockshon and Galloway, 1988). This
phenomenon may reflect a requirement for OBP, bound on each arm of the
palindrome, to be located on the same side of the DNA helix (Challberg and
Kelly, 1989).

An 11-bp motif containing the OBP-binding domain of site I is highly
conserved in oris and oriL of HSV-1 and HSV-2, as well as in VZV, equine
herpesvirus (EHV), and MDV origins of DNA replication (Stow and Davison,
1986; Baumann et al., 1989; Camp et al., 1991).

A third presumptive OBP-binding site (III) within oriS has been identified and
shown to contain sequence similarity to OBP-binding sites I and II, but no
sequence-specific OBP binding has been reported at site III (Elias et al., 1990; Weir
and Stow, 1990) (Fig. 4). Weir and Stow (1990) reported that the deletion of site
III within oris of HSV-1 resulted in only a moderate decrease in replication
efficiency. On the contrary, the deletion of part of a sequence corresponding to
site III in HSV-2 resulted in a dramatic decrease in DNA replication (Lockshon
and Galloway, 1986). A more recent report by Martin et al. (1991) indicated that
all three OBP-binding sites are required for optimal DNA replication, suggesting
the functional importance of all three OBP binding sites in HSV replication.

The lack of OBP binding to site 111 in in-vitro assays, and the decrease in DNA
replication by mutagenesis within site III, suggest that other factors may interact
with site 111. These "accessory factors" may change the overall origin structure,

thereby making site III more accessible to OBP in-vivo. Sequence-specific binding

18
of OBP to sites I and II within HSV origins, the loss of replication activity in

origins with mutagenized OBP binding sites, and the helicase and ATPase
activities contained within purified OBP, suggest that the OBP may represent an
initiator protein for HSV DNA replication.

Like other DNA viruses such as SV40, adenovirus, and bovine papillomavirus,
HSV origin sequences are flanked by an auxiliary component containing
transcriptional regulatory regions. OriL is located between the divergent
transcriptional start sites of the genes encoding the major DNA-binding protein
ICP8 and DNA polymerase (Weller et al., 1985; Polvino-Bodnar et al., 1987). Ori5
is positioned between the genes encoding immediate-early proteins ICP4 and
ICP22/47 (Preston and Tannahill, 1984; Preston et al., 1984; Preston et al., 1988).
Many transcription faCtor binding sites are located within the promoter regions
of ICP4 and ICP22/ 27, such as a VP16 transactivator recognition site
(TAATGARAT motif), Spl binding sites, and a nuclear factor III binding site.
Wong and Schaffer (1991) reported that mutations within these transcription
factor binding sites resulted in a significant reduction in oris replication activity.
In addition, Wong and Schaffer postulated that transcription factors which bind
to, and regulate immediate-early genes may promote a secondary effect on oris
activity, increasing accessibility of initiation protein complexes to oris.
Alternatively, these transcription factors may interact directly with DNA
replication proteins. Hubenthal-Voss and Roizrnan (1988) have identified two
transcripts which transverse ori5 sequences in both HSV-1 and HSV-2, and

postulate that these transcripts may have functions in DNA replication by either

19

inducing alterations in the superhelical density of the template or serving as
primers for DNA replication.

The current understanding of the mode of HSV DNA replication is primarily
based upon pulse-labeling HSV DNA and fractionation of resulting viral DNA
molecules of different buoyant densities. HSV DNA replicative intermediates
from buoyant density fractionation have been extensively studied by electron
microscopy Gacob and Roizman, 1977; Jacob et al., 1979). There are at least five
different types of HSV DNA structures: 1) double-stranded circular molecules, 2)
unit length linear molecules containing internal gaps and single-stranded termini,
3) lariat structures containing circular-linear molecules, 4) linear molecules with
"eye" and "D" loops, and finally 5) large "tangled masses" of DNA. With the
exception of lariat molecules, the same types of replicative intermediates were
observed in pseudorabies virus (an alpha-herpesvirus) infected cells (Ben-Porat
etal., 1976; Jean et al., 1977).

On the basis of these observations, it has been proposed that double-stranded
linear HSV DNA circularizes upon entry into host cells. As mentioned earlier, the
group E herpesvirus genome structure can be written as ab-UL-b’a’a’d-Us-ca.
Because the a sequence is repeated at both ends of the HSV genome, it is likely
that the linear HSV DNA can fold back upon itself to form circular molecules by
ligating ends through the a sequences. The resulting circular molecules are then
serve as templates to generate linear concatemers consisting of tandemly repeated
unit size viral genomes via a rolling circle mechanism (Iacob and Roizman, 1977).

Additional evidence to support concatemeric formation by a rolling circle

20
mechanism was reported by Jacob and co-workers (1979). Restriction enzyme

analyses of linear replicative intermediates revealed a decrease in the molar
concentration of termini fragments and an increase in the molar concentration of
the L-S Junction fragments suggesting that linear replicative intermediates exist
as head-to-tail concatemers. Furthermore, HSV was reported to induce an
amplification of integrated simian virus 40 (SV40) DNA by generating large,
tandemly reiterated SV40 DNA, which may have arisen by a rolling circle
mechanism (Matz, 1987). In a natural setting, SV40 replicates in a theta mode of
DNA replication, generating only circular monomers. This result suggests that
HSV contains signals directing the formation of concatemers, possible via rolling
circle mechanism.

Further evidence for rolling circle HSV DNA replication was reported by
Rabkin and Hanlon (1990) who performed an artificial in vitro DNA synthesis
using preformed replication forks consisting of a nicked, double-stranded, circular
DNA molecule with a 5’ single-strand tail as templates with partially purified
HSV infected cell protein extracts. Analysis of replicated DNA structures from
those templates revealed linear concatemers with HSV infected protein extracts,
but not with mock infected protein extract, suggesting that DNA synthesis
occurred via a rolling circle mechanism. The suggestion that HSV DNA
replication occurs via a rolling circle mechanism thus is very plausible and
supported by numerous studies (Jacob and Roizman, 1977;]acob et al., 1979; Matz,
1987; Rabkin and Hanlon, 1990). However, further studies are required to explain

the existence and functions of other types of replicative intermediates in HSV

21

infected cell nuclei. For example, it is difficult to explain the formation of
"tangled masses of DNA" by a rolling circle mechanism alone.

Difficulties associated with the study of herpesvirus DNA replication are due
primarily to the large genome size of the herpesviruses and, to the isomerization
of herpesvirus genomes, in which the L and 5 components invert relative to each
other. As a consequence, viral DNA extracted from virions or from infected cells
consists of four equirnolar isomers that differ in the orientation of the L-S
components relative to each other (Hayward et al., 1975).

It has been suggested that inversion of HSV genomes is mediated by
homologous recombination via the a sequence, present at both termini and the L-
S junctions of HSV genomes (Smiley et al., 1981). Further experiments revealed
that the a sequence contains cis-acting elements through which a site-specific
recombination occurs (Mocarski and Roizman, 1981; Chou and Roizman, 1985).
Chou and Roizrnan (1985) reported that DR4 elements of the a sequence are
important for site-specific recombination, whereas Smiley et al. (1990) concluded
that both ends of the a sequence are required for the inversion process (Fig. 1B,
p7). To the contrary, some investigators have suggested that the a sequence
inversion is mediated by a general, rather than a site-specific recombination
process (Pogue-Geile et al., 1985; Longnecker and Roizman, 1986; Weber et al.,
1988). However, numerous reports suggest that the a sequence may be
considerably more recombinogenic than other DNA fragments of equivalent
length and therefore, mediates high-frequency recombination events.

Furthermore, Poffenberger and Roizrnan (1985) constructed a mutant virus lacking

22

most of the internal inverted repeats including the a sequence. This mutant
(inverted repeat minus) virus was frozen in one arrangement suggesting an
inability of the mutant virus L and S components to invert.

A different approach was undertaken to quantitate HSV inversion frequency
by Dutch et al. (1992). To briefly explain, plasmid constructs containing the E. coli
lacZ gene flanked by directly repeating a sequences, with or without ori,, were
measured for its frequency of lacZ gene deletion. Deletion of the lacZ gene was
determined by using white and blue color selection of the resulting bacterial
colonies. In this system, lacZ gene deletion most likely occurred by homologous
recombination between two intramolecular a sequences. Dutch et a1 (1992)
concluded that the recombination frequency was higher with plasmids containing
an origin of HSV replication and two copies of the a sequences, compared to
plasmids containing only one copy of the a sequence with or without the origin
sequences. In addition, the processes of replication and recombination occur in
parallel.

An additional role of the HSV a sequence is in the virus maturation process.
The a sequence contains signals specifying a site-specific cleavage/ packaging
process (Spaete and Frenkel, 1982; Stow and McMonagle, 1983). As mentioned
previously, HSV generates replicative intermediates consisting of head-to-tail
concatemers. It has been proposed that concatemeric molecules are cleaved at the
a sequence thus ensuring that only unit length viral DNA is packaged into
nucleocapsids (Spaete and Frenkel, 1982). Comparative analysis of the a

sequences of other herpesviruses revealed two highly conserved sequence

23
elements, termed pad and pad (Davison, 1984; Matsuo et al., 1984; Albrecht et al.,

1985; Bankier et al., 1985; Tamashiro and Spector, 1986; Marks and Spector, 1988).
pad and pad sequences are located in the Ub and Uc regions of the HSV a
sequence, respectively (Fig. 1B, p. 7). Deiss et al. (1986) concluded that pad and
pad sequences play important roles in the site-specific cleavage/ packaging

process in HSV propagation.

iii). Beta-herpesvirus DNA replication.

Cytomegalovirus (CMV) is a beta-herpesvirus, and contains the largest
genome (240-kbp) among herpesviruses. Human cytomegalovirus (I-ICMV) can
cause serious clinical consequences in acquired immunodeficiency syndrome or
immunocompromised patients (Alford and Britt, 1985). Like other herpesviruses,
CMV exhibits both lytic and latent infection cycles. Although much less is known
about the molecular aspects of CMV DNA replication in comparison to alpha-
herpesviruses, there is some evidence to indicate that linear molecules of CMV
DNA undergo circularization, via terminal regions, upon initial infection of host
cells (Marks and Spector, 1988).

At the present, only a single copy of the lytic origin of CMV DNA replication
(oriLyt) has been identified and analyzed in some detail. Based on a novel
approach utilizing the ganciclovir-induced chain termination method, Hamzeh et
al. (1990) located a region encoding a potential lytic origin of CMV DNA
replication near the center of the CMV UL component (EcoRI-V fragment).

Subcloning and deletion analyses of the potential CMV lytic replication origin

24
revealed that the functional oriLyt was located on a 2.4-kbp fragment (Anders and

Punturieri, 1991). In contrast to alpha-herpesvirus lytic origins of replication, the
HCMV oriLyt is intriguing for its large size and complexity. Several novel
repeated sequence elements and known transcription factor binding sites such as
ATF/CREB, MLTF/USF, and Sp1 within the HCMV oriLyt were identified
(Anders et al., 1992). It is noteworthy that the overall base composition of the
HCMV oriLyt is approximately 62% G+C with an asymmetric nucleotide
distribution; the leftward boundary of oriLyt contains an AT-rich segment (up to
70%) and the rightward boundary contains extremely G+C rich sequences.

Several lines of evidence suggest that the HCMV oriLyt is functionally
analogous to lytic origins of alpha-herpesviruses. Moreover, the functionality of
the HCMV oriLyt appears to be similar to the oriL of HSV, based on several
observations; 1). HCMV oriLyt is localized near the center of the UL
(approximately 1-kbp upstream of U157 which encodes a single-stranded DNA
binding protein) as in the HSV oril, 2). Plasmids containing the 2.4-kbp HCMV
oriLyt can induce the generation of high-molecular-weight tandem oligomers, 3).
HCMV oriLyt requires the vitally encoded DNA polymerase (Hamzeh et al., 1990;
Anders and Punturieri, 1991; Anders et al., 1992).

The lytic origin of simian cytomegalovirus (SCMV) has also been located and
analyzed (Anders and Punturieri, 1991). As with the HCMV oriLyt, the structure
of the SCMV oriLyt appears to be complex. A core functional unit of the SCMV
oriLyt was localized to a 1.3-kbp fragment containing four distinct domains; a 9-

bp repeated sequence, an AT-rich region, an 11-bp direct repeat sequence, and a

25
47-bp direct repeat sequence.

The overall CMV DNA replication process is unclear. Hamzeh et a1. (1990)
proposed that CMV DNA replication occurs bidirectionally, and not by a rolling
circle mechanism. However, based on transient plasmid amplification assays,
HCMV oriLyt containing plasmids can induce the amplification of tandem
oligomers as in HSV lytic origins of replication. This result suggests that HCMV
DNA replication occurs via a rolling circle mechanism as in HSV (Anders et al.,

1991).

iii). Gamma-herpesvirus DNA replication.

EBV is a human herpesvirus belonging to the gamma-herpesvirus subfamily.
EBV infects and irnmortalizes human B lymphocytes in vitro, and is associated
with mononucleosis, nasopharyngeal carcinoma, and Burkitt’s lymphoma (zur
Hansen, 1981). The structure of EBV genome consists of five unique sequence
domains which are divided by four classes of internal repeat elements (IRs) and
a variable number of directly repeated 50pr sequences (TR) located at both ends
of the EBV genome (Dambaugh et al., 1980) (Fig. 5). Thus, overall structure of the
EBV genome can be represented as TR-UI-IRI -U2-IR2-U3-IR3-U4-IR4-U5-TR.

B-cells immortalized by EBV express relatively few EBV genes and are,
therefore said to be latently infected with EBV. There are two distinct cis-acting
elements required for EBV DNA replication during lytic and latent phases of the
EBV life cycle (Yates et al., 1984; Hammerschmidt and Sugden, 1988). In latently

infected cells, EBV genomes exist as multiple copies of plasmids containing the

26
plasmid origin of replication (oriP) which is derived from the U1 region of EBV

genomes (Fig. 5) ( Yates et al., 1984). OriP is approximately 1,700-bp in length,
and contains two cis-acting components required for activity; approximately 20
imperfect copies of a 30-bp sequence and a 65-bp region of dyad symmetry
located l-kbp away from the family of 30-bp sequences (Reisman et al., 1985).
Deletion analysis within both cis-acting components of oriP, results in loss of oriP
activity. The distance and orientation of two cis-acting elements did not alter the
oriP function, suggesting that neither a precise distance nor a particular
orientation of the two components relative to one another, is essential for oriP
activity (Reisman et al., 1985; Chittenden et al., 1989). The overall structure of
oriP is very complex and distinct from that of alpha-herpesvirus lytic origins of
replication. Most alpha-herpesviruses generate head-to-tail concatemers during
their replication, whereas the products of oriP replication are circular plasmid
molecules (Yates and Camiolo, 1988). In addition, replication of plasmids
containing oriP is synchronized with host cell proliferation, replicating only once
during S phase of the cell cycle (Adams, 1987).

In EBV latently infected B lymphocytes, four distinct EBV proteins are
expressed; EBV nuclear antigens 1 (EBNAl), 2 (EBN A2), 3(EBN A3) and a plasmid
membrane protein (LMP) (van Santen et al., 1981; Hennessy et al., 1984; Hennessy
et al., 1985; Hennessy and Kieff, 1985; Sample et al., 1986). The oriP domain and
its flanking sequences are of great interest, primarily due to the presence of 5’
regions for all three EBV nuclear antigen RN As. All three nuclear antigens are,

presumably, derived from a single primary transcript which is modified by

27

 

 

 

TR
TR 01 "it U2 IRZ U3 "13 U4 "14 US 4""
art? oriLyt

Fig. 5 Epstein—Barr virus genome structure. Terminal repeats (TR),

internal repeats (1R1, 182, "B, and ii“), uniqie encee (U1, U2, U3,
U4, and U5», origin of plasmid (oriP) and lytic (9’3” replication are

indicated.

 

 

alternate splicing. EBN A-1 has the ability to transactivate oriP by binding to the
family of 30—bp repeats within the enhancer region of oriP, allowing maintenance
and replication of EBV episomes (Reisman, et al., 1985).

Treatment of EBV latently infected cells with 12-O-tetradecanoyl-phorbol-13—
acetate ('TPA) can reactivate EBV and subsequently induce the lytic phase of the
EBV cycle (zur Hansen et al., 1981). During lytic infection, EBV utilizes a separate
replication origin, distinct from oriP (oriLyt). Hammerschmidt and Sugden (1988)
first described the oriLyt, which is present twice in genomes of most EBV isolates
and serves as a lytic origin of EBV DNA replication. One copy of oriLyt is
centrally located within the EBV genome, approximately 40-kbp away from oriP
(Fig. 5). The second copy of oriLyt is located at the right end of the EBV genome.
As in HSV orig EBV isolates containing only one copy of oriLyt can replicate as
efficiently as EBV isolates containing both copies of oriLyt. Several observations

suggest that EBV OriLyt is structurally and functionally different from EBV oriP,

 

28

but functionally analogous to HSV lytic origins of replication; 1). Replication
products via oriLyt are composed of concatemeric EBV DNA molecules, similar
to the products of HSV oris and oril, 2). Plasmids containing oriLyt induce a 100-
1000 fold amplification of the input plasmid molecules in cells that support the
lytic life cycle of EBV, whereas templates containing oriP replicate only once
during S phase of the cell cycle, and 3). OriLyt requires viral DNA polymerase as
do HSV lytic origins of replication, whereas oriP containing plasmids require only
EBN A-1 for replication.

The structure of oriLyt is more complex than HSV oriS and oriL consisting of
arrays of direct and inverted repeat sequences. Auxiliary sequences containing
potential transcription factor binding sites have been localized near oriLyt
sequences. These binding sites accentuate oriLyt replication activity. Therefore,
as for HSV DNA replication, a combination of cellular and virally encoded
transcription factors may play important roles in the lytic cycle of EBV DNA

replication.

iV). Marek’s disease virus DNA Replication.

Knowledge of the cis-acting sequences required for viral DNA replication is
essential to understand viral replication mechanisms. Utilizing a defective MDV
genome, a functional origin of DNA replication in serotype 2 MDV was identified
(Camp et al., 1991). The serotype 2 MDV replication origin is located in inverted
repeats flanking the L fragment, indicating the existence of at least two copies of

origin sequences within serotype 2 MDV genomes. The replication origin of

29

serotype 2 MDV was localized to a 90-bp region arranged in an imperfect
palindrome containing 30-bp of alternating AT sequences. The structure and
sequence of the serotype 2 replication origin is strikingly similar to that of alpha-
herpesvirus replication origins. In addition, the 90-bp sequence contains a 9-bp
motif highly conserved among origins of alpha-herpesviruses. The 9-bp sequence
is a subset of an 11-bp motif recognized by the HSV origin binding protein (Elias
and Lehman, 1988; Koff and Tegtmeyer, 1988). Although MDV shares biological
properties similar to those of gamma-herpesviruses, no significant nucleotide
sequence similarity was found between the serotype 2 replication origin and
origins of beta- or gamma-herpesviruses.

Based on nucleotide sequence analysis, a putative replication origin of
serotype 1 MDV was reported (Bradley et al., 1989). The putative serotype 1
MDV origin sequence is located in inverted repeats flanking the L component of
the serotype 1 MDV genome, and contains several transcription factor binding
sites (Sp1 and Oct-1 binding sites) within flanking regions of the presumed origin
of replication (Bradley et al., 1989; Morgan et al., 1991). As in HSV, CMV, and
EBV, these auxiliary components may play important roles in MDV DNA
replication. Morgan and co-workers (1991) observed a decrease in MDV plaque
formation when cells were co-transfected with standard viral DNA and a plasmid
containing the auxiliary component of the putative serotype 1 origin sequence.
Constructs containing deletions within those transcription factor binding sites did
not result in plaque inhibition, indicating that there may be competitions for

transcription factors between the standard virus and plasmids containing the

30

putative origin sequences.

Transcription factor binding sites located at close proximity to the putative
serotype 1 MDV replication origin are most likely utilized to express a family of
MDV genes; a gene encoding pp38 (Silva and Lee, 1984; Cui et al., 1991; Chen and
Velicer, 1992) and a family of transcripts thought to play roles in MDV
transformation (Bradley et al., 1989; Chen and Velicer, 1991). A gene encoding
pp38, a MDV phosphoprotein, is located within the junction of long unique and
long internal repeat regions of the serotype 1 MDV genome. The 5’-promoter
region of the pp38 gene is located upstream of the putative serotype 1 MDV
origin sequence, and transcribes right to left with respect to the putative origin
sequence. Expression of pp38 is insensitive to phosphonoacetic acid treatment
(inhibits viral DNA synthesis), suggesting that pp38 may belong to one of the
immediate-early or early proteins (Chen and Velicer, 1992). Based on
immunoprecipitation analyses, pp38 appears to be serotype 1 specific (Silva and
Lee, 1984; Ikuta et al., 1985). It has been proposed that pp38 plays a role(s) in
MDV tumorigenicity and latency due to its high expression in MDV
lymphoblastoid cell lines (Cui et al., 1991; Chen and Velicer, 1992).

Within the downstream region of the putative serotype 1 MDV replication
origin, a family of transcripts potentially responsible for MDV tumorigenicity was
identified (Bradley et al., 1989; Chen and Velicer, 1991 ). Expression of these
transcripts is very complex followed by extensive post-transcriptional
modifications. It is noteworthy that these transcripts are directly associated with

a heterogeneous region (het) which is presumed responsible for MDV

31

transformation. The het region consists of 132-bp repeats and undergoes
expansion (0.6 to 5.4-kbps) in different serotype 1 MDV isolates upon serial in
vitro passages (Hirai et al., 1981 ; I-Iirai, 1988; Fukuchi et al., 1985; Silva and Witter,
1985). In addition to those transcripts, a potential immediate early gene was
localized in the downstream region of the putative serotype 1 replication origin.
Construction of a cDN A library from serotype 1 MDV infected cell RNA revealed
a 1.3-kbp cDN A clone transcribing in the left to right direction away from the
putative origin sequence (Hong and Coussens, personal communication). It is
intriguing that multiple transcripts that are implicated in MDV transformation are
located near the putative serotype 1 MDV replication origin. It is of interest,
therefore, to study the affects of those transcripts and their promoter / enhancer
regions in MDV DNA replication.

Despite electron microscopic studies on MDV DNA structures were reported
(Nazerian and Burmester, 1968; Cebrian et al., 1982), the mode(s) and kinetics of
MDV DNA replication have not been determined. Based on the comparison of
genome structures between MDV and HSV, it is likely that MDV genomes
undergo isomerization and formation of concatemers during the MDV lytic
infection stage, but this has not been proven. Reilly and Silva (1993) observed
two different sizes of TRL fragments for serotype 3 MDV genomes, suggesting
that MDV genomes, may, in fact, isomerize similar to HSV genomes. Further
studies are required to understand the molecular events of MDV DNA replication

and genome arrangement.

32

4. Defective herpesviruses

Defective virus particles represent any virus genomes that are lacking the
complete genetic information of non-defective standard virus genomes. As a
mnsequence, such particles are not infectious and they need a "helper virus" to
provide functions necessary for replication and propagation. Defective viruses are
present in many animal, bacterial, and plant virus stocks that have gone through
in vitro undiluted serial passaging (Huang, 1973). The first description of
defective herpesvirus particles was reported by Bronson et al. (1973). Since then,
characterization of defective variants from other herpesviruses, such as
pseudorabies virus (PrV), equine herpesvirus (EHV), cytomegalovirus (CMV),
Epstein-ban (EBV) virus, herpesvirus of saimiri (HV S) and Marek’s disease virus
(MDV) has been reported (Fig. 6) (Frenkel, 1981; Carter and Silva, 1990). Because
defective HSV variants have been the most extensively studied, the current review
will primarily focus on defective HSV particles, and when appropriate, studies
reported on other defective herpesviruses will be included.

There are at least three important aspects to the study of defective virus
genomes. First, they represent deleted versions of their parental virus genomes
and yet retain cis-acting replication recognition sites (an origin of replication and
a cleavage—packaging signal). Defective virus genomes thus represent simpler
model systems in which to study cis-acting elements essential for virus
propagation, modes of viral DNA replication, and viral or cellular factors required

for viral DNA replication. Secondly, because of their ability to replicate in the

33

 

 

mg: ”L (Jo—O

Class I Justin,Patton
G strain —

 

 

 

 

 

 

anb UL b a'm c' U, on
Class II - -
.‘b b a'rn 6'
UL
"”VC) CID—LC)
— —
I. IR Us IR
PIV - H
- —
- _ —
L IR Us IR
W1 , O—O
_
H L H
HVS mm: mm
—

Fig.6 Blackbateerepleeentselectedviralseqtmretainedin
defective herpesrirusee; class i and class ll defective herpes simpler
virus (HSV), defective Marek's disease virus (MDV), defective
pseudorabies virus (PrV), defective equine herpesvirus type 1 (EHV-l)
and defective herpesvirus saimiri (HVS).

 

 

34

presence of trans-functions (enzymes and factors) provided by helper viruses,
defective virus DNA can be utilized as a shuttle vector to deliver and express
foreign gene products in eucaryotic cells. Finally, overall virus gene expression
and virus-host cell interactions are altered in cells infected with virus stocks
containing a large percentage of defective particles. Therefore, defective particles
have implications in the overall outcome of virus pathogenesis and the virus

evolutionary process.

A. Generation of defective particles.

Propagation of plaque purified virus in tissue culture cells at high multiplicity
of infection (M 01), often results in accumulation of defective viral variants. It
is noteworthy, however, that defective virus particles were frequently observed
in cells infected with herpes virus saimiri (HVS), a lymphotropic gamma-
herpesvirus, at low input multiplicities (Fleckenstein et al., 1975). In addition,
cells transfected with standard HVS DNA resulted in generation of defective virus
particles only after a few replication cycles. This result suggests that other factors
are involved in generation of defective particles other than high M.O.I. Defective
HVS DNA, which contains reiterated repetitive sequences, have been found in
autopsy materials from tumors induced by HVS and in lymphoblastoid cell lines
derived from lymphomatous HVS tumor organs (Fleckenstein et al, 1975). Thus,
results with HVS are raising questions in regard to a possible role(s) defective
particles play in overall viral pathogenesis and virus-host cell interactions.

In contrast to most other herpesviruses, MDV is highly cell-associated and,

35
therefore, it is impossible to establish high M.O.I in tissue culture cells. One

defective variant, however, was identified in CEF cells infected with 281MI/ 1
passage 94, a serotype 2 MDV strain (Silva et al., 1988; Carter and Silva, 1990).
As in HSV amplicons, defective MDV genomes (MDV replicon) exist as multiple
reiterated head-to-tail concatemers (Fig. 7). However, MDV replicon differs from
HSV amplicons in at least two aspects; 1). MDV replicon DNA is derived from
inverted repeats flanking the L component, whereas the analogous class I HSV
amplicons originate from repeats flanking the S component, 2). CEF cells co-
transfected with a plasmid construct containing a single repeat unit of MDV
defective genome and standard virus DNA resulted in the generation of a
recombinant replication competent virus (Silva et al., 1988; Silva and Witter, 1991),
whereas such a phenomenon has never been observed with HSV amplicons, and
3). Defective MDV particles were observed in cell culture passages of greater than -
90, whereas defective HSV particles are often observed in cell culture passages 10-
15. As in HVS, the occurrence of defective MDV genomes is not likely due to
high M.O.I, but rather may be due to repeated multiple passages of the virus
stocks and mechanisms that we do not understand at the present.

Most defective virus DNA contains selected viral sequences (cis-acting
replication elements) derived from parental virus genomes. A model suggesting
that, during high M.O.I., inter-molecular recombinational events generate
defective variants containing limited DNA sequences of viral origins. Intra-
molecular recombination events are favored over other types of genomic

rearrangements (Kwong and Frenkel, 1984). Conservation of cis-acting replication

36

elements retained in most defective virus DN As suggest that preferential selective

treatment of those DNA sequences during undiluted serial passages.

 

 

 

Replication
Id In! 3‘ II. I‘ u I‘ H u
Clone EcoRl Fragment

J3...

FIG. 7 Generation of Mill! replicon i985).

 

 

 

B. Structure of defective genomes
In general, most defective virus DNA contains multiple reiterations of selected

viral sequences, derived from parental virus genomes in head-to-tail

37

arrangements. The overall size of packaged defective virus DNA is
indistinguishable from that of standard virus DNA, with the exception of human
CMV defective virus particles, which are smaller than parental virus DNA (Stinski
et al., 1979).

There are two predominant species of defective variants in HSV type 1 and
2 strains. Class I defective HSV genomes contain viral sequences derived from
the right end of the S component of parental virus DNA in head-to-tail tandem
arrays (Bronson et al., 1973; Frenkel et al., 1976; Murray et al., 1975). Because of
the S component of HSV genomes contains relatively high G+C content, class I
defective HSV DNA display a higher buoyant density than that of parental virus
DNA. Class II defective HSV genomes contain multiple reiterated sequences
predominantly from within the central unique portion of the L segment,
covalently linked to DNA sequences from the end of the 5 segment (Cuifo and
Hayward, 1981; Stegman et al., 1978; Frenkel et al., 1981). Unlike class I series,
class II defective HSV variants display buoyant density that is indistinguishable
from that of parental virus DNA (Frenkel, 1981). Both classes of HSV defective
DNA contain relatively homogeneous populations of DNA molecules consisting
of head-to-tail concatemers. Different size repeat units are observed among
different defective viral isolates. Repeat size differences are primarily due to the
size of contiguous sequences flanking the 5 portion or the central L segment. In
certain cases, deletions or deletion/ substitutions were found in both types of
defective HSV DNA (Frenkel, 1981 ).

Pseudorabies virus (PrV) is an alpha-herpesvirus that infects porcine. In vitro,

38
rabbit kidney cells infected with PrV at high M.O.I have generated two different

types of defective PrV particles (Prl and Pr2) (Ben-Porat and Kaplan, 1976). Fri
and Pr2 defective particles differ from each other in terms of their overall DNA
composition. Yet Pri and Pr2 share some common structural characteristics
(Ben-Porate and Kaplan, 1976; Rixon et al., 1980; Rixon and Ben-Porate, 1979). Pr1
contains DNA sequences derived from two noncontigous regions of PrV genomes
(middle and end of the L segment), consisting of highly reiterated DNA sequences
that have a lower buoyant density than standard viral DNA (Rixon et al., 1980).
In contrast, Pr2 contain DNA sequences with only a few reiterated sequences and
have a buoyant density indistinguishable from that of standard virus DNA (Ben-
Porate and Kaplan, 1976; Rixon and Ben-Porate, 1979). However, the molecular
weights of both Pr] and Pr2 are very similar to each other, and only slightly

smaller than that of standard virus genomes.

C. Replication of defective particles.

Murray et al. (1975) reported that serial passages of undiluted HSV types 1
and 2 resulted in a "cyclic pattern" of infectious standard and defective virus
production. Moreover, cells infected with virus stocks containing a mixture of
standard and defective virus particles resulted in a loss of viral infectivity with
a concomitant increased level of defective virus DNA synthesis. It is noteworthy,
that interference with standard virus replication by defective particles was absent
in cells infected with standard virus prior to challenge with defective viruses,

suggesting that the level of interference may have occurred during an early event

39

of the viral replication process.

Time course infection studies with virus stocks containing a mixture of
standard and defective virus particles were reported by Vlazny and Frenkel
(1981). They proposed existence of a "temporal and enzymatic" relationship
between defective and standard viruses in DNA synthesis. Replication of
defective virus DNA occurs most actively during the late phase of infection, and
is inhibited by the treatment of infected cells with phosphonoacetic acid (an
inhibitor of HSV DNA polymerase), indicating that both defective and standard
virus DNA require the HSV DNA polymerase activity. In addition, they observed
an inverse relationship between the rate of DNA synthesis of defective and
standard viruses.

In contrast to the kinetics of defective HSV DNA synthesis, similar time
course infection studies on defective PrV synthesis indicate that defective PrV
genomes replicate preferentially at an early stage of infection (Wu et al, 1986).
Different kinetics-observed in defective HSV and PrV DNA synthesis are not well
understood. A possible answer may be due to enrichment of different types of
origin sequences that have different functions in DNA replication.

Becker et al. (1978) first proposed a model for defective virus DNA replication.
Based on electron microscopy studies, Backer et al. observed several "replicative
intermediates" in cell fractions containing only defective virus DNA; circular
molecules varying in size with contour lengths (10, 5, 2.5 and less than 2.5um),
linear DNA molecules with the length of intact virion DNA (52um), and circular-

linear DNA molecules, with circular components of either 2.5um or 5.0um, and

40

linear components varying in length from 1-50um. The observation of different
types of "replicative intermediates", and especially the presence of circular-linear
molecules suggest that the 5 component of standard virus DNA (common to all
defective HSV particles) is excised out, and fragmented into smaller circular
molecules during undiluted serial passages. These small circular molecules serve
as templates to generate linear concatemers via a rolling circle mechanism (Becker
et al., 1978).

Additional evidence supporting the rolling circle model of defective HSV
DNA replication was reported by Vlazny and Frenkel (1981). Co-transection of
rabbit skin cells with a monomeric repeat unit of defective virus DNA resulted
in re-generation of progeny defective virus genomes that were identical to the
structure of the original defective virus genome (Vlazny and Frenkel, 1981).
These results indicate that the single monomeric repeat unit can serve as a
template to generate full length head-to-tail concatemers via a rolling circle

mechanism.

D. Cis-acting recognition sites within defective virus genomes.

Successful regeneration of defective progeny DNA from a single repeat unit
parent, provided evidence for two important cis-acting replication elements within
the HSV genomes (Vlazny and Frenkel, 1981; Stow and McMonagle, 1983; Spaete
and Frenkel, 1985). Stow and McMonagle (1983) first reported the identification
of two cis—acting functions necessary for propagation of HSV class I amplicons (a

recombinant vector containing a single repeat unit). In addition, Spaete and

41
Frenkel (1985) have mapped oriL and a cleavage/ packaging signal site within the

amplicon pP2-102 containing a 3.9-kb repeat unit of HSV-1 class II defective
genomes. HSV class I defective genomes contain a replication origin derived
from repeat regions flanking 5 segments of the parental virus genome (oris).
Whereas, HSV class II defective genomes contain a replication origin derived from
a central portion of the L segment (oril).

Other defective herpesvirus particles were utilized to identify origins of DNA
replication, such as Equine herpesvirus I (EHV-I). EHV-l is an alpha-herpesvirus
that infects horses, and causes spontaneous abortions in infected pregnant mares
(O’ Callaghan et al., 1978). Defective EHV-l particles have been obtained from
serial undiluted passages of standard virus genomes. Defective EI-IV-l genomes
were utilized to map and determine nucleotide sequences of an EHV-l origin of
replication (Baumann et al., 1989). Thus, defective particles represent tools to
investigate cis-acting replication functions necessary for the replication of standard
and- defective viruses.

In addition to a functional origin of replication, both classes of HSV amplicons
retained sequences derived from 8 component termini, responsible for the
cleavage/ packaging process (a sequence). Detailed analysis of maturation and
cleavage of concatemeric defective genomes based on transfection-propagation
assays was reported by Deiss and Frenkel (1986). To summarize, functional HSV
amplicons and their deletion mutants were transfected into cells along with helper
virus DNA. Transfection-derived virus stocks were further propagated through

several serial passages and analyzed for the presence of regenerated progeny

42

defective genomes. Only the HSV amplicons containing a functional origin of
replication and a cleavage/ packaging signal site were expected to be recovered
in transection-derived virus stocks after serial propagation. In surrunary, the
results of Deiss and Frenkel led to the following conclusions; 1). The
cleavage/ packaging signal site resides within the a sequence of defective virus
genomes, 2). The cleavage process for viral DNA concatemers is coupled to the
packaging process, 3). The cleavage/ packaging event involves amplification of the

a sequence.

E. Role (5) of defective particles in pathogenesis and virus-host cell
interactions.

Accumulation of defective virus particles is often accompanied by a decrease
in standard virus production with reduction in virus titers. Since the first
description of this phenomenon by Von Magnus (1951), decrease in standard
virus titers have been recognized in the growth of almost all viruses in the
presence of defective particles. Based on available literature, it appears that the
interference of standard virus growth by defective particles is either due to the
ability of defective particles to compete with standard virus at the level of viral
DNA replication or transcription; defective genomes are enriched for origins of
DNA replication due to their concatemeric genomic arrangements, and often
contain several transcription factor binding sites. Alternatively, interference may
occur at the cleavage/ packaging process (Frenkel, 1981; Rixon and Ben-porate,

1980).

43
In HSV systems, the study of standard and defective genome replication

revealed a quantitative correlation between the capacity of various passages
containing different concentrations of defective particles to interfere with standard
virus replication (Schroder et al., 1975 / 76; Frenkel et al., 1975; Vlazny and Frenkel,
1981). In all cases, the degree of interference was dependent upon the initial
ratios of defective to standard virus particles present within a given virus stock.
However, other undetected small alterations within standard virus genomes that
may have arisen during undiluted serial passages were not taken into an account.
Seemingly small alterations in standard virus genome structure may have a
profound effect on the overall outcome of infectious standard virus production.

Analyses of viral transcripts and infected cell polypeptides (ICPs) in cells
infected with HSV virus stocks containing a mixture of standard and defective
viruses revealed the overproduction of ICP4 (Class I defective HSV series) and
ICP8 (Class II defective HSV series) (Murray et al., 1975; Frenkel et al., 1975).
Overproduction of certain ICPs was dependent upon the gene templates located
in each repeat unit of a given defective virus genome. In addition, Frenkel and
co-workers (1981) observed a linear relationship between the degree of ICP
overproduction and the number of reiterated sequences present in defective HSV
genomes. In contrast to overproduction of ICP4 and ICP8, most early (E) and late
(L) gene products were underproduced in cells infected with a mixture of
defective and standard HSV particles (Frenkel et al., 1981).

Development of defective particles from cells infected with Pseudorabies virus

(PrV) at high M.O.I occurs rather slowly when compared to defective HSV or

44
defective EHV-l particles (Ben-Porate et al., 1974). In addition, cyclical increases

and decreases in infectious virus titers observed upon infection of cells with virus
stocks containing defective HSV or EHV-l particles, were not observed with
defective PrV particles (Ben-Porate and Kaplan, 1976). In the PrV systems,
overproduction of some transcripts, for which the defective PrV DNA is enriched,
was observed in cells co-infected with standard and defective PrV viruses (Rixon
and Ben-Porate, 1980). However, the protein profile of the resulting co-infection
was indistinguishable from that observed in cells infected with standard PrV
alone. This indicates an apparent lack of translation of the RNA species that were
overproduced by defective PrV particles. Overproduction of IE gene products
such as HSV ICP4 was not observed and therefore, could not account for the
ability of defective Pr viruses to interfere with standard virus production. It is
noteworthy, that mechanisms controlling the kinetics of gene expression (E, E
and L) located within defective PrV genomes were identical to mechanisms
controlling standard PrV gene expression. Wu and co-workers (1986) reported
that interference of standard PrV by defective PrV particles occurs at an early
stage of infection and is accompanied by transient replicative advantages for
defective PrV, due to enrichment of origin sequences. Moreover, interference
occurs at the level of cleavage and encapsidation of standard virus. Precise
mechanisms regarding interference of standard virus growth by defective virus

particles, however, are yet to be clarified.

F. Defective herpesviruses as eucaryotic expression vectors

45

In general, proteins expressed in mammalian cells require post-translational
modifications that are unique to animal cells. Consequently, the development of
eucaryotic expression vectors that can deliver a high production of authentic,
biologically active protein products is of great interest.

Utilization of defective HSV particles as eucaryotic expression vectors stems
from several considerations; 1). defective HSV particles contain an origin of DNA
replication and a cleavage/ packaging signal site that are essential for virus
propagation (Frenkel, 1981; Spaete and Frenkel, 1985), 2). a single monomeric
repeat unit of defective HSV genomes can regenerate full-length defective
genomes consisting of head-to-tail reiterations, offering the potential for
amplification of foreign DNA ((Vlazny and Frenkel, 1981; Vlazny et al., 1982), 3).
defective HSV genomes appear to be relatively stable upon serial passage in vitro
(Kwong and Frenkel, 1984; Spaete and Frenkel, 1982).

Defective HSV particles (HSV amplicons) have been utilized as an efficient
cloning-amplifying. vector for expression of a chimeric chicken ovalbumin gene
in rabbit skin cells (Kwong and Frenkel, 1985). Briefly, a chimeric amplicon was
constructed by cloning a modified chicken ovalbumin gene, under control of the
HSV-1 ICP4 promoter sequences, followed by co-transfection of cells with the
chimeric amplicon and helper virus DNA. Analysis of polypeptides synthesized
in cells infected with the resultant virus stocks containing regenerated defective
virus genomes revealed an abundant chicken ovalbumin polypeptide of the
correct size.

Because of the ability of HSV-1 to infect and establish latency in neurons, HSV

46

amplicons have attracted much attention in the development of systems to deliver
foreign genes in post-mitotic neuronal cells. Infection of neurons with a chimeric
HSV amplicon vector containing the E. coli lacZ gene driven by the HSV-1
immediate early 4/ 5 promoter in the presence of a helper virus resulted in a high
expression of stable beta-galactosidase activity (Geller and Breakefield, 1988).

As mentioned earlier, the replication of the EBV genome in latently infected
human B-cells requires only oriP and EBN A-1 nuclear antigen. Several reports
indicate that plasmids containing oriP and EBNA-l can stably replicate as
episomes in Various uninfected mammalian cells without helper viruses (Yates et
al., 1985; Lupton and Levine, 1985). Thus, EBV derived replicons have been
utilized as cloning vehicles for the stable episomal replication of cDN A expression
libraries in human lymphoblastoid cells (Margolskee et al., 1988).

Jalanko et al. (1988) conducted extensive comparative studies on mammalian
cell expression vectors, one containing EBV replicon sequences fused to the SV40
early promoter directing the expression of the bacterial chloramphenicol
acetyltransferase (CAT) gene (EBV-SV40-CAT) and the other with the same
construct without the EBV replicon sequences (SV40-CAT) in various mammalian
cell lines. Three important observations were reported; 1). Overall, cells
transfected with the EBV-SV40-CAT construct have clearly higher CAT expression
levels than the SV40-CAT plasmid, 2). The level of CAT expression was highly
variable among tested cell lines, 3). Cells expressing a low level of CAT activity
with the EBV-SV40-CAT plasmid contained mostly oligomeric forms of the

transfected DNA, possibly resulting from the integration of vector DNA into host-

47

cell chromosomes.

Defective MDV genomes (MDV replicon) represent excellent tools to deliver
foreign gene products in avian species. The MDV replicon has advantages over
other defective herpesviruses in several respects. First, the use of natural host
will greatly facilitate study of gene dosages in in viva settings. Secondly, a
naturally occurring non-oncogenic virus can be used as a helper virus to provide
trans-acting functions required for the replication of MDV replicons. Further
characterization of MDV replicon, therefore, will increase our understanding of
molecular events involved in MDV DNA replication. Additionally, basic
knowledge concerning the molecular nature of the MDV replicon can be applied
to construction of delivery systems for expression of foreign genes in vitro as well
as in vivo.

The present dissertation is primarily focused on four objectives: 1). To locate
MDV replicon sequences within the parental virus genome, 2). To locate and
define a functional origin of MDV DNA replication, 3). To examine and compare
MDV replication origin with other herpesvirus origins, 4). To study viral or

cellular origin binding proteins.

Chapter II
Identification and characterization of a Marek's disease virus origin of

replication.

Heidi S. Camp, Paul M. Coussens and Robert F. Silva

J.Virol. 65, 6320-6324 (1991)

48

ABSTRACT

Previously, we isolated a replicon from a defective Marek’s disease virus
(MDV) which was shown to be analogous to defective herpes simplex viruses
(amplicons). Most defective viruses contain cis-acting elements required for DNA
synthesis and virus propagation such as an origin of DNA replication and a
packaging/ cleavage signal site. In this report, the MDV replicon was utilized to
locate an origin of MDV DNA replication. A comparison of MDV replicon
sequences to other herpesvirus replication origin sequences revealed a-90 bp
sequence containing 72% identity to the lytic origin (ori,) of Herpes simplex virus
type 1 (HSV-1). This 90 bp sequence displayed no similarity to beta-herpesvirus
or gamma-herpesvirus replication origins. The 90-bp sequence is arranged as an
imperfect palindrome centered around an A-T rich region. This sequence also
contains a 9-bp motif (5’CGTI‘CGCAC3’) highly conserved in alpha-herpesvirus
replication origins.
. To test functionality of the 90-bp putative MDV replication origin, m1
replication assays were conducted with subclones generated from the 4 kbp MDV
replicon. Results of these experiments revealed a 700-bp MDV replicon
subfragment containing the 90-bp putative MDV replication origin sequence,
capable of replicating in chicken embryo fibroblast (CEF) cells co-transfected with
standard MDV DNA. In conclusion, we have identified a functional origin of
DNA replication in MDV. Similarity of MDV origin sequences to those of alpha-
herpesviruses supports the current contention that MDV is more closely related

to alpha-herpesvirses than gamma-herpesviruses.

49

INTRODUCTION

Marek’s disease virus (MDV) is a highly cell-associated avian herpesvirus.
In chickens, MDV is the etiologic agent of Marek’s disease (MD), a malignant T-
cell lymphoma (Calneck and Witter, 1991). There are three MDV serotypes:
Serotype 1 includes oncogenic MD viruses and their attenuated derivatives,
serotype 2 includes all naturally occurring non-oncogenic chicken herpesviruses
and serotype 3 is the antigenically related non-oncogenic turkey herpesviruses
(HVT). MDV pathology has been extensively characterized (Payne, 1989).
Molecular analysis of MDV, however, has lagged behind that of other
herpesviruses, primarily due to technical difficulties presented by the tightly cell-
associated nature of MDV infection.

The MDV genome, a double-stranded linear DNA molecule of approximately
160-180 Kbp, consists of a unique long (UL) and a unique short (U,) segment
flanked by inverted repeats (T RU IRU IR, TR). MDV has been classified as a
gamma-herpesvirus based on its lymphotropism (Roizman, 1982). However, the
overall genomic structure and colinearity of many MDV genes to those of alpha-
herpesviruses such as herpes simplex virus (HSV) and Varicella-zoster virus
(VZV), suggest that MDV should be re-classified (Buckmaster, 1988).

Previously, we reported the isolation and characterization of a defective
serotype 2 MDV (Carter and Silva, 1990). The defective MDV genome exists as
a high molecular weight head-to-tail concatemer consisting of 4 kbp viral
monomeric repeats. The defective serotype 2 MDV virus (replicon) is analogous

to HSV amplicons (Frenkel et al., 1976).

50

51

HSV amplicons contain multiple head-to-tail reiterations of monomeric repeat
units derived from either end of the Us fragment or from two noncontiguous
regions within the U, and UL fragments of HSV genomes (Frenkel, 1980). HSV
amplicons replicate and propagate in the presence of a helper virus (Frenkel, 1980;
Spaete and Frenkel, 1985; Weller et al., 1985). Molecular analysis of HSV
amplicons revealed three lytic origins (oriL and two copies of mi.) of HSV DNA
replication and a packaging/ cleavage signal site (Vlazny and Frenkel, 1981; Spaete
and Frenkel, 1985).

In this study, MDV replicon DNA was utilized to locate an origin of MDV
DNA replication. We have identified a 90-bp replicon sequence that shares 72%
identity to the HSV-1 ori, and oriL. No significant similarity has been found with
replication origins from either beta- or gamma-herpesviruses (Gahn et al., 1989;
Hamzeh et al., 1990). The 90-bp sequence is arranged in an imperfect palindrome
with alternating A-T-rich sequences. Within its 90-bp sequence is a 9-bp motif
(CGTTCGCAC) that is highly conserved in alpha-herpesvirus origins of
replication and is known to be recognized by the HSV-1 origin binding protein
(Elias and Lehman1988; Koff and Tegtrneyer, 1988). Based on functional
replication assays, we have identified a 700-bp subfragment of the replicon that
contains the 90-bp sequence. This 700-bp subfragment was capable of replicating

in CEF cells in the presence of helper virus.

MATERIALS AND METHODS

Cells and virus stock

Strain 281MI/ 1, a serotype 2 MDV field isolate (Witter, 1983), was plaque
purified from cell-free virus following the fifth cell-culture passage. Preparation
and maintenance of primary or secondary chick embryo fibroblast (CEF) cultures
have been previously described (Silva and Lee, 1984). Secondary CEF were

infected with cell-associated 281MI/1 passage 15 virus stocks.

Isolation of high-molecular weight viral DNA

High—molecular weight viral DNA was isolated in association with cellular
DNA from CEF infected with cell-associated 281MI/1 virus stocks. Upon
evidence of extensive cytopathic effect (CPE), cells were lysed in 150 mM N aCl,
100 mM EDTA, 1% sodium dodecyl sulfate and 100 pg/ ml proteinase K for 424
hr at 37°C. DNA was extracted twice with an equal volume of phenol-chloroforrn-
isoamyl alcohol (25:24:1 vol/ vol/ vol), once with chloroform-isoamyl alcohol (24:1
vol/ vol), and precipitated with 2.5 volumes of 100% ethanol. The DNA was

resuspended in TE (1 mM EDTA, 10 mM Tris-HCl, pH 8.0) and stored at 4°C.

DNA sequence analysis

Isolation and cloning of pAS was described previously (Carter and Silva,

52

53
1990). pASI was constructed by cloning the 4 kbp replicon into pUC18 in the

inverse orientation with respect to pA5. Unidirectional deletion mutants were
created from pA5 and pA5I with exonuclease III (Boehringer Mannheim Corp.,
Indianapolis, IN.) and Si nuclease (Boehringer Mannheim Corp., Indianapolis,
IN .), followed by re-circularization of deleted clones using T4 DNA ligase
(Henikoff, 1987). DNA sequencing was performed using double stranded plasmid
templates and the dideoxy chain termination method (Sanger et al., 1977). In all
sequencing reactions, the Sequenase enzyme (United States Biochemical Corp.,
Cleveland, OH.) was used as recommended by the manufacturer. The MDV
replicon sequences were used to search for homologous sequences from
GENBANK data base using the MacVector (International Biotechnologies, Inc.,

New Haven, CT.) computer program.

Generation of subclones from pA5

pNOT A5 was derived from pA5. In essence, a 2.5 kbp Hygl and C_laI
fragment was replaced with N951 linkers (a gift from A. Finkelstein). p281MI-1
was constructed by cloning a 2 kbp l_3_a__mI-II fragment from the 281MI/1 genome
into pUC18 at the 13me site. pCK300 contains a 300 bp C_la_I to MI
subfragment of pA5 inserted between the Q41 and MI sites in pUC18. A 700
bp 133551 to Iggy subfragment of pA5 was isolated and cloned into pUC18 using

the 13st} and Kpnl sites, to create pA700.

Dpnl resistance assay

22g] resistance assay

4 kbp E_coRI replicon fragment was isolated from pA5 and re-ligated.
Subclones of pA5 as well as the circular form of the 4 kbp replicon were tested
for replication activity by conducting Qp_r1l resistance assays (Stow and
McMonagle, 1983). Secondary CEF were co-transfected with 500 nanograms (ng)
of plasmid DNA and 20 pg of 281MI/1 DNA by calcium phosphate co-
precipitation (Morgan et al., 1990), with the following modifications: Dulbecco
minimal essential medium (DMEM) containing 5% calf serum, 5% fetal bovine
serum and 25% tryptose phosphate buffer was used to seed 5 x 10‘ secondary
CEF cells in 60 mm tissue culture plates. Four hours after adding the calcium
phosphate-DN A co-precipitates, cells were treated with 15% glycerol in serum
free DMEM for 3 min. Cells were rinsed carefully three times with serum free
DMEM, and fresh DMEM was added. 24 hours following co-transfection, cells
were maintained with Flo-199 media containing 4% calf serum. When extensive
CPE (5-8 days post co-transfection) was evident, total high-molecular weight
cellular DNA was isolated as described before.

Approximately 6 pg of total cellular DNA was digested with EQQRI and Dml
(International Biotechnologies, Inc., New Haven, CT.) using conditions
recommended by the manufacturer. Digested DNA fragments were resolved on
0.8% agarose gels and transferred to Zeta Probe membranes (Bio-Rad, Inc.,
Richmond, CA.) as recommended by the supplier. Probes were prepared with
[alpha-”PldCTP using a random primed labeling kit (Bethesda Research

Laboratories, Gaithersberg, MD.) as specified by the manufacturer. Nylon

55
membranes were subsequently hybridized with appropriate 32p-labeled probes

(Budowle and Baechetel, 1990).

GENBANK ACCESSION NUMBER
The nucleotide sequences reported in this paper have been submitted to the

GenBank database. The accession number is M64132.

RESULTS

DNA sequence analysis

We attempted to locate a potential origin of MDV DNA replication by
initiating DNA sequencing of the 4 kbp replicon. As previously reported (Carter
and Silva, 1990), the MDV replicon was derived by cloning a 4 kbp monomeric
repeat unit from a defective MDV genome into the mm site of pUC18. The
restriction endonuclease (RE) map of the resulting clone, designated pA5, is
shown in Fig. 1.

DNA sequences of the MDV replicon were compared with published
sequences of herpesvirus replication origins. A £s_tl to Baal-II fragment of pA5
contained a 90-bp sequence homologous to consensus replication origins of HSV
and VZV replications (Fig. 2A). The 90- bp sequence was arranged in a A-T rich
imperfect palindrome and contained a 9 bp sequence (CGTTCGCAC) that is
highly conserved in alpha-herpesvirus replication origins. Overall, the 90-bp
sequence shares 72% identity to core regions of the HSV-1 ori, and oriL and also
contained 66% identity to origins of VZV and Equine herpes virus type 1 (EHV-I)
replication (Fig. 2B). Neither origins of Epstein-Barr virus (EBV) (9_ri_§ and
9% nor the origin of cytomegalovirus (CMV) replication contained specific
sequences similar to the putative origin of serotype 2 MDV replication (Anders

et al., 1991; Hammerschmidt and Sugden, 1988; Hamzeh et al., 1990).

56

57
Functional replication analysis of the 4 kbp replicon and pA5

Qp_n_l resistance assays were performed to test functionality of the putative
origin of MDV replication identified by DNA sequencing analysis. Dpnl cleaves
only methylated GATC sequences, therefore DNA propagated from dam+ strains

of E. coli is methylated and susceptible to 2,31 digestion. DNA replicated in

 

eucaryotic cells is not methylated at GATC sequences and therefore is resistant
to 2931 cleavage.

CEF cells were co-transfected with 281MI/1 DNA and either pA5 or a circular
form of the 4 kbp replicon contained within pA5. When extensive CPE were
evident on transfected monolayers, total cellular DNA was isolated and digested
with EQRI and M. The resulting fragments were resolved by agarose gel
electrophoresis and transferred to nylon membranes. Membranes were
hybridized with a 32P-labeled 4 kbp replicon probe. _Ec_oRI and 2931 cleavage of
281MI/ 1 infected CEF DNA produced 5.5, 5.2, and 2.8 kbp fragments which
hybridized to the 4 kbp replicon probe (Fig. 3, lane 2). The 4 kbp replicon probe,
however, did not hybridize to uninfected CEF DNA (Fig. 3, lane 1). Co-
transfection of CEF cells with circular E_chI 4 kbp DNA and 281MI/1 DNA,
resulted in a M resistant E_coRI fragment of 4 kbp (Fig. 3, lanes 3), in addition
to the three EQQRI fragments from 281MI/1 helper virus genome. DNA isolated
from CEF cells co-transfected with intact pA5 and 281MI/1 DNA also yielded a
4 kbp E_coRI fragment resistant to ngl cleavage (Fig. 3 lane 4) in addition to the
three 1:1;qu fragments. Methylated input pA5 DNA mixed with 281MI/1

infected CEF DNA was cleaved by anl, indicating that M resistance is not a

58
result of inhibition of Qprrl activity (Fig. 3 lane 5).

Functional replication analysis of pA5 subclones

Based on DNA sequence analysis, a series of subclones from pA5 were
constructed to further localize MDV sequences important for replication of pA5
(Fig. 4). Following co-transfection of CEF cells with each of the pA5 subclones
and 281MI/1 DNA, total cellular DNA was isolated and subjected to ml
replication analysis as described in Materials and Methods. Replication of pA5
would be indicated by the presence of unit-length 2,31 resistant plasmid DNA
following digestion of transfected cell DNA with E_coRI and ngl (mm
introduces a single cut within pA5 subclones) by hybridization to a 32P-labeled
pA5 probe. As expected, intact pA5 was able to replicate in the presence of
helper virus DNA (Fig. 5, lane 5) generating ngl resistant fragments of 4 kbp
and 2.7 kbp of viral and plasmid portions respectively. However, pA5 alone did
not show replication activity (Fig. 5, lane 3). Among tested subclones, only
pA700, which contains the putative MDV origin of replication identified by DNA
sequence analysis, replicated as indicated by a 2,31 resistant fragment of 3.4 kbp
(2686 bp pUC18 plus 700 bp 31 to §§_mI-II MDV DNA) (Fig. 5, lane 9). CEF cells
co-transfected with 281MI/ 1 DNA and either pNOTA5, p281MI-1 or pCK300 (Fig.
4) representing the remainder of MDV replicon DNA did not contain any 12ml
resistant plasmid sequence, suggesting they do not contain an origin of replication
(Fig. 5, lanes 6, 7 and 8). Methylated input pA5, used as a control, was

susceptible to Qpfl cleavage (Fig. 5, lane 1).

DISCUSSION

Localization and characterization of MDV replication origins will allow
construction of efficient MDV replicons for use as shuttle vectors. For example,
HSV amplicons have been utilized as cloning vectors to deliver and express
foreign genes in gigg as well as in .VLVQ (Geller and Breakefield, 1988; Kwong and
Frenkel, 1984; Sawtell and Thompson, 1990). In addition, isolation of an MDV
replication origin will help to identify MDV or cellular proteins required for
efficient replication of MDV genomes in infected cells.

DNA sequencing and functional replication assays of the MDV replicon
suggest that we have identified a functional origin of DNA replication in MDV.
The 4 kbp circular form of the replicon is able to replicate in CEF cells in the
presence of 281MI/1 helper virus DNA. pA5, containing the 4 kbp replicon
inserted into the EQQRI site of pUC18, is also replicated as an episome in CEF cells
co-transfected with 281MI/ 1 helper virus DNA. Therefore, sequences spanning
the EggRI site within the MDV replicon are not essential for amplification of
replicon DNA. However, failure of pA5 to replicate in the absence of helper virus

DNA, suggests that functions provided in trans by helper virus are essential for

 

pA5 replication.

Results of 2,31 resistance assays indicated that the functional origin of MDV
replication contained within pA5 is located in a 700-bp $31 to Karl fragment of
MDV replicon DNA. However, the exact localization of the MDV replication
origin in the 281MI/ 1 viral genome is unknown, due to the lack of complete maps

or genomic clones of serotype 2 MDV. Studies from our laboratory indicate there

59

60

are two copies of replicon sequences within the 281MI/1 viral genome (Silva et
al., 1988). This suggests there may be at least two replication origins in MDV.
A complete RE map of the serotype 2 MDV genome will allow precise localization
of the serotype 2 MDV replication origin within the viral genome.

HSV contains two copies of ori, within TR, and IR, localized by utilizing HSV
amplicons (Frenkel et al., 1976; Frenkel, 1980). By analogy with HSV, the MDV
replicon sequence is most likely to be present in repeats flanking the U, segment
of the MDV genome; assuming that the structure of serotype 2 MDV DNA is
similar to serotype 1 and 3 MDV DNA. However, based on DNA sequence
identity to lytic origins of alpha-herpesvirus replication, Bradly et al. (1989) and
Morgan et al. (1990) have located a putative serotype 1 MDV origin of replication
in repeats flanking the UL region. MDV serotype 2 functional replication origin
sequences, identified in this paper, share 82% identity to putative serotype 1 MDV
replication origin sequences.

The serotype 2 MDV replication origin displays an imperfect palindrome
containing 30 bp of alternating A-T residues, whereas the ori, and oriL of HSV
contain a nearly perfect palindrome with an 18 bp A-T rich region (Deb and
Doelberg, 1988; Stow and McMonagle, 1983). Lockshon et al. (1988) reported that
the center of palindrome containing the 18 bp A-T rich sequence is essential for
HSV-2 replication. However, expansion of the 18 bp A-T rich region to 52 bp by
introducing alternating A-T sequences did not abolish HSV-2 replication function.

Interestingly, MDV replication origin also contains a 9-bp sequence,

CG'I’I'CGCAC, shown to be recognized by the HSV-1 origin-binding protein, the

61
product of UL9 gene expression (Elias and Lehman, 1988; Koff and Tegtrneyer,

1988). This 9-bp sequence is a subset of an 11-bp motif (CGTTCGCACTT), highly
conserved among HSV and VZV replication origins. The lytic origin of EHV-l
replication also contains this 9-bp sequence, suggesting that conservation of this
region must be essential in replication functions among alpha-herpesviruses.
Recently, Bruckner el al. (1991) reported that the HSV-1 origin-binding protein
contains helicase activity, possibly used for unwinding DNA duplex at the origin.
Others have recently reported that proper interaction of origin-binding protein
with orig containing the 11 bp motif, is essential for DNA replication (Hernandez
et al., 1991).

Although MDV is classified as a gamma-herpesvirus, analysis of the MDV
replicon sequences has revealed significant sequence identity to the consensus
replication origins of HSV, VZV and EHV-l and no identity to EBV replication
origins. Our data supports recent findings by others (Brunovskis and Velicer,
1992; Buckmaster et al., 1988) that MDV is more closely related to the alpha-

herpesvirus, than the gamma-herpesvirus.

ACKNOWLEDGMENTS

We thank Dr. A. Finkelstein for providing the pNOTA5 construct. We also
thank Dr. L. F. Velicer and Dr. J. D. Reilly for critical review of the manuscript.
This research was supported in part by Competitive Research Grants (Nos. 85-
CRCR-1-1709 and 88-37266-3983) awarded to R. F. Silva and P. M. Coussens,
respectively under the Competitive Research Grants Administered by the U. S.
Department of Agriculture and the Michigan Agricultural Experiment Station. H.
S. Camp is supported in part by a Michigan State University Biotechnology

Research Fellowship award.

62

63

Figure 1. RE map of pA5.

xur (66403919111 (6647) Sinai (6652)
Sail (6635) Kpni (6656)
Purl (6629) . (6662)

l-iindlll (661 um (666s)
" “Pill (558)

.. (658)
Pet! (958)

      
 
  
  
 
 
     
 
  

BarnHI (1356)

 

e- . (1956)
Secl (1973)

Hindm (2179)

EooRI (3982)

Pet! (3109)

Figure 2. Identification of a putative origin of MDV serotype 2

replication by DNA sequence analysis of the MDV replicon.

DNA sequencing was performed by the dideoxy chain-termination method. A.
Location and DNA sequence of the 90 bp putative origin sequence of MDV
serotype 2 replication. Arrow indicates the 9 bp sequence shown to be recognized
by the HSV-1 origin binding protein (9). B. Alignment of the alpha-herpesvirus
lytic origins of replication with MDV serotype 2 replication origin. Boxed gray

areas indicates identical sequences among other origins of herpesvirus replication.

65

we no we no mo
v. 503. gaggggggflfi

 

mo 40 mo mo
255.388 9585 g 858

no no wo .0 no no 49 0o
0. E0 H» §§§8§§§§§§
max-HO no» g- fag-H8030 ..8 00,5.8- .>i8>8>8>§8 --.838? .3 ->>-..§50 ..8.0 8.
SETHMHMerB. .8003. ..0 18>>8gi§858>8>8i-8§-..8 9.03758- ...0Q0 0
<N<OH¢v ----- ...0’00HA8050 ......... >8>8ij>§§§8>8>g.0. .0 -g.0...00? ..0.".
2.3%. -.0.0.. ..0; 15:38:18 i.8.>8>?8-..§A0 ...gggg.

 

 

66

Figure 3. Replication of the 4 kbp replicon and pA5.

The 4 kbp replicon and pA5 were subjected to M resistance assay to test their
ability to replicate in CEF cells. CEF cells were co-transfected with either the
circular form of 4 kbp replicon DNA (lane 3) or pA5 (lane 4) and 281MI/1 DNA.
7 days post co-transfection, total cellular DNA was digested with §c_oRI and m1,
resolved through a 0.8% agarose gel and transferred to a nylon membrane. The
membrane was probed with 32P-labeled 4 kbp replicon DNA and visualized by
autoradiography. Lanes: (1) Mock-infected CEF DNA; (2) DNA from CEF
transfected with 281MI/1 DNA; (3) DNA from CEF co-transfected with 281MI/1
DNA and the circular form of the 4 kbp replicon DNA; (4) DNA from CEF co-
transfected with 281MI/1 DNA and pA5; (5) 281MI/1 DNA was mixed with

methylated input pA5.

67

 

.‘
«in
5

28>

 

68

Figure 4. Generation of pA5 subclones.
All clones were inserted into pUC18. The location of the putative MDV origin of
replication is indicated as an open circle (0 ). E = §_c_o_RI, le = HpaI, Sm = SmaI,

P=E§tLBH=BamHLK=KanCl=_C_l_a_IandH3=Eu_1_dIII.

69

pA5 subclones

E le-Sm P an K Cl H3 le an PSm HIE

pA5

 

pA7OO a

pCK300 H

E H1 BH PSm HIE
9281MI-1 H LIL—U

Sm

BH
Cl H3 le l l I HEI E

l
pNOTA5

[11
L
'6

70

Figure 5. Replication analysis of pA5 subclones.

Each of the pA5 subclones were tested for replication activity based on p.951
resistance assay as described in Fig. 3. Each lane contains total transfected
cellular DNA digested with @RI and [_)p_111. ”P-labeled pA5 was used to detect
plasmid DNA as well as 281MI/1 DNA. Lanes: (1) pA5 Darn+; (2) Mock-infected
CEF DNA; (3) DNA from CEF cells transfected with pA5; (4) DNA from CEF cells
transfected with 281MI/1 DNA; (5) DNA from CEF cells co-transfected with
281MI/1 DNA and pA5; (6) DNA from CEF cells co-transfected with 281MI/1
DNA and pNOTA5; (7) DNA from CEF cells co-transfected with 281MI/1 DNA
and p281MI-1; (8) DNA from CEF cells co-transfected with 281MI/1 DNA and

pCKOO; (9) DNA from CEF cells co-transfected with 281MI/1 DNA and pA700.

71

4.09

u
.

 

2.7->

 

Chapter III

Marek’s disease virus DNA contains a gene encoding a potential nuclear
DNA binding protein and a HSV a—like sequence.

Heidi S. Camp, Robert F. Silva and Paul M. Coussens

(Submitted to Virology)

72

ABSTRACT

Four RNA transcripts from chicken embryo fibroblast cells infected with
Marek’s disease virus (MDV) strain 281MI/1 hybridized to the 4-kbp MDV
replicon DNA. In an attempt to identify open reading frames encoding the four
transcripts, we determined the nucleotide sequences of 4-kbp replicon DNA
(represents a single monomeric repeat unit of defective MDV genome). Computer
analysis indicates that the 4-kbp MDV replicon DNA contains two intact open
reading frames (ORFs) with common promoter regulatory elements. ORF-A
codes for a putative 204 amino acid protein that shares 21% and 36% amino acid .
sequence identity to nuclear DNA binding proteins such as the EBNA-l of
Epstein-Barr virus and galline, a chicken sperm histone protein, respectively.
ORF-B encodes for a potential 350 amino acid protein, which did not show any
significant amino acid sequence similarity to known protein sequences wthin
Swiss-Protein data base. ORF-B may, therefore, encode a MDV specific protein.
The 5’-region of MDV replicon DNA revealed seven reiterated copies of an 11-bp
motif sharing 8 out of 11 nucleotide sequence identity to DR2 elements of the

herpes simplex virus strain USA-8 a sequence.

73

INTRODUCTION

Marek’s disease, a lymphoproliferative disorder of chickens, is caused by
Marek’s disease virus (MDV), a highly contagious oncogenic avian herpesvirus.
MDV has been classified as a gamma-herpesvirus based on its lymphotropism
(Roizman, 1984). However, the overall genomic structure of MDV and colinearity
of MDV genes to those of alpha-herpesviruses such as herpes simplex virus (HSV)
have led to reclassification of MDV as an alpha-herpesvirus (Roizman e_t al.,
1992). The MDV genome, a doublestranded linear DNA molecule of
approximately 160 to 180 kilobase pairs (kbp), consists of unique long (11.) and
unique short (Us) segments, each flanked by inverted repeats (TRU 1R1, IRS, TR,).
Based upon agar-gel precipitation and virus neutralization, MDV is divided into
three groups (Bulow and Biggs, 1975; Bulow and Biggs, 1975b). Serotype 1 MDV
includes oncogenic strains and their attenuated forms. Serotype 2 MDV consists
of naturally occurring non-oncogenic chicken herpesviruses. Serotype 3 MDV is
an antigenically related non-oncogenic turkey herpesviruses (HVT).

A defective MDV variant, analogous to HSV amplicons, was isolated from a
high-passaged (HP) stock of 281MI/1, a serotype 2 MDV strain (Carter and Silva,
1990). MDV defective genomes exist as high-molecular-weight head-to-tail
concatemers consisting of 4—kbp viral monomeric repeat units derived from repeat
regions of 281MI/1 virus genomes (Carter and Silva, 1990; Silva et a1, 1988). Most
defective herpesviruses contain two important Q's-acting elements necessary for

DNA synthesis and virus propagation; an origin of DNA replication and a

74

75
packaging / cleavage signal site (Frenkel, 1980; Frenkel and Roizman, 1976; Spaete

and Frenkel, 1985; Vlazny and Frenkel, 1981). Because of genetic complexity of
parental virus genomes, defective viruses represent simpler model systems to
study the process of viral DNA replication. Furthermore, defective genomes can
be engineered as shuttle vectors to introduce and express foreign gene products
in eucaryotic cells. Defective genomes replicate as episomes, and properly package
defective DNA into virions in the presence of helper virus. (Kwong and Frenkel,
1985; Geller and Breakfield, 1988; Margolskee gt_ g1., 1988; Jalanko gt_ a_l., 1988; Silva
et al., 1988; Jalanko gt a_l., 1989).

Utilizing defective MDV genomes (MDV replicon), we have identified a
functional origin of MDV DNA replication, based on DNA sequencing analysis
and 129111 resistance replication assays (Camp gt_ a_l., 1991). Structure and
nucleotide sequence of serotype 2 MDV DNA replication origin are strikingly
similar to those of alphaherpesvirus origins (Camp gt a_l., 1991; Baumann gt_ a_l.,
1989; Deb and Doelberg, 1988; Lockshon and Galloway, 1988; Stow and
McMonagle, 1983). We now extend our earlier findings by reporting the analysis
of the entire 4-kbp MDV replicon DNA. Northern blot analysis of infected cell
RNA revealed four poly(A)+ RNA species (0.9, 1.4, 1.8 and 4.5-kbp) encoded
within or near MDV replicon sequences. MDV replicon DNA contains two intact
patential open reading frames (ORF-A and ORF-B). ORF-B may encode a 350
amino acid protein unique to MDV. Computer-aided translation of ORF-A

revealed a potential protein product of 204 amino acids sharing amino acid

sequence similarity to nuclear DNA binding proteins, such as the EBNA-l of EBV

76

and galline, a chicken sperm histone.

The 5’-end of the MDV replicon DNA contains 90% G+C content, and shares
significant nucleotide sequence identity to DR2 elements of the HSV a_ sequence.
The HSV g sequence is thought to play important roles in generating four
equimolar isomers by inverting L-S components relative to each other, and in
cleavage/ packaging processes of replicative intermediate forms of HSV DNA into
preformed virions. The linear arrangement of MDV replicon DNA within the
parental virus genome and strong nucleotide similarity to DR2 elements of the
HSV g sequence suggest that MDV replicon DNA contains a portion of an _a_-like

sequence.

MATERIALS AND METHODS

Cells and virus stocks

Strain 281MI/1, a serotype 2 MDV field isolate has been described previously
(Witter, 1983). Preparation and maintenance of primary or secondary chicken
embryo fibroblast (CEF) cells was essentially as described by Silva and Lee (1984).
Secondary CEF cells were infected with cell-associated 281MI/1 passage 25 (LP)
or passage 100 (HP). Virus stocks were determined to be free of
reticuloendotheliosis and avian leukosis viruses by either complement fixation or

ELISA (Smith e_t a_l., 1979).

Northern blot analysis

Using a guanidinium-phenolzchloroform isolation method (Chomczynski and
Sacchi, 1987), total RNA was isolated from uninfected CEF cells and CEF cells
infected with LP 281MI/1, HP 281MI/1 virus or 52 virus, a recombinant 281MI/1
virus containing two copies of pUC18 sequences (Silva and Witter, 1992). Total
RNA pellets were resuspended in 500 ul of diethylpyrocarbonate treated water
and subjected to poly(A)+ RNA isolation using the PolyATack mRN A kit
(Promega, Madison, WI) as recommended by the manufacturer. Approximately
0.5 ug of poly(A)+ RNA was resolved on 1.2% formaldehyde gels and transferred
to Hybondm-Nylon membranes (Amersham corp. ,Arlington Heights, IL) using
conditions recommended by the manufacturer. MDV replicon DNA was labeled

with [a-3’P]dCTP using a random primed labeling kit (Bethesda Research

77

78
Laboratories, Gaithersberg, MD.) as specified by the manufacturer. Northern

blots were pre-hybridized in 5 x SSPE, 5 x Denhardt’s solution, 0.5% (w/v) SDS
and 100 ug/ ml of sonicated salmon sperm DNA in a 65°C shaking water bath for
1 hour. Denatured, labeled probe was added and incubated at 65°C overnight.
High stringency washing conditions were employed by washing filters twice in
2x SSPE, 0.1% (w/v) SDS at room temperature, and twice in 1 x SSPE, 0.1% (w/v)
SDS at 65 C for 15 min. Filters were removed and subjected to autoradiography

at -70° C.

Nucleotide sequencing

Isolation and cloning of pA5 and pASI were described previously (Carter and
Silva, 1990; Camp gt _a_l., 1991). Unidirectional deletion mutants were constructed
from pA5 and pASI double digested with M and fighl using exonuclease III
(Boehringer Mannheim Corp., Indianapolis, IN.) and SI nuclease (Boehringer
Mannheim Corp.), followed by re-circularization of deleted clones using T4 DNA
ligase (Henikoff, 1987). DNA sequencing was performed using double-stranded
plasmid templates and the dideoxy chain-termination method (Sanger gt gt, 1977).
In all sequencing reactions, FSS]-dATP (N EN, Boston, MA) and the Sequenase
enzyme (United States Biochemical Corp., Cleveland, OH.) were used as
recommended by the manufacturer. Both strands of MDV replicon DNA were
sequenced completely. M13 forward and reverse sequencing primers were used
to determine nucleotide sequences of p281MI-1, a genomic subclone containing

the left flanking region of MDV replicon DNA within the parental virus genome.

79
Construction of p281MI-1 subclone has been described previously (Camp gt 91.,

1991).

Computer analysis

Nucleotide sequences of deletion mutants were entered and compiled into
contiguous segments using SigMan (DNAST AR, Inc, Madison, WI). MDV
replicon sequences were compared with sequences from GENBANK data base
using MacVector 3.5 (International Biotechnologies, Inc., New Haven, CT). Amino
acid sequences of potential ORFs were compared against entries in the Swiss-
Protein data base. Final alignment of amino acid sequences was performed using '
the GAP program of the University of Wisconsin Genetics Computer Group

(Devereus gt gt, 1984)

RESULTS

Detection of transcripts from within or near the MDV replicon sequences

The generation and amplification of a defective MDV genome is depicted in
Fig. 1 adapted from Carter and Silva (1990). pA5 is a recombinant vector
containing a single copy of the 4—kbp monomeric repeat unit derived from a
defective MDV genome. The 4—kbp replicon DNA was used as a probe to detect
transcripts encoded within or near replicon sequences of 281 MI/ 1 genomes. CEF
cells infected with LP 281MI/1 virus stock contained four transcripts of 0.9, 1.4,
1.8 and 4.5-kbp which specifically hybridized to MDV replicon DNA (Fig. 2, lane
2). CEF cells infected with HP 281MI/1 virus stock, from which the MDV
replicon DNA was initially isolated, did not contain 4.5 and 1.8-kbp transcripts
(Fig. 2, lane 3).

We have previously described isolation of a replication competent
recombinant virus (52—virus) that contains two copies of pUC18 sequences
integrated within the MDV replicon DNA at the same location as pUC18
sequences appear in pA5 (Silva et al., 1988; Silva and Witter, 1992). To examine
changes in transcription due to integration of pUC18 sequences, transcripts from
CEF cells infected with 52—virus were also examined. These transcripts were
indistinguishable from those observed in CEF cells infected with LP 281MI/1.
Thus, the insertion of the pUC18 sequences did not appear to disrrupt any local

genes.

80

81
Native structure and nucleotide sequence of MDV replicon DNA

To determine the nucleotide sequence of the MDV replicon DNA, overlapping
unidirectional deletion mutants were constructed from pA5 and pASI. DNA
sequence analysis (Fig. 3) indicated that the MDV replicon DNA is 3926-bp long
and has an overall 65% G+C content. To determine the true linear arrangement
of MDV replicon DNA within the parental LP 281MI/1 virus genome, a 507-bp
segment of p281-1 subclone (Fig. 48) containing the left flanking region of MDV
replicon DNA were aligned with the MDV replicon sequences. Based on this
alignment, MDV replicon DNA begins with a guanosine G , 15 bp upstream of
a _I-_ngl site (Fig. 43). Information on native linear arrangement was critical to
allow accurate assessment of potential ORFs by linearizing MDV replicon

sequences as they would arranged in the parental virus genome (Fig. 3).

Analysis of potential open reading frames in MDV replicon DNA

Computer-aided translation of MDV replicon sequences in all six reading frames
revealed various potential ORFs (Fig. 5). Nucleotide postions of potential
methionine initiation codon, TATA box, CAAT like sequences and
polyadenylation signal sequences (AATAAA homologies) of ORF-A are indicated
in the legend to Figure 3. Computer alignment of translation products from ORF-
B did not show any significant amino acid sequence similarity to known protein
sequences, suggesting that ORF-B may encode a MDV specific protein. ORF-A,
transcribed from left to right, away from the origin of DNA replication, could

potentially encode 204 amino acids (Fig. 5). Comparison of ORF-A with entries

82

in the Swiss-protein data base revealed that ORF-A could encode for a protein
that has amino acid sequence similarities to nuclear DNA binding proteins; 1)
amino acid residues 70 to 132 of ORF-A displayed 36% identity and 56% overall
amino acid sequence similarity to galline, a sperm histone of chicken (Fig. 6A),
and 2) the entire 204 amino acid sequence of ORF-A contained approximately
22% identity and 41% similarity to the 3’-exon of the nuclear antigen EBNA-l of
Epstein-Barr virus (EBV) (Fig. 6B). ORF-A contains a high concentration of
arginine, a feature shared by the sperm histone of chicken, which is composed
primarily of basic amino acid residues. A hydrophilicity plot (window size of 7)
indicated that ORF-A may encode a strongly hydrophilic protein, and did not
contain leader signal sequences which are normally found in virus glycoproteins
(data not shown). To identify a transcript(s) associated with ORF-A, pCK300
subclone (Camp et al., 1991) was used as a probe. The same filter from Fig. 2 was
stripped and re-probed with pCK300. Only the 0.9—kbp transcript hybridized to
labeled ka300 DNA, suggesting that MDV DNA comprising ORF-A encodes a

specific mRNA (Fig. 7).

HSV-1 g-like sequence at the 5’-end of the MDV replicon sequences

Overall G+C content of MDV replicon sequences was 65%. However, the 5’-
end (1 to 500-bp) of MDV replicon sequences were 90% G+C (Fig. 8A). Further
analysis revealed that this region is related to the HSV-1 g sequence. Comparison

of MDV replicon sequences with the HSV-1 9 sequence displayed overall 53%

83
nucleotide sequence identity in a 400 bp overlap (Fig. SB). In particular, seven

copies of an ll-bp repeat motif (CGGTCGTCGCC) within MDV replicon
sequences were strikingly similar to internal short direct repeats (DR2)
(CGCTCCTCCCC) of the HSV-1 strain USA-8 g sequence. Three cytosine residues
were converted to guanine in the DR2 sequence of serotype 2 MDV in comparison
to the DR2 element of HSV-1. In addition to the 11-bp motif, two copies of
direct repeat element of a 25-bp sequence was present at the 5’ end of the MDV
replicon DNA (Fig. 3). Although an _a_-like sequence in serotype 1 MDV was
recently reported by Kishi gt a_l (1991), we did not detect any sequence similarity
between the putative g-like sequence in serotype 1 MDV and MDV replicon

sequences.

DISCUSSION

Based on the recently published RE map of serotype 2 MDV (Ono gt a_l., 1992)
and our own partial mm map of the 281MI/1 genome, generated from a cosmid
library of 281MI/1 DNA (unpublished results), MDV replicon DNA originates
from inverted repeat regions (TRL and IRQ flanking the UL segment. Four
poly(A)+ RNA transcripts (0.9, 1.4, 1.8 and 4.5-kbp) were detected in CEF cells
infected with LP 281MI/1 using the 4-kbp replicon DNA as probe. However, the
1.8 and 4.5-kbp transcripts were absent in HP 281MI/1 virus infected cells.
Alterations in RE patterns of cells infected with 281MI/1 upon serial i3 gitr_o
passage have been extensively studied (Silva and Barnett, 1991). Except for the
hypermolar 4-kbp fragment representing the defective genome in HP 281MI/1,
no changes in E_chI fragments specifically hybridizing with 4—kbp MDV replicon
DNA (5.5, 5.2, 2.8-kbp), have been detected (Carter and Silva, 1990; Silva and
Barnett, 1991). Serial passage i_n_ vi_trg, however, induced alterations in overall RE
patterns of 281MI/1 genome (Silva and Barnett, 1991). In addition, visible
changes in plaque morphology of HP 281MI/1 in comparison to LP 281MI/1
were observed (data not shown). Altered expression of the 1.8 and 4.5-kbp
transcripts observed in HP 281MI/ l infected cell RNA may be the result of point
mutations within MDV replicon sequences. It is possible that the 1.8 and 4.5-kbp
transcripts are encoded partially within MDV replicon DNA and mutations which
may affect expression of these transcripts may have occurred upstream or

downstream of MDV replicon DNA during serial i_n_ vitro passage. Alternatively,

84

85
MDV gene products required for expression of 1.8 and 4.5-kbp transcripts may

be altered or absent in HP 281MI/1. For example, expression of MDV A antigen
(gp57—65) is greatly reduced in attenuated MDV strain of Md11 relative to LP
Mdll. Loss of gp57-65 expression at the transcription level occurs without any
gross alterations in gp57-65 gene structure (Wilson gt g, 1992).

Insertion of two copies of pUC18 at an EQQRI site of MDV replicon DNA
(recombinant 52-virus) did not alter the expression of transcripts associated with
the 4-kbp replicon DNA. The QRI/ pUC18 integration sites also occur outside
of ORF-A and ORF-B. Therefore these positions may represent non-essential sites
for virus gene expression and replication tr; 1119. Repeated serial _i_n_ 1i_trg
passages of 52-virus in cell cultures did not cause any deletions within 52-virus
DNA, suggesting pUC18 integration sites are stable (Silva, personal
communication). However, i; 11:19 passage of 52-virus resulted, in some cases,
in deletion of pUC18 sequences, suggesting that this site is not entirely stable _ig
m (Silva and Witter, 1992).

Based on DNA sequence analysis, we have located two intact potential ORFs
containing conventional promoter elements. Because of a striking nucleotide
sequence identity of serotype 2 MDV origin of DNA replication to those of alpha-
herpesvirus origins of DNA replication (Camp gt a_l., 1991) and biological
similarities between defective HSV and defective MDV, we expected that ORFs
flanking the MDV origin of replication would encode for MDV homologues of
HSV ICP4, ICP22, or ICP47. However, ORF-A, which is located approximately

20-bp upstream of the serotype 2 MDV origin of DNA replication, revealed

86

overall 41% amino acid sequence similarity to the 3’ EBNA-l exon (641 amino
acids) of EBV (Baer gt gl., 1984); Sample gt a_l., 1990). EBNA-l is a nuclear antigen
expressed in human B lymphocytes latently infected with EBV. EBNA-l functions
i_n ga_ns_ by binding to "oriP", the EBV episome replication origin, permitting
episome maintenance and enhancing transcription (Yates gt a_t., 1984; Reisman gt
a_l., 1985; Rawlins gt a_l., 1985). The size of EBNA-l varies considerably, from 68
to 85 kilodaltons among different EBV isolates (Hennessay and Kieff, 1983) and
is expressed from a multiply spliced mRN A. It is possible that ORF-A encodes
an EBNA-l like MDV protein with a smaller molecular weight than EBV EBNA-I.
Alternatively, ORF-A may represent one exon of a spliced gene. Based on
immunofluorescence microscopy and subnuclear fractionations, EBN A-1 was
localized in association with metaphase chromosomes (Petti _e_t_ gt, 1990). This
association may ensure proteins neccessary for episome maintenance are
immediately available to daughter cells.

A MDV nuclear DNA binding protein (MDNA), which may be responsible for
the maintenance of MDV genomes in latently infected lymphoblastoid cell lines,
was reported by Wen et al. (1988). MDNA is a 28-kilodalton (Kd) nuclear antigen
from MSB-l, a T—lymphoblastoid cell line established from MDV tumors. Because
of MDNA is only expressed in latently infected cell lines and not in cells
undergoing productive infection with MDV, it has been proposed that MDNA
may share some biological properties with EBV EBNA-l. Although the size of
MDNA is consistant with the predicted size of ORF-A (25 Kd), the functional

relationship of MDNA to ORF-A is, at present, unclear.

87
An 11-bp sequence, tandemly repeated seven times within the 5’-end of MDV

replicon DNA shares 8 out of 11 nucleotide sequence identity to DR2 elements
present in the g sequence of HSV-1 strain, USA-8. An unusual property of HSV
genomes is isomerization due to inversion of LS components (Hayward et al.,
1975). The HSV g sequence, located at the termini and L-S junction of HSV
genomes, contains a cis-acting site(s) for inversion of L-S components, and a
signal responsible for the packaging/ cleavage process during virus replication
(Vlazny et al., 1982; Mocarski and Roizman, 1981; Dutch gt gt, 1992). Size of HSV
g sequences (250 to 500-bp) is strain specific with size differences mainly due to
the number of DR2 elements (Davison and Wilkie, 1981; Varmuza and Smiley,
1984; Umene, 1991).

A potential g—like sequence (256-bp), structurally similar to the 3 sequence of
HSV-1, has been localized in the L-S junction of serotype 1 MDV (Kishi gt a_l.,
1991). Serotype 1 MDV contains five and ten copies of the g-like sequence within
both termini and the LS junction, respectively. Serotype 1 MDV g-like sequences,
however, do not contain any sequence similarity to the g sequence of HSV-1.
Potential DR2 elements of serotype 1 MDV contain a 6-bp sequence (GGGTTA)
tandemly repeated 17 times, where as the DR2 sequence of HSV-1 contains 19 to
22 copies of a 12-bp sequence (CGCI‘CCTCCCCC). On the contrary, the 6-bp
repeat sequences (GGGTTA) in serotype 1 MDV were also found in human
herpesvirus-6 DNA (Kishi et al., 1988). Our data indicates that the potential DR2
element of serotype 2 MDV is more closely related to the DR2 element of HSV-1

than to serotype 1 MDV.

88
Defective HSV genomes (amplicons), contain a functional origin of HSV DNA

replication and a packaging/ cleavage signal recognition site (g sequence) (Deiss
and Frenkel, 1986). It is noteworthy that HSV amplicon (justin) contains two
copies of an 11-bp sequence (CGCI‘CCTCCCC) identical to DR2 elements of strain
USA-8 of HSV-1. Based on location of MDV replicon sequences within the MDV
genome, similarity of molecular structures between MDV replicon DNA and HSV
amplicons, and the fact that MDV replicons are packaged into defective virus
particles, it is likely that MDV replicon DNA contains an g-like sequence.

Recently, Reilly and Silva (1992) have located a 1.2-kbp region of HVT that is
similar in structural arrangement to the HSV g sequence. The number of 1.2-kbp
fragment copies, directly repeated at both termini and in the HVT L-S junction
region, was variable among different HVT isolates. DNA sequences of the HVT
a-like region, however, is not yet available. However, size and nucleotide
sequence of potential a-like elements in the three MDV serotypes vary
considerably.

Nucleotide sequences between 650-750 of MDV replicon DNA encode a
functional replication origin of serotype 2 MDV (Camp at al., 1990). This region
also shares a significant sequence similarity to a putative origin of serotype 1
MDV DNA replication within the Bg_mI-II-H fragment (Bradly et al., 1988; Morgan,
gt gt, 1991). The putative serotype 1 origin of replication contains bi-directional
promoter regulatory elements presumably used to express pp38 (Cui gt gt, 1991),
a serotype 1 specific phosphoprotein and an uncharacterized immediate-early

transcript (Hong and Coussens; personal communication). However, putative

89
regulatory elements (Spl binding sites and an Oct-1 binding site) (Cui gt gt, 1991;

Morgan gt a_l., 1991) present within the serotype 1 bi-directional promoter/ origin
region are not present in serotype 2 MDV replicon sequences. These differences
accentuate previous contention that MDV serotype 1 and 2 have diverged more

than serotype 1 and 3 MDVs (Silva and Lee, 1984).

Acknowledgements

We wish to thank Dr. jay calvert for review of the manuscript and for their
helpful discussions. This project was supported in part by he Michigan
Agricultural Experiment Station and competitive research grant 85-CRCR-1-1709
awarded to R. F. Silva and grants 90-34116-5329 and 8420-7430 awarded to P. M.
Coussens under the competitive research grants administered by the U. S.

Department of Agriculture.

90

91

Fig. 1. Generation and amplification of a defective MDV genome (Carter and

Silva, 1990).

(A). Schematic diagram of regions of MDV genome consists of unique long (U1)
and unique short (Us) segments flanked by inverted repeats (TRU IRU IRS, TRS).
(B). Partial RE map of IRL and TRL regions of serotype 2 MDV genome. Black
boxes represent MDV replicon DNA. Grey shaded areas represent 281MI/1 virus
DNA flanking MDV replicon sequences. Vertical white bars indicate serotype
2 MDV origin of DNA replication. (C). Schematic diagram of generation of
defective MDV genomes in head-to-tail concatemeric forms during serial in vitro
passages of 281MI/1 virus stock. junction regions joining each monomeric 4-kbp
replicon DNA are depicted as I. (D). A single 4-kbp EcoRI monomer was

isolated and cloned into pUC18, to generate pA5.

92

 

\\

1V

   

 

Replication

 

V

mummmmmmmmmmmammmmawm

Clone EooRI Fragment

 

V

pA5

93

Fig. 2. Detection of transcripts using MDV replicon DNA as a probe.

Poly(A)* RNA was isolated from uninfected CEF cells (lane 1), CEF cells infected
with LP 281MI/1 (lane 2), HP 281MI/1 (lane 3) , or 52-virus (lane 4). 0.5 ug of
each mRN A was resolved in a 1.2% formaldehyde/agarose gel, transferred to
Hybond-Nylon membrane, and hybridized with the 4-kbp MDV replicon DNA

derived from EcoRI digested pA5.

45—»

1.8—-
1.4—+

09—-

94

 

95

Fig. 3. Complete nucleotide sequence of monomeric MDV replicon DNA.

Both strands of MDV replicon DNA were sequenced using the Sanger dideoxy
chain termination method (1976) from overlapping pA5 uni-directional deletion
mutants as described in materials and methods. Nucleotide sequence of MDV
replicon DNA is shown in the reverse complement orientation of pA5.
Translations of two intact ORFs are located under the specific coding nucleotide
sequences. Potential methionine initation codon, TATA box, CAAT like sequences
and polyadenylation signal sequences (AATAAA) are indicated by circles.
Underlines represent a functional serotype 2 replication origin. Arrors represent
25-bp direct repeat elements, and seven copies of 11-bp repeat motif are indicated

by rectangles.

61

121

181

241

301

361

421

481

541

601

661

721

781

841

901

GCCCG

CGCCC

GGGTC

GTCGC

CCGCE CCCC% GCCCG

CGCCC

 

dEGTc

GTCGC

cdcgi}rccrc

 

Gcccéicrccr

 

GTCGT

CCCCC

GCGGA

TCGTT

GGAGG

CGCCG

CGAAA

CGATC

AAGCG

CACGT

cccgg
TCCGT
TCAGC
CCGCC
GAGGG
GTGAG
CCTAC
GCTTA
GAGAT

GATGC

CGCCCIGGTCG

GGTCC

CCCCG

GGTCG

GGGCC

TAGCC

GAGGT

GCGTT

TCCTC

CCGGC

AATGC

GCGCC

GCTCG

GGTCG

GGTCG

GGCCG

TGTGG

CGCCG

TGCCG

GCCCT

AAACG

TCGCC

CGGTC

CTCTC

CGATC

GACGG

GCCTG

GGGCG

CGCTC

CCACC

cciiililii;
AGCGG AG

CQGTC GTCGC
CGCCT CCGGC
GCGCT CCGCA
GCGTC CGTCG
GAGAA GAAGG
CAGTT CGGGA

GGCGG ATGTG

CGATC

TTGCG

CGGGA

TCCGT

TGACC

ACGCG

cctcr
GTCGT
TCCCT
GATCG
GGGAG
AGAGC
TGGGG
TCGCT
TCGCG
GCGCC

TCAGC

CGTCG

TCTCT

CTCGG

GCGCC

GGGGG

GGGGG

GAAGG

CCCGA

GGAAG

GGTCG

GTTCG

ccdcc
CCGGC

CGTCC

GAAGG
AGGCG
GAAGG
CGGGC
CCGGA
TGAGG

CACCG

TCGTC

GTCCG

GGAGG

CCCCG

AGGAG

CCGTC

GGGAG

TCTCG

CCGGC

GCGTA

CGAAC

cccd§_

GGTCG
GGCGC
CCGCA
AGGGG
CGAGG
ACGGC
TTCGG
GCTCT
ACGAT

CAATA

 

TAAGA

GAAGC

GCCGC
P

GTCCG
S

TTATA

GGGAT

CCCCC

P P

GCGGG

G G

TATAT AATA?
(KT1IEIBCCAC
M P

TCCCA CGGGG GCGAT

S H

G

GACCG TCG'I'I‘
D R R
CCGCC GTCGC

ATTAT

GTGTT
R V

G D

CCGCT

S A

GGATC

TGGCG

CGTGT
R V
TCGGG
S G
GGCCG
G R
GGATT

CAAGG

CCGGC
R
GACCT

T

TGCGA

CGCGG
P R
CCGGC

S G

GCCCG CCGTC
P A V
TTCTG GTCGT

ACGCC

CCCGC
P A
CTACA
L Q
CGAAA
R K
TCTTT

GTCCG

GCCGG
P
AATAC

I

G.”

GGCTA
G L
GCGAG

R E

GCGCG GGACC
R G T
TACCG CCGGG

GAAAC
E T
CGGAG
R R
GCGGT
A V
CGAAC

961 AATAA AGCGC

1021

1081

1141

1201

1261

1321

1381

1441

1501

1561

1621

1681

1741

I
CGCGC
A
GTCCC
S
TGGGT
G
CGCGG
A
GATCG

K R

GGCGA

R R

GAGAA

R E

ACCGA

Y R

CGTCC

A S

TCCCG

I V P

CCGCT
R
CGAGG

CGCGG

S R

TCGGA

E V G

GTTCG

TCGGT

CGGGA

AGTTC

GAAGC

CCGTT

GCGGT

TAGCG

CGAGG AGAAT

GTCAG

TTCCG

P P
ACGAA
T N
CGTCG
R R
GGGCA
G H
GGAAC
G T
ATCCC
I P
AACAC
N T
ACCTC
T S
GGAGC

TCTGC

CCCGA

AACCC

AGCTT

AAGTG

S
CCCGT
P
TCTAC
L
TCGAA
R
CGTCG
V
TCGCA
R
GGTCG
V
GGGAA

CCTTG

GCTGT

GAAAG

ATCCG

ACTTT

R I

CCCGT

S R

GGCTC

R L

CTGGC

T G

GACGA

G R

CGAGA

T R

CGGGT

A G

TCGAT

D

CGGCG

AGGGA

GGGGG

CGATA

GTCTG

G F S
TGGGA TCGCA
W D R
GCGTT CGCGG
A P A
CGAGC TCGGC
R A R
GAACG AGGGA
E R G
ACGAG ATCCG
T R S
CGCCC CCCGA
R P P
TGAAA ACGGG

TTCGG CGTGA

CTCTA CGAAA
TCGGC TCGGT
GGTAC GAGGA
CGCCG CCCCC

GCCGG CTCGG

G R

GGCGG

R R

GGGTC

G V

ATCTC

H L

GGACG

R T

ACCCC

D P

CTTTC

T F

CCGGT

GCTCG

GAACC

AGTAA

AGCTT

TCAAC

CGCGA

S F
CCGGG
P G
GCCGA
A D
CGCCT
R L
CCCCC
P P
GCGTC
A S
CCCCC
P P
CGAGA

GGATG

CGATG

CTCGT

GGCCT

CGTTC

GCCGG

T

A G

AAGCG ATCGC

S
CGGGT

D R

GGAAG

G W K

TCGGG

CCGCG

R A A

TCGTT
R
GGCGA
A
GGCCC
A
TAGGA
AGTCT
AGGCC
CTCGG
GCCGC
AGCAT

GCTCG

CTCGT

S R

GCGCC

S A

GCACC

R T

CCGCT

CGGAC

GCGTT

GTTTC

AAAAA

TCACT

GAGAG

R T
GCGCC
A P
GGGGA
G D
ACCCT
T L
CCGCG
P R
GAACG
E R
CGCGG
R G
CTCTC

GCGCG

CTGTT

GATCG

GACTC

TCGGC

AACGC

1801

1861

1921

1981

2041

2101

2161

2221

2281

2341

2401

2461

2521

2581

2641

2701

2761

2821

2881

2941

3001

3061

3121

3181

3241

3301

3361

3421

3481

GTACA TAGGG

GTCCC GCGGT

AGAGC GGGGG

GTCCA TGTGG

TCCCG TGATC

GGTAG AGGGT
V E G
AAATC GTGAG
E N R
AGTTC AGTAT
E F S
ACCGC GGGCT
H R G
CCGAG TAGGG
R V G
GGTTG AGTGG
V E W
TATCC GGAAT
I R N
CAGCC ACCGG
S H R
CGTCT GGAAG
V W K
CTACG GACGA
Y G R
CGCGC CCGGG
A P G
CGCGT ACAGT
A Y S
CTCGT CGTAG

S S
GGTGT CTGGC

CCTCT

TGCGA

GGGGG

TACAG

ATATA

TCCCA

TGTCG

CTACG

CCCCA

CTTGC

CCTCT

TTAGT

AGGGG

TTTCG

AACAG

CGTAT

CGTAT

GAATC

CACAC

ACGGA

ATCGT

GGCTC

TGGGG

TAATG

AACGC

GTCGC
L A
CCCAG
E P
TCGAG
I R
CCGTC
L R
TTCGG
P G
GAAGG
E G
CGAGC
R A
CTCAA
L K
AGTCT
S L
CCGGA
P E
ACGCT
T L
GCGTC
A S
TGCGT

GATGG

CCCCG

GTTGT

CACGC

CGACG

CCCCG

AAGCA

AAGTG

AGGAA

ACCGC

CTCTC

GGGCT GGGCA
TGGAC GAGGC
CATGG TCGGT
CGGGG AATAG
ATATG CCACA

TTTGT TGCAC
L L H
ACCAA TTGCC
R P I
ATTGG TCGCT
D W S
AAACG GAACG
Q T E
CCGGC TATTT
R L F
TCTCG CCTGC
L A C
CGGAC CGGTT
G P V
GCCCT CAGAC
P S D
GTGGA GTTTG
W S L
GAGGG AGACG
R E T
CGTTT CTTGC
V S C
GATCG TTAAG
I V K

ATGAA

CGCGG

AGGCG

GGAGA

GCATC

GGGAG

CGGTC

CCACT

TTAAC

TCCCG

GGCTA

GTCGT

GGTGT

GCGGG

GATGT

ACGTA

ATAAG

GAGCC

CACCT

GACTT

GGCCC

TCCCG

CAGAG

AGTAC

97

ACGAG

GGAAC

GGAGG

AGTTC

ATAGG AGTAG

GAAAT GGGGG

GGTGT GGGAA

CCTGA CGGGG

TA1E::§DAAC ccccc

M
ACACA
T Q
CCTAT
A P
ATGGG
L W
CTCTC
R S
CCCTC
P S
CGGAT
R I
TACGT
Y V
CGCGT
R V
CTTTC
V S
CCGAA
P K
GAAAC
E T
GTGTA
V Y
TTATT

GGAGT
GCGAG
CCGCG
TCGCT
GTTAG
CCTTG
ACGCG
CCGTC
GCCGC
CGGCT

GTTCA

Q P R
AGCAC ACAGT
A H T
ATGTG TTAAC
Y V L
ATCGG GATCC
D R D
TTTTT GGAGC
L F G
CGCCC CCCCC
P P P
CGGGG GGGGG
G G G
TTCAG AGTTC
S E F
CCGCG CCCCG
R A P
GAGGG CCGTG
R A V
ACTGC GGTGT
L P C
GGGGT GGACT
G W T
CGTGG GTGTA
V R V

AAAAA

CCGTC

GGGGG

CCGAG

CGTGG

CTCAC

TTCGG

GCGAC

GGAAC

GCGCA

GTCGT

AATCG

GTCTT

GCGTA

GGGGC

GCCCG

ATCGT

CGGGC

GACCT

CGTGG

CGAAT

CACGC

AGTAC

GTTTA

AGGGA
AGGGG
ACGCC
AGGAG

GACCC
D P
CACTT
H F
GGCCT
T A
GGAAT
P E
GATGT
A M
CCCGA
P E
AGGGG
R G
GAACA
E H
GGCCA
G Q
CGGGG
R G
AAACG
K R
GCTCG
A R
CGAGA
R D
TCTCG

AGTCG

GACGG

GTGCG

CGAGG

CCGCG

GCTAC

AAGTC

TCCAC

ACACA

GTATA

TTAAA

TGGGG AAAAA

ACGGC CTCTC

GCGAC GCCGC

TTGTC AGCTC

TCTCC TCACG
L L T
TTGCG GCGAC
C G D
CCGAC GGAAC
S D G
CCCGT GCGGG
S R A
CGTTT GGGTC
S F G
AGCCG TGCGG
A V R
AAAAG ACGTT
K D V
GGCTC TCACG
A L T
GCCTA GCGTC
P S V
AAACG CAGAG
N A E
AATGC TTAGC
M L S
CCCGC TCTTA
P L L
CCGCG TGCGA
R V R

CTTCG

GTCCC

GGGAA CGTAC

GCGAC

CGCGA

TGTTT

GGAGA

TTAGC

GCGAG

ATTCG

CACAC

GGCGC

GCCCT

CGGGA

AGGGT

GTGCG

CGCAG

CTCGC

TTCGG

AGATG

CCCCA

GACGC

CATGC

GAACG

GGCTC

ACCGG

AACAG
N R
AATGA

AACGG
T T
TACGC
G T
CGGGC
S G
ACCGA
T E
TGCGT
C V
AGCTG
S C
GCAAA
A N
GGCT
G S
GGACG
G R
GCCCT
A L
GCGGT
A V
GTTTC

ATGCG

TAGTT

GGGGA

ATGCG

TTCGT

ACCAG

AAGTG

CACAC

GAACA

TCAGA

3541

3601

3661

3721

3781

3841

3901

CTCCG

CTGCG

AGAAA

CGGAC

TGCAA

AACGC

CGAGG

TGTCG ATGGC

ACGCT CCCGA

ACGTC CGCCG

GCTTA GAGAG

AAAAA AAAGA

GCGGT TATCT

TTAAC TCCCT

GGGGG

GAGGA

TTCGC

CGCGG

ATGTA

CTACG

GCCCG

GGGGG

GGTGT

CCGGC

GCGTT

CGGAG

CCCCC

CGCCC

98

GGGGG

GCACC

CAAGC

TGCGG

AGACC

GACCC

GGGCG

GGGTC

GCCCG

CTGCG

CGCCC

GCACA

GTGTG AGTGA CACGA

GTTTT CTCCG AGACT

GACGC ACCGG TGTCC

GCGTT TAGGC GCGCA

GCGAG CCTCT CCGGC

TGCGA ATGCG CGGTC

ACTTA CTTTC

GAAAA GAGAA

GTACG GCCTT

GCGGC GCGTT

CCGGA GCGGA

GGTCG CATCG

99

Fig. 4. Linear arrangement of MDV replicon DNA within the parental virus
genome.

(A). 281MI/1 sub-genomic map containing MDV replicon DNA (black bar). The
white rectangle represents serotype 2 MDV replication origin. p281MI-1 is a
genomic subclone containing a 2-kbp w fragment which hybridizes to the
4-kbp replicon DNA (see materials and methods). (B). 507-bp nucleotide
sequence of p281MI-1 was compared to the 3926-bp MDV replicon sequence.
Sequences from p281MI-1 identical to the MDV replicon sequences are

underlined.

  

p281 MI-I

13. rrcccurccc

AAAAGGCGGC
CGGCCCGCGG
CGACCGCCCG
CTTCTTCCCC
CCAGTTGCGC
GACTTCCGCG

GGAGGGCGCG
CTGCGG‘TCG
GCTCCACGGG
GCTGCTG

AACCGTGCGG
ACGGGGCCCT
ATTCGGATCC
GAACGCCGCT
GATTCCGCCC
TTCCTCGACG
ATCGCGATCG

CGCAGGGACT
GGGCGTACAC
CBGGTCTCTC

CGGGAAGGCT
CAACGCACAG
CGCTCIIIII
ACTTACICGC
CGGCGCTTCT
TATTCAATCT

cwccccn
It- I

TAACCTCGCG

ATAACGCCGC
CCTACATTCT

100

CCGAGTCCGA
CGGATAGCAA
CCTTTTCTCC
TGGAGAAGCC
GCCTCTGCGC
AAACAAGGAA
AACAACGCCA

ATCCGACCGC
GTTTCCGTCC
TTTTTTTTAG

Smut

 

CCGGAAACGA
GAATTTCCCT
ECCGACTCGC
CCGTATTACT
GCCGGACAIA
GTCGGCCTCC
CGGGGAGCGC

CCATTCGGAT
GGGCCGGAGA
CAAACGCGCC

Pvufl

 

Pbtl Bani! ilndfll Bani! Pt"

101

Fig. 5. Potential ORFs encoded within MDV replicon DNA.

Various of ORFs containing greater that 85 amino acids were located in all three
reading frames (1, 2, and 3) of both replicon strands using the universal start and
stop codons (McVector, IBI). ORFs in the plus strand are depicted as black boxes,

whereas ORFs in the minus strand are indicated as white boxes.

102

 

 

 

 

 

500 1000 1500 2000 2500 3000 3500
— l 1
ORF-A ORF-B
_—
—————
E'-

 

 

 

 

 

It=uj

 

 

 

 

 

 

103

Fig. 6. Amino acid alignments of ORF-A to nuclear binding proteins.

(A). Amino acids 70 to 132 of ORF-A shares 36% amino acid identity to the
galline sperm histone of chicken, and (B). the entire ORF-A shares 22% amino
acid identity to the 3’ exon of EBNA-1 protein. Vertical lines represent identical
amino acids between two aligned sequences while colons and dots represent

similar amino acid sequences calculated according to the BESTFIT program.

104

>.

ompwwnm
owmnv
omppwsm

mel>

w.

me>H
OwMI>
mmZVH
Cwml>
mmZPH
owwnv
meVH
owmn>
me>H

owmuw

N wflwwmwewmwmwwmwwwwwwmmwwwmwwwwww. .Nomvwwmwwm<flaw

do mwefiowevwweZwmmzdwwwv0m0w>mmwwwwrwb>m>0<fibozxaca<w

pm www<mmwwwwwww

wmm

bpw

an
pmw

mu
mom
wa
mmm
Hum

HMO Omwaflwbwmrwrw

mwmw>w0w0wmwomxwwwmwmmommmma. .mwvwwwmwmwwwmmmw<fim>

.L L. L. L. L. L. L. :. .LL LL. LL.
ww<w<wwwwwmmrmemwmmmmmbmoemorOHwmwwmmmbwwm ........

UKmmKIOMQQwUOmmU<va>HmomwbDUvaQwmeflmwQOQUOwaxe
L L L .L. L . L LL.
........ vawm><wxw06><Hxwvvmemememev. .QwefiwwHZm

OSWQwaOOQQmeK ........ QMZH>MwaVbbwwmm<mwaefim062<
”uuuuuL ...... L. L. ..L.L... L.L
mwzdwwwwOmUNDvmwmwwbxF>W§Q<§UOZXQUQ<WQmweQw>wmerw

>O<m<<oomxemr<2rww06>r>HwonwbewrmwrwmoszflwmvowmVb

L. L . LL . u . .L.. L. .L .L. ......
>>6r>>moe<ow. .mwowemmwmwwwH<wHwwewemewmeMvmwwm

wmmH<O<mz<wb06me>m<be>Hx
L. ..... .L ..u. LL .."L.
w2e<>owww6mww>wawom<06m0HO

105

Fig. 7. Detection of transcripts using pCK300 as a probe.
The northern blot filter from Fig. 2 was stripped and re-probed with pCK300

DNA which is located at nucleotide positions from 1143 tol400.

106

 

107

Fig. 8. (A). Percent G+C chart of MDV replicon DNA. (B). Nucleotide sequence
alignment of the 5’-end of MDV replicon DNA containing the 90% G+C content
with the HSV-1 a sequence. Dots represent mismatched nucleotide sequences and
horizontal bars indicate gaps. Alignment of DR2-like elements in MDV replicon

DNA is underlined.

 

 

 

 

 

 

100 7 7
. I I | I l
t: 60 ' ‘m I..P_ ‘WTLH- LL 2“. I it ‘7. Ill~
(D ‘ ‘ m ‘ 1 ‘V I l
‘5 4o -
u I
20
o I

 

 

 

 

 

 

 

 

l
500 1000 1500 2000 2500 3000 3500

108

10 20 30 40 50 60
e

e e e e e
HDV'I'SBQ GCCCGICGCCC GGGTC GTCGC CCGCG GCCGC GCCCG CGCCC GGGflC GTCGC CCQQQ_IQ§IQ

HSV-1 I 370 380 390 400 010
I 500 l OCCCO c-00e m-e GT... eC‘eG eeece «COO .0000 0%.. “-.0. CW

70 00 90 100 110 120
e e e e e e
MDV 'I “830 G CG‘I‘C C
HSV-1 I ‘20 430 440 630 460 470

l 500 l -W

130 140 150 160 170 180
0 0 0 0 0 0
nova-839 Wcmcmmmmmrmmm
HSV-1 I 480 490 500 510 520 530
I 500 ] 0W Com e--e" eeme Geode T-oCT meec We“ .0...)
190 200 210 220 230 240
0 t 0 0 I Q

HDV'I'SIQ CGCCC TCGGT GCCCG GCTCG CTCTC CCGCT CCGCA.!CCCT‘CTCGG CGTCC GGNOO GGCCC
HSV-1 I 500 550 560 570 580 590
‘ 500 1 ..00C Moe exec 0a.. 0.0!. 60“! «0°. 0“.” CT. 00 0%.. m... eeCeD

250 260 270 200 290 300
. 0 0 0 0 .
HDV'I'SBQ mmwcmcmcmmmmcamccoccmcocccccoca
HSV-1 I 600 610 620 630 640
l 500 1 .CG.. Toe'e GOING W... Coeec .CGTC 06.00 00.00 -ecec 9..” COO“ CeeCe>
310 320 330 340 350 360
0 0 . 0' O 0
HDV'I'SBQ TCGET GCGCC GGGCC GGTCG GAOGG GAGAA.GAAGGIGGGAGIGGGGG GAAGG.AGGNG 30000

HSV-1 I 650 660 670 680 690 I 700
I 500 l ec-ee C-.CC 000cc eee'e G.C.G ee-ee -..G. 0.0.6 06006 000” eGeeG eeeGe>

370 380 390 4001 410 420
e e e e e e

HDV'I'SBQ GGAGG GAGGG TAGGG GCCCG GGCTC CAST? CGGGA.AGAGC GGGGG ACGCG CCGTC CGAGG

HSV-1 I 710 720 730 7‘0 750
l 500] eeeee eeGe- ..GCC GeCCe Moo Cecee eGeee .G..C Geeee m

53‘ overall nucleotide sequence identity

Chapter IV
Cellular and viral protein interactions with Marek’s disease virus

replication origin

Heidi S. Camp, Robert F. Silva and Paul M. Coussens

109

ABSTRACT

Previous reports indicate that a serotype 2 Marek’s disease virus (MDV)
replication origin contains a region highly homologous to the herpes simplex
virus (HSV) type 1 origin-binding protein (OBP) recognition site 1 (Camp et al.,
1991). Origin-specific cellular and viral proteins have been investigated using a
synthetic 22 base-pair (bp) oligonucleotide containing a potential OBP recognition
site of serotype 2 MDV. Three protein-DN A complexes were identified in
uninfected chicken embryo fibroblast (CEF) cell extracts. Incubation of the labeled
22-bp MDV ori oligonucleotide with MDV-infected cell extracts revealed two
virus specific protein-UN A complexes (complexes A and B). Competition analysis
indicates that only the complex B is sequence-specific. In addition, we have
detected a protein-DN A complex (complex K) between the 22-bp MDV DNA and
HSV-1 infected cell extracts suggesting the interaction of HSV-1 specific protein(s)

with MDV origin sequences.

110

INTRODUCTION

As with most herpesviruses, genes in MDV appear to be sequentially
regulated in a cascade fashion. Genes of MDV are divided into immediate-early
(IE), early (E) and late genes (L) (Maray et al., 1988). In herpesvirus systems, IE
genes are the first set to be expressed upon initial infection, and transcription of
these genes does not require de novo protein synthesis. The expression of E
genes requires functional IE proteins, whereas the expression of L genes requires
both the functional IE proteins and viral DNA synthesis.

In a previous report, Camp et al. (1991) provided evidence that the structure
and properties of serotype 2 MDV DNA replication origin are strikingly similar
to HSV lytic origins of replication (oriS and oril). Within the serotype 2 MDV
replication origin, a 9-bp motif, sharing 9 out of 11 nucleotide sequence identity
to the 11-bp consensus sequence which is recognized by the herpes simplex virus
type 1 origin binding protein (OBP) (Elias and Lehman, 1988), was identified
(Camp et al., 1991). OBP is the product of UL9, an E gene (McGeoch et al., 1988;
Wu et al., 1988). Based on the following observations, it has been proposed that
OBP is an initiator of HSV DNA replication: 1). In DNase I footprinting and
methylation interference assays, OBP recognizes an overlapping small binding
domain of 8-bp (5’-G'I'TCGCAC-3’/ 3’-CAAGCGTG-5’) located within flanking
regions of oris palindrome (Koff and Tegtmeyer, 1988), 2). Site-specific deletion
analysis of HSV OBP-binding site I revealed a consensus sequence
(CGTTCGCACIT) which is required for specificity and binding affinity to OBP,

3). Mutations within the OBP recognition sites inhibit HSV DNA replication in

111

112
vitro, 4). OBP contains DNA-dependent nucleoside 5’-triphosphatase and ATP-

dependent helicase activities, and 5). OBP exists primarily as a homodimer in
solution and that these dimers specifically interact with lytic origins of I-ISV DNA
replication (Hernandes, et al., 1991; Debrowski and Schaffer, 1991; Koff et al.,
1991; Bruckner et al., 1991).

As with many other DNA viruses, it is likely that MDV encodes proteins
which specifically interact with replication origin(s) during DNA synthesis. In
this report, a 22-mer synthetic oligonucleotide containing a potential OBP binding
site present within the serotype 2 MDV replication origin was used as a probe to
detect origin-specific viral or cellular proteins using mobility-shift assays. '
Incubation of MDV-infected whole cell extracts with the 22-mer MDV ori
oligonucleotide resulted in formation of two virus-specific protein-DNA
complexes (complex A and B), while incubation with uninfected cell extracts
resulted in formation of three protein-DN A complexes (complex M, M’, and M").
Based on competition analysis, only the protein-DN A complex B from the MDV
infected cell extracts was formed by in a sequence-specific manner. In addition,
whole cell extracts isolated from vero cells or HSV (KOS strain) infected vero cells
contained proteins capable of sequence-specific binding to the labeled 22-mer
MDV ori oligonucleotide. Preliminary data suggests that MDV may encode a

protein product analogous to HSV-1 OBP during MDV DNA replication.

MATERIALS AND METHODS

Cells and virus stocks

Strain 281MI/1, a serotype 2 MDV field isolate (Witter, 1983), was plaque
purified from cell-free virus following the fifth cell-culture passage. Preparation
and maintenance of primary or secondary chick embryo fibroblast (CEF) cultures
have been previously described (Silva and Lee, 1984). Secondary CEF were
infected with 281MI/1 passage 15 virus stocks.

African green monkey kidney (vero) cells were grown and maintained as
described (Weller et al., 1983). Stocks of the HSV-1 wild-type virus, strain KOS, -

were grown and assayed as described (Debrowski and Schaffer, 1991).

Preparation of whole cell extracts.

Whole cell extracts were prepared from uninfected or infected CEF cells with
a serotype 2 MDV strain, 281MI/ 1. Cell extracts prepared from MDV infected
CEF cells were isolated from different time points post-inoculation (P1). In
addition, whole cell extracts were isolated from vero cells or HSV-1 strain KOS
infected vero cells. Isolation of whole cell extracts and mobility-shift DNA
binding assays were performed essentially as described (Dabrowski and Schaffer,

1991). Final cell suspensions were aliquoted and stored at -70", until use.

113

114
Preparation and labeling of oligonucleotide and DNA probes.

A 22-mer synthetic oligonucleotide (CAGCGTTCGCACCGCGAACCAA)
containing an HSV OBP binding site I homologe (bold letters) from the serotype
2 MDV replication origin, was synthesized at the Michigan State University
Macromolecular Structure, Synthesis, and Sequence facility. For mobility-shift
assays, double-stranded 22-mer MDV ori oligonucleotide was annealed by heating
to 65° C for 5 min. followed by slow cooling to room temperature. Radiolabeled
oligonucleotide was prepared by incubating approximately 100 ng of the annealed
22-mer oligonucleotide with [y -32P]ATP (7000 Ci / mmol, Dupont New England
Nuclear, Boston, MA) and T4 polynucleotide kinase (Boehringer-Mannheim). '
Unincorporated labeled nucleotides were separated from probe DNA by spin-
column chromatography (Maniatis et al. 1982) through Sephadex G—50
(Pharmacia-LKB, Piscataway, N] ) followed by ethanol precipitation. DNA pellets
were resuspended in TE [10 mM Tris-hydrochloride, pH8.0 and 1 mM
ethylenediaminete-tracetate (EDTA)] at 1 ng/ul and stored at -20° C.

A plasmid construct, p130.HC, containing the entire 90-bp core serotype 2
MDV origin sequence was constructed from the parental construct, pA5 (Carter
and Silva, 1990). First, pPB400 was created by cloning a 400-bp ml to B_am_,I-II
subfragment of pA5 and inserted between the Egg and MI sites in pUC18.
A 130-bp 23:11 fragment containing the origin sequences was then isolated from
pPB400 and cloned into the Sim} site of pUC18 using a blunt-end ligation
procedure (Maniatis et al., 1989). For mobility-shift assays, the ends of the 130-bp

fragment were dephosphosphoryated by treating DNA with 1 unit of alkaline

115
phosphatase (Boehringer Mannheim Corp., Indianapolis, IN.) for 30 min. at 37° C

, followed by extraction of DNA with an equal volume of phenol-chloroform-
isoamyl alcohol (25:24:1 vol/vol/ vo/ ), once with chloroform-isoamyl alcohol (24:1
vol/ vol), and precipitation with 2.5 volumes of 100% ethanol. Recovered DNA
was resuspended in TE (1 mM EDTA, 10 mM Tris—HCI, pH 8.0) and stored at 4°

C. DNA probe was end-labeled by T4 DNA polymerase as described above.

Mobility-shift assays

DNA binding assays were performed as described by Debrowski and Schaffer
(1991), except the N-a-p-tosyl-L-lysine chloromethyl ketone (TLCK) was omitted. ‘
Briefly, 1-2 ng of labeled probes were incubated with approximately 10 ug of
uninfected or infected cell extracts in the presence of 2 ug of poly (dI-dC)
(Pharmacia, Bern, Switzerland) in DNA binding buffer (10% glycerol, 50 mM N -2-
hydroxyethyl peperazine-N’-2-ethanesulfonic acid (HEPES, pH 7.5), 0.1 mM
EDTA, 0.5 mM dithiothreitol, 100 mM Nacl) in a final volume of 20 ul.
Competition experiments were performed by combining labeled probe with 100-
fold excess of cold specific or non-specific competitor DNA prior to initiation of
binding reactions. N on-specific competitor DNA contains a 20-mer synthetic
oligonucleotide derived from I-ISV IE promoter gene sequences. All reactions
were analyzed by polyacrylamide gel electrophoresis (PAGE) on 5%
nondenaturing gels buffered with Tris-borate-EDTA buffer (TBE) (Maniatis et al.,
1990; Garner and Revzin, 1981). Bands of unbound and protein-bound DNA

complexes were visualized by autoradiography.

RESULTS

Detection of protein-DNA complexes with zz-mer MDV ori oligonucleotide and

whole cell extracts isolated from uninfected or MDV infected CEF cells.

Conservation of a potential HSV OBP binding site within the origin of
serotype 2 MDV replication prompted us to initiate experiments to detect cellular
or viral proteins that may interact directly with the MDV replication origin.
Initially, the p130.HC construct containing the entire 90-bp MDV origin sequence
was used as a probe in mobility-shift gel assays. Two protein-DNA complexes, .
M and M’ were detected with mock infected whole cell extracts (Fig. 1, lane 2).
Incubation of the labeled 130-bp DNA with infected cell extracts, isolated 12 hrs
post infection (PI) resulted in formation of a distinct protein-DNA complex
(complex A) (Fig. 1, lane 4), which was absent in mock infected or 5 hrs PI MDV
cell extracts (Fig. 1, lane 3). Protein-DNA complexes from 12, 24, and 48 hrs PI
MDV cell extracts were essentially identical (Fig. 1, lanes 4-6).

Conservation of a potential OBP binding site I within the MDV core origin of
replication suggested that MDV origin binding proteins may interact with this
sequence. To analyze this possibility, a 22-mer oligonucleotide containing the
potential MDV OBP binding site was constructed (Fig. 2). Mobility-shift assays
were performed with the labeled 22-mer oligonucleotide as described in materials
and methods. Three protein-DNA complexes were detected in reactions

containing mock infected cell extracts (M, M’ and M”) (Fig. 3, lane 2). Incubation

116

117
of labeled 22-mer oligonucleotide with MDV infected cell extracts, isolated at

different time points PI, revealed two viral specific protein-DN A complexes (A
and B), which were absent in mock infected cell extracts (Fig. 3, lanes 36).

To test the specificity of protein-DN A complexes observed with MDV infected
cell extracts, competition mobility-shift assays were performed. As expected,
protein-DN A complexes, M, A and B were observed with 12 hrs PI cell extracts
(Fig. 4, lane 1). Unlabeled 22-mer oligonucleotide competed effectively for
binding of all three protein-DN A complexes to the labeled 22-mer MDV
oligonucleotide (Fig. 4, lane 2). Unlabeled non-specific competitor, however, also
competed for binding of complexes M and A. N 0 competition was observed for -
complex B formation. These results suggest that only protein-DNA complex B

was formed in a sequence-specific manner.

Interaction of MDV origin sequences with [(08 infected cell proteins.

A potential HSV OBP-binding site (CGTTCGCACCG) within MDV
replication origin shares 9 out of 11 nucleotide sequence identity (BOLD) to
consensus sequences recognized by HSV-1 OBP. To determine if there are any
interactions between the potential OBP-binding site found in MDV DNA and KOS
infected cell proteins, the radiolabeled 22-mer MDV DNA was incubated with
whole cell extracts isolated from vero cells or KOS infected vero cells. Two
protein-DN A complexes were identified (V’ and V") with either mock-infected or
KOS-infected vero cell extracts (Fig. 5, lane 2). One protein-DN A complex, K, was

observed specifically in KOS- infected cell extracts (Fig. 5, lane 3).

118

To determine if the formation of protein-DNA complex K was due to the
interaction of the OBP in HSV-1 infected cell extracts, different concentrations of
antiserum against a carboxy terminal peptide of HSV-1 OBP (a gift from M.
Challberg) were added to DNA binding reactions for antibody supershift assays.
Although the intensity of the protein-DNA complex K was less with the
increasing concentrations of OBP antiserum, the expected supershift banding
patterns did not occur (data not shown). In addition, immunoprecipitation of
MDV-infected cell proteins with the OBP antiserum did not detect OBP-like
protein in MDV (data not shown). Further experiments are necessary to analyze
the interaction between the HSV OBP and a putative OBP-binding site from MDV A

replication origin.

DISCUSSION

The formation of complexes between viral or cellular proteins with origins of
DNA replication has been reported and characterized in a number of virus
systems (Challberg and Kelly, 1989). T antigens of Simian virus 40 (SV40) and
polyomavirus, the terminal protein of adenovirus, and the OBP of HSV-1 interact
directly with correspponding origins of virus DNA replications (Borowiec etal.,
1990; Gaudray et al., 1981; Debrowski and Schaffer, 1991).

As in other herpesviruses, MDV likely encodes many DNA metabolic proteins,
necessary for DNA replication. The expression of MDV late genes is prevented
by phosphonoacetic acid (PAA) which is known to specifically inhibit herpesvirus .
DNA polymerase activity (Lee et al., 1976). These results suggest a functional
similarity between MDV polymerase and those of other herpesviruses.
Identification of a conserved OBP recognition site within a functional MDV
replication origin, similar to the HSV-1 OBP binding sequences, strongly suggests
existence of an OBP-like protein that may interact directly with MDV DNA
replication origins.

Initially, Elias and Lehman (1988) identified and purified the HSV-1 OBP
which interacts directly with Orig. Since then, it has been proposed that OBP is
one of the required components for origin-dependent initiation of HSV-1 DNA
replication. Interestingly, the 11-bp motif which was shown to interact physically
with the HSV-1 OBP is highly conserved among origins of alphaherpesvirus
replication, such as VZV, EHV-l, and MDV (Stow and Davison, 1986; Baumann

et al., 1989; Camp et al., 1991). The 22-mer MDV oligonucleotide containing

119

120
a potential OBP recognition site is derived from the left end of the serotype 2

MDV origin sequence palindrome. At least one virus-specific protein-DNA
complex (complex B) is formed via a sequence-specific binding with labeled 22-
mer MDV oligonucleotide, suggesting the existence of a HSV OBP-like protein in
MDV infected cells. However, failure of HSV-1 OBP antiserum to interact with
protein-DNA complexes in MDV infected cell extracts suggests that MDV may
encode an OBP that has different antigenic properties than HSV-1 OBP.
Alternatively, the specificity of HSV-1 OBP antiserum, which is directed against
the carboxy termini of the OBP, is not sufficient to bind to MDV OBP-like protein-
DNA complexes.

Based on the work of Debrowski and Schaffer (1991), conservation of the 9-bp
OBP binding sequence in MDV should be sufficient for interaction with the HSV-1
OBP. In support of this hypothesis, a virus-specific protein-DN A complex, K,
was observed in reactions containing HSV-l strain KOS-infected cell extracts and
the 22-mer MDV oligonucleotide. However, further investigation is required to

determine if protein-DN A complex K is related to the HSV-1 OBP.

121

P: p130.HC

 

123456

Figure 1. Mobility-shift assay using 1 ng og end-labeled 130-bp fragment probe
and 10 ug of CEF whole cell extracts (lane 2), or 10ug of MDV-infected whole cell
extracts from 5, 12, 24 and 48 hrs PI (lanes 3-6). Control reaction (lane 1)

contained labeled probe only without cell extracts.

122

Figure 2. Serotype 2 MDV origin sequences. 22-mer oligonucleotide containing

a potential HSV OBP binding site is indicated.

123

ZU< OE

 

5084500 9.3.89an Strong—E. q>>0>d>a> a>e>a>>a>a
”p.58. O—mmo ,

 

>H4>HHOOOD EOOHODG> >OQOOODHOD 0009x5160

124

Figure 3. Mobility-shift assay using end-labeled 22-mer ds MDV oligonucleotide
probe (1 ng) and 10 ug of CEF whole cell extracts (lane 2), or 10 ug of MDV-
infected whole cell extracts from 5, 12, 24 and 48 hrs PI (lanes 3-6). Control

reaction (lane 1) contained labeled 22-mer DNA only without cell extracts.

125

22—merOri oligo

p:

 

3456

2

126

Figure 4. Competition mobility-shift assay using end-labeled 22-mer ds MDV
oligonucleotide probe (1 ng/ lane). Probe DNA was incubated with 100 fold
excess of unlabeled specific (lane 2) or non-specific (lane 3) double-stranded
oligonucleotides at 37° C for 30 min in the presence MDV-infected protein
extracts. Control reaction (lane 1) contained labeled probe with MDV-infected

protein extracts without any competitor DNA.

127

o
.m.
..o.
a
e
_m
2
2

P:

A»

 

B»

    

123

Figure 5. Mobility-shift assay using end-labeled 22-mer ds MDV oligonucleotide
(1 ng) with 5 ug of uninfected vero cell extracts (lane 2), or with 5 ug of KOS

infected vero cell extracts. (lane 3). Lane 1 contained labeled probe only as a

control.

Chapter V

Summary and Conclusions

Defective MDV genomes represent a useful system to study MDV DNA

replication and for construction of a shuttle vector system for delivery of foreign

gene products into eucaryotic cells. Previous studies indicated defective MDV

genomes exist as high-molecular-weight viral DNA consisting of head-to-tail

concatemers (Silva et al., 1988; Carter and Silva, 1990). A single monomeric

repeat unit of a defective MDV genome was isolated and cloned into a bacterial
plasmid, designated pA5 (Carter and Silva, 1990).

By analogy with defective HSV particles (HSV amplicons), a single monomeric
repeat unit of defective MDV genomes should contain at least a subset of the
same cis-functions (replication origin and cleavage/ packaging signal site)
operating in the replication and packaging of standard virus. In search of cis-
acting elements necessary for standard and defective MDV DNA replication, a 4-
kbp MDV replicon DNA was completely sequenced and analyzed. Comparisons
of MDV replicon DNA sequences with other viral replication origin sequences
revealed a potential serotype 2 MDV replication origin within a 90-bp fragment.
Sequences within this fragment are arranged in an imperfect palindrome, centered
around 30-bp of alternating A-T residues. Primary structure of this MDV
replication origin is very similar to alpha-herpesvirus replication origins; the 90-bp
sequence shares 72% nucleotide sequence identity to core regions of HSV-1 oriS
and oril, and 66% nucleotide sequence identity to replication origins of VZV and
EHV-l. Additionally, replication origin sequences of serotype 2 MDV were

129

130

similar (75% nucleotide sequence similarity) to a 76-bp region in the inverted
repeats flanking the unique long region of the serotype 1 MDV genome (Bradley
et al., 1988). Although MDV is a lymphotropic herpesvirus like EBV, neither
origins of EBV (oriP and oriLyt) contained significant sequence similarity to the
potential origin of serotype 2 MDV replication.

To test functionality of the putative serotype 2 MDV replication origin
identified by DNA sequence analysis, ngI resistance assays were performed.
2,31 recognizes and cleaves only methylated GATC sequences. Progeny DNA
molecules from dam+ strains of E. coli are methylated at GATC sequences and
thus susceptible to _ngl digestion. In contrast, progeny DNA molecules
replicated in eucaryotic cells are not methylated at GATC sequences and therefore
are resistant to M cleavage. Several subclones (pNOT A5, p281MI-1, pCK300
and pA700) of pA5 were constructed and tested for their replication activity via
12ml resistance assays. Among tested subclones, only pA700, which contains the
90-bp putative MDV replication origin replicated in the presence of a helper virus.
Therefore, the MDV replicon DNA contains a functional origin of MDV DNA
replication.

In addition to the functional replication origin sequence, MDV replicon DNA
revealed various potential open reading frames (ORFs). Two ORFs (ORF-A and
ORE-B) containing common promoter elements were further analyzed. ORE-B
encodes a potential MDV-specific 350 amino acid protein product; computer
alignment of the translation products of ORF-B did not show any significant

amino acid similarity to known protein sequences. Due to significant nucleotide

131
sequence similarity between the serotype 2 MDV replication origin and HSV lytic

replication origins (oris and oriL), and biological similarity between defective MDV
and defective HSV led us to expect that the ORFs flanking the MDV origin of
replication may encode for MDV homologues of HSV ICP4, ICP22, or ICP47. On
the contrary, ORF-A, which is located approximately 20-bp upstream of the
serotype 2 MDV replication origin, revealed a potential 204 amino acid protein
product sharing overall 21% amino acid sequence identity to the 3’ exon of
EBNA-l encoded by Epstein-Barr virus. EBNA-l is one of three nuclear antigens
expressed only in cells latently infected with EBV. Although functional
relationships between EBN A-1 and the potential protein product of ORF-A are '
unknown, several lines of observation indicate that ORF-A could code for a
potential nuclear DNA binding protein. First, MDV and EBV are both
lymphotropic oncogenic herpesviruses, and therefore, can infect and establish
latency in lymphocytes. A population of MDV and EBV genomes exists as
circular forms in latently infected lymphocytes, or integrate into host cell
chromosomes. EBNA-l functions in trans by binding to "oriP", allowing
maintenance of the circular EBV genome structure and enhancing transcription
of other nuclear antigens. Secondly, a 28-kilodalton (Kd) MDV nuclear antigen
(MDNA) was isolated from lymphoblastoid cell line established from MDV
tumors (Wen et al., 1988). The MDN A was not detected in lytically infected with
MDV cells, suggesting that MDNA may share some biological properties with
EBV EBNA-l. The functional relationship of MDNA to ORF-A is unclear at

present, however, it is likely that MDV encodes a protein product(s) required for

132

maintenance of MDV genomes as circular forms. Thirdly, although defective
MDV and defective HSV genomes share some biological similarity, they differ in
several respects. Based on recently reported restriction endonuclease maps of a
serotype 2 MDV genome (Ono et al., 1992), MDV replicon DNA originates from
repeats flanking the UL component, whereas HSV class I amplicons originate from
the terminal repeat region of the 5 component. Therefore, gene templates located
within defective MDV and HSV genomes are most likely different.

Overall %G+C content of MDV replicon DNA was 65%, significantly greater
than the proposed G+C content of standard virus genome (45%). The 5’ end
region of MDV replicon DNA is composed of approximately 90% G+C. Further '
analyses revealed that the 5’ end region of MDV replicon contained considerable
nucleotide sequence identity to the HSV a sequence, which is located in the L-S
junction and both termini of the HSV genome (overall 53% sequence identity).
The HSV a sequence plays important roles in inversion of L-S components
relative to each other, and in packaging/ cleavage process of HSV replicative
intermediates into nucleocapsids. Although further investigation is required to
test the functionality of a potential a-like region of MDV replicon DNA, it is likely
that MDV replicon DNA contains a cis-acting element that controls the
cleavage/ packaging process of head-to-tail concatemers of defective MDV DNA.
The highly cell-associated nature of MDV infection, however, will cause
difficulties in performance of functional propagation assays, in which the
production of cell-free virus is essential. Additionally, MDV infection often results

in cell-to-cell fusion, presenting a different route of MDV infection other than via

133

infectious enveloped particles.

Most herpesviruses encode proteins necessary for viral DNA replication.
Within the serotype 2 MDV replication origin, two copies of a 9-bp sequence
present in the flanking regions of AT-rich sequences share a significant nucleotide
sequence identity to the HSV origin binding protein (OBP) recognition site I. A
synthetic 22-bp oligonucleotide containing a single copy of the 9-bp sequence was
constructed (ori 22), and used as a probe to detect virus-specific proteins
interacting with the serotype 2 MDV replication origin. Mobility shift and
competition assays were performed with labeled ori 22 and whole cell extracts
prepared from uninfected or MDV infected cells. A virus-specific protein .
(protein-DN A complex B) interacting specifically with ori 2 probe DNA was
identified. To test if protein-DN A complex B was related to the HSV OBP,
antibody super-shift assays and irnmunoprecipitations were performed using anti-
serum directed against the carboxyl terminus of HSV OBP. Super-shifted bands
representing the protein-DN A complex B were not observed even with a high
concentration of the HSV OBP anti-serum. Additionally, HSV OBP antiserum
failed to precipitate a protein from infected cells. Several factors are implicated
in these results. First, anti-serum directed against HSV OBP may not recognize
any epitope in an MDV OBP homologue. The HSV OBP recognizes two
additional sites (site II and 111) within lytic origins of HSV DNA replication. OBP
recognition site II is similar in sequence to site I of HSV, whereas the potential
OBP binding site I of MDV is repeated twice in the MDV replication origin. The

OBP binding site III has not been observed in MDV. Therefore, MDV may

134
encode a protein product(s) similar in function but different structurally when

compared to the HSV OBP. Further studies are necessary to investigate MDV-
specific protein(s) that may function as a initiator during MDV DNA replication.

Optimization of MDV whole-cell or nuclear extract preparations is pertinent
to continue MDV OBP research. MDV-infected cell extracts degrade rapidly when
compared to HSV-infected cell extracts. Addition of proteinase inhibitors in
extract preparations, thus will be helpful to isolate, and stOre intact MDV-infected
cell extracts.

Additionally, future experiments should include the identification of MDV
UL9 gene homologue based on low stringency hybridizations using HSV UL9 ’
gene as a probe, which will facilitate understanding of MDV DNA repliation
process. Based on gene colinearity of MDV genes and HSV and VSV, it is likely
that MDV UL9 gene homologue is located in the unique long region of MDV
genomes.

An intriguing aspect of this report is that MDV contains properties of both
alpha- and gammaherpesviruses at the molecular level. Although MDV has been
re-classified as an alpha-herpesvirus based on MDV gene colinearity and sequence
homology to alpha-herpesviruses, the results obtained from this study suggest
that further investigation is required to clearfy the lineage of MDV within the
herpesvirus family.

Study of MDV DNA replication is far behind those of other herpesviruses.
In essence, MDV DNA replication research has just begun. Results obtained from

this research project can be utilized for future experimentation and research

135
direction. A systematic deletion/ substitution analyses across the serotype 2 MDV

replication origin sequences will required to delimit the core MDV origin
sequences necessary for its replication activity. Additionally, mutational analyses
within the potential OBP recognition site I of MDV and its affect on MDV DNA
replication are of interest. Comparisons of alphaherpesvirus replication origins
indicate that they contain variable numbers of alternating A-T residues in the
center of palindromic origin regions. A putative serotype 1 MDV replication
origin contains 30-bp alternating A-T sequences, whereas HSV lytic origins of
replication contain 18-bp A-T sequences. Functions implicated for this common
AT-rich sequence are not clear. However, it is likely that the AT-rich region may .
play an important role in formation of secondary structure that is essential for the
initiation of DNA replication. Deletion analyses of the AT-rich region of serotype
2 MDV replication origin may help us to understand the role of AT-rich regions
play during herpesvirus replication.

Based on nucleotide sequencing analysis, the MDV replicon DNA contains at
least two potential open reading frames. ORF-A is especially interesting because
of its potential amino acid sequence similarity to EBNA-l of EBV. Generation of
a fusion protein product and corresponding anti-sera from ORF-A will be helpful
in examining a functional relationship between EBN A-1 and MDV ORF-A.

Identification of a potential a-like sequence in MDV suggests that the MDV
replicon DNA is packaged into nucleocapsids as in standard virus genome,
suggesting that defective MDV genomes can be propagated continuously in the

presence of a helper virus. Because of a highly cell-associated MDV infection, it

136
is difficult to perform standard functional virus propagation assays. Additionally,

the densities of MDV DNA and host cell DNA are very similar and thus, isolation
of pure viral DNA from total infected cell DNA is difficult. Nucleotide
sequencing analysis of MDV replicon DNA indicated that the overall % G+C
content of MDV replicon DNA is 65%, which is greater than the % G+C content
of standard virus DNA (45%). This result offers a possibility to isolate pure
defective MDV DNA, separated from host and standard virus DNA by standard
density centrifugation methods. Isolation and analysis of intact defective MDV
genomes are essential to understanding the native structure of defective MDV
genome, and for manipulation of MDV replicon DNA as a shuttle vector to ‘
deliver foreign genes in vitro and in viva.

Only 281MI/ 1 strain has been used as a helper virus to propagate the MDV
replicon DNA. Identification of other MDV strains which may function as helper
viruses will be of great interest in light of recombinant virus technology. For
instance, serotype 3 MDV (HVT) is the most widely used vaccine against MD.
Also, bivalent vaccines containing 584 have been reported to offer a greater
protective immunity against MD. Thus, it would be useful to introduce
recombinant MDV replicon DNA, containing genes of interest, in chickens

inoculated with HVT vaccine or bivalent vaccine strains.

LIST OF REFERENCES

Adams, A. (1987). Replication of latent Epstein-Barr virus genomes in Raji cells.
J. Virol. 61, 1743-1746.

Albrecht, M., Darai, G., and Flugel, RM. (1985). Analysis of the genomic termini
of tupaia herpesvirs DNA by restriction mapping and nucleotide sequencing. J.
Virol. 56, 466-474.

Alford, C.A., and Britt, WJ. (1985). Cytomegalovirus, p. 629-660. _Ir_1_ B. N. Fields
et al. (ed.), Virology. Raven press, New York.

Anders, D.G., Kacica, M.A., Pari, G., and Punturieri, SM. (1992). Boundaries and
structure of Human cytomegalovirus oriLyt, a complex origin for lytic-phase DNA
replication. J. Virol. 66, 3373-3384.

Anders, D.G., and Punturieri, SM. (1991). Multicomponent origin of
cytomegalovirus lytic-phase DNA replication. J. Virol. 65, 931-937.

Anderson, A.S., Francesconi, A., and Morgan, R.W. (1992). Complete Nucleotide
sequence of the Marek’s disease virus ICP4 gene. Virology 189, 657-667.

Anvret, M., Karlsson, A., and Bjursell, G. (1984). Evidence for integrated EBV
genomes in Raji cellular DNA. Nucleic Acids Res. 12, 1149-1161.

Bacon, L.D., Witter, KL, and Fadly, AM. (1989). Augmentation of retrovirus-
induced Lymphoid Leukosis by Marek’s disease herpesviruses in white leghorn
chickens. J. Virol. 63, 504-512.

Baer, R., Bankier, A.T., Biggin, M.D., Deininger, P.L., Farrell, P. 1., Gibson, T.J.,
Hatfull, G., Hudson, G.S., Satchwell, S.C., Ssguin, C., Tuffnell, RS. and Barrell,
BC. (1984). DNA sequence and expression of the B95-8 Epstein-Barr virus
genome. Nature. 310, 207-211.

Bankier, A.T., Dietrich, W., Baer, R., Barrell, B.G., Colbere-Garapin, F.,

Fleckenstein, B., and Bodemer, W. (1985). Terminal repetitive sequences in
herpesviruses simiri virion DNA. J. Virol. 55, 133-139.

137

137

Baumann, R.P., Yalamanchili, RR. and O’Callaghan, DJ. (1989). Functional
mapping and DNA sequence of an equine herpesvirus 1 origin of replication. J.
Virol. 63, 1275-1283.

Becker, Y., Weinberg-Zahlering, Y.A.E. and Rabkin, S. (1978). Defective herpes
simplex virus DNA: Circular and circular-linear molecules resembling rolling
circles. J. Gen. Virol. 40, 319-335.

Ben-Porat, T., Demarchi, J.M. and Kaplan, AS. (1974). Characterization of
defective interfering viral particles present in a population of pseudorabies
virions. Virology 61, 29-37.

Ben-Porat, T., Lonis, B. and Kaplan, AS. (1975). Further characterization of a
population of defective, interfering pseudorabies virions. Virology 65, 179-186.

Ben-Porat, T. and Kaplan, AS. (1976). A comparison of two populations of
defective, interfering pseudorabies virus particles. Virology 72, 471-479.

Ben-Porat, T., Kaplan, A.S., Stehn, B., and Rubenstein, AS. (1976). Concatemeri .
forms of intracellular herpesvirus DNA. Virology 69, 547-560.

Borowiec, J.A., Dean, R.B., Bullock, RA, and Hurwits, J. 1990. Binding and
unwinding-how T antigen engages the SV40 origin of DNA replication. Cell 60,
181-184.

Bradley, G., Hayashi, M, Lancz, G., Tanaka, A., and Nonoyama, M. (1989).
Structure of the Marek’s disease virus BamHI-H gene family: Genes of putative
importance for tumor induction. J. Virol. 63, 2534-2542.

Bronson, D.L., Dreesman, G.R., biswal, n. and Benyesh-Melnick, M. (1973).
Defective virions of herpes simplex virus. Intervirology 1, 141-153.

Bruckner, R C., Crute, 1.1., Dodson, MS, and Lehman, LR. (1991). The herpes
simplex virus 1 origin binding protein: A DNA helicase. J. Virol. 266, 2669-2674.

Brunovskis, P., and Velicer, LP. (1992). Genetic organization of the Marek’s
disease virus unique short region and identification of Us-encoded polypeptides,
p. 74-77. In proceedings of World’s poultry congress, Vol. 1, Amsterdam, the
Netherlands.

Buckmaster, A. E., Scott, S.D., Sanderson, M.J., Boursnell, M.E.G., Ross, N .L.J.,
and Binns, M.M. (1988). Gene sequence and mapping data from Marek’s disease
virus and herpesvirus of turkeys: Implications for herpesvirus classification. J.
Gen. Virol. 69, 2033-2042.

138

Budowle, B., Baechetel, ES. (1990). Modifications to improve the effectiveness of
restriction fragment length polymorphism typing. Appl. Theor. Electro. 1, 181-187.

Bulow, V.v., and Biggs, RM. (1975). Differentiation between strains of Marek’s
disease virus and turkey herpesvirus by immunofluorescence assays. Avian
Pathol. 4, 133-146.

Bulow, V.v., and Biggs, P.M. (1975b). Precipitating antigens associated with
Marek’s disease virus and a herpesvirus of turkeys. Avian Pathol. 4, 147-162.

Camp, H.S., Coussens, P.M., and Silva, RF. (1991). Cloning, sequencing, and
functional analysis of a Marek’s disease virus origin of DNA replication. J. Virol.
65, 6320-6324.

Calnek, B.W. (1985). Marek’s disease-model for herpesvirus oncology. CRC
Critical Reviews in Microbiology 12, 293-320.

Calnek, B.W., and Adldinger, H.K. (1971). Some characteristics of cell-free
preparations of Marek’s disease virus. 508-517. .

Calnek, B.W., Adldinger, H.K., and Kahn, DE. (1970). Feather follicle epithelium:
a source of enveloped and infectious cell-free herpesvirus from Marek’s disease.
219-233.

Calnek, B.W., Carlisle, J.C., Fabricant, J., Murthy, KK., and Schat, K.A. (1979).
Comparative pathogenesis studies with oncogenic and nononcogenic Marek’s
disease viruses and turkey herpesvirus. Am. J. Vet. Res. 40, 541-548.

Calnek, B.W., Ubertini, T., and Adldinger, H.K. (1970). Viral antigen, virus
particles, and infectivity of tissues from chickens with Marek’s disease. J. Nat.
Cancer Inst. 45, 341-351.

Calnek, B.W., and Witter, R.L. (1991). Marek’s disease, p. 342-385. In B. W. Calnek,
H. J. Barnes, C. W. Beard, W. M. Reid, and H. W. Yoder Jr (ed.), Diseases of
poultry, Ames, Iowa.

Carter, J.K., and Silva, RF. (1990). Cell culture amplification of a defective Marek’s
disease virus. Virus genes. 4:225-237.

Challberg, M., and Kelly, T.J. (1989). Animal virus DNA replication. Annu. Rev.
Biochem. 58, 671 -71 7.

Cebrian, J., Kaschka-Dierish, C., Berthelot, N ., and Sheldrick, P. (1982). Inverted
repeat nucleotide sequences in the genomes of MDV and herpes virus of turkeys.
Proc. Natl. Sci. USA 79, 555-558.

139

Chen, X., and Velicer, LP. (1991). Multiple bidirectional initiations and
terminations of transcriptions of transcription in the Marek’s disease virus long
repeat regions. (1991). J. Virol. 65, 2445-2451.

Chen, X., and Velicer, LE. (1992). Identification of a unique Marek’s disease virus
gene which encodes a 38-Kilodalton phosphoprotein and is expressed in both
lytically infected cells and latently infected lymphoblastoid tumor cells. J. Virol.
66, 85-94.

Chittenden, T., Lupton, 5., and Levine, A. (1989). Functional limits of oriP, the
Epstein-Barr virus plasmid origin of replication. J. Virol. 63, 3016-3025.

Chomczynski, P., and Sacchi, N. (1987). Single-step method of RNA isolation by
acid guanidinium thiocyanate-phenol-chloroform extraction. Analytical Biochem.
162, 156-159.

Chou, J., and Roizman, B. (1985). Isomerization of herpes simplex virus type 1
genome: identification of the cis-acting and recombination sites within the domain
of the a sequence. Cell 41, 803-811. ‘

Churchill, A.E., and Biggs, RM. (1967). Agent of Marek’s disease in tissue culture.
Nature 215, 528-530.

Churchill, A.E., Chubb, RC, and Baxendale W. (1968). The attenuation, with loss
of oncogenicity, of the herpes-type virus of Marek’s disease(strain HPRS-16) on
passage in cell culture. J. Gen. Virol. 4, 557-564.

Cui, A., Lee, L.F., Liu, J.L., and Kung, H]. (1991). Structural analysis and
transcriptional mapping of the Marek’s disease virus encoding pp38, an antigen
associated with transformed cells. J. Virol. 65, 6509-6515.

Cuifo, D.M. and Hayward, GS. (1981). Tandem repeat defective DNA from the
L segment of the HSV genome. p. 107-128. in Y. Becker (ed). Herpesvirus DNA.
Martinus N ijhoff Publishers. The Hague, The Netherlands.

Dambaugh, T., Heller, M., ZRaab-Traub, N ., King, W., Cheung, A., Bei

sel, C., Hummel, M., van Santen, V., Fennewald, S., and Kieff, E. (1980). DNAs of
Epstein-Barr virus and herpesvirus papio, p. 85-90. In A. N ahrnias,W. Dondle, and
R. Schinazi (ed.), The human herpesvirus. Elsevier, New York.

Davison, A.J. (1984). Structure of the genome termini of varicella-zoaster virus. J.
Gen. Virol. 65, 1969-1977.

Davison, A.J., and Wilkie, N .M. (1981). Nucleotide sequences of the joint between
the L and S segments of Herpes simplex virus type 1 and 2.]. Gen. Virol. 55, 315-

140
331.

Deb, S., and Deb, SP. (1989). Analysis of oriS sequence of HSV-1: identification
of one functional DNA binding domain. Nucleic Acids, Res. 17, 2733-2752

Deb, S., and Deb, SP. (1991). a 269-amino-acid segment with a pseudo-leucine
zipper and a helix-turn-helix motif codes for the sequence-specific DNA-binding
domain of herpes simplex virus type 1 origin-binding protein. J. Virol. 65, 2829-
2838.

Deb, S., and Doelberg, M. 1988. A 67 base-pair segment from the ori-s region of
herpes simplex virus type 1 encodes origin function. J. Virol. 62, 2516-2519.

Deiss, L.P., Chou, J., and Frenkel, N. (1986). Functional domains within the a
sequence involved in the cleavage-packaging of herpes simplex virus DNA. J.
Virol. 59, 605-618.

Deiss, L.P., and Frenkel, N. (1986). Herpes simplex virus amplicon: cleavage of
concatemeric DNA is linked to packaging and involves amplification of the-
terminally reiterated a sequence. J. Virol. 57, 933-941.

Delecluse, H.J., and Hammerschmidt, W. (1993). Status of Marek’s disease virus
in established lymphoma cell lines: Herpesvirus integration is common. J. Virol.
67, 82-92.

Devereus, J., Haeberli, P. and Srnithies, O. (1984). A comprehensive set of
sequence analysis programs for the VAX. Nucleic Acids Research 12, 387-395.

Dutch, R.E., Bruckner, R.C., Mocarski, ES, and Lehman, LR. (1992). Herpes
simplex virus type 1 recombination: Role of DNA replication
and Viral a sequences. J. Virol. 66, 277-285.

Elias, P., Gustafsson, C.M., Hammarsten, O. (1990). The origin binding protein of
herpes simplex virus 1 binds cooperatively to the viral origin of replication oris.
J. Biol. Chem. 265, 17167-17173.

Elias, P., and Lehman, LR. (1988). Interaction of origin binding protein with an
origin of replicatin of herpes simplex virus 1. Proc. Natl. Acad. SCi. USA. 85,
2959-2963.

Fleckenstein, B., Bornkarnrn, G.W., and Ludwig, H. (1975). Repetitive sequences
in complete and defective genomes of Herpesvirus saimiri. J. Virol. 15, 398-406.

Frenkel, N ., Jacob, R.W., Honess, R.W., Hayward, G.S., Locker, H. and Roizman,
B. (1975). Anatomy of herpes simplex virus DNA. III. Characterization of defective

141

DNA molecules and biological properties of virus populations containing them.
J. Virol. 16, 153-167.

Frenkel, N ., Locker, H., Batterson, W., Hayward, GS, and Roizman, B. (1976).
Anatomy of herpes simplex virus DNA. VI. Defective DNA originates from the
5 component. J. Virol. 20, 527-531.

Frenkel, N. (1981). Defective interfering herpesviruses, p. 91-120. I_n A. J. N ahmias,
W.R. Dowdle, and RF. Schinazi (ed.), The human herpesviruses, Elsevier, New
York.

Fukuchi, K., Tanaka, A., Schierrnan, L.W., Witter, R.L., and N onoyama, M. (1985).
The structure of Marek’s disease virus DNA: the presence of unique expansion
in nonpathogenic viral DNA. Proc. Nat. Acad. Sci. USA 82, 751-754.

Fynan, E., Block, T.M., DuHadaway, J., Olson, W., and Ewert, D.L. (1992).
Persistence of Marek’s disease virus in a subpopulation of B cells that is
transformed by Avian Leukosis virus, but not in normal bursal B cells. J. Virol.
66, 5860-5866. '

Gahn, T.A., and Schildkraut, C.L. (1989). The Epstein-Barr virus origin of plasmid
replication, orip, contains both the initiation and termination sites of DNA
replication. Cell 58, 527-535.

Gaudray, P., Tyndall, C., Kamen, R., and Cuzin, F. (1981). The high affinity
binding site on polyoma virus DNA for the viral large-T protein. Nucleic Acids
Res. 9, 5697-5710.

Geller, AI. and Breakefield, X0. (1988). A defective HSV-1 vector expresses
Escherichia coli beta-galactosidase in cultured peripheral neurons. Science 241,
1667-1669.

Gibbs, C.P., N azerian, K., Velicer, L.P., and Kung, H]. (1984). Extensive homology
exists between Marek’s disease herpesvirus and its vaccine virus, herpesvirus of
tyrkeys. Proc. Natl. Acad. Sci. USA 81, 3365-3369.

Glaubiger, C., N azarian, K., and Velicer, L. F. (1983). Marek’s disease herpesvirus.
IV. Molecular characterization of Marek’s disease herpesvirus A antigen. J. Virol.
45, 1228-1234.

Hamdy, F., Sevoian, M., and Holt, SC. (1974). Biogenesis of Marek’s disease
(Type-II leukosis) virus in vitro-electron microscopy and immunological study.
Infecc. Immun. 9, 740-749.

Hammerschmidt, W., an d Sugden, B. (1988). Identification and characterization

142
of oriLyt, a lytic origin of DNA replication of Epstein-Barr virus. Cell 55, 427-433.

Hamzeh, F.M., Lietrnan, P.S., Gibson, W., and Hayward, GS. (1990). Identification
of the lytic origin of DNA replication in human cytomegalovirus by a novel
approach utilizing ganciclovir-induced chain termination. J. Virol. 64, 6184-6195.

Hayward, G.S., Jacob, R.J., Wadsworth, SC, and Roizman, B. (1975). Anatomy of
herpes simplex virus DNA: Evidence for four populations of molecules that differ
in the relative orientations of their long and short components. Proc. Natl. Acad.
Sci. USA 72, 4243-4247

Henikoff, S. (1987). Unidirectional digestion with exonuclease III in DNA
sequencing analysis. Methods enzymol. 155, 156-159.

Hennessy, K., and Kieff, E. (1983). One of two Epstein-Barr virus nuclear antigens
contains a glycine-alanine copolymer domain. Proc. Natl. Acad. Sci. USA 80, 5665-

5669.

Hennessy, K, Fennewald, S., Hummel, M., Cole, T., and Kieff, E. (1984). A“
membrane protein encoded by Epstein-Barr virus in latent growth-transforming
infection. Proc. Natl. Acad. Sci. USA 81, 7202-7211.

Hennessy, K., Fennewald, S., and Kieff, E. (1985). A third nuclear protein in
lymphoblasts immortalized by Epstein-Barr virus. Proc. Natl. Acad. Sci. USA 82,
5944-5948.

Hennessy, K., and Kieff, E. (1985). A second nuclear protein is encoded by
Epstein-Barr virus latent infection. Science 227, 1239-1240.

Hernandez, T.R., Dutch, R.E., Lehman, I.R., Gustafsson, C., and Elias, P. (1991).
Mutations in a herpes simplex virus type 1 origin that inhibit interaction with
origin-binding protein also inhibit DNA replication. J. Virol. 65, 1649-1652.
Hirai, K., Ikuta, K., Kitamoto, N ., and Kato, S. (1981). J. Gen. Virol. 53, 133-143.
Hirai, K. (1988). Molecular biology of Marek’s disease virus, p. 21-42. In S. Kato,
T. Horiuchi, T. Mikami, and K. Hirai (ed.), Advances in Marek’s disease research,
Osaka, Japan.

Hirai, K., Ikuta, K., and Kato, S. (1979). Comparative studies on Marek’s disease
virus and herpesvirus of turkey DNAs. J. Gen. Virol. 45, 119-131.

Huang, AS. (1973). Defective interfering viruses. Ann. Rev. Microbiol. 27, 101-117.

Hubenthal Voss, J., and Roizman, B. (1988). Properties of two 5’-coterrninal RNAs

143

transcribed part way and across the 5 component origin of DNA synthesis of the
herpes simplex virus I genome. Proc. Natl. Acad. Sci. USA 85, 8454-8458.

Igarashi,T., Takahashi, M., Donovan, J., Jessip, J., Smith, M., Hirai, K., Tanaka, A.,
and N onoyama, M. (1987). Restriction enzyme map of herpesvirus of tyrkey DNA
and its colinear relationship with Marek’s disease virus DNA. Virology 157, 351-
358.

Ikuta, K., Nakajima, K., Naito, M., Ann, S.H., Ueda, S., Kato, S., and Hirai, K
(1985). Identification of Marek’s disease lymphoblastoid cell lines using
monoclonal antibody against virus-specific phosphorylated polypeptides. Int. J.
Cancer 35, 257-264.

Jacob, R.J., and Roizman, B. (1977). Anatomy of herpes simplex virus DNA. VIII.
Properties of the replicating DNA J. Virol. 23, 394-411.

Jacob, R.J., Morse, L.S., and Roizman, B. (1979). Anatomy of herpes simplex virus
DNA. XII. Accumulation of head-to-tail concatemers in nuclei of infected cells and
their role in the generation of the four isomeric arrangement of viral DNA. J."
Virol. 29, 448-457.

Jalanko, A., Kallio, A., Ruohonen-Lehto, M., Soderlund, H. and Ulmanen, I. (1988).
An EBV-based mammalian cell expression vector for efficient expression of cloned
coding sequences. Biochim. Biophys. Acta 949, 206-212.

Jalanko, A., Kallio, A., Salminen, M. and Ulmanen, I. (1989). Efficient synthesis of
influenza virus hemagglutinin in mammalian cells with an extrachromosomal
Epstein-Barr virus vector. Gene 78, 287-296.

Jean, J.H., Blankenship, ML, and Ben-Porat, T. (1977). Replication of herpesvirus
DNA. 1. Electron microscopic analysis of replicative structures. Virology 79, 281-
291.

Kaschka-Dierich, C., Bornkamm, G.W., Thomssen, R. (1979). No homology
detected between Marek’s disease virus (MDV) DNA and herpesvirus of turkey
(HVT) DNA. Med. Microbiol. Immunol. 165, 223-239

Kato, S., and Hirai, K. (1985). Marek’s disease virus. Adv. Virus Res. 30, 225-277.
Kemble, G.W., and Mocarski, ES. (1989). A host cell protein binds to a highly

conserved sequence element (pac-2) within the cytomegalovirus a sequence. J.
Virol. 63, 4715-4728. ‘

144

Kishi, M., Bradley, G., Jessip, J., Tanaka, A., and N onoyama, M. (1991). Inverted
repeat regions of Marek’s disease virus DNA possess a structure similar to the a
sequence of herpes simplex virus DNA and contain host cell telomere sequences.
J. Virol. 65, 2791-2797.

Kishi, M., Harada, H., Takaahashi, M., Tanaka, A., Hayashi, M., N onoyama, M.,
Josephs, S.F., Buchbinder, A., Schachter, F., Ablashi, D.V., Wong-Staal, F.,
Salahuddin, 8.1, and Gallo, RC. (1988). A repeat sequence, GGG'I'I‘A, is shared
by DNA of human herpesvirus 6 and Marek’s disease virus. J.Virol. 33, 912-914.

Koff, A., and Tegtmeyer, P. (1988). Characterization of major recognition
sequences for a herpes simplex virus type 1 origin binding protein. J. Virol. 62,
4096-4103.

Kwong, A.D., and Frenkel, N. (1984). Herpes simplex virus amplicons: Effect of
size on replication of constructed defective genomes containing eucaryotic DNA
sequences. J. Virol. 51, 595-603.

Lee, L.P., Kieff, E.D., Bachenheimer, S.L., Roizman, B., Spear, P.G., Burmester,
B.R., and Nazerian, K. (1971). Size and compostion of Marek’s disease virus
deoxyribonucleic acid. J. Virol. 7, 289-294.

Lee, L.F., Nazerian, K., Leinbach, 5.5., Reno, J.M., and Boeizi, J.A. (1976). Effect
of phosphonoacetate on Marek’s disease virus replication. J. Natl. Cancer Inst. 56,
823-827.

Lockshon, D., and Galloway, DA. (1988). Sequence and structural requirements
of a herpes simplex viral DNA replication origin. Mol. Cell. Biol. 8, 4018-4027.

Longnecker, R, and Roizman, B. (1986). Generation of an inverting herpes
simplex virus 1 mutant lacking the L-S junction a sequences, an origin of DNA
synthesis, and several genes including those specifying glycoprotein E and the
alpha-47 gene. J. Virol. 58, 583-591.

Lupton, 5., Levine, AS. (1985). Mapping genetic elements of Epstein-Barr virus
that facilitate extrachromosomal persistence of Epstein-Barr virus-derived
plasmids in human cells. Mol. Cell. Biol. 5, 2533-2542.

Malkinson, M., Davidson, 1., Strenger, C., Weisman, Y., Maray, T., Levy, H., and
Becker, Y. (1989). Kinetics of the appearance of Marek’s disease virus DNA and
antigens in the feathers of chickens. Avian Pathology 18, 735-744.

Maniatis, T., Fritsch, BF and Sambrook, J. (1989). Molecular cloning: A laboratory
manual. Cold Spring Harbor laboratory, Cold Spring Harbor, New York.

145

Marek, J. (1907). Multiple nervenentzundung (polyneuritis) bei Huhnem. Dtsch.
tierarztl. Aschr. 15, 417-421.

Margolskee, R.F., Kavathas, P. and Berg, P. (1988). Epstein-Barr virus shuttle
vector for stable episomal replication of cDN A expression libraries in human cells.

Mol. Cell. Biol. 8, 2837-2847.

Marks, JR, and Spector, DH. (1988). Replication of the murine cytomegalovirus
genome: structure and role of the termini in the generation and cleavage of
concatenates. Virology 162, 98-107.

Martin, D.W., Deb, S.P., Klauer, J.S., and Deb, S. (1991). Analysis of the herpes
simpelx virus type 1 oris sequence: Mapping of functional domains. J. virol. 65,
4359-4369.

Martin, S.L., Aparisio, DI, and Bandyopahyay, PK. (1989). Genetic and
biochemical characterization of the thyrnidine kinase gene from herpesvirus of
turkeys. J. Virol. 63, 2847-2852.

Matsuo, T., Heller, M., Petti, L., O’Shiro, E., and Kieff, E. (1984). Persistence of the
entire Epstein-Barr virus genome integrated into human lymphocyte DNA.
Science 226, 1322-1325.

Matz, B. (1987). Herpes simplex virus infection generates large tandemly
reiterated simian virus 40 DNA molecules in a tranformed hamster cell line. J.
Virol. 61,1427-1434.

McGeoch, D.J., Dalrymple, M.A., Dolan, A., McN ab, D., Perry, L.J., Taylor, R,
Challberg, MD. (1988). Structures of herpes simplex virus type 1 genes required
for replication of virus DNA. J. Virol. 62, 444-453.

Mocarski, E.S., Deiss, L.P., and Frenkel, N. (1985). Nucletide sequence and
structural features of a novel Us-a junction present in a defective herpes simplex
virus genome. J. Virol. 55, 140-146.

Mocarski, E.S., and Roizman, B. (1981). Site-specific inversion sequence of the
herpes simplex virus genome: Domain and structural features. Proc. Natl. Acad.
Sci. USA 78, 7047-7051.

Mocarski, E.S., and Roizman, B. (1982). Herpesvirus dependent amplification and
inversion of cell-associated viral thmidine kinase gene flanked by viral sequences
and linked to an origin of replication. Proc. Natl. Acad. Sci. USA 79, 5626-5630.

Morgan, R.W., Cantello, J.L., and McDermott, CH. (1990). Transfection of chicken
embryo fibroblasts with Marek’s disease virus DNA. Avian Dis. 34, 345-351.

146

Morgan, R.W., Cantello, J.L., Claessens, J.A.J., and Sondermeyer, P. (1991).
Inhibition of Marek’s disease virus DNA transfection by a sequence containing an
alphaherpesvirus origin of replication and flanking transcription regulatory
elements. Avian Dis. 35, 70-81.

Murchie, M.-J., and McGeoch, DJ. (1982). DNA sequence analysis of an
immediate-early gene region of the herpes simplex virus type 1 genome (map
coordinates 0.95 to 0.978). J. Gen. Virol. 62, 1-15.

Murray, B.K, Biswall, N ., Bookout, J.B., Lanford, RE, Courtney, R]. and Melnick,
J.L. (1975). Cyclic appearance of defective interfering particles of herpes simplex
virus and the concomitant accumulation of early polypeptide VP 175.
Intervirology 5, 173-184.

N azerian, K 1973. Oncogenesis of Marek’s disease. Cancer Research 33, 1427-1430.

Nazerian, K (1970). Attenuation of Marek’s disease virus and study of its
properties in two different cell cultures. J. Nat. Cancer Inst. 44, 1257-1267.

N azerian, K and Burmester, BR (1968). Electron microscopy of a herpes virus
associated with the agent of Marek’s disease in cell culture. Cancer Research 28,
2454-2462.

N azerian, K., Lee, L.F., Witter, RL., and Burmester, BR. (1971). Ultrastructural
studies of a herpesvirus of turkeys antigenically related to Marek’s disease virus.
Virology 43, 442-452.

O’Callaghan, D.J., Allen, GP. and Randall, CC. (1978). The equine herpesviruses.
In "Equine infectious diseases: (J.T. Bryan and H. Gerber, eds.). Vol. 4, p 1-32.
Veterinary Publications, Princeton, N .J.

Oliva, R., Goren, R., and DIxon, CH. (1989). Quail (Coturnix japonica) protamine,
full-length cDN A sequence, and the function and evolution of vertebrate
protamines. J. Biol. Chem. 264, 17627-17630.

Oliva, R., Mezquita, J., Mezquita, C., and Dixon, CH. (1988). Haploid expression
of the rooster protamine mRN A in the postmeiotic stages of spermatogenesis.
Dev. Biol. 125, 332-340.

Olivo, P.D., Nelson, N .J., and Challberg, MD. (1989). Herpes simplex virus type
1 gene products required for DNA replication: Identification and overexpression.
J. Virol. 63, 196-204.

Ono, M., Katsuragi-Iwanaga, R., Kitazawa, T., Kamiya, N ., Horirnoto, T., N iikura,
M, Kai, C., Hirai, K and Mikami, T. (1992). The restriction endonuclease map of

147

Marek’s disease virus (MDV) seritype 2 and collinear relationship among three
serotypes of MDV. Virology 191, 459-463.

Payne, L.N. (1985). Pathology, p. 43-75. In Marek’s disease, Payne, L.N. (ed.).,
Martinus Jijhoff Publishing, Boston.

Payne, L.N. (1988). Pathogenesis of Marek’s disease virus, p. 307-316. In Advences
in Marek’s disease virus research, S. Kato, T. Horiuchi, T. Mikami, K Hirai (ed.),
Japanese association on Marek’s disease, Osaka, Japan.

Payne, L.N., and Biggs, RM. (1967). Studies on Marek’s disease. II. Pathogenesis.
J. Nat. Cancer. Inst. 39, 281-302.

Petti, L., Sample, C., and Kieff, E. (1990). Subnuclear localization and
phosphorylation of Epstein-Barr virus latent infection nuclear proteins. Virology
176, 563-574.

Poffenberger, KL, and Roizman, B. (1985). A noninverting genome of a viable
herpes simplex virus 1: presence of head-to-tail lingages in packaged genomes
and requirements for circularization after infection. J. Virol. 53, 587-595.

Pogue-Geile, KL., Lee, G.T.Y., and Spear, PG. (1985). Novel rearrangements of
herpes simplex virus DNA sequences resulting from duplication of a sequence
within the unique region of the L component. J. Virol. 53, 456-461.

Polvino-Bodnar, M., Orberg, P.K., and Shaffer, RA (1987). Herpes simplex virus
type 1 oriL is not required for virus replication or for the establishment and
reactivation of latent herpesvirus in mice. J. Virol. 61, 3528-3535.

Preston, C.M., Cordingley, M.G., and Stow, N .D. (1984). Analysis of DNA
sequences which regulate the transcription of a herpes simplex virus immediate-
early gene. J. Virol. 50, 708-716.

Preston, C.M., and Tannahill, D. (1984). Effects of orientation and position on the
activity of a herpes simplex virus immediate early gene far-upstream region.
Virology 137, 439-444.

Preston, C.M., Frame, MC, and Campbell, EM. (1988). A complex formed
between cell components and an HSV structural polypeptide binds to a viral
immediate early gene regulatory DNA sequence. Cell 52, 425-434.

Pulaski, J.T., Tieber, V.L., and Coussens, RM. (1992). Marek’s disease virus-
mediated enhancement of Avian leukosis virus gene expression and virus
production. Virology 186, 113-121.

148

Purchase, H.G. (1974). Marek’s disease virus and the herpesvirus of turkeys.
Progr. Med. Virol. 18, 178-197.

Rabkin, S.D., and Hanlon, B. (1990). Herpes simplex virus DNA synthesis at a
preformed relication fork in vitro. J. Virol. 64, 4957-4967.

Rawlins, D., Milman, G., Hayward, 5., and Hayward, G. (1985). Sequence-specific
DNA binding of the Epstein-Barr virus nuclear antigen EBN A-1 to clustered sites
in the plasmid maintenance region. Cell 42, 859-868.

Reilly, J.D., and Silva, RF. (1993). The number of copies of an a-like region in the
serotype 3 Marek’s disease virus DNA genome is variable. Virology (In press).

Reisman, D., Yates, J., and Sugden, B. (1985). A putative origin of replication of
plasmids derived from Epstein-Barr virus is composed of two cis-acting
components. Mol. Cell. Biol. 5, 1822-1832.

Rixon, F., Feldman, LT. and Ben-Porat, T. (1980). Expression of the genome of
defective intefering pseudorabies virions in the presence or absence of helper
functions provided be standard virus. J. Gen. Virol. 46, 119-138.

Roizman B. 1982. The family herpesviridae: general description, taxonomy, and
classification, p. 347-431. In B. Roizman (ed.), The herpesviruses vol 1. Plenum
press, New York.

Roizman, B., and Balterson, W. (1985). Herpesviruses and their replication, p.497-
526. Q B.N.F"1elds, D.M.Knipe, R.M.Chanock, J.L.Melnick, B. Roizman, and RE.
Shope (ed.), Virology. Raven press, New york.

Roizman, B. (1990). Herpesviridae: a brief introduction, p. 1787-1793. In B. N.
Fields, and D. M. Knipe (ed.), Fields Virology. Raven press, New York.

Roizman, B. (1992). The family Herpesviridae: an update. Arch. Virol. 123, 425-449.
Sample, J., Hummel, M., Braun, D., Birkenbach, M., and Kieff, E. (1986).
Nucleotide sequences of mRNAs encoding Epstein-Barr virus nuclear proteins: A
probable transcription initiation site. Proc. Natl. Acad. Sci USA 83, 5096-5100.

Sanger, F., Nicklen, S., and Coulson, AR. (1977). DNA sequencing with chain-
terminating inhibitors. Proc. Natl. Acad. Sci. USA. 74, 5463-5467.

Sawtell, N .M., and Thompson, R.L. (1990). HSV-1 amplicons as gene transfer
vectors in Vivo. p. 334 Abstr. 15th International herpesvirus workshop.

Schat, K.A., and Calnek, B.W. (1978). Characterization of an apparently

149

nononcogenic Marek’s disease virus. J. Natl. Cancer Inst. 60, 1075-1082.

Schat, K.A., Calnek, B.W., Fabricant, J., and Graham, D.L. (1985). Pathogenesis of
infection with attenuated Marek’s disease virus strains. Avian Pathology 14, 127-
146.

Schroder, C.H., Stegmann, B., Lauppe, H.F. and Kaerner, HQ (1975/ 76). An
unusual defective genotype derived from herpes simplex virus strain AN G.
Intervirology 6, 270-284.

Sevoian, M., Chamberlain, M.D, Larose, RN. (1963). Avian Lymphomatosis. V.
Air-born transmission. Avian. Dis. 7, 102-105.

Sheldrick, P., and Berthelot, N. (1975). Inverted repetitions in the chromosome of
herpes simplex virus. Cold Spring Harbor Symp. Quant. Biol. 39, 667-678.

Silva, R.E., and Barnett, J.C. (1991). Restriction endonuclease analysis of Marek’s
disease virus DNA: Differenciation of viral strains and determination of passage
history. Avian Dis. 35, 487-495. .

Silva, RE, and Lee, LE. (1984). Monoclonal antibody-mediated
immunoprecipitation of proteins from cells infected with Marek’s disease virus
or turkey herpesvirus. Virology. 136, 307-320.

Silva, R.F., Carter, J., and Witter, R.L. (1988). Construction of a shuttle vector and
its utilization as a vehicle to insert foreign DNA into the Marek’s disease virus
genome, p. 50-55. In S. Kato, T. Horiuchi, T. Mikami, and K. Hirai (ed.), Advances
in Marek’s disease research, Osaka, Japan.

Silva, R. F., Witter, R. L. (1992). A non-defective recombinant Marek’s disease
virus with two copies of inserted foreign DNA. In "Proceedings 19th World’s
Poultry Congress" Vol. 1, pp.140-143. Ponsen 8: Looijen, Wageningen,
Netherlands.

Smiley, J.R., Fong, 8.5., and Leung, WC. (1981). Construction of a double-jointed
herpes simplex virus DNA molecules: inverted repeats are required for segment
inversion and direct repeats promote deletion. Virology 113, 345-362.

Smiley, J.R., Duncan, J., and Howes, M. (1990). Sequence requirements for DNA
rearrangement induced by the terminal repeat of herpes simplex virus type 1 K05
DNA. J. Virol. 64, 5036-5050.

Smiley, J.R., Lavery, C., and Howes M. (1992). The herpes simplex virus type
1(HSV-1)a sequence serves as a cleavage/ packaging signal but does not drive
recombinational genome isomerization when it is inserted into the HSV-2 genome.

150
J. Virol. 66, 7505-7510.

Spaete, R.R., and Frenkel, N. (1982). The herpes simplex virus amplicon: a new
euccccaryotic defective-virus cloning-amplifying vector. Cell 30, 295-304.

Spaete, R.R., and Frenkel, N. (1985). The herpes simplex virus amplicon: Analysis
of gi_s-acting replication functions. Proc. Natl. Acad. Sci. USA. 82, 694-698.

Stegmann, B., Zentgraf, H., Ott, A. and Schroder, CH. (1978). Synthesis and
packaging of herpes simplex virus DNA in the course of virus passages at high
multiplicites. Intervirology 10, 228-240.

Stinski, M.F., Mocarski, E.S., and Thomsen, DR. (1979). DNA of human
cytomegalovirus: Size heterogeneity and defectiveness resulting from serial
undiluted passage. J. Virol. 31, 231-239.

Stow, N .D. (1982). Localization of an origin of DNA replication within the
TRs/IRs repeated region of the herpes simplex virus type 1 genome. EMBO J. 1,
863-867. ‘

Stow, N .D., and Davison, J. (1986). Identification of a varicella-zoster virus origin
of DNA replication and its activation by herpes simplex type 1 gene products. J.
Gen. Virol. 67, 1613-1623.

Stow, N .D., and McMonagle, EC. (1983). Characterization of the TRs/IRs origin
of DNA replication of herpes simplex type 1. Virology. 130, 427-438.

Tamashiro, J.C., and Spector, DH. (1986). Terminal structure and heterogeneity
in human cytomegalovirus strain Ad169. J. Virol. 59, 591-604.

Tamashiro, J.C., Filpula, D., Friedmann, T., and Spector, DH. (1984). Structure of
the heterogeneous L-S junction region of human cytomegalovirus strain AD169.
J.Virol. 52, 541-548.

Tanaka, A., Silver, S., and Nonoyama, M. (1978). Virology 88, 19-24.

Tieber, V.L., Zalinskis, L.L., Silva, R.F., Finkelstein, A., and Coussens, RM. 1990.
Transactivation of the Rous sarcoma virus long terminal repeat promoter by
Marek’s disease virus. Virology 179, 719-727.

Umene, K (1991). Recombination of the internal direct repeat element DR2
responsible for the fluidity of the _a_ sequence of Herpes simplex virus type 1. J.
Virol. 65, 5410-5416.

151

van Saten, V., Cheung, A., and Kieff, E. (1981). Epstein-Barr virus RNA. VII. Size
and direction of transcriptionof virus specified cytoplasmic RNA in a cell line
transformed by EBV. Proc. Natl. Acad. Sci. USA 78, 1930-1934.

Varmuza, S. L., and Smiley, J. R. (1985). Signals for site-specific cleavage of HSV
DNA: maturation involves two separate cleavage events at sites distal to the
recognition sequences. Cell 41, 793-802.

Velicer, LE, and Brunovskis, P. (1992). Genetic organization of herpesviruses in
the Marek’s disease system, p. 33-30. In Proceedings of World’s poultry congress,
Vol. 1, Amsterdam, the Netherlands.

Vlazny, D.A., and Frenkel, N. (1981). Replication of herpes simplex virus DNA:
localization of replication recognition signals within defective virus genomes.
Proc. Natl. Acad. Sci. USA. 78, 742-746.

Vlazny, D.A., Kwong, A., and Frenkel, N. (1982). Site-specific cleavage/ packaging
of herpes simplex virus DNA and the selective maturation of nucleocapsids
containing full-length viral DNA. Proc. Natl. Acad. Sci. USA 79, 1423-1427.

Von Magnus, P. (1951). Propagation of the PR8 strain of influenza virus in chick
embryos. Ill. Properties of the incomplete virus produced in serial passages of
undiluted virus. Acta pathol. Microbiol. Scand. 29, 156-181.

Wadsworth, S. R., Jacob, R. J., and Roizman, B. (1975). Anatomy of herpes simplex
virus DNA. II. Size, compositio and arrangement of inverted terminal repetitions.
J. Virol. 15, 1487-1497.

Weber, P.C., Challberg, M.D., Nelson, N J., Levine, M., and Glorioso, J.C. (1988).
Inversion events in the HSV-1 genome are directly mediated by the viral DNA
replication machinery and lack sequence specificity. Cell 54, 369-381.

Weir, H.M., and Stow, N .D. (1990). Two binding sites for the herpes simplex type
1 OBP protein are required for efficient activity of the oris replication function. J.
Gen. Virol. 71, 1379-1385.

Weller, S.K., Spadaro, A., Schaffer, J.E., Murray, A.W., Maxam, A.M., and Shaffer,
RA. (1985). Cloning, sequencing, and functional analysis of ori-L, a herpes
simplex virus type 1 origin of DNA synthesis. Mol. Cell. Biol. 5, 930-942.

Wen, L-T., Lai, P.K, Bradley, G., and Tanaka, A., and Nonoyama, M. (1990).
Interaction of Epstein-Barr viral (EBV) origin of replication (oriP) with EBNA-l
and cellular anti-EBNA-1 proteins. Virology 178, 293-296.

152

Wilson, MR, and Coussens, RM. (1991). Purification and characterization of
infectious Marek’s disease virus genomes using pulsed field electrophoresis.
Virology 185, 673-680.

Wilson, M.R., Tieber, V. L., Mehigh, C. 5., Pulaski, J. T., Southwick, R. A. and
Coussens, P. M. (1992). Molecular basis for reduced expression of glycoprotein C
(GP57-65) in attenuated Marek’s disease virus. I_n "Proceedings 19th World’s
poultry congress", Vol. 1. pp49-53. Ponsen 8: Looijen, Wageningen Netherlands.

Witter, R. L. (1983). Characteristics of Marek’s disease viruses isolated from
vaccinated commercial chicken flocks: association of viral pathotype with
lymphoma frequency. Avian Dis. 27, 113-132.

Witter, R.L. (1985). Principles of vaccination, p.203-250. In Marek’s disease,
developments in veterinary virology, Payne, L.N. (ed.)., Martinus Nijhoff
Publishing, Boston / Dordrecht

Witter, R. L. (1987). New serotype 2 and attenuated serotype 1 Marek’s disease
vaccine viruses: comparative efficacy. Avian Dis. 31, 752-765.

Witter, R.L. (1991). Attenuated revertant serotype 1 Marek’s disease viruses: safety
and protective efficacy. Avian Dis. 35, 877-891.

Witter, R.L., Lee, LE, and Sharma, J.M. (1990). Biological diversity among
serotype 2 Marek’s disease viruses. Avian Dis. 34, 944-957.

Witter, R.L., Silva, RF, and Lee, LP. (1987). New serotype 2 and attenuated
serotype 1 Marek’s disease vaccine viruses: selected biological and molecular
characteristics. Avian Dis. 31, 829-840.

Wong, 5. W., and Schaffer, P. A (1991). Elements in the transriptional regulatory
region flanking herpes simplex virus type 1 oris stimulate origin function. J. Virol.
65, 2601-2611.

Wu, C.A., Harpe, L., and Ben-Porat, T. (1986). Qié functions involved in
replication and cleavage-encapsidation of pseudorabies virus. J. Virol. 59, 318-327.

Wu, C.A., Nelson, N .J., McGeoch, D.J., Challberg, MD. (1988). Identification of
herpes simplex virus type 1 genes required for origin-dependent DNA synthesis.
J. Virol. 62, 435-443.

Yates, J.L., and Camiolo, SM. (1988). Dissection of DNA replication and enhancer
activation functions of the Epstein-Barr virus nuclear antigen 1. In Cancer cells,
Vol. 6, B. Stillman and T. Kelly, (eds.), 6, 197-205 Cold Spring Harbor Laboratory,
Cold Spring Harbor, New york.

153

Yates, J.L., Warren, N ., Reisman, D., and Sugden, B. (1984). A Q's -acting element
from the Epstein-Barr viral genome that permits stable replication of recombinant
plasmids in latently infected cells. Proc. Natl. Acad. Sci. USA. 81, 3806-3810.

Yates, J.L., Warren, N ., Sudgen, B. (1985). Stable replication of plasmids derived
from Epstein-Barr virus in various mammalian cells. Nature 313, 812-815.

zur Hansen, H. (1981). Oncogenic herpesviruses, p. 747-795. _I_n_ J. Tooze (ed.),
DNA tumor viruses. Cold Spring Harbor Laboratory, Cold Spring Harbor, N .Y.

 

"I'llll'lllllllllll