u.“
x». .

«m...

4

“z: ‘ "a
._: :4 ‘t-
n m!‘

3’? fires}:-

‘;a

‘

infig

'. .94: ~—
‘ :2'
r1

laia' 535

3-4.

5'
id:
I
,1
‘1 :
vé
g: ..

‘Iv
i
3:“
1

 

- I

:4.4
«.42‘

— *1

le

.3
")ar’) .1
A .. /

LIBRARY
Michigan State
University

This is to certify that the
dissertation entitled

THE ROLE OF INTERLEUKlN-8 LIKE GENE (VlL-8) IN THE
PATHOGENESIS OF MAREKS’S DISEASE VIRUS

presented by

Xiaoping Cui

has been accepted towards fulfillment
of the requirements for the

PhD. degree in Pathology

 

 

 

Major Professor’s Signature

laZ/g/OB
/ /

Date

MSU is an Affirmative Action/Equal Opportunity Institution

-.-.-.-.-.-.-.-.-.-.-._.-.-.-.-.-.-.-.-...

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

"08 29 ‘ ’5‘
“12272005

Whirligig

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/ClFiC/DateDue.p65—p. 15

 

THE ROLE OF INTERLEUKlN-8 LIKE GENE (vlL-8) IN THE
PATHOGENESIS OF MAREK’S DISEASE

By

Xiaoping Cui

A DISSERTATION

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

DOCTOR OF PHILOSOPHY
Department of Pathobiology and Diagnostic Investigation

2003

ABSTRACT

THE ROLE OF INTERLEUKlN-8 LIKE GENE (vlL-8) IN THE PATHOGENESIS
OF MAREK'S DISEASE

By

Xiaoping Cui

Marek’s disease (MD) is a lympho-proliferative disease of chickens caused
by an alpha-herpesvirus called Marek’s disease virus (MDV). MDV can interact
differently with different types of cells, resulting in productive infection, latent
infection or cell transformation. The molecular basis of pathogenesis of this virus
has not been clearly defined. The oncogenic serotype 1 MDV encodes a virokine
vlL-8, with general homology to cellular CXC chemokines such as lL-8 and Gro-a.
To study the function of the vlL-8 gene, we deleted both copies of the vlL-8 gene
residing in the TRL and IRL region of the viral genome and generated a vlL-8
deleted virus er5/AvlL-8. Growth kinetics studies showed that the vlL-8 gene is
dispensable for virus replication in cell culture. In vivo, vlL-8 gene is involved in
early cytolytic infections in lymphoid organs as indicated by less viral antigen
expression detected by immunohistochemical staining. The er5/AvlL-8 virus is
unimpaired in virus spreading. Similar viremia titers as the er5 virus at 6 and 8
dpi indicated that vlL-8 gene is not necessary for virus reactivation and is not

involved in latency. Nevertheless, deletion of the vIL-8 gene compromised

transformation of the virus with reduced number of transformed cells at 5 weeks
post inoculation and less gross tumors developed in vivo. The revertant virus
that restored the expression of vlL-8 gene showed the same phenotype as the

er5.

The role of vlL-8 in affecting the virulence and pathogenesis of MDV was
also elucidated by er5/AvlL-8. The er5/AvlL-8 virus not only had decreased
virulence but also showed the ability to protect against challenge with the very
virulent plus (w+) MDV strain 648A. Studies on cell tropism between the virus
er5/AvlL-8 and the parental virus er5 inoculated chickens were carried out
using flow cytometry assays (FACS). During the early cytolytic infection, there
was significantly more B cells in the er5/AvlL8 virus inoculated chickens than
the parental er5. Significant differences were also demonstrated in the virus

induced numbers of activated T cells.

One of the interesting sequential differences between the vlL-8 gene and
other cellular homologue is that vlL-8 encodes a “DKR” motif instead of “ELR”.
To study the role of the variation of this motif in MDV pathogenesis, we
generated recombinant MDV, er5/vlL—8-ELR, carrying “ELR” motif by
mutagenesis. Both in vitro and in vivo studies showed that er5/vlL-8-ELR has

the same phenotype as er5.

DEDICATION

To my family, my advisors, friends ...........

ACKNOWLEDGEMENTS

I wish to express my gratitude to my mentors, Dr. Lucy F. Lee, Dr. Willie M.
Reed, and Dr. Sanjay M. Reddy, for providing me the opportunity to study at
Michigan State University and work at the Avian Disease and Oncology
Laboratory. This research work would not be possible without their unselfish
indoctrination of their countless knowledge, their patience and their

encouragement.

Profound appreciation goes to all my committee members Dr. Scott Fitzgerald,
Dr. Richard Fulton, Dr. Katheryn Meek for their stimulating discussions and

advice.

I also would like to thank every member of the Avian Disease and Oncology
Laboratory for their support and friendship. I deeply appreciate Dr. Richard
Witter for his generosity in providing me professional knowledge in research and
also his patience in discussing with me about the future and life in general.
Thanks to Drs. Blanca Lupiani, Isabel Gimeno, and Arun Pandiri for their critical
comments and suggestions on my research project. My gratitude also goes to
Mr. Barry Coulson, for his wonderful technique support in cell culture and
assistance in necropsy. Special thanks to Dr. Robert Silva’s lab and Mr. Lonnie
Milam for my initial trainings in molecular biology and offering experimental

instruments and reagents to conduct my experiments. I would like to

acknowledge Dr. Henry Hunt’s Lab and Terry Lemme, for their patient assistance
with my flow cytometry and migration studies. I also would like to thank Ms.
Barbara Riegle for preparing different strains of Marek’s disease virus for my
research. I am grateful for the help from the staffs working on the farm of Avian
Disease and Oncology Laboratory, with special thanks to Raj Kulkarni, for

making sure that each of my experiments went smoothly.

I would like to acknowledge Dr. Hsing-Jien Kung for his kindness of providing

reagents and valuable suggestions throughout my research studies.

I cannot have today without my family. I am grateful to my grandmother and
my parents in China for their unconditional love and encouragement in all of my
lifetime. I thank my sister and brother for their support in making me stronger
and more confident. Lastly, my deepest gratitude belongs to my beloved
husband, Yan Du, for his love, encouragement, incredible patience and most of

all for his faith in me which has helped me handle stress and any crisis.

vi

TABLE OF CONTAINS

LIST OF TABLES ................................................................. x
LIST OF FIGURES ................................................................ xii

LIST OF ABBREVIATIONS ........................................................ xiv
INTRODUCTION .................................................................. 1

CHAPERTER 1. LITERATURE REVIEW ........................................ 3

I. History of Herpesvirus .......................................................... 3

ll. Marek’s Disease and Marek’s Disease Virus (MDV) ..................... 5

1. History of the disease ......................................................... 5

2. Biological aspects of MD and MDV ........................................ 6

1) Classification and pathotyping ....................................... 6

a) Serotypes ......................................................... 7

b) Pathotypes ......................................................... 8

2) Morphology and ultra-structure ....................................... 8

3) Cultivation and propagation of MD virus .............................. 9

4) Host range ................................................................. 10

3. Pathology aspects of MD ............................................... 11

1) Clinical signs ........................................................ 11

2) Gross and histology lesions ...................................... 12

3) Differential diagnosis ............................................... 13

4) Pathogenesis ........................................................ 14

3) Early cytolytic, productive-restricted infection ........... 14

b) Latency ........................................................ 16

c) Productive infection in FFE ...................................... 16

d) The second phase of cytolytic infection .................... 17

e) Transformation and lymphoma ............................. 17

5) Virus genes associated with oncogenicity .................... 19

a) Phosphorylated protein 38 (pp38) ............................ 19

b) Meq gene ........................................................ 20

4. MDV Immunity ................................................................. 21

1) Cell-mediated immune response ...................................... 21

2) Humoral immune response ...................................... 22

3) lmmunosuppression ............................................... 23

4) Vaccinal immunity ....................................................... .24

vii

5) Types of vaccines and virulence trends ............................. 24

5. Molecular Biology of MDV ............................................... 26
1) Physical MDV map and genomic structure .................... 26

2) Cosmid and Bacterial Artificial Clone (BAC) systems for
recombinant MDV ........................................................ 28
Ill. Chemokine family and Virokines ................................................ 3O
1. Introduction to chemokines family ....................................... 31
2. Chemokine and virus ......................................................... 32
3. Virokines ........................................................................... 33
4. Chemokine and angiogenesis ................................................ 34
5. MDV encoded vlL-8 gene ................................................ 36

CHAPTER 2. MDV ENCODED VIL-8 GENE IS INVOLVED IN EARLY
CYTOLYTIC INFECTION AND TRANSFORMATION BUT DISPENSABLE

TO LATENCY .................................................................. 48
1 . Abstract ................................................................... 48
2. Introduction .................................................................. 5O
3. Material and Methods .......................................................... 53
4. Results .................................................................. 60

1) MDV vlL-8 is dispensable for in vitro replication, latency and in vivo
transmission but important for early in vivo cytolytic infection and

transformation ......................................................... 60
2) Generation and in vitro characterization of an ELR mutant virus,
er5/vlL-8-ELR. .......................................................... 63
3) Mutant vlL-8 protein carrying the ELR motif induces strong
angiogenic activity in CAM assay. ............................... 64
4) ELR mutations have no effect on the pathogenic properties of
er5/vlL-8-ELR. ........................................................... 65

5. Discussion ................................................................... 66

CHAPTER 3. AN ATTENUATED VIL-8 DELETION MAREK’S DISEASE VIRUS

(MDV) MUTANT CONFERS PROTECTION AGAINST CHALLENGE OF A
VERY VIURLENT STRAIN OF MDV

............................................................................ 84
1. Abstract ............................................................................ 84
2. Introduction ................................................................... 86
3. Material and Methods .......................................................... 88
4. Results .................................................................... 93
1) The recombinant mutant virus, er5/AvIL-8, has an attenuated
virulence. ................................................................... 93
2) Phenotype of the target cells for transformation is not affected by
vIL-8. ................................................................... 94

viii

3) Deletion of vlL-8 gene delayed MEQ expression and impaired pp38

expression in lymphoid organs. ....................................... 94

4) Change of lymphocytes proportion at early cytolytic infection by
er5 and er5/AvlL-8. ................................................ 95

5) Recombinant virus er5/AvlL-8 protected chickens against the
challenge of the very virulent virus strain ..................... 96
5. Discussion .................................................................. 98
CHAPTER 4. CONCLUSION AND FURTHER RESEARCH ..................... 113
LIST OF REFERENCES ........................................................ 120

LIST OF TABLES

Table 1 Classification of MDV strains by serotypes and prototypes ............ 38

Table 2 Differential diagnosis of MD and other chicken lymphoma diseases.

.................................................................................... 39
Table 3 Classification of oncogenic herpesviruses ............................... 44
Table 4 CXC chemokines, their target cells and angiogenesis activity ........... 45

Table 5 Herpesviruses encoded Chemokine or Chemokine receptor homologues
....................................................... 46

Table 6 Virus titer in peripheral blood lymphocytes of er5, er5/vIL-8-ELR and
er5/AvlL-8 inoculated chickens 35 days post inoculation.
..................................................................................... 79

Table 7 Comparison of the pathogenecity of er5, er5/AvIL-8, er5/vlL-8-ELR
and er5/AvlL-8-RV in 15x7 antibody negative chickens.
.................................................................................... 81

Table 8 Pathological lesions in MDV maternal antibody negative chickens
inoculated with er5, er5/AvIL-8, and er5/AvlL-8-RV viruses

Table 9 Pathological lesions in MDV maternal antibody positive chickens

inoculated with recombinant er5, er5/AvlL-8, and er5/AvIL-8-RV
viruses ........................................................................... 103

Table 10 In vivo expression kinetics of the viral protein pp38 and Meq
.. ............................................................................................... 107

Table 11 Comparative protective efficacy of vlL-8 deletion virus in MD maternal
antibody positive chickens 110

xi

LIST OF FIGURES

Figure 1 Possible sequential events in the pathogenesis of Marek’s disease.
.................................................................................... 40

Figure 2 The continued evolution of MDV towards greater virulence parelled with
the foregoing succession of MDV vaccines ............................... 41

Figure 3 Schematic map of MDV genome and the locations of genes Meq, pp38

and vlL-8. .................................................................. 42
Figure 4 Schematic structure and restriction map of the MDV genome. ....... 43
Figure 5 Structure comparison of lL-8. ................................................ 47
Figure 6 Construction of recombinant virus with deletion of vlL-8 gene. ........ 70

Figure 7 Generation of er5/AvlL—8 revertant virus by homologous recombination
..................................................................................... 71

Figure 8 lmmunofluorescence staining of parental and recombinant MD viruses
infected DEF using anti-pp38 and anti-vlL-8 antibodies ............. 72

Figure 9 Western blot analysis of supernatant from recombinant viruses.
.................................................................................... 73

Figure 10 Southern blot analysis of DNA isolated from nucleocapsids of wild
type and recombinant MDV viruses ........................................ 74

Figure 11 In vitro growth kinetics of er5, er5/vIL-8-ELR-1, er5/vlL-8-ELR-2,
er5/AvlL-8-1, and er5/AvIL-8-2 viruses ............................... 75

xii

Figure 12 lmmunohistochemistry of lymphoid organs ...................... 76

Figure 13 lmmunohistochemical analysis of Feather Follicle Epithelium (FFE)
cells from inoculated chickens ................................................ 78

Figure 14 Viral titers, at 6, 8, 14, and 35 days post-inoculations, in peripheral
blood lymphocytes of 15x7 chickens inoculated with er5, er5/vlL—8-
ELR and er5/AvlL-8. ......................................................... 80

Figure15 Mutation of DKR to ELR motif in the MDV vlL-8 gene. ................ 82

Figure 16 Angiogenic effect of supernatants from uninfected DEF, er5 and
er5/vIL-8-ELR infected DEF ................................................. 83

Figure 17 Histopathological lesions derived from infection with the recombinant
viruses. ................................................................ 104

Figure 18 lmmunohistochemical staining of tumor tissues derived from all three
recombinant viruses with Mab against chicken CD4 and CD8
.......................................................................... 106

Figure 19 lmmunohistchemical staining of spleens using antibody against MEQ.
......................................................................... 108

Figure 20 Flow cytometry analysis of spleen cells from virus infected chickens.
......................................................................... 109

Figure 21 The possible roles of vlL-8 in the sequential pathogenesis of MDV
infection. ................................................................ 1 1 1

xiii

INTRODUCTION

Marek’s disease (MD), a contagious lymphoma of chickens, is considered a
major disease problem in commercial poultry. Marek’s disease virus (MDV) is
classified as a member of the family Herpesviridae. Based on its genomic
organization and significant sequence similarity, it is classified into subfamily
Alphaherpesvirinae (Buckmaster et al. 1988; Roizman 1992a). The MDV as well
as the disease itself has been studied for several decades. Many MDV genes
have been characterized and the mechanisms of the pathogenesis of the disease
have been explored step by step. There are three serotypes of MDV of which,
only serotype 1 MDV strains are pathogenic and causing lymphomas. The most
specific difference in the genomic information among different serotypes is clearly

seen in the inverted repeat regions.

The vIL-8 is one of the unique genes located in this inverted repeat region.
The studies described in this dissertation were designed to characterize the
function of the virus encoded chemokine like gene, vIL-8, in the viral replication
and pathogenesis of MDV. Chapter 1 provides a review of information on the
involvement of MDV and chemokines in host immune responses. Different
mechanisms served by large DNA virus to defend against the host inflammatory
responses are also discussed. Chapter 2 describes a detailed analysis of
mutagenesis and pathogenesis of three recombinant MDVs, er5/vlL-8-ELR
which carries ELR (Glu-Leu-Arg) motif in the vlL-8 gene, vlL-8 knockout virus

er5/AvIL-8 and its revertant, er5/AvlL-8-RV. It is demonstrated in this chapter,

that MDV encoded vIL-8 gene is involved in early cytolytic infection and
transformation but dispensable to latency. Various in vivo studies in Chapter 3
showed that the er5/AvlL-8 virus had attenuated virulence, and was also highly
protective against challenge with very virulent MDV, 648A (w+). Meanwhile, the
immunohistochemical staining demonstrated that the deletion of vlL-8 gene
impaired the expression of pp38 gene and delayed the expression of Meq gene in
vivo. In the final Chapter, future directions are suggested to further characterize
the protein and its function in the mechanisms responsible for the MDV replication

and transformation.

CHAPTER ONE

LITERATURE REVIEWS

1. History of Herpesvirus

The Herpesviridae family has numerous large DNA viruses. More than 80
viruses have been reported in different hosts including humans, mammals, and
vertebrates, and in one case, an invertebrate (Le Deuff 1994; Minson 2000).
Current molecular analysis have tracked herpesvirus evolution back to more than
200 million years, parallel to vertebrate evolution (McGeoch 1994, 1995). The
genomes of herpesviruses of mammals and birds clearly indicate descent from a
common ancestor, but with a great range of variation in terms of nucleotide

substitution, gene content, and genomic arrangement (McGeoch 1999).

All herpesviruses have similar morphology with a large enveloped icosahedral
nucleocapsid which contains double-stranded DNA. Despite the similar
morphology, the genome varies widely with the genomic size range from 120 to
240 kbp and numbers of genes varies from 70 to 200. The base composition
and pattern of repeated sequences also vary considerably (McGeoch 1989a;
Roizman 1992a). It is common that all herpesviruses establish latency following
primary infection and the latent infection may last throughout the life of the host,
but the biological and the pathological aspects of different herpesviruses vary
greatly. Herpesviruses have been classified into the three subfamilies, the

Alpha-, Beta-, and Gammaherpesvirinae, based on their distinct biological

properties and genomic attributes (McGeoch 1995; Roizman 1992a).
Alphaherpesvin'nae are neurotropic viruses which have the capacity to establish
latent infections primarily in ganglia. The examples of this subfamily include
herpes simplex virus types 1 and 2 (HSV-1, HSV-2), varicella zoster virus (VZV),
bovine herpesvirus and pseudorabies virus (PRV). Betaherpesvirinae have a
narrow host range and establish latency in the secretory glands and
lymphoreticular cells. Members of this subfamily are human cytomegalovirus
(HCMV), human herpesvirus 6 (HHV-6) and human herpesvirus 7 (HHV—7). The
third group, Gammaherpevirinae cause mostly Iymphoproliferative diseases and
the host range is highly restricted. Virus latency is established with lymphoid
cells. Members of this subfamily include Epstein-Barr virus (EBV), Herpesvirus
saimiri (HVS), and Kaposi’s sarcoma herpesvirus or human herpesvirus 8 (HHV-8)

(Table 3).

Because of the lymphotropism and oncogenicity, Marek’s disease virus (MDV)
was originally classified as a gammaherpesvirus along with EBV. However,
analysis at the molecular level suggests that MDV is more closely related to the
members of alphaherpesviruses (Brunovskis et al. 1995). MDV was then

re-classified as an alphaherpesvirus based on its genomic structure.

II. Marek’s Disease Virus

1. History of the disease

The disease was first recognized and described by Joseph Marek, a
Hungarian veterinarian, in 1907 (Marek 1907). The lesions were restricted to the
nervous system; therefore, he named the condition a polyneuritis. Later, this
syndrome was also described in USA (Kaup 1921) and in the Netherlands (Van
der Walle 1924). Their descriptions of MD suggested that the pathological
changes were restricted to the central and peripheral nervous system. The
neoplastic condition of the disease was not recognized until 1926, when
Pappenheimer et al. (Pappenmeimer 1926, 1929a, 1929b) described
Iymphomatous tumor-like masses in 10% of the cases. The significance of these
findings was the recognition that the disease was a Iymphoproliferative process
resulting in lesions in peripheral nerves. These lesions were essentially similar
to the visceral tumors that occurred in some of the cases. However, the visceral
tumors found in many cases of MD appear similar to those found in lymphoid
leukosis caused by a group of RNA tumor viruses. It was the contribution of
Campbell (Campbell 1956, 1961) and Biggs (Biggs 1968) that identified certain
histological differences of the lymphoid tumors, and distinguished the neural
lymphomatosis from other neoplasm and established it as MD. Successful
experimental transmission of the disease was accomplished in early 1960’s by
inoculating young chicks with blood or lymphoma cells from chickens with MD

(Biggs 1963; Sevoian 1962a; Sevoian 1962b). This was the first hallmark in

modern MD research, and the concept of MD was widely accepted by this time.
Further studies on the etiology of MD by Biggs and his colleagues clearly defined
the differences between MD and lymphoid leucosis (LL) (Biggs 1963, 1967).

The real breakthrough came when two laboratories simultaneously isolated a
herpesvirus from MD tumors by co-cultivating in chick kidney cells CKC (Churchill
1967) and in duck embryo fibroblasts (DEF) (Nazerian 1968; Solomon 1968). It
was also confirmed that this infectious agent is cell-associated (Biggs 1967; Biggs
1968; Spencer 1967). The MD virus (MDV) was propagated in cell culture and
established as the etiological agent of MD. Detailed studies by Calnek and his
co-workers demonstrated that the antigens of the herpesvirus were consistently
present in the superficial layers of the feather follicler epithelium and that
enveloped herpesvirus particles were present at this site (Calnek 1969). They
also showed that the cell-free extracts of skin were infectious both for chicken and
cultured cells (Calnek BW 1970a; Calnek 1970b). It was also found that poultry
dust and litter were infectious for long periods (Beasley 1970; Carrozza 1973;
Jurajda 1970). These observations provided a satisfactory explanation as to
why and how the etiological agent of MD is highly contagious and the model of its

transmission.

2. Biological aspects of MD and MDV

1) Classification and pathotyping

MDV is classified as a member of the family Herpesviridae (Roizman 1992b).
Because of its lymphotropism, which is similar to Epstein Bar virus (EBV), MDV
was originally classified as a y—herpesvirus (Roizman 1981 ). However, later on,
based on its genomic organization and significant sequence similarity to other
a—herpesviruses, such as herpes simplex virus (HSV) and varicella-zoster virus
(VZV), it was classified in the subfamily Alphaherpesvirinae (Buckmaster et al.

1988; Roizman 1992a).

a) Serotypes

In a series of studies using the agar gel precipitation test (AGP), indirect
immunofluorescence assay (IFA) and virus neutralization, it was found that there
were antigenic differences between MDV isolates (Blilow 1975a, 1975b). Based
on these results, the MDV group was classified into three serotypes (Biilow 1975a,
1975b; Calnek et al. 1997): Pathogenic strains of MDV and their attenuated
derivatives are the prototype viruses of the MDV group, and are designated as
serotype 1 MDV (MDV-1). The non-pathogenic strain of MDV from chicken
origin is serotype 2 MDV (MDV-2), and that of turkey origin (HVT) is serotype 3
MDV (MDV-3) (TABLE1). The division of the MDV group into three serotypes
has been supported by the differences of the restriction endonuclease digestion
patterns among the genomes of viruses of the three serotypes (Hirai 1979; Ross
et al. 1983; Silva et al. 1991 ). Furthermore, the serotype specific monoclonal
antibodies have been developed which confirmed the specificity of the three

serotypes and aided in viruses classification (lkuta et al. 1982; Lee et al. 1983a).

b) Pathotypes

Based on the ability of different isolates to cause disease in vaccinated
chickens, Witter (Witter 1983; Witter et al. 1985) proposed to classify the isolates
with mild (mMDV), virulent (vMDV), very virulent (wMDV) and very virulent plus
(w+MDV) (Table 1). Calnek et al. (Calnek 1998) found a correlation between
early cytolytic infection and pathotype resulting in the establishement of an
alternate criteria for pathotyping based on lymphoid organ atrophy. There was
also a correlation between pathotype and neurological signs in SPF chicken lines

(Gimeno 2002).
2) Morphology and ultra-structure

The morphology and structure of MDV virions have been studied by many
laboratories (Kato et al. 1985; Payne 1976; Schat 1985). It was shown that
viruses of all three serotypes have similar characteristics typical of other
herpesviruses. Electron microscope studies of thin sections of infected cell
cultures reveal naked hexagonal particles measuring 85-100 nm in the nucleus,
and the presence of some small ring-shaped structures about 35 nm in diameter
(Epstein et al. 1968; Hamdy et al. 1974; Morgan 1959; Nazerian et al. 1968;
Okada et al. 1972). In negatively stained preparations, the unenveloped virion
measures about 100 nm in diameter and has 162 capsomeres (Churchill 1967;
Epstein et al. 1968; Nazerian et al. 1968). Usually the centrally located nucleoid,

50-60 nm in diameter, appears as an electron-dense toroid that surrounds a less

dense cylindrical mass (Nazerian 1974; Nii et al. 1975). The cylindrical mass
probably consists of protein. The electron-dense toroidal structure consists of a
pair of fibrils in a cohelical configuration (Okada et al. 1980). The fibrils are

presumed to be viral DNA, because of the 2-3 nm diameters.

3) Cultivation and propagation of MD Virus

Both chickens and cell cultures are used for the propagation of cell-free and
cell-associated MDV to provide stocks of infectious material for experimental
studies. All three serotypes of MDV can be cultured in a number of types of
tissue cultures. Chick kidney cells (CKC) or duck embryo fibroblasts (DEF) are
the best culture condition for the propagation of low passage serotype-1 MDV
isolates, especially for primary isolation and quantification of infected cells derived
from chickens (Churchill 1967, 1968a; Churchill 1968b; Solomon 1968). In both
cell systems, infection results in a cytopathic effect in the form of characteristic
plaques which can be used for quantitative assays. The attenuated serotype-1
MDV, serotype-2 MDV and HVT propagate well in chick embryo fibroblasts (CEF)
(Biggs 1972; Schat et al. 1978a). The developed discrete focal cytopathic
plaque effects (CPE) consist of clusters of rounded, refractile degenerating cells.
Differences in morphology of MDV-1 plaques in chick and duck cells (Biggs 1972;

Schat 1985; Spencer 1969; Witter et al. 1969) have been described.

MDV can also develop virus pocks on the chorioallantoic membrane (CAM)

of chicken embryo as a result of the yolk sac inoculation with cell-associated MDV

preparations (Biggs et al. 1971; Bfilow 1977). The number of the pocks on the
CAM is directly proportion to the dose of the virus and can be used to quantitate
virus titer. The chicken embryos are also used for the evaluation of the MD
vaccines. At 18th day of hatch, vaccine viruses are inoculated in the amnionic

sac for in ovo vaccination administration (Sharma 1987).

4) Host range

By far, chickens are the most important host system for the MDV in both
natural and experimental infections. Quail is another host for MDV infection.
MDV isolated from either chickens or quail have been used to reproduce MD in
both species (Kenzy et al. 1969; Pradhan et al. 1985; Pradhan et al. 1987).
However, quail-origin MDV appeared to be more pathogenic than chicken-origin
MDV for quail (lmai 1991). Furthermore, the pathogenesis of infection was also
different between chickens and quail, and turkey herpesvirus (HVT) vaccine failed
to protect quail against MDV challenge (Kaul et al. 1991). Chicken-quail hybrids
were found susceptible to MD (Powell 1984). Experimental transmission of MD
was possible in pheasants (Dandapat et al. 1994) but not in ducks. Sparrows as
well as variety of mammalian species were refractory to infection with virulent
MDV (Churchill et al. 1968; Hlozanek et al. 1974; Rispens et al. 1972b; Sharma et

al. 1973; Sharma et al. 1972).

10

3. Pathology of MD

1) Clinical signs

MD induced clinical signs are characterized by asymmetric progressive
paresis and later, the complete paralysis of one or more of the extremities. The
incidence of MD is quite variable in commercial flocks and depends on strain,
dose of virus, age at exposure, maternal antibody, host gender, genetics and
several environmental factors including stress. A particular characteristic of MD
is the unilateral paresis or paralysis of the leg resulted in a typical presentation
where one leg is stretched forward and the other backward (Biggs 1968). With
the extensive use of vaccines, the transient paralysis syndrome is only seen in
experimental trial and seldom in the field. The affected chickens display varying
degrees of ataxia and partial or whole body paralysis beginning 8-15 days after
virus inoculation and last for 1-2 days. During this stage a few affected birds
may die without specific viscera lesions and result in nonspecific MD mortality.
Most of the birds recover only to succumb from clinical MD a few weeks later. It
is clear now that the transient paralysis syndrome appear to be the result of
vasogenic brain edema (Kornegay 1983; Swayne et al. 1988; Swayne 1989).
Infection of MDV in the iris results in blindness in chickens and is another specific
clinical sign of MD. This is normally caused by neoplastic mononuclear cell

infiltration (Calnek 1997).

11

2) Gross and histological lesions

Lesion distribution appears to be similar for naturally occurring or experimental
disease (Pappenmeimer 1926; Payne 1967). Nerve enlargements are one of
the most consistent gross lesions in affected birds. Various peripheral nerves,
particularly the vagus, brachial, and sciatic, become enlarged and lose their
striations (Goodchild 1969; Sevoian 1962b). Diffuse or nodular lymphoid tumors
may be seen in various visceral organs, particularly the liver, spleen, gonads,
heart, lung, kidney, muscle, and proventriculus (Payne 1985; Payne 1976). Both
the genetic strain and the virus strain can influence the location of lesions.
Enlarged feather follicles may be noted in broilers after defeathering during
processing and are a cause for condemnation. But the skin lesions are usually
associated with, but not limited to, feather follicles (Benton 1957).

Non-neoplastic lesions of MD include severe atrophy of the bursa of Fabricius and
thymus as well as degenerative lesions in the bone marrow and various visceral
organs (Jakowski 1970; Witter 1980b). These are the result of intense cytolytic
infections that can result in death of chickens at an early age before the

development of lymphomas.

Histologically, the lesion consists of a mixed population of small, medium, and
large lymphoid cells plus plasma cells and large anaplastic lymphoblasts. These
histopathologic changes associated with MD appear in both the peripheral nerves
(Lawn 1979) and the visceral organs (Payne 1967). The mixed cell populations

in MD associated lesions undoubtedly include both tumor cells and reactive

12

inflammatory cells. The bursa rarely develop tumor, but when it is involved, the

infiltrated cells typically appear in interfollicular areas.

3) Differential diagnosis

Depending on the etiologic agent, virus induced neoplasms of poultry are
divided into two main groups: 1) Herpesvirus induced Marek’s disease, and 2)
retroviruses induced avian leucosis/sarcoma (ALV) and reticuloendotheliosis virus
(REV). Differential diagnosis has been considered difficult in the field, because
MD gross lesions may resemble those of lymphoid leucosis (LL), or
reticuloendotheliosis (RE). RE, although rare, can easily be confused with MD
because both diseases feature enlarged nerves and T-cell lymphomas in visceral
organs. However, LL can usually be differentiated from MD by its common
involvement of the bursa of Fabricius and uniform blast cell morphology. Also,
MD occurs at any age >3 wk (most commonly in birds <16 wk), whereas LL
occurs in older birds. Histochemical assays to detect cellular and viral antigens
in sections of tumor tissue are techniques for differential diagnosis of MD and
other avian viral tumors. Specifically, MD can be diagnosed by the
demonstration of predominant T-cell populations and viral specific antigens, such
as MEQ and MATSA on tumor cells by the histochemistry. Furthermore, MD
lymphomas usually lack evidence of integration or alteration of the cellular

oncogene c-Myc as they are in the cases of avian retroviruses (Table.2).

13

4) Pathogenesis

The pattern of events which occur sequentially in antibody-free, genetically
susceptible chickens which ultimately die from lymphoid tumors after infection
with an oncogenic strain of MDV can generally be divided into four phases (Fig.1):
(1) early cytolytic, productive-restrictive infection, (2) latent infection, (3) late
cytolytic, productive-restricted infection, and (4) transformation. Although
essentially sequential, these are not necessarily discrete phases. A fairly sharp
line demarcates the first two stages, and latent infection in certain cell types is
prerequisite to transformation, but both transforming and latent infections may
exist intermixed with cytolytic infections in different cell populations as lymphomas

are developing in later stages (Calnek 1986, 2001 ).

a) Early cytolytic, productive-restricted infection

The Iymphotropic nature of MDV is illustrated by the fact that the first
significant sites of infection following exposure to the virus are in the major
lymphoid organs. Virus generally gains entry via the respiratory tract, where it is
probably picked up by phagocytic cells, but little or no active infection can be
detected in the trachea, lungs, or air sacs. The productive, cytolytic infection can
be detected with the identification of large amounts of the viral antigen expression
in the lymphoid organs (spleen, bursa of Fabricius, and thymus) peaking at 3-6
days post inoculation (DPI). Few or no enveloped virions are produced; rather,
naked intranuclear particles are present in infected cells. Like all productive or

semi-productive herpesvirus infections, the consequence is cytolysis. The

14

necrotizing effects of this early infection provoke an acute inflammatory reaction
with infiltration of various cells including macrophages, granulocytes, both
immunologically committed and uncommitted lymphocytes. Ultimately, the
consequence is a transient atrophy of the lymphoid organs, especially the thymus
and the bursa. Depending on the virulence of the challenge strain, birds may
recover between 8 to14 DPI or the atrophy may become permanent (Calnek et al.
1998). Using a variety of techniques to separate lymphocyte populations, Shek
et al. (Shek et al. 1983) found that bursa-derived lymphocytes (B-cells) are the
primary target cells for the MDV infection in all lymphoid organs. Apparently,
thymus-derived lymphocytes (T-cells), which are in the resting state, either in the
central organ from which they originate or in peripheral locations are relatively
refractory to infection during this early phase. However, using a sensitive dual
fluorescence technique with monoclonal antibodies to deteCt surface markers on
circulating T—cells, it was later identified that cytolytically infected T-cells also
present during the early infection period. The massive number of apoptotic
thymocytes may possibly be the consequence of viral-induced cytokine changes.
Recently, the presence of the vial homologue of lL-8 (vlL—8) has been reported in
serotype 1 MDV strains. Schat and Xing speculated that production of vlL-8
during the lytic infection in B-cells may be essential in attracting T-cells to the
areas of virus replication, and transfer of infection from B—cells to activated T-cells

(Schat et al. 2000).

Several factors can modify the early pathogenesis. Prior vaccination or the

presence of maternal antibodies reduces the cytolytic infection. The reduction in

15

cytolytic infection will also reduce the number of latently infected cells and reduce

tumor development.
b) Latency

Herpesvirus latency is defined as the presence of the viral genome without
detection of infectious virus except during episodes of reactivation (Feldman 1991;
Fraser 1992; Garcia-Blanco 1991; Stevens 1989; Wager 1991). Maintenance of
a reservoir of latently infected cells within an immunocompetent host is critical to
viral persistence. Like other o—herpesviruses, MDV establishes a life-long latent
infection, but differs from other members of the subfamily in that it goes latent not
in sensory nerve ganglia but in activated T-cells (Calnek et al. 1984; Calnek et al.
1981; Calnek 1997). MDV can be reactivated by cell culture propagation of
T-cells isolated from infected chickens (Calnek et al. 1984). At about 6-8 days,
the infection becomes latent when cytolytic infection can be no longer
demonstrated and tumors are not yet detectable. The development of latency
coincides with the development of immune responses. T- cell mediated
immunity (CMI) plays a central role in this switch (Calnek and Spencer, 1985;
Payne, 1985). The interactions between virus and cells during the induction of

latency are not completely understood.
c) Productive infection in FFE.

MDV replication in FFE is also a cytolytic infection. It is the only site that

produces cell-free infectious virus due to complete virus replication (Calnek et al.

16

1970). MDV is most likely transferred to the FFE by infected lymphocytes. The
lymphoid aggregates consist of small nuclear inclusion in FFE. These
aggregates can either develop into a necrotic area with feather follicle epithelial

cells or degenerated lymphocytes, or into cutaneous tumors.

d) The second phase of cytolytic infection

The second phase of cytolytic infection does not always happen. The
development and extent of this phase depends on genetic resistance of the host
and the virulence of the MDV strain. The second phase of cytolytic infection
coincides with permanent lmmunosuppression. It is much more severe than the
early cytolytic infection and affects not only lymphoid organs but also other
visceral including kidney, ovary, pancreas, adrenal gland, proventriculus, etc.

The details of the second lytic infection have not been clearly defined.

e) Transformation and lymphoma

Susceptible chicken lines will progress past the latent stage and develop a
second wave of cytolytic infection. It is at this stage that Iymphoproliferation and
T—cell tumors would be developed (Buscaglia et al. 1988; Calnek 1997). The
composition of lymphomas consists of a mixture of neoplastic, inflammatory and
immunologically committed and noncommitted cells. Both T and B-cells are
present, but T-cells are the predominant type (Hudson et al. 1973; Ross et al.
1973). The usual target cells for transformation are CD4+, activated T-cells

(Schat et al. 1991). But under certain experimental conditions, other T-cell

17

subsets, including 004*, CD8+ and CD4'ICD8' cells are transfonnable (Schat et al.
1991). But continuous cell lines of MD tumor cells provided uniform population
of cells for study (Matsuda et al. 1976a; Nazerian et al. 1975; Ross et al. 1975;
Schat et al. 1989). Several antigens are found associated with MD tumors cells
or cell lines established from tumors, including Marek’s disease tumor-associated
surface antigen (MATSA) (Matsuda et al. 1976b; Powell 1984; Witter et al. 1975),
now known as a marker associated with normal activated T-cells (McColl et al.
1987), chicken fetal antigens (Murthy et al. 1979; Powell et al. 1983a), and
heterophil and Forssman antigens (lkuta et al. 1981). Because of MATSA’s
association with activated T-cells, it is reasonable to expect that transformed cells
would express this antigen. Ross et al. recently described another MD tumor
associated antigen, AV37 (Ross et al. 1997). The authors speculated that AV37

is linked with activation and may actually play a role in pathogenesis.

Lymphomagenesis is the consequence of MDV infection that is most
commonly associated with MD. Lymphomatous tumors can appear as early as
12-14dpi following infection of young, genetically susceptible chickens by a highly
virulent strain of MDV. Tumors may not appear until several weeks or months
after infection with less virulent viruses or under influencing factors such as
vaccinal immunity, maternal antibodies, genetic resistance, and age of infection.

Lymphomas may occur in almost any visceral organ.

18

5) Virus genes associated with oncogenicity

The MD system provides a number of approaches to identify viral genes that
are responsible for, or contribute to, the transformation of T-cells and the
maintenance of that state. First, the genomes of virulent serotype 1 viruses can
be compared with their attenuated derivatives. Second, genes present in
virulent serotype 1 viruses but are absent in serotype 2 viruses and HVT can be
identified for further study. Third, genes that are expressed in lymphoblastoid
cell lines derived from MD lymphomas and in lymphomas cells isolated from the
chicken can be examined. Using these approaches, a number of genes and

alterations in gene structure have been identified.

a) Phosphorylated protein 38 (pp38)

MDV phosphprylated protein pp38, is an unique serotype 1 gene within
internal repeat long region. It was expressed in a lymphoblastoid cell line and in
lymphoma cells in MD tumors (Naito 1986; Nakajima 1987). This finding
suggests that it might have a role in transformation of lymphoid cells. However,
pp38 homolog gene has been reported in an attenuated serotype serotype 1
strain (Ross et al. 1993) and non-oncogenic serotype 2 and 3 strains (Ono et al.
1994; Ono et al. 1995; Smith 1995). The fact that the pp38 gene is expressed in
productive infections and is found in attenuated virus suggests that it may not play
a direct role in oncogenecity. It was shown that pp38 is involved in the
maintenance of latency and early cytolytic infection and is essential for

transformation (Reddy et al. 2002; Xie et al. 1996).

19

b) Meq gene

A 339 amino acid protein is enoced with the MDV goo Q fragment, therefore it
is designated Meq. Meq gene is located in the repeat long regions, thus two
copies of the Meq gene are presented in the MDV genome. Like other genes in
the repeat long regions, Meq gene is encoded by serotype-1 MDV but not by
serotype-2 (SB-1) or serotype-3 (HVT). Meq is consistently expressed in MDV
latently infected or tumor cells (Jones et al. 1992). It codes for a basic-leucine
zipper (bZlP) domain at the N-terminus (Qian 1996) and is closely related to the
Jun/Fos oncoproteins (Liu 2000). The C-terminal proline—rich domain structurally
resembling the WT-1 tumor suppressor gene (Qian 1995). Because of its
structural properties and specific function, Meq gene has been characterized by
different groups as a transcription factor, an oncogene, andas an immunogen.
As a reflection of its multifunctional roles, Meq has an interesting subcellular
localization pattern (Kung 2001). The cytoplasmic location of Meq is
cell-cycle-dependent and is only being detected in S phase. Meq has two
stretches of basic amino acids which signal the translocation of Meq into nucleus
and nucleolus (Liu 1997). Meq is also localized in the coiled body, which is
correlated with transformation and increased metabolic activity (Liu 1999).
Several factors have defined Meq as an oncogene and transformation factor.
Over-expression of Meq induces morphological transformation and
anchorage-independent growth of rat-2 cell (Kung et al. 2001). Similarly,

inhibitors of Meq in MDV-transformed T-cells results in decreased growth (Xie et

20

al. 1996). This and other evidence strongly support the role of Meq in

transformation.

4. MDV Immunity

MDV immunity has conversed significances: Immunological responses of the
host to MDV infection may be the basis for resistance but the virus can also
induce immunosuppression. Vaccinal immunity is the prime means of control,
but the immunologic response may contribute to the cellular mass of the

lymphoma (Calnek 1997).

1) Cell-mediated immune response

Infection with pathogens normally results in the activation of nonspecific and
specific immune responses. MDV infection has been controlled effectively by
vaccination using nononcogenic and/or attenuated oncogenic MDV strains.
Thus far, there is little knowledge on the role of cell-mediated immune (CMI)

responses during MDV infection or vaccination.

MDV infections cause nonspecific responses involving macrophages, Natural
Killer (NK) cells, and production of specific cytokines and interleukins (IL). Thus
far, there is limited information on cytokine activation after MDV infection.
Infection or vaccination with MDV results in an early activation of NK cells

(Lessard et al. 1996; Schat 1996). NK cells may also play a role in tumor

21

regression (Calnek 1997). Several groups have reported that macrophages can
inhibit virus replication in vitro (Haffer et al. 1979; Kodama et al. 1979; Lee 1979;
Powell et al. 1983b). Depletion of macrophages from spleen cell preparations
enhanced the level of virus replication significantly. The mechanism for the
inhibition of virus replication has not been elucidated, but it has been suggested
that Interferons (lFNs) or macrophage activating factor (MAF) could be involved
(Billow 1984). Specific immune responses are antigen-dependent and require
lymphocyte activation to produce specific antibodies and antigen-specific CD4+
and CD8+ T-cells. MDV specific responses can be detected as early as 5—7 DPI
with the appearance of antibodies and antigen-specific CTL. Omar reported that
virus specific CD8+ CTL but not CD4+ effector cells are responsible to the MDV
specific syngeneic cell-mediated immune response (Omar et al. 1996; Omar et al.

1998).
2) Humoral immune response

MDV induced precipitating and virus-neutralizing antibodies can be detected
within 1-2wk and persist throughout the life of the chicken. The presence of
maternal antibodies may delay or reduce virus replication (Ball at al. 1971; Chubb
et al. 1969) and may interfere with vaccine-induced immunity, especially when
cell-free vaccines are used (Schat 1987). The protective role of the passively
acquired humoral antibody in chickens is thought to cause immune resistance to

MD by neutralizing and reducing the level of infection but may not be able to

22

exclude virus replication. The exact mechanism of how the humoral antibody

can cause immune resistance to MD is not clear.

3) lmmunosuppression

Much of the impact of MD in broiler chickens is considered to be due to
immunosuppression induced by MDV. These infections result in atrophy of the
thymus and bursa of Fabricius leading to reduced numbers of circulating T
lymphocytes and B lymphocytes. It is also noticed that MDV
immunosuppression greatly increased susceptibility to E. coli infection but
reduced infectious bursa disease (IBD) antibody titer. Calnek et al. proposed
three criteria of immunosuppression in MDV: 1) persistence of early cytolytic
infection (Le, a delay or failure to enter latency) in lymphoid organs, 2) atrophy of
the bursa of Fabricius and thymus as measured by organ weight proportional to
body weight at 8 and 14 DPI, and 3) histopathologic evidence of necrosis and
atrophy in lymphoid organs (Calnek et al. 1998). By comparing the relationship
between the immunosuppressive potential and the pathotype of MDV isolates,
they suggested that the degree of immunosuppression is linked to virulence and
that a simple measure of atrophic changes (relative organ weights) in the bursa of
Fabricius and thymus might be useful in determining the pathotype classification
of new MDV isolates. The MDV serotype-3 virus HVT alone caused mild
depletion of T and B lymphocytes but had no effect on immune organ weight or

IBD titer.

23

4) Vaccinal Immunity

Although MD vaccines have been used since 1970, the mechanism for
protection is still poorly understood. Vaccination results in a decrease in viral
replication of pathogenic MDV strains during the early lytic infection as well as a
lower level of infected lymphocytes aftenivards (Lee et al. 1999; Volpini et al.
1995). It has been assumed that this is caused by the generation of
antigen-specific CTL, but recent work from Schat’s group suggested that
nonspecific responses may also be involved (Schat et al. 2000). Vaccinal
immunity is largely against viral antigens, and possibly tumor antigens.
Therefore, vaccination has high efficacy in protecting against tumor induction and
mortality but does not prevent viral replication or shedding and can not eradicate

the disease.

5) Types of vaccines and evolution of virulence

Prior to vaccination, MD was economically the most costly poultry disease.
For the past 30 years, a series of avirulent or attenuated live virus vaccines have
provided protections to chickens in both commercial layer and breeder flocks
(Churchill et al. 1969; Rispens et al. 1972a). The development of successful MD
vaccines is a significant achievement both in avian disease and basic cancer

research. It is now the most effective strategy for prevention and control of MD.

24

Since early 19705, several MD vaccines have been used with success in the
field and generally, they could be grouped into four generations. The first
effective vaccine was attenuated serotype1 virus, HPRS-16att (Biggs et al. 1970;
Churchill et al. 1969), it was then soon replaced by the second generation, the
naturally non-oncogenic turkey herpesvirus (HVT) of serotype 3 MDV (Okazaki et
al. 1970; Witter et al. 1970). The third generation introduced in 1983 was a
bivalent vaccine (Schat et al. 1978b) composed of HVT and 88-1 (serotype-2,
non-oncogenic chicken herpesvirus). About 10 years after using the bivalent
vaccine, more virulent virus strains appeared, which are named very virulent plus
(w+) strains. In mid-19903, an attenuated serotype 1 vaccine virus, strain
CVI988 (also called Rispens) was introduced in US, either alone or in combination
with other vaccines, against w and w+ MDV strains (Witter 1992). But, CVI988

has been used in Europe for much longer timer.

However, the foregoing succession of MD vaccines paralleled the continued
evolution of field strains of serotype 1 MDV to greater virulence (Witter 1997)
(Figure 2). Until now, MD vaccines have been able to keep pace with the
challenges of continued evolution of MDVs towards greater virulence. The future
of sustainable poultry production would be in jeopardy if MD vaccines are unable
to keep pace with evolving MDV. As it was stated, after about 8-10 years of
efficient protection from one generation of vaccine, more virulent viruses emerge
that require the introduction of more efficientive vaccines. The evolution of the
virulence appears to be caused by the mutations of MDV. And it is possible that

the selection pressure created by intensive vaccination could be responsible for

25

MDV evolution. Concerns exist that the continued evolution of MDV toward
greater virulence could break through the CVI988 vaccine-induced immunity and
have devastating effects on poultry. In order to meet the challenges raised by
the continued increases in virulence of MDV strains, the mystery of the cycle of
mutation of MDV around vaccinal immunity and how to prevent future cycle of

mutation is critical to eradicate the disease.

5. Molecular Biology of MDV

1) Physical MDV map and genomic structure

Like other herpesviruses, MDV contains a linear double-stranded DNA
genome that has a density of 1.705 g/cm3 in CsCI, and a base composition of
46% guanine and cytosine. The molecular weight of MDV DNA is about 108-120
x 106 Dalton (Da) (Lee et al. 1971), equivalent to 166-184 kilo base pairs (kbps)
(Cebrian et al. 1982; Fukuchi 1984; Hirai 1979). The composition and density of
HVT DNA is similar to MDV-1 DNA (Lee et al. 1971 ). The genomic structure and
gene arrangement of MDV DNA are similar to that of d- herpesviruses, including
HSV and varicela-zoster virus (VZV) (Buckmaster et al. 1988; Davison 1986;
Karlin et al. 1994; Roizman 1992a; Wu et al. 1988), although biological properties

of MDV are close to y—herpesviruses, such as Epstein-Barr virus (EBV).

The genomic sequences of all three serotypes of MDV have been completely

sequenced which include two serotype 1 MDVs, GA (virulent strain) (Lee et al.

26

2000) and Md5 (very virulent strain) (Tulman et al. 2000), one serotype 2 MDV,
HPRS-24 strain (Izumiya et al. 2001), and the serotype 3 MDV, FC126 (HVT
strain) (Afonso et al. 2001). The DNA sequence data indicates that the genomic
size differs between serotypes. The MDV genome consists of unique long (UL)
and unique short regions (Us), each bounded by inverted repeats, called terminal
repeat long (TRL), internal repeat long (IRL), internal repeat short (IRS) and

terminal repeat short (TRs) (Figure 3).

Several direct repeats have also been identified in the MDV genome (Hirai
1988). These direct repeat sequences are mostly located within the internal and
terminal repeat regions. A heterogeneous expansion region, containing multiple
tandem 132 bps repeats, has been identified in attenuated MDV and mapped to
the inverted regions flanking the UL region of the genome, TRL and IRL (Fukuchi et
al. 1985a; Maotani et al. 1986; Silva 1985), adjacent to the MDV origins of
replication (Ori), and are designated as DR1. These DR1 can be expanded in
vitro by passages of virulent MDV-1 strain in primary CEF cell culture (Fukuchi et
al. 1985b; Maotani et al. 1986). There exist two copies of 132 bps in virulent
MDV-1, and multiple copies in their attenuated derivates. Kawamura et al.
suggested that the function of this 132 bp repeats may be associated with virual
oncogenicity (Kawamura 1991). With the establishment of cosmid clone of MDV
Md5 strain, Silva et al. constructed the recombinant virus with the deletion of two
copies of DR1 repeats. The deletion virus has similar virulence and

tumorgenicity as the parental virus Md5 (Witter 1980a).

27

Based on the physical maps of BamHl restriction endonuclease (RE)
fragments constructed from MDV-1 (Figure 3) (Fukuchi et al. 1985a), MDV-2 (Ono
et al. 1992), and HVT (lgarashi et al. 1987), it was shown by cross hybridization
data (Yoshida et al. 1994) that all three serotypes differ in their RE digestion
patterns (Silva 1991), but share significant homology at the DNA level. The RE
maps have become a basis for most gene identification and localization (Figure 4),

and also are useful for comparative studies.

2) Cosmid and Bacterial Artificial Clone (BAC) systems for recombinant

MDV.

Although the members of the Herpesvirus family are responsible for a wide
variety of human and animal diseases, advances in the understanding of viral
molecular mechanisms of pathogenesis have been hampered by the large size of
herpesvirus genomes, rendering the viruses difficult to experimentally manipulate.
In recent years, manipulation of the large herpesvirus genomes have been
facilitated by using cosmid DNA library and bacterial artificial chromosome (BAC)
vectors (Arvin 2001; Cohen 2001; Schumacher et al. 2000). The genomes of
murine and human cytomegaloviruses (HCMV) (Ehsani et al. 2000; Messerle
1997), herpes simplex virus type 1 (Cohen et al. 1993; Suter 1999), pseudorabies
virus (PRV) (Smith 1999), and Epstein-Barr virus (Delecluse 1998; Tomkinson et
al. 1993) have been cloned either as overlapping cosmid clones or infectious

BACs, or both. These techniques also facilitated mutagenesis of herpesvirus

28

genomes, allowing for the assessment of the role of specific viral gene products in

replication, immunity, and pathogenesis.

Studies on MDV have been carried out for decades, but because of its strong
cell-associated features and lack of efficient tools, MDV gene characterizations
have been limited to in vitro methods (Cui et al. 1990; Jones et al. 1992). To
understand the viral oncogenicity of MDV, it is necessary to have extensive
knowledge about MDV genes functions, especially their roles in vivo. The best
way to characterize the role that proteins play in oncogenesis is to introduce

site-specific mutations in the viral genome.

The finding in early 19903 that purified MDV DNA was infectious in cell culture
(Wilson et al. 1991) indicated that MDV recombinants could be generated by
cotransfection with selectable markers. From then on, generation of mutant
MDV has always been done by the marker rescue method (Bublot et al. 1999;
Parcells et al. 1995). The disadvantages of using this method in MDV are
obvious. Since this method requires the introduction of a selectable marker in
the mutant virus as well as several rounds of plaque purification, the whole
process of selection of recombinant viruses is laborious which can cause the
introduction of unexpected mutations elsewhere in the genome and even
attenuation of the recombinant virus. In addition, the introduction of a foreign
marker gene in the viral genome could also generate the effect to the viral

oncogenecity, although it has not been reported thus far (Parcells et al. 2001).

29

Recently, overlapping MDV cosmid clones have been generated and was
successfully used to introduce mutations into a wMDV strain, Md5 (Reddy et al.
2002). This method does not require repeated rounds of plaque purification and
insertion of selective marker genes. Thus far this is the only tool available to
introduce site-specific mutations into oncogenic MDV. Using this recombinant
technology it is shown that pp38 is not essential for replication in cultured cells
and dispensable for tumor induction, but deletion of pp38 severely impaired in
vivo replication in lymphoid organs (Reddy et al. 2002). Several other MDV
specific genes have been targeted for characterization now using this method. It
is believed that easy generation of mutant viruses with this technique will lead to a

better understanding of the mechanisms of MDV pathogenesis.

Bacterial artificial chromosome (BAC) technique is the other useful tool to
introduce mutations to large DNA virus genomes. Thus far, all of the MDV BAC
clones were derived from attenuated virus, such as 584Ap80C (Schumacher et al.
2000; Tischer et al. 2002), and vaccine strain CVI988 (Petherbridge et al. 2003).
MDV BAC clones could not be used for pathogenesis studies presently. But it
has been reported that MDV BAC clone DNA is capable of inducing protection as
a vaccine, possibly through the reconstitution of live virus in vivo. This
represents a new generation of DNA-based vaccines against herpesviruses

(Petherbridge et al. 2003; Tischer et al. 2002).

III. Chemokine family'and Virokines

30

1. Introduction to chemokines family

Chemokines constitute a growing superfamily of intercellular messengers
which play multiple roles in the development and homeostasis of different organ
systems, particularly the hematopoietic system, as well as the generation of both
innate and adaptive immune responses. It is now known that chemokines and
their receptors are expressed by a wide variety of nonhematopoietic cells, and
that chemokine function extends far beyond leukocyte physiology. For example,
the knowledge of the relation among chemokines, their receptors and human
immunodeficiency virus (HIV) infection broadens the previously narrow focus on
chemokines as mere chemoattractants. Chemokines are small molecules, with
molecular weights in the range of 8 to 12 KDa, but there are exceptions which
involve proteins comprised of multiple domains, (Bazan 1997; Ernst 1994).
Chemokines have four highly conserved cysteine residues. Based on the
number of amino acids which separate the first two cysteine residues, chemokine
families are classified into four groups: CXC (a class), CC (8 class), CX3C (7 class)
and C (8 class) (Baggiolini 1997). CXC family can be further divided into two
subgroups of molecules (ELR‘ and non-ELR) according to the presence or
absence of an “ELR” (Glu-Leu-Arg) motif located immediately before the first
cysteine residue. The presence of “ELR” seems to correlate to the neutrophil
chemoattractant (Rollins 1997b). But there are also some exceptions, such as

chicken Chemotaxis and Angiogenic Factors (cCAF), 9E3/CEF. The chicken

31

cCAF gene has an “ELR” motif, but it does not attract neutrophils, only monocytes

(Martins-Green 2001)(Table 4).
2. Chemokine and virus

The successful propagation of viruses within the host requires skillful evasion
or manipulation of the host immune arsenal (Alcami 2000; Lalani 1999a;
McFadden 2000; Murphy 2001). Large DNA viruses, particularly the poxviruses
and herpesviruses, provide some of the most extensive inventories of gene
products that serve to defend these viruses against the aggressive assault
executed by the host inflammatory response. The chemokine network represents
a significant target for disruption or exploitation by these viruses (Lalani et al. 2000;
Murphy 2001; Rosenkilde et al. 2001). From a functional standpoint, two major
groups of chemokines can be distinguished: housekeeping (HK) chemokines
(Baggiolini et al. 1994; Butcher 1996), which are generally expressed
constitutively under physiological conditions and play essential roles in
development and homeostasis; and proinflammatory (Pl) chemokines (Cook
1995), which are typically inducible and participate in the generation of both
innate and adaptive immune responses. The relation between virus and the
chemokine system is characterized by a complex blend of enmity and attraction.
Chemokines are key regulators of innate and adaptive immune responses against
invading microorganisms, including viruses (Baggiolini 1998; Baggiolini et al.
1995). As immune system “traffic officers”, chemokines control leukocyte

migration under either physiological or pathological conditions, meanwhile,

32

chemokines could modulate the induction, amplification, and cytokine-secretion
pattern of antiviral responses. Indeed, the physiological effects of
chemokine-elicited intracellular signaling are not only limited to the activation of
Chemotaxis. Chemokines also serve as co-stimulators that amplify both
proliferative and cytokine-secretive T-cell responses which allow antiviral immune
responses to proceed beyond a critical threshold under conditions of low antigenic

load or during infection by non-cytopathic viruses (Taub 1996).
3. Virokines

Viruses have succeeded in turning the chemokine system into an ally.
During the course of a long parallel evolution, viruses have captured from their
hosts the genetic information for encoding chemokines and chemokine receptors
and have reprogrammed it for evading the control of the immune system. The
large DNA viruses, particularly the poxviruses (Lalani 1999b) and herpesviruses,
have developed several mechanisms to corrupt the normal functioning of the
chemokine network by either mimicking chemokines or chemokine receptors
(Seet 2002). These include U83 of HHV-6 (Zou 1999), leP-l, —II and -III of
HHV-8 (Chen 1998; Dairaghi 1999) and M131 of murine CMV (Table 5). All
these chemokines are CC chemokines. Liu et al. first reported the identification
of vlL-8 in MDV (Liu et al. 1999) and about the same time, Penfold et al. reported
the discovery of UL146 and 147, two lL-8-like chemokines in human CMV
(Penfold et al. 1999). These are the only viral CXC chemokines identified thus

far. Among all these Virokines thus far identified, many of them have been

33

characterized. Knowledge of how viruses function to oppose chemokines is
providing important insights into how the immune system combats viral infection
and contributes to a better understanding of how to better control unwanted
chemokine-mediated inflammation. Because of the elucidation of the complex
relationship between HIV and the chemokine system, a cautious optimism on the
possibility of developing effective strategies for the control of HIV infection has

been raised (Baba 1999; Bashoff 1997; Faure 2000).
4. Chemokine and angiogenesis

Several members of the CXC family are among the first chemokines identified
as regulators of angiogenesis, acting either as angiogenic or angiostatic factors
(Strieter et al. 1995a). Angiogenesis, defined as the growth of new capillaries
from preexisting vessels, is a pervasive biological phenomenon that is at the core
of many physiologic and pathologic processes. An opposing balance of
angiogenic and angiostatic factors regulates angiogenesis. Examples of
physiologic processes that depend on angiogenesis include embryogenesis,
wound repair, and the ovarian/menstrual cycle (Strieter et al. 2002). A common
characteristic of all solid tumor growth is the presence of neovasculan’zation.

This is a basic requirement for solid tumor progression, since the absence of new
vessels does not allow tumors to grow beyond a few millimeters. In addition,
blood-bome metastasis cannot be initiated without neoplastic cell access to blood
vessels. Tumor cells may induce angiogenesis via release of growth factors,

such as prostaglandins and fibroblast growth factor 2 (FGF-2), and by the

attraction of inflammatory cells which in turn release multiple angiogenic stimuli

(Bemardini et al. 2003).

Interestingly, the presence or absence of an “ELR” motif in the amino acid
sequence seems to correlate with an angiogenic or angiostatic activity,
respectively (Strieter et al. 1995b). The requirement of the “ELR” motif has been
demonstrated by site-directed mutagenesis substitution of the “ELR” motif of an
angiogenic chemokine (i.e. lL-8/CXCL8) with a non-ELR motif of the angiostatic
molecule (Mig/CXCL9) and vice versa. The shift in angiogenic properties of the
mutated chemokines in both in vitro and in vivo assays, strongly supports the
importance of the “ELR” motif as a structural domain for angiogenic activity

(Strieter et al. 19950).

The angiogenic members of the CXC chemokine family include interleukin
(lL)—8 (lL-8/CXC ligand [CXCL8), epithelial neutrophil activating protein (ENA)-78
(ENA-78/CXCL5), growth-related genes (GROs) [GRO-a, GRO-B, and GRO-«y;
CXCL1, CXCL2, and CXCL3, respectively], granulocyte chemotactic protein
(GCP)-2 (GCP-2/CXCL6), and NHz-terminal truncated forms of platelet basic
protein, which include connective tissue activating protein-III, IS-thromboglobulin,
and neutrophil activating protein (NAP)-2 (NAP-2/CXCL7) (Strieter et al. 1995d)

(Table 4).

The angiostatic members of the CXC chemokine family include platelet
factor-4/CXCL4, monokine induced by lFN-y (MIG) [MlG/CXCL9], and

lFN-y—inducible protein (lP)-10 [lP-10/CXCL10] (Zlotnik 2000). All three

35

IF N-inducible ELR- CXC chemokines specifically bind to the CXC chemokine
receptor, CXCR3 (Zlotnik 2000).

Angiogenesis is regulated by an opposing balance of angiogenic and
angiostatic factors. CXC chemokines comprise a unique cytokine family which
contain members that exhibit on a structural/functional basis either angiogenic or
angiostatic biological activity. As a family, the CXC chemokines appear to be
important in the regulation of angiogenesis associated with the pathogenesis of
chronic inflammatory/fibroproliferative disorders. These findings support the
notion that therapy directed at either inhibition of angiogenic or augmentation of
angiostatic CXC chemokines may be a novel approach in the treatment of chronic

fibroproliferative disorders.

5. MDV encoded vIL-8 gene

The MDV encoded vIL-8 gene is located within the repeated sequences
flanking the unique long (UL) region of the MDV genome. The vlL-8 gene is
produced by multiple splicing with a conserved exon-intron organization. The
protein is 134-amino-acid long. It is designated as IL-8 like gene based on the
high sequence and structure similarity compared with the cellular lL-8 genes,
including both mammalian and avian lL-8 (Figure 5) (Hughes et al. 2000; Kaiser et
al. 1999; Martins-Green 2001; Sick et al. 2000). Similar to other cellular
homologs, vlL-8 has a signal peptide at the N-terminal end which signals the
secretion of the vlL-8 molecule and vlL-8 does have the four conserved cysteine

residues that function in forming two disulfide bridges in the protein tertiary

36

structure that might have effects on the stability of the structure and also be
essential in maintaining biological function of the protein (Rajarathnam et al.
1999). However, there are interesting structural differences between vlL-8 and
cellular lL-8. Most notably, vlL-8 carries extra sequences in the C-terminal
domain. Secondly, the NH2 terminus of the majority of the CXC chemokines
contains three amino acid residues (Glu-Leu-Arg: the “ELR” motif), which
precedes the first cysteine amino acid residue of the primary structure of these
cytokines, while MDV encodes vlL-8 protein it does not have an “ELR” motif,
instead it is Asp-Lys-Arg (DKR). Based on this sequence feature, vlL-8 belongs
to ELR’CXC chemokine family. Because of its chemokine-like properties, it is
hypothesized that the function of vlL-8 is to attract cells for MDV as target cells for
virus infection and replication. Since MDV is a cell associated virus, target cells
have to come in contact with the infected cells to become infected. It is possible
that vlL-8 plays a critical role in attracting naive cells for infection. The role of
vIL-8 in virus replication in vitro and in vivo, and the role of vlL-8 in lytic infection,

latency, or transformation, is evalulated in the studies reported herein.

37

 

 

 

 

 

 

 

 

 

Serotype Pathotype 5:33),” Oncogenicity
Serotype 1
1 mild (mMDV) CU2 Chicken
(MDV-1)
virulent (vMDV) GA Chicken
very virulent (wMDV) M d5 Chicken
very virulent plus (w+MDV) 648 A Chicken
Serotype 2
(MDV-2) 2 SB1 no
Serotype 3
3 HVT no
(MDV-3)
Table 1. Classification of MDV strains by serotypes and pathotypes.

38

 

MD LL REIBursal REINonbursal

 

Virus type herpesvirus retrovirus retrovirus retrovirus
Chicken age
at onset 3wk >14wk >14wk
Gross
lesion:
nerve + - - -
bursa atrophy enlarge enlarge
Histology Pleomorphic Homogeneity Homogenity pleomorphic
B cell
marker ' + + '-
T cell
marker + ' ' +
pp38 + - - -
Meq + - - -
MATSA + - - -
c-Myc - + + +

 

Table 2. Differential diagnosis of MD and other chicken lymphoma
diseases.

MD: Lymphoma by Marek’s disease virus; LL: Lymphoid leucosis by ALV;
RE/Bursal: Bursa lymphoma by Reticuloendotheliosis virus (REV); RE/Nonbursal:

Non bursa lymphoma by REV; +2 Present; -: Absent.

39

MDV Infection

  

 

MDV
Replicationl

  
  

 

 

 

ApoptOSIs Apoptosrs
l
I L fl] _ 1I l
ELI)? toI tic infection and >l-2 week post inoculation:l >3-4 wk post infection '
y y y Latent infection Iymphoproliferative In
immune responses nerves and visceral
organs

Figure 1. Possible sequentlal events In the pathogenesis of Marek’s disease.
MDV infected cells or dander are phagocytized by macrophages followed by
infection of B-cells. The infected B-cells initiate the early cytolytic infection.
Some of these B-cells undergo apoptosis. Infected B-cells may induce an early
immune response, releasing other cytokines and chemicals such as lFN-y or NO,
which will assist in activating T-cells. Only the activated T-cells (aT) but not the
resting T-cells are the targets for MDV infection. Some of the aT-cells undergo
apoptosis. others enter into the latency (aT-Latency) phase or become

transformed (aT-Transformed), finally leading to the development of tumors.

40

Rispens
A

 

 

 

HVT+SB1 I
i W... ” 0‘...
:0 ‘0’.
9.
DD
5' HVT W
0
S.
a 1
3 v
0
(D
__ M
l l l l l l I
1940 1960 1980 2000

Figure 2. The continued evolution of MDV towards greater virulence parelled
with the foregoing succession of MDV vaccines. MDV virulence has
increased from low virulent mild (mMDV) strain, to the virulent (vMDV) strain, to
high virulence of the very virulent (wMDV) strain and, finally to the very virulent
plus (w+MDV) strain. MDV vaccines have also been kept pace with the
continued evolution of MDV virulence. HVT was used in 19703, and was only
efficient against vMDV. Bivalent vaccine (HVT+SB1) was introduced in mid
1980’s, and was used to protect chickens against wMDV strains. With the
emergence of w+MDV, Rispens strain has become the most efficient vaccine

against wMDV and w+MDV strains since 19903.

41

TRL IRL U TR

II:l———--I_I_l—l\

+— —> >-> —>
pp38 Meq

 

 

 

 

 

Figure 3. Schematic map of MDV genome and the locations of genes Meq,
pp38 and vlL-8. Double stranded DNA structure of MDV consists of unique long
(UL) and unique short (Us) regions each bounded by inverted repeats, named
terminal repeat long (T RL), internal repeat long (IRL), internal repeat short (IRS)
and terminal repeat short (TRs). MDV-1 unique genes, such as pp38, Meq and

vlL-8, are located in the repeat long region as indicated in the figure.

This figure Is In color.

42

0 2 4 6 8 10 12 14 16 180K

Kb n n n n n 4 n n n n n n n n n n n n n

 

 

 

 

 

 

 

 

TR; IR; IR. umoue TR.
..........—. "m :EZI’"—-I:I
onemlzmo

3 S T N
0 CI 01 f O
smut ‘I Rf A
MAP I I I II I I II I III I I II II I
mt I2 0 G c F mm E 92:: J a MK1 H 12
EcoRl W 111111 1 11111 11 n I II [I In F I
“A" o F K J G TMN e L1L2 o c e x F o o I

Figure 4. Schematic structure and restriction map of the MDV genome.
The size of MDV genome and its organization are shown in the top panel. The
location of BamHI and EcoRI restriction endonucleus sites are shown in the

bottom (adapted from Fukuchi et al., 1984; Silva and Witter, 1985)

43

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. Natural Associated Other
Virus subfamily Host disease Oncogenic species
Marek's disease . T-cell Turkey,
virus alpha chicken Lymphoma yes quail
HumaJiirzgmplex alpha human -
varicella zoster Vericella
' 9
virus alpha human zoster .
pseudorabies . Aujeszky’s 9
virus alpha pig disease ‘
Human T-cell
7
herpesvirus 6 beta human lymphomas '
Human T-cell
9
herpesvirus 7 beta human lymphomas '
human
Cytomegalovirus beta human
Epstein-Barr Burkitt's 9
virus gamma human lymphoma yes .
Nasopharyngeal
carcinoma
Kaposllizfsrcoma gamma human Kaposi sarcoma yes ?
Herpesvirus a m m a squirrel l m h om a no other
saimiri g monkey y p monkeys
Herpesvirus spider other
ateles gamma monkey lymphoma no monkeys
Hergaegigrus gamma Baboon lymphoma yes Marmosets
Herpesvirus Cotton 9
sylvilagus gamma rabbit 'Ympmma yes '
Lucke Renal
7 7
herpesvirus ' Frog carcinoma yes ‘
Chenoid sea .
7 9 7
herpesvirus . turtle Skin tumors. yes .

 

 

 

 

 

 

 

 

Table 3. Classification of oncogenic herpesviruses (Adapted from (McGeoch

1999; McGeoch et al. 2000).

44

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Name Target Cells Angiogenic Angiostatic
ELR
Neutrophils, T-cells,
lL-8 basophils yes
?endothelial cells
Neutro hils,
GRO'O‘ (MGSA) melanopma cells yes
?endothelial cells
Neutrophils,
GRO'B (MIP'Z) ?endothelial cells yes
Neutrophils,
GROW (WP-2) ?endothelial cells yes
ENA-78 Neutrophils yes
GCP-2 Neutrophils yes
Mononuclear cells
9E3/CEF ?endothelial cells yes
Mononuclear cells
K60 ?endothelial cells yes
Platelet basic _ _
rotein
CATP-III Fibroblasts ?
Thromboglobulin Fibroblasts ?
Neutrophils, 9
NAP-2 basophils '
Non-ELR
Fibroblasts,
Platelet factor 4 endothelial cells yes
activated T-cells,
?endothelial cells,
'P'10 ?NK cells yes
MIG activated T-cells, yes
SDF-1 T-cells,?B-cells yes

 

 

 

 

 

 

Table 4. CXC chemokines, their target cells and angiogenesis activity

(Adapted and edited from (Rollins 1997a)).

45

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Cellular .
Virus Coding gene homologges Proposed functions
MDV vIL-8 lL-8 ?
"1131 chemokine
MCMV MCK 1 CC Chemokines agonist(target cell
( ' I ' 7
recruntment . )
Functional receptor
m33 CCR1 (disseminate infected
cells)
chemokine agonist
HCMV (9&3) lL-8 (target cell
recruitment?)
UL147
- 9
(vcxc-2) 'L 8 '
U827 ? ?
Functional CC
U828 CCR1 chemokine receptor
H lV-1 corecgptor
HHV-6 UL33 CCR1 ?
chemokine agonist
UL83 CC Chemokines (target cell
recruitment?)
UL12 CC Chemokines Functional CC
receptors chemokine receptor
UL51 CC Chemokines chemokine
receptors sequestraion?
CC Chemokines
- 9
HHV 7 UL12 receptors .
chemokine agonist
HHV-8 K6(VMlP-l) MIP-1 on CCR8,
Angiggenesis
chemokine agonist
K4(VMlP-II) MlP-1 on CCR3,
Angiggenesis
chemokine agonist
K4.1(VMlP-Il) MIP-1 on CCR4,
Angiogenesis
constitutive signaling,
HSV ORF74 CXCR2 cellular
transformation
functional CXC
ORF74/ECRF3 CXCR2 chemokine receptor

 

Table 5. Herpesviruses encoded chemokine or chemokine receptor

homologues (adapted from (Lusso 2000)).

46

 

hIL-8

102\bp 271bp 2 425bp 84%;)

21a WI 83 an I 588

M313?" ...... ...........

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

QE3ICEF4
75bp 596D
\ 20“ 774bp 46“ 64pr 28“ J 11“ /790b
‘ \ .
m. ....... \\\\\\ ........ \\\. ..... .\
MDV-VIL-3
16b
p\l\ 2088 I 98bp 68aa 34bp
31W
I:I Untranslated Translated -------- introns

Figure 5. Structure comparison of lL-8. Cellular lL-8s inculding hlL-8 and

chicken lL-8 9E3/CEF consist of four exons and three introns.

The MDV

encoded vIL-8 gene only has three exons and two introns, and it codes for a much

longer exon 3. The number of nucleotides for each intron or untranslated region

and the number of amino acids for each translated region are marked in the

figure.

47

CHAPTER TWO

MDV ENCODED VIL-8 GENE IS INVOLVED IN EARLY CYTOLYTIC

INFECTION AND TRANSFORMATION BUT DISPENSABLE TO LATENCY

ABSTRACT

Marek’s disease, a lympho-proliferative disease of chickens, is caused by an
alpha-herpesvirus MDV (Marek’s disease virus). This virus encodes a virokine
vlL-8, with general homology to cellular CXC chemokines such as IL-8 and Gro-a.
To study the function of vlL-8 gene, both copies of vlL-8 residing in the TRL and
IRL regions of the viral genome were deleted. The vlL-8 deleted virus
er5/AvIL-8 was generated. Growth kinetics studies showed that vlL-8 gene is
dispensable for virus replication in cell culture. In vivo, vlL-8 gene is involved in
early cytolytic infections in lymphoid organs as indicated by less viral antigen
expression detected by immunohistochemical staining. But the er5/AvlL-8
virus is unimpaired in virus spreading. Similar viremia titers as the er5 virus at
6 and 8 dpi indicated that vlL-8 gene is not necessary for virus reactivation and is
not involved in latency. Nevertheless, deletion of vlL-8 gene compromised
transformation of the virus with reduced number of transformed cells at 5 week
post inoculation and less gross tumors development in vivo. The revertant virus
that restored the expression of vlL-8 gene showed the same phenotype as the

er5.

48

One of the interesting sequence differences between the vlL-8 gene and other
cellular homologue is that vlL-8 encodes a “DKR” motif instead of “ELR”. To
study the role of the variation of this motif in MDV pathogenesis, we generated
recombinant MDV, er5/vlL-8-ELR, carrying ELR motif by mutagenesis. Both in

vitro and in vivo studies showed that er5/vlL-8—ELR has the same phenotype as

er5.

49

INTRODUCTION

Marek’s disease (MD) is a contagious, Iymphoproliferative disease of domestic
chickens in which mononuclear infiltration, demyelination of peripheral nerves,
and T-cell lymphomas are common features (Biggs 1975). The etiological agent
of MD is a lymphotropic, oncogenic herpesvirus, MD virus (MDV). The MDV
genome is about 180 kbps in length and is most related genetically and
structurally to the genomes of alpha-herpesviruses, such as herpes simplex virus
and varicella-zoster virus (Buckmaster et al. 1988; Lupiani et al. 2001). Recently,
the complete nucleotide sequences have been determined for all serotypes of
MDV(Afonso et al. 2001; Lee et al. 2000). The data showed that MDV and other
alpha herpesviruses are colinear in the unique long and short regions but differ
substantially in the adjacent repeats (Lee et al. 2000; Tulman et al. 2000). MDV
is grouped into three serotypes: serotype 1 consists of all pathogenic virus strains,
serotype 2 comprises the naturally occurring, non-oncogenic strains in chickens,
and serotype 3 includes the non-pathogenic herpesvirus of turkeys (HVTs)
(Becker et al. 1992; Btilow 1975c; Kaaden 1977; Lee et al. 1983b). MD
incidence has largely been controlled by vaccination with all three serotypes of
MDV, often in bi- and multivalent combinations since the 19703 (Witter 1991,
2001). However, there is a continuation of an apparent evolutionary trend of
MDV towards greater virulence, which results in recent increased losses from
Marek's disease (MD) in vaccinated flocks (Calnek 1997). A thorough
understanding of the genes involved in the replication, immune modulation and

oncogenesis holds keys to the development of improved live-vaccines, based on

50

targeted mutations of the MDV genome. We have focused on genes specific to
the oncogenic serotypes of MDV and developed a cosmid-based recombinant
virus approach to study their functions in vivo (Reddy et al. 2002). In this study,
the findings on vIL-8, a virokine encoded by an oncogenic serotype of MDV is

reported.

vIL-8 is located in the repeat region of the MDV genome and like other
Virokines of herpesviruses, may be involved in viral replication and/or host
immune-modulation (McGeoch 1989b). MDV vlL-8 shares significant homology
to cellular CXC chemokines such as lL-8 and GRO-a, and is the only one found in
alpha-herpesvirus. Cytomeglavirus, a beta-herpevirus, encodes two CXC
Virokines (i.e., UL146 and 147) (Penfold et al. 1999). Most other Virokines
belong to the CC family of chemokines. Mutagenesis studies of ELR+
chemokine (IL-8) and ELR- chemokine (Mig) revealed that the presence of the
ELR motif correlated well with the chemokines” ability to attract neutrophils during
inflammation (Baggiolini et al. 1997) and induce angiogenesis in tumorigenesis

(Strieter et al. 1995b).

Liu et al. reported the identification of MDV vlL-8 and the initial
characterizations of this virokine (Liu et al. 1999). It was found that vlL-8 has a
DKR motif in place of ELR, and in a chemotaxis assay, the major cell types
targeted by vlL-8 are mononuclear cells rather than heterophils (chicken
equivalent of neutrophils). A vlL-8 deletion mutant in the genetic background of

RB1 B strain of MDV was constructed by inserting a soluble-modified green

51

fluorescent protein (smGFP) expression cassette at the site of deletion. This
vlL-8 virus was found to replicate well in cell culture, but much less so in vivo and
had a weak oncogenic phenotype (Parcells et al. 2001). In this early study,
revertant virus was not developed and inadvertent mutations responsible for
some of the observed phenotypes cannot be completely ruled out. Nevertheless,
the results have provided an important framework for the understanding of the

general properties of MDV vlL-8.

In this dissertation, an extended study was carried out by using the newly
established MDV cosmid DNA library (Reddy et al. 2002) to construct
recombinant MDVs. This approach permits the efficient construction of revertant
virus, unattainable by the previous approach. The in vivo infection course of the
mutant virus was also characterized in more detail. The results demonstrated
that vlL-8 plays an important role in the establishment of early infections,
presumably functioning to recruit target cells for MDV infection. To test whether
the DKR is critical in target cell selection, a mutant virus where DKR is replaced

by ELR was also developed and the properties of the mutants is discussed.

52

MATERIALS AND METHODS

Antibodies. Monoclonal antibody (Mab) H19 specific to viral protein pp38
(Cui et al. 1990) was used to detect pp38 by IFA and IHC. Rabbit antiserum

specific to viral protein vlL-8 was used to detect vlL-8 in IFA and western blot.

Chickens. Chickens used in the study were specific pathogen free MD
susceptible F1 progeny (15x7) of Avian Disease and Oncology Laboratory line
15l5 males and line 71 females. All the chickens were wing-banded at hatch, and
randomly sorted into different experimental groups (17 chickens per group) and

held in modified HorsfalI-Bauer isolators.

Cells and viruses. Primary duck embryonic fibroblasts (DEF) were used for
virus propagation and DNA transfections. Recombinant viruses were generated

from cosmids derived from a very virulent MDV strain, Md5 (Witter 1980a).

Plasmids. A 3.1kb fragment (nt. 1,451 to 4,543) containing the entire
sequence of the vlL-8 gene was obtained by digesting the 8N5 cosmid with
BamHI and was cloned into the same site of pUC19 generating the transfer vector
pUC19/SNSBamHI. Subsequently, the pUC19/SN5BamHI was digested with
Clal (nt. 2,807) and Ncol (nt. 3,605), blunt ended and religated to generate the

vlL-8 deletion transfer vector, pUC19/SN58amHl/AvlL-8.

Substitution of the DKR (Asp-Lys-Arg) motif of vlL-8 with the ELR
(Glu-Leu-Arg) motif in the plasmid pUC19/SN58amHl was carried out using the

QuickChangeTMXL site-directed mutagenesis kit (Stratagene, CA), according to

53

the manufactures’ instructions. The primers used were: primer 1: 5’G AGT CTC
GCT GTC GAQ §_TG AGG TGC AAG TGC G 3’ and primer 2: 5’0 GCA CTT GCA
CCT C,A_G QT C GAC AGC GAG ACT C 3’ where the introduced mutations are

indicated by bold and underline. The presence of the corresponding mutations in

the plasmid pUC19/SN53amHl/ELR was confirmed by sequencing.

Cosmids. MDV cosmid clones SN5, P89, SN16, A6 and B40, from the very
virulent strain, Md5 (Witter 1997), encompassing the entire MDV genome were
used to generate recombinant Md5 viruses (Reddy et al. 2002) (Figure 6).
Cosmid clones A6 and SN5, containing a copy of the complete coding sequence
of the MDV unique gene vIL-8 in opposite orientation, were used to introduce

vlL-8 specific mutations.

The RecA-assisted restriction endonuclease (RARE) cleavage method (Ferrin
et al. 1991) was used to delete the vlL-8 gene from the SN5 and A6 cosmid DNA.
Briefly, the SN5 and A6 cosmids were incubated with RecA protein, ADP/ATP 73
and two oligonucleotides, vlL-8blkF (5’-GCC CGC ATC TCG CAG CCC CCG GAT
CCG ATC CCG CAG ACC C-3’) and vIL-8blkR (5’-TCC CCT GCT AGC CCT GCC
CTA GGT AAT GCA TTT TAA ATC T-3’), overlapping the two BamHl sites flanking
the vlL-8 sequence (nt. 1,451 to 4,543) to protect these sites from methylation.
The protected cosmid DNAs were methylated with BamHI methylase, denatured
and digested with BamHI to generate SN5/ABamHl and A6/ABamHI. These
cosmid DNAs were treated with CIP (calf intestinal alkaline phosphatase) and

were purified by phenol-chloroform extraction followed by ethanol precipitation.

To introduce the vlL-8 deletion into the SN5 and A6 cosmids, the
pUC19/SN5BamHl/AvlL-8 transfer vector was digested with BamHI and the
SNSBamHl/AvlL-8 fragment ligated to SN5/ABamHl and A6/ABamHl, generating

SN5/AvlL-8 and A6/AvIL-8, respectively.

Similarly, to introduce the ELR motif into the vlL-8 coding sequence, the
pUC19/SN58amHl/ELR transfer vector was digested with BamHI and the
SNSBamHI/ELR fragment ligated to SN5/ABamHl and A6/ABamHl to generate

SN5/vIL-8-ELR and A6/vlL-8-ELR, respectively.

Transfections. Parental, P89, SN16 and B40 and mutant, SN5/AvlL-8,
A6/AvlL-8, SN5/vlL-8—ELR and A6/vlL-8-ELR, cosmid DNA were digested with
Notl and purified by phenol chloroform extraction and ethanol precipitation. To
generate a mutant virus with vIL-8 gene deletion, er5/AvIL-8, 500 pg of each
digested cosmid DNA (P89, SN16, B40, SN5/AvIL-8, and A6/AvlL-8) along with 2
pg of sheared salmon sperm DNA were used to transfect 5x105 DEF in 35mm
dishes by the calcium phosphate method (Morgan et al. 1990). Four days after
transfection, cells were trypsinized, seeded onto a 100 mm dish and monitored
daily for cytopathic effect (CPE). Viral stocks were subsequently made in DEF
for further analysis. A MDV mutant virus carrying the “ELR” motif,
er5/vlL-8-ELR, was generated in a similar method using P89, SN16, B40,

SN5/vlL-8-ELR and A6/vlL-8-ELR cosmid DNA.

Revertant Virus. To generate a revertant virus from er5/AvlL-8 containing

the vlL-8 gene, transfer vector pUC19/SNSBamHl was digested with BamHI and

55

co-transfected onto DEF cells with the purified er5/AvlL-8 viral DNA. After the
CPE was evident, transfected cells were overlayed with 1.25% of Bacto-Agar and
more than 400 viral plaques picked by trypsinization. Cells from each plaque
were divided into two aliquots, one was used to re-infect a fresh 60mm dish of
DEF and the other was used for PCR analysis. Integration of vlL—8 gene into the
er5/AvlL-8 genome was detected by PCR using primers CIaIF: 5’-GGC GCA
GCA CTG AAT AAG CC-3’, and BamHoriR: 5’-GGA GTA ATC TGC GTT-3’ which
would generate a 2200bp and 1400bp fragments in the revertant and deletion

mutant virus, respectively (Figure 7).

Indirect immunofluorescence assay (IFA) and lmmunohistochemistry
(IHC). IFA of cosmid transfected DEF cells was carried out as previously
described (Cui et al. 1988). For IHC, lymphoid organs (thymus, spleen, bursa of
Fabricius) and feather follicles of infected and uninfected chickens were
embedded in O.C.T. (optimal cutting temperature) compound (Sakura Finetek
USA, Inc. CA), immediately frozen in liquid nitrogen and stored at -80°C until
use. Four to eight micron thick cryostat sections of tissue blocks were prepared,
fixed with cold ethanol for 5 minutes and air—dried. lmmunostaining was carried out
using the Vectastain ABC Kit (Vector Laboratories, Inc. CA) according to the
manufacture’s instructions. In IFA, Mab H19 and rabbit polyclonal anti-vlL-8 were
used at 1:300 and 1:50 dilutions, respectively. In IHC, Mab H19 was diluted at

1 :3000.

56

Growth kinetics. Growth kinetics of er5, er5/AvlL-8 (clone 1 and 2) and
er5/vIL-8-ELR (clone 1 and 2) viruses were studied as described (Cohen et al.
1993). Briefly, 100 plaque forming units (pfu) of the different viruses were
inoculated onto DEF cells seeded on 60-mm plates. On days 1, 2, 3, 4 and 5 post
inoculation the infected cells were typsinized, serial dilutions were inoculated on
fresh DEF cells seeded on 35-mm plates, and plaques of different dilutions

counted 7 days post infection.

Southern blot. DNAs from er5, er5/AvlL-8, er5/vlL-8-ELR and
er5/AvlL-8-RV infected or uninfected DEF were isolated as previously described
(Reddy et al. 2002). Five micrograms of each viral DNA were digested with
EcoRI, or BamHl, or double digested with BamHI and Sail. The DNA fragments
were then separated on a 1% agarose Tris-borate/EDTA (TBE) gel and
transferred to nylon membranes. Two individual [32P]dCTP-labeled DNA
probes, one from the total genomic viral DNA (SN5, P89, SN16, A6 and B40
cosmid DNA fragments) and one from the 3.1kb BamHI fragment containing the
vlL-8 gene were generated by random priming and hybridization was carried out

using standard protocols.

Western blot. Supernatants from er5, er5/AvIL-8, er5/vlL-8-ELR and
er5/AvlL-8-RV infected DEF as well as the supernatant from the non-infected
DEF were collected 3 days post infection. Sixteen microliters of each
supernatant were separated on 15% SDS-PAGE gel using standard procedures.

Proteins were then transferred to nitrocellulose membranes, blocked with 5% dry

57

milk for 1hr at room temperature, and probed with rabbit anti-vlL-8 polyclonal
antibody (1:200 diluted in TBS buffer(20mM Tris-HCI (pH 7.5), 0.8%NaCl) at 37°C
for 1hr. After 3 washes with TBS buffer, horseradish peroxidase conjugated goat
anti-rabbit antibody (1 :3000) (Kirkegaard & Perry Laboratories, Inc.) was added to
the blots and incubated at 37°C for 1 hour. Following 3 washes with TBS,

LTM

antibody bound specific antigens were detected by incubating with EC western

blotting detection reagent (Amersham Bioscience, UK) and exposed to X-ray film.

Chicken Chorioallantoic Membrane assay. Recombinant er5 and
er5/vlL-8-ELR viruses were inoculated onto fresh DEF at the same multiplicity of
infection (MCI) and supernatants collected 72 hours post infection. Gelatin
sponges (1 mm3; Upjohn Company Kalamazoo, USA) were absorbed with 100ul
supernatant from each virus and placed on the chorioallantoic membrane (CAM)
of 8 day old embryonating eggs. Phorbol 12-myristate 13-acetate (TPA) (0.1 ug)
was also absorbed to gelatin sponges and used as positive control. Sponges
alone and sponges absorbed with supernatant from non-infected DEF cells were
used as negative controls. On day 12 of embryonation, CAMs were

photographed in ovo.

Pathogenesis studies. Specific pathogen free MD-susceptible F1 progeny
(15x7) of the Avian Disease and Oncology Laboratory line 15l5 males and line 71
females were used in all the studies. These progeny were free of maternal
antibodies against MDV. Chickens were wing-banded at hatch, and randomly

sorted into different experimental groups (17 chickens per group) and held in

58

modified Horsfall-Bauer isolators. Day old chickens were inoculated with 2000
pfu of er5, er5/AvlL-8, er5/vlL-8-ELR or er5/AvlL-8-RV, subcutaneously.
All the chickens that died during the trial or were killed at the end of the
experiment (8 weeks post inoculation) were necropsied and evaluated for tumor

incidence.

Viremia assay. To examine in vivo virus replication and reactivation, 5 birds
from each group were randomly selected and bled at 6, 8, 14 and 35 days post
inoculation. Buffy-coats were obtained by centrifuging at 700 rpm for 5 minutes.
Lymphocytes were then counted and diluted to 106cells/ml. For each chicken
sample, duplicated 35-mm plates of freshly seeded DEF monolayers were
inoculated with both 105 and 106 lymphocytes and plaques were counted 7 days

post inoculation.

59

RESULTS

MDV vlL-8 is dispensable for in vitro replication, latency and in vivo
transmission but important for early in vivo cytolytic infection and
transformation. In order to determine the role of MDV vlL-8 in viral replication
and pathogenesis, er5/AvIL-8 virus was constructed in which the entire coding
sequence of the vIL-8 gene was deleted (Figure 6C). Cosmids SN5/AvlL-8 and
A6/AvlL-8, lacking the entire coding sequence of vIL-8, were transfected, along
with parental SN16, P89, and B40, into DEF and transfected cells observed for
CPE. To confirm the deletion of vlL-8 gene, transfected cells showing CPE were
examined by IFA with Mab H19 (anti-pp38) and rabbit anti-vlL-8 polyclonal sera.
As expected, er5 virus expressed both pp38 and vlL-8 (Figure 8A and B) while
er5/AvlL-8 expressed only pp38 (Figure 80 and D). Similarly, Western blot
analysis of er5 and er5/AvlL-8 infected DEF supernatants indicated that a 18
kDa band, corresponding to vlL-8, was present in supernatants from er5 but
absent in supernatants from er5/AvlL-8 infected cells (Figure 9). To verify that
er5/AvlL-8 had the expected genome structure, Southern blot of er5 and
er5/AvlL-8 genomic DNA digested with EcoRI was performed. As shown in
figure 10A, both viruses showed no detectable difference in pattern of DNA
fragments, suggesting that there were no gross rearrangements in the
er5/AvlL-8 genome (Figure 10A lanes 5-6). In addition, Southern blot analysis
of viral DNA digested with BamHI and probed with the 3.1 Kb fragment purified

from pUC19/SN5BamHI resulted in a 3.1 Kb band in er5 viral DNA (Figure 108

60

lane2) and a 2.3 Kb band in the er5/AvlL-8 DNA, reflecting the deletion of 798

bp spanning the vlL—8 gene (Figure 10B lanes 5-6).

To determine if vlL-8 plays a role in the in vitro replication of MDV in DEF, the
growth kinetics of er5, er5/AvIL-8-1 and er5/AvlL-8-2 were compared with
each other. As seen in Figure 11, viral titers at all time points tested were very
similar for all three viruses indicating that expression of vlL-8 is dispensable for

viral replication in cell culture.

To examine if vlL-8 plays a role in the in vivo viral replication, day old 15x7
chickens were inoculated with er5 and er5/AvlL-8 viruses. Six days
post-inoculation, lymphoid organs (bursa, thymus and spleen) from 3 chickens
from each group were collected and examined for virus replication by
immunohistochemistry. As seen in figure 12, there was a high level of
expression of pp38 in the lymphoid organs of er5 inoculated chickens (Figure
120, E and F). However, there was significantly lower level of pp38 expression
in er5/AvIL-8 infected chickens (Figure 12G, H and I). These results indicated

that vlL-8 gene is important for early cytolytic infection in lymphocytes.

MDV virus transmission takes place by virus replication in the feather follicle
and release of infectious virus in the dander. To examine if vIL-8 is necessary
for virus transmission, viral replication in the feather follicle of chickens infected
with er5, and er5/AvlL-8 were examined 2 weeks post inoculation. As shown
in Figure 13, both er5, and er5/AvIL-8 viruses had similar level of pp38

expression in the feather follicle suggesting that er5/AvlL-8 can be transmitted

61

horizontally like parental MDV. In addition, sentinel chickens housed in the
same isolator as er5/AvlL-8 inoculated chickens developed high titer anti-MDV
antibodies (data not shown), indicating that deletion of the vlL-8 did not affect

shedding of viral particles into the environment.

To examine if vlL-8 affects MDV latency, viral titers in peripheral blood
lymphocytes of infected chickens were tested at 6, 8, 14 and 35 days post
inoculation. As shown in the Figure 14, both er5 and er5/AvlL-8 viruses
reach the highest viral titer 8 days post-inoculation indicating that vlL-8 is neither
involved in latency nor in reactivation from latently infected cells. However, at
both 14 days and 35 days post-inoculation, there were significantly less
reactivated latently infected PBL in er5/AvlL-8 infected chickens (Table 6,
Figure 14). This may be due to reduced number of transformed PBL in
er5/AvlL-8 infected chickens compared to parental er5 at later stages of

infection.

In order to determine if the deletion of vlL-8 affects the pathogenic properties
of MDV, chickens inoculated with er5 or er5/AvlL-8 were examined for gross
tumors and mortality for a period of 8 weeks. As indicated in Table 7, the
incidence of mortality was significantly lower in the group inoculated with
er5/AvlL—8 (4.3%) than in the group inoculated er5 (88.2 %). In addition, in
agreement with the low level of replication of er5/AvlL-8 in lymphoid tissues, no
atrophy of bursa and thymus was observed in this group of chickens compared

with massive atrophy of these organs in the er5 inoculated group (data not

62

shown). Similarly, the tumor incidence in the group inoculated with er5/AvlL-8
was much lower (17.6%) than that observed in the group inoculated with er5
(76.7%). These data, all together indicate that the deletion of the gene vlL-8

significantly decreases the virulence of the recombinant virus, er5/AvlL-8.

To verify that the phenotypic changes observed in the in vivo replication and
pathogenesis of er5/AvlL-8 were only due to the deletion of vlL-8, a revertant
virus, er5/AvlL-8-RV, was generated by cotransfection of er5/AvIL-8 viral DNA
with a transfer vector, pUC19/SN5BamHI, containing the vlL-8 gene. Revertant
viruses were selected by plaque purification and screened for the presence of
vlL-8 gene by PCR (Figure 7). In addition, expression of vlL-8 in supernatant of
DEF infected cells was confirmed by Western blot (Figure 9 lane 1). As seen in
table 7, the pathogenic properties of the revertant er5/AvlL-8-RV virus were very
similar to those of parental virus with regards to mortality (100%) and tumor
incidence (76.7%). These results confirm that vlL-8 plays and important role in

MDV pathogenesis.

Generation and in vitro characterization of an ELR mutant virus,
er5IvlL-8-ELR. Most CXC chemokines present an ELR motif that has been
associated with attraction of neutrophils. However, MDV vlL-8 presents a DKR
motif in place of the ELR motif and attracts mononuclear cells instead of
heterophils (chicken neutrophils). In order to determine the role of the DKR motif
in the biological properties of vlL-8, a recombinant MDV virus, er5/vlL-8-ELR,

was generated, in which the DKR motif was substituted by ELR. Using site

63

directed mutagenesis and RARE cleavage, two cosmid DNA, SN5/vlL-8-ELR and
A6/vlL-8-ELR, were modified, in which the nucleotides CAA were changed to GCT
resulting in two amino acids substitutions in the vlL-8 gene from DK (Asp-Lys) to
EL (Glu-Leu). Interestingly, these mutations resulted in the loss of a Sal I site in
the vlL-8 gene and this feature was used for the selection of recombinant clones
prior to sequencing (Figure 15A). Transfection of DEF with cosmids, P89, SN16,
B40, SN5/vlL-8-ELR and A6/vlL-8-ELR resulted in a recombinant virus with in vitro
growth properties similar to those of wild type er5 (Figure 11). In addition,
Southern blot analysis of EcoRl (Figure 10A, lane3-4) or BamHl digested viral
DNA (Figure 10B, Ian93-4) showed the same DNA pattern as the parental virus
while the BamHI /SaIl double digested DNA (Figure 10B, lane3’-4’) showed the
lost of Sall restriction enzyme site indicating the presence of the introduced

mutations and no other major rearrangements.

Immunofluorescence analysis of er5/vIL-8-ELR infected DEF (Figure 8E and
F) and Western blot analysis of supernatants of these cells (Figure 9, lanes-6)

indicated that vlL-8-ELR was expressed and secreted like parental virus.

Mutant vlL-8 protein carrying the ELR motif induces strong angiogenic
activity in CAM assay. The presence of an ELR motif of CXC chemokines has
been shown to correlate well with their ability to induce angiogenesis in
tumorigenesis (Strieter et al. 1995b). CAM assay has been used to study
angiogenic and anti-angiogenic properties of compounds. In order to examine if a

change from DKR to ELR induces angiogenic activity in vIL-8, gelatin sponges

64

absorbed with 100ul supernatant from er5 and er5/vlL-8-ELR infected DEF
were placed on the CAM of 8 day old embryonating eggs. On day 12 of
embryonation, macroscopic examination showed that sponges treated with 0.1ug
of TPA (positive control) were surrounded by numerous allantoic vessels that
developed radially towards the implant in a ‘spoked-wheel’ pattern (Figure 16A).
A similar picture was produced by sponges loaded with supernatant of virus
er5/vlL-8-ELR infected DEF (Figure 163). However, when sponges were
treated with supernatant from parental virus er5 infected DEF (Figure 16C); no
vascular response was detectable around the sponges, as was the case with

control specimens containing supernatant from non-infected DEF (Figure 160).

ELR mutations have no effect on the pathogenic properties of
er5/vlL-8-ELR. To examine if the DKR to ELR mutation had any effect on the
pathogenicity of er5, er5/vIL-8-ELR was inoculated into day old 15x7 chickens
and its effects in viremia, early cytolitic infection, mortality and tumor induction
were examined. As seen on table 2 and figure 14, chickens infected with
er5/vlL-8-ELR presented similar viremia titers as er5 at all four time points
tested. Similarly, viral replication in lymphoid organs (bursa, thymus and spleen)
determined by pp38 expression at 6 dpi, showed levels similar to those of er5
(Figure 12) indicating that the mutations did not have any effect on the early
cytolytic infection of er5/vlL-8-ELR. In addition, mortality (94.1%) and tumor
incidence (88.2%) induced by er5/vlL-8-ELR were very similar to those
observed with er5 indicating that the mutations had no effect on the pathogenic

properties of the virus.

65

DISCUSSION

In this report, a detailed analysis of pathogenesis of three recombinant MD
viruses, er5/vlL-8-ELR which carries ELR motif in the vlL—8 gene, vlL-8
knockout virus er5/AvlL-8 and its revertant, er5/vlL-8-RV is described. The
use of comid-based strategy significantly facilitated the construction of these
viruses(Reddy et al. 2002). The results demonstrate that vlL-8 is not required for
in vitro replication but plays an important role in in vivo propagation and
pathogenesis. The gross MD tumor incidence caused by the vlL-8 deletion
mutant is reduced to 17.6 % of infected birds (compared to 76.7% for chickens
infected with the parental virus). lmportantly, the revertant completely restores
pathogenecity, conclusively demonstrating the crucial role of vlL-8 plays in this
process. The vlL-8-ELR mutant has a pathogenic pattern similar to that of the
wild type, indicating ELR does not significantly shift the tropism or affect the

angiogenic properties of vIL-8.

While all the mutant viruses replicate equally well on fibroblasts in vitro, the
infection patterns in vivo are quite different. A time-course analysis of infection in
different organs was conducted. MDV induces an early phase of cytolytic
infection in bursa during the first week, which is followed by latency entry and
reactivation, resulting in the second phase of cytolyic infections occurring around
two weeks. It was found that er5/AvlL-8 virus is significantly impaired in the
early phase of cytolytic infections in lymphoid organs (6 days Pl). It is well

documented that reduction or absence of early cytolytic infection correlates with

66

absence or reduced incidence of lymphomas (Calnek 1972; Calnek et al. 1979;
Calnek et al. 1983; Payne et al. 1973). This may account for the low virulence
and lymphoma incidence of the vlL-8 deletion mutant. The role of vlL-8 in the
second lytic infection of MDV was studied at 14 days post infection which
correlates with the beginning of MDV latency. It was showed that er5/AvlL-8
virus replicated well in pheripheral blood lymphocytes and feather follicle
epithelium (FF Es). FFE is the only site for productive infection resulting in
cell-free infectious viral particles, which are transmitted to contact birds. It thus
seems that vlL-8 plays a minor, if any, role in the second lytic infection phase as

well as virus shedding and spreading.

The development of latency in MDV is not fully understood, but it is known that
MDV can be reactivated by DEF cells co-cultivated with lymphocytes isolated
from infected chickens (Calnek et al. 1981), a standard way of measuring viremia
titers. Therefore, viremia titers in chickens can reflect both the degree of virus
reactivation from latency and the number of latently infected cells. It has been
shown that a successful cytolytic infection of B and T-cells is a prelude to
latent-infection and transformation of T-cells. At 8 dpi, both vIL-8 deletion virus
and the parental virus reached a similar peak of viremia titer, indicating that the
vlL-8 gene is not essential for virus reactivation from latency. Nevertheless, at
14 and 35 days post inoculation, the viremia titers were significantly reduced in
the deletion virus inoculated chickens compared to that of the pathogenic viruses.

This data indicated that deletion of the vlL-8 gene compromised virus

67

transformation and reduced the number of transformed cells and was also

confirmed by the lower tumor incidence.

The vlL-8 deletion mutant not only was less pathogenic in chickens, but also
can induce protective immunity, when challenged with the very virulent plus strain,
648A virus (data not shown). Witter showed that a partially attenuated serotype
1 virus can induce protection when challenged with highly virulent viruses (Witter
2001). In this regard, er5/AvlL-8 might be considered a good candidate for the

development of an improved vaccine for MDV.

The ELR mutant, er5/vIL-8-ELR, has both in vitro and in vivo properties
similar to pathogenic parental virus, er5, a finding initially surprising to us. In
mammals, ELR+ CXC chemokines engages CXCR2, a G-protein coupled receptor
expressed in endothelial cells. It is speculated that the presence of ELR in vlL-8
would induce angiogenic activity, thus facilitating tumor growth. The preliminary
study using the chicken chorioallantonic membrane assay (CAM) also showed that
mutation to ELR motif is essential in inducing angiogenic activity by the mutated
vlL-8 protein. It was thus speculated that vlL-8-ELR mutant virus would induce
more aggressive tumors. The fact that it did not suggests that either the
interaction of chicken chemokine or its respective receptor is significantly different,
the tissue distribution of chicken CXCR2 is different from that of mammals, or
infection of MDV induces significant level of cellular angiogenic factors such as
cellular IL-8 or VEGF, such that the contribution by vlL-8 is inconsequential. At

present little knowledge was known about the chicken CXCR3, nor their

68

expression patterns. It is also not known whether vlL-8 is able to trigger signals
necessary for angiogenesis. A possible scenario is that vIL-8 is able to bind
certain receptors and attract target cells, but unlike its cellular counterpart lacks the
ability to trigger intracellular signals required for the proliferation of endothelial
cells. Further investigation is required to sort out these questions. It is, however,
known that MDV infection results in the release of cellular lL-8 homologs, 9E3/CEF
and K60 (Carol Cardona personal communication), both of which contain the
“ELR” motif and at least for 9E3/CEF, its angiogenic effect has been demonstrated
(Martins-Green 2001; Martins-Green et al. 1998). At the same time, ELR
containing chemokines are known to be chemoattractants for neutrophils, which
might reduce virus load and impede tumorigenesis. The results would argue that
the replacement of DKR by ELR does not change significantly its tropism toward
target cells. It is noted that vIL-8 has a significantly long c-terminal domain, which
is also considered important in chemotactic functions. The presence of this
domain may diminish its ability to attract or activate neutrophils, even in the

presence of ELR.

In summary, findings in this study are consistent with a model indicating that
MDV encoded vIL-8 gene is involved in the early phase of cytolytic infections,
presumably through the recruitment of B or T lymphocytes. Deletion of this gene
has little impact on either virus reactivation from latency or virus
spreading/shedding. Impaired early cytolytic infection due to the deletion of vIL-8
leads to weak inflammatory / immune responses, and reduced numbers of

transformed cells, significantly decreased pathogenecity and tumor incidence.

69

IRL TRS

U1.

 

   

B SN5 SN16 B40

 

 

 

SgrAI(42294) SgrAI(67529) SgrAI(lO94l6) BlpI(I4OI88)

 

 

 

P89 A6
Nar 1(35537) PmeI (78402) Ascl(1023l4) SanDI(l45776)
C
BamHl BamHl SNSAvlL-s AaAvIH BamHl BamHI
Z9; 2...:
C10 I NCOI NCOI Clal
(2803) (3605) (138011) (138810)

Figure 6. Construction of recombinant virus with deletion of vIL-8 gene.

A. The MDV genome consists of terminal repeat long (TRL) and short (TRs),
internal repeat long (IRL) and short (IRS), unique long (UL) and unique short (Us)
DNA segments. B. Schematic representation of overlapping clones generated
to reconstitute an infectious virus from a very virulent strain of MDV (Md5). The
restriction enzymes used to generate each cosmid clones and their positions are
indicated. C. Cosmids SN5/AvlL-8 and A6/AvIL-8 have the vIL-8 coding
sequences deleted by Clal and Ncol digestion. The locations of the restriction

enzymes used to introduce the deletions are indicated.

70

A
BamHl fraament

BBEHI Clal Ncol BB'TH'
u—i

‘-‘

 

‘
primer Call F

 

 

 

 

 

B.
er5/AvlL-8
BamHl Ba mHI
’1 Ale-L;8- “
// 4 l4 \\
primer Ca/IF primer BamHoriR
¥ J
C. V
er5/AvlL-8-RV “Wm
B1mHl(1451) ('lul Mal BarnHl(4543)
// A“ \\
// 4 ‘ ‘ 1‘ \Y
primer Call F VIL-8 primer BamHoriR
¥ 4
y.
2200bp:

Figure 7. Generation of er5/AvlL-8 revertant virus by homologous
recombination. (A) 3.1 kb BamHI fragment (1451 -4543) of er5 which
contains the complete vlL-8 coding sequence. (B) BamHI fragment of
er5/AvlL-8 lacking the vIL-8 gene, but the remaining sequences in the BamHI
region are homologous to the BamHI fragment in panel A. Homologous
recombination between the DNA in A and B would generate a revertant virus
er5/AvlL-8-RV (C) which includes a 3.1 kb BamHI fragment (1451-4543)
contains the complete vlL-8 coding sequence. The location of CallF and
BamHoriR primers and the size of the PCR product in each one of the virus is

indicated.

This figure is in color.

71

er5 er5/AvlL-8 er5lvlL-8-ELR

9938
c
VIN-.-
I

Figure 8. lmmunofluorescence staining of parental and recombinant MD

 

 

viruses infected DEF using anti-pp38 and antl-vlL-8 antibodies. PP38 could
be detected in all three viruses infected DEF, er5 (A), er5/AvlL-8 (C), and
er5/vlL-8-ELR (E). The vIL-8 can only be detected in er5 (B) and

er5/vlL-8—ELR (F), but not in the vlL-8 deleted virus er5/AvlL-8 (D).

This figure Is In color.

72

43—}

33_'

18—9

 

Figure 9. Western blot analysis of supernatant from er5/AvlL—8-RV (lane 1),
er5/AvlL-8 (lanes 2 and 3), er5 (lane 4), er5/vIL-8-ELR (lanes 5 and 6) and
mock infected (lane 7) DEF using rabbit sera against vIL-8. The size of vlL-8

protein is about 18kDa.

73

A B

M123456

M 12 34 5 61’2’3’4’5’6’

3.1kb

2.3kb
I .8kb

1.3kb

   

¥_.V__J
BamHl BamHl/Sall
EcoRl digestion double digestion
digestion

Figure 10. Southern blot analysis of DNA Isolated from nucleocapsids of
wild type and recombinant MDV viruses. (A) Viral DNA was digested with
EcoRI and probed with all five radiolabeled cosmids. (B) Viral DNA was digested
with BamHI (lanes 1-6) or double digested with BamHl/Sall (lanes 1'-6’) and
hybridized with the 3.1 kbps MDV BamHI fragment containing the vlL-8 gene.
Lanes: 1 and 1' uninfected CEF; 2 and 2' er5; 3-4 and 3’-4' er5/vlL-8-ELR; 5-6
and 5'-6' er5/AvlL-8. BamHl single digestion produces a 3.1 kb band in both
er5/vlL-8-ELR and er5 viruses and a 2.8 kb fragment in er5/AvlL-8 virus.
BamHl/ Sall double digestion results in two bands (1.8 kb and 1.3 kb) in er5 and

a single band (3.1 kb) in er5/vlL-8-ELR.

74

IOOOOr

1000 ~

  
   

+ erS
-— erS/AvlL-8-1
—*— erS/AVIL-8-2
—¢— erS/vlL-S-ELR-l
+ erS/vIL-8-ELR-2

Numbers of plaques
’s‘

 

 

10 _1_ _L '
0 I 2 3 4 5

Days post infection
Error bars in the figures show the standard deviation of the mean (S.D.).
Figure 11. In vitro growth kinetics of er5, er5IvlL-8-ELR-1,
er5IvIL-8-ELR-2, er5lAvlL-8-1, and er5IAvIL-8-2 viruses. DEF were
infected with aproximately 100 PFU of the indicated viruses and at 1, 2, 3, 4, and
5 days post-infection, the cells were harvested and their titer determined in fresh
DEF after infection and titered on fresh DEF. The experiment was performed in
duplicate, and the titer (logarithm of the mean number of plaque-forming units per
dish) is indicated. Error bars in the figures show the standard deviation of the

mean (S.D.).

This figure Is in color.

75

 

76

Figure 12. lmmunohistochemistry of lymphoid organs (bursa, spleen and
thymus) of 15x7 MDV maternal Ab-negative chickens 6 days after inoculation
with: no virus (A, B, C), er5 (D, E, F), er5/AvlL—8 (G, H. l) er5/vlL-8-ELR (J, K,
L), or er5/AvlL-8-RV (M, N, O) Monoclonal antibody against pp38 (H19) was
used for the staining. Antigen expression in lymphoid organs is severely
impaired only in er5/AvlL-8, showing that vlL-8 is involved in early cytolytic

infection in lymphocytes.

This figure is in color.

77

 

Figure 13. ' L' ‘ L ‘ ' analysis of Feather Follicle Epithelium
(FFE) cells from Inoculated chickens. FFEs were sampled at 2 weeks post
inoculation. All of the recombinant viruses er5 (A), er5/vIL—8-ELR (B), and
er5/AvlL-8 (C) expressed viral antigen in FFEs, indicating that the second lytic
infection is not impaired in either the vlL-8 gene mutations or the vIL-8 gene

deletion. No viral antigen was detected in the control chickens (D).

This figure Is In color.

78

 

Virus Viremia§

 

 

6de 8de 14dpi 35dpi
Mock 0 0 0 0
er5 7.251487 803513372 223251387 183512715
er5/vIL-8-ELR 2.25105 12731336 139164 25291992

er5/AvIL-8 5.125135 128013494 79.751109" 355151.96'"

 

TABLE 6. Virus titer in peripheral blood lymphocytes of er5,
er5IvIL-8-ELR and er5/AvlL-8 inoculated chickens 35 days post
inoculation.

§: The values represent the average of the 5 birds from each inoculated group.

The results shown are the mean value 1 SD.

*2 The value in er5/AvIL-8 inoculated birds is smaller than the other two
viruses inoculated birds. Differences in viremia were compared among the
viruses using the Student's t test. A significant difference in viremia was

observed at p5 0.05 (2 asterisks) and p5 0.001 (3 asterisks).

79

 

 

 

10000 F

U)

o

H

>3

3 1000 r

1:

G.

E

2‘
,9 100 r

G

9-1

1—

‘5.

D 10 b —0—er5

E +er5/AvlL-8

+er5/vIL-8-ELR
l I I I J
6 8 14 35

Days post inoculation

Figure 14. Viral titers, at 6, 8, 14, and 35 days post-Inoculations, In
peripheral blood lymphocytes of 15x7 chickens Inoculated with er5,
er5IvIL-8-ELR, and er5IAvIL-8. Three birds from each experimental group
were tested and titrations performed in duplicate. The titer is indicated as the

logarithm of the mean number.

This figure Is in color.

80

 

Virus‘I Mortality Tumor incidence (%)

 

Mock 0/17 (0) 0/17 (0) '°

er5 15/17 (88.2) 13/17 (76.7)
er5/AvlL-8 1/17 (4.3)“ 3/17 (17.6)'
er5/vlL-8-ELR 16/17 (94.1) 15/17 (88.2)
er5/AvlL-8-RV 17/17 (100) 13/17 (76.7)

 

TABLE 7. Comparison of the pathogenecity of er5, er5IAvlL-8,

er5leL-8-ELR and erSlAvIL-B-RV in 15x7 antibody negative chickens.§

§: This experiment was repeated two times separately.
'°: Data in parentheses indicate percentage of affected chickens.

1': All chickens were inoculated with 2000 PF U of the indicated viruses.
*: Student's ttest (p 5 0.001) analysis indicated that these values were

significantly different from the other groups.

81

Sall

l

G AGT CTC GCT GTC GA'C AAG AGG TGC AAG TGC G
s L A v 1) K R c K c
Primer 1: 5’ c AGT CTC GCT GTC GAG crc AGG TGC AAG TGC G 3’
s L A v 1: L R c K c
Primer 2: 5’ c GCA C'I'I‘ GCA CCT CAG crc GAC AGC GAG ACT c 3’

 

 

 

 

3. I kb
2.7kb
I.8kb

1.3kb

 

Figure 15. Mutation of DKR to ELR motif in the MDV vIL-8 gene.

A. Wild type DNA sequence of the DKR motif. Primers 1 and 2 were used to
introduce 3 nucleotides changes in the vlL—8 coding sequence, which results in
the loss of the Sall restriction endonuclease site and a change from DKR to ELR.
B. BamHI and Sall double digestion of ELR mutant (pUC19/SN5—ELR) (lanes
1-3) and wild type (pUC19/SN5) (lane 4) transfer vectors confirms the absence of
Sall site in the ELR mutant vector. Mutagenesis results in the loss of Sall RE
site, and in turn changes the digestion pattern of the plasmid DNA. Plamid
vector (lane 5) alone is used as control. The size of the restriction fragments

obtained is indicated on the right in kb.

This figure is in color.

82

 

Figure 16. Angiogenic effect of supernatants from uninfected (A) and er5
(B) and er5/vlL-8-ELR (C) Infected DEF. Chorioallantoic membranes (CAMs)
of 10-day-old chicken embryos were incubated for 5 days with sponges absorbed
with the indicated supernatants. Supernatant from er5/vIL-8-ELR infected cells
showed a “spoked-wheel" patterned neovascularization reaction similar to the
TPA positive control (D). Supernatants from er5 infected and uninfected DEF

produced no neovascularization reaction.

This figure Is In color.

83

CHAPTER THREE

AN ATTENUATED VlL-8 DELETION MAREK’S DISEASE VIRUS (MDV)
MUTANT CONFERS PROTECTION AGAINST CHALLENGE WITH A VERY

VIRULENT STRAIN OF MDV

ABSTRACT

Marek’s disease virus (MDV) encoded vlL-8 gene, a chemokine-like gene,
was studied by using a recombinant virus, er5/AvIL-8, in which both copies of
the vlL-8 gene were deleted. Deletion of the vlL-8 gene attenuated the virulence
of the virus as demonstrated by the pathological studies in chickens. Both the
MD maternal Ab" and Ab’ chickens inoculated with er5/AvlL-8 developed a
much lesser number of gross lesions compared to the parental virus er5.
Similarly, the mortality and tumor incidence in groups of chickens inoculated with
er5/AvlL-8 were also significantly lower than er5 inoculated chicken groups.
The revertant virus, er5/AvlL-8-RV, which restored the vlL-8 gene back in the
er5/AvlL-8 viral genome, also restored the pathological phenotype as the
parental virus. With the successful generation of revertant virus, one can make a
more convincing explaination of the in vitro and in vivo phenotype characteristics

of vlL-8 deleted virus.

Interestingly, the er5/AvIL-8 virus also had the ability to protect chickens
against challenge with a MDV very virulent plus (w+MDV) strain, 648A. This

finding provides new optimism regarding the generation of recombinant virus

84

vaccines for control of MDV. Using the flow cytometry assay, the T and B
lymphocytes populations were studied along with the splenocytes of virus
inoculated chickens. Compared to the non-infected chickens or chickens with
the inoculum of er5/AvlL-8, the numbers of B lymphocytes in the er5
inoculated chickens were dramatically reduced, while the numbers of activated
T-cells were increased at 7 DPI. The loss of B-cells may due to cytosis or
apoptosis resulting from the effective cytolytic infection of the er5. The early
cytolytic infection resulted in stimulation of the immune response and activation of
T-cells to become the next major target cells. With the deletion of vIL-8, the early
cytolytic infection was impaired, therefore, the number of B-cells remained similar
to the non-infected chickens, and activation of T-cells was also reduced. Since
vIL-8 is a chemokine-like gene, it is also reasonable to speculate that the
presence of vlL-8 may be functioning as a chemoattractant for B-cells and
activated T-cells in assisting MDV infection and replication. A hypothesized

model for the possible role of vlL-8 in the pathogenesis of MDV is proposed.

85

INTRODUCTION

MDV is classified as a member of the family Herpesviridae (Roizman 1992b).
Because of its lymphotropism, which is similar to Epstein Bar virus (EBV), MDV
was originally classified as a y—herpesvirus (Roizman 1981). However, later on,
based on its genomic organization and significant sequence similarity to other
a—herpesviruses, such as herpes simplex virus (HSV) and varicella-zoster virus
(VZV), it was classified in the subfamily Alphaherpesvirinae (Buckmaster et al.
1988). The MDV group was divided into three serotypes (Biilow 1975a):
Pathogenic strains of MDV and their attenuated derivatives are the prototype
viruses of the MDV group, and are designated as serotype 1 MDV (MDV-1). The
nonpathogenic strains of MDV are designated serotype 2 MDV (MDV-2) and HVT,

a nonpathogenic herpesvirus of turkeys is grouped into serotype 3 MDV (MDV-3).

The pathogenesis of MDV is divided into four distinct phases, the early
cytolytic infection, a second lytic infection, latent infection, and transformation.
The molecular mechanisms of all these phases have not yet been fully
understood. MD incidence has largely been controlled by vaccination with all
three serotypes of MDV, often in bi- and multivalent combinations (Witter 1991,
2001). However, there is continuation of an apparent evolutionary trend of MDV
towards greater virulence which has resulting in increased losses from MD in
vaccinated flocks (Witter 1997). The attenuated MDV-1 strain virus, CVI988, is
currently the most protective vaccine in use. If the MDV evolutionary trend

continues, vaccine may fail even with CVI988. The live virus vaccines for MDV

86

are efficient in controlling disease by protecting chickens against tumor induction
and mortality, but do not prevent viral replicating or shedding. To effectively
control this disease, understanding the mechanisms of the virus replication and

virus interaction with the host cells is the key.

Recently, the successful establishment of MDV cosmid clones has provided a
molecular tool to introduce targeted mutations into the genome of a pathogenic
strain of MDV. This recombinant technology will be extremely helpful in the
exploration of the mechanisms of MDV pathogenesis and identify the roles of
individual genes in the four distinct pathogenesis phases. Based on this
technique, a recombinant MDV, er5/AvlL-8 was generated. And using
homologous recombination, the revertant virus that restored the vIL-8 gene in the
er5/AvlL-8 viral genome was also successfully generated and provided
comparative information to the er5/AvlL-8 and parental virus. It is known that
vlL-8 is involved in the early cytolytic infection and transformation, but not in
latency or virus reactivation from latency. The absence of vIL-8 does not impair
virus spreading and shedding either. In this study, the pathologic studies of
er5/AvlL-8 and er5/AvIL-8-RV were carried out. The results demonstrated
that the virulence of er5/AvlL-8 was attenuated and for the first time in marek's
disease, it is reported that an attenuated recombinant virus has the ability to

protect chickens against challenge from w+MDV strains.

87

MATERIALS AND METHODS

Antibodies. Phosphoprotein 38 (pp38) specific Mab, H19 (Cui et al. 1990),
and rabbit antiserum against MEQ (Lucy Lee, Avian Disease and Oncology
Laboratory, East Lansing, MI.) were used in this study. Monoclonal antibodies
(Mab) CT4 and CTLA3 reacts specifically to chicken CD4 and CD8 respectively
(Southern Biotechnology Associate, lnc., Birmingham, AL) were used to detect
the transformed cell type of the tumors. Mab anti-chicken MHC class l-R-PE
(R-phycoerythrin), Mab anti-chicken class ll-R-PE, Mab anti-chicken CD3-FITC
and Mab anti-chicken Bu1a/1b-FITC (Southern biotechnology associate, Inc,

Birmingham, AL) were used in flow cytometry experiments.

Cells and viruses. Primary duck embryonic fibroblasts (DEF) were used for
virus propagation and DNA transfections. Both the MDV vaccine strain CVI988
and challenge virus 648A were kindly provided by Richard Witter, Avian Disease
and Oncology Laboratory (East Lansing, MI). Recombinant viruses were
generated from cosmids derived from a very virulent MDV strain, Md5 (Reddy et

al. 2002).

Chickens. Chickens used in the study were specific pathogen free (SPF)
MD susceptible F1 progeny (15x7) of Avian Disease and Oncology Laboratory
line 15l5 males and line 71 females. All the chickens were wing-banded at hatch,
and randomly sorted into different experimental groups (17 chickens per group)

and held in modified Horsfall-Bauer isolators.

88

Pathogenesis studies. Chickens from either unvaccinated breeder hens
free of antibodies to all three MDV serotypes (Ab-negative) or from breeder hens
vaccinated with all three serotypes of MDV (Ab-positive) were used in these
studies. Day old chickens were inoculated with 2000 plaque forming units (PF U)
of er5, er5/AvlL-8, or er5/vlL-8-RV, subcutaneously. Chickens that did not
receive any inoculum were used as negative controls. Chickens were raised in
isolators for 8 weeks, each group comprised of 17 chickens. The chickens in Ab'
groups were observed daily for signs of transient paralysis. All the chickens that
died during the trial or were killed at the end of the experiment (8 weeks after
inoculation) were necropsied and evaluated for gross and histological lesions and
mortality. Lymphoid organs and feather follicles were collected in a time course
(4, 6, 14, 35 DPI) for the expression kinetics of MDV viral proteins (pp38 and Meq)
by immunohistochemical staining. Tumor tissues were also collected freshly at
termination by embedding in O.C.T. (optimal cutting temperature) compound
(Sakura Finetek USA, Inc., CA) and immediately frozen in liquid nitrogen.
lmmunohistochemical staining of frozen sections were carried out to identify the

phenotype of the transformed cells.

Immunohistochemical staining. Lymphoid organs (thymus, spleen, bursa
of Fabricius) and feather follicles of infected and uninfected chickens were
collected and embedded in O.C.T. compound, immediately frozen in liquid
nitrogen and stored at -80°C. Four to eight micron thick cryostat sections of
tissue blocks were mounted on poly-L-lysine (Sigma Diagnostics, St. Louis,

MO)«coated sides and fixed for 5 minutes in ethanol, air dried at -20°C for at least

89

30 minutes, and stored at -80°C until staining. An avidin-biotin-peroxidase
complex (ABC) method (Vectastain® ABC kit, Vector Laboratories, Inc.,
Burlingame, CA) was used for immunohistochmistry. Samples were equilibrated
for 15 minutes in isotonic phosphate-buffered saline (PBS), pH 7.4. Between
each remaining step, sections were washed three times for 10 minutes each in
PBS. Sections were preincubated for 20 minutes with normal blocking serum
(provided with the kit) to decrease nonspecific background staining. Samples
were then incubated at room temperature with primary antibodies in an
appropriate concentration for 30 minutes. Mab H19 (Cui et al. 1990) specific to
antigen pp38 was used at a working dilution of 123200, rabbit antiserum against
Meq was used at a working dilution of 112500. Mab CT4, which is specific to CD4
and reacts with T-cell subsets expressing CD4, and Mab CTLA3, which reacts
with the T-cells subsets expressing CD8 were diluted at theconcentration
recommended by the manufacturers (Southern Biotechnology Associate, lnc.,
Birmingham, AL) to examine the MDV tumor cells. A biotinylated secondary
antibody (provided with the kit) was incubated with the slides for another 30
minutes at room temperature after the primary antibody incubation. The working
concentration of the secondary antibody was also used according to
manufacturers’ Instruction. Following a PBS wash, the sections were incubated
for 30 minutes with the ABC. The immunohistochemical reaction was visualized
following incubation with a solution of hydrogen peroxide and 3,
3’-diaminobenzidine (DBA) (Vector® SK-4100, Vector Laboratories) for 7 minutes.

All sections were then lightly counterstained with Gill’s hematoxylin N02 (Fisher

90

Scentific, Fair Lawn, NJ), dehydrated, and mounted in DPX mountant (BDH

Laboratory Supplier).

Flow cytometry assay. Antibody negative chickens (line 15x7) were
inoculated with 2000 PFU of parental virus er5 or mutant virus er5/AvlL-8.
Uninoculated chickens were used as negative controls. On 4 and 7 days post
inoculation, splenocytes from 5 chickens from each group were examined for the
presence of activated T-cells (MHC class ll*/CD3*) and B-cells (MHC class
l‘lBu1a”/1b"). To determine the number of the activated T-cells, 107 splenocytes
were double stained with mouse anti-chicken MHC class ll-R-PE (red) and mouse
anti-chicken CD3-FITC (green). The numbers of B-cells were determined by
double staining splenocytes with mouse anti-chicken Bu1a/1b-FITC (green) and
mouse anti-chicken MHC class-l-R-PE (red). After 1 hourrincubation at 40°C,
the cells were washed three times with PBS and analysed by Fluorescent
Activated Cell Sorter, BD FACSCalibur system (BD Bioscience, San Jose, CA).
The significance of cell population between each group was analyzed by

student-t-test. Difference were considered to be significant when p< 0.05.

Protection experiments. The protection efficacy of er5/AvIL-8 was
compared to that of vaccine virus CVI988. A w+MDV strain, 648A, was used as a
challenge virus. MDV maternal Ab‘ chickens were used for protection

experiments.

Groups of 17 chickens were inoculated subcutaneously at 1 day of age with

2000 PFU per chicken of vaccine virus. Two groups of chickens were inoculated

91

with the test vaccine strain, er5/AvlL-8. One group of chickens was inoculated
with vaccine strain CVI988. The vaccinated groups and one unvaccinated virus
control group were challenged with very virulent plus MDV strain, 648A, 500 PFU
per chicken subcutaneously at 5 days post-vaccination. One more group of

chickens was set as control with no vaccination and no challenge.

Chickens that died during the experiment or were killed at the termination at 8
weeks post-challenge were examined for gross MD lesions, which include

enlarged peripheral nerves, visceral lymphomas, and bursal/thymic atrophy.

MD responses were expressed as a percentage (% MD) based on the
chickens with MD lesions divided by the number of chickens at risk and multiplied
by 100. The percentage protection (% Protection) was calculated as the % MD
in unvaccinated challenged controls minus the % MD in vaccinated challenged
group divided by the % MD unvaccinated challenged controls multiplied by 100.
Data pooled from the two replicates were also summarized for % MD and %

Protection.

92

RESULTS

The recombinant mutant virus, er5IAvlL-8, has an attenuated virulence.
The pathological lesions induced from the mutant viruses (er5/AvlL-8 and the
revertant virus er5/AvIL-8-RV) was compared with the parental virus er5 in
both MDV maternal Abiand Ab" chickens (Table 8, 9) by evaluation for transient
paralysis (TP), gross lesions, mortality and microscopic lesions. During the 8
weeks post inoculation with virus er5/AvlL-8, no mortality was observed in the
antibody positive chickens and only one bird died in the antibody negative group,
while in the groups inoculated with either the parental virus er5 or the revertant
virus er5/AvlL-8-RV, high mortality was observed in both Ab' chicken groups
(88.2-100%) and Ab+ chicken groups (64.7%). The incidence of nerve lesions
and visceral tumors in er5/AvlL-8 inoculated groups was also much lower in
both Ab' chickens (17.6%) and Ab” chickens (5.9%) compared to the groups
inoculated with parental virus er5 or er5/AvlL-8-RV (76.7-88.2%). It was also
observed that none of the er5/AvIL-8 inoculated chickens developed TP in Ab‘
chickens. There was severe atrophy in the bursa of Fabricius and thymus in
er5 and er5/AvlL-8-RV inoculated chickens. However, no atrophy was
observed in chickens inoculated with er5/AvlL-8. Based on the results of gross
lesions, the deletion of the vlL-8 gene significantly attenuated the virulence of the
recombinant virus, er5/AvIL-8. Nevertheless, it was observed that er5/AvlL-8
still induced microscopic lesions in nerves and visceral organs (Figure 17). The
lesions consisted of a mixed population of small, medium, and large lymphoid

cells including plasma cells and large anaplastic lymphoblasts.

93

Phenotype of the target cells for transformation is not affected by vIL-8.
During the pathological studies, only one bird developed gross tumor in the group
of chickens inoculated with er5/AvlL-8. The populations of transformed cells in
the gross tumor were characterized by immunohistochemical staining with
specific T-cell markers. Tumors developed from er5 inoculum were used as
controls. The immunohistochemical results revealed that both er5 and
er5/AvlL-8 virus transformed CD4’ICD8' subset of T-cells (Figure 18). The
results suggested that the phenotype of the tumor cells was not influenced by the

absence of vlL-8.

Deletion of vlL-8 gene delayed MEQ expression and impaired pp38
expression in lymphoid organs. A time course study on the expression of
MDV viral proteins Meq and pp38 in the er5/AvlL-8 inoculated chickens was
compared to those in the parental virus er5 inoculated chickens (Table 10). In
the er5 inoculated chickens, pp38 expression was detected from 4 to 35 DPI in
lymphoid organs, with peak detection at 6 DPI. In the FFEs of er5 inoculated
chickens, pp38 expression was positive from 6 DPI and reached the peak at 14
DPI. Compared to er5, er5/AvlL-8 inoculated chickens had almost no
detectable pp38 expression in the lymphoid organs while a normal level of pp38
expression was detected in the FFEs at 14 DPI. This data shows that deletion of

vlL-8 gene impairs the expression of pp38 in lymphoid organs, but not in the FFE.

The expression of Meq in the er5 inoculated chickens was detected from 6 to

35 DPI in both lymphoid organs and FF Es. No Meq expression was detected in

94

er5/AvlL-8 inoculated chickens at 4 and 6 DPI, however, Meq expression was
detected later at 14 and 35 DPI in both lymphoid organs and FFEs of er5/AvlL-8
inoculated chickens. The numbers of Meq positive cells in the tissue of
er5/AvlL-8 inoculated chickens were comparatively lower than that in the er5

inoculated chicken tissues.

The nucleus morphology of Meq positive cells was changed from single lobed
at 6 DPI, to multi-lobular like pattern at 14 DPI and later (Figure 19). The same
multi—lobular like staining pattern was observed in er5/AvlL-8 inoculated
chickens at 14 DPI and later. This indicated that at 14 DPI or later, MDV infected
cells were undergoing transformation (Liass 1985; Russo et al. 1988). The

er5/AvlL-8 induced a lower level of transformation.

Change in lymphocyte population following early cytolytic infection with
er5 and er5IAvIL-8. The proportions of activated T-cells and B-cells in the
splenocytes of virus inoculated chickens were analyzed using the FACS. The
data showed significant differences between chickens that received er5
inoculum and chickens inoculated with er5/AvIL-8 at 7 DPI but not at 4 DPI.
The number of B-cells dramatically decreased at 7DPl in er5 inoculated
chickens because of a robust early cytolytic infection. However, the number of
B-cells in the er5/AvlL-8 inoculated chickens remained significantly higher than
er5 inoculated chickens and no significant difference to the non-infected group
indicating that vlL-8 is essential for a robust early cytolytic infection (Figure 20).

The number of activated T-cells was increased at 7 DPI in er5 inoculated

95

chickens, but not in the er5/AvlL-8 virus inoculated chickens or the non-infected
groups. Activated T-cell was the other target cells for MDV infection. Deletion
of vlL-8 reduced the number of the activated T-cells indicating that vlL-8 plays an
important role in the activation of T-cells, which is critical for establishment of

MDV infection.

Recombinant virus er5IAvlL-8 protected chickens against very virulent
MDV. The protective efficacy was compared between er5/AvIL-8 and strain
CVI988. Two replicates were conducted in the experiment. The MD responses
and percentage of protection for each group were evaluated in each replicate and

summarized (Table 11).

The results showed that chickens vaccinated with er5/AvlL-8 and challenged
with 648A had a lower percentage incidence of MD than that of CVI988
vaccinated chickens. As shown in Table 11, with vaccination with er5/AvlL-8,
there was only 23.5% of MD incidence in replicate 1 and 17.6% MD incidence in
replicate 2, while vaccination with CVI988 the MD incidence was 47.1% and
41.2% in replicate 1 and 2, respectively. Without vaccination, 648A induced a
100% incidence of MD in both replicates. So, based on MD incidence, the
percentage of protection of er5/AvlL-8 was 76.5% in replicate 1 and 82.4% in
replicate 2, and the percentage of protection with CVI988 was 52.9% in replicate 1
and 58.8% in replicate 2. Data pooled from two replicates were summarized in

Table 11 and demonstrated that er5/AvlL-8 induced about a 79.4% protection

96

index against 648A, and CVI988 induced about a 55.9% of protection index

against 648A.

97

DISCUSSION

This study reports the pathological characterization of the attenuated
recombinant vlL-8 deleted MDV, er5/AvlL-8, and its ability to protect chickens

against challenge with a w+MDV, 648A.

Since the first report of reconstitution of MDV with cosmid clones, only the
pp38 gene deleted virus was studied and reported (Reddy et al. 2002). This
vlL-8 deleted virus, er5/AvlL-8, is the first cosmid clone reconstituted mutant
virus characterized by detailed in vivo studies. Results presented here from
experiments in both Ab“ and Ab‘ chickens indicate that the virulence of
er5/AvlL-8 was attenuated and caused a much lower incidence of gross lesions
and mortality. Moreover, a revertant virus that restored the vIL-8 gene back to
the er5/AvIL-8 viral genome was also successfully generated. The
successfully restored pathological phenotype of the revertant virus strongly
assisted in demonstrating that the impairments in er5/AvlL-8 pathogenicity

specifically resulted from the deletion of the vlL-8 gene.

In order to investigate the interaction between vIL-8 and other viral proteins in
vivo, lymphoid tissues and FFEs from er5/AvlL-8 inoculated chickens were
examined for pp38 and Meq expression level in a time course study with IHC and
compared to that of er5 inoculated chickens. The results demonstrated that
deletion of vlL-8 impaired the expression of pp38 in lymphoid organs, but not in
FFEs. Furthermore, deletion of vlL-8 delayed Meq expression in vivo, and the

number of Meq positive cells was also reduced in comparison to that of er5

98

inoculated chickens. Meq is expressed in the nucleus, from 6 DPI to 14 DPI, and
later. The multi-Iobular—Iike nuclear staining pattern of Meq-positive cells at later
stages of MDV infection suggested that the cells were undergoing transformation.
Although MEQ positive stained cells were detected in er5/AvlL—8 inoculated
chickens at 14 DPI with the same multi-lobular like nucleus, the percentage of
positive cells in each sample was lower than the er5 inoculated chickens. The
pathogenic studies also showed that er5/AvIL-8 inoculated chickens developed
a lesser number of gross tumors. Collectively, the results may indicate that vIL-8
gene is involved in pathogenecity and transformation of the disease. Moreover,
neither pp38 nor Meq expression in lymphoid organs at 6 DPI confirmed again the

er5/AvlL-8 induced impairment of early cytolytic infection in lymphoid organs.

B-cells and activated T-cells are the major target cells for MDV infection and
replication. As the primary target cells, infected B-cells at the early cytolytic
infection phase will either undergo apoptosis or cytosis, which will induce early
immune responses and release cytokines and other immune factors such as
lFNay, Nitric Oxide (NO) etc, to activate resting T-cells (Schat et al. 2000). The
activated T-cells then become the major target cells. It was also observed that
deletion of vlL-8 resulted in a reduction in the number of B-cells infected and also
a reduced number of activated T-cells. It was demonstrated earlier that
er5/AvIL-8 had impaired early cytolytic infection, which could result in reduced
infection of B -cells, followed by less apoptosis or cytosis, and therefore induce a
weak immune response to activate T-cells. Furthermore, as a chemoattractant,

it is also reasonable to speculate that vIL-8 may function to attract target cells

99

 

 

(B-cells and activated T-cells) and facilitate virus infection and dissemination.
Although it has been reported that in vitro over-expressed vlL-8 protein attracted
mononuclear cells, further studies are necessary to characterize the specific

mononuclear cell type.

Calnek has established a model for the sequential events during the
pathogenesis of MDV infection (Calnek 1985). Based on our studies of the vIL-8
gene, a modified model of MD pathogenesis is proposed with reference to the
function of vlL-8 (Figure 21). Specifically, during the early phase of MDV
replication (4-7 DPI), vlL-8 is critical in attracting B-cells for MDV infection and
replication, and consequently, vlL-8 gene functions in activating the resting T-cells
to become MDV susceptible activated T-cells and attract activated T-cells to the
infection sites. Similarly, the MDV infected activated T-cells secrete large
amounts of vlL-8 which further aid in activating and attracting more T-cells to

establish a robust early cytolytic infection.

The other important finding in this study is that this recombinant virus has the
ability to protect chickens against challenge with w+MDV strain, 648A. The
protection index was even higher than what was observed with the commercial
vaccine strain, CVI988. The mechanism of how the MDV vaccine protects
chickens against challenge with a highly virulent virus strain is not quite clear, but
Calnek et al. have speculated that vaccination could alter the pathogenesis of
MDV infection by severely curtailing or eliminating the early cytolytic phase, thus

decreasing the number of target cells for the establishment of latency or

100

transformation (Calnek et al. 1997). As it was proved in this study and the earlier
work reported in this dissertation, deletion of vlL-8 gene resulted in depletion of
viral antigen expression in the lymphoid organs at 6 DPI, and no bursa/thymus
atrophy in Ab‘ chickens during the experimental trial. So it is reasonable to
conclude that er5/AvlL-8 could act like the vaccine virus strains in impairing the
early cytolytic infection, and therefore block the latent infection and transformation
of T-cells and protect chickens. This is the first report of a recombinant serotype

1 MDV vaccine.

Attenuation of serotype 1 MDV vaccines was achieved by virus passages in
cell culture. In order to adapt to the in vitro growth environment, the virus would
introduce mutations in the genome resulting in attenuation of virulence. The
mutations happened randomly in certain viral copies but not all. After several in
vitro passages, attenuated virus in fact consists of a mixture of mutant and intact
viral genomes. Witter recently reported that partially attenuated serotype 1 MDV
vaccines induce stronger protection (Witter 2002). The disadvantage of using
partially attenuated serotype 1 MDV vaccine is also apparent. The copies of
intact viral genome in the mixture of virus have the advantage to replicate in vivo,
so after backpassage of the attenuated virus in the chickens for several
generations, the virus will gain back the virulence and cause disease. Because
the attenuated recombinant virus such as the er5/AvlL-8 was produced from a
pure clone, backpassage in chickens would not change the virulence of the virus.
So, it is likely that recombinant virus vaccines will possibly become the next

generation of MD vaccines in the near future.

101

 

 

Virus Transient Gross Lesions§ Microscopic Total
Inoculum Paralysis Nerves Visceral Lesions Mortality
None 0/17(0)° 0/17 0/17 0

erS 12/17(70.6) 13/17(76.7) 10/17(58.8) NT1| 15/17(88.2)
erS/AvIL-S 0/17(0) 1/17(4.3) 2/17(13.3) 9/17(52.9) 1/17(4.3)
erS/AvlL-B 11/17(64.7) 13/17(76.7) 11/17(64.7) NT 17/17(100)

-RV

 

Table 8. Pathological lesions in MDV maternal antibody negative chickens

inoculated with er5, er5lAvlL-8, and er5/AvlL-8-RV viruses.

‘Data in parentheses indicate percentage of chickens.

1'Not all chickens were examined by histopathology. NT: Not Tested.

§Gross lesions include nerve enlargement and visceral tumors.

102

 

 

Virus Gross Lesion§ Microscopic Total

 

inoculum Nerve Visceral Lesions Mortaliy
None 0 0 0 0

erS 13/17(76.5). 6/17(35.3) 6/6(100) 11/17(64.7)
erS/AvIL-8 O/17(0) 1/17(5.9) 9/17(52.9) 0/17(O)

erS/AvlL-S-RV 15/17(88.2) 7/17(41 .2) 4/4(100) 13/17(76.5)

Tabel 9. Pathological lesions in MDV maternal antibody positive chickens
inoculated with recombinant er5, er5/AvIL-8, and er5/AvIL-8-RV

viruses.

*Data in parentheses indicate percentage of chickens.

§Gross lesions include nerve enlargement and viceral tumors.

103

 

 

104

Figure 17. Histopathological lesions derived from infection with the
recombinant viruses. The deletion virus er5/AvlL-8 induced a much lower
frequency of gross lesions, but histological lesions were present in various organs
(Ovary, testis, kidney, liver). A mixed population of infiltrating small, medium,

and large lymphoid cells infiltration replaced the normal structure of the tissue.

This figure is in color.

105

 

er5 er5IAVIL-8

 

CD8

   

Figure 18. “ ‘ L ’ ' staining of tumor tissues derived from all
three recombinant viruses using a Mab against chicken CD4 and 008.
Cells in the tumor tissues derived from er5 (A, B), and er5/AvlL-8 (C, D)
infected chickens express CD4 (A, C) but not CDB (B, D). The transformed cell

types in both viruses are the CD4*/CD8' subset of T cells.

This figure Is In color.

106

 

pp38 MEQ

 

 

LO FFE LO FFE

erS

4 DPI + - - -

6 DPI ++++ + +++ +

14 DPI +++ ++++ +++ +++

35 DPI +++ +++ ++++ +++
erS/AvIL-S

4 DPI - - - -

6 DPI - - - -

14 DPI - ++++ ++ ++

35 DPI - +/. ++ ++

 

Table 10. In vivo expression kinetics of the viral protein pp38 and MEQ.
With parental virus er5, the expression of viral protein (pp38) in lymphoid
organs (L0) is present at all time points examined (4, 6, 14, and 35 DPI), whereas
there was no pp38 expression in er5/AvlL-8 inoculated chickens. pp38
expression in FFE (feather follicle epithelium) was similar to that observed in er5
inoculated chickens. MEQ expression in er5/AvlL-8 inoculated chickens was
delayed in both lymphoid organs and FFE until 14DPI compared to er5, and the

level of Meq expression was lower in er5/AvlL-8 inoculated chickens.

Note: the level of viral protein expression was scored as negative (-), low (+),

medium (++), high (+++), very high (++++), minimal to none (+/—).

107

er5 14 DPI

 

er5/AvIL-8 6 DPI er5/AvlL-8 14 DPI
Figure 19. ' L’ ‘ L ’ ' staining of spleens using antibody
against MEQ. In the MDV infected cells, Meq has a nuclear localization. At 6
DPI with parental virus er5, during the early cytolytic infection phase, MEQ
positive cells appear to have a single lobe staining pattern in the nucleus (A),
while at 14 DPI, the Meq positive cells have multi-lobular-like staining pattern in
the nucleus (B). No viral expression was detected in the er5/AvlL-8 virus
inoculated chickens at 6 DPI (C), but MEQ was expressed at 14 DPI in

er5/AvlL-8 inoculated chickens (D).

This figure Is In color.

108

 

B Cells

Percent
N
D

 

T Cells

Non-infected
III er5/AvIL-8
I er5

Percent

 

4 DPI YDPI

Figure 20. Flow cytometry analysis of spleen cells from virus infected
chickens. The parental virus, er5 (black) destroyed large numbers of B-cells
in chickens at both 4 and 7 DPI, and induced a higher level of activated T-cells.
The levels of B-cells and activated T-cells in chickens infected with the deletion
virus er5/AvlL-8 (white) have no significant difference to the non-infected

control (strip) chickens.

‘ = p<0.05 between indicated group and the others.

109

 

 

 

 

Vaccine Challenge" Replicate 1 Replicate 2 Summary
lesions MD% P119. lesions MD% PI% MD% PI%
None - 0 0 0 O 0
erS/AVIL-S - 0/17 0 0/17 0 O
erS/AvIL-8 648A 4/17 23.5 76.5 3/17 17.6 82.4 20.6 79.4
CVI988 648A 8/17 47.1 52.9 7/17 41.2 58.8 44.1 55.9

- 648A 17/17 100

17/17 100

 

Table 11. Comparative protective efficacy of vlL-8 deletion virus in MD

maternal antibody positive chickens.

* Challenge was done at 6 days of age with 500 PFU of 648A.

110

 

MDV Infection

4— vIL-8

/' lFN-y,\N‘

 

 

    
  

MDV "l
Replicatio

   
 

MDV
Replicatlo

 

V

 

 

 

Apoptosis Apoptosis TumorIQGHGSIS
L 1 ,
I 4-7 DPI l J I i
. . >l-2 week post inoculation ' >3-4week post infection
Early cytolytic infection and Latent infection Iymphoproliferative
immune responses lesion in nerves and
visceral organs

111

Figure 21. The possible roles of vIL-8 in the sequential pathogenesis of
MDV infection. MDV infected cells are phagocytized by macrophages followed
by infection of B-cells. MDV replicates in the infected B-cells and initiate an early
cytolytic infection. Some of these B-cells undergo apoptosis. Infected B-cells
may induce an early immune response, release other cytokines and chemicals

such as IFN-y or NO, which activate the T-cells.

Only the activated T-cells (aT) but not the resting T-cells are the target cells for
MDV infection. Infected aT-cells express viral antigens, such as vlL-8, which
results in attracting more B and aT-cells for virus infection. Some of the aT-cells
undergo apoptosis, some aT enter into the latency (aT-Latency) phase or become

transformed (aT-Transformed), finally leading to the development of tumors.

This figure is in color.

112

 

CHAPTER FOUR

DISCUSSION AND FUTURE RESEARCH

DISCUSSION

As a member of the alpha herpesvirus family, MDV has a genomic
organization and protein composition homologus to VZV and HSV. The major
difference between MDV and other alpha herpesvirus is viral tropism. VZV and
HSV are neurotropic, in contrast, MDV is lymphotropic, causing latent infection

and transformation of lymphocytes.

Several MDV proteins have been identified based on sequence homology to
HSV proteins, although very few have been functionally characterized. The most
intriguing MDV proteins are those encoded in the invert repeat regions, unique to
MDV serotype 1 strains. The vlL-8 gene described in this dissertation is one of
the MDV-1 unique genes. Virus encoded chemokines or chemokine receptors
have been reported in different viruses, especially in herpesviruses (Seet et al.
2002) and poxviruses (Haig 2001). Some of them have characterized functions.
It is likely that during the long course of evolution, viruses have been successful in
taking advantage of their hosts’ genetic information and reprogramming these

chemokines or chemokine receptors as viral allies to aid in viral immune evasion.

To study the function of vlL-8 gene, both copies of vlL-8 residing in the TRL
and IRL region of the viral genome was deleted and generated the vlL-8 deleted

virus, er5/AvlL-8. Growth kinetics study showed that the vlL-8 gene is

113

 

 

dispensable for virus replication in cell culture. In vivo, er5/AvlL-8 inoculated
chickens have much less viral antigen expression in lymphoid organs at 6 DPI,
but normal level of viral antigen expression in the feather follicle epithelium at 14
DPI, indicating that vlL-8 gene is involved in early cytolytic infection in lymphoid
organs but not in lytic infection in the feather follicle epithelium. The er5/AvlL-8
and the wild type virus also showed similar viremia titers at 6 and 8 DPI, a period
where the virus titer comes primarily from reactivated latent genomes,
demonstrating that vIL-8 gene does not appear to be important for virus
reactivation or latency entry. Nevertheless, because of the impaired cytolytic
infections, the overall transformation efficiency of the vIL-8 deleted virus is much
lower, as reflected by the reduced number of transformed cells at 5 weeks post
inoculation and fewer gross tumors. The virulence of er5/AvlL-8 was also
dramatically reduced in both MDV maternal antibody negative and positive
chickens. It has been documented that a reduction or absence of an early
cytolytic infection with MDV correlates with the absence or reduction in the
incidence of lymohomas. Impairment of the early cytolytic infection caused by
deletion of the vIL-8 gene may also have an essential role in reducing the
virulence of er5/AvlL-8. Another interesting finding was that the deletion virus
protected chickens against challenge with a very virulent plus MDV strain.
Although it is well known that MD can be efficiently prevented and controlled by
vaccination, little is known about the mechanisms of how vaccines protect

chickens. lmportantly, the revertant virus that restored the expression of vlL-8

114

gene also restored the wild type phenotype, indicating the deficient phenotypes

are the result of vlL-8 deletion.

Since vlL-8 is highly homologous to the members of the chemokine family, it is
speculated that the function of vlL-8 is that of a chemoattractant. It was first
hypothesized that vlL-8 may function in MDV to attract target cells for virus
infection and replication. Preliminary in vitro migration assays carried out in
Dr.Kung’s laboratory identified that vIL-8 attracts cells different than their cellular
homologues. They found that unlike the cellular lL-8, vlL-8 does not attract
neutrophils, but attracts mononuclear cells instead. The lymphocyte population
change following early cytolytic infection with er5 and er5/AvlL-8 was also
studied using the flow cytometry assay. Because er5 virus is able to cause a
robust early cytolytic infection, the numbers of B lymphocytes in the er5
inoculated chickens were dramatically reduced, while the numbers of activated
T-cells were increased at 7 DPI. The loss of B-cells may due to cytosis or
apoptosis resulting from the effective cytolytic infection of the er5. The early
cytolytic infection stimulates the immune response and activates T-cells to
become the next major target cells. With the deletion of vIL-8, the early cytolytic
infection was impaired. Therefore, the number of B-cells in infected chickens
was not significantly different from the non-infected chickens. It is speculated
that the vIL-8 gene may play a necessary role in attracting B-cells, which are the
primary target cells for virus infection. This could also account for the weak
inflammatory/immune responses, resulting in less activation of T-cells and

significant decreased number of transformed cells, and low virulence.

115

 

The vlL-8 gene was transcribed in all tested serotype 1 MDV genomes by
Northern blot probing. The signal peptide on the N-terminal of the gene was also
functional. Sequence and structural analysis indicates that, there are two notable
differences among this vlL-8 gene and other cellular lL-8 gene. First, vlL-8 gene
carries a “DKR” motif but not an “ELR” motif which is strictly reserved in most of the
cellular lL-8 genes. As was mentioned earlier, vlL-8 and other cellular lL-8 attract
different types of cells. It is speculated that this motif variation could be one of
the factors to explain the difference in attraction of different cell types. The
“ELR” motif is also interesting because it is related to angiogenesis which is
extremely important to tumor formation, while a chemokine protein without “ELR”
motif normally will have the opposite, angiostatic reaction, which does not favor
tumor growth. The other difference is that vIL-8 has a much longer C-terminal
sequence. The siginificance of this is yet to be identified, but may be involved in

cell signaling unique to vlL-8.

To study the significance of this variation, recombinant MDV, er5/vlL-8-ELR,
was generated carrying “ELR” motif. Both in vitro and in vivo studies revealed
that the “DKR” motif is as competent as “ELR” in sustaining the vlL-8 functions.
Preliminary studies using the CAM assay demonstrated that the mutant
vIL-8-ELR protein induced clear neovascularization in the chicken
chorioallantonic membrane. But, the mutant virus er5/vlL-8-ELR did not show
any difference in in vitro replication and in vivo pathogenesis compared to the
parental virus, which suggests that, an “ELR” motif in the virus does not facilitate

more aggressive tumor growth. With these surprising results, sereval hypotheses

116

 

 

have been made here. Although vlL-8 is highly homologous to the cellular lL-8,
the role of of extra C-terminal domain in vIL-8 is not known. This unique feature
could be responsible for signaling events essential for establishment of a robust
infection during the early cytolytic infection. The other hypothesis is that vIL-8
does not have a role in inducing angiogenesis, but only functions in attracting
target cells. With stimulation by infection with MDV, the release of cellular
chemokines or cytokines (Xing et al. 2000) may serve as the factors that
contribute to the angiogenic reaction. Since few studies have been done on the
chicken cellular chemokines and its receptors, the signaling pathway in this area

is unclear. Further investigations are necessary.

117

 

FUTURE RESEARCH

1. Construct and characterize mutant MDV-2 and MDV-3 viruses containing

the vIL-8 gene.

Based on the data reported in this dissertation, deletion of the vIL-8 gene
results in impaired cytolytic infection and virus attenuation. This result indicates
that vlL-8 is necessary for the pathogenecity of the virus. It is suggested that
vlL-8 plays a role in attracting target cells for virus infection and replication. It is
also known that the vIL-8 gene is only encoded in MDV-1 viruses, which replicate
at a much higher level in vivo than serotype 2 and 3 viruses. Future research
that could be carried out is to insert the vIL-8 gene into MDV-2 and MDV-3. Two
goals could be tested after insertion of the vlL-8 gene into MDV-2 and MDV-3: 1)
Evaluation of the in vivo and in vitro replication level of the mutant MDV-2 and
MDV-3; and 2) Determine the virulence of the mutant MDV-2 and MDV-3 carrying
vlL-8 gene. This would help determine whether vlL-8 is a necessary gene for
MDV replication in vivo and whether or not it is essential to the virulence of the
virus. If the presence of the vIL-8 gene increases virus replication in chickens,
MDV-2 and MDV-3 recombinants may offer better protection. Comparisons
among the mutant MDV-2 and MDV-3 viruses coding the vIL-8 gene with the
parental viruses and mutant MDV-1 virus with no vIL-8 may also assist to further

understand the role of vlL-8 in MDV pathogenesis.

118

F- _r_MJ

2. Characterize the relation of vIL-8 and cIL-8 during infection in vivo.

With the work described herein, it is possible to elucidate the agonist or
antagonist role of vlL-8 gene to the cellular lL-8 (clL-8) gene. First, by using the
vIL-8 deletion virus, the transcription levels of both vIL-8 and cIL-8 could be
studied by RT-PCR from the infected chicken tissues. Since replication starts in
lymphoid organs, an analysis to monitor the transcript levels at different times
after inoculation could be performed. If vIL-8 plays an agonist role, a high level
of both vlL-8 and clL-8 transcripts in parental virus would be expected. The
clL-8 should have a similar level in both parental and the vlL-8 deletion viruses.

If clL-8 transcription level is higher in the vlL-8 deletion virus than in the parental
virus, this may mean that vIL-8 suppresses the function of clL-8 in parental virus.
To study the important function clL-8 may play in MDV infection, another mutant
virus with the insertion of the clL-8 in place of vlL-8 could be generated. A
comparison of the sequence and structural differences between the clL-8 and
vlL-8 could be done to explain the difference in attracting different target cells, or
other biological properties. This information would also help in answering
several questions regarding the MDV encoded vlL-8 gene. Why does MDV-1
encode this chemokine-like gene? Is a chemokine gene necessary for the
virulence and pathogenecity of MDV? Can other homologous chemokine genes
replace its function in MDV infection and replication? Will the replacement of
cIL-8 change the cell tropism in MDV infection? Will this change be related to
MDV induced transformation or tumorigenesis? With these questions answered,

a greater elucidation of the pathogenic mechanisms of MDV will be obtained.

119

References

Afonso, C. L. et al. (2001). "The genome of turkey herpesvirus." J Virol 75(2):
971-8.

Alcami, A., and Koszinowski, UH. (2000). "Viral mechanisms of immune evasion."
W 8(9): 410418.

Arvin, A. M. (2001). "Varicella-zoster virus: molecular virology and virus-host
interactions." Curr Ogin Microbiol 4(4): 442-9.

Baba, M., Nishlmura, O., Kanzaki, N., Okamoto, M., Sawada, H., Lizawa, Y.,
Shiraishi, M., Aramaki, Y., Okonogi, K., Ogawa, Y., Meguro, K., and Fujino,
M. (1999). "A small-molecule, nonpeptide CCR5 antagonist with highly
potent and selective anti-HlV-1 activity." Proc Natl Acafd Sci U S A 96:
5698-5703.

Baggiolini, M. (1998). "Chemokines and leukocyte traffic." Nature 392: 565-568.

 

Baggiolini, M. et al. (1994). "Interleukin-8 and related chemotactic cytokines—CXC
and CC chemokines.” Adv Immunol 55: 97-179.

Baggiolini, M. et al. (1997). "Human chemokines: an update."fl1_nu Rev lm_munol
15: 675-705.

Baggiolini, M., Dewald, B. and Moser, B. (1997). "Human chemokines: an
update.” Annu. Rev. Immunol. 15: 675.

Baggiolini, M. et al. (1995). "Interleukin-8 and the chemokine family.” Int J
lmmunopharmacol 17(2): 103-8.

Ball, R. F. et al. (1971). "The resistance to Marek’s disease of chicks from
immunized breeders." Poult Sci 50(4): 1084-90.

Bashoff, C., Endo, Y., Collins, PD., Takeuchi, Y., Reeves, JD., Scweickart, VL.,
Siani, MA., Sasaki, T., Williams, TJ., Gray, PW., Moore, PS., Chang, Y.,
and Weiss, RA. (1997). "Angiogeneic and HIV-inhibitory function of
KSHV-encoded chemokines." Science 278: 290-294.

 

Bazan, JF., Bacon, KB., Hardiman, G., Wang, W., 800, K., Rossi, 0., Greaves,
DR., Zlotnik, A., and Schall, TJ. (1997). "A new class of membrane-bound
chemokine with a CX3C motif.” Nature 385: 640.

 

Beasley, JN., Patterson, LT., and McWade, DH. (1970). "Transmission of Marek's
disease by poultry house dust and chicken dander.” Am J Vet Res 31:
339-344.

120

 

 

Becker, Y. et al. (1992). "Polymerase chain reaction for differentiation between
pathogenic and non-pathogenic serotype 1 Marek's disease viruses (MDV)
and vaccine viruses of MDV-serotypes 2 and 3." J Virol Methods 40(3):
307-22.

Benton, WJ., and Cover, MS. (1957). "The increased incidence of visceral
lymphomatosis in broiler and replacement birds." Avian Dis 1: 320-327.

Bernardini, G. et al. (2003). "Analysis of the role of chemokines in angiogenesis."
J Immunol Methods 273(1-2): 83-101.

Biggs, P. M. (1975). "Marek's disease—the disease and its prevention by
vaccination." Br J Cancer 31 SUPPL 2: 152-5.

Biggs, P. M. et al. (1971). "Use of the embryonating egg in studies on Marek's
disease." Am J Vet Res 32(11): 1795-809.

Biggs, P. M. et al. (1970). "Field trials with an attenuated cell associated vaccine
for Marek’s disease." Vet Rec 87(23): 704-9.

 

Biggs, PM and Payne, LN (1963). "Transmission Experiments with Marek's
disease isolates." Vet Rec 75: 177-179.

 

Biggs, PM and Payne, LN (1967). "Studies on Marek’s disease. I. Experimental
transmission." J Natl Cancer Inst 39: 267-280.

Biggs, PM. (1968). "Marek's disease-Current state of knowledge." Curr Topic
Microbiol lmunol 43.

Biggs, PM., and Milne, BS (1972). Biological properties of a number of Marek's
disease virus kisolates. Oncogenesis and Hemesvirus. P. a. P. Biggs, LN.
Lyon, France, IARC: 88-94.

Brunovskis, P. et al. (1995). "The Marek's disease virus (MDV) unique short
region: alphaherpesvirus-homologous, fowlpox virus-homologous, and
MDV-specific genes." Virology 206(1): 324-38.

Bublot, M. et al. (1999). "Non-essential loci in the BamHl-l and -F fragments of the
HVT FC126 genome.” Acta Virol 43(2-3): 181-5.

Buckmaster, A. E. et al. (1988). "Gene sequence and mapping data from Marek's
disease virus and herpesvirus of turkeys: implications for herpesvirus
classification." J Gen Virol 69 ( Pt 8): 2033-42.

Billow, V. (1977). "Further characterization of the CVI988 strain of Marek's
disease." Avian Path. 6: 395-403.

121

 

Biilow, V., and Biggs, PM. (1975a). "Differentiation between strains of Marek's
disease." Avian Path. 6: 395-403.

Biilow, V., and Biggs, PM. (1975b). "Precipitating antigens associated with
Marek's disease viruses and a herpesvirus by immunofluorescence
assays." Avian Path. 4: 147-162.

Blilow, V., Biggs, PM., and Frazier, JA. (19750). Characterization of a new

serogge of Marek's disease hemesvirus. International Agency for research

in Cancer., Lyon.

Biilow, V. V., Weiler, H., and Klasen, A. (1984). "Activating effects of interferons,
lymphokines and viruses on cultured chicken macrophages." Avian Pathol
13: 621-638.

Buscaglia, C. et al. (1988). "Maintenance of Marek's disease herpesvirus latency
in vitro by a factor found in conditioned medium." J Gen Virol 69 ( Pt 11):
2809-18.

Butcher, EC., and Picker, LJ. (1996). "Lymphocyte homing and homestasis."
Science 272: 60.

Calnek, B. W. (1972). "Effects of passive antibody on early pathogenesis of
Marek's disease." Infect lmmun 6(2): 193-8.

Calnek, B. W. (1986). "Marek's disease—a model for herpesvirus oncology." gm,
Rev Microbiol 12(4): 293-320.

Calnek, B. W. (2001). "Pathogenesis of Marek's disease virus infection." Curr Top
Microbiol lmmim 255: 25-55.

Calnek, B. W. et al. (1997). Marek's disease. Ames, Iowa, USA, Iowa state
University press.

Calnek, B. W. et al. (1970). "Feather follicle epithelium: a source of enveloped and
infectious cell-free herpesvirus from Marek's disease." Avian Dis 14(2):
219-33.

Calnek, B. W. et al. (1979). "Comparative pathogenesis studies with oncogenic
and nononcogenic Marek’s disease viruses and turkey herpesvirus." Am J
Vet Res 40(4): 541-8.

Calnek, B. W. et al. (1998). "Relationship between the immunosuppressive
potential and the pathotype of Marek's disease virus isolates." Avian Dis
42(1): 124-32.

Calnek, B. W. et al. (1983). "Field trials with a bivalent vaccine (HVT and SB-1)
against Marek's disease." Avian Dis 27(3): 844-9.

122

 

Calnek, B. W. et al. (1984). "Further characterization of Marek's disease
virus-infected lymphocytes. I. In vivo infection." Int J Cancer 33(3): 389-98.

Calnek, B. W. et al. (1981). "Latent infections with Marek's disease virus and
turkey herpesvirus." J Natl Cancer Inst 66(3): 585-90.

Calnek, B.W. (1985). Pathogenesis of Marek's disease. Proc. Int. Symp. Marek’s
Disease American Association of Avian Pathologists,, Kennett Square,
USA.

Calnek BW, Adldinger HK, Kahn DE (1970a). "Feather follicle epithelium: a
source of enveloped and infectious cell-free herpesvirus from Marek's
disease." Avian Dis 14: 219-233.

Calnek, BW., and Hitchner, SB. (1969). "Localization of viral antigen in chickens
infected with Marek's disease herpesvirus." J Natl Cancer Inst 43: 935-949.

Calnek, BW., and Witter, RL. (1997). Neoplastic diseases: Marek's disease._
Disease of Poultgy. B. Calnek. Ames, Iowa, USA., Iowa State University
Press,.

Calnek, BW., Harris, RW., Buscaglia, C, Schat, KA., and Lucio, B. (1998).
"Relationship between the immunosuppressive potential and the pathotype
of Marek's disease virus isolates." Avian Dis. 42: 124-132.

Calnek, BW., Ubertini, T., and Adldinger, HK. (1970b). "Viral antigen, virus
particle, and infectivity of tissues from chickens with Marek's disease." J_
Natl Cancer Inst 45: 341-351.

Campbell, JG (1956). "Leucosis and fowl paralysis compared and contrasted."
Vet Rec 68: 527-529.

Campbell, JG (1961 ). "A proposed classification of the leucosis complex and fowl
paralysis." Brit Vet J. 117: 316-325.

Carrozza, JH., Fredrickson, TN., Prince, RP., and Luginbuhl. (1973). "Role of
desquamated epithelial cells." Avian Dis 17: 767-781.

Cebrian, J. et al. (1982). "Inverted repeat nucleotide sequences in the genomes of
Marek disease virus and the herpesvirus of the turkey." Proc Natl Acad Sci
U S A 79(2): 555-8.

Chen, 8., Bacon, KB., Li, L., Garcia, G.E., Xia, Y., Lo, 0., Thompson, DA., Siani,
lVlA., Yamamoto, T., Harrison, JK., and Feng, L. (1998). "In vivo inhibition
of CC and CX3C chemokine-induced leukocyte infiltration and attenuateion
of glomerulonephritis in Wistar-Kyoto (WKY) rats by vMIP-ll." J Exp Med
188: 193-198.

123

Pummm‘filgl

Chubb, R. C. et al. (1969). "Effect of maternal antibody on Marek's disease." Vit
R_e_c_: 85(11): 303-5.

Churchill, A. E. et al. (1968). "Herpes-type virus isolated in cell culture from
tumors of chickens with Marek's disease. ll. Studies in vivo." J Natl Cancer
flit 41(4): 951-6.

Churchill, A. E. et al. (1969). "Immunization against Marek's disease using a live
attenuated virus." Nature 221(182): 744-7.

 

Churchill, AE and Biggs, PM (1967). "Agent of Marek's disease in tissue culture."
Nature 215: 528-530.

Churchill, AE and Biggs, PM (1968a). "Herpes-types virus isolated in cell culture
from tumors of chickens with Marek's disease. ll. Studies in vivo." J Natl
Cancer Inst 41: 951-956.

Churchill, AE. (1968b). "Herpes-type virus isolated in cell culture from tumors of
chickens with Marek's disease. I. Studies in cell culture." J Natl Cncer Inst
41: 939-951.

Cohen, J. I. (2001). "Mutagenesis of the varicella-zoster virus genome: lessons
learned." Arch Virol Su_9p_l(17): 91-7.

Cohen, J. I. et al. (1993). "Generation of varicella-zoster virus (VZV) and viral
mutants from cosmid DNAs: VZV thymidylate synthetase is not essential
for replication in vitro." Proc Natl Acad Sci U S A 90(15): 7376-80.

Cook, D.N., Beck, M.A., Coffman, TM., Kirby, SL., Sheridan, JF., Pragnell, IB.,
and Smithies, O. (1995). "Requirement of MIP-1 alpha for an inflammatory
response to viral infection." Science 269: 1583-1585.

 

Cui, Z. 2. et al. (1988). "Monoclonal-antibody-mediated enzyme-linked
immunosorbent assay for detection of reticuloendotheliosis viruses." Avian
% 32(1): 32-40.

Cui, Z. 2. et al. (1990). "Marek's disease virus gene clones encoding virus-specific
phosphorylated polypeptides and serological characterization of fusion
proteins." Virus Genes 3(4): 309-22.

Dairaghi, D.J., Fan, RA., McMaster, BE., Hanley, MR., and Schall, TJ. (1999).
"HHV8-encoded leP-1 selectivity engages chemokine receptor CCR8.
Agonist and antagonist profiles of viral chemokines." J Biol Chem 274:
21569-21 574.

Dandapat, S. et al. (1994). "Anti-idiotype antibodies to Marek’s

disease-associated tumour surface antigen in protection against Marek's
disease." Vet lm_munol Immunopathol 40(4): 353-66.

124

 

Davison, A.J. and Scott, J.E. (1986). "The complete DNA sequence of
varicella-zoster virus." J Gen Virol 67: 1759-1816.

Delecluse, H. J., T. Hilsendegen, D. Pich, R. Zeidler, and W. Hammerschmidt.
(1998). ".Propagation and recovery of intact, infectious Epstein-Barr virus
from prokaryotic to human cells." Proc. Natl. Acad. Sci. USA 95:
8245-8250.

Ehsani, M. E. et al. (2000). "Generation of mutant murine cytomegalovirus strains
from overlapping cosmid and plasmid clones." J Virol 74(19): 8972-9.

Epstein, M. A. et al. (1968). "Structure and development of the herpes-types virus
of Marek's disease." J Natl Cancer Inst 41(3): 805-20.

Ernst, CA., Zhang, YJ., Hancock, PR., Rutledge, BK., Corless, CL., and Rollins,
BJ. (1994). "Biochemical and biological characterization of murine MCP-1:
identification of two functional domains." J Immunol 152: 3541.

Faure, 8., Meyer, L., Costagliola, D., Vaneensberghe, C., Genin, E., Autran, B.,
Delfraissy, JF., McDermott, DH., Murphy, PM., Debre, P., Theodorou, l.,
and Combadiere, C. (2000). "Rapid progression to AIDS in HIV+
individuals with a structural variant of chemokine receprtor CX3CR1."
Science 287: 2274-2277.

 

Feldman, LT. (1991). The molecular biology of herpes simplex virus latency. Boca
Raton.

Ferrin, L. J. et al. (1991). "Selective cleavage of human DNA: RecA-assisted
restriction endonuclease (RARE) cleavage." Science 254(5037): 1494-7.

 

Fraser, NN., Block, TM., and Spivck, JG. (1992). "The latecy-associated
transcripts of herpes simplex virus: RNA in search of function." Virology
191: 1-18.

Fukuchi, K. et al. (1985a). "Map location of homologous regions between Marek's
disease virus and herpesvirus of turkey and the absence of detectable
homology in the putative tumor-inducing gene." J Virol 53(3): 994-7.

Fukuchi, K., Sudo, M., Lee, Y. S., Tanaka, A., Nonoyama, M. (1984). "Structure of
Marek’s disease virus DNA: detailed restriction enzyme map." J Virol 51(1):
102-9.

Fukuchi, K. et al. (1985b). "The structure of Marek disease virus DNA: the
presence of unique expansion in nonpathogenic viral DNA." Proc Natl
Acad Sci U S A 82(3): 751-4.

Garcia-Blanco, MA., Cullen, BR. (1991). "The molecular basis of latency in
pathogenic human viruses." Science 254: 815-820.

 

125

Gimeno, lM., Witter, RL., and Neumann, U. (2002). "Neuropathotyping: A new
system to classify Marek's disease virus." Avian Diseases 46(4): 909-918.

Goodchild, WM. (1969). "Some observations on Marek's disease (fowl paralysis)"
Vet Rec(84): 87-89.

 

Haffer, K. et al. (1979). "The role of the macrophages in Marek's disease: in vitro
and in vivo studies." Int J Cancer 23(5): 648-56.

Haig, D. M. (2001). "Subversion and piracy: DNA viruses and immune evasion."
Res Vet Sci 70(3): 205-19.

Hamdy, F. et al. (1974). "Biogenesis of Marek's disease (type II leukosis) virus in
vitro: electron microscopy and immunological study." Infect lmmun 9(4):
740-9.

Hirai, K., Horiuchi, T., and Mikami, T. (1988). Molecular biology of Marek's
disease virus. Osaka, Japan, Japanese Association on Marek's disease.

Hirai, K., lkuta, K., and Kato, S. (1979). "Comparative studies on Marek's disease
virus and herpesvirus of turkey DNAs." J Gen Virol 45(1): 119-31.

Hlozanek, I. et al. (1974). "Lack of pathogenicity of Marek's disease herpes virus
and herpes virus of turkeys for mammalian hosts and mammalian cell

cultures." Folia Biol (Praha) 20(1): 51-8.

Hudson, L. et al. (1973). "An analysis of the T and B cells of Marek's disease
lymphomas of the chicken." Nat New Biol 241(106): 52-3.

Hughes, S. et al. (2000). "The gene encoding the chicken chemokine K60 maps
to chromosome 4." Anim Genet 31(6): 418-9.

lgarashi, T. et al. (1987). "Restriction enzyme map of herpesvirus of turkey DNA
and its collinear relationship with Marek’s disease virus DNA." Virology
157(2): 351-8.

lkuta, K. et al. (1982). "Monoclonal antibodies specific to and cross-reactive with
Marek's disease virus and herpesvirus of turkeys." Biken J 25(4): 171-5.

lkuta, K. et al. (1981). "lmmunoprecipitation of Marek's disease virus-specific
polypeptides with chicken antibodies purified by affinity chromatography."
Virology 114(1): 277-81.

lmai, K., N.Yuasa, K. Furuta, M. Narita, H. Banba, S. Kobayashi, and T. Horiuchi.
(1991). "Comparative studies on pathogenicalmm virological and

serological properties of Marek's disease virus isolated from jpanese quail
and chicken." Avian Path 20: 57-65.

126

 

lzumiya, Y. et al. (2001). "A complete genomic DNA sequence of Marek's disease
virus type 2, strain HPRS24." Curr Too Microbiol Immunol 255: 191-221.

Jakowski, RM., Fredrickson, TN., Chomiak, TW., and Luginbuhl. RE. . (1970).
"Hematopoietic destruction in Marek's disease." Avian Dis 14: 374-385.

Jones, D. et al. (1992). "Marek disease virus encodes a basic-Ieucine zipper gene
resembling the fos/jun oncogenes that is highly expressed in
lymphoblastoid tumors." Proc Natl Acad Sci U S A 89(9): 4042-6.

Jurajda, V., and Klimes, B. (1970). "Presence and survival of Marek's disease
agent in dust." Avian Dis 14: 188-190.

Kaaden, O. R. (1977). "[Structure and function of virus-induced antigens in
cultured cells infected with Marek's disease and turkey herpes viruses. I.
Alteration of the immunological specificity of plasma membranes in
infected cells (author's transl)]." Med Microbiol Immunol (Berl) 163(3):
141-56.

Kaiser, P. et al. (1999). "The chicken 9E3/CEF4 CXC chemokine is the avian
orthologue of IL8 and maps to chicken chromosome 4 syntenic with genes
flanking the mammalian chemokine cluster." mmunogenetics 49(7-8):
673-84.

Karlin, S. et al. (1994). "Molecular evolution of herpesviruses: genomic and
protein sequence comparisons." J Virol 68(3): 1886-902.

 

Kato, S. et al. (1985). "Marek's disease virus." Adv Virus Res 30: 225-77.

Kaul, L. et al. (1991). "lmmunopathology of Marek’s disease in quails: presence of
antinuclear antibody and immune complex." Vet Immunol lmmunopathol
28(1): 89-96.

Kaup, BF (1921 ). "Paralysis of the domestic fowl." J Amer Ass Instructors and
Invest in Poult_rv Husbandy 7: 25-31.

Kawamura, M., Hayashi, M., Furuichi, T., Nonoyama, M., lsogai, E., Namioka, S.
(1991). "The inhibitory effects of oligonucleotides, complementary to
Marek's disease virus mRNA transcribed from the BamHl-H region, on the
proliferation of transformed lymphoblastoid cells, MDCC-MSB1." J Gen
yjr_ol72 ( Pt 5): 1105-11.

Kenzy, S. G. et al. (1969). "Transmission of classical Marek's disease by affected
and carrier birds." Avian Dis 13(1): 211-4.

Kodama, H. et al. (1979). "Inhibitory effects of macrophages against Marek's

disease virus plaque formation in chicken kidney cell cultures." J Natl
Cancer Inst 63(5): 1267-71.

127

 

1'- ,3'
K.,
, V

Komegay, JN, Gorgacz, EJ, Parker, MA, Brown, J, and Schierrnan, LW (1983).
"Marek's disease virus-induced transient paralysis: Clinical and
electrophysiological findings in susceptible and resistant lines of chickens."
Am J Vet Res 44: 1541-1544.

Kung, H. J. et al. (2001). "Meq: an MDV-specific bZIP transactivator with
transforming properties." Curr Top Microbiol Immunol 255: 245-60.

Kung, H. J., Xia, L., Brunovskis, R, Li, D., Liu, J. L., Lee, L. F. (2001). Meq: An
MDV-specific bZlP transactivator with transforming properties. Marek's
Disease. 255: 245-260. '

 

Lalani, A. S. et al. (2000). "Modulating chemokines: more lessons from viruses."
Immunol Today 21(2): 100-6.

Lalani, AS., and McFadden, G. (1999a). "Evasion and exploitation of chemokines
by viruses." Cvtolgjne Growth Factor Rev 10(3-4): 219-233.

Lalani, AS., Masters, J., Zeng, W., Barrett, J., Pannu, R., Everett, H., Arendt, CW.,
and McFadden, G. (1999b). "Use of chemokine receptors by poxviruses."
Science 286: 1968-1971.

 

Lawn, AM., and Payne, LN. (1979). "Chronological study of ultrastructural
changes in the peripheral nerves in Marek's disease." Neuropathol Appl
Neurobiol 5: 485-497.

Le Deuff, R. M., Nicolas, J. L., Renault,T., and Cochennec, N. (1994).
"Experimental transmission of a herpes-like virus to axenic larvae of Pacific
oyster Crassostrea gigas." Bull. Eur. Assoc. Fish Pathol. 14: 69-72.

Lee, L. F. (1979). "Macrophage restriction of Marek's disease virus replication and
lymphoma cell proliferation." J Immunol 123(3): 1088-91.

Lee, L. F. et al. (1971). "Size and composition of Marek's disease virus
deoxyribonucleic acid." J Virol 7(3): 289-94.

Lee, L. F. et al. (1983a). "A monoclonal antibody reactive with Marek's disease
tumor-associated surface antigen." J Immunol 130(2): 1007-11.

Lee, L. F. et al. (1983b). "Monoclonal antibodies with specificity for three different
serotypes of Marek's disease viruses in chickens." J Immunol 130(2):
1003-6.

Lee, L. F. et al. (2000). "The complete unique long sequence and the overall
genomic organization of the GA strain of Marek's disease virus." Proc Natl
Acad Sci U S A 97(11): 6091-6.

128

 

 

Lee, S. I. et al. (1999). "Re-isolation of Marek's disease virus from T cell subsets
of vaccinated and non-vaccinated chickens." Arch Virol 144(1): 45-54.

Lessard, M. et al. (1996). "Influence of Marek's disease virus strain AC-1 on
cellular immunity in birds carrying endogenous viral genes." Avian Dis
40(3): 645-53.

Liass, L. A. (1985). "[Role of the cytoskeleton in the morphological normalization
of transformed cells in cultural." Biull Eksp Biol Med 100(7): 86-9.

Liu, J. L., Kung, H. J. (2000). "Marek's disease herpesvirus transforming protein
MEQ: a c-Jun analogue with an alternative life style." Virus Genes 21(1-2):
51-64.

Liu, J. L., Lee, L. F., Ye, Y., Qian, Z., Kung, H. J. (1997). "Nucleolar and nuclear
localization properties of a herpesvirus bZlP oncoprotein, MEQ." J Virol
71(4): 3188-96.

 

Liu, J. L. et al. (1999). "MEQ and V-lL8: cellular genes in disguise?" Acta Virol
43(2-3): 94-101.

Liu, J. L., Ye, Y., Qian, Z., Qian, Y., Templeton, D. J., Lee, L. F., Kung, H. J.
(1999). "Functional interactions between herpesvirus oncoprotein MEQ
and cell cycle regulator CDK2." J Virol 73(5): 4208-19.

 

Lupiani, B. et al. (2001). "Protein-coding content of the Sequence of Marek's
disease virus serotype 1." Curr Top Microbiol Immunol 255: 159-90.

Lusso, P. (2000). "Chemokines and viruses: the dearest enemies." Virology
273(2): 228-40.

Maotani, K. et al. (1986). "Amplification of a tandem direct repeat within inverted
repeats of Marek’s disease virus DNA during serial in vitro passage." J_
Virol 58(2): 657-60.

Marek, J. (1907). "Multiple Nervenentzundung (Polyneuritis) bei Huhem." Dtsch
Tierarztl Wschr 15: 417-421.

Martins-Green, M. (2001). "The chicken Chemotactic and Angiogenic Factor
(cCAF), a CXC chemokine." Int J Biochem Cell Biol 33(4): 427-32.

Martins-Green, M. et al. (1998). "The chicken chemotactic and angiogenic factor
(9E3 gene product): its angiogenic properties reside in the C-terminus of
the molecule." Cflokine 10(11): 819-30.

Matsuda, H. et al. (1976a). "Detection of T-cell surface determinants in three
Marek's disease lymphoblastoid cell lines." Biken J 19(1): 29-32.

 

129

 

 

Matsuda, H. et al. (1976b). "Demonstration of a Marek's disease
tumor-associated surface antigen (MATSA) on six cell lines derived from
Marek's disease lymphomas." Biken J 19(3): 119-23.

McColl, K. A. et al. (1987). "Expression of a putative tumor-associated surface
antigen on normal versus Marek's disease virus-transformed
lymphocytes." J Natl Cancer Inst 79(5): 991-1000.

McFadden, G., and Murphy, PM. (2000). "Host-related immunomodulators
encoded by poxviruses and herpesviruses." Qprr Opin Microbiol 3(4):
371-378.

McGeoch, D. J. (1989a). "The genomes of the human herpesviruses: contents,
relationships, and evolution." Ann4u Rev Microbiol 43: 235-65.

McGeoch, D. J., and Davison, A. J., Ed. (1999). The molecular evolutionaLy
histou of the herpesviruses,. Origin and evolution of viruses. London,
England, Academic Press.

McGeoch, D. J., Cook, S. (1994). "Molecular phylogeny of the alphaherpesvirinae
subfamily and a proposed evolutionary timescale." J Mol Biol 238(1): 9-22.

McGeoch, D. J., Cook, 8., Dolan, A., Jamieson, F. E., Telford, E. A. (1995).
"Molecular phylogeny and evolutionary timescale for the family of
mammalian herpesviruses." J Mol Biol 247(3): 443-58.

McGeoch, D. J. et al. (2000). "Toward a comprehensive phylogeny for
mammalian and avian herpesviruses." J Virol 74(22): 10401-6.

McGeoch, DJ. (1989b). "The genomes of the human herpesviruses: contents,
relationships, and evolution." Annu Rev MicrotLol 43: 235—265.

Messerle, M., l. kaovi, W. Hammerschmidt. H. Ziegler, and U. H. Koszinowski.
(1997). "Cloning and mutagenesis of a herpesvirus genome as an
infectious bacterial artificial chromosome." Proc. Natl. Acad. Sci. USA 94:
14759-14763.

Minson, A. C., Davison,A. J., Desrosiers,R. C., Fleckenstein, B., McGeoch, D. J.,
Pellett, P. E.,Roizman, B., and Studdert., D. M. J. ,, Ed. (2000.).
Hemesviridae. Virus taxonomy. New York, NY, Academic Press.

Morgan, C., Rose, H.M., Holden, M., and Jones, ER (1959). "Electron
microscopic observations on the development of herpes simplex virus." J_
Exp Med 110: 643-656.

Morgan, R. W. et al. (1990). "Transfection of chicken embryo fibroblasts with
Marek's disease virus DNA." Avian Dis 34(2): 345-51.

130

 

 

 

iy

Murphy, PM. (2001). "Viral exploitation and subversion of the immune system
through chemokine mimicry." Nat Immunol 2(2): 116-122.

Murthy, K. K. et al. (1979). "Demonstration of chicken fetal antigen (CFA) on
normal splenic lymphocytes, Marek's disease lymphoblastoid cell lines and
other neoplasms." Int J Cancer 24(3): 349-54.

Naito, M., Nakajima, K., lwa, N., Ono, K., Yoshida, l., Konobe, T., lkuta, K., Ueda,
8., Kato, S., and Hirai, K. (1986). "Demonstration of Marek's disease
virus-specific antigen in tumor lesions of chickens with Marek's disease
using monoclonal antibody against a virus phosphorylated protein." Avian
Path. 15: 503-510.

Nakajima, K., lkuta, K., Naito, M., Ueda, 8., Kato, S., and Hirai, K. (1987).
"Analysis of Marek's disease virus serotype-1 specific phosphorylated
polypeptides in virus-infected cells and Marek's disease lymphoblastoid
cells." J. Gen. Virol. 68: 1379-1389.

Nazerian, K. (1974). "DNA configuration in the core of Marek's disease virus." J_
Virol 13(5): 1148-50.

Nazerian, K. et al. (1968). "Electron microscopy of a herpes virus associated with
the agent of Marek's disease in cell culture." Cancer Res 28(12): 2454-62.

Nazerian, K. et al. (1975). "Detection of T-cell surface antigens in a Marek's
disease lymphoblastoid cell line." J Natl Cancer Inst 54(1): 277-9.

Nazerian, K., Solomon, .lJ., Witter, RL., and Burrnester, BR. (1968). "Studies on
the etiology of Marek’s disease. ll. Finding of a herpesvirus in cell culture."
Proc Sic Exp Bio Med 127: 177-182.

Nil, 8. et al. (1975). "Detection of viral cores having toroid structures in eight
herpes viruses." Biken J 18(1): 41-6.

Okada, K. et al. (1972). "The fine structure of Marek's disease virus and
herpesvirus of turkey in cell culture." Jpn J Vet Res 20(3): 57-68.

Okada, K. et al. (1980). "Cohelical arrangement of the DNA strand in the core of
Marek's disease virus particles." Arch Virol 64(1): 81-5.

Okazaki, W. et al. (1970). "Protection against Marek's disease by vaccination with
a herpesvirus of turkeys." Avian Dis 14(2): 413-29.

Omar, A. R. et al. (1996). "Syngeneic Marek's disease virus (MDV)-specific
cell-mediated immune responses against immediate early, late, and unique
MDV proteins." Virology 222(1): 87-99.

131

 

Omar, A. R. et al. (1998). "Cytotoxic T lymphocyte response in chickens
immunized with a recombinant fowlpox virus expressing Marek's disease
herpesvirus glycoprotein B." Vet lrpmunol lmmunopathol 62(1): 73-82.

Ono, M. et al. (1992). "The restriction endonuclease map of Marek's disease virus
(MDV) serotype 2 and collinear relationship among three serotypes of
MDV." Virology 191(1): 459-63.

Ono, M. et al. (1994). "Nucleotide sequence analysis of Marek's disease virus
(MDV) serotype 2 homolog of MDV serotype 1 pp38, an antigen associated
with transformed cells." Virology 201(1): 142-6.

Ono, M. et al. (1995). "Expression of Marek's disease virus (MDV) serotype 2
gene which has partial homology with MDV serotype 1 pp38 gene." Virus
gas 35(2): 223-9.

Pappenmeimer, AM., Dunn, LC., Cone, V (1926). "A study of fowl paralysis
(neuro-lymphomatosis gallinarum)." Storrs Agric Ecp Sta Bull No 143:
186-290.

Pappenmeimer, AM., Dunn, LC., Cone, V (1929a). "Studies on fowl paralysis
(neurolymphomatosis gallinarum). l.Clinical features and pathology." J Exp
Med 49: 63-86.

Pappenmeimer, AM., Dunn, LC., Cone, V (1929b). "Studies on fowl paralysis
(neurolymphomatosis gallinarum). ll. Transmission experiments." J Exp
Med 49: 87-102.

Parcells, M. S. et al. (1995). "Retention of oncogenicity by a Marek's disease virus
mutant lacking six unique short region genes." J Virol 69(12): 7888-98.

Parcells, M. S. et al. (2001). "Marek's disease virus (MDV) encodes an
interleukin-8 homolog (vIL-8): characterization of the vIL-8 protein and a
vlL-8 deletion mutant MDV." J Virol 75(11): 5159-73.

Payne, L. N. et al. (1973). "Pathogenesis of Marek's disease in chicks with and
without maternal antibody." J Natl Cancer Inst 51(5): 1559-73.

Payne, LN and Biggs, PM. (1967). "Studies on Marek's disease. ll.
Pathogenesis." J Natl Cancer Inst 39: 281-302.

Payne, LN. . Ed. (1985). Pathology. Mareks's Disease. Martinus Nihjhoff, Boston,
MA.

Payne, LN., Frazier, JA., and Powell, PC. (1976). "Pathogenesis of Marek's
disease." n Rev Exp Pathol 16: 59-154.

132

Penfold, M. E. et al. (1999). "Cytomegalovirus encodes a potent alpha
chemokine." Proc Natl Acad Sci U S A 96(17): 9839-44.

 

Petherbridge, L. et al. (2003). "Replication-competent bacterial artificial
chromosomes of Marek's disease virus: novel tools for generation of
molecularly defined herpesvirus vaccines." J Virol 77(16): 8712-8.

 

Powell, P. C. et al. (1983a). "The occurrence of chicken foetal antigen after
infection with Marek's disease virus in three strains of chicken." Oncodev
Biol Med 4(4): 261-71.

Powell, P. C. et al. (1983b). "Studies on the role of macrophages in Marek's
disease of the chicken." J Reticuloendothel Soc 34(4): 289-97.

Powell, PC., and Rennie, M. (1984). "The expression of Marek's disease
tumor-associated surface antigen in various avian species." Avia_n Pathol
13: 345-349.

Pradhan, H. K. et al. (1985). "Marek's disease in Japanese quails (Coturnix
cotumix japonica): a study of natural cases." Avian Dis 29(3): 575-82.

Pradhan, H. K. et al. (1987). "Experimental studies on Marek's disease in
Japanese quail (Coturnix cotumix japonica)." Avian Dis 31(2): 225-33.

Qian, Z., Brunovskis, F., Lee, L., Vogt, P. K., Kung, H. J. (1996). "Novel DNA
binding specificities of a putative herpesvirus bZlP oncoprotein." J Virol
70(10): 7161-70.

Qian, Z., Brunovskis, P., Rauscher, F., 3rd, Lee, L., Kung, H. J. (1995).
"Transactivation activity of Meq, a Marek's disease herpesvirus bZlP
protein persistently expressed in latently infected transformed T cells." J_
.V_irpl 69(7): 4037-44.

Rajarathnam, K. et al. (1999). "Disulfide bridges in interleukin-8 probed using
non-natural disulfide analogues: dissociation of roles in structure from
function." Biochemistpy 38(24): 7653-8.

Reddy, S. M. et al. (2002). "Rescue of a pathogenic Marek's disease virus with
overlapping cosmid DNAs: use of a pp38 mutant to validate the technology
for the study of gene function." Proc Natl Acad Sci U S A 99(10): 7054-9.

Rispens, B. H. et al. (1972a). "Control of Marek's disease in the Netherlands. I.
Isolation of an avirulent Marek's disease virus (strain CVI 988) and its use
in laboratory vaccination trials." Avian Dis 16(1): 108-25.

Rispens, B. H. et al. (1972b). "Control of Marek's disease in the Netherlands. ll.
Field trials on vaccination with an avirulent strain (CVI 988) of Marek's
disease virus." Avian Dis 16(1): 126-38.

133

 

Roizman, B. (1992a). "The family Herpesvirudae: an update." Archive Virology
123: 425-449.

Roizman, B., Carmichael, LE., Deinhardts, F., de The G., Nahmias, AJ., Plowright,
W., Rapp, F., Sheldrick, P., Takahashi, M., and Wolf, K. (1981).
"Herpesviridae: definition, provisional nomenclature, and taxonomy."
lntervirol 16: 201 -21 7.

Roizman, B., Desrosiers, RC., Fleckenstein, B., Lopez, C., Minson, AC., and
Studdert, MJ. (1992b). "The family Herpesviridae: an update." Arch Virol
23: 425-449.

Rollins, B. J. (1997a). "Chemokines." Blood 90(3): 909-28.
Rollins, BJ. (1997b). "Chemokines." Blood 90(3): 909-928.

Rosenkilde, M. M. et al. (2001). "Virally encoded 7TM receptors." Onppgene
20(13): 1582-93.

Ross, L. J. et al. (1975). "Serological relationship between a pathogenic strain of
Marek's disease virus, its attenuated derivative and herpes virus of
turkeys." J Gen Virol 28(1): 3747.

Ross, L. J. et al. (1973). "Purification and properties of the 'A' antigen associated
with Marek's disease virus infections." J Gen Virol 18(3): 291-304.

Ross, L. J. et al. (1983). "Restriction endonuclease analysis of Marek's disease
virus DNA and homology between strains." J Gen Virol 64 ( Pt 12):
2785-90.

Ross, N. et al. (1993). "Alterations in DNA sequence and RNA transcription of the
Barn HI-H fragment accompany attenuation of oncogenic Marek's disease
herpesvirus." Virus Genes 7(1): 33-51.

Ross, N. et al. (1997). "Marek's disease virus EcoRl-Q gene (meq) and a small
RNA antisense to lCP4 are abundantly expressed in CD4+ cells and cells
carrying a novel lymphoid marker, AV37, in Marek's disease lymphomas."
J Gen Virol 78 ( Pt 9): 2191-8. -

Russo, J. et al. (1988). "Expression of phenotypical changes by human breast
epithelial cells treated with carcinogens in vitro." Cancer Res 48(10):
2837-57.

Schat, K. A. et al. (1989). "Partial transcription map of Marek's disease

herpesvirus in lytically infected cells and lymphoblastoid cell lines." Int J
Cancer 44(1): 101-9.

134

 

Schat, K. A. et al. (1978a). "Characterization of an apparently nononcogenic
Marek's disease virus." J Natl Cancer Inst 60(5): 1075-82.

Schat, K. A. et al. (1978b). "Protection against Marek's disease-derived tumor
transplants by the nononcogenic SB-1 strain of Marek's disease virus."
Infect lmmun 22(1): 225-32.

Schat, K. A. et al. (1991). "Transformation of T-Iymphocyte subsets by Marek's
disease herpesvirus." J Virol 65(3): 1408-13.

Schat, K. A. et al. (2000). "Specific and nonspecific immune responses to Marek's
disease virus." Dev Com_p Immunol 24(2-3): 201-21.

Schat, K.A. (1987). Immunity in Marek's disease and other tumors; Boca Raton.
USA, CRC Press,.

Schat, KA. (1985). Characteristics of the virus. Marek's disease. L. Payne. Boston,
MA, Martinus Nijhoff: 77-112.

Schat, KA. (1996). Immunity to Marek's disease, lymphoid leukosis and
reticuloendotheliosis. Abingdon, England, Carfax Publishing Company,.

Schumacher, D. et al. (2000). "Reconstitution of Marek's disease virus serotype 1
(MDV-1) from DNA cloned as a bacterial artificial chromosome and
characterization of a glycoprotein B-negative MDV-1 mutant." J Virol
74(23): 11088-98.

Seet, B. T. et al. (2002). "Viral chemokine-binding proteins." J Leukoc Biol 72(1):
24-34.

Seet, BT., and McFadden, G. (2002). "Viral chemokine-binding proteins." _._l_
Leukoc Biol 72(1): 24-34.

Sevoian, M. and Chamberlain, DM. (1962a). "Avian lymphomatosis. Experimental
reproduction of the ocular form." Vet Med 57: 608-609.

 

Sevoian, M, Chamberlain, DM., and Counter, FT. (1962b). "Avian lymphomatosis.
I. Experimental reproduction of the neural and visceral forms." Vet Med 57:
500-501.

Sharma, J. M. et al. (1973). "Public health implications of Marek's disease virus
and herpesvirus of turkeyes. Studies on human and subhuman primates."
J Natl Cancer Inst 51(4): 1123-8.

Sharma, J. M. et al. (1972). "Lack of pathogenicity of Marek's disease virus and

herpesvirus of turkeys in marrnoset monkeys." J Natl Cancer Inst 49(4):
1191 -7.

135

Shanna, JM. (1987). "Delayed replication of Marek's disease following in ovo
inoculation during late stages of emvryonal development." Avian Dis 31:
570-576.

Shek, W. R. et al. (1983). "Characterization of Marek's disease virus-infected
lymphocytes: discrimination between cytolytically and latently infected
cells." J Natl Cancer Inst 70(3): 485-91.

Sick, C. et al. (2000). "Novel chicken CXC and CC chemokines." Cytokine 12(3):
181-6.

Silva, R. F. et al. (1991). "Restriction endonuclease analysis of Marek's disease
virus DNA: differentiation of viral strains and determination of passage
history." Avian Dis 35(3): 487-95.

Silva, R. F., Barnett, J. C. (1991). "Restriction endonuclease analysis of Marek's
disease virus DNA: differentiation of viral strains and determination of
passage history." Avian Dis 35(3): 487-95.

Silva, R. F ., Witter, R. L. (1985). "Genomic expansion of Marek's disease virus
DNA is associated with serial in vitro passage." J Virol 54(3): 690-6.

Smith, G. A., and L. W. Enquist. (1999). "Construction and transposon
mutagenesis in Escherichia coli of a full-length infectious clone of
pseudorabies virus, an alphaherpesvirus." J. Virol. 73: 6405-6414.

Smith, GD., Zelnik, V., and Ross, LJN. (1995). "Gene organization in herpesvirus
of turkeys: identification of a novel open reading frame in the ling unique
region and a truncated homologye of pp38 in the internal repeat." Virology.
207: 205-216.

Solomon, JJ., Witter, RL., Nazerian, K., and Bumester, BR. (1968). "Studies on
the etiology of Marek's disease. I. Propagation of the agent in cell culture."_
Proc 800 Exp Biol Med 127: 173-177.

Spencer, J. L. (1969). "Marek's disease herpesvirus: in vivo and in vitro infection
of kidney cells of different genetic strains of chickens." Avian Dis 13(4):
753-61.

Spencer, JL., and Calnek, BW (1967). "Storage of cells infected with Rous
Sarcoma virus or JM strain of avian lymphomatosis agent." Avian Dis 11:
274-287.

Stevens, JG. (1989). "Human herpesviruses: A consideration of the latent state."
Microbiol Rev 53: 318-332.

Strieter, R. M. et al. (2002). "CXC chemokines in angiogenesis related to
pulmonary fibrosis." Chest 122(6 Suppl): 2988-3018.

136

Strieter, R. M. et al. (1995a). "Interferon gamma-inducible protein 10 (lP-10), a
member of the C-X-C chemokine family, is an inhibitor of angiogenesis."

Biochem Biophys Res Commun 210(1): 51-7.

Strieter, R. M. et al. (1995b). "The role of CXC chemokines as regulators of
angiogenesis." Shock 4(3): 155-60.

 

Strieter, R. M. et al. (19950). "Role of C-X-C chemokines as regulators of
angiogenesis in lung cancer." J Leukoc Biol 57(5): 752-62.

Strieter, R. M. et al. (1995d). "The functional role of the ELR motif in CXC
chemokine-mediated angiogenesis." J Biol Chem 270(45): 27348-57.

Suter, M., Lew,A. M., Grob, P., Adema,G. J., Ackermann, M., Shortman, K.,
and Fraefel, C. (1999). "BAC-VAC, a novel generation of (DNA) vaccines:
a bacterial artificial chromosome (BAC) containing a replication-competent,
packaging-defective virus genome induces protective immunity against
herpes simplex virus 1." Proc. Natl. Acad. Sci. USA 96: 12697-12702.

Swayne, D. E. et al. (1988). "Marek's disease virus-induced transient paralysis in
chickens: alterations in brain density." Acta Neuropathol (Berl) 76(3):
287-91.

Swayne, DE., Fletcher, OJ., and Schierrnan, LW. (1989). "Marek's disease
virus-induced transient paralysis in chickens: Demonstration of vasogenic

brain oedema by an immunohistochemical method." J Comp Path 101:
451-462.

Taub, DD., Ortaldo, JR., Turcovski-Corrales, SM., Key, ML., Londo, DL., and
Murphy, WJ. (1996). "Beta chemokines costimulate lymphocyte cytolysis.
proliferation, and lymphokine production." J. Leukocfle Bio 59: 81-89.

Tischer, B. K. et al. (2002). "A DNA vaccine containing an infectious Marek's
disease virus genome can confer protection against tumorigenic Marek's
disease in chickens." J Gen Virol 83(Pt 10): 2367-76.

Tomkinson, B. et al. (1993). "Epstein-Barr virus recombinants from overlapping
cosmid fragments." J Virol 67(12): 7298-306.

Tulman, E. R. et al. (2000). "The genome of a very virulent Marek's disease virus."
J Virol 74(17): 7980-8.

Van der Walle, N and Winkler-junius, E (1924). "De Neuritisepizootie bij Kippen te
Barnveld in 1921 ." T. Vergeliik Geneesk Gezondhleer 10: 34-50.

Volpini, L. M. et al. (1995). "Stages of Marek's disease virus latency defined by
variable sensitivity to interferon modulation of viral antigen expression." m
Microbiol 47(1-2): 99-109.

137

Wager, EK. (1991). Herpesvirus transcription- general aspects. Boca Raton.

Wilson, M. R. et al. (1991). "Purification and characterization of infectious Marek's
disease virus genomes using pulsed field electrophoresis." Virology 185(2):
673-80.

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(1): 113-32.

Witter, R. L. (1991). "Attenuated revertant serotype 1 Marek's disease viruses:
safety and protective efficacy." Avian Dis 35(4): 877-91.

Witter, R. L. (1997). "Increased virulence of Marek's disease virus field isolates."
Avian Dis 41(1): 149-63.

Witter, R. L. (2001). "Protective efficacy of Marek's disease vaccines." Curr Top
Microbiol Immunol 255: 57-90.

Witter, R. L. (2002). "Induction of strong protection by vaccination with partially
attenuated serotype 1 Marek's disease viruses." Avian Dis 46(4): 925-37.

Witter, R. L. et al. (1975). "Oncogenesis by Marek's disease herpesvirus in
chickens lacking expression of endogenous (gs, chick helper factor,
Rous-associated virus-O) and exogenous avian RNA tumor viruses." J Natl
Cancer Inst 55(1): 215-8.

Witter, R. L. et al. (1970). "Isolation from turkeys of a cell-associated herpesvirus
antigenically related to Marek's disease virus." Am J Vet Res 31(3):
525-38.

Witter, R. L. et al. (1985). "Field trials to test the efficacy of polyvalent Marek's
disease vaccines in layer and broiler breeder chickens." Poult Sci 64(12):
2280-6.

Witter, R. L. et al. (1969). "Cell culture techniques for primary isolation of Marek's
disease-associated herpesvirus." Avian Dis 13(1): 101-18.

Witter, RL. (1992). Sam and comparative efficacy of the CVI988/Rispens vaccine

strain. 4th International Sympoisium on Marek's Disease, 19th World's
Poultry Congress, Amsterdam, Wodd's kPoultry Science Assoc.

Witter, RL., Sharma, JM., and Fadly, AM (1980a). "Pathogenicity of Variant
Marek's Disease Virus lsolants in Vaccinated and Unvaccinated
Chickens." Avian Dis 24: 210-232.

138

Witter, RL., Sharma, JM., and Fadly, AM. (1980b). "Pathogenicity of variant
Marek's disease virus isolants in vaccinated and unvaccinated chickens."
Avian Dis 24: 210-232.

Wu, C. A. et al. (1988). "Identification of herpes simplex virus type 1 genes
required for origin-dependent DNA synthesis." J Virol 62(2): 435-43.

Xie, O. et al. (1996). "Marek's disease virus (MDV) ICP4, pp38, and meq genes
are involved in the maintenance of transformation of MDCC-MSB1
MDV-transformed lymphoblastoid cells." J Virol 70(2): 1125-31.

Xing, 2. et al. (2000). "Expression of cytokine genes in Marek's disease
virus-infected chickens and chicken embryo fibroblast cultures."

Immunology 100(1): 70-6.

Yoshida, S. et al. (1994). "The glycoprotein 8 genes of Marek's disease virus
serotypes 2 and 3: identification and expression by recombinant fowlpox
viruses." Virology 200(2): 484-93.

Zlotnik, A., and Yoshie, O. (2000). "Chemokines: a new classification system and
their role in immunity." mmunig 12: 121-127.

Zou, P., lsegawa, Y., Nakano, K., Haque, M., Horiguchi, Y., and Yamanishi, K.

(1999). "Human herpesvirus 6 open reading frame U83 encodes a
functional chemokine." J Virol 73: 5926-5933.

139

 

   

I"EIIIIIIIIIIIIIISIIUIIljljlljll