‘2‘ 3. Sr.
.Jvmuhzu .JW. .1»:
. s .53..“ . .

14.4 o , «Witt:
f. rLunL? iminaduwfi
10.3 {1r} ‘3'
gunfire}.

2.4%.».
“was... mu. ,
gm“. ..

PS... “X

gsftr «an»...
I . $.55..-
: i: 1!...

«£13? . .
- %

gawk

!‘

V... ._ I105»:
.3
$1. I:
r}... In.
a. £923...
. . :.
$1....) I ‘ €13.13"!
I3: ti. .3 {its
. l: isfiufi! 545...?“
tv E 3:13“;
N 13": ‘tlxinlkl.
\ui . «I; ‘II.lInL‘.-it3.
.Iifluflrluules
f. \Iull
512;.i
I133»!!! 35 731:1: ‘.
.353? Iii: it
1.5-2.5... I 5...;

a 3.2.}..6
Ix ‘ £6.
. 4.... .l

:1! 1 3‘23. 2‘.
(525»!!! .
541.....ul}: i
‘téi
Cl) 1 \(

Vt“

I .93 m :4.U

 

936%....

 

C.) k- A

.o-v‘

a

 

LIBRARY
Michigan State

University

 

 

 

This is to certify that the
dissertation entitled

MAREK’S DISEASE VIRUS SUPERINFECTION:

THE EFFECT OF VIRAL DOMINANCE AND
EXPOSURE INTERVALS

presented by

John Richard Dunn

has been accepted towards fulfillment
of the requirements for the

Ph.D. degree in Pathology

 

 

)%§%¢%5§é%%%%{

Major Professof'gzéllgnature
/Z/%/07

Date

MSU is an Affirmative Action/Equal Opportunity Employer

.._.-._.—.-.-.-.-.-.-.-.-.-.-.-.-._.-.-.-.-

 

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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 KIIProjIAOOSIPresICIRCIDateDmJndd

 

MAREK’S DISEASE VIRUS SUPERINFECTION: THE EFFECT OF VIRAL
DOMINANCE AND EXPOSURE INTERVALS

By

John Richard Dunn

A DISSERTATION

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

DOCTOR OF PHILOSOPHY
Pathology

2009

ABSTRACT

MAREK’S DISEASE VIRUS DUAL INFECTION: THE EFFECT OF VIRAL
DOMINANCE AND EXPOSURE INTERVALS

By

John Richard Dunn

Marek’s disease vaccination protects chickens against the development of
disease, but does not prevent infection or transmission of infectious virus. Previous
studies have demonstrated the presence of multiple strains of Marek’s disease virus
(MDV) simultaneously circulating within poultry flocks, leading to the assumption that
individual birds are repeatedly exposed to a variety of virus strains in their lifetime.
Initial experiments were designed to evaluate whether chickens could be infected
simultaneously with two fully virulent serotype 1 MDV strains, using two similar
(er5 and er5//38CVI) and two different (J M/102W and er5//38CVI) virus pairs.
Bursa, feather follicle epithelium (FFE), spleen, and tumor samples were collected at
multiple time points to determine the frequency and distribution of each virus present
following simultaneous infection using pyrosequencing, immunohistochemistry and
quantitative PCR.

Following simultaneous dual challenge, mixtures of both virus strains were
present in bursa, F FE and tumors samples in 35 of 84 birds (42%). Dominance, as
defined by a majority frequency in samples regardless of inoculation order, was
consistently demonstrated for the more virulent er5// 3 8CVI strain when challenged
with the less virulent J M/102W strain. Dominance was not observed between the two

similar strains. Both viruses were readily detected in individual bursa and feather

 

follicles from 16 of 23 birds (70%), indicating that the presence of one virus did not
exclude the second virus. Results from tumors demonstrated differences between the
predominance of each strain within multiple tumors from individual birds, suggesting
but not proving a possible polyclonal origin. These effects may be relevant under field
conditions as no significant differences in dominance were detected in vaccinated.

Exposure to multiple MDV strains is presumably more likely to occur over time
in natural conditions, so an additional set of experiments were designed to evaluate the
effect of time between challenges on the effect of superinfection using the same virus
pairs and similar sampling protocol. The effect of time interval was strong as
superinfection was observed in 82 of 149 (55%) FF E samples following short interval
challenge (24 hours) compared to only 6 of 121 (5%) samples following long interval
challenge (13 days). Significant dominance was detected again using I M/ 102W and
er5//3 8CVI following a short interval challenge, but detection was unapparent
following the longer challenge interval. Tumor samples generally consisted of a single
predominant virus, but both viruses were detected again within a small portion of
tumors. No significant differences were detected between Virus frequencies in
vaccinated versus unvaccinated chickens.

Viral dominance and temporal relationships may be important factors that
influence the outcome of coinfection under field conditions, including the potential
outcome of emergence or evolution of more virulent strains. Understanding what
conditions exclude superinfection may help slow the evolution and establishment of

new strains within poultry flocks.

This work is dedicated especially to my wife, Mary, my kids, Isaac, Paloma and Eliana,
and my mom, Mary Ann. Thank you for your inspiration and endless support leading

up to and during writing of “the book.”

iv

ACKNOWLEDGEMENTS

I would like to thank the USDA Agricultural Research Service for their financial
support of this project. Thank you to all of the staff at the USDA Avian Disease and
Oncology Laboratory (ADOL) who helped with sampling or in necropsy at any given
time during the project, including Evelyn Young, Laurie Molitor, Melanie F lesberg,
Barry Coulson, Jody Mays, Noah Keller, and Tom Goodwill. This study also would not
have been possible without the dedicated farm staff and animal caretakers, including
Raj Kulkarni, Deb Ferguson, Chuck Clement, David White, Cecil Kirchen, and Jim
Yanz. Thank you, Barbara Riegle for teaching me how to use the cryostat, as well as to
perform immunohistochemistry, virus isolation and fluorescent antibody testing. Lucy
Lee provided one of the recombinant viruses in this study and all of the monoclonal
antibodies, Henry Hunt offered his support with flow cytometry and several samplings
and Hans Cheng introduced me to pyrosequencing, a seminal part of this study. Matti
Kiupel and Tom Wood at MSU DCPAH graciously allowed me to use their equipment
for automated immunohistochemistry staining in later experiments.

I owe special gratitude to Lonnie Milam within Bob Silva’s laboratory. Thank
you, Lonnie for willingly offering countless hours of support between teaching
techniques, helping with samplings and necropsies, sharing reagents, ordering supplies,
and providing one of the virus strains used in this study.

Four MSU veterinary students assisted at various times during multiple summers

with sampling and running assays. Thank you, Shari Gross, James Finlay, Bryan

Marker, and Jillian Glover for your dedication and patience as this project was
unfolding.

My supervisor, Aly F adly, has been extremely supportive since my first day of
work at ADOL. Thank you especially for the effort you have put into my professional

development, including supporting my travel to meetings and introductions to our

industry stakeholders.
My committee was supportive from the beginning to the end of the project. E.
Thank you, Scott Fitzgerald for serving as academic and major advisor, and occasional F

conference roommate! Thank you, Mick Fulton for your expertise and instruction about
poultry diseases and industry. Thank you, Steve Bolin for your expertise and assistance
in virology and immunology. Thank you, John Kaneene for your conversations and
contribution to the analysis and statistics.

Finally, I would like to especially thank my remaining two committee members,
Bob Silva, head of ADOL’s Marek’s disease group, and Dick Witter, my research
advisor. The level of mentorship and one-on-one time I’ve had with both of you is truly
a reflection of your pride and dedication to this field. Working with you has been

inspiring and a true pleasure.

vi

TABLE OF CONTENTS

LIST OF TABLES .................................................................................................... ix
LIST OF FIGURES ................................................................................................... xii
KEY TO ABBREVIATIONS .................................................................................... xv
CHAPTER 1
LITERATURE REVIEW .......................................................................................... 1
Relevance ...................................................................................................... 1
Marek’s disease background .......................................................................... l
Marek’s disease pathology ............................................................................. 2
Evolution of Marek’s disease ......................................................................... 5
Marek’s disease transmission ........................................................................ 9
Evidence of multiple strains within a flock ................................................... 9
Evidence of superinfection following 82 or S3 infection ............................. 11
Superinfection following attenuated 8] infection .......................................... l3
Superinfection with two fully virulent MD strains ........................................ 13
Superinfection in other herpesviruses ............................................................ 14
Prior obstacles ................................................................................................ 16
Goals of current study ...................................... . .............................................. 19
Hypotheses ..................................................................................................... 19
CHAPTER 2
PILOT STUDIES (I-III) ............................................................................................ 21
Pilot I. Methods to qualitatively and quantitatively distinguish selected
serotype 1 MDV in dual-infected chickens ................................................... 21
Abstract .............................................................................................. 21
Introduction ........................................................................................ 21
Materials and Methods ....................................................................... 22
Chickens ................................................................................. 22
Viruses ................................................................................... 23
Immunohistochemistry testing ............................................... 24
Monoclonal antibodies: H19, T65 ......................................... 25
Pyrosequencing assay ............................................................ 27
Pyrosequencing validation ..................................................... 28
Experimental design ............................................................... 30
Results ................................................................................................ 31
Summary and Conclusion .................................................................. 32
Pilot II. Pathotyping of er5 and er5//38CVI .......................................... 34
Abstract .............................................................................................. 34
Introduction ........................................................................................ 34
Materials and Methods ....................................................................... 36

vii

Chickens ................................................................................. 36

Viruses ................................................................................... 36
Experimental design ............................................................... 36
Results ................................................................................................ 37
Trial 1 ..................................................................................... 37
Trial 2 ..................................................................................... 38
Summary and Conclusion .................................................................. 38

Pilot III. A comparison between MDV production in feather pulp versus
feather follicle epithelium (F FE) ................................................................... 40
Abstract .............................................................................................. 40
Introduction ........................................................................................ 40
Materials and Methods ....................................................................... 41
Chickens ................................................................................. 41
Viruses ................................................................................... 41
Quantitative PCR assay .......................................................... 42
Experimental design ............................................................... 46
Results ................................................................................................ 46
Summary and Conclusion .................................................................. 48

CHAPTER 3
DEMONSTRATION OF VIRAL DOMINANCE FOLLOWING SIMULTANEOUS
CHALLENGE OF TWO FULLY VIRULENT MAREK’S DISEASE VIRUS

STRAINS ................................................................................................................... 76
Abstract .......................................................................................................... 76
Introduction .................................................................................................... 77
Materials and Methods ................................................................................... 81

Chickens ............................................................................................ 81
Viruses ............................................................................................... 82
Monoclonal antibodies ....................................................................... 83
Immunohistochemistry testing .......................................................... 83
Virus Isolation .................................................................................... 84
Pyrosequencing assay ........................................................................ 85
Quantitative PCR assay ...................................................................... 86
Experimental design ........................................................................... 87
Data analysis ...................................................................................... 89
Results ............................................................................................................ 90
Bursa and FFE ................................................................................... 90
Tumors ............................................................................................... 91
Effect of vaccination .......................................................................... 92
Discussion ...................................................................................................... 93
CHAPTER 4

THE EFFECT OF THE TIME INTERVAL BETWEEN EXPOSURES ON THE

SUSCEPTIBILITY OF CHICKENS TO SUPERINFECTION WITH MAREK’S

DISEASE VIRUS STRAINS .................................................................................... 111
Abstract .......................................................................................................... lll

viii

Introduction .................................................................................................... 1 12

Materials and Methods ................................................................................... 116

Chickens ............................................................................................. 116

Viruses ............................................................................................... 117

Monoclonal antibodies ....................................................................... 1 l8

Immunohistochemistry testing .......................................................... 1 l8

Virus isolation .................................................................................... 1 19

Pyrosequencing assay ........................................................................ 120

Quantitative PCR assay ...................................................................... 121

Experimental design ........................................................................... 122

Data analysis ...................................................................................... 125

Results ............................................................................................................ 126

Feather follicle epithelium ................................................................. 126

Tumors ............................................................................................... 128

Effects of vaccination ........................................................................ 130

Summary Data ................................................................................... 132

Discussion ...................................................................................................... l 33
CHAPTER 5

SUMMARY AND FUTURE STUDIES ................................................................... 154

REFERENCES .......................................................................................................... 160

ix

LIST OF TABLES

Table 2-1: Reconstruction mixtures used for pyrosequencing validation ................. 50
Table 2-2: Challenge outline ..................................................................................... 50
Table 2-3: Sample outline ......................................................................................... 51
Table 2—4: Gross lesions at termination .................................................................... 52
Table 2-5: IHC and pyrosequencing results from samples collected 6 dpi .............. 53
Table 2-6: IHC results from F FE samples collected 21 dpi ..................................... 54
Table 2-7: Challenge outline for pathotyping assay .................................................. 55
Table 2-8: %MD in replicates 1 and 2 (R1 & R2) from trials 1 and 2 ...................... 56
Table 2-9: %Protection from trials 1 & 2 plus combined results .............................. 57
Table 2-10: Challenge outline ................................................................................... 58
Table 2-11: Results of paired t-test statistics for the mean difference of FF E minus

feather pulp VCN/host cells by virus and by days post challenge ............................. 59
Table 3-1: Challenge outline ..................................................................................... 98

Table 3-2: Experiment 1A. Pyrosequencing and IHC results from bursa (6 dpi)
and F FE (21 dpi) samples in chickens single or dual-challenged with J M/102W (J)
and/or erS//38CVI (M2) ......................................................................................... 99

Table 3-3: Experiment 1B. Pyrosequencing and IHC results from bursa (6 dpi)
and FFE (21 dpi) samples in chickens single or dual-challenged with er5 (M1)
and/or er5//38CVI (M2) ......................................................................................... 100

Table 3-4: Experiments 2A & 3A. Virus isolation and pyrosequencing results
from single and dual-challenged chickens with JM/102W (J) and/or
er5//38CVI (M2) .................................................................................................... 101

Table 3-5: Experiments 2B & 3B. Virus isolation and pyrosequencing results from
single and dual-challenged chickens with er5 (M1) and/or er5//3 8CVI (M2)... 102

Table 3-6: Experiment 4A. Pyrosequencing results from multiple FFE and/or
tumor samples in vaccinated and unvaccinated chickens challenged simultaneously
with J M/102W (J) and er5//38CVI (M2) ............................................................... 103

Table 3-7: Experiment 48. Pyrosequencing results from multiple FFE and/or
tumor samples in vaccinated and unvaccinated chickens challenged simultaneously

with er5 (M1) and er5//38CVI (M2) .................................................................. 104
Table 3—8: Summary — viral predominance .............................................................. 105
Table 4-1: Outline of treatment groups for experiments 1 and 2 .............................. 138
Table 4-2: Outline of treatment groups for experiments 3 and 4 .............................. 139

Table 4-3: Experiment 1 - FFE. IHC and pyrosequencing results of FFE samples
from single-challenge, dual-challenged, and contact chickens exposed to JM/102W (J)
and/or er5//38CVI (M2) [Exp 1A] or er5 (M1) and/or M2 [Exp 1B] ................ 140

Table 4-4: Experiment 3 - FF E. Number of birds with FFE samples positive by
pyrosequencing based on order of inoculation for each virus in single, dual-challenged
or contact chickens exposed to J M/102W (J) and/or er5//38CVI (M2) [Exp 3A] or
er5 (M1) and/or M2 [Exp 3B] ................................................................................ 142

Table 4-5: Experiments 3 & 4 — Failed initial results. Number of birds positive
by pyrosequencing for each virus inoculated in long-interval dual-challenged chickens
from those that had failed results from the first sampling taken immediately before
inoculation of the second virus .................................................................................. 143

Table 4-6: Experiment 2A — Tumors. Pyrosequencing (tumor) and virus
isolation (spleen) results from single and dual-challenged chickens challenged
with J M/102W (J) and/or er5//38CVI (M2) and contact recipient chickens ......... 144

Table 4-7: Experiment 28 - Tumors. Pyrosequencing (tumor) and virus
isolation (spleen) results from single and dual-challenged chickens challenged
with er5 (M1) and/or er5//38CVI (M2) and contact recipient chickens ............. 145

Table 4-8: Experiment 2 — Comparison of multiple tumors. Pyrosequencing
results of birds with multiple tumors in experiment 2. Birds were dual-challenged
with either J M/102W (J) and er5//38CVI (M2) or er5 (M1) and M2 ................ 146

Table 4-9: Experiment 3 - Tumors. Number of birds with tumor samples positive
by pyrosequencing based on order of inoculation for each virus in single, dual-
challenged or contact chickens exposed to J M/ 102W (J) and/or er5//38CVI (M2)
[Exp 3A] or er5 (M1) and/or M2 [Exp 3B] ............................................................ 147

xi

Table 4-10: Experiment 4 - FFE. Number of birds positive by pyrosequencing
for each virus inoculated in vaccinated and unvaccinated dual-challenged chickens
by order of inoculation for each virus ........................................................................ 148

Table 4-11: Experiment 4 - Tumors. Number of birds with tumor samples
positive by pyrosequencing based on order of inoculation for each virus in single,
dual-challenged or contact chickens exposed to JM/102W (J) and/or

er5//38CVI (M2) [Exp 3A] or er5 (M1) and/or M2 [Exp 38] ........................... 149

Table 4-12: Summary table. Combined pyrosequencing data displaying the
number of samples where the second-challenged virus was present in F F E or
tumor samples ............................................................................................................ 150

xii

LIST OF FIGURES

Figure 2-1: Sequence of pp38 gene indicating pyrosequencing primer binding

Sites ............................................................................................................................ 60
Figure 2-2: Gel analysis of pp38 gene cloned into plasmid .................................... 61
Figure 2-3: Averaged pyrosequencing results from two sets of independent

plasmid mixtures (trial 1 & 2) each run in duplicate ................................................. 62
Figure 2-4: Gel analysis of pp38 gene plasmid serial dilutions ............................... 63

Figure 2-5: Pyrosequencing results following nested PCR of low plasmid
Quantities ................................................................................................................... 64

Figure 2-6: Pyrosequencing results following nested PCR comparing the effect
of using samples diluted in chicken DNA ................................................................. 65

Figure 2-7: Serial bursa samples stained with monoclonal antibodies H19 and
T65 at 6 dpi from a bird singly infected with erS and er5//38CVI ..................... 66

Figure 2-8: Serial FFE samples stained with monoclonal antibodies H19 and
T65 at 21 dpi from a bird singly infected with er5 and er5//38CVI ................... 67

Figure 2-9: Serial bursa samples stained with monoclonal antibodies H19 and
T65 at 6 dpi fi'om a bird dually infected with erS and er5//38CVI .................... 68

Figure 2-10: Serial FFE samples stained with monoclonal antibodies H19 and
T65 at 21 dpi from a bird dually infected with er5 and er5//38CVI .................. 69

Figure 2-11: Average copy number for all viruses combined comparing feather
pulp and F FE by sampling time ................................................................................. 70

Figure 2-12: Laser microdissection results from sections of frozen feather
follicles collected at 31 days post challenge .............................................................. 71

Figure 2-13: Average isolator virus copy number isolated from dust based on
virus and sampling time ............................................................................................. 72

Figure 2-14: Viral copy number group comparisons for each virus based on
sampling time ............................................................................................................. 73

Figure 2-15: Virus copy number over time in individually-housed chickens ........... 74

xiii

Figure 2-16: Virus copy number comparison for individually housed chickens
at 21 days, challenged with er5 .............................................................................. 75

Figure 3-1: Experiment 1. Immunohistochemistry staining of serial FF E
sections 21 dpi from a chicken simultaneously challenged with J M/ 102W and
er5//3 8CVI ............................................................................................................. 106

Figure 3-2: Experiment 1. Immunohistochemistry staining of serial bursa
sections 6 dpi from a chicken simultaneously challenged with erS and
er5//38CVI ............................................................................................................. 107

Figure 3-3: Experiment 1. Average quantitative PCR results from DNA isolated
from bursa sections sampled at 6 dpi from birds singly or dually infected with
J M/ 102W (J), er5 (M1), or er5//38CVI (M2) ..................................................... 108

Figure 3-4: Experiment 1. Average quantitative PCR results from DNA isolated
from F FE sections sampled at 21 dpi from birds singly or dually infected with
JM/102W (J), er5 (M1), or er5//38CVI (M2) ..................................................... 109

Figure 3-5: Experiment 4. Average quantitative PCR results from DNA

isolated fiom F FE sections sampled at 13, 26, and 63 dpi from birds vaccinated

or unvaccinated with HVT and dually infected with either J M/102W (J) and
er5//38CVI (M2), or er5 (M1) and er5//38CVI (M2) ..................................... 110

Figure 4-1: Experiment 1 - Distribution of short interval virus frequencies
within exclusively predominant and mixed samples analyzed by IHC ..................... 151

Figure 4-2: Experiment 1 - Distribution of short interval virus frequencies
within exclusively predominant and mixed samples analyzed by pyrosequencing... 152

Figure 4-3: Experiment 4 — Results of qPCR from F FE in vaccinated or

unvaccinated chickens dually-challenged with J M/102W (J) and er5//38CVI
(M2), or er5 (M1) and er5//38CVI (M2) ............................................................ 153

xiv

 

KEY TO ABBREVIATIONS

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ADOL Avian Disease and Oncology Laboratory
dpi Days post infection
DNA Deoxyribonucleic acid
F FE Feather follicle epithelium
HVT Turkey herpesvirus (vaccine)
IHC Immunohistochemistry
J J M/102W
M1 er5
M2 er5//38CVI
Mab Maternal antibody
MD Marek’s disease
MDV Marek’s disease virus
pp38 MDV phosphoprotein pp38
qPCR Quantitative polymerase chain reaction
S1 Serotype 1 MDV
82 Serotype 2 MDV
S3 Serotype 3 MDV
SNP Single nucleotide polymorphism
VCN Virus copy number
VMDV Virulent MDV
vaDV Very virulent MDV

 

XV

 

Chapter 1
LITERATURE REVIEW
Relevance
Superinfection is defined as an additional infection occurring during the course

of an existing infection. Superinfection with two virulent Marek’s disease virus (MDV)
strains is likely an important step in the pathway towards establishment of evolved
strains within a population. Demonstrating whether superinfection occurs and the
conditions that promote infection and transmission of a second MDV strain would

provide knowledge to help slow evolution of the virus within poultry flocks.

Marek’s disease background

Marek’s disease (MD) is a lymphoproliferative disease in chickens caused by an
alphaherpesvirus, MDV. The disease is typically characterized by enlarged nerves and
visceral tumors composed of transformed T cells. Marek’s disease (MD) remains an
important problem in poultry throughout many parts of the world, due to condemnation
losses and the cost of vaccination. As recently as 2002, an outbreak of Marek’s disease
in Indonesia was reported to have killed 2.8 million chickens fi'om two major chicken
suppliers with an estimated loss of $21M for the breeders (90). MDV is classified into
three serotypes: serotype 1 (S 1) includes pathogenic strains and their attenuated
variants, serotype 2 (S2) includes all apathogenic chicken strains, and serotype 3 (83)
includes turkey herpesvirus (HVT), which is also apathogenic in chickens (12).

Serotypes 2 and 3 have been used as vaccines, as well as several attenuated

strains of 81. Although vaccination has successfully been used to prevent disease, the

vaccines do not prevent against infection or transmission of the virus. Witter et al.
reported that in birds vaccinated with HVT at hatch challenged with J M/ 102W at day
15, MDV was not isolated from vaccinated and challenged birds until 14 days after
challenge, compared to 3 days in non-vaccinated challenged birds (104). Calnek et al.
similarly reported on the effects of vaccination with HVT (500 pfu) at hatch followed
by J M-lO virus (500 pfu) infection 17 days later (21). According to the authors,
splenomegaly associated with J M-10 infection was prevented, viremia levels were
markedly reduced, and expression of virus internal antigens in lymphoid organs was
completely prevented. Thus, if vaccinated birds are less susceptible to virus replication,

can one still demonstrate S1 superinfection in such populations?

Marek’s disease pathology

Factors that influence pathogenesis of MD include age at exposure, maternal
antibody, genetic constitution, age resistance, strain of virus, sex, and preceding or
concurrent infections with immunosuppressive agents. Superinfection may be an
additional factor that influences the pathogenesis of MD. Infection with MDV can be
divided into four phases: 1) Early cytolytic infection, 2) Latent infection, 3) Late
cytolytic infection, and 4) Transformation. The stages of infection with serotype 1
MDV have been described in multiple reviews (7, 66, 70, 71).

Early cytolytic infection occurs in the primary lymphoid organs, consisting of
the bursa of Fabricius (bursa), thymus and spleen. The cytolytic infection appears at
approximately 3 days post infection, peaking at 5-7 days and resolving by about 2

weeks (72). Schat et al. reported that embyronal bursectomy eliminated early cytolytic

infection, although not latent infection (83). Shek et a1. supported this finding by
demonstrating that B cells were the principal targets of MDV during early cytolytic
infection (85). Early cytolytic infection is subdued or absent in birds with maternal
antibody (15, 72) or prior vaccination (17, 82, 87).

At approximately 7-8 days post infection, early cytolytic infection switches to
latent infection, defined as the presence of viral DNA in the absence of viral transcripts
and proteins. Viremia refers to the latently infected peripheral blood lymphocytes. The
viral genome persists with no viral or tumor antigen expression, and there is no virus
production except following reactivation. Interestingly, the predominant lymphocyte
for latent infection are activated T cells, with only a few latently infected B cells (22).
It is unclear what causes the switch from cytolytic infection to latency, but reactivation
of the viral genome can occur by both in vitro cultivation and by inoculation of
susceptible chickens.

A second cytolytic phase of infection has been reported from 14-21 dpi,
affecting the thymus, bursa, kidney, adrenal gland, proventriculus, and some epithelial
tissues including F F E (16). Calnek and Hitchner first reported the predilection of MDV
to the FFE, describing fluorescent staining of antigen varying from a few stained cells
in the superficial layers to heavy staining several cell layers deep surrounding the
feather shaft (20). Replication of MDV in the FFE is fully productive and creates large
numbers of enveloped, fully infectious virions (18). As early as 2 weeks after
inoculation or exposure, MDV is spread in the feather dust to other chickens as the
keratinized layer of infected epithelial cells is sloughed and shed (17). Shedding

appears to peak at approximately 3-5 weeks and continues indefinitely (94, 106).

 

Infected feather dust can remain infectious in a poultry house for at least several months
at 20-25°C and for years at 4°C (17, 58). The fully productive replication of MDV in
FF E has made feather tips a useful source for virus detection (18, 35, 36, 74).

Within feathers, lymphoid lesions have been described that developed in the
feather pulp. Fujimoto et al. described two types of lesions: R-type (non-tumorous

response) and T-type (tumorous proliferation) (39). Ten years later, Moriguchi et a1.

further classified the lesions into three types: Rl-type — variable infiltration by small

lymphocytes; ’Rz-type — edema and infiltration by plasma cells and small lymphocytes;

and T-type — proliferation usually by medium lymphocytes or blast cells (64). Cho et
al. have confirmed similar types of lesions using feather pulp cytology, and concluded
feather pulp was an effective diagnostic and prognostic tool for MD incidence (29).
The relationship between virus load in feather pulp and pathogenesis has not been
demonstrated, but Bai gent et a1. recently reported quantitation of MDV genomes in
feather tips with copy numbers considerably greater than in the spleen and peripheral
blood lymphocytes (3).

Lymphoid proliferation can begin as early as 1 week post infection and can
produce grossly visible tumors from as early as 2 weeks post infection (73). Tumors
typically consist of pleomorphic lymphoid cells, ranging from small lymphocytes to
large blast cells. Common sites affected include spleen, heart, gonads, liver, kidneys,
proventriculus, intestine, lung, thymus, bursa, skeletal muscles, skin, and peripheral
nerves. The skin, therefore, is associated with both dermal lyrnphocytic infiltration and
FF E late cytolytic infection. Ross et al. reported that although nearly all T cells in MD

lymphomas contain the MDV genome, most cells appear latent because only a small

number of cells produce transcripts (80). The authors reported that among others, Meq
transcripts were abundant in purified lymphoma cell populations, but pp38 transcripts
were not. Using a knockout recombinant virus, Lupiani et al. demonstrated that Meq is
involved in lymphocyte transformation, but not lytic replication in chickens (62).
Gimeno et al. showed that MD tumors have much higher amounts (~100-fold) of MDV
DNA compared to latently infected tissues (42).

There is current debate on the clonality of MD tumors. Delecluse et al.
performed in situ hybridization to examine MDV genome structure and its relationship
to host DNA in MD lymphomas (3 7). The authors reported that tumors were
monoclonal based on the identical number of hybridization signals of interphase nuclei
and the identical pattern of chromosomal integration in metaphase spreads. In contrast,
Burgess and Davison have suggested a polyclonal tumor origin based on their analysis
of T cell receptor (TCR) variable [3 chain gene family expression (13). A recent abstract
presented by Cheng et al., also suggested that MD tumors were not necessarily clonal
based on TCR mRNA spectratyping (24). In an experiment evaluating expansion of a
132-bp repeat region, Silva et al. demonstrated differences in virus populations in two
tumors from one bird (86). Pathology also favors polyclonal tumors, as there are
lymphoid proliferations in many tissues at a very early time without evidence for a

single focus and subsequent metastasis.

Evolution of Marek’s disease
A major concern facing the poultry industry is the evolutionary trend of MDV

towards greater virulence (95). Two waves of evolution have been described in a recent

review (65). The first wave of virus evolution from mild (mMDV) to virulent (vMDV)
strains was during the 1950’s and was likely due to the transformation of the industry to
highly intensive poultry practices. The continuous availability of large populations of
susceptible birds is certainly a major contributor. Subsequent evolution has been
attributed to the introduction of successive generations of MD vaccines (92). Because
MD vaccines fail to produce a sterilizing immunity, viruses are able to replicate and
shed within vaccinated hosts.

Evolution of strains within a flock could begin from an internal or external
source. A more virulent mutated virus may develop spontaneously within a resident
bird, or a more virulent virus could arrive from an outside source, such as introduction
of new birds or from a neighboring flock via fomites or contaminated personnel. By
either source, this new virus needs the proper fitness traits to become established and
for transmission to other birds. Once multiple strains become established in a flock,
birds may become repeatedly exposed to superinfection.

What properties are necessary for one virus strain to have a competitive
advantage over another strain? Virulence has historically been defined by oncogenicity.
The pathotyping classification has been used to designate differences in virulence of
field viruses based on frequency of MD lesions induced in vaccinated and unvaccinated
birds. Techniques for pathotyping have evolved and were recently reviewed by Witter
et al. (98). Pathotyping became an important classification of viruses not long after
increased virulent forms of MDV were initially reported (5). Early classifications by
Biggs et al. differentiated strains as classical or acute, with the term acute based on

higher morbidity and mortality, shorter time until onset, and a higher incidence of

visceral tumors (11). Biggs et al. later included additional criteria for this designation
based on histological lesion scores (9). Witter first designated the virulence of MDV
strains based on responses in vaccinated chickens, using a cut-off value of HVT
protection (77%) (93). Several years later he included prototype isolates as a
comparison against field isolates to better standardize the pathotype designations from
other laboratories (97). Specifically, WMDV (very virulent) isolates were those with
MD lesions in HVT-vaccinated, susceptible birds at a rate greater than a prototype
vMDV(virulent) isolate, such as J M or GA. Other groups began using these techniques,
but the procedures were variable with different prototype viruses and chicken lines used
for making the designation of vaDV (14, 50, 51, 89). Witter recognized an increased
virulence of MDV strains in the early 1990’s in bivalent-vaccinated chickens, leading to
four current pathotype designations: m (mild), V (virulent), vv (very virulent), and w+
(very virulent plus) (95). It still remains difficult to distinguish between similar strains
or to be certain that two strains do not differ.

Perhaps virulence is not defined by oncogenicity, but by the increased potential
for replication and/or transmission. Citing the work of Gandon et al., Nair has
suggested that since MDV vaccine does not provide sterile immunity, a partially
effective vaccine may be leading to virus evolution with increased transmission
potential and adaptation to vaccinated hosts (40, 65). The topic of replication potential
and its relationship to virulence has been the topic of several recent papers. Yunis et al.
infected resistant and susceptible chickens with v and vv+ MDV strains and measured
copies of DNA or RNA transcripts during 1-10 days post infection (dpi) using qPCR

and qRT-PCR (109). Viral DNA increased with initiation of viral transcription for both

virus strains and chicken strains. The virulent virus became latent at 6 dpi, but the vv+
virus never went into latency in both chicken lines during the 10 days. The prolonged
virus replication and presence of viral transcripts for the vv+ virus throughout 10 days
could cause more severe damage and atrophy of lymphoid organs.

Using quantitative RT-PCR and PCR to measure lytic viral transcription and
DNA replication, J arosinski et al. showed significantly higher levels of viral replication
for the vv+ strain RK-l in two chicken lines compared with the v strain J M-16 (57).
Calnek et al. demonstrated a prolonged phase of cytolytic infection at 7-8 dpi for three
vv+ isolates (RK-l , 584A, 648A) compared with two v isolates (J M16, GAS) (19).
Gimeno et al. compared higher and lower protective vaccine pairs originating from the
same MDV strain and found better in viva replication for the higher protective strains
(45).

Past studies have correlated viremia to tumor development (106). Recent work
using quantitative PCR assays have confirmed the associations. Gimeno et al. evaluated
the load of MDV in peripheral blood at 3 weeks with subsequent development of MD
tumors and found chickens that developed tumors had significantly higher MDV load in
peripheral blood compared to chickens that did not develop tumors (41). The authors
also noted that HVT virus load in peripheral blood at 3 weeks had no correlation with
subsequent tumor development. In a similar study using quantitative PCR, Islam et al.
reported the mean MDV copy number in peripheral blood was greater in chickens
subsequently exhibiting gross MD lesions than in chickens with no lesions at all four
time points of 14, 21, 28, and 35 days post challenge (55). The bottom line may be that

evolution is driven by the ability of replication, which is related to oncogenicity.

Marek’s disease transmission

Transmission of MDV occurs through the production of fully infectious virions
in the feather follicle epithelium (F FE) that are spread in the feather dust as early as 2
weeks after exposure and then inhaled by other birds (18). Carrozza et al. demonstrated
the virus in the debris of dead stratified epithelial cells and moulted feathers with
infected cells attached (23). The authors proposed that highly infectious but labile cell-
free virus particles were associated with skin debris, whereas keratin-wrapped particles

were less infectious but more stable in the environment. Shedding appears to peak at
approximately 3-5 weeks and continues indefinitely at quantities as high as 109 virus

copies per chicken per day, irrespective of vaccination status (53, 94, 106). Infected
feather dust can remain infectious in a poultry house for at least several months at 20-
25°C and for years at 4°C (17, 58). The continuous transmission, long survivability of
the virus, and large population of birds favor a significant viral load within a poultry

house.

Evidence of multiple strains within a flock

There is evidence that multiple strains of MDV can circulate simultaneously
within the same flock. Biggs et al. reported on the epidemiology of MD in six flocks
from a broiler female grandparent line (10). Virus isolates were propagated from five
flocks, and then 25 isolates were examined and classified for their pathogenicity as

either apathogenic, classical or acute. Interestingly, four of the five flocks had strains

present from more than one level of pathogenicity. Two flocks, including a single house
from one flock, had strains present from all three levels of pathogenicity.

In a follow-up study the next year, Bi ggs et al. recorded mortality from a flock
of 8000 female broiler breeders ages 8-22 weeks to determine the effect from supply
flock, rearing house/pen, and production house/pen (8). The study was based on two
supply flocks, placed in pens at three rearing houses, and then at 8-9 weeks each pen
was divided and placed in one of two production houses. The authors reported that only
the rearing house pen of origin significantly affected the incidence of MD. Further, the
authors concluded the levels of mortality were likely determined by the events
occurring at the time of primary infection and the variable incidence of MD was due to
a complex interrelationship between strains of virus of variable pathogenicity. This
mixture resulted in natural vaccination by apathogenic strains prior to exposure to
infection with pathogenic strains in some groups and not in others.

Jackson et al. followed up on the interrelationship between strains of variable
pathogenicity (56). The authors confirmed the presence of multiple pathotypes
circulating within the same flock and that the environment during the first 8-9 weeks of
life was of greatest importance in subsequent mortality from MD. MD mortality was
closely associated with the sequence of infection and frequency of isolation of viruses
of differing virulence. For example, three pens that experienced high mortality had
early isolates that were of the acute type, whereas two different pens with low incidence
of MD had apathogenic strains that predominated throughout early and late isolations.

Witter et al. evaluated the virulence of multiple isolates (2-3) from two separate

flocks and found very similar pathotypes (95). These isolates were all from tumor-

10

bearing birds which may have created a bias towards isolation of more virulent strains.
This also suggests, however, that a single Virus strain may have the ability to become

dominant. The studies described above provide evidence for multiple strains of MDV
circulating with a flock, but don’t address the possibility of multiple strains within the

same bird.

Evidence of superinfection following 82 or S3 infection

Superinfection has been demonstrated following initial infection with MD
vaccine strains. Witter et al. showed that 8] virus was isolated when MDV was
challenged after initial infection with S3 turkey herpesvirus (HVT) (104). Okazaki et
al. evaluated the temporal relationship between HVT vaccination and MDV challenge
(69). The authors concluded that HVT was fully protective when administered at least
one week before challenge, and the vaccine gave partial protection when administered
simultaneously with challenge at one day of age (25% MD vs 90% MD in unvaccinated
challenged group).
Calnek et al. infected birds with 82 SB-l and S3 HVT-4, followed by JM-lO, to study
the effect of vaccination on early pathogenesis (21). Viremia was present, although
reduced following subsequent challenge.

Cho et al. reported evidence of coinfection with acute (Id-1) and mild (HN)
MDV strains following various intervals between challenge, although HN was later
classified as a nonpathogenic 82 strain (28, 81, 93) Cho and Kenzy reared several
chicken lines in pens with contaminated litter from two groups of birds that had been

challenged with either Id-l or HN MDV strains (27). Blood was collected weekly from

11

1-6 weeks post exposure and strains were identified by viremia based on plaque
morphology. The acute strain alone was first to be identified in all chicken lines at 1 or
2 weeks post exposure. By three weeks, dual infection was identified in several birds
which generally persisted through 6 weeks. In a few cases the mild strain was singly
identified beginning at 3 weeks. The authors concluded that infection with one strain,
whether acute or mild, did not exclude subsequent infection by the other strain.

In another study, Cho challenged birds by subcutaneous injection singly and
dually with ld-l and HN strains of MDV to compare viremia responses between
resistant and susceptible lines of chickens (26). Based on plaque morphology, Cho
identified both strains in susceptible and resistant birds. In susceptible birds, isolates
with both strains were present at 1 week continuing through termination at 8 weeks,
with gradual reduction from 6 mixed isolates down to 1 mixed isolate at 8 weeks. In the
resistant line of birds, the mixed isolates were present at 1 week but only lasted through
4 weeks, after which only the mild strain was isolated at 6 and 8 weeks. Interestingly,
in the resistant line, the viremia persisted in all birds singly challenged with the mild
strain, but challenge with the acute strain resulted in a decrease in viremic birds after 2
weeks.

Cho then studied the relationship of MDV simultaneous challenge in blood and
feather follicles using the same strains described above, Id-l and HN (25). Based on
plaque morphology, dual viremia was detected in all of the susceptible chickens at 2, 4,
and 6 weeks post challenge, but in less than half of the resistant chickens. Not all

feather tip extracts had detectable virus, but consistent with his prior experiment,

12

feathers with dual virus maturation had more pathogenic than apathogenic MDV in

susceptible chickens and more apathogenic MDV in resistant chickens.

Superinfection following attenuated 8] infection

In addition to S1 superinfection following initial infection with S2 or S3 strains,
superinfection has also been demonstrated following initial infection with attenuated S1
strains. Churchill et al. illustrated that 81 viruses can cause infection by infecting birds
with live attenuated Sl virus (HPRS-l6/att) followed by more virulent strains (3 O).
Rispens soon after demonstrated infection after initial challenge with the avirulent
CVI988 strain (79). Witter and Kreager also recently demonstrated infection and MD
lesions after initial infection with a series of fully attenuated S1 strains, with increased

MD prevalence proportionate to the virulence of subsequent challenge strains (100).

Superinfection with two fully virulent MD strains

Laboratory-based evidence that superinfection can occur with two fully virulent
S1 MDV strains is limited. Ianconescu et al. attempted to evaluate the effect of early
low virulent natural exposure on subsequent challenge with virulent J M MDV (48).
Nonvaccinated farm-reared chickens were removed from the farm and challenged with
J M at 5, 7 or 12 weeks of age. Uninoculated groups developed MD tumors and
antibodies after isolation, confirming natural exposure on the farm. At 12 weeks, fann-
reared chickens inoculated with J M had no difference in tumor incidence compared to
unchallenged birds. Thus, there was no clear evidence that superinfection occurred or

had an effect.

13

Witter and Gimeno studied the affect of MDV superinfection in adult birds,
specifically asking the question if chickens exposed to MDV early in life were
susceptible to a highly virulent strain as an adult (99). Virus exposed chickens were
vaccinated with HVT at hatch plus challenged with JM/102W at 5 weeks. When virus
exposed chickens were challenged with a highly virulent strain as adults at 18 weeks,
they generally failed to develop transient paralysis or tumors. Not only were the virus
exposed groups refractory to lesion indication and transient paralysis, but the viremia
responses and transmission to contacts appeared minimal at best, indicating that these
birds had a much reduced susceptibility to infection. The authors concluded that late
outbreaks in the field are not likely due to superinfection, but are more likely triggered
by unknown environmental factors. Thus, with a long time interval between exposures,
both studies failed to detect any significant superinfection.

Reports from the field have suggested that superinfection of virulent MDV
strains is possible, based on apparent adult-to-adult transmission of w+ MDV with very
late outbreaks occurring, even after molt (5 9). It would seem that superinfection must

occur for the establishment of new strains within flocks.

Superinfection in other herpesviruses

Superinfection has been studied using other herpesvirus models. Richter et al.
recently evaluated feline herpesvirus type 1 (FeHV-l) superinfection in cats latently
infected with wild-type FeHV-l (77). Nine SPF cats were inoculated with wild-type
FeHV—l and superinfected 8 months later with a recombinant F eHV-l virus with green

fluorescent protein substituted for glycoprotein G (rFeHV-l AgG/GFP). The

14

recombinant virus produced a temporary local lymphadenopathy, but otherwise
produced no clinical or ocular signs of infection. Cyclophosphamide-dexamethasone
treatment was administered 16 months after superinfection to stimulate reactivation,
which was successful for the underlying theHV-l infection but no reactivation of the
superinfecting recombinant virus was detected. The study unfortunately lacked a
control group of cats infected only with rFeHV-l AgG/GFP to confirm both strains were
fully pathogenic.

A study of patients in Malawi infected with human herpesvirus-8 (HHV-8), also
known as Kaposi sarcoma-associated herpesvirus, analyzed sequence variation at 3 loci
of the virus genome (6). The authors determined there was significant
intraperson/intersample and intrasample sequence polymorphisms in 60% of patients
with amplifiable HHV-8 DNA, implying HHV-8 superinfection.

A review of multiple infection of Epstein-Barr virus (EBV), a y-herpesvirus, in
healthy humans cited 9 references documenting multiple infections ranging from 2-
100% prevalence among tested individuals (91). In the current study, the authors
developed a genotyping technique based sequence variation in the EBV LMP-l gene
and found 2 of 9 subjects harboring multiple EBV infections. One subject typed for two
different genotypes among simultaneously collected saliva and blood specimens at two
time points.

Rickabaugh et al. constructed a recombinant murine gammaherpesvirus-68
(MHV-68) that overexpresses a viral replication and transcription activator (78). Along
with a reduction in disease, the recombinant MHV-68 had faster replication kinetics in

vitro and in vivo, and when simultaneously challenged to mice with wild-type MHV—68

15

was dominant and protected from secondary infection. This reference suggests that
inherent dominance, or an inherent fitness trait, of one virus over another may be
important in the outcome of infections with more than one virus strain.

A study evaluating in vitro recombination potential in cells superinfected with
two bovine herpesvirus-1 (BoHV— 1) recombinants demonstrated a relatively high level
of superinfection and recombination when viruses were administered within two hours,
but very little superinfection or recombination after two hours (63). The authors
suggested there was an establishment of a barrier which reduced or prevented
superinfection within the same cell, thus preventing the generation of recombinant
viruses. This is an important fact that may influence the rate of virus evolution
depending on infection interval in vivo. In addition to potential effect of inherent
dominance, temporal effects may also be an important factor governing the outcome of

superinfection.

Prior obstacles

Prior obstacles have made the study of virulent Sl coinfection difficult using
DNA and antigen as methods for critical analysis, particularly because of the difficulty
distinguishing between S1 viruses. Previous studies have relied on plaque morphology
or changes in total MD lesions to make assumptions about coinfection. The recent
development of mutant strains with slight differences in the MDV phosphoprotein pp38

gene has provided a convenient method for distinguishing between two similar (and two

different) fully virulent strains. The function of the pp38 gene, located in the IRL and

UL region, is not completely understood, but several studies have contributed to our

16

current knowledge. Xie et al. concluded that pp3 8, in addition to ICP4 and meq were
involved in maintenance of MDV-transformed MSBl lyrnphoblastoid cells (1 O7). Cui
and Qin described immunodepressive effects of pp38 (32). Several experiments have
indicated that pp38 may be an important part of MDV reactivation. Infection of QT35
cells, a cell line latently infected with serotype 1 MDV, with HVT activates pp38
expression (108). The gene is also expressed in productively infected cells including the
FFE (33, 49, 67, 68, 76). Gimeno et al. used a mutant virus lacking pp38 (er5App38)
to demonstrate that pp38 was necessary to establish cytolytic infection in B cells, but
not FF E (43). The authors found that pp38 was also necessary to produce an adequate
level of latently infected T cells and to maintain the transformed status in viva by
preventing apoptosis.

The recombinant er5 strain was generated by Reddy et al. using overlapping
cosmid clones produced fi'om wild type Md5, a vaDV strain (75, 93, 103). The
recombinant virus er5//38CVI was subsequently produced by Lee et al. using the
er5 cosmid clones, substituting the pp38 gene from CVI988/Rispens (60). Thus,
these two viruses share the same er5 backbone and are only different in the pp38 gene
which differs by two single nucleotide polymorphisms (SNPs). Differentiation between
these two viruses is possible by detecting the SNPs coupled with the availability of
monoclonal antibodies that detect the difference. The two strains were also usefirl
based on their similar pathogenicity, as reported by Gimeno et al., in which there were
no differences in the frequency of lymphoproliferative lesions between the viruses at
either 6 or 15 wpi (43). Nearly all other S1 MDV strains share the same pp38 sequence

as er5 and therefore could all be differentiated from er5//38CVI. In this study,

17

J M/102W was chosen to pair with er5//38CVI as the second virus pair because of
differing pathogenicity. The parental strain, J M, was isolated by Sevoian in 1962 (84),
and later cloned by 3X endpoint passage in DEF and designated as JM/102W (88).

J M/102W was pathotyped as vMDV by Witter and is a principle prototype strain
frequently used in MD experiments (93, 95).

Monoclonal antibodies (Mab) were the other critical technology used to detect
and differentiate the viruses, in addition to sequence alone. According to early
published pp38 sequence data, all serotype 1 MDV strains have identical bases in the
pp38 gene except bases #320 and #326 (31, 38). CVI988, however, is the only serotype
1 MDV strain that is not reactive with pp38-specific Mab H19 (61, 105). Using DNA
sequencing from the pp38 gene of additional strains, Cui et al. determined the molecular
basis for the difference in the antigenic specificity of H19 and T65 antibodies for the
pp38 gene (34). Mab H19 was specific to base pair #320, and was only positive when
adenine (A) was present. All serotype 1 MDV strains except CVI988 were positive for
H19 and had A at bp #320. Mab T65 was specific to base pair #326, and was only
positive when guanine (G) was present. T65 was positive for CVI988 which had G
present at base pair #326. Most serotype 1 MDV strains were positive for H19 only,
having A at both base pairs #320 and #326, however, a few field strains have A at base
pair #320 and G at base pair #326, such as GA22, and thus are positive for both H19
and T65. Pyrosequencing was designed to detect the differences at base pair #320, thus
measuring the frequency of A (J M/ l 02W or er5) vs. G (er5//38CVI) in DNA

samples.

18

Goals of current study

Superinfection of multiple virulent MDV strains seems necessary to allow the
establishment of evolved MDV strains within a flock. Understanding what events are
necessary for superinfection to occur will increase our understanding of the conditions
necessary to allow viruses to interact within a host and evolve to greater virulence or
prevalence in a population.

This current study had several primary goals. 1) To demonstrate whether two
virulent Sl MDV strains could successfully co-infect and replicate within individual
birds. This goal was evaluated in the chapter 2 pilot study and validated in chapter 3.
2) To determine whether one virus could dominate another in terms of replication,
transmission, and tumor formation. Understanding the potential phenomenon of
dominance under the conditions of simultaneous infection is essential to evaluate data
from non-simultaneous superinfection studies. 3) To demonstrate whether
superinfection was still possible after delaying the time between challenges to 24 hours
and 13 days. Detection of dominance was evaluated with respect to increasing time
interval. 4) To compare effects in maternal antibody positive birds vaccinated with

HVT to determine the relevance under field conditions.

Hypotheses
Based on these goals, the following hypotheses were made:

I Following simultaneous challenge with two fully virulent 81 strains, both strains
will replicate, transmit to other birds and lead to tumor formation.
I Following simultaneous challenge, both strains will have equal chance of

transmission and tumor formation.

19

I Following short and long interval challenge, the first virus will be present in
greater frequency in bursa, feather follicles, and tumors relative to the second
inoculated virus.

I Vaccination will mute the effects of both viruses leading to equal mixtures in

bursa, feather follicles and tumors following short and long interval challenges.

20

Chapter 2
PILOT STUDIES (I-III)

Pilot 1. Methods to qualitatively and quantitatively distinguish selected serotype 1
MDV in dual-infected chickens
Abstract

Marek’s disease virus (MDV) can spread between vaccinated birds within a
flock, leading to potential superinfection by the same or multiple strains of MDV. This
study was designed to evaluate methods to distinguish two nearly identical recombinant
strains of serotype 1 MDV (er5 and er5//38CVI) in the same tissue sample, with
focus on the sensitivity and specificity of each assay. Methods used to differentiate the
viruses included immunohistochemistry (IHC) using monoclonal antibodies specific for
each strain, as well as pyrosequencing. 112 chickens were infected with one or both
viruses. IHC and pyrosequencing were 100% specific for samples from single-infected
birds. Using IHC, both virus strains were able to be detected in birds dually infected,
however, the relative intensity of staining for each virus was variable between birds.
Pyrosequencing was also able to detect both viruses in dually infected birds and results
correlated well with IHC. IHC and pyrosequencing were successful for detecting both
viruses in dually infected birds and will be used in further studies to understand the

effects of serotype 1 MDV superinfection.

Introduction

Marek’s disease (MD) is a lymphoproliferative disease of poultry caused by MD

virus (MDV), an alphaherpesvirus. The pathogenesis of infection with a single virulent

21

strain of MDV has been extensively studied; however, the pathogenesis of infection
with multiple virulent MDV strains has not been studied in depth. Infection with an
initial serotype 1 MDV followed by a second infection with a serotype 1 MDV
(superinfection) is assumed to occur in commercial poultry raising conditions.
Chickens are presumably being repeatedly exposed with MDV from their environment.
It is unclear what effect each virus has on the other during coinfection. Studying the
relationship between dual infection with multiple viruses in viva would further increase
our understanding of MDV pathogenesis. Current technologies such as
immunohistochemistry (IHC) and pyrosequencing would be key elements in
characterizing the effects of one virus on the other.

This project incorporated two different virus pairs to study the effect of two
similar viruses on each other. The first virus pair included er5 and a recombinant
Md5 virus containing pp38 from MDV Rispens strain (er5//38CVI) developed by Lee
(60). Her study found that there were no significant differences between the two strains
with in viva replication, antibody response, or tumor inductions. Specific monoclonal
antibodies exist for detection of each strain with IHC.

The goal of this pilot study was to evaluate specific methods to distinguish the
selected viruses using the tests described in the following section. If successful, these
methods would be used for an extensive set of studies examining the effect of

superinfection at multiple time intervals, and comparing the effect of vaccination.

Materials and Methods

Chickens

22

This experiment used 1515x71 white leghom chickens, a F1 hybrid cross of MD

susceptible 1515 males and 71 females (2). This particular cross has been a common

choice of birds for Marek’s disease experiments because both parental lines are >99%
inbred and are highly susceptible to MD, while the F1 birds have strong hybrid vigor.
Chicks were maternal antibody negative, as birds were hatched from an in-house SPF
breeder flock that had not been exposed to MDV vaccination. Chicks were challenged

at day of age before placement in Horsfall-Bauer negative pressure isolators.

Viruses

Two recombinant MD viruses were used for this experiment: er5 and
er5//38CVI. The origins of the two recombinant viruses come from the wild-type
Md5 virus, isolated by Witter in 1977 (103). Md5 is a vaDV, and is considered a
principle prototype strain used commonly for MD experiments, including as a control in
the ADOL pathotyping assay (93). Reddy et al. used Md5 to construct a library of
overlapping cosmid clones which generated a recombinant Md5 (er5) when
transfected into MDV-susceptible cells (75). The authors concluded that in vitra
grth properties were the same between Md5 and er5, and development of visceral
tumors were similar between groups.

The recombinant virus er5//38CVI was produced by Lee et al. using the erS
cosmid clones to substitute the pp38 gene from CVI988/Rispens in order to determine
whether the pp38 gene from CVI988/Rispens was related to the protective properties of
the vaccine (60). The authors, however, found the biological properties of

er5//38CVI were similar to the wild-type er5 virus with regards to in viva

23

replication, antibody response and tumor induction. Gimeno et al. reported no
differences in the frequency of lymphoproliferative lesions between er5 and
er5//38CVI at either 6 or 15 wpi (43). Based on these two studies, the virus pair
er5 and er5//3 8CVI were chosen for the experiments in the following chapters to
simulate re-infection with an identical virus on the assumption that there was no

significant differences in the pathology of the two viruses.

Immunohistochemistry testing

Immunohistochemistry (IHC) is the process of localizing specific proteins in a
tissue section by using antibodies specific to antigens in the tissue. There are several
different methods of visualizing the antigen-antibody interaction, most commonly by
conjugating the antibody or a secondary antibody to an enzyme such as peroxidase, or
by tagging with a fluorophore such as fluorescein. Both methods were used in these
experiments and will be described below.

Peroxidase staining of thymus, bursa, spleen, and F FE tissue sections was
performed using the Vectastain ABC kit.1 The kit was designed on the principles of the
high affinity of the avidin glycoprotein for the biotin vitamin (46, 47). The kit includes
biotinylated affinity-purified horse anti-mouse IgG (H + L) antibody, blocking serum
(normal horse serum), avidin DH, and biotinylated horseradish peroxidase H.
Colorizing was achieved using the Vectastain DAB substrate kit for peroxidase.2 This
kit includes 3, 3'- diaminobenzidine (DAB), buffer stock solution, and hydrogen

peroxide. Frozen samples were dried under vacuum and fixed in acetone before

 

l . .
Vector Laboratories, Inc., Burlrngame, CA

2 . .
Vector Laboratories, Inc., Burlrngame, CA

24

staining. Slides were later rehydrated and blocking serum was added to the fixed tissue
from the serum of the same species the secondary biotinylated antibody was made from
(horse) to reduce background or unspecific staining. The primary antibody was then
added, which was H19 (IgG monoclonal antibody specific for all 81 MDV pp38 genes
except CVI988 Rispens) or T65 (IgG monoclonal antibody specific for CVI988 Rispens
pp38 gene). After washing, the biotinylated secondary antibody was added, incubated,
and washed. Next, the avidin and biotinylated horseradish peroxidase were added
which form high affinity complexes between each other and the biotin conjugated to the
secondary antibody. After incubating and rinsing, the DAB solution was added
including the buffer and hydrogen peroxide which reacts with the horseradish

peroxidase and produces a dark brown color.

Monoclonal antibodies: H19, T65

The MDV phosphoprotein pp38 gene, located in the IRL and UL region, was
utilized for virus differentiation in the following chapters of experiments. The function
of the pp38 gene is not completely understood, but several studies have contributed to
our current knowledge. Xie et a1. concluded that pp3 8, in addition to ICP4 and meq
were involved in maintenance of MDV-transformed MSBl lyrnphoblastoid cells (107).
Cui and Qin described immunodepressive effects of pp38 (32). Several experiments
have indicated that pp38 may be an important part of MDV reactivation. Infection of
QT35 cells, a cell line latently infected with serotype 1 MDV, with HVT activates pp38
expression (108). The gene is also expressed in productively infected cells including the

_ FFE (33, 49, 67, 68, 76). Gimeno et al. used a mutant virus lacking pp38 (erSApp3 8)

25

 

to demonstrate that pp38 was necessary to establish cytolytic infection in B cells, but
not F F E (43). The authors found that pp38 was also necessary to produce an adequate
level of latently infected T cells and to maintain the transformed status in viva by
preventing apoptosis.

According to early published pp38 sequence data, all serotype 1 MDV strains
have identical bases in the pp38 gene except bases #320 and #326 (31, 38). CVI988,
however, is the only serotype 1 MDV strain that is not reactive with pp38-specific Mab
H19 (61, 105). Using DNA sequencing from the pp38 gene of additional strains, Cui et
al. determined the molecular basis for the difference in the antigenic specificity of H19
and T65 antibodies for the pp38 gene (34). Mab H19 was specific to base pair #320,
and was only positive when adenine (A) was present. All serotype 1 MDV strains
except CVI988 were positive for H19 and had A at bp #320. Mab T65 was specific to
base pair #326, and was only positive when guanine (G) was present. T65 was positive
for CVI988 which had G present at base pair #326. Most serotype 1 MDV strains were
positive for H19 only, having A at both base pairs #320 and #326, however, a few field
strains have A at base pair #320 and G at base pair #326, such as GA22, and thus are
positive for both H19 and T65.

The experiments described in the following chapters utilized this difference in
the pp38 gene for detection of individual viruses from the virus pairs described above.
J M/ 102W and er5 have A at both base pair #320 and #326 and thus were positive for
H19 only. er5//38CVI has G at both base pairs #320 and #326 and thus was positive

for T65 only. Pyrosequencing was designed to detect the differences at base pair #320,

26

thus measuring the frequency of A (J M/ 102W or er5) vs. G (er5//38CVI) in DNA

samples.

Pyrosequencing assay

Pyrosequencing is a DNA sequencing technique that relies on the following
steps to detect pyrophosphate released from nucleotide incorporation: 1) DNA
fragments containing the sequence of interest are amplified using standard PCR. One of
the primers used for amplification is biotin-labeled for immobilization to sepharose
beads. The sequencing primer is designed within five bases from the single nucleotide
polymorphism (SNP) of interest (Figure 2—1). 2) The amplified PCR product is mixed
with a sequencing primer and incubated with DNA polymerase, ATP sulfurylase,
luciferase, apyrase, adenosine 5’ phophosulfate (APS) and luciferin. 3)
Deoxyribonucleotide triphosphates (dNTPs) are added one at a time and if the dNTP is
complementary to the template strand, DNA polymerase catalyzes the incorporation
which releases pyrophosphate (PPi) in an amount equimolar to the amount of
incorporated dNTP. 4) The PPi in combination with APS is converted to ATP by ATP
sulfurylase. The ATP activates luciferase which mediates the conversion of luciferin to
oxyluciferin which generates visible light proportional to the amount of ATP. The light
is detected by a charge coupled device chip which produces a peak on the pyrogram
proportional to the number of dNTPs incorporated. Apyrase degrades unincorporated

nucleotides and ATP before the next dNTP is added.

27

Pyrosequencing was performed on a PSQ 96MA system.3 Three primers
specific for the pp38 gene were used for each reaction, consisting of the forward,

biotinylated reverse, and sequencing primers (Figure 2-1). The sequencing primer was

designed using SNP Primer Design fi'om Pyrosequencing AB version 1.0.1.4 Samples
results are reported as failed when no sequence is detected. The samples may either not
have adequate virus present (e. g. negative control samples) or there may have been

either insufficient or excessive total DNA present.

Pyrosequencing validation

Pyrosequencing was validated for sensitivity and accuracy using plasmid
constructs. The validations included 1) testing the accuracy of known mixtures of
plasmids, 2) determining the lowest level of virus copy number for detection on gel
electrophoresis 3) determining the accuracy of DNA amplified by nested PCR from
known mixtures of low quantity, and 4) comparing the accuracy of pure plasmid
mixtures compared to plasmid mixtures diluted in chicken DNA.

To test the accuracy of known mixtures of plasmids, two plasmids were
generated containing the pp38 genes of er5 and er5//38CVI using the A6 cosmid
from er5 and DNA isolated from er5//38CVI, respectively. Cloning was performed

using the TOPO Cloning Reaction Kit and Transforming One Shot Competent E. coli,

according to manufacturers instructions.5 PCR and gel electrophoresis were performed

fiom three bacterial colonies of each clone to confirm positive reaction, using MdCV-F

 

3 Qiagen, Hilden, Germany

4 . .
Applied Brosystems Inc., F oster City, CA
Invitrogen Corporation, Carlsbad, CA

28

and MdCV-R2 primers (Figure 2-2). Plasmid DNA was purified with the QIAprep
MiniPrep using the manufacturer’s protocol for a microcentrifirge column tube.6 DNA

was then sequenced on ABI 3100 Genetic Analyzer to confirm correct SNPs.7 Two
plasmids were selected containing the pp38 genes from er5 and er5//38CVI,
respectively. The first two sets of independent reconstructions (trial 1 & 2) were
performed using only the plasmids, each diluted to 10ng/ul (Table 2-1). From each
dilution, 2ul was used in a 50ul PCR reaction volume, and 40ul was used for
pyrosequencing. Pyrosequencing was run in duplicate for both reconstruction mixtures
and indicated that in some cases greater than 10% of the lower frequency virus was
needed for detection. er5 was over-represented in both mixtures (Figure 2-3).

To determine the lowest level of detection for passing results, a similar
reconstruction mixture was created using a different set of plasmids provided by Robert
Silva which included perS pp38-l and L233-6 pp38 Rispens, specific for pp38 genes
of er5 and er5//38CVI, respectively. Stocks were serially diluted in chicken DNA
(isolated from DF-l cells; Sng/ul) from 1x106 copies/u] to l copy/ul for both plasmids,
to mimic the conditions of the experiment when DNA samples will contain both virus
and chicken DNA. Pyrosequencing will fail without the quantity of PCR product
necessary to produce a band on gel electrophoresis. A gel was run following PCR and
the lightest band present for both plasmids indicated 1000 copies/u] necessary for band
detection (Figure 2-4). Pyrosequencing analysis confirmed passing results for 1000

copies/u] and failure to detect sequence at higher dilutions (data not shown).

 

6 Qiagen, Hilden, Germany
7
Applied Biosystems Inc., Foster City, CA

29

The same set of mixtures was used to determine the accuracy of DNA amplified
by nested PCR from known mixtures of low quantity. The next two lower dilutions
(100 copies/ul & 10 copies/ul) were mixed together in known frequencies similar to
Table 2-1. A nested PCR was run for the mixture using pp3 8-R and MdCV-R outside
primers and MdCV-F and MdCV-BiotinR inside primers. The same mixtures were also
run on a single PCR using the inside primers, and as expected all samples failed
pyrosequencing since bands were not detected on the gel. However, double
amplification using the nested PCR assays led to passing results. Results were less
accurate then after a single PCR assay, but both viruses were still able to be detected in
all samples (Figure 2-5).

Finally, an additional reconstruction was made using plasmid DNA and tested
by pyrosequencing following nested PCR both with and without background chicken
DNA (5ng/ul). The plasmids were mixed together in the same mixtures as described
above, and lOul of each was used by itself in the nested PCR or mixed with 10ul of
chicken DNA. The same PCR primers were used as described above for the nested
PCR. The presence of chicken DNA did not alter the results, although results again

were less accurate after nested PCR than following single PCR (Figure 2-6).

Experimental design

Twenty chickens were singly-infected with er5 or er5//38CVI, or dually-
infected with both viruses at hatch (Table 2-2). Tissue samples (spleen, bursa, thymus)
were collected from 5 birds (3 control birds) at both 4 and 6 dpi for IHC and DNA

isolation (Table 2-3). Feather follicle epithelium samples were similarly collected from

30

5 birds (3 control birds) at 21 days post infection for IHC only. Serial sections of
thyms, bursa, spleen, and F F E were cut and stained with H19 (specific for erS) and
T65 (specific for er5//38CVI) monoclonal antibodies. IHC results were scored
subjectively as negative, or 1-4+ based on the amount of staining within follicles and
total number of follicles. Gross necropsy and histopathology were performed on the

remaining 5 birds (3 control birds) at 8 weeks post infection to compare lesions between

groups.

Results

The pathogenicity of single and dually-infected chickens was compared between
the three groups at termination (T able 2-4). Every bird from each of the three groups
was diagnosed with Marek’s disease based on lymphoid atrophy with nerve
enlargement and visceral tumors. Heart and spleen lymphomas were prevalent, and all
infected birds had one or both of this type of tumor. All three groups plus the control
group were terminated early at 41 dpi due to the death of all five birds from the
er5//38CVI group by 41 days. The er5 group had 2 deaths by 41 days, and the dual
infected group had one death.

Immunohistochemistry staining of sections from 4 dpi lacked sufficient quality
to interpret the results, but did not appear to have any staining. Staining at 6 and 21 dpi
was specific, as predicted, for birds singly infected (Figures 2-7, 2-8; Tables 2-5, 2-6).
Both virus strains were able to be detected in birds dually infected, however, the relative
intensity of staining for each virus was variable between birds (Figures 2-9, 2-10).

There were birds within each group that did not stain for either antibody and was likely

31

 

due to inoculation failure or slow virus replication. N/A indicates that no sample was
available for that particular bird.

Pyrosequencing results were similarly 100% specific in singly-infected birds.
All birds that had no staining by IHC subsequently failed with pyrosequencing, since no
virus sequence was detected. In dual infected birds, pyrosequencing results correlated
with IHC results, except for one bird that had weak IHC staining and failed by
pyrosequencing. Both viruses were present by pyrosequencing from the bird that had

dual IHC staining.

Summary and Conclusion

Pyrosequencing validation of sensitivity demonstrated the detection of virus at
quantities as low as 1000 copies/ul. The next two higher log dilutions, however, were
also detected by using nested PCR before pyrosequencing. Results from plasmid
mixtures illustrated a linear relationship of pyrosequencing results with slight over-
representation of er5. Results from chapters 2 and 3 categorize the pyrosequencing
data according to which virus was predominant, thus minimizing the effect of over-
representation. Since not all reconstruction replicates successfirlly identified mixtures
with frequencies of 10% suggests results indicating the presence of only one virus may
not reveal small proportions of the second virus.

Immunohistochemistry and pyrosequencing were both 100% specific for er5
and er5//38CVI based on results from singly infected birds. Both strains were
detected in dually infected birds, although relative intensity was variable between birds.

Interestingly, both virus strains were detected by IHC in dually infected birds in only

32

one of four positive staining birds at both 6 and 21 dpi. Results from one bird at 6 dpi
(M6392) were positive for IHC and not by pyrosequencing, which indicated that IHC
staining may be more sensitive than pyrosequencing. The methods that were developed
and tested in this pilot experiment were considered sufficient for differentiating between

er5 and er5//38CVI and were used for all subsequent experiments.

33

Pilot II. Pathotyping of erS and erS/l38CVI

Abstract

Pathotyping was performed to confirm that both Md5 recombinant viruses used
in the following studies were of similar virulence. Two trials of two replicates each
were performed because the prototype v and W MDV strains had uncharacteristic
results in trial 1. The prototype control strains were adequately protected against in trial
2, and results showed statistically similar virulence between erS and er5//38CVI,
although both were significantly less protected using bivalent vaccination compared to
prototype vv strain Md5. Pathotype designations were not made due to prototype virus
inconsistencies between replicates and trials. Results of virulence rank were nearly
identical between erS, er5//38CVI and historical results of Md5. Despite
inconsistencies, er5 had the highest percentage of MD lesions from all MDV strains

in both replicates fiom trials 1 and 2.

Introduction

The proposed model to study. the effects of superinfection and simultaneous
dual infection with the “same” virus will use er5 and er5//3 8CVI. It is therefore
important to confirm that these two viruses are as similar in pathogenicity as possible.
Two previously reported studies found that there were no significant differences
between the two strains with in viva replication, antibody response, or tumor inductions

(43, 60).

34

Pathotyping is one additional method for evaluating the similarity between two
virus strains. The philosophy and a description of pathotyping methods was recently
reviewed by Witter et al. (98). Pathotyping became an important classification of
viruses not long after increased virulent forms of MDV were initially reported (5). The
classification has been used to designate differences in virulence based on frequency of
disease induced in vaccinated and unvaccinated birds. Early classifications by Biggs et
al. differentiated strains as classical or acute, with the term acute based on higher
morbidity and mortality, shorter time until onset, and a higher incidence of visceral
tumors (11). Bi ggs et al. later included additional criteria for this designation based on
histological lesion scores (9). Witter first designated the virulence of MDV strains
based on responses in vaccinated chickens, using a cut-off value of HVT protection
(77%) (93). Several years later he included prototype isolates as a comparison against
field isolates to better standardize the pathotype designations from other laboratories
(97). Specifically, vaDV (very virulent) isolates were those with MD lesions in
HVT-vaccinated, susceptible birds at a rate greater than a prototype vMDV(virulent)
isolate, such as J M or GA. Other groups began using these techniques, but the
procedures were variable with different prototype and chicken lines used for making the
designation of vaDV (14, 50, 51, 89). Witter recognized an increased virulence of
MDV strains in the early 1990’s in bivalent-vaccinated chickens, leading to four current
pathotype designations: m (mild), v (virulent), vv (very virulent), and vv+ (very virulent
plus) (95).

er5//38CVI and er5 are assumed to be the same pathotype as Md5 (vv)

based on the previous studies, but neither have been officially pathotyped. The purpose

35

of this subphase was to pathotype erS and er5//38CVI in order to give further

evidence to the assumption that erS and er5//38CVI are pathogenically similar.

Materials and Methods

Chickens

This experiment used maternal antibody positive1515x71 white leghom

chickens, a F 1 hybrid cross of MD susceptible 1515 males and 71 females (2). Birds

were hatched from an in-house SPF breeder flock, separated into treatment groups and

placed in Horsfall-Bauer negative pressure isolators.

Viruses

The viruses pathotyped included er5 and er5//38CVI, which were described
in pilot 1. Prototype control viruses used were J M/ 102W and Md5 which were also
described in pilot 1. Briefly, the two recombinant Md5 viruses being tested share the
same er5 backbone and are only different in the pp38 gene. The viruses were
previously shown to have no differences in the frequency of lymphoproliferative lesions

at either 6 or 15 wpi (43).

Experimental design
The ADOL pathotyping assay was used for this study. Two prototype MDV

strains, (J M/ 102W and Md5) and the unknown strains (er5 and er5//38CVI) were

inoculated in three lots (17 birds each) of 1515x71 antibody positive chickens:

36

unvaccinated, HVT-vaccinated, and bivalent-vaccinated (HVT + SB-l) (Table 2-7).
Vaccination (2000 pfu) was administered at hatch and challenge (500 pfu) at 5 days
post vaccination. Termination and necropsy were performed at 56 days post challenge
and a diagnosis of MD was made based on gross lesions consisting of nerve
enlargement and/or visceral tumors. Suspicious lesions were confirmed by
histopathology. One isolator (6 birds) of unvaccinated/unchallenged control birds was
included in each replication. A second full replication was performed following the first
challenge. Two complete trials were completed (four total replicates).

Chi-square analysis was used to compare the similarity of responses between

replicates as well as to measure the significance of response in vaccinated birds

compared to the prototype control viruses.8 Virulence rank was calculated for each

virus, defined as the mean of the % protection by HVT and bivalent vaccines subtracted

from 100.

Results

Trial 1

The percentage of MD positive birds was consistent between all replicates except for
two lots, where higher mortality was seen in the second replicate (Table 2-8). The
percentage of MD positive birds challenged with the prototype control viruses varied
somewhat from previously reported studies under the same conditions. Due to
variability in prototype virus results, viruses were not assigned a pathotype designation

and a second trial was performed.

 

8 SAS V9.1, Cary, NC

37

Trial 2

The percentage of MD positive birds was consistent between all replicates
(Table 2-8). Bivalent protection against Md5 was unusually high, so er5 and
er5//38CVI were both significantly less protected by bivalent vaccination (Table 2-
9). When compared with each other, there was no significant difference in protection
between erS and er5//38CVI in each group of vaccinated birds. Virulence rank was
calculated using the summary of results from trials 1 and 2. Results from er5 (57.5)
and er5//38CVI (58.5) were well in line with historical results reported from Md5
(58) by Witter (95). At the same time, these results also confirm the significant
difference in virulence between JM/102W and er5//38CVI, an additional virus pair

used in chapters 3 and 4.

Summary and Conclusion

Trial 2 suggested there was no significant difference in pathogenicity between
erS and er5//38CVI in vaccinated birds, while also confirming significant
difference in virulence between the virus pair JM/102W and er5//3 8CVI used in
chapters 3 and 4. Although statistically similar to each other, er5 and er5//38CVI
both were significantly more virulent than Md5 in trial 2, suggesting the recombinant
strains might not be pathogenically equivalent to the parent strain. Both viruses were
nearly identical to historical results from Md5 based on virulence rank. It is not known
what caused the uncharacteristic reduced protection of the prototype virus strains in trial

1, but further trials should be conducted to confirm results from trial 2 before officially

38

designating pathotypes for erS and er5// 3 8CVI. Despite the inconsistencies, er5
had the highest percentage of MD lesions from all MDV strains in both replicates from

trials 1 and 2, even though there were no statistically significant differences in trial 2.

 

39

Pilot III. A comparison between MDV production in feather pulp versus feather
follicle epithelium (FFE) .

Abstract

Although Marek's disease viral load in chickens can be assessed by quantifying
the viral DNA in homogenized feather tips, this value may not correlate to the viral load
being shed. This study was designed to determine what portion of the feather tip tissue
is most reflective of virus load in the shed dander by comparing feather follicle
epithelium (F FE) cells, adhered to the exterior of the shaft, to interior cells within the
feather pulp. Using multiplex quantitative PCR, the FF E cells from infected birds had
significantly higher viral copy number (VCN) per host cell than from feather pulp, and
were more comparable to the VCN measured from feather dander. These results
suggest that DNA isolated from FFE is more reflective of actual shed virus than VCN

measured from the feather pulp.

Introduction

Several recent studies have been performed by other groups analyzing the MDV
load in feather tips (1, 3, 4, 35, 36, 52, 54). Despite the success quantifying DNA in
these studies, there has been no evidence that the quantity of DNA in the feather pulp is
related to the quantity of virus that is shed via the FFE.

Islam et al. have reported on MDV DNA extracted from isolator dust (54). Dust
samples were collected between days 9 and 58 after placement and DNA was extracted
fiom 5mg of dust and quantified as viral copy number per mg of dust. MDV was first

detected in dust at 7 dpi and quantity peaked around 35-40 dpi. The quantity of virus

40

shed in dust, however, has not been directly correlated with the quantity of virus within
the feather follicle epithelium.

The objectives from chapters 3 and 4 were to characterize the difference in
pattern and distribution of specific virus antigen for single infection vs. superinfection
in F FE, and to compare the quantity and frequency of shed viruses based on DNA
isolated from FFE. This experiment was designed to help understand the relationship of
DNA quantity isolated from feather pulp and FF E, and whether DNA isolated from FFE
correlates to the DNA isolated from dust. The results were used to determine which
samples would be collected in the subsequent set of experiments to accurately represent

virus being shed to other birds.

Materials and Methods

Chickens

This experiment used maternal antibody negative1515x71 white leghom

chickens, a F1 hybrid cross of MD susceptible 1515 males and 71 females (2). Maternal

antibody positive chickens were used for one group that was challenged with a vv+
MDV strain to protect the birds from early mortality. Birds were hatched from an in-
house SPF breeder flock, separated into treatment groups, challenged and placed in

Horsfall-Bauer negative pressure isolators.

Viruses
Four viruses representing three pathotypes (V, W, vv+) were compared in this

experiment, including a partially attenuated vv+ strain. The lower virulent strains

41

JM/102W (v) and erS (W) have already been described (see chapter 1). The two
higher virulent strains used in this study were derived from 648A, a field strain isolated
in 1995 that has been designated as a prototype vv+ strain (95). A highly virulent low
passage working stock (p8) was used, as well as a partially attenuated strain at passage
50 (96). Gimeno et al. reported that 648A p50 had 100% persistent neurological disease
but slightly reduced tumors and absence of transient paralysis, compared to lower
passage strains (44). Three contact birds exposed to chickens challenged with 648A
p50 had no viremia, suggesting a lack of transmission from the partially attenuated

strain.

Quantitative PCR assay

The standard polymerase chain reaction (PCR) has been adapted to allow
quantitative measurements of amplification following each cycle in a technique known
as quantitative PCR (qPCR), or real-time PCR. The standard PCR is a molecular
technique that amplifies a selected portion of DNA exponentially with each repeated
cycle that includes denaturation, annealing, and elongation steps. The denaturation
step (~95°C) melts the double-stranded DNA strand into single-stranded DNA. The
annealing step cools down the DNA (~65°C) which allows the primers to bind to the
single-stranded DNA. The DNA is then heated (~72°C) to activate Taq polymerase, a
DNA polymerase which synthesizes a new DNA strand complementary to the single-
stranded DNA template. A final elongation step is generally programmed which

elongates any remaining single-stranded DNA. Thirty PCR cycles are commonly used,

42

followed by agar gel electrophoresis to confirm that the correct sized PCR fragment has
been amplified.

qPCR follows the same principles as standard PCR except the target amplified
DNA fragment is quantified following every cycle. This quantification can be absolute
or relative by normalizing to DNA input or another gene, such as a host cellular gene.
Detection of the DNA fragment for quantification is enabled by the use of an
oligonucleotide probe that fluoresces when bound to the complimentary DNA.
Experiments described in the following chapters used a TaqMan® probe (fluorogenic 5'
nuclease chemistry) which has a reporter fluorescent dye on the 5’ end and a quencher
dye on the 3’ end. The close proximity of the quencher dye reduces fluorescence from
the reporter dye when intact. During the annealing step, the probe binds downstream
from one of the primer sites and is cleaved during elongation when the upstream primer
is extended. Cleavage of the quencher dye fiom the reporter dye increases the
fluorescence from the reporter dye, and removes the probe for primer extension to
continue to the end of the template. The advantages of using TaqMan chemistry is that
specific hybridization is required between the probe and DNA strand to allow
fluorescence which reduces background signal, and probes can be labeled with different
reporter dyes to allow more than one type of quantification in the same reaction.

The experiments described in this and the following chapters used a multiplex
reaction performed on an ABI 7500 Real-Time PCR Systemg. One probe was specific

to the MDV pp38 gene and the other specific to an endogenous reference,

glyceraldehyde-3-phosphate dehydrogenase (GAPDH). Both probes used TAMRATM

 

9 . . .
Applied Brosystems Inc., Foster City, CA

43

(tetramethylrhodamine) quencher dyes. The probe specific to MDV used a green
emitter (500-549nm) reporter dye, 6FAMTM (6-carboxyfluorescein, excitation 492,
emission 515nm) whereas the probe specific to GAPDH used a yellow emitter (550-
584nm) reporter dye, VICTM (excitation 53 8, emission 554nm). The following
conditions were used: 1) 50°C — 2 minutes; 2) 95°C — 10 minutes; 3) 95°C — 15
seconds; 4) 60°C — 1 minute with steps 3-4 repeated 40 times. The primers and probes
used for the pp38 and GAPDH sequences are listed below.
pp38 primers and probe:
Mde-F: 5’-GTGATGGGAAGGCGATAGAA-3’
Mde-TMR: 5’-TCGTCAAGATGTTCATTCCCTG-3’
MC-TMP: 5’FAM-TACCGCCTGAGCCCCGGAGG-3’TAMRA
GAPDH primers and probe:
GDH-TM2.5: 5’-ACAGAGGTGCTGCCCAGAA-3’
GDH-TM2.3: 5 ’ -ACTI‘TCCCCACAGCCTTAGCA-3 ’

GDH-TMP2: 5’VIC—TCATCCCAGCGTCCACT-3’TAMRA

To isolate DNA from the F F E and feather pulp, the following procedure was
used. Approximately 56 feathers were plucked from the feather tract and placed intact
in a microcentrifuge tube (Tube #1) containing 300uL of cell lysis buffer. Once all
samples were collected, the microcentrifuge tubes were shaken at a moderate speed for
approximately 2 hours on an orbital shaker at room temperature to allow lysis of F FE
cells adhered to the outside of the feather shaft. After shaking, feathers were removed

from the lysis buffer and the feather shaft was cut into small pieces to expose the feather

44

pulp and placed in a second microcentrifuge container (Tube #2) containing 300uL of
cell lysis buffer. Both microcentrifuge tubes were incubated overnight. Tube #1 was
used to represent DNA isolated from F FE cells and Tube #2 represented DNA isolated

from the feather pulp. Additional DNA purification was done using the PureGene DNA

Purification Kit, according to manufacturer instructions.10 Results were reported as
viral copy number per host cell (VCN/host cell) by dividing quantity results from virus
gene pp38 by quantity results from host gene GAPDH, factoring for two copies of each
host gene per cell.

DNA was isolated from feather dust using a commercially available handheld
shop vacuum. Individual collections were made using a standard commercially
available paper coffee filter placed over the end of the collection hose. Birds were
vacuumed together as groups (dust pooled) or individually for isolators 6-17. The
negative pressure from the vaccum created a small cup in the middle of the coffee filter
containing the feather dust. This cup containing the feather dust was inverted into the
top of a 50mL centrifuge tube and the lid closed. The outer portions of the coffee filter
outside of the lid were torn off. At the end of the collections, the tubes were centrifuged
at approximately 500rpm to collect the dust at the bottom of the tubes. After spinning,
the paper filter was removed and 300uL of cell lysis buffer was added to the tubes.
Tubes were mixed and the contents were transferred to a microcentrifuge tube where
overnight incubation and DNA purification proceeded in the same way as the FFE and
feather pulp samples. Results were also reported as VCN/host cell, assuming that a

significant number of intact keratinized epithelial cells were present in samples.

 

10 . .
Gentra Systems, Inc., Minneapolis, MN

45

Experimental design

Birds were inoculated with one of four different challenge viruses (or control)
and sampled at 11, 21, and 31 days post-challenge (Table 2-10). Birds with the same
challenge virus were housed in isolators with 10 birds each (lots 1-5) and in addition 10
birds were infected with er5 and housed individually in their own isolator unit (lots 6-
17). For individually housed birds, an extra bird control bird was kept in each isolator
for socialization until the first sampling was initiated. Birds were housed individually
to prevent contamination of dander samples with infected dander from other birds.
Sampling consisted of pulling feathers from the cervical and subhumeral feather tracts
and collecting dust at each time point, as well as skin sections saved for
immunohistochemistry. Dust samples were pooled from birds grouped in the same
isolator, whereas feather samples were kept separate from individual birds. Multiplex
quantitative PCR was performed as described above.

A paired t-test was calculated to determine the statistical significance in the
difference of the VCN/host cells between each group (F FE vs feather pulp). The
relationship between FF E, feather pulp and dust in individual birds was measured using
Pearson correlation, along with a t-test to determine the difference in viral quantity

between averages for FFE, feather pulp and dust.

Results

Based on the average virus copy number (VCN) per host cell for all samples

separated by sampling time, the F F E cells had a significantly higher virus load

46

compared to cells isolated from the feather pulp (Figure 2-11). These differences were
statistically significant based on a paired t-test for all viruses tested at all three time
points (Table 2—11). Results from day 31 post-challenge were inconsistent with the first
two samplings with respect to VCN from F FE and feather pulp cells (Figure 2-11).
The protocol for processing of the feathers was deviated for the day 31 samples, and
feathers remained in cell lysis buffer overnight, instead of 2 hours, before removing and
cutting the shaft to isolate DNA from feather pulp. Results suggest that this prolonged
interval may have caused disintegration of the feather shaft and dilution of FFE DNA
with feather pulp DNA. An attempt to validate the day 31 samples was made using
laser microdissection to isolate and quantitate DNA from individual cells from the FFE
and feather pulp (methods not described), but results were inconsistent (Figure 2-12).
Dust samples were analyzed in the same way as feather tissue samples. In
general, VCN/host cell was low at 11 days post-challenge, peaking by 21 days and
decreasing again by 31 days based on averages from the isolators with multiple birds
infected with the same virus. Birds infected with the vv+ MDV strains, however,
increased linearly from 11 to 31 days without peaking at 21 days post-challenge
(Figure 2-13). The VCN/host cell was considerably higher for the FFE and dust,
compared to the feather pulp (Figure 2-14).

Analysis from birds infected with the same virus but housed individually
showed variable results at each of the three samplings, but a similar average trend in
dust samples as birds housed together (Figure 2-15). Results from birds housed alone
also showed a much higher quantity of VCN/host cell in the FFE and dust compared to

feather pulp, although all three were significantly different at 21 days post challenge

47

(FF E*Dust p=0.0037; F FE*Pulp p<0.0001; Dust*Pulp p=0.0060) (Figure 2-16).
Interestingly, VCN/host cell was higher in the F FE cells in the individual birds than
from dust samples which was the opposite relationship seen in pooled samples from
groups of birds housed together (Figure 2-14).

A Pearson correlation analysis in individually—housed birds indicated a
significant positive correlation between FFE and dust (0.39; p=0.03 87), including all
three time points. A partial correlation with the-effect of sampling day removed had
even a high correlation between F FE and dust (0.64, p=0.0007). A significant
correlation was not observed with feather pulp, likely due to low values. If the day 31
samples are removed, the Pearson correlation was quite strong among all three sample
types (FF E*Dust 0.72, p=0.0005; FFE*Pulp 0.54, p=0.0262; Dust*Pulp 0.72;

p=0.0018).

Summary and Conclusion

Based on the current data, the VCN/host cell appears to be quite high from DNA
isolated from F FE and dust, compared to the lower quantity present from cells within
the feather pulp. These results suggest that based on techniques used in this study, FFE
VCN is more reflective of actual shed virus than VCN measured from the feather pulp,
and may be the preferable method for measuring true quantity of shed virus from
individual chickens rather than by using homogenized feather tips. As a result of these
findings, sampling from subsequent experiments was done as described in this

experiment, isolating DNA from cells adhered to the outside of the feather shaft, to

48

reflect the frequency and quantity of shed virus from individual birds following dual

infection.

49

 

 

 

 

 

 

eémasém com 382%? + m2): om 4 an

esteem 8m 38 $8 $2.? 8 m “3

25% com 38 was: om N as

Ease 8m 2880 Beads: 2 2 “3
925 5.2: owns—=30 25:35 a. we‘—

 

 

 

 

oefiso oweozmau "ué eEah

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

o o 2 2:

a S a 8

N on w 8

m 8 e E

a. a. e 8

m cm m 8

a 8 a. 9.

e E m 2

w ow N 8

a so _ S

2 2: o o

:3 aeao> .x. :3 0828/ .x.
an: 303553 N was; 3.25 2.33 _ Bea:

 

 

 

 

noumwzg wfiogsgmohxm 8m wow: man—ta: soropnmegom ”TN 035—.

 

50

 

ESQ

 

H>Ow QEEZH

 

2V2“

 

_O.E_HOU

>38er 305

amp cm

 

ES

 

H>Ow QREZH

 

m3)?

 

75:00

mnE

whee “m

 

Eda

 

H>me\\m2>i

 

93$?

 

85:00

coo—mm £33m £5521

mane o

 

E59

 

~>me\\w3>:

 

m3)?

 

MWWWMWWWMWWWMWWW

65:00

macaw .35m .385:

9mg 4

 

 

achm a

 

ammo—=30

 

052:.

 

San— wfi-mfiam

 

 

cacao oanam "TN 933,—.

51

8553 +4 8 825 i 80¢ 28m a so woman—o 225 EoEowEEo etc: can Emobm 8:5. 98 SEE: .Eamoa 555 N
age 52% .05 :40: .646 seams are: £15485

ammoommeoz Amie 52%me .Qomv Eofiowsfio can: ocfiom Afimv EoEowSEo ate: 3:85 .356 “scammed—do eta:
mamm> A<mv 43988 35m A<C 439:8 38439 AOmQV 58% mo owco=mno “mom when ADDS 85 Mo 3:9 umcoufi>o£n< _

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

92 am am i : +N +4 +4 4 M causes Ed 332
Q: am a: +N +N +4 +4 _4 M across 35 NNSE
92 am .5 +N +m +4 4 M 8:8? age :32
a: am +m +N +N +4 MN 0 canoes and ESE
92 am .5 +_ +4 +4 _4 M causes ago $32
92 am +N +N +4 +4 em a Sammie/e 3%:
oz am a: +_ +4 +4 mm a 50354,: $32
a: a: +N +N +4 +4 2 a 5822:): £52
92 am +N +4 +4 NN a 582%? £32
a: am a: +N +4 +4 _4 a 50354,: 4&4:
o2 am t +4 +4 2 a was: NwSE
a: 3m +N +N +N +4 +4 NN a was: $32
a: E +N +N +m +N +N _4 M was: 832
92 a: +N +4 +4 4 M 34,: 832

x .4 M :58 8%:

M _4 M assoc 8%:

x 4 M 6580 8%:
men £085 am am we, <m N5 95 _o\M awas—=26 4.5m

 

 

 

 

 

 

 

 

 

 

coumEEB “a gamma" $80 "TN o-Aah

52

529m n Em .355 u Sm .msfihfir u that Aoocoscom 03 295% 83¢ H mm ”323305344

N

Awfififim ciaoiov +4 9 35“:an cocotmomv +_ Eofi 28m a no 83896 as $853 M

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

tom mom m H mo i i i +N +N +N 538%: :35 332
m m m - - - - - - couoomi :35 4082
O52 92: 052 - - - +m +~ +~ 550%: :50 882
m m m i - i - - - 838%: .25 $82
053 053 one? - - - +m +N +m eouoamfi :35 332
oofio 09 Ho oofio +~ +N +m - - - Sommtmgfi N302
m m m - - - - - - H>me>m3>t ESE
oofio oofio oofio +m +m +N - - - Smehmeu omnoE
oofio oofio 2: Ho +N +N +m - - - Sommhmgfi $32
oofio oofio 2: Ho +m +m +N - - - Sommhmwzc wmnoE
m m m - - - - - m3): Que—2
once O52 O63 - - - +N +N +N 03>? SSE
once one? 92: - - - +N + + 33: 0.902
one? 903 052 - - - +N +N +N mes: $822
once one? 053 - - - i - +_ 32c 832
m m m - - - - - - 3550 04502
n m m - - - - - - 35:00 3.32
m m m - - - - - - 65:00 wmncE
am :5 5 :5 Sn ea. Em Sm NEE. 85:45 44am
Sowgmgfi .X. n m3): .x. 3% SE
mfionosromebfi MzuquoAuoumEeN—nfifim
.Eu 0 380:8 838$ 82w 33mph memofisgmoba can Dim umé 25¢.

53

052a: 52 u 4.2 sees? 22:: see: H mm: accuses“? N

3:883 o>mm:2xov

+4 8 35:83 wobumomv i Sch 038 a :0 43835 8: 4:284 _

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

i +N :28? as: $32
+N - cases as: 44%:
- +m 8:83 4:: N48:
- i 85:: :3: N45:
- - causes as: 4%:
+N - 58:54: 8%:
+N - 58:54: SSE
i - SENSE): was:
<2 <2 58:54: 522
+N - 58 $34: 332
- +N was: 3%:
- +N was: 452
- +N was: $82
- +N 84: N52
- +N was: :32
- - assoc SSE
- - sauce SSE
- - 65:8 832
Se 2: 85:29 44::
NE

 

 

 

 

.5: a 380:8 mafia—am mm...“— Eom 33mph Emgnoowmflogg ueé ozah

54

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2,9332%]: Tmm + H>E .2 2
~>me\\m3\€ H>m C N—
H>me§m2>t 0:02 S _ _
m3): me + :53 C S
mug: :5: S o
m3)? 0:02 S w
32 Tmm + H>E : N.
mug E>E : 0
$2 0:02 E m
3832—. Tmm + H>E : v
38 5‘»: P>E S m
38 5):. 0:02 b H N
205:8 0080.383 0:0 Z c _
50.50 ”vi-EU 55.25005» 050—050 u «0‘—

 

 

.Egofia

003 800330: 03% :80: “mo: 9?: m :0 mg mo 3: com HEB ngzfio 0:03 300620 :80: :0. Sim + Ev 0:60? 05:35
.8 H>$ :056 .00 a: ooom 53» 0000500? 8 0000:6003: 0:03 300620 $0000 wE§853 :8 0:52: 0w:0=0:0 K-" 03:.

55

a GE 0:00:00: (:0 :05 80:”: 88: 0:03??? :3 30:02:: EESGEwmm GE E0809.n

 

 

 

 

 

 

 

G mm mm 8 2: 2: 58:02,: ,
3 mm 2 on 2: 2: 82:

M: M: E a 2: 2: $2

0 N: o M: 2: 2: 38:):

E 2 g a 2: 2: $0352:
*8 a 8 K 2: 2: 202:

an x mm mm 2: 2: 82

mm mm m :V 2: mm 38:92?!
N: E N: E N: 2 ‘fig 1
70m + g: S: 25:

 

 

 

 

55> wouwfiwoom> mCOvHOEO Gm Q: .X:

 

 

.N 23 _ was EB: 2.93. 3: 39:

8:00:90: 3 RES 3086 m. 0:02 $ £-N 030,—.

 

‘ {will

56

.2: 80:”: 00:00:53 00::000>
::0_0>E 0:0 by: :3 80008:: .x. 0:: :0 000:: 0:: 00 00:000 0: 00:0: 00:03:; N
0:030: £50003 ::0:0.t:0
3 00:00:08 0:0 095% :0::0::000> £505 m0w0::00:00 :0 0005:0000 0:02:35 _

 

 

 

 

 

 

1,000 4, < S 0 0: 303002: 0 :1
20 h m S 0 mm 002: _ ,
0.00 < E 0 mm 022 ,
20 2 < 0: 385): :2
0 mm m 00 5003002: ,
m :0 0 mm 00:: fl
< S 0 mm 002 W
a < a 302:): 0 N
, 0
m :0 m 0 5000002: ,
< mm m< wm m02: i
< 00 m 0: 002 7
00 < mm 38%: :
700 + 5: 5: 328: :00:
mg“ 2.315 Bumcmoomxr
uofi®~3fi> mCDv—Omfio S COCOOHOHQ 2X0

 

 

 

 

 

3300: 0025800 02: N 0% : $0.5 80¢ :0.:00:0:m o\o noun 030,—.

57

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

:00 :0 :01: 3080 -02): €32 : 0:
:00 :0 :0 .:: 30:50 -05: €32 : 2
:00 :0 :0 .:: 0220.: 2:0 00:): -03): €32 : 2
:00 :0 :0 ,:: 002.0: 2: 002: -22 €32 : E
:00 :0 :0 ,:: 0003:: 2:0 00:: A22 €32 : 0:
:00 :0 :0 .:: 02:20:: 2: 00:2: -052 €32 : 0:
:00 :0 :0 .2 00.20:: 2: 00:2: -,:<2:000:0: : ::
:00 :0 «:0 0:: 02:20:: 2: 002: -032 €32 : c:
:00 :0 :0 ,:: 0020:: 2:0 00:): -802 €32 : 0
:00 :0 :0 .: : 2:20: 2:0 002: -03): €32 : 0
:00 :0 :0 .:: 0:520: 20 00:2: -03): €32 : :
:00 :0 :0 .:: 00:20:: 2: 002: -22 €32 : 0
:00 :0 “:0 ,:: 30:50 -303: €32 0 0
:00 :0 :0 a: 2:20: 2:0 80:: <20 .03: €32 :: 0
:00 :0 :0 ,:: 0220: 2:0 :00: <20 +22 €32 c: 0
:00 :0 :0 .:: 0220:: 2:0 002: -03): €32 :: 0
:00 :0 :0 .:: 0:520: 2:0 30:: :2: -03): €32 0: :
Wins—am 0w:0=0:U 50:00 0w:0=0:U £0.50 0:00—0:00 0:00.050 0: :01—

 

 

2008 0002:0000 "2-0 0:000

58

:82 08:00 00:03:.

 

 

 

 

 

 

 

:8 :09: ZU>*
:0 :00 E: 00 8:00:00 ::< 000 00000
0:80 000 00 8:00:00 :2 00 40000
:80? 0:0 00 3:00:00 :00 0020
Soodv ohm; m N 0290000 =< Bmo : \S:
3086 mm: m m :m 03:03 =<
25qu N : m“: w m :N 000003 =<
Soodv 33 mm 2 000003 =<
3030?,— 0315 I mm": 3:20:20 :02): moi—=00 69:00— 0: wee—=30 «man @009 2:;

 

 

owco=0so “mom

00000 3 9:0 0:03 3 0:8 :mo£ZU> Ea 00500.0 035:: mm“: mo 8:20.006 :02: 0:: 08 00320000 002-: 3:000 00 £38m 3 TN 030,—.

59

Figure 2-1: Sequence of pp38 gene indicating pyrosequencing primer binding sites

ATGGAATTCGAAGCAGAACACGAAGGGCTGACGGCGTCTTGGGTCG
CCCCCGCTCCCCAGGGTGGAAAAGGGGCGGAGGGCCGCGCAGGGGTCGCC
GACGAGGCAGGGCATGGGAAAACAGAAGCGGAATGCGCCGAGGACGGCGA
GAAATGCGGGGACGCCGAGATGAGCGCTTTGGATCGGGTCCAGAGGGACC

GGTGGAGA’ITCAGTTTCTCCGCCCCCTCACTCfiIGGAGTCACGGGGAAGGGG
GCTATTCCAATAAAGdGTGATGGGAAGGCGATAGAAJ‘I'TGCCAGGAGCTAAC

CGGAGAGLGGAGAGTGGCTGTC/sfdX/GFGTGGGWGGAGCTACCGCCTGA
GCCCCGGAGGTCAGGGAATGAACATCTTGACGAAAGTCGGTATGCGAAACA
AACCGAAAGGGGTAGCTCTACGGGGAAAGAAGAGGIGAGATGGTATGAAGC]

-‘GGGGAGCTTGCCCAGCAGTGCIGAAGGAGGAACATATGCGG/q’CTT
GC'ITGTCGAAGCAGAGCAAGCTGTTGTACATTCCGTTCGCGCATTAATGCTG
GCCGAAAGACAAAACCCAAATATAT‘TGGGGGAGCATI'I‘GAATAAAAAACG
GGTI‘CTTGTACAACGACCCCGTACTATTCTATCCGTGGAGTCAGAGAATGCA
ACAATGCGTTCTTATATGCTGGTTACATTGATCTG'ITCTGCAAAATCATTATT
ACTAGGATCGTGCATGTCA’I‘T’ITI‘CGCTGGTATGTTAGTCGGTAGAACGGCA
GACGTAAAAACACCATTATGGGATACTGTATG'I‘TTGTTAATGGCT’ITCTGTG
CAGGCATTGTCGTTGGGGGAGTGGATTCTGGGGAGGTGGAATCTGGAGAAA
CAAAATCTGAATCAAAT

 

 

 

 

 

 

Standard pyrosequencing reaction:
2Forward primer (Mde-F): 5’-GTGATGGGAAGGCGATAGAA-3’
3Sequencing primer (Mde-PyroSeq): 5’-GGAGAGTGGCTGTCA-3’
4 SNP used in pyrosequencing assay (bp #320)
SAdditional SNP in pp38 gene (bp #326)
6Reverse primer (Mde-BiotinR): 5’-CATCTGCTTCATACCATCTC-3’

Nested PCR pyrosequencing reaction:
Additional outside primers

'Forward primer (pp38-R): 5'-TCTCCGCCCCCTCACTCT-3‘
7Reverse primer (MdCV-R): S’-TCCGCATATGTTCCTCCTTC—3’

 

60

Figure 2-2: Gel analysis of pp38 gene cloned into plasmid. Lane 1 = molecular-weight
marker of 100-bp ladder. Lanes 2—4 = er5 pp38 plasmid. Lanes 5—7 = er5//38CVI
pp38 plasmid

 

61

.. mucoLmMom ......

_>me=mus_._ :x. : mus: 0x.

L

0300L
OL 06
08308

D
8
O

9
.0
.V
O

C._ .V

01509

.0 .0.
9 9
0 O

08 OZ
— 0630L
b 00L 0

 

omm0m>< IT _.

 

o
or
0 ON
om
ov
. om
om
on
ow
.. om

SPWJ %

 

 

 

.. 8:

.8039: 80 0:302 mo flow 58 2: 50380: 0:030:66 wacgum

0:00:96 5 5: £000 3 a. _ .05: 08302:: 280.03 005050505 mo 38 9.3 89a 0:502 wfionosvomoim cow80>< ”Wu 0.5M?—

62

Figure 2-4: Gel analysis of pp38 gene plasmid serial dilutions. Lanes 1 & 16 =
molecular-weight marker of IOO-bp ladder. Lanes 2-7 = serial log dilutions of erS
pp38 plasmid in chicken DNA (100,000 copies/ul in lane 2 down to 1 copy/ul in lane 7).
Lanes 8-15 = serial log dilutions of er5/l38CVI pp38 plasmid in chicken DNA (same
dilutions as lanes 2-7).

 

1 234567 8910111213141516

63

OIOOL

02508

08:01

SUQQBUE. .x. : mus... :x.

9

m
.7
0

.7
my
9
0

0!. 08
08:02
06 0L
3 00F0

 

 

 

 

 

 

 

 

i-r

 

 

 

 

 

 

3:883. .......
Eon: 0:0: 3:00:08 3 LT

:50 0:0: 3:00:08 0:: Iml

 

 

 

 

0000:0505 2500—0: 32 mo MUm @800: wEBozom 0:33.: wcmonoscomoim ”min 95w?—

SPWJ %

0:00L

5032322.: :x. : was... ex.

9
m
.7
0

0F06
02308
08 0!.

.7
m
9
0

09:09

01 09
08:02
06:0L

00L10

 

 

 

 

 

 

 

 

 

 

 

 

 

wochmem .......

220 03.020 0:: $53.: 0:50.20 LT

3.35:. EEmaE lml

 

 

 

 

<75 80:03:. E 082% 029000 was: we Soto 05 waEmEoo Mom 023: 9:36:90 3308 wfioaoscomobhm "eé 95m:—

SPW.I %

65

IADSE/ / S PIAIJ

SPWJ

 

000 0::

€002: 00: 500 000020 000 :00:
0e: 000,:0 .0.; 08800: 0:000 0:0 0 080 :00 0 a 000 05 0:0 8:02:05. 08:88:: 5:3 00000m 8:058 802: 00:00 0.-" 200:0

66

IAD8€//SPIAIJ

SPWJ

 

00002: 00: 5000000020 000 :09
8: 00:2H :00; 02800: 0:000 0:0 0 080 :00 :0 a 000 05 0:: 8:02:05 0028020 .005 000000. 3:000a 0:00 00.00 "0-0 2.0:..—

67

spwJ + IADSE/ISPIAIJ

 

5000000020
05 00:20 5:3 02800: 0:80 00:0 0 0:00 :00 0 a 000 000 0:: 8:08:05 00.2880: 0:3 02:30 8:052 0m.50 0000 ”0.0 200E

68

SPIAIJ + IADSSUSPWJ

 

m3. E I

500000020 08
0020 0:05 02800: 0:030 00:0 0 080 :00 :0 a 000 05 0E 8:000:000 002888 003 08:00 8:00:00 000 00:50 50-0 200:0

69

2:: oiEnm

 

 

 

 

 

 

 

 

 

 

 

 

00 >00 :0 >00 2 :00:
i 0
i 00
S.

0 m
A
O
O
m.

i 00 0
W
O
w
0
w

i 00 I

r 02

i 02

 

 

 

 

0:5 $03600 >0. mm: 0:0 3:: 5500.0 9:00:88 35:88 00033 :0 :8 50:5: :38 owfio>< "Sun 95»:—

70

 

 

 

029000 02:00. 3 >00

 

 

 

 

 

 

 

 

rom

-ioor

i of.

 

 

 

 

 

 

com

.--000

 

"33 ISOLIINOA

 

 

 

ow:o=0:o :8: @000 :m :0 00:00:00 022:3 000000.: 5:30 we 0:00on :80 000000: :26320828 00004 fiTN ouswfi

71

:00: 200 05:00:00

 

 

 
   

 

 

 

:0 >00: :0 >00: 5 000:
i . o , o
aha—=00
- 00
2:2: 4.0..
m 8:
00:
:lmwmflwfiilnll m
0000 <30!!! . 80 m
: 002:3: ,. W
:ENOJEE. LT: - 000 m
:8 not u
-_l lo l L 38:20
n,
- 80
000 <30 \.
s\»
x . 000.
/./ .\
./ \
x. x - 8:.
00 <03 / . n.
i 1 ll}. l. i i It i- l :5. ill- :zlltlln om?

35830003 0: 080 $5990 0:0 033 :o 0000: 0:29:00 00:0 0200: :80 08062 <29 5 000850.83: :38 25> ”2A 0:0—ww—

72

000 <0.00lmfl

00 <03!

, 002: 0::
2:00:20!

 

 

 

 

 

 

 

 

 

 

 

:.\

 

7
ii :

02.

"33 ISOHINOA

Si 80

L. 000

\ ens.

0:8: 2:0 :83 :0 00:5 :0 :80 2:80.: 0088: 0:8 :8 00000: 80 0000:0000 00:5 .980» 800 :80 029800 00:03:00: :8 00000
mow08>0 80 0806:: :38 0:0: 0:0 mm“: 6:00 95:80.0 :8 00000. 25> 0000 8.: 0:020:88 98% 80:8: :38 30> "VTN 0:0—ME

73

ER 2% 9:33
5 >8 5 >8 : >8

0 , ‘ > O

 

TON

om

ow

gm 1 I a _
6:80 .9 . 8 - o3

"99 ISOHINOA

r ON?

1 03.

 

, 00..

‘ of.

 

 

 

£2838 @8985 28 8885 EB 8km 2:5
wowaofiaofigzh N. can 259 8.980 m 82m wowfigu £5830 womaoa$=maE>€E S we: B>o bags ~38 25> fiTN charm

74

Figure 2-16: Virus copy number comparison for individually housed chickens at 21 days, challenged with er5. Significant results

based on the average are indicated by different lowercase letter.

 

IPulpjj

u]
u.
u.
I

I ‘EDustfl

0 Ill|Ill||||||||||||||I|I|llll|

.o\m

\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\

llllllllllll
.\\\\\\\\\\\\\\\\\\\V\\\\\\\\\\\\\\V

WW

llllllllllllllllllllllllllllllIlllllllllllllllllllll|||||||||
mm

mm

W

W

IIIIIIII
WW

 

l l
O O
0 LO
(0 N

O
O
(\l

150 e
100 a

||9315°qlN3A

75

50~
0

Avg

1

Control Control

 

 

 

Chapter 3
DEMONSTRATION OF VIRAL DOMINANCE FOLLOWING SIMULTANEOUS
CHALLENGE OF TWO FULLY VIRULENT MAREK’S DISEASE VIRUS
STRAINS
Abstract
Marek’s disease vaccination protects chickens against the development of

disease, but does not prevent infection or transmission of infectious virus. Previous
studies have demonstrated the presence of multiple strains of Marek’s disease virus
simultaneously circulating within poultry flocks, leading to the assumption that
individual birds are repeatedly exposed to a variety of virus strains in their lifetime. A
series of four experiments were designed to test whether coinfection was possible with
two fully virulent SI MDV strains, using two similar (er5 and er5//38CVI) and two
different (J M/102W and er5//38CVI) virus pairs. Bursa of Fabricius, feather follicle
epithelium, spleen, and tumor samples were collected at multiple time points to
determine the frequency and distribution of each virus present following simultaneous
infection using pyrosequencing, immunohistochemistry and quantitative PCR. Virus
dominance was significant between the second virus pair, but was not consistent
between the two similar strains. Both viruses were readily detected in individual bursa
and feather follicles in 16 of 23 birds (70%), indicating that the presence of one virus
did not exclude the second virus. Results from tumors demonstrated differences
between the predominance of each strain within multiple tumors fi'om individual birds,
suggesting a possible polyclonal origin. These effects may be relevant under field
conditions as no significant differences were detected between vaccinated and

unvaccinated chickens. Viral dominance may be an important factor that influences the

76

outcome of coinfection under field conditions, including the potential outcome of

emergence or evolution of more virulent strains.

Introduction

Marek’s disease (MD) is a lymphoproliferative disease in chickens caused by an
alphaherpesvirus, Marek’s disease virus (MDV). The disease is typically characterized
by enlarged nerves and visceral tumors composed of transformed T cells. MDV is
classified into three serotypes: serotype 1 (SI) includes pathogenic strains and their
attenuated variants, serotype 2 (82) includes all apathogenic chicken strains, and
serotype 3 (S3) includes turkey herpesvirus (HVT), which is also apathogenic in
chickens (12). 82 and S3 have been used as vaccines, as well as several attenuated
strains of 81. Although vaccination has successfully been used to prevent disease, the
vaccines do not prevent against infection or transmission of the virus. Thus, as early as
2 weeks after inoculation or exposure, MDV is spread in the feather dust to other
chickens as the keratinized layer of infected epithelial cells is sloughed and shed (17).
Shedding appears to peak at approximately 3-5 weeks and continues indefinitely (94,
106). Infected feather dust can remain infectious in a poultry house for at least several
months at 20-25°C and for years at 4°C (17, 58).

As a result of continual shedding and virus stability in the poultry house
environment, it is assumed that most poultry houses are rich in infectious virus.
Chickens raised in a commercial setting are likely exposed to MDV repeatedly and
continuously over their lifetime. It is unknown whether chickens are susceptible to

more than one virulent strain (coinfection), either simultaneously or later in time

77

(superinfection). Superinfection may be in the form of re-infection with the same strain
or with a novel strain. Although superinfection may be more likely, if the virus load is
high enough, newly placed chicks may be exposed to multiple strains virtually
simultaneously.

Evidence exists for the simultaneous presence of more than one strain of MDV
in a chicken flock. Biggs et al. reported on the epidemiology of MD in six flocks from
a broiler female grandparent line (10). Virus isolates were propagated from five flocks,
and then 25 isolates were examined and classified for their pathogenicity as either
apathogenic, classical or acute. Four of the five flocks had strains present from more
than one level of pathogenicity. Two flocks, including a single house from one flock,
had strains present from all three levels of pathogenicity. A study by Jackson et al.
confirmed the presence of multiple pathotypes circulating within the same flock, and
concluded that the environment during the first 8-9 weeks of life was of greatest
importance in subsequent mortality from MD (56). Witter et al. evaluated the virulence
of multiple isolates (2-3) from two separate flocks and found very similar pathotypes
(95). These isolates were all from tumor-bearing birds which may have created a bias
towards isolation of more virulent strains. This also suggests, however, that a single
virus strain may have the ability to become dominant.

Vaccine studies have shown that coinfection can occur in chickens with virulent
and avirulent MDV strains. Witter et al. showed that 81 virus was isolated when MDV
was challenged afier initial infection with S3 HVT (104). Okazaki et al. concluded that
HVT was fully protective when administered at least one week before challenge, and

the vaccine gave partial protection when administered simultaneously with challenge at

78

one day of age (25% MD vs 90% MD in unvaccinated challenged group). (69). Calnek
et al. infected birds with 82 88-1 and S3 HVT-4, followed by J M-lO, to study the effect
of vaccination on early pathogenesis (2 l ). Viremia was present, although reduced
following subsequent challenge. Cho et al. reported evidence of coinfection with acute
(Id-1) and mild (HN) MDV strains following simultaneous challenge, although HN was
later classified as a nonpathogenic 82 strain (28, 81, 93) After contact exposure to both
strains, Cho and Kenzy concluded that infection with one strain, whether acute or mild,
did not exclude subsequent infection by the other strain (27). After simultaneous
challenge by injection, Cho et al. reported isolates from both strains were present in
susceptible birds at 1 week continuing through termination at 8 weeks, with gradual
reduction from 6 mixed isolates down to 1 mixed isolate at 8 weeks (26). Cho also
reported the presence of both viruses in feather tip extracts following simultaneous
challenge (25). Coinfection with attenuated 81 strains and virulent 81 strains has also
been documented. Churchill et a1. infected chickens with live attenuated SI virus
(HPRS-l 6/att) and demonstrated that subsequent challenge with more virulent Sl
strains can cause infection (3 0).

Given the evidence of multiple strains of MDV within a flock, the question
remains whether simultaneous exposure with fully virulent strains will result in
simultaneous infection. Possible outcomes of simultaneous infection include
independent growth of each strain, total exclusion of a lesser strain, or dominance
without complete exclusion. If the phenomenon of dominance exists between different
strains, this may lead to decreased replication and/or transmission of the subordinate

strain and may affect the contribution of each virus in tumors. The potential selection

79

of dominant strains within individual chickens may be an important mechanism in the
evolution of virus strains within a flock. Simultaneous infection of multiple strains and
subsequent dominance of one strain over the other may lead to transmission of only the
dominant virus to neighboring chicken flocks or future flocks exposed to contaminated
litter.

Prior obstacles have made the study of virulent Sl coinfection difficult using
DNA and antigen as methods for critical analysis, particularly because of the difficulty
distinguishing between SI viruses. Previous studies have relied on plaque morphology
or changes in total MD lesions to make assumptions about coinfection. The recent
development of mutant strains with slight differences in the pp38 gene has provided a
convenient method for distinguishing between two similar fiilly virulent strains (er5
and er5//38CVI). The recombinant erS was generated by Reddy et al. using
overlapping cosmid clones produced from wild type Md5 (75). The recombinant virus
er5//38CVI was produced by Lee et al. using the erS cosmid clones, substituting
the pp38 gene from CVI988/Rispens (60). Thus, these two viruses share the same er5
backbone and are only different in the pp38 gene which differs by two single nucleotide
polymorphisms (SNPs). Differentiation between these two viruses is possible by
detecting the SNPs coupled with the availability of monoclonal antibodies that detect
the difference. The viruses were also useful based on their similar pathogenicity, as
reported by Gimeno et al., in which there were no differences in the frequency of
lymphoproliferative lesions between the viruses at either 6 or 15 wpi (43).

This study had two hypotheses. First, following simultaneous challenge with

two fully virulent 81 strains, both strains will replicate, transmit to other birds and lead

80

to tumor formation. Second, following simultaneous challenge, both strains will have
equal chance of transmission and tumor formation. Understanding the potential
phenomenon of dominance under the conditions of simultaneous infection is essential to
evaluate data from non-simultaneous superinfection studies. All experiments were
carried out using two virus pairs, the first being a pair with similar pathogenicity (erS
and er5//38CVI, described above) and the second being a pair of different
pathogenicity (J M/102W and er5//38CVI). Samples were collected and analyzed
from bursa during early cytolytic infection, feather follicle epithelium during late
cytolytic infection, and spleen and tumors following transformation. Maternal antibody
positive, vaccinated chickens were used in the last set of experiments to test any

observed effects in conditions relevant to the normal poultry environment.

Materials and Methods

Chickens

White leghom 1515x71 chickens, a Fl hybrid cross of MD susceptible 1515

males and 71 females, were used in these experiments (2). Maternal antibody negative

(Ab—) chickens were used in experiments 1 and 2 to prevent any potential masking of
significant effects. Ab- chickens were reared from an SPF breeding flock with no MD
vaccinations or exposure. The flock was negative for MDV antibodies by routine
surveillance tests. Maternal antibody positive (Ab+) chickens were used in experiments
3 and 4 to reduce mortality and simulate field conditions. Ab+ chickens were reared
from breeder hens vaccinated at hatch with 2000 plaque forming units (pfu) of HVT and

at 25 weeks with 2000 pfu of SBl and Mdl l/75C viruses for exposure to all three

81

serotypes. Both flocks were also negative for exogenous avian leukosis virus and
reticuloendotheliosis virus from routine surveillance testing. All birds were maintained
in negative pressure Horsfall-Bauer isolators. Experiments were approved by the

ADOL Animal Care and Use Committee.

Viruses

Two virus pairs were used for simultaneous dual infection: 1) J M/ 102W and
er5//38CVI; 2) er5 and er5//38CVI. The recombinant erS was generated by
Reddy et al. using overlapping cosmid clones produced from wild type Md5, a vaDV
strain (75, 93, 103). The recombinant virus er5//3 8CVI was produced by Lee et al.
using the er5 cosmid clones to substitute the pp38 gene from CV1988/Rispens in
order to determine whether the pp38 gene from CVI988/Rispens was related to the
protective properties of the vaccine (60). Thus, these two viruses share the same er5
backbone and are only different in the pp38 gene. The two pp38 genes differ by two
single nucleotide polymorphisms (SNPs) which was the basis for differentiation by IHC
and pyrosequencing. The viruses were also chosen based on their similar pathogenicity,
as reported by Gimeno et al., in which there were no differences in the frequency of
lymphoproliferative lesions between the viruses at either 6 or 15 wpi (43).

The third virus used was J M/ 102W. The parental strain, J M, was isolated by
Sevoian in 1962 (84), and later cloned by 3X endpoint passage in DEF and designated
as J M/102W (88). J M/ 102W was pathotyped as vMDV by Witter and is a principle

prototype strain frequently used in MD experiments (93, 95). JM/102W was chosen to

82

pair with er5//38CVI to study dual infection of two MDV strains of differing
pathotype.
Experiment 4 compared superinfection results between vaccinated and

unvaccinated chickens. For this experiment, chicks were vaccinated at hatch with 2000

pfu of S3 HVT (strain FC126/2) vaccine (101, 102).

Monoclonal antibodies
The MDV pp38 gene, located in the IRL and UL region, was utilized for virus

differentiation by immunohistochemistry (IHC) and pyrosequencing. All S1 MDV
strains reportedly have identical bases in the pp38 gene except bases #320 and #3 26
(31, 38). Cui et al. determined that monoclonal antibody (Mab) H19 was specific to
base pair #320, and was only positive when adenine (A) was present, whereas Mab T65
was specific to base pair #326, and was only positive when guanine (G) was present
(34). J M/ 102W and er5 used in this study have A at both base pair #320 and #326
and thus were positive for H19 only. er5//38CVI has G at both base pairs #320 and

#326 and was positive for T65 only.

Immunohistochemistry testing

Serial sections of frozen bursa and feathered skin were cut at -25°C and dried
overnight under vacuum at room temperature. Slides were then fixed in acetone for 45
minutes, air dried, and stored at room temperature. Peroxidase staining of bursa and

FFE tissue sections was performed using the Vectastain ABC kit, as described by the

83

manufacturer.ll Slides were loaded onto a DakoCytomation Autostainer, where they
were rinsed with PBS and then incubated for 20 minutes with 150 uL of normal

blocking sera (horse) from the same species the secondary biotynlyated antibody was

made from.‘2 Slides were rinsed once with PBS and serial sections of each tissue were
incubated for 30 minutes with either 150 uL of H19 (1 :3200) or T65 (1 :2000)
monoclonal antibody. Slides were rinsed three times with PBS and then incubated for
30 minutes with 150 uL of the biotinylated secondary antibody. Slides were again
rinsed three times with PBS followed by a 30 minute incubation with 150 uL of the
avidin and biotinylated horseradish peroxidase (ABC complex). Slides were then rinsed
three times with PBS and then stained for 7 minutes with 150 uL of the DAB solution
including the buffer and hydrogen peroxide which reacts with the horseradish
peroxidase and produces a dark brown color. Two rinses followed and then slides were

counterstained with hematoxylin and coverslipped.

Virus Isolation

Viral plaque assays were performed using spleen cell suspensions plated on
duck embryo fibroblasts (DEF). Whole spleens were collected at termination (day 56)
and placed in 2 mL of Leibovitz McCoy (LM) cell media. Samples were kept on ice
and processed immediately following the termination. Spleen samples were rinsed in
PBS, placed inside folded gauze pads, and mashed using a syringe handle.
Approximately 5 mL of LM media was rinsed through the cheese cloth to collect cells.

100 uL of the cell suspension was added to 1 'mL freezing media (55% - 4% LM media,

 

H Vector Laboratories, Inc., Burlingame, CA
I2
Dako, Glostrup, Denmark

84

25% - calf serum, 20% - Dimethyl sulfoxide [DMSO]) and the samples were stored at -
80°C until later plated on DEF.

Cell suspensions were thawed 2-6 weeks later, counted and plated on 2-4
secondary DEF cultures. After 7 days, cultures were fixed with an alcohol-acetone
mixture (6 parts acetone and 4 parts 95% alcohol) and stored at -20°C until they were
stained and read. Fixed cell culture dishes were rehydrated and incubated for 30
minutes with the same primary antibodies used for immunohistochemistry (H19 or
T65), half of the plates with one antibody and half with the other. Plates were washed
with three rinses of distilled water followed by three rinses with PBS and then incubated

for 30 minutes with a fluorescein-conjugated goat anti-mouse IgG (H + L) secondary

antibody”, purified by gel filtration. Plates were washed again with three rinses of

distilled water and PBS and then plaques were counted using a fluorescent microscope.

Pyrosequencing assay

Pyrosequencing was performed with bursa, FFE, spleen and tumor samples on a
PSQ 96MA system, according to the manufacturer’s protocol. 14 Briefly, biotinylated
PCR product specific to the MDV pp38 gene was generated using a standard PCR with
a biotinylated reverse primer (forward primer 5’-GTG ATG GGA AGG CGA TAG
AA-3’, reverse primer 5’ Biotin-CAT CTG CTT CAT ACC ATC TC-3’). The PCR
product was then captured on streptavidin-coated beads by shaking binding buffer (10
mM Tris-HCl, 2 M NaCl, 1 mM EDTA, 0.1% Tween 20), streptavidin sepharose beads,

and PCR product at 1400 rpm for 10 minutes. Vacuum probes were used to collect the

 

13’ MP Biomedicals, LLC, Solon, OH
14 Qiagen, Hilden, Germany

85

beads for transfer to 70% ethanol, denaturation solution (0.2M NaOH), and wash buffer
(IOmM Tris-Acetat, pH 7.6) to denature and remove the non-biotinylated DNA strand.
The beads were then transferred to a new 96-well plate containing annealing buffer (20

mM Tris-Acetat, 2 mM MgAcz) and the sequencing primer (5’-GGA GAG TGG CTG

TCA-3’), designed using SNP Primer Design from Pyrosequencing AB version 1.0.1.]5
Nucleotides were added one at a time and incorporation was measured and quantified
by light generated by successfiil elongation. The resulting sequence reported the
frequency of each nucleotide that was incorporated at the location of the SNP.
Pyrosequencing results were analyzed from all samples with passing results, meaning
from samples in which sequence was detected. Samples that failed either did not have
adequate virus present (e.g. negative control samples) or there was either insufficient or

excessive total DNA present.

Quantitative PCR assay

Quantitative PCR (qPCR) using a multiplex reaction was performed on .F FE

samples using an ABI 7500 Real-Time PCR System16. One TaqMan® probe
(fluorogenic 5 ' nuclease chemistry) was specific to the MDV pp38 gene and the other
specific to an endogenous reference, glyceraldehyde-3-phosphate dehydrogenase
(GAPDH). Both probes used TAMRATM (tetramethylrhodamine) quencher dyes. The
probe specific to MDV used a green emitter (500-549nm) reporter dye, 6FAMTM (6-
carboxyfluorescein, excitation 492, emission 515nm) whereas the probe specific to

GAPDH used a yellow emitter (550-584nm) reporter dye, VICTM (excitation 53 8,

 

15 Applied Biosystems Inc., Foster City, CA
l6 ' .
Applied Biosystems Inc., Foster City, CA

86

emission 554nm). The following conditions were used: 1) 50°C —— 2 minutes; 2) 95°C —
10 minutes; 3) 95°C — 15 seconds; 4) 60°C — 1 minute with steps 3-4 repeated 40 times.
Each well contained primers and FAM-TAM probe for the MDV pp38 gene (forward
primer 5’-GTG ATG GGA AGG CGA TAG AA-3’, reverse primer 5’-TCG TCA AGA
TGT TCA TTC CCT G-3’, TaqMan probe 5’FAM-TACCGCCTGAGCCCCGGAGG-
3’TAMRA) and primers and VIC-TAM probe for the cellular gene GAPDH (forward
primer 5'-ACA GAG GTG CTG CCC AGA A-3', reverse primer 5'-ACT TTC CCC
ACA GCC TTA GCA-3', TaqMan probe 5' VIC-TCA TCC CAG CGT CCA CT-3'
TAMRA). A standard curve with 10-fold serial dilutions of plasmids containing MDV
pp38 or GAPDH was used to calculate absolute copy numbers of each gene per sample.

For the calculation of averages, outliers were removed if the value was greater than 1.5
times the interquartile range (the difference between the 25th and 75th percentile). No

more than two values were removed per group. Results were expressed as the viral

copy number per host cell.

Experimental design

Four experiments were designed to characterize coinfection in simultaneously
challenged chickens (Table 3-1). All four experiments utilized J M/102W, erS and
er5//38CVI virus strains challenged alone or in combination. Data from each virus
pair was analyzed separately as parts A & B of each experiment. Single-challenged
control birds were used to confirm the specificity of each assay and as a comparison for

quantitative differences compared to dual-infection. All virus and vaccine inoculations

87

were administered intra-abdominally (IA), with each dose divided into two sites.

Specifics of each experiment are detailed below.

Experiment 1 — Twenty 1515x71 Ab- chickens were singly-challenged with each virus

or dually-challenged with both viruses at hatch, except one group of uninfected
chickens used as controls. Bursa samples were collected from 6 birds that were
removed and euthanized from each lot at 4 and 6 days post infection (dpi). The
remaining 8 birds from each lot were euthanized at 21 dpi and feathers and skin sections
were collected from the subhumeral feather tract for DNA and IHC testing,

respectively.

Experiments 2 & 3 — These two experiments were identical except experiment 3 used
Ab+ chickens to reduce early mortality. Groups of seventeen chickens were singly-
challenged with each virus or dually-challenged with both viruses at hatch to increase
the probability of acquiring visceral tumors. Surviving chickens were euthanized at 56
dpi and samples were collected. Tumors were collected for pyrosequencing and qPCR
testing, and spleens were collected for virus isolation. In experiment 2, only visceral
tumors were collected for analysis. In experiment 3, all chickens with visceral tumors
were sampled plus enlarged nerves were randomly collected from additional chickens
without visceral tumors in order to collect samples from a minimum of eight birds per

lot.

88

Experiment 4 — This experiment consisted of two lots of seventeen 1515x71 Ab+

chickens dually-challenged with virus pair 1, and another two lots dually-challenged
with virus pair 2. Vaccinated lots received IA inoculation of 2000pfu of HVT at hatch.
Virus challenge was administered at 6 days for both vaccinated and unvaccinated
groups. Feathers were collected from each chicken at three time points: 13, 26, and 63
dpi. Tumors were also collected at the time of the last feather collection. All feathers
and tumor samples were analyzed from tumor-bearing chickens. In addition, feather
samples were analyzed from remaining non tumor-bearing chickens chosen in ranking

order of wing band numbers to have a minimum of eight birds per lot.

Data analysis

IHC results were analyzed by comparing individual bursa and feather follicles
from serial sections stained with H19 or T65 monoclonal antibodies. The percent of
follicles staining for each virus or both were contrasted. Pyrosequencing results
compared the frequency of each virus present within all bursa, FFE or tumor DNA
samples that had passing results (sequence was detected). Differences in frequency
(dominance) between viruses in each experiment were analyzed by a paired t-test.
Virus isolation results were analyzed using the Wilcoxon rank-sum test due to low
numbers of samples in most groups. The analysis compared plaque counts for each
virus in dual-challenged chickens with single-challenged chickens challenged with the
same virus. Differences in virus copy numbers from qPCR results were compared by
ANOVA for significant differences between viruses based on challenge group and

sampling day. Significant results were followed by Duncan post hoc analysis to

89

determine which groups were different. P-values less than 0.05 were considered
significant. Data were analyzed with the statistical program SAS 9.1 (SAS Institute

Inc., Cary, NC).

Results
Bursa and FFE

Experiment 1 was designed to compare the differences between virus frequency
and distribution for both virus pairs during early and late cytolytic infection in the bursa
and F FE, respectively. Data was not included from day 4 bursa samples due to a lack of
staining in all but one IHC sample and lack of virus sequence detection in all but 4
samples by pyrosequencing. In experiment 1A, virus mixtures were detected in 2 of 5
bursa samples by pyrosequencing, and 2 of 3 samples with staining by IHC (Table 3-2).
Seven of eight FF E samples appeared as mixtures by at least one assay, although
pyrosequencing did not detect mixtures from two of these birds (M6890 & M6882). A
majority of individual feather follicles were found to stain for both viruses following
dual challenge and viruses were frequently present within the same region of the follicle
(Figure 3-1). The frequency of er5//38CVI was significantly greater than JM/102W
in bursa and FFE samples from dual-challenged chickens (p=0.0100).

In experiment 1B, virus sequence was detected from 4 bursa samples from
chickens dual-challenged with erS and er5//38CV1 (Table 3-3). Both viruses were
detected from 3 of the 4 by both pyrosequencing and IHC. In most cases, only one
virus was detected within individual bursa follicles, although both viruses were detected

in a small number of follicles (Figure 3-2). Both viruses were also detected from 5 of 7

90

FF E samples by both pyrosequencing and IHC. The frequency of erS was
significantly greater than er5//38CVI in bursa and FFE samples (p=0.0006).

The individual contributions of each virus to qPCR results from each sample
were calculated based on pyrosequencing frequencies. Combined qPCR results from
experiments 1A & 1B showed that in the bursa, dual-challenge did not change the
average copy number/host cell of each virus compared to single challenged birds based
on ANOVA (p=0.0968) (Figure 3-3). In contrast, qPCR results fi'om F FE samples
illustrated a significant decrease in average copy number/host cell for both viruses in

experiment 1A, but only one virus in experiment 1B (p=0.0009) (Figure 3-4).

Tumors

The purpose of experiments 2 and 3 were to look for evidence of virus
dominance in tumors. Experiments 2 and 3 were identical except Ab+ birds were used
in the latter experiment to increase survival until sampling. Virus sequence was
detected from 14 tumors from 8 birds dually-challenged with J M/102W and
er5//38CVI in experiments 2 and 3 (Table 3-4). Of these 14 tumor samples, 10
samples were positive for both viruses. Seven tumors from 5 birds favored J M/102W
and 7 tumors from 5 birds favored er5//38CVI. Two of four birds with multiple
tumors had virus frequencies that were markedly different between tumors and in fact
were predominant for different viruses. Virus plaque counts were unusually low from
spleen cell suspensions, but of 6 samples that stained positive, two stained for only

JM/102W and four stained only for er5//38CVI. There were no significant

91

differences between plaque counts for each virus in single versus dual-challenged
chickens (p=0.3152).

In chickens dually—challenged with erS and er5//38CVI in experiments 2
and 3, virus was detected in 17 tumor samples from 11 chickens (Table 3-5). Both

viruses were detected from 3 samples from 2 birds and erS was favored in all three.

 

In the remaining 14 tumors, er5 was detected exclusively. More than one tumor was '
tested for 5 birds and unlike with samples from virus pair 1, all frequencies were similar i
within individual birds. Virus plaques stained only for erS from spleen cells isolated if

from 6 of 7 chickens. The remaining chicken had plaques present from both viruses,
although twice as many were positive for er5. There were no significant differences
between plaque counts for each virus in single versus dual-challenged chickens

(p=0.2413).

Effect of vaccination

Experiment 4 was designed to confirm potential virus dominance in vaccinated
birds. After simultaneous challenge with JM/102W, the frequency of er5//38CVI was
significantly higher compared to J M/ 102W in the FF E and was nearly exclusively
present in both unvaccinated and HVT-vaccinated chickens (p<0.0001) (Table 3-6).
There was no significant difference in virus frequencies between vaccinated and
unvaccinated chickens in FFE samples (p=0.3466). er5//38CVI was dominant in all
14 tumors, including 4 tumors in which both viruses were present. Interestingly, four

chickens had mixed infections in heart tumors, in contrast to the exclusive

92

er5//38CVI detected from FFE, and frequencies differed between multiple tumors
from the same bird.

Both viruses were detected in DNA isolated from F F E in 6 of 9 unvaccinated
chickens and 3 of 9 vaccinated chickens dually-challenged with erS and
er5//38CVI (Table 3-7). Mixtures frequencies were quite consistent between the
three time points in individual birds. In contrast to experiments 1-3, however,
er5//38CVI was dominant compared to erS as it was present in higher frequency for
all birds within mixtures and those with only one virus present (p<0.0001). Similar to
virus pair 1, there was no significant difference in virus frequencies between vaccinated
and unvaccinated chickens in FFE samples 33:0. 1602). Three of eight tumors had
substantial mixtures of both viruses detected and similar virus frequencies were
detected between all tumors and FF E samples in respective birds.

Quantitative PCR results from both virus pairs were highly variable, but except
for three day 26 samples of er5//38CVI, there were no significant differences between

virus quantity in vaccinated versus unvaccinated chickens (Figure 3-5).

Discussion

Data from experiments 1-4 provided evidence that two virulent S1 MDV strains
could be present simultaneously in chickens. Mixtures of both strains were relatively
common in bursa and feather follicles in birds challenged with either virus pair.
Interestingly, virus mixtures were also present within at least 21 of 53 individual
tumors. The phenomenon of viral dominance was also demonstrated. For the first time,

certain virulent 8] virus strains were shown to have a competitive advantage over

93

others (Table 3-8). In birds simultaneously challenged with JM/102W and
er5//38CVI, all bursa samples and 6 of 8 FFE samples from experiment 1A favored
er5//38CVI whether samples were composed of one or both virus strains. The same
was true in experiment 4A in which er5//38CVI was nearly exclusively dominant in
F FE samples from both vaccinated and unvaccinated chickens, although it is unclear
why the predominance was more one-sided compared to experiment 1. JM/102W used
in experiments 1 and 4 came from different lots, and perhaps minor variations in titer
estimates could have significant effects in this type of challenge model. The dominance
of er5//38CVI was statistically significant in both experiments.

Dominance may be more difficult to detect using two strains of similar
virulence. The majority of bursa and F FE samples from chickens simultaneously
challenged with erS and er5//38CVI in experiment 1 consisted of virus mixtures,
with 6 of 7 favoring erS. The remaining non-mixtures also favored er5. In
experiment 4, both viruses were again detected from most FF E samples (6 of 9 birds),
although in this experiment er5//38CVI was present in higher proportions in both
vaccinated and unvaccinated chickens. Although one virus tended to have a slight edge
of the other within an experiment, the predominant virus was not consistent between
experiments. If both viruses had similar inherent dominance, this may account for the
fact that mixtures of both viruses in bursa and F F E samples were more prevalent with
the similar viruses (63% of birds) compared to the more different viruses (28% of birds)
in all experiments.

The presence of one virus strain within bursa and feather follicles apparently

provided no exclusion fi'om infection with the second virus. IHC confirmed the

94

presence of both viruses within individual follicles. In feather follicles, a very high
percentage of follicles stained for both viruses and many times the positive cells
occupied the same proximity (Figure 3-1). In most cases, dominance of a particular
virus within a chicken as evident by pyrosequencing results was consistent with
proportions of each strain within individual follicles (data not shown).

Viral dominance was less evident in tumor samples, although tumors originate
from rare individual transformation events and may be less appropriate for measuring
dominance. It was interesting that with the erS and er5//38CVI virus pair, the
frequency of viruses in tumors was very consistent between multiple tumors in
individual birds, and was also very consistent compared to F FE samples from the same
bird. On the other hand, the JM/102W and er5//38CVI virus pair led to some
differences in the virus frequency between tumors in the same bird, suggesting that
these tumors may have a polyclonal origin. There is current debate on the clonality of
MD tumors. Some have reported that tumors were monoclonal based on the identical
number of hybridization signals of interphase nuclei and the identical pattern of
chromosomal integration in metaphase spreads (37). In contrast, other reports have
suggested a polyclonal tumor origin based on their analysis of T cell receptor (TCR)
variable [3 chain gene family expression and spectratyping (13, 24). Another study also
found differences in 132-bp repeat expansion between two tumors from one bird,
suggesting two separate transforming events (86). It is difficult to conclude why the
two virus pairs in the current study had different results in tumors, but in the case of the
er5 and er5//38CVI virus pair, the results would suggest that tumors may be the

result of multiple transforming events, possibly from transformed cells containing more

95

than one virus. The presence of virus of both viruses within a tumor sample, however,
does not necessarily identify the virus responsible for transformation. These
experiments may be a useful model for future studies of clonality of tumors.

Vaccination seemingly had no effect on the detection of dominance between
virus strains for either pair. Since FC126 (HVT) protects better against J M/102W than
Md5, it seemed likely that vaccination would benefit er5//38CVI (95). This was
difficult to assess, however, since er5//38CVI was already significantly dominant to
J M/ 102W in the absence of HVT. The two similar viruses were likely protected to the
same degree by HVT, preventing any advantage of one virus over the other.

The topic of replication potential and its relationship to virulence has been the
topic of several recent papers. Following challenge with v and vv+ MDV strains in
resistant and susceptible birds, Yunis et al. found that the virulent virus became latent at
6 dpi, but the vv+ virus never went into latency in both chicken lines during 10 days.
(109). The prolonged virus replication and presence of viral transcripts for the vv+
virus throughout 10 days could cause more severe damage and atrophy of lymphoid
organs.

J arosinski et al. showed significantly higher levels of viral replication for the vv+ strain
RK-l in two chicken lines compared with the v strain J M-16 (57). And Calnek et al.
demonstrated a prolonged phase of cytolytic infection at 7-8 dpi for three vv+ isolates
(RK-l , 584A, 648A) compared with two v isolates (J M16, GAS) (19).

If virulence is partially defined by replication potential, this would give a clear
advantage to whichever virus can replicate faster during coinfection. Presumably when

administered simultaneously, both viruses have equal chances for replication within

96

bursa follicles. The collection of FF E samples from the same birds at three time points
showed the relative stability of virus frequencies over time, indicating that any
competition that occurred early on was finished and established by 13 days. The
proportions of each virus that were present at day 13 in the FF E changed very little
during the next two samplings, regardless of which virus was more predominant at the
first samplings. The two similar viruses in this study did not have a significant
advantage over each other, yet coinfection with the two differing viruses was
consistently dominated by the virus of higher virulence. Whether an advantage is the
result of preventing coinfection in the same cells is unclear and was not answered by
this study. The possible correlation between dominance and replication ability requires
additional confirmation.

Viral dominance may be an important factor that influences the outcome of
MDV coinfection under field conditions. This effect may be especially important in the
selection and propagation of more virulent strains within a poultry flock. Additional
studies are necessary to elucidate the effect of dominance in strains of similar virulence
and to evaluate the effect in conditions of non-simultaneous challenge (superinfection).
Although a weak vaccination was intentionally used, the results suggest that the effect

of dominance may be relevant under field conditions.

97

22.222 ooom 3.028 222 22223 223822 28 2882822228288 283 0:823 252

 

 

 

 

 

 

2.
2:222 8822 22028 228 2829288 883 828.25 22 M
222828 2m 228 2822:2228 mum 28:28 28888 883 82825 w wc2:2mEo2 8222 28:8 .2228 c 28:22 2. :0 828.2222 0 80.2.2 28282888 8283 3.5m m
2 5&2: 2228228 883 8:820:20 22 2
8.288 8 .8m .2 2.228.823.8822 88 2828 8 2522288322 + 28322 522 83828.2 2 8
83828 88 .8 .2 2822822233222 88 2828 8 2522228322 + 28322 82228.2 2 2.
8:88 8 .8m .2 2822822233222 88 8288 8 56222828322 + 38:3: 2.522 82228.2 2 m
2828 :8 .8m .2 282288323822 88 .3828 8 2.6828832. + 38:3: 83828.2 2 m
mm~AmU MO .ON .M— m—Obfloo UDHOUMGMG: 0>Cmmom Om _ V
288 88 2588238222 88 2828 2 258228328 + 88322 82228.2 2 8
8:88 88 28228833332 888 :88 2 2582828322 + 38:3: 82228.2 2 8
8828 88 28222388 88 :88 2 .2225 258228322 82228.2 2 2.
.388 88 8228:2822 88 3828 2 3288 28322 82228.2 2 m
83828 88 2:28:88 88 .2828 2 3288 38:3: 82228.2 2 N
8.328 on 828280 2888,2525 932228022 3 2 m
288 88 8829822282888 :88 2 258828322 + 28322 8228882 2 8
83.28 88 282288333222 88 :88 2 2582828322 + 38:3: 82288872 2 8
388 88 282283822 88 :88 2 3288 258828322 8228882 2 2.
2828 88 825.2822 88 .2828 2 3225 828322 8.288882 2 m
8388 88 2:28.588 88 :88 2 :280 38:3: 8228882 2 N
@2828 cm 8282280 2888,2525 83228372 02 2 m
8.288 2m .8 .2. 82282823388 88 :88 2 25822822 + 8832.2 8.388882 on 8
8388 2m .8 .2. 282288233822 88 .288 2 258888322 + 38:3: 8328882 8. 8
288 2m .8 .8 2:28.328 88 3828 2 3288 258228322 83288872 on 2.
2828 2m .8 .2. 28.22.2328 88 3.8 2 3288 28322 83288822 cm 8
8.2828 2m .8 .2. 2828:2822 88 :88 2 3225 38:3: 8228882 8 N
~23: 2m .8 .2. 82922288 288288222225 o>22mwoz on 2 2
an
a 0 2:2 :8 a
on: ”22222255 3.25 888222220 22222.5 away—2220 822235» 2 28 2222 2a 2 2 8228292229 2222 8.— 52m—

 

 

 

 

 

 

 

 

 

 

o:222:o 888228220 “TM 2222:.

98

 

Table 3-2: Experiment 1A. Pyrosequencing and IHC results from bursa (6 dpi) and
FFE (21 dpi) samples in chickens single or dual-challenged with JM/102W (J) and/or
er5//38CVI (M2). Pyrosequencing results show the ratio of each virus present in

DNA isolated from the tissue. IHC results are the percentage of positive-staining

follicles for one or both viruses. Samples that failed pyrosequencing are indicated by

 

 

 

 

 

 

 

 

 

an F.
Pyrosequencing IHCI
20 E” (Frequency 0f (% follicles staining
g E ‘3. =11: g each v1rus) for each virus)
2 g S e E .2 J M2
0 m m 'u :1: E" %J:%M2 only Mixed only
Controls (averaged per group)
None 6 Avg (6 birds) Bursa F 0% 0% 0%
J 6 Avg (3 birds) Bursa 97:3 8% 0% 0%
M2 6 Avg (4 birds) Bursa 0: 100 0% 0% 12%
Dual-challenged (individual birds)
J, M2 6 M6898 Bursa 11:892 0% 1% 2%
J, M2 6 M6886 Bursa 6:94 1% 2% 16%
J, M2 6 M6899 Bursa 0:100 0% 0% 30%
J, M2 6 M6895 Bursa 0:100 0% 0% 0%
J, M2 6 M6896 Bursa F 0% 0% 0%
J, M2 6 M6884 Bursa 0:100 0% 0% 0%
Controls (averaged per group)
None 21 Avg (8 birds) F FE F 0% 0% 0%
J 21 Avg (8 birds) FFE 98:2 98% 0% 0%
M2 21 Avg (8 birds) FFE 0:100 0% 0% 98%
Dual-challenged (individual birds)
J, M2 21 M6890 FFE 0:100 0% 29% 71%
J, M2 21 M6882 FFE 0:100 0% 23% 77%
J, M2 21 M6893 FFE 99:1 86% 0% 0%
.1, M2 21 M6881 FFE 17:83 0% 92% 8%
J, M2 21 M6888 FFE 45:55 0% 100% 0%
J, M2 21 M6883 FFE 70:30 0% 100% 0%
J, M2 21 M6900 FFE 43:57 0% 0% 0%
J, M2 21 M6892 FFE 7:93 0% 80% 20%

 

 

1 Number of bursa and feather follicles analyzed per sample averaged 286 (range 138-

43 3) and 9 (range 2-18), respectively.

2 Frequency of er5//38CVI was significantly greater than J M/102W in bursa and FFE
samples from dual-challenged birds by paired t-test (p=0.0100).

99

 

Table 3-3: Experiment 1B. Pyrosequencing and IHC results from bursa (6 dpi) and
FFE (21 dpi) samples in chickens single or dual-challenged with er5 (M 1) and/or
er5//38CVI (M2). Pyrosequencing results show the ratio of each virus present in
DNA isolated from the tissue. IHC results are the percentage of positive-staining
follicles for one or both viruses. Samples that failed pyrosequencing are indicated by an
F.

 

l

 

 

 

 

 

 

 

 

 

 

Pyrosequencing IHC
:0 E” (Frequency 0f (% follicles staining
g E E t g each virus) for each virus)
a g E a .2 2 M1 M2
9 w ‘0 '° “9 9' %Ml:%M2 only Mixed only
Controls (averaged per group)
None 6 Avg (6 birds) Bursa F 0% 0% 0%
M1 6 Avg (6 birds) Bursa 97:3 25% 0% 0%
M2 6 Avg (4 birds) Bursa 0: 100 0% 0% 12%
Dual-challenged (individual birds)
M1,M2 6 M6911 Bursa 83:172 11% 1% 0%
M1, M2 6 M6909 Bursa F 0% 0% 0%
M1, M2 6 M6912 Bursa F 0% 0% 0%
M1, M2 6 M6901 Bursa 76:24 7% 1% 1%
M1, M2 6 M6913 Bursa 97:3 40% 0% 0%
M1, M2 6 M6905 Bursa 21 :79 0% 3% 3%
Controls (averaged per group)
None 21 Avg (8 birds) FFE F 0% 0% 0%
M1 21 Avg (6 birds) FFE 97:3 67% 0% 0%
M2 21 Avg (8 birds) FFE 0:100 0% 0% 98%
Dual-challenged (individual birds)
M1, M2 21 M6908 FFE 91 :9 8% 92% 0%
M1, M2 21 M6902 FFE 84:16 0% 100% 0%
M1, M2 21 M6903 FFE 98:2 No sample
M1, M2 21 M6907 FFE 88:12 25% 75% 0%
M1, M2 21 M6920 FFE 82:18 15% 85% 0%
M1, M2 21 M6915 FFE 88:12 0% 100% 0%
M1, M2 21 M6910 FFE 98:2 100% 0% 0%

 

1 Number of bursa and feather follicles analyzed per sample averaged 286 (range 138-
433) and 9 (range 2-18), respectively.

2 Frequency of er5 was significantly greater than er5//38CVI in bursa and FFE
samples from dual-challenged birds by paired t-test (p=0.0006)

100

 

8822228822 22222223 22882282: 88 22282 22080 .8 80282588 82223 8 282282082 822.23 828% 8.22:8 .8 88825:: 822.2. 8

@022 882802 .9222 832 .289 288280 .2823 8225 6.22.22 2882.2 A822 38: 28w82:m ”82822282852228 883.2. N
28228232822225 22882888888222 N2 mo 282828280 8282:2288 nomm 2822222082 83.22:, 8.2 288: 883 8:82am 2

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

82 Ho 8822 8N 8 82220 88 N32 .2
o N 820 88 N32 .2.
82 8 82220 88 N32 .2
882 RumN 2.888 8 8 2.2220 88 N32 .:
292.. 2222 88 222 o N820 88 N32 .2.
82 Ho 8 8 8820 88 N32 .:
2 8N”: 8820 88 N32 .8
2 N 8 22820 88 N32 .:
N88 2888 2 8820 88 N32 .8 8
2 28 88832 88 N32 .:
82 Ho 2.8832 88 N32 .2. N
22:28 222.22.322.25 22882282222228-2822:
2:828 2:828 288828 622 228 88 N32
22.28882 €882 228282 2228 2:8 88 :
282 ”2 2822 o 2822 o 88 6252 8
282 8828 88 N32
222 882 22 28 22.8 2.88 88 . 8
22 .2 88 682 N
m 22225.2» .222 28928222 828.22an
252 5 280 2228 2:2 N282 N32 2. 222:2: 82. 82.2228 22228.2
N288” 2.2x. 28223 822 2 >580 922222823 emu—8:220
88:23 :2 822.23 2.208 .8 30282228282 82222822282 w><
wfionoaaomeur 2828—82 822.22.?

 

 

 

 

 

 

 

.82 :8 .882 28280222822 88 82028222208823 28228.2 2282.22 8022283 83.22:, 22028 8.2 8228 82: >58 222828222 822.22.»
$28838 0222 8228 8222.288 822282082 22.25 8228822 8222 228222 2882282082 <75 :2 2:88:22 82223 2228 me 0228 8222 32.28 8222282 82082228822882
88282888 2802828 883 2202223 8282280 32283: 22288 $808382»: 8.2 2888: 083 88:82. .2822 2>me\\m2>22 88:8 5 >982 228
22223 88820320 28w822282292838 28:8 o2wq2m 2282.2 8222288 wfiocozwomoba 28:28 82228282 25> .<m 8% <N 8288288282m "vim 0222.;

101

8880588888888 £583 @880me 088 099 some .8 8282888 8:23 8 8858888 8:23 98.8w 8288 .8 85882:: 2: 8

GO: 288ch Aid 8034 A859 8880 damv 588m .925 8.88: A8086 who: 8838885 ”8888282862858 883. N
.uhéxwofifia 8882280888888 N; 8 88882888 288888 comm 48888888 88:; 28 wow: 2883 8:82am _

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

8882 82 88 88880 88 82 .22
8888 8888 8 8 8885 88 82 .22
882 882 8 88 888.0 88 82 .22
2 882 8 S 8885 88 82 .E
8882 882 :88 8 88 8885 88 82 .:z
2 $85 88 .82 .:z
8882 882 8 8 385 88 82 .22
888 8882 8 8 8885 88 82 £2 8
888 88888 88 82.22
2888 88888 88 82 .22
882 88882 88 82 ,2:
8882 88882 88 82 .22 8
3:: 38823888 8885:2288:
38878 38828 8888828 E8 58 88 82
8: 882 :8 8882 888 :88 888 _ 888 8 88 22
888 2 88: 8 a: 8 88 252 8
888 882 ”8 88 82
:8 :88 88 22
888 2 88 8882 8
m 388% .88— 83823 88.8885
.5. >5 888. .28 :82 88872 82 82 88:88 888 852.8 ea
N288” 8828\8 _8=oo 898 >580 mam—98:5 $5530
823 some .8 8885:3282 828872 w><
wfiuuoacomcuzm 883.88% 8:888»

 

 

 

 

 

m :8 3 8888885 828 82058838082888 80:88 885 $8.5m .8523 some 8.8 8:80 82 >58 88:88—88 8:83

$828828 85 0.8 8:388 coca—0% 88:5 .0388: 85 Soc 8888882 <75 5 888888 8323 some 8 0:82 2: 30:8 828802 3658088882
8288888 52% 0.53 88033 88288 gunmen 8888 .wEocoavomoSm 88 com: 8203 88589 ANS: 50w 3822 8888.8 23:
88822 523 888—25 woman—3:048:88 98 Emma 80.8.8 8:588 8285588088888 288 8288888 8:2; .mm 8% mm mucofitomxm "Wm «Bah

102

 

Gwvfioufi 888-8 .3 8:382 38888828: 8.28 88888? 8.83888. 8.820.888 88088838 02 .Cooodvmv 888-8 88022.88.
.3 8828 38028888488 :82”. 828.888 mm... 88 88285 E 8585):. 8.85 28.89% >2§o¢8w88 8? 5083.822 .8 88.88332... 8
.80.. 888.5. .93. 88>... .8808 88880 .8888 888.88 .88... .8888. H88.88.5588 88858 8
owm 8 8.88.8 8 88 58288 omen—.28 8.888 .8 :8 com .3 830:8 38 25$ :38: 88 3828888888 883 H>E .

 

 

 

 

 

 

 

 

 

8888 888. 88.8 88.8 88888. 5.2 82 .8
88. 8 88.8 88.8 .8888. 822 8.). .8
88.8 88. H8 88. U8 88888. .58. 82 .8
2 88.8 88.8 88.8 88888. 82. 8.... .8
88.8 88.8 88.88 88888. 888.2 ms. .8
88. .8 88.8 88.8 88888. 5.2 82 .8
88.8 88.8 88.8 88. ”8 88888. 2.2. 88,. .8
88.8 88.8 88.8 88.8 88.8 88.8 88888. 2.22 8.2.8
88. H8 88.8 2 88888. .58. 8:. .8
88. H8 2 88.8 88.8 88888. 58. 82 .8
88.8 88. ”8 88.8 88.8 88888. 88,. .8
88. H8 888 88.8 88888. 82 .8
88.8. 88. 8 88.8 88.8 .8888. 8.). .8
888. 88.8 88.8 88888. 8.). .8
88.8 88.8 88.8 88.8 88.8 88.8 88.8 88888. 82.8
88. ”8 88.8 88.8 8888.2 8.). .8
88. H8 88.8 88.8 88888. 8.). .8
88.8 88.8. 88.8 88.8 88.8 88888. 82.8
88.8 88. ”8 888.8 88888. 88,. .8
.5. _ >5 _ 8888 a .88 _ :8. 88 888. _ 88 888. H 8. 888. 8.88.8. 18.88588.» 85888888535
88885;. mm“—
~2€8n888
88:23 .2888 8 8:88.383 name—88:389.?—

 

 

.m 88 .3 888808.: 088 wcmocoscomoba 8828.8 88.: 828.8% .8388: 05 .8828 888888. <75 5 80808

8:23 .2080 .8 2882 85 30:8 88880.. wfiocosvomobnm .ANEV 8>me<m822 .858 CV 38. \28. £23 388828888888 Ema—88:0 8:08.028
882.858.8288. .888 828.888.; 5 8298888 883 8888 mum 2858:. .8888 828.802 wfiocoscumobhm .<v 80:82.8me 8-8” 288,—.

103

.Amoc _ dun: 888-8 .3 32:38 83589:: can 32282883 8802509 ooaobmfiw 88888.22:me
0 Z .Cooodvmv 38.8 3823 .3 £8.85 wowco:m:o-_m=w 88808.8. 829:3 man 5 :2 SW8: 883% 358$:me 33 NE 90 .8888on 8
so: 82.858 .958 55 .309 8.280 .88ng 82% .EE :8: ”22355.88 5&8; 8
own we @328 o 8a 58% 095220 :03 20 PE com .3 3.50:0,“ 25mm ooomv 822m: :8 BBEEER m8; H>E _

 

 

 

 

 

 

cox: oofio oofio vomwd HE N2 .32

mcfim $an aha Emma H>E 92 .22

oofio 0070 0070 mama {—52 Q2 .:2

am“: 36 Nouw 8.3mm FE NE .22

noun oofio 0025 U2 09mm .25: N2 .32

oofio oofio oofio 3232 g: 32.32

oofio 0010 “2 wand .25: S2 .22

mm”? 00.3. 382m «Ho—mm F>I Q2 .32

oofio oofio oofio 32mm .25: $2.22

91% m mnKN Eumm owfim NE .32

$82 8;. oofio 0010 2:3— 32 .32

mnnmm $6M mm”: mm”? 3:?— m2 .:2

oofio 0070 0070 m Nwfim $2.32

Sun 00—5 oofio oofio oofio wind NE .22
mqu: owuom momma 8.23. .22 .22

xnumm 08.82.. 08.83. 02mm 92 .32

vonc x92 :86— maM N2 .32

oofio mm”: mm”: menf «:2 S2 .32

EM _ >5 _ 850 _ Em _ 8.8: men _ on .35 _ m— zan— ucbm 8.35.8583» 2:95.». mew-3:830
N32888:. Hr:
32.x.” 32.x.
GE? some 980 hogsvofiv wanna—.2883:

 

 

 

 

 

m S... 83 vofiflwfi 08a $653883 3:882 8885 82988.8 .253 on... Sch @8282 <ZQ 5 8.83882
95> some mo 288 05 32?. $338 wamocoavomobf .366 50382328 98 :3: 2.228 523 Emsoocazzfii wom=o=ano many—0.30
BKEooga: was 38m58m> 5 8:253 883 80893 mum 2.538 :82 $388 mfiocoswomoim .mv 8552.80.58.82 “Wm 033,—.

104

 

2000.0V
2000.0v
womvd

00 200

2:88.»-

2000.0v
_000.0V
womfo
0020.0

 

 

 

fiww
Ax.0w

Rh
$2

$80
$00
.wim

 

$3
$8
$3
€on

$0
.XL
$.va

   

£0859

mum
€083.

mum 8% 35m

 

 

2.3:.

£9883.

mum
20:885.

mum 8% $885

 

can;

 

 

 

888885800080 28883 I $8886 wcmocosaomobhm ”04‘. 2:8;

105

 

50288088

£20255 38828888 m8. 058 08m 505 028.88.». 203 Em: 05 02 05 :o 828% 2:. .SDw $2828 058 38522 82.5
woman—230 $385382... 888—020 a Sch Ev 8N £8008 92.2 3.80m 00 was... Emmqfisoofifiogg .8 «Fountain 87m 953%

106

107

 

.3383 Son .88 02836 8386
3030882823 058 .3880 3.83 50388328 .88 02883 86820 SE? .388 3.83 9:28 80 02883 880 mug—2:8 088858885 82986 xoflm
5258088 £06335... Eco—cocoa m8. 088.» 0:2 52.5 02885 803 Emu 05 £2 05 no 2803on 2:. .~>me\\m228 23 2:28 523
vows—230 338883—888? 888—050 a Set Ev 0 28:08 3.885 3.8% no mEESm Efizsgoofifiocdfics .2 8:288:22an "unm 95mm”—

 

N_>_+_.S_

8

N20
:2n
1.

N
\\
.8
88
.8
8.
, 88
, ,\\
88

 

«2+0

25> 00982.25
8:5 ms.

2:0 :2

k\\\\\1\\\\V\\\\\\\‘

 

k\\\\\\\\\\\\\\\\\

8.8 8.

use :sou/NoA 95min!

 

 

$30.81: <>oz< 8.8 853..

0.803 350% 880853 ggbbmu 880830 02 .00 owaq 880 000608 05 8 000.580 .3 80:80 3 090808 0.83 838% 0N

mo 03.8. 8088300 0.308888 280 33808 an 808m .2920 50382328 .80 .2820 2.2.8 .5 38528. 583 088,88 3286 .80 >380
00.882 80.0 80 0 «a 00383 880an «885 80.0 02208 <ZQ 80.0 $888 MUm $388.30 09803.. .8 80888098...“ 6.0 0.5»?—

108

i N§.+F§

N2 +0

m:.:> 8000:2350

2:0 N82 2:0 :2 2:0 8,

 

 

 

 

00008

0000N

 

. 808

 

. 0000v

00000

”as Isoq/NQA afieJaAv

 

00000 4

 

 

11%-.;,. ..: .. 00005

 

00000

.aooodne <>oz< 8.8 885588880

888088.888 888808 808808 880.8880 583 8880080 .00 0mg 80 880088808 88.. 8 80088888080 8 808880 8 800.8088 803 8888 mm
.80 0888.8. 808888.508. 8888808 0880 88808808 888 808m .ANEV 50033828 80 A8820 20828 .5 3N0 88828 88883 800800.888 88888380 80 >888
80.8888 808.8 8880 8m 888 0088888 808808 mum 808.8 800888888 <75 808,8 $888 #8088 038888888880 ow80>< .8 8808Co88xm "v-0 PEME

109

 

 

 

 

 

 

 

N5— + :3 N2 + .2
0

coo —.
88 w
u
6
NE D M
:2 8 Doom m
a I w
o
.e...
88 m

000m

0000

 

 

 

008.93 <>oz< E 8085.00 .805.“me 3865 95:2 “same as, 3:80 .3 Oman 8 $3.3

09 a 30.880 3 32:8 8 3682 203 sass 2: mo 2 dofisuu Bags; 25 28:05 as sum .50 50% 05% Ba
:20 82“ 8 £20 :6” $83: Ea S ESE: has as, 8.83 326 2a E as, Engage. s Eases, «ES 88
Eu 8 2a ,8 .S a Baa: 288m mum 83 3280 <20 a3 5:52 «00 2:303 owns: .0“ “fiauofim "Wm 25E

110

Chapter 4
THE EFFECT OF THE TIME INTERVAL BETWEEN EXPOSURES ON THE
SUSCEPTIBILITY OF CHICKENS TO SUPERINFECTION WITH MAREK’S
DISEASE VIRUS STRAINS
Abstract
Repeated exposure to Marek’s disease virus (MDV) is likely in commercial

chickens because vaccination does not prevent infection or transmission of infectious
virus. Superinfection of multiple strains may lead to interaction of viruses within a host
and encourage the evolution of more virulent strains within the flock. Four experiments
were performed to evaluate the effect of time between challenges on the effect of
superinfection using two pairs of fillly virulent MDV strains: 1) J M/ 102W and
er5//38CVI , and 2) erS and er5//38CVI. Feather follicle epithelium, spleen, and
tumor samples were collected at single or multiple time points from the same bird to
determine the frequency and distribution of each virus present following superinfection
using pyrosequencing, immunohistochemistry, and virus isolation. The effect of time
interval was strong. Superinfection was observed in 82 of 149 (55%) FFE samples
following short interval challenge (24 hours) compared to only 6 of 121 (5%) samples
following long interval challenge (1 3 days). Dominance was detected between the first
virus pair following a short interval challenge, but was unapparent following the longer
challenge interval. Tumor samples generally consisted of a single predominant virus,
but both viruses were detected within a small portion of tumors. Vaccination with HVT
had no significant effect on the virus frequency for either virus pair or challenge time
interval, suggesting these results may be significant in field conditions. In cases where

birds are exposed within shorter periods of time or where the initial exposure is weak or

111

delayed, mixed infections are likely to occur regardless of vaccination status or
inoculation order. These studies demonstrate that mixed infections under some
conditions can lead to shedding of the second virus a prerequisite for the establishment

of the second virus strain in the population.

Introduction

Marek’s disease virus (MDV) is an alphaherpesvirus that causes
lymphoproliferative lesions in chickens. One of the greatest concerns facing the poultry
industry is the evolutionary trend of MDV towards greater virulence (95). The
evolution of Marek’s disease (MD) was recently reviewed (65). The first wave of virus
evolution from mild (mMDV) to virulent (vMDV) strains was during the 1950’s and
was likely due to the transformation of the industry to highly intensive poultry practices.
Subsequent evolution has been attributed to the introduction of successive generations
of MD vaccines (92). The more recent evolution is likely due to the fact that MD
vaccines fail to produce a sterilizing immunity, allowing viruses to replicate and shed
within vaccinated hosts. The early and later evolution waves, however, both rely on the
continuous availability of large populations of susceptible birds.

Transmission of MDV occurs through the production of fiilly infectious virions
in the feather follicle epithelium (F FE) that are spread in the feather dust as early as 2
weeks after inoculation or exposure (18). Shedding appears to peak at approximately 3-
5 weeks and continues indefinitely at quantities as high as 109 virus copies per

chicken per day, irrespective of vaccination status (53, 94, 106). Infected feather dust

112

can remain infectious in a poultry house for at least several months at 20-25°C and for
years at 4°C (17, 58). The continuous transmission, long survivability of the virus, and
large population of birds favor a significant viral load within a poultry house.

Previous studies have confirmed that multiple strains of MDV can circulate
simultaneously within the same flock. One study propagated 25 isolates from five
poultry flocks and found that four of the five flocks had strains present from more than
one level of pathogenicity (apathogenic, classical, or acute) (10). Two flocks, including
a single house from one flock, had strains present from all three levels of pathogenicity.
Another study examined isolates from a flock at different time points and found that
Marek’s disease (MD) mortality was closely associated with the sequence of infection
and frequency of isolation of viruses of differing virulence (56). For example, three
pens that experienced high mortality had early isolates that were of the acute type,
whereas two different pens with low incidence of MD had apathogenic strains that
predominated throughout early and late isolations.

The studies described above provide evidence for multiple strains of MDV
circulating with a flock, but don’t address the possibility of multiple strains within the
same bird. Superinfection is defined as an additional infection occurring during the
course of an existing infection. In this study, the term is meant to describe infection
with one MDV strain followed by a later infection with a different MDV strain.
Establishment of a second strain not only implies infection, but requires the necessary
fitness traits for that strain to compete with the first virus. After the strain infects one
cell, it must be able to replicate and spread to other cells. Conditions that may facilitate

superinfection include internal sources of new mutant strains from within a bird or

113

external sources of new strains brought in from other flocks. Superinfection with two
virulent Marek’s disease virus (MDV) strains is likely an important step in the pathway
towards establishment of evolved strains within a population. Chapter 3 demonstrated
that coinfection was possible when viruses were administered simultaneously and
dominance was observed between viruses of different pathotypes. In field conditions, a
time interval between exposures is more likely than simultaneous exposure.
Laboratory-based evidence that superinfection can occur with two fully virulent
Sl MDV strains is limited. Ianconescu et al. attempted to evaluate the effect of early
natural exposure on subsequent challenge with virulent J M MDV (48). Nonvaccinated
farm-reared chickens were removed from the farm and challenged with J M at 5, 7 or 12
weeks of age. Uninoculated groups developed MD tumors and antibodies after
isolation, confirming natural exposure on the farm. At 12 weeks, farm-reared chickens
inoculated with J M had no difference in tumor incidence compared to unchallenged
birds. Thus, there was no clear evidence that superinfection occurred or had an effect.
A more recent study evaluated the effect of superinfection in adult birds using known
virus challenge for early and late exposure (99). Virus exposed chickens were
vaccinated with HVT at hatch plus challenged with JM/102W at 5 weeks. When virus
exposed chickens were challenged with a highly virulent strain as adults at 18 weeks,
they generally failed to develop transient paralysis or tumors. Not only were the virus
exposed groups refractory to lesion indication and transient paralysis, but the viremia
responses and transmission to contacts appeared minimal at best, indicating that these
birds had a much reduced susceptibility to infection. The authors concluded that late

outbreaks in the field are not likely due to superinfection, but are more likely triggered

114

by unknown environmental factors. Thus, with a long time interval between exposures,
both studies failed to detect any significant superinfection. Reports from the field have
suggested that superinfection of virulent MDV strains is possible, based on apparent
adult-to-adult transmission of vv+ MDV with very late outbreaks occurring, even after
molt (59). Since most birds in a commercial flock are presumably infected with MDV
early after placement, it would seem that superinfection must occur for any new strain
to gain access and become established within a bird and flock.

Superinfection has been studied using other herpesvirus models. A recent study
evaluated feline herpesvirus type 1 (F eHV-l) superinfection in cats latently infected
with wild—type FeHV-l (77). Nine SPF cats were inoculated with wild-type FeHV-l
and superinfected 8 months later with a recombinant F eHV-l virus with green
fluorescent protein substituted for glycoprotein G (rFeHV-lAgG/GFP). The
recombinant virus produced a temporary local lymphadenopathy, but otherwise
produced no clinical or ocular signs of infection. Cyclophosphamide—dexamethasone
treatment was administered 16 months after superinfection to stimulate reactivation,
which was successful for the underlying theHV-l infection but no reactivation of the
superinfecting recombinant virus was detected. The study unfortunately lacked a
control group of cats infected only with rFeHV-l AgG/GFP to confirm both strains were
fully pathogenic.

This study had two hypotheses. First, following short (24 hours) and long (13
days) interval challenge, the first virus will be present in greater frequency in bursa,
feather follicles, and tumors relative to the second inoculated virus. Second,

vaccination will mute the effects of both viruses leading to equal mixtures in bursa,

115

feather follicles and tumors following short and long interval challenges. The short
interval was potentially more reflective of rapid exposure to more than one strain in
field conditions, whereas the long interval was considered sufficient time to illustrate
the effect of delayed challenge. The experiments in this chapter also were designed to
determine whether the virus dominance observed in chapter 3 was still evident when
one virus was given a head start. The final experiment in this chapter included maternal
antibody positive birds vaccinated with HVT to determine the relevance of these events
to field conditions.

Superinfection of multiple virulent MDV strains seems necessary to allow the
establishment of evolved MDV strains within a flock. Understanding what events are
necessary for superinfection to occur will increase our understanding of the conditions
necessary to allow viruses to interact within a host and evolve to greater virulence or

prevalence in a population.

Materials and Methods

Chickens

The experiments used 1515x71 white leghom chickens, a F1 hybrid cross of MD

susceptible 1515 males and 71 females (2). Maternal antibody negative (Ab-) chickens

were used in experiments 1 and 2, which were reared from an SPF breeding flock
housed in isolators that have received no MD vaccinations or exposure. The flock was
negative for MDV antibodies by routine surveillance tests. Maternal antibody positive
(Ab+) chickens were used in experiments 3 and 4, reared from breeder hens vaccinated

at hatch with 2000 plaque forming units (pfiJ) of HVT and at 25 weeks with 2000 pfu of

116

$31 and Md] 1/ 75C viruses for exposure to all three serotypes. Both flocks were also
negative for exogenous avian leukosis virus and reticuloendotheliosis virus from routine
surveillance testing. All birds were maintained in negative pressure Horsfall-Bauer

isolators. Experiments were approved by the ADOL Animal Care and Use Committee.

Viruses

Three viruses were used to create two virus pairs: 1) J M/ 102W and
er5//38CVI; 2) er5 and er5//38CVI. JM/102W is a clone originating from the
parental strain JM, and has been pathotyped as a vMDV (84, 93, 95). The two
recombinant viruses originate from the wild type Md5 strain, a vaDV strain (93, 103).
er5 was produced from the wild type strain using overlapping cosmid clones (75).
J M/102W and erS were both paired against er5//38CVI, produced using the erS
cosmid clones with substitution of the pp38 gene from CVI988/Rispens (60). The
ability to differentiate—the pp38 gene of CVI988/Rispens from other SI MDV strains,
allowed differentiation of these two virus pairs within a mixture. The first virus pair
was chosen based on differences in pathotype. The second virus pair was chosen based
on their similar pathogenicity, as reported by Gimeno et al., in which there were no
differences in the frequency of lymphoproliferative lesions between the viruses at either
6 or 15 wpi (43).

Experiment 4 compared superinfection results between vaccinated and
unvaccinated chickens. For this experiment, chicks were vaccinated at hatch with 2000

pfu of S3 HVT (strain FC126/2) vaccine (101, 102).

117

Monoclonal antibodies
The MDV pp38 gene, located in the IRL and UL region, was utilized for virus

differentiation by immunohistochemistry (IHC) and pyrosequencing. According to
early published pp38 sequence data, all serotype 1 MDV strains have identical bases in
the pp38 gene except bases #320 and #326 (31, 38). Cui et al. determined that
monoclonal antibody (Mab) H19 was specific to base pair #320, and was only positive
when adenine (A) was present, whereas Mab T65 was specific to base pair #326, and
was only positive when guanine (G) was present (34). JM/102W and er5 used in this
study have A at both base pair #320 and #326 and thus were positive for H19 only.
er5//38CVI has G at both base pairs #320 and #326 and thus was positive for T65

only.

Immunohistochemistry testing

Immunohistochemistry (IHC) testing was performed using previously described
methods (see Chapter 3). Briefly, feathered skin sections were cut at -25°C, dried
overnight, and fixed in acetone for 45 minutes before storage. Peroxidase staining for

pp38 antigen was performed using the Vectastain ABC kit, as described by the
manufacturer.'7 A DakoCytomation Autostainer was used for staining, where slides
were rinsed (all rinses were done with PBS) and then incubated for 20 minutes with 150

uL of normal blocking sera (horse).l8 Slides were rinsed once and then incubated for

30 minutes with either 150 uL of H19 (1 :3200) or T65 (132000) monoclonal antibody.

 

17 Vector Laboratories, Inc., Burlingame, CA
‘8 Dako, Glostrup, Denmark

118

 

Slides were rinsed three times and then incubated for 30 minutes with 150 uL of the
biotinylated secondary antibody. Slides were rinsed three times followed by a 30
minute incubation with 150 uL of the avidin and biotinylated horseradish peroxidase
(ABC complex). Slides were then rinsed three times and stained for 7 minutes with 150
uL of the DAB solution including the buffer and hydrogen peroxide which reacts with
the horseradish peroxidase and produces a dark brown color. Two rinses followed and

then slides were counterstained with hematoxylin and coverslipped.

Virus isolation

Viral plaque assays were performed using spleen cell suspensions plated on
duck embryo fibroblasts (DEF), as previously described (See Chapter 3). Briefly,
whole spleens were collected at termination and mashed using a syringe handle to
produce a single cell suspension. Approximately 100 uL of the cell suspension, in
Leibovitz McCoy (LM) media, was added to 1 mL freezing media (55% - 4% LM
media, 25% - calf serum, 20% - Dimethyl sulfoxide [DMSO]) and the samples were
stored at -80°C until later plated on DEF.

Cell suspensions were thawed 2-6 weeks later, counted and plated on 2-4
secondary DEF cultures. After 7 days, cultures were fixed with an alcohol-acetone
mixture (6 parts acetone and 4 parts 95% alcohol) and stored at -20°C until they were
stained and read. Fixed cell culture dishes were rehydrated and incubated for 30
minutes with either H19 or T65 monoclonal antibodies. Plates were washed with three
rinses of distilled water followed by three rinses with PBS and then incubated for 30

minutes with a fluorescein-conjugated goat anti-mouse IgG (H + L) secondary

119

:-—-—=|

antibody”), purified by gel filtration. Plates were washed again and then plaques were

counted using a fluorescent microscope.

Pyrosequencing assay

FFE and tumor samples were analyzed by pyrosequencing to determine the
percentage of each nucleotide present in an A/G SNP at base pair #320 of the MDV
pp38 gene. This SNP was used to distinguish er5//38CVI (dGTP) from J M/102W

and erS MDV strains (dATP). DNA samples were purified using the Gentra

Puregene Genomic DNA Purification Kit.20 The assay was performed using a PSQ

96MA system, according to the manufacturer’s protocol, as described previously (See

Chapter 3).” Briefly, biotinylated PCR product specific to the MDV pp38 gene was
generated using a standard PCR with a biotinylated reverse primer (forward primer 5’-
GTG ATG GGA AGG CGA TAG AA-3’, reverse primer 5’ Biotin-CAT CTG CTT
CAT ACC ATC TC-3’). In experiments 3 and 4, DNA degradation led to low yield of
PCR product and failing pyrosequencing results. A nested PCR was performed on these
samples first using an additional set of outside primers (forward primer 5'-TCT CCG
CCC CCT CAC TCT-3', reverse primer 5’-TCC GCA TAT G'IT CCT CCT TC—3’).
The PCR product was then captured on streptavidin-coated beads by shaking binding
buffer, streptavidin sepharose beads, and PCR product at 1400 rpm for 10 minutes.
Beads were transferred to 70% ethanol, denaturation solution, and wash buffer to

denature and remove the non-biotinylated DNA strand. The beads were then

 

19 MP Biomedicals, LLC, Solon, OH
20 Qiagen, Hilden, Germany
21

Qiagen, Hilden, Germany

120

 

transferred to a new 96-well plate containing annealing buffer and the sequencing

primer (5’-GGA GAG TGG CTG TCA-3’), designed using SNP Primer Design from

Pyrosequencing AB version 1.0.1.22 Nucleotides were added one at a time and
incorporation was measured and quantified by light generated by successful elongation.
The relative percentage of each nucleotide incorporated at the location of the SNP was
reported and interpreted as the frequency of each virus within the DNA sample.
Pyrosequencing results were analyzed from all samples with passing results, meaning
from samples in which sequence was detected. Samples that failed either did not have
adequate virus present (e. g. negative control samples) or there was either insufficient or

excessive total DNA present.

Quantitative PCR assay

Quantitative PCR (qPCR) using a multiplex reaction was performed on an ABI

7500 Real-Time PCR System, as described previously (see Chapter 3)”. Briefly, each
well contained the purified DNA sample, primers and F AM-TAM probe specific for
the MDV pp38 gene (forward primer 5’-GTG ATG GGA AGG CGA TAG AA-3’,
reverse primer 5’-TCG TCA AGA TGT TCA TTC CCT G-3’, TaqMan probe S’FAM-
TACCGCCTGAGCCCCGGAGG-3’TAMRA) and primers and VIC-TAM probe
specific for the cellular gene GAPDH (forward primer 5'-ACA GAG GTG CTG CCC
AGA A-3', reverse primer 5'-ACT TTC CCC ACA GCC TTA GCA-3', TaqMan probe
5' VIC-TCA TCC CAG CGT CCA CT-3' TAMRA). The following conditions were

used: 1.) 50°C — 2 minutes; 2) 95°C — 10 minutes; 3) 95°C — 15 seconds; 4) 60°C — 1

 

22 Applied Biosystems Inc., Foster City, CA
23 Applied Biosystems Inc., Foster City, CA

121

 

minute with steps 3-4 repeated 40 times. A standard curve with lO-fold serial dilutions
of plasmids containing MDV pp38 or GAPDH was used to calculate absolute copy
numbers of each gene per sample. For the calculation of averages, outliers were

removed if the value was greater than 1.5 times the interquartile range (the difference
between the 25th and 75th percentile). On average, one value was removed per group

(range 0-3). Results were expressed as the viral copy number per host cell.

Experimental design

 

A total of four experiments were designed to analyze superinfection following
short and long time intervals. Each experiment consisted of two parts (A/B); Part A
used virus pair 1 (JM/102W and er5//38CVI) and part B used virus pair 2 (erS and

er5//38CVI).

Experiments 1 & 2

Experiment I examined superinfection in F FE using Ab- chickens. Experiment
1A consisted of 19 groups of 10 chickens, examining the frequency of J M/102W and
er5//38CVI in feathered skin sections (Table 4-1). Viruses were administered alone
or together following short (1 day) or long (13 day) intervals in forward and reverse
order. Samples were collected 28 days post infection of the second virus challenge.
Skin sections were frozen for IHC and DNA isolated from the outside of the feather
shafts for pyrosequencing. Contact chickens were also included as an additional
method to determine which viruses were being shed in mixed infections. Three “donor”

birds were transferred from each of the four lots of dually infected chickens and

122

transferred to the contact “recipient” isolator of newly hatched chickens. The donor
birds remained in the contact isolators for 1 week and were then removed for sampling
along with the rest of the main group of birds. Thus, donor birds were transferred at
approximately 21 days after the second challenge during the time of peak shedding and
were removed at 28 days post infection of the second challenge. Contact recipient birds
were sampled at 28 days post exposure. Experiment 1B was identical to 1A, except
using virus pair 2 (er5 and er5//38CVI).

Experiment 2 examined superinfection in tumors using Ab— chickens.
Experiment 2 was the same basic design as experiment 1, except this experiment was
designed to determine the frequency of each virus in tumors and latently infected
lymphocytes (Table 4-1). Samples consisted of DNA isolated fiom tumors and
enlarged nerves for pyrosequencing and spleen suspensions for measuring viremia.
Samples were collected 56 days post inoculation of the second virus challenge.
Seventeen birds were used in challenged groups instead of 10 birds because it was
expected that at least half of the infected chickens would not produce visceral tumors.
Tumors and enlarged nerves were collected from all surviving birds at the time of
termination with visceral tumors. Multiple tumors were collected when present. If less
than 10 birds per group had visceral tumors, nerves were sampled from birds with nerve
enlargement randomly in order of low to highest wing band number until at least 10
birds were sampled from each group. Some groups did not have enough surviving birds
to meet this requirement. Spleen was sampled from uninfected control chickens.
Contact chickens were hatched and exposed in the identical procedure as experiment 1,

except donor birds were returned to their original isolator following one week of

123

housing with recipient birds since sampling of the main group of birds was not until 4
weeks later. Contact recipient birds were sampled at approximately 56 days post
exposure. Experiment 2A was terminated early at 5.5 weeks (short and long intervals)
due to higher than expected mortality in groups challenged with er5//38CVI.
Experiment 2B was also terminated early due to high mortality with the short interval

and long interval groups terminated at 4.5 and 5.5 weeks, respectively.

Experiments 3 & 4

Experiment 3 examined superinfection over three time points in Ab+ chickens,
comparing multiple FFE and tumor samples collected from individual birds (Table 4-
2). Multiple sampling from the same bird allowed changes to be analyzed over time in
virus frequency within individual birds. Viruses were again administered alone or
together following short (1 day) or long (13 day) intervals in forward and reverse order.
Feathers were sampled at three time points (day 14, 27, and 70) and tumors were
collected on day 70 at termination. DNA isolated from the outside of feather shafts was
analyzed by pyrosequencing and qPCR, and tumors were analyzed by pyrosequencing.
The first sampling on day 14 was used to confirm the presence of the first inoculated
virus before inoculation of the second virus. Seventeen birds were used in singly
infected control groups, but dually infected chicken groups were increased to 34 to
further increase the number of visceral tumors, since five of eight dual-challenged
groups from experiment 2 had 3 tumors or less. Multiple tumors were sampled when
present and sampling only consisted of visceral tumors. Contact birds were used again

to provide further evidence of the frequency of virus shedding in mixed infections, but

124

 

the method of exposure was altered. Instead of physically transferring three donor
birds, the air outtake tubing from the isolator of inoculated donor birds was re-routed to
the air intake from the isolator of the uninfected recipient chicks at 21 days post
infection of the second virus in donor birds. Contacts were only exposed to isolators of
birds challenged with the long interval in this experiment, and were only sampled for
feathers and tumors at one time point (day 75). Contact chicks were three weeks old at
initial exposure, compared to 1 day old in experiments 1 and 2, due to availability of
chickens.

Experiment 4 used vaccinated Ab+ chickens to determine the relevance of
effects in conditions more closely similar to field conditions. Experiment 4 had similar
inoculation and sampling schedule as experiment 3, except due to vaccination at hatch,
all virus inoculation and sampling dates were delayed 5 days compared to experiment 3.
Since experiment 4 was used to compare results from vaccinated and unvaccinated

dual-challenged chickens, there were no single-challenged control groups.

Data analysis

The pyrosequencing procedure resulted occasionally in some background noise
that was reflected in virus frequencies from single-challenged control birds that were
less than 100%. Therefore, results from single-challenged control birds were used as
standards for interpretation of results from dual-challenged birds. To determine if one
virus was exclusively predominant in a sample from dual-challenged birds, cut-off
values were assigned by age-matched single-challenged control birds based on the

lowest frequency value at each sampling day for each experiment. For example, on day

125

 

14 of experiment 3A, the frequency of dATP from chickens single-challenged with
J M/lOZW ranged from 95-100% and samples single-challenged with er5//38CVI
were all 100%. Samples from dual-challenged chickens were then considered
predominant for JM/102W or er5//38CVI if the frequency of dATP was >95% or
dGTP was 100%, respectively. Samples not predominant for either virus were
considered mixtures.

Chi-square analyses were used to analyze 1) the association between virus
challenge interval and predominance of the first inoculated virus in FFE samples in
experiments 1-4; 2) The association between vaccination status and virus predominance
in F F E samples in experiment 4; and 3) the association between virus predominance in
multiple tumors from individual birds in experiments 3 & 4. The Fisher’s exact test was
used when expected values were less than 5 in any group. Significant results were
based on p-value <0.05.

Virus isolation results from experiment 2 were analyzed using the Wilcoxon
rank-sum test due to low numbers of samples in dual-challenged chickens. P-values
were reported based on two-tailed tests comparing plaque counts for each virus in dual-
challenged chickens with single-challenged chickens challenged with the same virus on
the same day. One-tailed tests were used to analyze the reduction of plaque numbers

for the second virus inoculated following long interval challenge.

Results

Feather follicle epithelium

126

Data from uninfected and single-challenged control birds indicated that both
IHC and pyrosequencing assays were highly specific in FF E samples from all
experiments. In experiment 1, the presence of both virus strains was common for both
virus pairs following short interval challenge (Table 4-3). In experiment 1A, mixtures
were common in individual follicles following challenge with JM/102W and
er5//38CVI, with relatively equal distribution of each strain within mixed samples
(Figure 4-1). In experiment 18, erS was more prevalent than er5//3 8CVI, within
single and also in the few dual-staining follicles. Pyrosequencing results from the same
birds were generally very similar to that of IHC except in experiment 1B with
er5//38CVI challenged first, in which most samples that were positive for
er5//38CVI only by IHC were detected as mixtures by pyrosequencing (Figure 4-2).
Short interval contact birds had virus frequencies very similar to the frequencies seen in
the donor bird groups for both IHC and pyrosequencing. The contact group from
donors inoculated with er5//38CV1 first in experiment 1B had a higher frequency of
the first virus (er5//3 8CVI) compared to the inoculated group. Results from qPCR
analysis were extremely variable within short and long interval challenge groups (data
not shown). There was no clear pattern or significance between virus quantities in
single-challenged compared to dual-challenged chickens.

Long interval challenge in experiment 1 led to markedly different results
compared to the short interval (Table 4—3). Based on IHC, all follicles stained only for
the first virus in both experiment 1A and 1B, regardless of which virus strain was
administered first. Pyrosequencing results were similar although the second virus was

detected in mixture with the first virus from one bird in two groups. IHC results from

127

 

contact recipient birds similarly were positive for the first virus only, although virus
staining was very low in 3 of 4 groups. Virus sequence was undetected by
pyrosequencing from all three groups with low staining.

Experiment 3 introduced samplings at multiple times to (1) investigate the
influence of time on the interaction between viruses in FFE and (2) confirm the results
of experiment 1. Overall, 102 of 106 samples (96%) had the same virus predominance
at all three sampling time points (Table 4-4). Following short interval challenge, the
er5//38CVI strain was present as the majority strain when administered with
J M/102W, as J M/ 102W was only present in F F E from two birds. Mixtures were also
rare following short interval challenge with erS and er5//38CVI strains, but both
strains had relative equal chance of exclusive majority within F FE of individual birds.

Following long interval challenge, the first virus administered was the exclusive
majority for both virus pairs in all but one sample (Table 4-4). In contact recipient
birds to long interval challenge, however, the second virus was much more prevalent.
Contact birds were exposed to donor birds that both passed and failed initial FFE testing
used to confirm the presence of the first virus challenged. In donor birds that failed the
initial test, the second virus was present in 42 of 95 birds (44%) (Table 4-5).

Quantitative PCR results from experiment 3 were again highly variable, but of
(particular interest was the low replication of J M/ 102W strain in all single and dual-

challenged birds (data not shown).

Tumors

128

 

Experiment 2 was similar in design to experiment 1 but focused on the
virological analysis of tumors. Data from uninfected and single-challenged control
birds in experiment 2 indicated that both pyrosequencing and virus isolation assays
were highly specific (Tables 4-6 & 4-7). Due to high mortality, the number of samples
per group was quite low in many cases. Following short interval challenge, 8 of 10
tumors were positive for both viruses among both virus pairs (Tables 4-6 & 4-7).
Tumors from short interval contact recipient birds consisted of mixtures as was seen in
short interval donor birds, but there were more tumors that were predominant for a
single virus. Virus isolation results were similar to pyrosequencing results for both
virus pairs in dual-challenged and contact recipient birds with nearly identical
proportions of samples positive for each virus as in pyrosequencing. Due to low sample
numbers in dual-challenged chickens, average plaque counts for each virus were
compared between single and dual-challenged chickens using the Wilcoxon rank-sum
test. In experiment 2B, there were no significant differences between the plaque counts
from viruses challenged on day 1 or day 2 in single versus dual-challenged chickens.

Following the long interval challenge, the first virus was exclusively
predominant in 16 of 16 (100%) tumors in experiment 2A by pyrosequencing, whereas
experiment 2B included 3 of 8 (37%) tumors that were positive for only the second
challenged virus. Whereas long interval contact birds were similar to donor birds in
experiment ZB with the second virus present in 10 of 24 tumors (42%), the presence of
the second virus in 3 of 22 tumors in experiment 2A was unexpected. Similar as with
FFE results, contact birds were exposed to donor birds that failed initial testing in which

the second virus was prevalent (Table 4-5). In virus isolation, all but one sample in

129

experiment 2A were positive for the first virus only and all contact recipient birds were
positive for this first virus only. In experiment 2A, there were no significant differences
in plaque counts for viruses administered on day 1 in single vs. dual challenged birds,
however there was a significant reduction in plaque count for viruses administered on
day 14 in dual-challenged birds compared to single control birds (p-values 0.0333 &
0.0476).

Multiple tumors were compared from individual birds either dual-challenged or
contact exposed from experiment 2 (Table 4-8). Only 1 of 24 birds with multiple
tumors had differences in virus frequency >10% between tumors.

Experiment 3 was designed as a replicate of experiment 2, except using only
pyrosequencing (Table 4-9). A major difference in experiment 3 was the use of nested
PCR prior to pyrosequencing due to DNA degradation. The result was a reduction in
specificity as high levels of background noise were detected in 13 of 84 tumors (15%).
Following short interval challenge er5//38CVI was exclusively predominant in 19 of
21 tumors when challenged with J M/102W, whereas both virus were present in tumors
following challenge with erS and er5//3 8CV1 with mild favor towards the latter
strain. Tumors were not present from the long interval groups in experiment 3B, but
only the first inoculated virus was present following long interval challenge in
experiment 3A. Unlike results from donor birds, tumors from long interval contact
birds in experiment 3A tended to favor er5//38CVI as opposed to the first virus

inoculated.

Effects of vaccination

130

 

Experiment 4 compared the effects of vaccination on results observed from the
first three experiments. In the F FE, there were no significant differences between
vaccinated and unvaccinated chickens for either virus pairs or challenge intervals
(Table 4-10). Results were consistent with experiment 3, demonstrating a preference
for er5//38CVI in short interval challenge with JM/102W and a loss of this effect
following long interval challenge.

As expected, fewer tumors were obtained from the vaccinated chickens than
unvaccinated chickens, yet after short interval challenge the presence of the second
challenged virus was relatively consistent within each challenge group (Table 4-11).
er5//38CVI was present in greater frequency in virus pair 1, but dominance was not
detected with virus pair 2. Following long interval challenge, 3 of 23 tumors were
detected as mixtures, but all others were exclusively composed of the first virus
challenged. Quantitative PCR results fiom experiment 4 were highly variable, and
similar to experiment 3 was a noticeable low replication of JM/102W strain compared
to erS and er5//38CVI strains. There were no significant differences in virus
quantity between vaccinated and unvaccinated birds (Figure 4-3). -

Unvaccinated birds in experiment 4 also served as a replicate for earlier
experiments, except challenge was administered at day 6 instead of day 1. Short
interval FFE results were most similar to experiment 3 results, demonstrating a strong
dominance of er5//38CVI over JM/102W. When J M/102W was administered first,
er5//38CVI was present in 10 of 10 samples with J M/102W only present in 5 of 10.
When er5//38CVI was challenged first, 10 of 10 were positive for er5//38CVI and

zero for J M/ 102W. Neither virus was favored with the er5 and er5//38CVI virus

131

.t{_

pair. Similar to both earlier experiments, nearly every sample was exclusively
predominant for the first challenged virus following long interval challenge for both
virus pairs (41 of 43 samples). Tumor samples were nearly the same as FFE samples,

as was the case in earlier experiments.

Summary Data

A chi-square analysis was performed to test the independence between time
interval of challenge and the exclusive predominance of the first virus in the FFE by
pyrosequencing. Results were stratified by virus pair and combined from all four
experiments comparing the number of samples with the second inoculated virus present
based on inoculation order (Table 4-12). This table focused on the second virus as its
presence was the main determinant for judging the effect of time interval and the
detection of dominance. In the FFE, the presence of the second virus was significantly
different following challenge with J M/102W and er5//38CVI for both short and long
interval challenge. This difference was based on order of inoculation as the second
virus was only present in significant quantity when er5//38CVI was the second virus.
No significance was seen following challenge with er5 and er5//38CVI. The
presence of the second virus in F FE samples was strongly dependent on the challenge
interval for both virus pairs, regardless of the order of inoculation (p<0.0001). Results
were nearly identical in tumors, except significant differences for the J M/ 102W and
er5//3 8CVI virus pairs were only seen following short interval challenge.

Some variation was detected between experiments in this study. Most

noticeable of these variations was the difference in the degree of dominance for

132

 

er5//38CVI following short interval challenge between experiment 1 and subsequent
experiments. In general, er5//38CVI was generally present in greater frequency in

later experiments within both virus pairs.

Discussion

Chapter 3 demonstrated that dominance should be considered as a factor that
influences the outcome of superinfection. Virulence, as measured by-pathotype, was a
significant factor that may have defined dominance. The J M/102W strain, a vMDV
strain, was subordinate when challenged simultaneously with the er5//38CVI strain, a
recombinant of a vaDV strain. When viruses of similar virulence were challenged
simultaneously (er5 and er5//38CVI), there was no clear strain that was
consistently dominant. The experiments in the current chapter were designed to 1)
evaluate the effect of time with the outcome of superinfection and 2) evaluate the
strength of dominance when the same two virus pairs were challenged following
different time intervals.

With the short interval challenge (24 hours), dominance measured in the F FE
was detected when J M/102W was paired with er5//38CVI (Table 4—12). The
dominance of er5//38CVI over JM/102W was most evident in experiments 3 and 4,
although it was also weakly evident in experiment 2 when analyzed with results from
contact recipient birds. Similar to chapter 3 results, dominance was weak and
inconsistent when birds were challenged with erS and er5//38CVI.

The effect of time was strong compared to dominance in F FE. Any dominance

detected following the short interval was inapparent following long interval challenge

133

(13 days) with both virus pairs. F F E samples were nearly all exclusively predominant
for the first challenged virus, regardless of virus pair or which virus was challenged
first. Chi-square analysis confirmed a strong association between challenge interval and
predominance of the first virus challenged (p<0.0001).

The second virus had opportunities for infection, even following a long interval
between challenges. In experiments 1, 3 and 4, there were a total of five birds in which
the second virus was present in the F FE by pyrosequencing after the long interval. Of
greater importance, however, was the presence of the second challenged virus in cases
when birds failed early infection with the first virus challenged on day 1. Experiments
3 & 4 focused on the probability of superinfection following confirmation of infection
with the first challenged virus, but analysis of samples that failed this confirmation
revealed an abundance of the second challenged virus in FFE samples (Table 4-11).
These birds were most likely the source of the second challenged virus that was present
in many of the long interval contact birds in experiment 3. The reason for the lack of
response to the first challenged virus in some birds is unknown. Viruses were
challenged with two separate injections to minimize administration failures, yet
somehow some birds appeared to have the effective dose diminished. Thus, conditions
that favor the second challenged virus in the field include short interval between
exposures and in chickens with low dose or delayed exposure to the first virus. Birds
infected in the field are likely to be exposed to different levels of virus. If the initial
exposure is weak enough, superinfection and establishment of the second virus may be

more likely.

134

 

Results from contact birds showed a close relationship between virus
frequencies present in the FFE of donor birds and frequencies present in the FF E of the
contact recipient birds. This indicated that the F FE pyrosequencing results from donor
birds were reflective of the frequency of viruses being shed and exposed to contact
chickens. Viruses shed in the highest frequency from donor birds usually had the
advantage in contact birds. Data from additional passages through contact birds would
be very useful for tracking the competition in fiJture generations to see if mixtures
remain or if one virus becomes mostly exclusive.

Tumors were collected and analyzed in experiments 2, 3 and 4. Results of virus
frequencies were similar to results seen in FFE with the presence of both viruses
following short interval challenges and the predominance of the first virus with long
interval challenges. Three tumors were predominant for the second challenged virus
following long interval challenge in experiment 2B, but the infection status of these
birds at the time of the second challenge was not tested. Similarly, the second
challenged virus was present in contact birds from experiments 2 and 3 exposed to long
interval challenged donor birds, but again, the initial infection status of the donor birds
was unknown. In experiments 3 and 4, the necessity of using a nested PCR for tumor
samples unexpectedly reduced the specificity of the assay in single-challenged control
chickens. Overall, the data following dual-challenge was consistent with results from

experiment 2, but mixtures were interpreted with caution. Results were included in the

overall summary, but not for the comparison of multiple tumors within individual birds.

Early studies suggested that MDV tumors were of a monoclonal origin (3 7), but more

recent studies have suggested a polyclonal tumor origin (13, 24, 86). The presence of

135

1

lymphoid proliferations in many tissues early in infection would favor polyclonal
tumors as well as the presence of both viruses within a single tumor. And although
most chickens with multiple tumors in experiment 2 were predominant for the same
virus in each tumor (24 of 25), there was one case of tumors of different predominance
within individual birds. However, the presence of more than one virus in a tumor does
not necessarily provide enough evidence to imply both viruses were causative agents for
the tumor. Rather, if a single cell containing both viruses became transformed, it would
be difficult to know which strain was involved in the transformation. The existence of a
virus-specific transformation marker would be necessary to answer this question.

It was interesting to note that vaccination had no apparent effect on virus
frequencies in short or long interval dual infections. HVT was expected to mute the
effects of the first challenged virus significantly enough to reduce differences detected
between the first and second virus. Yet dominance was still detected between
J M/102W and er5//3 8CVI following short interval challenge, and the first virus
remained significantly predominant following long interval challenge. The results
suggest that these effects may both be relevant under field conditions although this has
only been demonstrated with the use of weak vaccine.

This study provides evidence that MDV superinfection can occur following
short periods of time between exposures, but the probability is drastically reduced
following a longer interval of two weeks between challenges. In cases where birds are
exposed within shorter periods of time or where the initial exposure is weak or delayed,
mixed infections are likely to occur regardless of vaccination status or inoculation

order. These studies demonstrate that mixed infections under some conditions can lead

136

to shedding of the second virus a prerequisite for the establishment of the second virus
strain in the population.

Superinfection is likely an important step in the establishmentof evolved strains
within a population. The sequence of events may begin by the generation of a mutant
strain within an infected chicken or from an outside source introduced into a poultry
flock. By either origin, evolution won’t occur unless conditions exist for replication
and transmission of the second virus strain. This work provides examples of conditions
that would promote establishment of the second virus. Future work should include
additional virus pairs of similar and different virulence to confirm the phenomenon of
dominance. In addition, gaining a better of understanding of how delayed challenge
excludes the second virus and determining a more precise time interval for the
exclusion would provide important knowledge that may be used to help slow virus
evolution in the field. The lack of effect by vaccination should be tested with stronger

vaccines to confirm relevance under field conditions.

137

0000500000 5 00000005 000 N 0000:0098 00.0 0mm 0:29:00 4
.300 <3: .00 000—0 E 000: 003 mg): 3088. 300002

003 £053 Summing/t 0:0 9020 000: mm a. m: 00:0Et0mxm 00:30 0:0 E 030% ._>me\\m02._ 0:0
3mg 0):. 000: <m 0% <_ mucofiwsaxm .0000 95> 000000.00 wEms 3.80 025 00; 0000 N 000 y magfitoaxm m

_ 532 5000 0003 0000—030 =< N
0000500000 5 003205 0.8 N EOEt0ax0 000 0000800 5on0 5 0000000005 _

 

 

 

 

 

 

 

 

 

 

 

:0 ON 8280 0 888 E .3 880.82 :0 S 2
:0 ON 880 _ 888 2 .3 8882 :0 2 0
:0 on 8980 0 888 : .3 8882 :0 S t
:0 cm 8980 _ 888 0 .3 2,882 :0 2 2
00 ON 3988 8805:: 2,882 2 2
80 NV 2823300 0% E 3 3803: 35000023 9,882 m0 2 E
80 N... 3338338 0% E ._ 5000502. .3802: 8882 :0 o. 2
80 N4 Bean 0% 3 38 5000523 2,882 :0 S N_
80 N4 Bean 0% E 38 38:3: 03882 :0 S :
80 N4 2538 80 _ 38 3000820 2,882 :0 2 2
80 N4 25308 80 0 38 380:: ”>8tz :0 2 8
ACC NV £03200 680005.03 OZEMOZ A: w
00 ON 2553300 0% N ._ BN2: 35003820 880.32 :0 S :
:0 2 23023308 0% N ._ 5000082. .3803: 8882 :0 2 0
:0 9.. Bean. 0% N 38 5000032. 3882 :0 2 m
:0 ON 8&8 8m. N 38 38:3: 3882 :0 2 4
:0 ON Bean 80 _ 38 5003820 2,882 :0 S N
00 om 8.88 gm 2 38 BN2; 3882 :0 2 N
:0 om 3088 8.808: 2,882 S _
v A933 own A0303 00¢.
0.50800 0.25 0000—12—0 NEE: 005—125 N30003:.“ :3.—0.02 use—030 n 004

 

 

0000500000 5 008065 0.8
N 0008500 000 0800 00:00:00 000 00035:: 0000000 5 000080.05 .N 000 _ 00:08t00x0 00.0 anew 00080000 .00 030:0 "TV 030,—.

138

022200333222 00m 003 08G .3N02 N22. .20 0002.2 :2 0000 003 m02>20
88.... 28882 a; 82.23 $000282. 8. 82. 2.3... m2. .0 mm $888.5 8.2.8 .8 a 8.22. 3500.82.
000 3N2 N22. 000: 2. 0% <m 020082000xm2 .0223 00.22.» 20020.20 0020: 02.282 032 0022 22000 0 000 m 020082.200xm N
00202 0.200 m 00220200002 200.2 .3 00302202 .2520 000m 20 000 00 2 .30 00 2000020202800 003 0 20082002208 :2 00020023 N
2 22022 02000 80.5 000020200 :3. 2

 

 

 

 

 

 

 

 

 

 

 

m0 .3 .2 2 .0 3002222. .2>meNNm0S20 .25: 0322002 2 02

mN. .mm .2 2 .0 38:22. .2>meNNn02>2._ 0322000 2 0

mN. .mm .2 N. .0 382:): 23002202.. .2.>E 03220052 2 w

mm .mm .2 w .0 382:): .2>meNNm02>20 0322002 2 N.

mN. .Nm .2 2 .0 2>meNNm02>2.2 .3820): H>22 0322002 2 0

mm .mm .2 2 .0 2>meNNw02>2.2 .3822. 03220052 2 m

mN. .mm .2 N. .0 300200320 .3822: 25: 03000.2 2 0

mm .mm .2 N. .0 500.3020 .3820): 0322002 2 m

mN. .mm .2 0200000 0020000225 .25: 0222002 02 m

mN. .mm .2 0200000 002000020: 030002 2 2 2.
ms mm 02002000 22 20.2 0322002 2 2
mm mm 02002000 2 20.2 0302002 2 m2

0N. Km .2 2 .2 382:): .2>meNNm02...2H 03000.2 0m 22

2. .N.~ .2 2 .2 5002020 .3822: 0322002 21m 2

ON. .N.N .2 N .2 33222 .2>meNNm02>20 0322002 21m o,

2. .N.~ .2 N .2 5002022 .3820): 0302002 21m w

2. .N.N .2 2 .2200 23002022 0322002 2 N.

2. .N.m .2 2 2:0 38222. 0322002 2 0

2. Km .2 m .2200 2>Uw§m0§h 0322002 2 m

00 Km .2 N .2200 382:): 0322002 2 v

on .2.. .2 2 .2200 2>UmmNNm02>22 0322002 2 m

o: :N .3 2 38 3802. 03.8.2 2 N

2. .N.~ .2 0200000 0020020225 022200; 2 2 m

20.2000
00.. 0:229:00 20.2000 000 .2220 NEE: 000022220 N05000:» 2.200.222:N 2050002 0000—02—20 02 20.2 92....—

 

 

.0 00.0 m 0200820098 00.2 820000 20082000 .20 0022200 "N4. 0200.2.

139

Table 4-3: Experiment 1 — FFE. IHC and pyrosequencing results of FFE samples
from single-challenge, dual-challenged, and contact chickens exposed to JM/102W (J)
and/or er5//38CVI (M2) [Exp 1A] or er5 (M1) and/or M2 [Exp 1B]. Results are
shown from birds with passing results only (except groups in which all samples failed).
Samples with failed virus sequence detection are indicated by an F.

Footnotes:

1 Based on pyrosequencing results from experiment 1A single-challenged control
birds, samples were considered predominant for J if frequency was >90% for J
and predominant for M2 if frequency was 100% for M2. Samples were
otherwise considered mixtures.

2 Based on pyrosequencing results from experiment 18 single-challenged control
birds, samples were considered predominant for M1 if frequency was >90% for

M1 and predominant for M2 if frequency was 100% for M2. Samples were I
otherwise considered mixtures.

 

140

Table 4-3 continued:

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

IHC Pyrosequencing
(% follicles staining for (% birds predominant for
each virus, averaged by each virus)
group)
3» . - a"
5 E E 2 3 a a;
A T.“ g 8 g 5 'U ls! lst
5 m g (I) 15! & 2nd ls! & 2nd
11 only 2"‘1 only n oily 2"d only
Controls
1 None 30 10 0 0 0 9 F F F
2 J 1 30 9 100 0 0 8 100 0 0
3 J 2 30 4 100 0 O 5 100 0 0
4 M2 1 30 6 83 0 O 7 100 0 0
5 M2 2 30 4 100 0 O 5 100 0 O
6 None 42 10 O 0 0 10 F F F
:1. 7 J l 42 10 99 0 0 6 83 17 O
p 8 J 14 42 10 96 0 0 10 100 0 0
Z 9 M2 1 42 8 100 0 0 8 100 O 0
E 10 M2 14 42 6 100 0 0 5 100 0 0
E Dual-challengedl
2 11 J, M2 1,2 30 9 12 55 33 9 11 33 56
[a] 12 M2, J 1, 2 30 7 14 29 57 7 14 29 57
13 J, M2 1,14 42 8 95 0 0 9 89 11 0
14 M2, J 1,14 42 4 100 0 0 6 100 0 0
Contact exposed to lot
11 29 7 44 37 29 9 22 56 22
12 29 10 8 41 51 10 20 30 50
13 29 10 100 0 0 0 100 O O
14 29 9 22 0 0 7 F F F
Controls
1 None 30 6 0 0 0 10 F F F
2 M1 1 30 10 100 0 0 9 100 0 0
3 M1 2 30 4 100 0 O 4 100 0 0
4 M2 1 30 7 100 0 0 7 100 0 0
5 M2 2 30 3 100 O O 3 100 0 0
6 None 42 10 0 0 O 10 F F F
E 7 M1 1 42 7 99 0 0 5 100 O 0
[— 8 M1 14 42 9 99 0 O 9 100 0 0
Z 9 M2 1 42 4 92 0 O 4 100 0 0
E 10 M2 14 42 10 85 0 0 9 100 0 0
g Dual-challenged2
a 11 M1,M2 1,2 30 5 77 8 O 4 75 0 25
h: 12 M2, M1 1, 2 30 7 0 0 100 7 0 86 14
13 M1,M2 1,14 42 5 100 0 0 7 100 O 0
14 M2, M1 1,14 42 5 95 0 O 7 86 14 0
Contact exposed to lot
11 29 5 96 4 0 4 100 0 0
12 29 8 63 9 28 8 63 12 25
13 29 10 10 0 0 4 F F F
14 29 20 0 O 4 F F F

 

 

141

 

Table 4-4: Experiment 3 - FFE. Number of birds with FF E samples positive by pyrosequencing based
on order of inoculation for each virus in single, dual-challenged or contact chickens exposed to JM/102W
(J) and/or er5//38CVI (M2) [Exp 3A] or er5 (M1) and/or M2 [Exp 3B]. Results are shown from

birds with passing results (except groups in which all samples failed). Samples with failed virus sequence

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

detection are indicated by an F.
# birds predominant for # samples that
each virus on day 70 changed
1 5‘ predominance
Challenge Inoculation Sampling lSt & 2nd between three
Lot strains da 5 day 11 only 2"d only sampling dates
Controls
1 None 70 3 F F F
2 J l 70 6 6 0 0 0
3 J 2 7O 6 6 O O 0
< 4 J 14 70 5 5 0 0 0
m 5 M2 1 70 6 6 0 0 0
g 6 M2 2 70 7 7 0 0 0
E 7 M2 14 70 3 3 0 0 0
E Dual-challenged2
H 8 J, M2 1, 2 70 16 0 0 16 0
a 9 M2, J 1, 2 7O 8 6 2 0 2
10 J, M2 1, 14 7O 5 4 l 0 1
11 M2, J 1, 14 70 12 12 O 0 0
Contact exposed to lot
10' J,M2 1,14 I 70 I 9 ‘ o 1 8 o
11 M2, J 1, 14 70 9 9 0 0 0
Controls
1 None 3 F F F
2 M1 1 70 7 7 0 0 0
3 M1 2 70 7 7 O 0 O
a: 4 M1 14 7O 11 11 0 0 O
m 5 M2 1 70 6 6 0 0 0
E 6 M2 2 7o 8 8 0 0 0
[=3 7 M2 14 7O 9 9 0 0 0
E Dual-challenged3
E 8 M1, M2 1, 2 70 10 2 1 7 1
E 9 M2, M1 1, 2 70 8 4 0 4 0
10 M1,M2 1,14 70 12 12 O O 0
1 1 M2, M1 1, 14 7O 0
Contact exposed lot
10 M1,M2 1,14 70 ‘ 9 ‘ 7 1 1 o
11 M2, M1 1, 14 70 8 8 O 0 O

 

 

 

 

1 Long interval analysis only includes results from FFE samples that passed at all three time

points

Based on results from single-challenged control birds, FFE samples were considered
predominant for J if frequency was >95% on day 14. >90% on day 27 or >87% on day 70 for J.
FF E results were considered predominant for M2 if frequency was 100% on day 14, >97% on
day 27, or 100% on day 70 for M2. Samples were otherwise considered mixtures.

Based on results from single-challenged control birds, FFE samples were considered
predominant for M1 if frequency was 100% on day 14, >97% on day 27 or >86% on day 70 for
M 1. FFE results were considered predominant for M2 if frequency was 100% on day 14, day
27, or day 70 for M2. Samples were otherwise considered mixtures.

142

 

as: 50553: .5 A :6 was: .5 3.0.234 653 8% mafia owfizfio N
.wczofiwm mum BE 05 um oozfl 35 838% mum Sow 3:58 mocfiofi Eco mmemc/x _

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

o o o o 3 .3 mm 2 6 gm :2 ,NE
o o o o E E mm 2 d :4 ,NE
_ o o _ c N c 4 a mu 2 .o ,5: 22 c2
N o o N o m o o .6. mm 3 6 NE 42 me
o o o o S 2 mm 3 o H>m _. .NE
o o H _ v n m m 2 mm 9 6 ES NE E.
_ o _ N m m m E n mm 3 6 N2 4 $6
o o _ o n w on E 4 NE E2 mm
o m N v N o _ 3 m _ on 3 E _. ,NZ
v o m o m m“ m m a on E J 32 .n <m
Eco Em Eco c mwczcfiam Eco Em Eco c % m. m. M m, mu. ma
EN 6% at “L BE“ ecu ocooom EN 6% at “L .w M; m m m. M d
c8369 oococEcoooa H ( p.16. m. sz m
vomcfio 35 moEcSm 4% cm W. W cans
Sb Ewe co mar» sumo Sn E6 co mar» :03 m. u m
com acacmEooocc £255 3 com “cwEEooocm $59 3
wfiococcomoc: wfiococoomobmm
who—cob Em,"—

 

 

 

 

 

.mc.c> ocooow 2: mo
conflcoofi 8on3 Eocflooacc: :83 wczcccwm $5 05 «com 3:58 82mm on: $5 omofi Bot mcoono wowcofisoéco 3%?“ch
cm 33:88 m3? some com wcmococoomoba 3 03963 mobs mo conch: Z .338.— EEE goo—man— I v Q m mucoEEoA—xm— ”mtv 2a.;

143

.mncocw E898 mcoca EEEScoE EcoEoono cit SAME 9 0ch E30 vocal—EB ma? Ecoatonxm m
.EEQEUonEEcu EmcoEoocoHEmocmEm N-E mo EooEmEcho oEmEmm comm N
.3533: E55238 8:225 953 838%
.22 com $03 $3 zococooa .EE NE SE EcmcEEoooE ecu H 8E fowoA mg» hococcoa EE 75E EcmcEEoEooa E55288 Bo? 338% £35 Eobcou

vowcoEEflEooEmcEm Bob 33ch co vommm .Eonmccwm cooEnEm Eobcoo voEoochcc Fm BEBE Enooxov mEEcmoc mcEmmfiE 5:» £23 80cm 850% 0.8 mEEcmoyE E

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

o o o o o w o o m E E NE. E E

0 EE o o > El E N o a NV E:

2 E E E o o E N E. S NV 0

m 5N N N E o m m N w NV w
.2 8 E5893 3350

0 EN o o N m o o m m Ev E ,E H .NE>E EE

5 o E o o a o o E 2 E» E .E NE .E 2

o o o o o E o N o N Ev N .E E. NE a

o o o o o m E v o m Ev N .E NE>E A w
comco—EacoEacn

o v o o E. 2 o o E6 v Ev 3 NE 5

o m o o E N o Ev N NE>E o

0 NE 0 o o E o o v v Ev E NE m

0 CE 0 o o o o o o o Ev E E v

o N o o 0 E: o o o m Ev N H m

o E o o o E: o o h n Ev E E. N

o o o o 0 E: ”E “E “E S Ev 282 E
m—ObflOU

EN a E Eco EN Eco c Eco EN Eco c mENEo Amino mcEwbm 84
EN as e E EmE EN as a E chEEEEow coEEmEcoocE owcoEEfiEU
smE
5.3 :03 AmEE> AmEE> some
com mEEoo cc E \Eccoo some com o>EEEmom 838% com EcocEEoEooE .2053 3
0:33 m>< Eo E Ncorn—omE 25> Ewfiococoomoc:

 

.mceroEo EcoEnEEooc Eoficoo Eoca ANEE E>Omm\\m32.c 8B5“ 5 3352a 55 camcoEEfiEo mceroEo EoomcoEEEEo
-Emch was oEmcEm EoEE 3E3“: @333 coEEmEomE 95> Eocm Cocccv chocochomobEnE .Eofich I <N EcoEEuomxm 64 03:,

144

 

.on E8 co chEaSmm 05 80¢ 052846 083 $5 83ch 52% 02

v
m

.mmcoEm EEo>om mcocue EEEotoE E58098 55 .5:ch 9 26 E30 vaccine 33 Ecoatomxm

.EEEEEooxEEEqa EacoEoocoE\moEmEa NE .3 voEmEcho oEnEEdm comm

N

.8538 E85238 omEEofio 203 .83ch
NE 8E .32: was» Eocozoocm EE NE SE EcccEEoEookE v5 ESE com .xemoA ma? EococooEE EE ESE BE Ecacccocoa 35238 803 8383 docs Eobcoo
camcoEESEoéEmcEm 88E mEEomE co vomum .EonnEcam 82% Eon—coo EuoEooEcEcc En: voEEfl 38on mEEcmoc mcEmmfiE 5E3 mEEnE 88E cBoEEm 8m mEEcmoyE E

 

 

 

 

 

 

 

 

0 vv 0 o v o N N o E Nv EVE
EN MN m o m n o o m E E NV 2
o o o o o o v o N o Ev NE
E E E: E o m w o o v v Ev E E
E2 3 E5898 6550
m2 N o m m om E ,E ESE .NE 3
m2 E o N m om E .E N2 .22 NE
E: 3; o m o m E E o N Nm N.E ESE.NE>E NE
Elmo RN 0 E o E o E o E Nm N,E NE>E.EE>E EE
couco=§EoEa=Q
m2 0 o N N mm 3 NE 9
m2 o o v v om E NE 0
m2 0 o E E am 3 EE m
mZ o o v v om E :2 n
vwz “E “E ”E m om 282 o
0 SN 0 o n w o o v v Nm N NE m
o EEN o o E. n o o c 0 NM E NE>E EV
o va o o o m o o w w Nm N :2 m
o 3E o o o a o o o o Nm E E2 N
o o o o o m E “E “E c Nm 282 E
23280
EN HmE a EN 3 c Eco Em was a «an $56 mecca “3
EN as 3 3E EN as :E .mE mcEEaSmm coEEmEcoocE owcoEEflEo
mEE> :08 SE Emit, some SE 03:” on $23 some SE ccacEEoEeoa

mEEoo 02 5550 occmEnE w><

 

8383 E0 no NcouanmE 25>

 

@583 3 act—5:63.33

E

 

 

 

 

 

 

L

.chEoEo EcoEmEoE .8280 Evan GEE Sowmtmgfi cOEEocm EEE>EE 331 5:5
vowcoEEflEo mcoonEo newcoEEflEoEach ecu oEmcEm Boa 3:58 9333 coEEmEOmE 25> Eocm Cocccv chocochomobEE .2255. I MEN EcoEEconnm "We wish.

 

145

Table 4-8: Experiment 2 - Comparison of multiple tumors. Pyrosequencing results
of birds with multiple tumors in experiment 2. Birds were dual-challenged with either
JM/102W (J) and er5//38CV1 (M2) or er5 (M1) and M2. Results from contact
birds are indicated. Pyrosequencing results show the ratio of each virus present in DNA
isolated from the tumors.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Pyrosequencing
Frequency of each virus in
tumors
%1St virus:%2nd virus
Challenge Challenge Sampling
Exp strains day Bird# Nerve Heart Gonad
J, M2 1,2 58 Q2024 38:62 68:32 47:53
Contact to J, M2 1,2 58 Q2102 0:100 0:100
Contact to J, M2 1,2 58 Q2108 91 :9 92:8 80:20
M2, J 1,2 58 Q2054 48:52 57:43
Contact to M2, J 1,2 58 Q2116 0:100 0:100
Contact to M2, J 1,2 58 Q2117 0:100 0:100
Contact to M1, M2 1,2 58 Q1695 0:100 0:100
Contact to M1, M2 1,2 58 Q1700 0:100 0:100
Contact to M2, M1 1,2 58 Q1707 0:100 0:100
Contact to M2, M1 1,2 58 Q1717 100:0 100:0
J, M2 1,14 70 Q2035 100:0 100:0
J, M2 1,14 70 Q2037 100:0 100:0
J, M2 1,14 70 Q2045 100:0 100:0 100:0
J, M2 1,14 70 Q2046 100:0 96:4
Contact to J, M2 1,14 70 Q2136 83:17 75:25
Contact to J, M2 1,14 70 Q2137 100:0 99:1
Contact to M2, J 1,14 70 Q2153 O: 100 0:100
Contact to M2, J 1,14 70 Q2157 0:100 0:100
Contact to M2, J 1,14 70 Q2161 0: 100 0:100
Contact to M2, J 1,14 70 Q2162 0:100 0:100
Contact to M1, M2 1,14 70 Q1735 100:0 100:0
Contact to M1, M2 1,14 70 Q1738 99:1 100:0
Contact to M2, M1 1,14 70 Q1748 99:1 100:0
Contact to M2, M1 1,14 70 Q1749 0:100 0:100

 

l Spleens were used for virus isolation. Each sample consisted of 1-2
plates/monoclonal antibody/bird.

2 Tumor abbreviations: Enlarged nerve (Ner), Heart (Hrt), Spleen (Spl), Gonad (Gon),
Liver (Liv), Kidney (Kid)

3 The numbers of control group virus isolation or tumor samples of each type are
indicated within parentheses

146

 

Table 4-9: Experiment 3 - Tumors. Number of birds with tumor samples positive by
pyrosequencing based on order of inoculation for each virus in single, dual-challenged
or contact chickens exposed to JM/102W (J) and/or er5//38CVI (M2) [Exp 3A] or
er5 (M1) and/or M2 [Exp 3B]. Results are shown from birds with passing results
(except groups in which all samples failed). Samples with failed virus sequence
detection are indicated by an F. A nested PCR was used as described on page 120.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

# tumors predominant for
each virus on day 70]
Challen ' ‘ 3‘ 1‘“ & “d
ge Inoculation Samplmg 1 2
Lot strains day(s) day 11 only 2nd only
Controls
1 None 70 2 F F F
2 J l 70 2 1 l 0
3 J 2 70 3 1 2 0
< 4 J 14 70 2 0 1 l
m 5 M2 1 70 l 1 0 0
E 6 M2 2 70 7 7 0 0
g 7 M2 14 70 6 6 0 0
a Dual-challenged 2
E 8 J, M2 1, 2 70 9 0 0 9
E 9 M2, J l, 2 70 12 10 0 2
10 J, M2 1, 14 70 2 2 o o
11 M2, J l, 14 70 5 5 0 0
Contact exposed lot
10 J, M2 I 1, 14 ’ 70 t 5 . [ 0 2 3
11 M2, J 1, 14 70 8 7 l 0
Controls
1 None 3 F F F
2 M1 1 70 14 13 1 0
3 M1 2 7O 7 5 l 1
m 4 M1 14 70 16 14 2 0
m 5 M2 1 70 6 5 l 0
E; 6 M2 2 7o 9 8 1 o
E 7 M2 14 70 11 10 l 0
a Dual-challenged 3
E1 8 M1, M2 1, 2 70 19 4 2 13
L1: 9 M2, M1 1, 2 70 13 8 1 4
10 M1, M2 1, 14 70 0
11 M2, M1 1, 14 70 0
Contact exposed lot
10 M1, M2 1, 14 J 70 J o ‘
1 1 M2, M1 1, 14 70 O

 

 

 

I Long interval analysis only includes results from tumors with passing FF E
results at all three time points.
Tumor samples were considered predominant for single viruses if frequency of J
was >98% or M2 was 100%. Tumors were otherwise considered mixtures.
Tumor samples were considered predominant for single viruses if frequency of
M1 was >87% or M2 was >96%. Tumors were otherwise considered mixtures.

147

 

as: E53522 5 Es 82E .5 38:3: as; was was: 85:26 N
850m 0:5 3:: EEm Ea comma 65 838mm mEnEnE SEE 3:58 89:05 Eco £9388 BEBE weed E

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

83 o o o o fl 2 6 SEE E2 .NE 2
o o NE NE mm E .o :2 .82 m E
. o o w w mm E .c ,El>E.E NE .22 E
E ~88 E o EE NE we 2 .0 922 ME
oooE o o m m mm 3 .0 H>EE E. NE NE
0 O Q E mm E d H .NSE EE
E . o E o E mm E d .E1>E.E NE .E. 2
E :8 o o E m c E 2 .0 NE .E a
mEONd o o o 0 mm E. 6 H>EE ESE nNE>E m
N E 5 2 mm m .o :2 NE n
N mEde m E 0 2 mm \16 ,E.>I NE ”:2 c
E N o n 0 mm B 6 NE>E .22 m
oooE o o EE EE mu m .o FEE E. .NE>E v
E o 0 CE 2 mu m .c a .NE>E m
N . w m 0 EE mb 5 6 E1>EE NE>E .E. N
N EN 0 m m o E: E a .8 NE .E E
moat wEEnEEmm osEm>AE %Emo EVEEN bee a New Amine mane NmEmbm SE
085 50355 Nx EEN 6% E EmE wEEEEEam 20:23er 0588/ ownoEEwEEU
vocaeEEoon Emu Eng mo
woweano me; :23 E8 ESEEEEEOESOEEE £85 3
ES: moEQFam a Mia—5:70.883
.35 05mm 05 89¢ £083

5 3:58 Hubbub .8 3:86 no 8me Bug 8m 3E3“: 583. .08: b>o wowEEEEo 35.. 085 was wEEEEEum :08 En Egfifiocoa we?
we? 083 05 :03? 305 .3 3:8 2a 338% mum 40183 some 8m 832305 .8 EEO .3 meooEoEo 332350-36 3338ng
new BEaeEoog E @8232: 95> :08 8E wfioconwomoba .3 033mg 83 mo #5832 Hum - 1 Santana "3;. 03:.

148

$2 8 83A 83 E. E0 83:35 .5 885> oEw5m EOE 8858038 38588 083 838mm 888 .E 58 85> EonE N
.353 25E 85E EEEE Em 338E mnEnE @588 553 888 805 5:85 8365 Eco 5388 E83885 was E

8.858 38588 85358 083 8083. .ExcooA 83 NE 8 EwaA 83
E2 .8 83:85 .5 885> oEw5m E8 8858038 38588 083 8388 .888 .N 58 85> 81E a\°ooE 83

 

 

 

 

 

 

 

 

 

 

 

 

 

o o N N 3. QE 6 .E.>m :2 NE E
o o m m E. E .c :2 NE E
o E m v m E .0 .E1>E.E N2 .32 E
o o n E. mm GE 6 N2 .EE 2
o mm 2 6 .E.>EE E. .NE NE
0 N N v mm QE .0 E. .NE>E EE
o mm 3 , .E1>EE NE .E. 3
o c m m mu 2 6 NE .a a
o o m m 3. n .0 E1>I ESE .NSE w
m o E v 3. n 6 ESE .NE 5
E E E m mm m 6 E.>E.E N2 .32 o
m E m o 2. n 6 NE “:2 m
o E N m mm m .o E1>I E. .NE>E v
o o N N mm m d E. .NE>E m
m E o o mu m d .E1>E.E NE .E. N
v o E m 3. n 6 NE .E. E
Eco EVEEN >an c .98 8388 A585 355305 @958 o53w> W588 owcoEEmEEU 85
am a. ...,E ..E
Emu >58 :0
85> 508 85 888838 8088 E...

 

 

149

.ONE 085 no 35385 8 3m: 83 mom 388 < ME 8 .3 38055 8m 805385 8333...” 85> 3E5 553 888mm _
A355 83:8 :8 5553 5 850% 333 3:88 8583 553 855 805 832.? 88 533E .Emm 9m: NE>E E838 :3: m2): .5

7.5. 95m: ENE): E >Uw MERE): 8588 AD BNO E h): 8 3858 805550 383 8 3wEEoEE8EoE8EV hoEw5m 5 85> 33 55 8055305
.3 858 so 385 w5o33E3mo§EE .3 83585 888mm 888 553 555 mo 858: Z 888$. .. v «nan—Euoaxm "=4. 55:.

Table 4-12: Summary table. Combined pyrosequencing data displaying the number
of samples where the second-challenged virus was present in FFE or tumor samples.

 

 

 

 

 

 

Number of samples with second inoculated virus present /
Total samples
Virus X2
challengel Exp 1 Exp 3 Exp 4 Exp 4 HVT Totals (%12 p-value
FFE Short
J, M2 8/9 16/16 10/10 11/11 45/46 (98%)* <0 0001
M2, J 6/7 2/8 0/10 0/11 8/36 (22%)* '
M1, M2 1/4 8/10 2/9 4/10 15/33 (45%) 0 7238
M2, M1 7/7 4/8 3/10 0/9 14/34 (41%) '
Group total: 82/149 (55%)A
F F E Long
J, M2 1/9 1/5 1/6 1/1 4/21(19%)* 00152
M2, J 0/6 0/12 0/13 0/5 0/36 (0%)* '
M1, M2 0/7 0/12 1/12 0/8 1/39 (3%) 1 0000
M2, M1 1/7 - 0/12 0/6 1/25 (4%) '
Group total: 6/121 (5%) B <0.0001
Virus X2
challengel Exp 2 Exp 3 Exp 4 Exp 4 HVT Totals 1%) p—value
Tumors Short
J, M2 5/5 9/9 4/5 6/6 24/25 (96%)* <0 0001
M2, J 2/2 2/12 0/2 1/3 5/19 (26%)* '
M1, M2 1/1 15/19 6/9 2/3 24/32 (75%) 0 2422
M2, M1 2/2 5/8 3/4 0/3 10/17 (59%) '
Group total: 63/93 (68%) A
Tumors Long
J, M2 0/13 0/2 0/3 - 0/18 (0%) 01517
M2, J 0/3 0/5 2/4 - 2/12 (17%) '
M1, M2 1/3 - 0/7 1/4 2/14 (14%) 1 0000
M2, M1 2/5 - 0/3 0/2 2/10 (20%) '
Group total: 6/54 (11%) B <0.0001

 

I Viruses challenge are listed in order of inoculation with the following viruses:
(J), er5 (M1), or er5//38CVI (M2)

JM/lOZW

2 Significant differences by chi-square analysis between pairs of results are indicated by
(*). Significant differences between group totals based on sample interval are

indicated by different superscript letters.

150

 

E23 3 382$ muu E 88 N 95>; 25>

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2:5 88 85¢ 88 one» 98,
:2 .8- #
as. .28 M
E. N:- w.
us. in
813 3585

8888 8x8 58 E§a8ouo5E 305888 8553 888835 85> €855 505m .8 85855559 - E 80585me "E... 28E

151

 

r: .2...
NE .FEMm
E. .NEI
NE .2“.

 

 

 

NE .E.
E. .NE
NS. 5.2
:2 .NE 1

:23 3. 882$ mu". 5 2.2 u 9:5; 25>

ccwno

owucw

oonov

3. ”co

 

 

9N5”

 

 

 

 

 

 

902

1o

1F

-N #
m
J

In W

1v

1m

85888853 55 58838“

838mm 858 58 E88808 385838 55.53 865835 85> E8855 55% .8 858555Q - E 85585me "Nuv ouaumm

152

 

, NED,
E i

iii

E2 069 :2 + a .89 N:

uoEmEoom>
E.>I

mm mm m_.

5885088:

no

ON

8 $9 «2 + 6 .69 :2

558608
._.>_._

2 mm mm 9

5586883

mm mm

5: ‘69 E. + 5... 089 8

598608
52..

m? 8 5N 9
80 80 E60 80 80 80 80 80 >50 80 80 80 Ema E60 xmo 80 80 Ema Ema 80 E80 80 86

E5 39 N2 + 8 >80 a

 

885008
8850883 ._.>I 50858ch
no 5N 2. mo 5N 2 mm MEN 9

E 80

 

 

 

E

 

 

 

 

 

 

 

 

flHHH HHH 1

 

 

 

o

ooow

ooom

ooom

000Eo

ooom

oooo

coon

ooom

ooom

"99 ISOLI/NOA

 

80558508586 80 80882 8mm .NNE 88 :0 80588

2: E 888% a 8an a 8382 22> 33:5 EN Eo mm .820 30888 as :20 88 5 .820 508588 Es
CE 3N0 E :2 E. 553 58858595850 88550 8850085: 80 885008 5 mum 805E MEUmEo .50 388% I v 88588me "mé 98w:—

153

Chapter 5
SUMMARY AND FUTURE STUDIES

The first chapter provided an introduction to Marek’s disease with a review
focused on the likelihood of multiple exposures to MDV during the lifetime of a
chicken. Evidence was presented demonstrating multiple virus strains simultaneously
present within a flock. The presence of multiple strains in a flock increases the
probability of multiple exposures to chickens and the possibility of infection with more
than one strain (superinfection). Despite anecdotal field evidence, there is little
laboratory evidence, demonstrating that two fully virulent Sl MDV strains can both
replicate and transmit within individual birds. Recent technology using recombinant
viruses gives us the opportunity to now identify individual virulent strains within mixed
populations. Superinfection is likely an important mechanism in the establishment of
evolved strains within a population; therefore, understanding the conditions that
promote infection with a second virus may be important for knowing how to slow the
evolution of MDV in the field.

The second chapter included three pilot studies evaluating the methods and
conditions used in the following chapters. Pilot 1 validated the sensitivity and
specificity of pyrosequencing and IHC, confirming the use of these techniques for
accurately distinguishing between virus pairs. Superinfection studies were based on
two virus pairs — one with similar virulence and the other of different virulence. Pilot 2
evaluated the pathogenicity of strains selected for the two virus pairs by pathotyping,
confirming the similarity between erS and er5//38CVI as well as the dissimilarity

between J M/ 102W and er5//38CVI. Pilot 3 validated the sampling methods used in

154

 

later chapters, isolating DNA from cells adhered to the outside of the feather shafi, to
reflect the frequency and quantity of shed virus from individual birds following dual
infection.

The third chapter demonstrated the phenomenon of viral dominance, as detected
in simultaneously dual-challenged chickens. For the first time, certain virulent 8] virus
strains were shown to have a competitive advantage over others. There was no
consistent dominance detected between the two similar strains, but the higher virulent
er5//38CVI was significantly dominant over the less virulent J M/102W strain. For
both virus pairs, the presence of one virus strain within bursa and feather follicles
apparently provided no exclusion from infection with the second virus. Interestingly,
virus mixtures were present in a large portion of tumors. Simultaneous challenge with
J W] 02W and er5// 3 8CVI strains led to some differences in the virus frequency
between tumors in the same bird, suggesting that these tumors may have a polyclonal
origin. Results from HVT-vaccinated birds suggest that the effect of dominance may be
relevant under field conditions.

The fourth chapter evaluated the effect of dominance following a short (24
hours) and long (13 days) challenge interval. Dominance was again detected for
er5//38CVI when challenged with J W] 02W following the short interval, regardless
of which strain was inoculated first. Any detection of dominance was lost, however,
following the long challenge interval as most FF E and tumor samples were exclusively
predominant for the first inoculated virus. Tumor samples generally consisted of a
single predominant virus, but both viruses were detected within a small portion of

tumors. Vaccination with HVT had no effect on the virus frequency for either virus pair

155

or challenge time interval, suggesting these results may be significant in field

conditions.

Several significant findings were demonstrated in these experiments.

1)

2)

3)

4)

This study established that superinfection with two virulent Sl MDV strains is
possible and indeed prevalent following simultaneous or short interval dual-
challenge. Both viruses can be simultaneously present within bursa, FF E and tumor
tissues, and frequently found within the same individual bursa and feather follicles.
Many times both strains were present within the same overlapping regions of each
follicle.

The phenomenon of virus dominance was demonstrated in at least one virus pair,
establishing a potential mechanism for the establishment and transmission of
evolved strains within individual birds.

The effect of delayed challenge was strong and masked the effect of dominance
observed in shorter intervals. The presence of the second virus in certain conditions
indicates superinfection may be especially significant in instances when time
between exposures is short and viruses are of similar virulence. Understanding
what events are responsible for the eventual exclusion of superinfection may help to
slow the progress of evolution within the field.

Many tumor samples were composed of both virus strains and in a few instances
multiple tumors within individual birds were composed of differing strain
frequencies. These results are consistent with but do not prove a polyclonal origin

of tumors.

156

5)

Although a weak vaccine was used, vaccination had no effect on virus dominance or
the effect of challenge interval on virus frequencies. This suggests that the effects

in these studies may be relevant to field conditions.

The experiments described in this study have prompted several additional questions that

deserve follow-up studies.

1)

2)

3)

The confirmation of viral dominance using more virus pairs of differing and similar
virulence. The evidence of dominance between er5//38CVI and JM/102W was
compelling, but only represents one virus pair. An evaluation of additional virus
pairs would help confirm which characteristics contribute most to dominance.
Replication appears to be closely related to virulence and should be evaluated with
reference to dominance.

Narrowing down the time interval necessary to exclude the second challenged virus.
We observed a significant effect between the presence of the second virus after
short interval (24 hours) versus long interval (13 days) challenge. An experiment
designed to evaluate the effect with challenge intervals between 2-12 days would
help to contribute to determining the cause of exclusion.

Serial passages through contact recipient birds. Results from contact recipient birds
were correlated with the donor recipient birds, but tended to include fewer mixtures.
In field conditions, establishment of new strains may take place over several
generations as less frequent strains grow towards greater fiequency. Evaluating
potential changes over multiple generations may provide further knowledge on

mechanisms for viral evolution.

157

 

4)

5)

6)

7)

Increased or varied dosages. Results from chapters 3 and 4 demonstrated
differences between the extent of dominance for er5//38CVI when challenged
with J W] 02W. Experiments were performed over a period of several years and the
later experiments used different virus lots compared to earlier experiments. The
increased apparent dominance of er5//38CVI may have been related to differences
in effective dosages between the first and second virus. New experiments with
higher total dosages and differing dosages for each strain would help elucidate the
effect of virus fitness compared to starting dosage.

Superinfection within individual cells. Superinfection was demonstrated within
individual tissues and both strains were observed in overlapping regions within
bursa and feather follicles. The methods used in the study do not provide
information on whether virus was present within the same cells. Flow cytometry
would be a suitable method to demonstrate multiple strains within individual cells if
a suitable marker was available for differentiation. Unfortunately, MDV pp38 is
poorly expressed in tumor cells, so this could not be evaluated in the current study.
Recombination of MDV strains. The evolution of MDV strains may be caused by
the establishment of dominant strains following mutation or introduction of new
strains into a poultry flock. Recombination between multiple strains may be another
source of virus evolution within birds. There is little or no evidence of
recombination within the MD system.

Determining the clonality of MD tumors. The experiments designed in this study

may be an appropriate model for studying clonality of MD tumors. Simultaneous

158

dual challenge with both virus pairs produced mixed tumors that could be analyzed

with additional assays, such as TCR spectratyping.

159

REFERENCES

1. Abdul-Careem, M. F ., B. D. Hunter, E. Nagy, L. R. Read, B. Sanei, J. L.
Spencer, and S. Sharif. Development of a real-time PCR assay using SYBR Green
chemistry for monitoring Marek's disease virus genome load in feather tips. J Virol
Methods. 133:34-40. 2006.

2. Bacon, L. D., H. D. Hunt, and H. H. Cheng. A review of the development of

chicken lines to resolve genes determining resistance to diseases. Poult Sci. 79:1082-
1093. 2000.

3. Baigent, S. J ., L. J. Petherbridge, K. Howes, L. P. Smith, R. J. Currie, and V. K.
Nair. Absolute quantitation of Marek's disease virus genome copy number in chicken
feather and lymphocyte samples using real-time PCR. J Virol Methods. 123:53-64.
2005.

4. Baigent, S. J ., L. P. Smith, R. J. Currie, and V. K. Nair. Replication kinetics of
Marek's disease vaccine virus in feathers and lymphoid tissues using PCR and virus
isolation. J Gen Virol. 86:2989-2998. 2005.

5. Benton, W. J ., and M. S. Cover. The increased incidence of visceral
lymphomatosis in broiler and replacement birds. Avian Dis. 1:320-327. 1957.

6. Beyari, M. M., T. A. Hodgson, R. D. Cook, W. Kondowe, E. M. Molyneux, C.
M. Scully, C. G. Teo, and S. R. Porter. Multiple human herpesvirus-8 infection. J Infect
Dis. 188:678-689. 2003.

7. Biggs, P. M. Marek's Disease. In: The Herpesviruses. A. S. Kaplan, ed.
Academic Press, New York. pp 557-593. 1973.

8. Biggs, P. M., C. A. Jackson, and D. G. Powell. The epizootiology of Marek's
disease, 2. The effect of supply flock, rearing, house and production house on the
incidence of Marek's disease. Avian Pathol. 2:127-134. 1973.

9. Biggs, P. M., and B. S. Milne. Biological properties of a number of Marek's
disease virus isolates. In: Oncogenesis and herpesviruses. P. M. Biggs, G. De-The and
L. N. Payne, eds. International Agency for Research on Cancer, Lyon, France. pp 88-
94. 1972.

10. Biggs, P. M., D. G. Powell, A. E. Churchill, and R. C. Chubb. The epizootiology
of Marek's disease: 1. Incidence of antibody, viraemia and Marek's disease in six flocks.
Avian Pathol. 125-25. 1972.

ll. Biggs, P. M., H. G. Purchase, B. R. Bee, and P. J. Dalton. Preliminary report on
acute Marek's disease (fowl paralysis) in Great Britain. Vet Rec. 77: 1339-1340. 1965.

160

12. Bulow, V. V., and P. M. Biggs. Differentiation between strains of Marek's

disease virus and turkey herpesvirus by immunofluorescence assays. Avian Pathol.
4:133-146. 1975.

13. Burgess, S. C., and T. F. Davison. Identification of the neoplastically
transformed cells in Marek's disease herpesvirus-induced lymphomas: recognition by
the monoclonal antibody AV3 7. J Virol. 76:7276-7292. 2002.

14. Buscaglia, C., P. Nervi, and M. Risso. Characterization of four very virulent
Argentinian strains of Marek's disease virus and the influence of one of those isolates on
synergism between Marek's disease vaccine viruses. Avian Pathol. 33:190. 2004.

15. Calnek, B. W. Effects of passive antibody on early pathogenesis of Marek's
disease. Infect Immun. 6:193-198. 1972.

 

l6. Calnek, B. W. Marek's disease--a model for herpesvirus oncology. Crit Rev
Microbiol. 12:293-320. 1986.

17. Calnek, B. W. Marek's disease virus and lymphoma. In: Oncogenic
Herpesviruses, lst ed. F. Rapp, ed. CRC Press, Boca Raton, FL. pp 103-143. 1980.

18. Calnek, B. W., H. K. Adldinger, and D. E. Kahn. Feather follicle epithelium: a
source of enveloped and infectious cell- free herpesvirus from Marek's disease. Avian
Dis. 14:219-233. 1970.

19. Calnek, B. W., R. W. Harris, C. Buscaglia, K. A. Schat, and B. Lucio.
Relationship between the immunosuppressive potential and the pathotype of Marek's
disease virus isolates. Avian Dis. 42:124-132. 1998.

20. Calnek, B. W., and S. B. Hitchner. Localization of viral antigen in chickens
infected with Marek's disease herpesvirus. J Natl Cancer Inst. 43:935-949. 1969.

21. Calnek, B. W., K. A. Schat, and J. Fabricant. Modification of Marek's Disease
Pathogenesis by In Ovo Infection or Prior Vaccination. In: Viruses in Naturally

Occurring Cancers. M. Essex, G. Todaro and H. zur Hausen, eds. Cold Spring Harbor
Laboratory. pp 185-197. 1980.

22. Calnek, B. W., K. A. Schat, L. J. N. Ross, W. R. Shek, and C.-L. H. Chen.
Further characterization of Marek's disease virus-infected lymphocytes. 1. In vivo
infection. Int J Cancer. 33:389-398. 1984.

23. Carrozza, J. H., T. N. Fredrickson, R. P. Prince, and R. E. Luginbuhl. Role of
desquamated epithelial cells in transmission of Marek's disease. Avian Dis. 17:767-781.
1973.

24. Cheng, H. H., H. Hunt, and M. Delany. Molecular Characterization of MD
Tumors with Respect to Clonality, Loss of Heterozygosity, and MDV Integration. In:

161

4th International Workshop on Molecular Pathogenesis of Marek's Disease Virus.
Newark, DE. p 6. 2006.

25. Cho, B. R. Dual virus maturation of both pathogenic and apathogenic Marek's
disease herpesvirus (MDHV) in the feather follicles of dually infected chickens. Avian
Dis. 21 :501-507. 1977.

26. Cho, B. R. Viremic responses of genetically susceptible and resistant chickens to
experimental infection with acute, mild, or both strains of Marek's disease herpesvirus.
Avian Dis. 19:67-74. 1975.

27. Cho, B. R., and S. G. Kenzy. Dual infection of chickens with acute and mild
strains of Marek's disease herpesvirus. Avian Dis. 17:390-395. 1973.

28. Cho, B. R., and S. G. Kenzy. Isolation and characterization of an isolate (HN) of
Marek's disease virus with low pathogenicity. Appl Microbiol. 24:299-306. 1972.

29. Cho, K. 0., N. Y. Park, D. Endoh, K. Ohashi, C. Sugimoto, C. Itakura, and M.
Onuma. Cytology of feather pulp lesions from Marek's disease (MD) virus- infected
chickens and its application for diagnosis and prediction of MD. J Vet Med Sci. 60:843-
847. 1998.

30. Churchill, A. E., L. N. Payne, and R. C. Chubb. Immunization against Marek's
disease using a live attenuated virus. Nature. 221 :744-747. 1969.

31. Cui, Z.-Z., L. F. Lee, J. L. Liu, and H. J. Kung. Structural analysis and
transcriptional mapping of the Marek's disease virus gene encoding pp3 8, an antigen
associated with transformed cells. J Virol. 65:6509-6515. 1991.

32. Cui, Z., and A. Qin. lmmunodepressive effects of the recombinant 38 kd
phosphorylated protein of Marek's disease virus. In: Current Research on Marek's
Disease. R. F. Silva, H. H. Cheng, P. M. Coussens, L. F. Lee and L. F. Velicer, eds.
American Association of Avian Pathologists, Kennett Square, PA. pp 278-283. 1996.

33. Cui, Z., D. Yan, and L. F. Lee. Marek's disease virus gene clones encoding
virus-specific phosphorylated polypeptides and serological characterization of fusion
proteins. Virus Genes. 3:309-322. 1990.

34. Cui, Z. Z., Z. Zhang, A. J. Qin, and L. F. Lee. Analyzing the H19- and T65-
epitopes in 38 kd phosphorylated protein of Marek's disease viruses and comparing

chicken immunological reactions to viruses point-mutated in the epitopes. Sci China C
Life Sci. 47:82-91. 2004.

35. Davidson, I., and R. Borenshtain. Novel applications of feather tip extractsfiom

MDV-infected chickens; diagnosis of commercial broilers, whole genome separation by
PFGE and synchronic mucosa] infection. FEMS Immunol Med Mic. 38:199-203. 2003.

162

 

36. Davidson, 1., and I. Skoda. The impact of feathers use on the detection and study
of DNA viral pathogens in commercial poultry. World Poultry Sci J. 61 :407-417, 516,
521, 525-526, 530-531, 536. 2005.

37. Delecluse, H. J ., S. Schuller, and W. Hammerschmidt. Latent Marek's disease

virus can be activated from its chromosomally integrated state in herpesvirus-
transforrned lymphoma cells. EMBO J. 12:3277-3286. 1993.

38. Endoh, D., M. Niikura, K. Hirai, N. Inagaki, M. Hayashi, K. Nakajima, and F.
Sato. Expression and purification of recombinant Marek's disease virus serotype 1
specific phosphorylated protein pp38 in E. coli. J Vet Med Sci. 56:823-826. 1994.

39. Fujimoto, Y., M. Nakagawa, K. Okada, M. Okada, and K. Matsukawa.
Pathological studies of Marek's disease. I. The histopathology on field cases in Japan.
Jpn J Vet Res. 19:7-26. 1971.

40. Gandon, S., M. J. Mackinnon, S. Nee, and A. F. Read. Imperfect vaccines and I
the evolution of pathogen virulence. Nature. 414:751-756. 2001.

41. Gimeno, I. M., A. L. Cortes, and R. F. Silva. Load of challenge Marek's disease
virus DNA in blood as a criterion for early diagnosis of Marek's disease tumors. Avian
Dis. 52:203-208. 2008.

42. Gimeno, I. M., R. L. Witter, A. M. Fadly, and R. F. Silva. Novel criteria for the
diagnosis of Marek's disease virus-induced lymphomas. Avian Pathol. 34:332-340.
2005.

43. Gimeno, I. M., R. L. Witter, H. D. Hunt, S. M. Reddy, L. F. Lee, and R. F.
Silva. The pp38 gene of Marek's disease virus (MDV) is necessary for cytolytic
infection of B cells and maintenance of the transformed state but not for cytolytic

infection of the feather follicle epithelium and horizontal spread of MDV. J Virol.
79:4545-4549. 2005.

44. Gimeno, I. M., R. L. Witter, H. D. Hunt, S. M. Reddy, and U. Neumann.
Differential attenuation of the induction by Marek's disease virus of transient paralysis

and persistent neurological disease: a model for pathogenesis studies. Avian Pathol
30:397-409. 2001.

45. Gimeno, I. M., R. L. Witter, H. D. Hunt, S. M. Reddy, and W. M. Reed.
Biocharacteristics shared by highly protective vaccines against Marek's disease. Avian
Pathol. 33:59-68. 2004.

46. Hsu, S. M., L. Raine, and H. Fanger. A comparative study of the peroxidase-
antiperoxidase method and an avidin-biotin complex method for studying polypeptide
hormones with radioimmunoassay antibodies. Am J Clin Pathol. 75:734-738. 1981.

163

47. Hsu, S. M., L. Raine, and H. F anger. Use of avidin-biotin-peroxidase complex
(ABC) in immunoperoxidase techniques: a comparison between ABC and unlabeled
antibody (PAP) procedures. J Histochem Cytochem. 29:577-580. 1981.

48. Ianconescu, M., H. G. Purchase, W. Okazaki, and B. R. Burmester. Influence of
natural exposure on response to subsequent challenge with virulent Marek's disease
virus. Avian Dis. 15:745-752. 1971.

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

50. Imai, K., and N. Yuasa. Pathogenicity of Marek's disease virus isolated in Japan
in two genetically different strains of chickens, PDL—l AND P-2 lines. In: Advances in
Marek's disease research. S. Kato, T. Horiuchi, T. Mikarni and K. Hirai, eds. Japanese
Assoc. on Marek's Dis., Osaka. pp 414-418. 1988.

51. Imai, K., N. Yuasa, H. Iwakiri, K. Nakarnura, H. Hihara, T. lshita, A. Inamoto,
l. Okamoto, K. Ohta, and M. Maeda. Characterization of very virulent Marek's disease
Viruses isolated in Japan. Avian Pathol. 21:119-126. 1992.

52. Islam, A., B. F. Cheetharn, T. J. Mahony, P. L. Young, and S. W. Walkden-
Brown. Absolute quantitation of Marek's disease virus and Herpesvirus of turkeys in

chicken lymphocyte, feather tip and dust samples using real-time PCR. J Virol
Methods. 132:127-134. 2006.

53. Islam, A., and S. W. Walkden-Brown. Quantitative profiling of the shedding
rate of the three Marek's disease virus (MDV) serotypes reveals that challenge with

virulent MDV markedly increases shedding of vaccinal viruses. J Gen Virol. 88:2121-
2128. 2007.

54. Islam, A. F. M. F., P. J. Groves, G. J. Underwood, and S. W. Walkden-Brown.
Dynamics of Marek's disease virus and herpesvirus of turkey shedding in feather dander
of broiler chickens. In: Proceedings of the 17th Australian Poultry Science Symposium,
Sydney, New South Wales, Australia, 7-9 February 2005. T. A. Scott, ed. Poultry
Research Foundation, Sydney. pp 105-108. 2005.

55. Islam, A. F. M. F., S. W. Walkden-Brown, A. Islam, G. J. Underwood, and P. J.
Groves. Relationship between Marek's disease virus load in peripheral blood

lymphocytes at various stages of infection and clinical Marek's disease in broiler
chickens. Avian Pathol. 35:42-48. 2006.

56. Jackson, C. A., P. M. Biggs, R. A. Bell, F. M. Lancaster, and B. S. Milne. The
epizootiology of Marek's disease. 3. The interrelationship of virus pathogenicity,
antibody and the incidence of Marek's disease. Avian Pathol. 5:105-123. 1976.

164

57. Jarosinski, K. W., B. L. Njaa, H. O'Connell P, and K. A. Schat. Pro-
inflammatory responses in chicken spleen and brain tissues after infection with very
virulent plus Marek's disease virus. Viral Immunol. 18:148-161. 2005.

58. Kenzy, S. G. Excretion of the Marek's disease agent by infected chickens. Vet
Record. 80:565-568. 1967.

59. Kreager, K. Industry Concerns Workshop. In: Current Research on Marek's
Disease. R. F. Silva, H. H. Cheng, P. M. Coussens, L. F. Lee and L. F. Velicer, eds.
American Association of Avian Pathologists, Kennett Square, PA. pp 509-511. 1996.

60. Lee, L. F., X. Cui, Z. Cui, I. Gimeno, B. Lupiani, and S. M. Reddy.
Characterization of a very Virulent Marek's Disease Virus Mutant Expressing the pp38
Protein from the Serotype 1 Vaccine Strain CVI988/Rispens. Virus Genes. 31 :73-80.
2005.

61. Lee, L. F., X. Liu, and R. L. Witter. Monoclonal antibodies with specificity for

three different serotypes of Marek's disease viruses in chickens. J .Immunol. 130:1003-
1006. 1983.

62. Lupiani, B., L. F. Lee, X. Cui, I. Gimeno, A. Anderson, R. W. Morgan, R. F.
Silva, R. L. Witter, H. J. Kung, and S. M. Reddy. Marek's disease virus-encoded Meq

gene is involved in transformation of lymphocytes but is dispensable for replication.
Proc Natl Acad Sci U S A. 101:11815-11820. 2004.

63. Meurens, F., F. Schynts, G. M. Keil, B. Muylkens, A. Vanderplasschen, P.
Gallego, and E. Thiry. Superinfection prevents recombination of the alphaherpesvirus
bovine herpesvirus 1. J Virol. 78:3872-3879. 2004.

64. Moriguchi, R., Y. F ujimoto, and H. Izawa. Chronological observations of
feather pulp lesions in chickens inoculated with Marek's disease virus. Avian Dis.
26:375-388. 1982.

65. Nair, V. Evolution of Marek's disease -- a paradigm for incessant race between
the pathogen and the host. Vet J. 170:175-183. 2005.

66. Nair, V. K., and L. N. Payne. Molecular pathogenesis of Marek's disease - recent
developments. Avian Pathol. 24:597-609. 1995.

67. Naito, M., K. Nakajima, N. Iwa, K. Ono, I. Yoshida, T. Konobe, K. Ikuta, S.
Ueda, S. Kato, and K. Hirai. Demonstration of a Marek's disease virus-specific antigen
in tumour lesions of chickens with Marek's disease using monoclonal antibody against a
virus phosphorylated protein. Avian Pathol. 15:503-510. 1986.

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

165

 

69. Okazaki, W., H. G. Purchase, and B. R. Burmester. The temporal relationship
between vaccination with the herpesvirus of turkeys and challenge with virulent
Marek's disease virus. Avian Dis. 15:753-761. 1971.

70. Payne, L. N. Pathogenesis of Marek's disease - A review. In: Oncogenesis and
herpesviruses. P. M. Biggs, G. de-The and L. N. Payne, eds. International Agency for
Research on Cancer, Lyon, France. pp 21-37. 1972.

71. Payne, L. N., J. A. Frazier, and P. C. Powell. Pathogenesis of Marek's disease.
Int Rev Exp Pathol. 16:59-154. 1976.

72. Payne, L. N., and M. Rennie. Pathogenesis of Marek's disease in chicks with and
without maternal antibody. J Natl Cancer Inst. 51:1559-1573. 1973.

73. Payne, L. N., and M. Rennie. The proportions of B and T lymphocytes in
lymphomas, peripheral nerves and lymphoid organs in Marek's disease. Avian Pathol.
5:147-154. 1976.

74. Rangga-Tabbu, C., and B. R. Cho. Marek's disease virus (MDV) antigens in the
feather follicle epithelium: difference between oncogenic and nononcogenic MDV.
Avian Dis. 26:907-917. 1982.

75. Reddy, S. M., B. Lupiani, I. M. Gimeno, R. F. Silva, L. F. Lee, and R. L. Witter.
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 fimction. P Natl Acad Sci
USA. 99:7054—7059. 2002.

76. Reddy, S. M., B. Lupiani, R. F. Silva, L. F. Lee, and R. L. Witter. Genetic
manipulation of a very virulent strain of Marek's disease virus. In: Current Progress on
Marek's Disease Research. K. A. Schat, R. W. Morgan, M. S. Parcells and J. L.

Spencer, eds. American Association of Avian Pathologists, Montreal, Canada. pp 55-58.
2001.

77. Richter, M., L. Schudel, K. Tobler, F. Matheis, A. Vogtlin, A. Vanderplasschen,
B. Costes, B. Spiess, and M. Ackerrnann. Clinical, virological, and immunological

parameters associated with superinfection of latently with FeHV-l infected cats. Vet
Microbiol. 2009.

78. Rickabaugh, T. M., H. J. Brown, D. Martinez-Guzman, T. T. Wu, L. Tong, F.
Yu, S. Cole, and R. Sun. Generation of a latency-deficient gammaherpesvirus that is
protective against secondary infection. J Virol. 78:9215-9223. 2004.

79. Rispens, B. H., H. Van Vloten, N. Mastenbroek, H. J. L. Maas, and K. A. Schat.
Control of Marek's disease in the Netherlands. 1. Isolation of an avirulent Marek's

disease virus (strain CVI 988) and its use in laboratory vaccination trials. Avian Dis.
16:108-125. 1972.

166

 

80. Ross, N., G. O'Sullivan, C. Rothwell, G. Smith, S. C. Burgess, M. Rennie, L. F.
Lee, and T. F. Davison. Marek's disease virus EcoRI-Q gene (meq) and a small RNA
antisense to ICP4 are abundantly expressed in CD4+ cells and cells carrying a novel
lymphoid marker, AV37, in Marek's disease lymphomas. J Gen Virol. 78:2191-2198.
1997.

81. Schat, K. A., and B. W. Calnek. Characterization of an apparently nononcogenic
Marek's disease virus. J Natl Cancer Inst. 60:1075-1082. 1978.

82. Schat, K. A., B. W. Calnek, and J. F abricant. Characterisation of two highly
oncogenic strains of Marek's disease virus. Avian Pathol. 11:593-605. 1982.

83. Schat, K. A., B. W. Calnek, and J. Fabricant. Influence of the bursa of Fabricius
on the pathogenesis of Marek's disease. Infect Immun. 31 :199-207. 1981.

84. Sevoian, M., D. M. Chamberlain, and F. T. Counter. Avian lymphomatosis. 1.
Experimental reproduction of the neural and visceral forms. Vet Med. 57:500-501.
1962.

85. Shek, W. R., B. W. Calnek, K. A. Schat, and C.-L. H. Chen. Characterization of
Marek's disease virus-infected lymphocytes: Discrimination between cytolytically and
latently infected cells. J Natl Cancer Inst. 70:485-492. 1983.

86. Silva, R. F ., S. M. Reddy, and B. Lupiani. Expansion of a unique region in the
Marek's disease virus genome occurs concomitantly with attenuation but is not
sufficient to cause attenuation. J Virol. 78:733. 2004.

87. Smith, M. W., and B. W. Calnek. Comparative features of low-virulence and
high-virulence Marek's disease virus infections. Avian Pathol. 3:229-246. 1974.

88. Stephens, E. A., R. L. Witter, L. F. Lee, J. M. Sharma, K. Nazerian, and B. M.
Longenecker. Characteristics of J MV Marek's disease tumor: A nonproductively

infected transplantable cell lacking in rescuable virus. J Natl Cancer Inst. 57:865-874.
1976.

89. Sung, H. W. Recent increase of Marek's disease in Korea related to the virulence
increase of the virus. Avian Dis. 46:517-524. 2002.

90. Tri Suwarni, Y. Poultry disease wipes out 2.8 million chickens. In: The Jakarta
Post. Jakarta, Indonesia. 2002.

91. Walling, D. M., A. L. Brown, W. Etienne, W. A. Keitel, and P. D. Ling.
Multiple Epstein-Barr virus infections in healthy individuals. J Virol. 77:6546-6550.
2003.

92. Witter, R. L. Avian tumor viruses: persistent and evolving pathogens. Acta Vet
Hung. 45:251-266. 1997.

167

 

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

94. Witter, R. L. Epidemiology of Marek's disease - a review. In: International
Agency for Research on Cancer. P. M. Biggs, G. de-The and L. N. Payne, eds., Lyon.
pp 111-122. 1971.

95. Witter, R. L. Increased virulence of Marek's disease virus field isolates. Avian
Dis. 41:149-163. 1997.

96. Witter, R. L. Induction of strong protection by vaccination with partially
attenuated serotype 1 Marek's disease viruses. Avian Dis. 46:925-937. 2002.

97. Witter, R. L. Very virulent Marek's disease viruses: importance and control. In:
XVIII World’s Poultry Congress. Japan Poultry Science Association, Nagoya, Japan. pp
92-97. 1988.

98. Witter, R. L., B. W. Calnek, C. Buscaglia, I. M. Gimeno, and K. A. Schat.
Classification of Marek's disease viruses according to pathotype: philosophy and
methodology. Avian Pathol. 34:75-90. 2005.

99. Witter, R. L., and I. M. Gimeno. Susceptibility of adult chickens, with and

without prior vaccination, to challenge with Marek's disease virus. Avian Dis. 50:3 54-
365. 2006.

100. Witter, R. L., and K. S. Kreager. Serotype 1 viruses modified by backpassage or
insertional mutagenesis: approaching the threshold of vaccine efficacy in Marek's
disease. Avian Dis. 48:768-782. 2004.

101. Witter, R. L., K. Nazerian, H. G. Purchase, and G. H. Burgoyne. Isolation from
turkeys of a cell-associated herpesvirus antigenically related to Marek's disease virus.
Am J Vet Res. 31:525-538. 1970.

102. Witter, R. L., and D. W. Salter. Vertical transmission of reticuloendotheliosis
virus in breeder turkeys. Avian Dis. 33:226-235. 1989.

103. Witter, R. L., J. M. Sharma, and A. M. Fadly. Pathogenicity of variant Marek's
disease virus isolates in vaccinated and unvaccinated chickens. Avian Dis. 24:210-232.
1980.

104. Witter, R. L., J. M. Sharma, and L. Offenbecker. Turkey herpesvirus infection in
chickens: induction of lymphoproliferative lesions and characterization of vaccinal
immunity against Marek's disease. Avian Dis. 20:676-692. 1976.

105. Witter, R. L., R. F. Silva, and L. F. Lee. New serotype 2 and attenuated serotype

1 Marek's disease vaccine viruses: selected biological and molecular characteristics.
Avian Dis. 31 :829-840. 1987.

168

106. Witter, R. L., J. J. Solomon, L. R. Champion, and K. Nazerian. Long-term
studies of Marek's disease infection in individual chickens. Avian Dis. 15:346-365.
1971.

107. Xie, Q., A. S. Anderson, and R. W. Morgan. Marek's disease virus (MDV)
ICP4, pp3 8, and meq genes are involved in the maintenance of transformation of
MDCC-MSBI MDV-transformed lymphoblastoid cells. J Virol. 70:1125-1131. 1996.

108. Yamaguchi, T., S. L. Kaplan, P. Wakenell, and K. A. Schat. Transactivation of
Latent Marek's Disease Herpesvirus Genes in QT35, a Quail Fibroblast Cell Line, by
Herpesvirus of Turkeys. J Virol. 74:10176-10186. 2000.

109. Yunis, R., K. W. J arosinski, and K. A. Schat. Association between rate of viral
genome replication and virulence of Marek's disease herpesvirus strains. Virology.
328:142-150. 2004.

 

169

lIu1111111111111fillflllllllllmlull