1.2.1»:
1&5...“

tn]

ya

5%

uni“?

... i... .
I

r .
WW. .
11.x \
. V.
2.3;...
.2 .2

.25. ‘3
in“ wary...

 

 

3.. I.
1

5...?
:2 ...w.;sio-
5.... 2 z. .4:

,_..Ia (.1253. E .

 

!.l..‘l..-.V..AVI
.: ~rr;. a . .1.

 

~.~—.~
x

TLJuL

 

q
5063
{ll/£5359 9 L!BP.. " P.Y
Michigan. St. .-.e
University

 

 

 

This is to certify that the
dissertation entitled

Phosphorylation of the transcriptional activator VP16 during
Iytic infection by Herpes simplex type 1

presented by
Soren Ottosen

has been accepted towards fulfillment
of the requirements for the

Ph. D. degree in Biochemistry and Molecular

Biology
M4

Major ProfesKJr’Qjfi s ig a ure
f -//- 4:

Date

 

 

 

MSU is an Affirmative Action/Equal Opportunity Institution

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

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/CIRCIDateDue.pG$-p. 15

PHOSPHORYLATION OF THE TRAN SCRIPTIONAL ACTIVATOR VP16 DURING
LYTIC INFECTION BY HERPES SIMPLEX TYPE 1

By

Soren Ottosen

A Dissertation
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Department of Biochemistry

2003

ABSTRACT

PHOSPHORYLATION OF THE TRANSCRIPTIONAL ACTIVATOR VP16 DURING
LYTIC INFECTION BY HERPES SIMPLEX TYPE 1

By

Seren Ottosen

Phosphorylation of a protein frequently serves to regulate the activity of the
protein. Because the understanding of its regulation can be the key to understanding the
function of the protein, efforts are often made to map the patterns of phosphorylation.
VP16, the activator of the immediate early genes of herpes simplex, type 1, is
phosphorylated during infection. This activator is viewed as a key factor not only in
activation of the immediate early genes, but also in the control of the virion host shut-off
function and in the assembly of virion. Because these functions occur at different stages
in infection, it raises the question of how the virus program manages them during
infection. Phosphorylation is a likely possibility and the identification of the
phosphorylation sites of VP16 during infection could improve our understanding of how
herpes simplex bypasses the host cells defense mechanisms and how it recruits the
cellular systems to ensure its own survival.

The main question addressed in this thesis was whether specific phosphorylation
events were involved in regulation of the fimction of VP16 during infection. In order to
answer that question, it was first necessary to map sites of phosphorylation during
infection. The mapping positively identified three sites of phosphorylation (Ser18,

Ser353, Ser452) and indicated a fourth (Ser411) as a possible site. A site that had been

reported to be phosphorylated outside of the context of infection (Ser375) was found not
to be phosphorylated under the conditions tested.

The functions of the Ser375 and Ser411 in infection were tested. A virus strain
with the Ser375Ala mutation displayed a significant reduction of immediate early
activation when protein synthesis was inhibited. In contrast, when protein synthesis was
allowed, the immediate early gene expression was only marginally affected, as compared
to the wild type. The strain did not exhibit discernible deficiencies in virus proliferation,
nor did the mutation of the site affect the phosphorylation of VP16 derived from infection
by this virus. It was concluded that while the mutation affects one mode of activation by
VP16, the IE genes are redundantly activated. The results favors a model in which
Ser375 plays a functional role but does not serve as a phosphorylation site, and is
therefore probably not as a site of regulation.

Virus strains carrying alanine, glutamate and threonine substitutions at Ser411
were found to be unaffected in proliferation. The strains were also found to support wild
type levels of immediate early gene activation, regardless of whether protein expression
was allowed in infection or not. The over-all phosphorylation of VP16 from these strains
was comparable to wild type. It was concluded that Ser411 was not an essential site for

VP16 function in infection, regardless of the phosphorylation state of the site.

To Karen

iv

ACKNOWLEDGMENTS

First, I would like to thank my thesis advisory committee for their support and
assistance. I would especially like to thank Steve, for sharing his scientific philosophy
and for showing that, at the heart of it, scientific inquiry is simple. Also for his patience

and his personal commitment to me and all the people in his laboratory.

Then, I would like to thank my fiiends and colleagues of the Triezenberg lab, as
well as other labs in the department and other departments. Peter, John, Rath, Eugene,
Lee, Marty, Billy, Sue, Mao, Kirsten, Dawn, Francisco, Dean, Heather, Amy, Karen and
so many more that I can not mention them all. I have treasured their friendship and their

support and I hope they have treasured mine.

Finally, I want to thank my family. My mother and my father, because of the
compassion and the ambition they have passed on to me and my siblings. My brother
and sisters, Annemette, Liselotte, Susanne and Peter, because they remind me that there
are things other than science to care about. I also want to thank my nieces and nephews,
Andreas, Jonas, Nikoline, Annekatrine, Aksel and Cecilie, for nothing in particular,
except that they are kids. Finally, I want to remember Stine and Astrid, for whom there

was so little time.

TABLE OF CONTENTS

LIST OF TABLES. ..................................................................................................... iiiv

LIST OF FIGURES ............................................................................................................ ix

KEY TO SYMBOLS AND ABBREVIATIONS ................................................................ x
CHAPTER ONE

INTRODUCTION AND LITERATURE REVIEW .......................................................... .1

The herpes Viruses .................................................................................................. 1

HSV-l Virion Structure .......................................................................................... 3

HSV—l Infection ...................................................................................................... 6

VP16 in HSV-l infection ...................................................................................... 11

Transcriptional Activation by VP16 ..................................................................... 19

Virion Host Shutoff ............................................................................................... 24

VP16 in Egress and cellular localization ............................................................... 25

VP16 homologs ..................................................................................................... 27

Regulation of VP16 ............................................................................................... 29
CHAPTER TWO

VP16 IS PHOSPHORYLATED IN THE INFECTED CELL AND IN VIRION ............ 31

Introduction ........................................................................................................... 31

Search for phosphorylation sites in vivo ................................................... 32

Results .................................................................................................................. 33

Computer-aided prediction of phosphorylation sites in VP16 .................. 33

Confirmation of VP16 phosphorylation in infected cells and in Virion 36
Peptide mapping of in viva [32P]-labelled VP16 isolated late

in infection using lysylendopeptidase C .................................................... 38

Peptide mapping ofin vivo [32P]-labelled VP16 isolated late

in infection using trypsin ........................................................................... 40

Phosphorylation of VP16 truncation mutants in viva ............................... 42

Peptide mapping of the phosphorylation sites of VP16 truncation

mutants using lysylendopeptidase C ......................................................... 45

Peptide mapping of the phosphorylation sites of VP16 truncation

mutants using trypsin ................................................................................ 50

Phosphoaminoacid analysis of VP16 from late in infection ..................... 51

Mass spectrometry of VP16 isolated late in infection ............................... 54

In vitro phosphorylation of the 375 CK2 site ............................................ 59
Materials and Methods .......................................................................................... 64

vi

Discussion ............................................................................................................ 78

CHAPTER THREE
MUTATIONAL ANALYSIS OF THE POTENTIAL PHOSPHORYLATION SITES
IN HERPES SIMPLEX INFECTION AT SERINE 375 AND SERINE 411 .................. 93
Introduction ........................................................................................................... 93
The function of neither Ser375 nor Ser411 has been adequately
described in the context of viral infection. ................................................ 93
Determining the effect of substitution of Ser375 and Ser411
on the proliferation and transcriptional phenotype in infection ................ 95
Results ................................................................................................................... 95
Function of Ser375 in vivo ........................................................................ 95
Phosphorylation of the EKC fragment of SJ 02 ....................................... 99
Phosphoaminoacid analysis of Ser375A1a .............................................. 100
Growth kinetics of SJ 02 ......................................................................... 102
IE gene expression assay ......................................................................... 104
Transcription of IE genes by VP16-Ser375Ala requires
protein translation .................................................................................... 107
Phosphorylation and function of Ser411 ................................................. 110
Generation of viral strains carrying point mutations at Ser411 .............. 110
Phosphorylation of VP16 lacking Ser411 ............................................... 115
Phosphoaminoacid complement in $011 ............................................... 116
Growth kinetics of SO] 1, 8012 and $013 ............................................. 117
Transcriptional activation of the IE genes by 801 1, 8012, SO] 3 ......... 122
Materials and Methods ........................................................................................ 126
Discussion ........................................................................................................... 131
CHAPTER FOUR
DISCUSSION AND FUTURE DIRECTIONS .............................................................. 143
Hypotheses .......................................................................................................... 143
Future research: Addressing the hypotheses experimentally .............................. 147
LIST OF REFERENCES ................................................................................................ 150

vii

LIST OF TABLES

Table 1: Phosphorylated and non-phosphorylated peptides detected by MS/MS
covering Ser18, Ser353, Ser375 and Ser452 ......................................................... 57

Table 2: Primers used for generating VP16 point mutations ........................................... 71

viii

Figure 1.
Figure 2.
Figure 3.
Figure 4.
Figure 5.
Figure 6.
Figure 7.
Figure 8.

Figure 9.

Figure 10.
Figure 11.
Figure 12.

Figure 13.

Figure 14.
Figure 15.
Figure 16.
Figure 17.

Figure 18.

LIST OF FIGURES

VP16 is a likely phosphorylation target ............................................................ 34
VP16 is labeled in the carboxyterrninal 80 amino acids. .................................. 37
Viral strains ........................................................................................... 44
VP16 is labeled outside of the activation domain ............................................ 47
VP16 is labeled on serines ........................................................................ 53
Peptide sequences identified by LC-MS/MS ................................................... 56
In vitro phosphorylation of the VP16 core domain ......................................... 62
Phosphorylation site conservation in herpes homologs ................................... 85
Alignment of VP16 and the Luman family ..................................................... 86
Loss of Ser375 does not affect the phosphorylation of VP16 ........................ 98
Loss of Ser375 does not affect the proliferation of HSV-1 .......................... 103
Loss of Ser375 has marginal effect on activation of the IE genes ................ 106
Loss of Ser375 severely diminishes IE gene activation in the absence of
protein synthesis ............................................................................................ 108
VP16-Ser411Ala, Glu and Thr are expressed and packaged in Virion ......... 114
Loss of Ser411 does not affect the phosphorylation of VP16 ...................... 119
Loss of Ser4ll does not affect the proliferation of HSV-1 .......................... 121
Loss of Ser411 does not affect the activation of IE genes ............................ 123

Loss of Ser411 does not affect the activation of IE genes in the absence
of protein synthesis ...................................................................................... 125

ix

CAT
CK2
CsTFA
DBD
DE
DMEM
FCS
HBM
hpi
HSV- 1
IE

KDa
LC-MS/MS, MS
moi
Vhs
VIC

VP16

Key to Symbols and Abbreviations

Amino acid

activation domain

chloramphenicol acetyltransferase
Casein kinase 2

Cesium trifiuoroacetate

DNA binding domain

Delayed early

Dulbecco’s modified essential medium
Fetal calf serum

HCF -1 binding motif

Hours post infection

herpes simplex, type 1

Immediate early

kilodalton

In-line liquid chromatography dual mass spectrometry
Multiplicity of infection

Virion host shutoff factor

VP16 induced complex

Virion protein 16

Chapter One

Introduction and literature review

Virion protein 16 (VP16) (vmw65, a-TIF) of herpes simplex type 1 (HSV-l) is a
transcriptional activator. Its carboxyterminal activation domain is frequently used in
studies of transcriptional activation and in in vivo protein interactions because it is active
in a variety of cells types and species ranging from yeast to human. In contrast, other
activities and domains of the protein have been less well studied. In particular, little is
known about regulation the of VP16 function during infection with HSV—l.
Phosphorylation of VP16 has been proposed to be part of a regulatory mechanism of
VP16, but this has not been shown conclusively. In this thesis, I will examine the

function of phosphorylation of VP16 during HSV-l infection.

The herpes viruses

Herpes viruses are among the largest known viruses. They carry genomes of 150
to 220 kbp, expressing more than 70 proteins in a temporally regulated cascade (Roizman
et al. 1993). In addition to the proteins encoded by their own genomes, the viruses use
factors and organelles from the host cell to produce and package virions. The Virion itself
is a complex structure, comprising dozens of viral proteins, as well as a membrane that
has been scavenged from the host cell. Herpesviruses also must evade the host organisms
systemic and cellular immune defense. Herpesviruses have been shown to suppress the

interferon response, (Leib 2002) intracellular transport of antigen, (Bauer et al. 2002)

mRNA processing (Sandri-Goldin 1994) and the expression of cytokines (Hukkanen et
al. 2002) to avoid destruction by the host. Yet, in order for the virus to proliferate
efficiently, the systems that govern gene expression, protein translation and modification
and nuclear transport must be regulated. The complexity of the virus program has
allowed the virus to adapt, making the herpesviruses robust viruses with a wide host
range, both in terms of cell types affected and in terms of the number of organisms that
are susceptible to one or more member of the herpes family (Roizman et al. 1993).

This complex interaction with the host cell makes these viruses an attractive
research subject. Understanding the general processes of viral infection could lead to
therapies against many different herpes-associated diseases. For instance, the human
herpes viruses are involved in human diseases ranging from the benignly annoying, like
the cold sores caused by herpes simplex type 1 and 2, (Roizman et al. 1993) to the
deadly, like human herpesvirus 8, which is involved in the development of Kaposi’s
sarcoma (Gandhi et al. 2002). The benefits would not only apply to human diseases.
Some herpes viruses can lead to severe losses in agriculture. Marek’s disease virus
causes $1 billion of losses in poultry farms worldwide, (Suszkiw 2001) due to death of
animals or failure to thrive (Payne et al. 2000). Similarly, the pseudorabies virus infects
and kills pigs, causing equally large losses (Mettenleiter 2000). Understanding the
weaknesses in the immune defense that the viruses use to evade detection and destruction
could also help in finding therapies against other pathogens.

In addition to understanding the diseases related to herpes infection, the study of

cellular responses to infection can help us understand the basic principles of the processes

of the host cell. By understanding how the virus uses these processes, we can also better
our understanding of how the host cell normally works, in the absence of infection.
Herpes simplex type I (HSV—l) is often used as a model for herpes viruses. The
simplex viruses (HSV-l and HSV—2) are the most widespread herpes viruses among
humans, with most studies finding approximately 70% of US citizens testing positive for
antigens against one or the other (Smith et al. 2002). The most common effect of HSV-1
infection is cold sores, which are annoying but hardly a major health threat (Roizman et
al. 1993). But the virus can also give rise to other, more serious and potentially lethal
diseases, both in normal and immunosuppressed individuals. Recurring ocular herpes
infections cause scarring of the cornea and are the most common cause of corneal
blindness worldwide (Liesegang 2001). Genital herpes is most often associated with
HSV-2, but can also be caused by HSV-l. Vaginal herpes can affect fertility and can be
transferred to infants in utero or during birth (Liesegang 2001). Congenital herpes
infections can cause severe birth defects such as micro-, hydro- or hydranencephaly,
blindness and premature birth (Liesegang 2001). Finally, and quite rarely, herpes can
cause herpes encephalitis (Kennedy et al. 2002). These infections are difficult to treat

and very frequently lethal.

HSV-l Virion structure

Like all herpes viruses, HSV-l is a double stranded DNA virus. It is classified as
a member of the alphaherpesvirinae subfamily, based on the structure of its genome.
HSV-l expresses approximately 70 open reading frames from a 152 kbp genome. The

viral genome is packaged in the Virion particle as linear double stranded DNA. It is

complexed with the polyamine sperrnine, which neutralizes the charge of the
polynucleotide-phosphate backbone and facilitates compact packaging of the DNA
within the capsid (Roizman et al. 1993). The viral DNA within the capsid is organized in
a structure resembling a spool with DNA spaced at 26 A (Zhou et al. 1999; Bhella et al.
2000). DNA is packaged into a preformed pro-capsid by the virally encoded factors UL6,
UL15, UL17, UL25, U128, UL32, and UL33 (White et al. 2003), some of which remain
associated with the mature Virion.

The mature capsid is an 1,250 A eicosahedral protein structure, comprising VP5,
VP26, VP23 and VP19C. The primary structural capsid component is VP5. VP5 forms
structures known as pentons and hexons. These are five- and six sided multimers of VPS,
which form the superstructure of the capsid. VPS protrudes outwards and interacts with
the tegument (Zhou et al. 1999; Bowman et al. 2003). VP5 is involved in capsid
maturation (Warner et al. 2000; Desai et al. 2003). VP23 and VP19C form structural
parts of the eicosadedron and are essential for capsid assembly. VP5 interacts directly
with the VP26 protein (Desai et al. 2003). This interaction is ATP-dependent in vitro
(Chi et al. 2000). The fimction of VP26 is unknown and it is dispensable for virus
proliferation (Desai et al. 1998). The structure and assembly of the capsid is reviewed in
(Homa et al. 1997) and in (Chiu et al. 2002).

The tegument is the compartment between the capsid and the envelope. The
mature Virion carries a number of viral proteins in this space. These proteins are
delivered to the host cells during initial infection. The tegument does not appear to be
organized, except for a limited interaction with VP5 of the capsid. The major proteins of

the tegument are VP16, VP22, VP13-14, VP1-2, the product of the US11 gene and the

Virion host shut-off factor (vhs). These factors are all involved in events occurring
during the initial stages of infection. Between 500 and 1000 copies of VP16 are present
in the tegument (Heine et al. 1974). The roles of VP16 in HSV-l infection, including
those associated with the formation of the Virion will be addressed later in this chapter.
Vhs is involved in degradation of mRNA early in infection. It has been suggested that
interaction of vhs with VP16 inhibits the activity of vhs later in infection (Knez et al.
2003). This interaction will be also addressed later in this chapter. The role of VP13/ 14
is less well understood. It is dispensable for growth but may be involved in activation of
the IE genes (Preston et a1. 1984; McKnight et al. 1987b; Zhang et al. 1991). A
spontaneous viral mutant, in which the genes for VP16 and VP13/14 have fused into one
open reading frame appears to have no growth defect (McKnight et a1. 1994) VP1-2 is an
essential protein, with a role in Virion maturation and egress (Desai 2000) and may
interact with the capsid protein VPS (McNabb et al. 1992) USll is an RNA binding
protein (Schaerer-Uthurralt et al. 1998) that is associated with the 60S proteasome
particle(Roller et al. 1996). and is involved in the viral repression of cellular response to
dsRNA (Poppers et al. 2000). VP22 is also an RNA binding molecule. It interacts
directly with VP16 (Elliott et al. 1995) and is capable of migrating between cells (Elliott
et al. 1997). It may also be involved in capsid maturation .

Recent reports have suggested that mRNA is also packaged in the Virion of HSV-
1 as well as other herpes viruses (Bresnahan et a1. 2000; Greijer et al. 2000; Sciortino et
al. 2001; Sciortino et al. 2002). The transport of these RNA molecules may occur

through the RNA binding proteins in the capsid. It can be demonstrated that the mRNAs

that are transported become expressed in the host cell, but the function of the mRNAs
remain unclear.

The Virion is enveloped by a lipid bilayer membrane. The lipid composition of
this envelope is consistent with that of membranes of the ER, the Golgi apparatus or
another cytoplasmic membrane (van Genderen et a1. 1994). In the mature Virion, it
contains a number of viral glycoproteins, including gD which has been shown to be
required for in cell surface binding and entry of the Virion into the host cell. The entry
into the cell is mediated by the HveA/HVEM receptor. The envelope is shed during entry
at the plasma membrane (Roizman et al. 1993). Three other glycoproteins, gB, gI-l, gL
are essential for membrane fusion during infection, although the mechanism is not
understood. Mutations in a fifth glycoprotein, gK, causes the formation of syncytial
infection. During these infections, infected cells fuse with adjacent cells, forming large

syncytia (Hutchinson et al. 1993; Dolter et al. 1994).

HSV—l infection

HSV-l infects most epithelial cells and neuronal cells in its natural host.
Infections in epithelial cells typically result in a lytic infection, leading to generation of
progeny virus and to the destruction of the infected cell. Viruses infecting neuronal cells

typically remain latent until reactivated by a trigger event (Roizman et al. 1993).

Lytic infection
The lytic pathway follows a typical viral gene expression cascade. The genes

required for the lytic pathway of the virus are expressed sequentially in three major

categories defined as the immediate early (IE) genes, the delayed early (DE) genes and
the late (L) genes. The expression of each set of genes is activated by viral transcription
factors and is regulated according to specific checkpoint events (Roizman et al. 1993).

The key events in the [E stage are the entry of the virus into the host cell and the
activation of the IE genes. During entry, the tegument proteins are released into the
cytosol and the capsid migrates to a nuclear pore complex and deposits the viral genome
into the nucleus. Injection of DNA from the capsid into the nucleus early in infection is
thought to occur through the nuclear pore complex in a manner requiring GTP (Sodeik et
a1. 1997; Ojala et al. 2000). VP16 is the primary activator of the IE genes during lytic
infection (Campbell et al. 1984; McKnight et al. 1987a; Triezenberg et al. 1988b). The
immediate early proteins are ICPO, ICP4, ICP22, ICP27 and ICP47. These proteins are
part of the initial steps of infection and they mount a very intricate and very efficient
subversion of the host cell defense system.

ICPO is a transcriptional activator which has been shown to activate the IE genes,
although no canonical response element has been identified. It is a RING-finger protein
and the RING-finger is required for transcriptional activation. In the absence of VP16
activation, ICPO is required for expression of the IE genes (Mossman et al. 1999). This
has led to the suggestion that ICPO belongs in a temporal class of its own, that might be
called pre—IE (Mossman et al. 1999). It is also involved in the dispersion of certain
nuclear structures know as NDlOs or PML nuclear bodies (Maul et al. 1993; Everett et
al. 1994) and can act as an E3 ubiquitin ligase (Van Sant et al. 2001). ICPO is capable of
ubiquitinylating p53 (Boutell et a1. 2003). One hypothesis proposes that the deletion of

ICPO reduces the infectivity of HSV-1 by lowering the probability of initiation of lytic

infection. In the absence of ICPO, the infecting virus is far more likely to enter a latent
state than in the presence (Everett 2000). This phenotype can be overcome by infection
at a high M01 or by expressing wild type ICPO from a separate promoter.

ICP4 is also a transcriptional activator, required for the expression of genes of all
three classes. Initially in infection, ICP4 is diffusely distributed throughout the nucleus,
but later in infection, it is found in nuclear structures adjacent to ICPO-containing
structures (Everett et a1. 2003). ICP4 acts as a transcriptional repressor on a few
promoters, including its own and that of ICPO (Lium et al. 1996). ICP4 is an essential
gene.

ICP22 is not an essential protein in infection of cultured cells (Ogle et al. 1999).
Some evidence suggests that ICP22 represses the expression of the IE genes, (Prod'hon et
al. 1996) but the most significant effect of deleting the ICP22 gene is the loss of
expression of DE and a subset of the L genes (Sears et al. 1985; Purves et al. 1993).
ICP22 appears to affect transcription by changing the phosphorylation pattern of the CTD
of RNAPII. The virally encoded kinase UL13 is also involved in this function (Rice et al.
1995; Long et al. 1999).

ICP27 is involved in the nuclear export of the intron-less HSV-l mRNAs, using
the TAP export pathway (Chen et al. 2002). ICP27 inhibits host cell mRNA splicing,
thereby preventing expression of the bulk of host mRNA (Sciabica et al. 2003). It is also
involved in expression of certain L genes. ICP27 is required for efficient viral DNA

replication (McMahan et al. 1990). It is an essential protein.

ICP47 is involved in avoidance of the host organisms immune response. It
prevents T-cell antigen transport and thereby prevents triggering the immune response
(Bauer et al. 2002).

The replication of the viral genome begins during the DE stage of infection. The
genes required for the replication, including the viral DNA polymerase, belong to the DE
group of genes. Although the recruitment of the VP16 activation domain to replication
origins in heterologous systems, such as polyoma, papilloma and baculovirus (Barn et a1.
1991; Bennett-Cook et al. 1991; Li et al. 1993; Pathakamuri et a1. 2002) can facilitate
replication, loss of VP16 binding sites at the HSV-l origin of replication does not affect
HSV—l replication (N guyen-Huynh et al. 1998).

The L genes are expressed after the replication of the viral genome has begun.
This group of genes is sometimes divided into two sub-groups termed leaky late and true
late (Roizman et al. 1993). The leaky late genes are expressed regardless of whether
viral DNA synthesis takes place. The true late genes do not get expressed unless the viral
genome is replicated. These genes include the structural genes of the capsid and the
glycoproteins of the membrane. They also include the tegument proteins, including
VP16, VP22 and vhs.

Expression of a number of host genes have been shown to increase during
infection with both wild-type HSV-l and with vhs, ich and VP16 mutants. The
expression of these genes may be the result of the cellular stress response. In the absence
of protein expression during infection, a group of interferon-responsive genes are turned
on (Mossman et al. 2001). Similar genes are also expressed in infection with virus

strains lacking VP16 and ICPO. The expression of these genes is part of a host defense

and would lead to destruction of both the virus and the cells that have been infected.
How the wild type virus inhibits the expression of those genes and prevents apoptosis of

the host cell remains to be determined (Khodarev et al. 1999; Taddeo et al. 2002).

Latency

HSV-l that infects neuronal cells enters latency. HSV—l typically infects the
trigeminal ganglia, but is also found in other sensory ganglionic neurons (Roizman et a1.
1993; Liesegang 2001). During latency, the viral genome exists as a nuclear episome (Su
et al. 2002) but the transcription of viral genes is largely repressed. The only HSV-l
RNA that is expressed from the latent HSV-l genome is the latency associated transcript
(LAT) (Roizman et al. 1993). The LAT transcript is partially complementary to the ICPO
gene and was initially thought not to be translated. This prompted the notion that the
LAT was an antisense repressor of ICPO, but the hypothesis was recently refuted (Burton
et al. 2003). Other data suggest that the LAT transcript is translated, and that the gene
products are present in the latently infected cell (Thomas et al. 1999; Thomas et al.
2002). They accumulate in the periphery of HSV-1 replication compartment and may be
important for both the establishment of and the emergence from latency.

HSV-l emerges from latency as a result of extracellular stress. Stressors include
heat shock and UV irradiation (Roizman et al. 1993). Luman, a cellular protein that
interacts with HCF-l in a manner similar to VP16, seems to enhance the establishment of
latency, possibly by sequestering HCF-l in the cytoplasm or by activating the LAT gene
(Lu et al. 2000a). The protein is also able to activate ICPO expression. This suggests that

Luman is also involved in the emergence from latency (Lu et al. 2000a). Also, deletion

10

of VP16 itself does not seem to affect establishment of latency or reactivation (Steiner et
al. 1990). However, the exogenous expression of not only VP16, but also of ICPO or
ICP4 will activate a latent virus in trigeminal ganglia (Halford et al. 2001).

The HSC-l genome may exist in several forms in the infected cell. Some reports
suggest that it exists as naked DNA, others that it is packaged as nucleosomes.
Micrococcal nuclease digestion of viral DNA from latent infection suggest that it is
predominantly nucleosomal (Deshmane et al. 1989). In contrast, micrococcal nuclease
digestion of lytic infection has shown that the viral genome is predominantly non-

nucleosomal (Leinbach et al. 1980; Lentine et al. 1990).

VP16 in HSV-l infection

VP16-mediated activation of the IE genes. The IE genes all have three
upstream regulatory elements in common, TAATGARAT elements, SP1 elements and
GA elements (Kristie et al. 1984; Jones et a1. 1985; Kristie et a1. 1986; Triezenberg et al.
1988a). The TAATGARAT element (R signifies a purine) is capable of regulating the IE
genes (Gaffney et al. 1985). It is also capable of binding VP16 in the context of Oct-1
and HCF-l (Kristie et al. 1989; Stem et al. 1989). Reviewed in. (Herr 1998; Wysocka et
al. 2003b).

In addition to the TAATGARAT, the IE promoters have at least two more cis
regulatory elements in common, the GA-box and the Spl binding site. Several published
results hint to a function of VP16 through these elements. The GA-element is involved in
VP16-mediated activation (Triezenberg et al. 1988b) as well as interact with HCF-l in

the cell (V ogel et al. 2000). Spl also interacts with HCF-l in vitro, (Gunther er al. 2000)

11

and the deletion of Spl sites from the IE promoters affect expression (Jones et a1. 1985;

Gu et a1. 1993).

VP16

The factor responsible for the transcriptional activation of the IE genes during
lytic infection is VP16. VP16 is a 490 a protein. VP16 belongs to the leaky late group
of HSV-1 genes. It is expressed late in infection, but does not require the replication of
the viral genome to be expressed. While VP16 is not strictly an essential gene, deletion
of the VP16 locus in the HSV-l strain 8MA does cause a severe reduction in
proliferation. Curiously, the strain does not appear to have any difficulties in replicating
its genome, (Weinheimer et al. 1992) suggesting that the strain is capable of expressing
viral genes required for replication. The phenotype of the null mutant is attributed to
defects in making mature virions. (Weinheimer et al. 1992; Mossman et al. 2000). In the
absence of VP16, empty capsids accumulate within the perinuclear space.
Transcriptional activation does not appear to be essential for viability, as viruses with
mutations in VP16 that reduce activation of the IE genes are viable, albeit with reduced
infectivity (Ace et al. 1989; Smiley et al. 1997; Tal-Singer et al. 1999).

The VP16 protein is typically divided into two major domains. The core domain
lies between amino acids 49 and 410 (Greaves et al. 1990; Liu et al. 1999). The
transcriptional activation domain AD resides at the carboxy terminus (aa 410-490)
(Triezenberg et al. 1988a; Werstuck et al. 1989). Specific functions have been defined
for each of these domains. VP16 does not have a traditional nuclear localization signal.

It appears to be transported to the nucleus through its interaction with HCF-l (La

12

Boissiere et al. 1999). The shuttling of this factor in and out of the nucleus by a factor
called HPIP may be crucial for the transport of VP16 to the nucleus (Mahajan et al.

2002)

The VP16 induced complex

Because VP16 does not have a DNA-binding domain, VP16 relies on a complex
comprising, in addition to VP16, the ubiquitous transcriptional activator Oct-l (Stem et
al. 1989) and HCF—l, a putative cell cycle regulator (Stern et a1. 1989; Wilson et al.
1993; Wilson et al. 1995a). Oct-1, a member of the POU family of transcription factors,
is a 100 KDa protein capable of activating both RNAPII and RNAPIII genes (Wysocka er
al. 2003a). Oct-1 activates responsive genes through the regulatory element

ATGCAAAT, also known as the octamer element.

Oct-l interaction

The minimal domain of VP16 required for interaction with Oct-1 is a 49-388
(Greaves et al. 1990). While the entire domain is required for the interaction, the
interaction surface appears to be much smaller. Single aa substitutions at aa 331, 371,
373 and 375 abolish interaction between VP16 and Oct-1 but do not affect interaction
with HCF-l (Greaves et al. 1990; Lai et al. 1997). The Oct-1 POU homeo domain is
sufficient for the interaction with VP16 (Kristie et al. 1989; Stem et al. 1989). Specific
mutations of the homeo domain map the interaction surface of Oct-l to the solvent
exposed groove between helix 2 and helix 3 of the homeo domain (Lai et al. 1992;

Pomerantz et a1. 1992). A glutamic acid within helix 2 is critical for the interaction, as

13

the replacement of it abolishes interaction with VP16. This model is supported by the
observation that VP16 does not bind Oct-2 (Amosti et al. 1993) or Oct-1 from the mouse
(Suzuki et al. 1993). Neither of these homologs of Oct-1 have a glutamic acid at this
critical position. The spacing of the hydrophobic residues suggests that a 370-375 may
form an amphipathic or-helix. The groove between these two helices is lined with
hydrophobic residues that may interact with the hydrophobic stretch of VP16 between
residue 375 and 385.

The interaction between Oct-1 and aa 370-375 of VP16 has been suggested to be
similar to the one that occurs between the two POU proteins MATal and MATaZ from
S. cerevisieae (Li et al. 1995; Lai et al. 1997). The crystal structure of this dimer bound
to its canonical DNA binding element has been solved (Li et al. 1995; Li et al. 1998). It
clearly reveals the formation of a distorted helix of MATor2 that interacts with the back
side of the MATal POU homeodomain. Additionally, the segment of this helix of
MATuZ can be replaced with the corresponding segment of VP16, suggesting that they
work in the same manner (Stark et al. 1999). Indeed, an alignment of the two motifs
reveal that the pattern of critical hydrophobic residues is conserved. A noticeable
difference is that the MATor2 helix is initiated by a proline residue, a very common
residue to occur at the Nl-position of the helix. In this model, the proline residue aligns
with Ser375 of VP16. Serines are also favored in the amino termini of helices, (Doig et
al. 1995) particularly when followed by an acidic residue two or three positions
carboxyterminal (Harper et a1. 1993; Doig 2002). This motif, incidentally, forms a
casein kinase 2 consensus site (S/TxxD/E). Additionally, the presence of a

phosphoserine at the N-terminus of an a-helix favors the formation of the helix (Andrew

14

et al. 2002). This all favors the model that the Oct-1-VP16 interaction is similar to the
interaction between Matal and Mata2. It also raises the possibility that the interaction

can be regulated by phosphorylation of VP16.

HCF-l interaction

The other interaction partner of the VIC complex is HCF-l (also known as C1 or
VCAF). HCF-1 is a very large protein of more than 2000 amino acid residues (Wilson et
al. 1993). The protein undergoes proteolytic cleavage afler translation to produce two
non-covalently linked subunits (Wilson et al. 1993; Wilson et al. 1995b). These subunits
remain tightly bound, despite the cleavage. The N-terminus of the protein (aa 1-380)
consists of a number of kelch or WD40 repeats. These repeats form a structure known as
a B-propeller (Adams et al. 2000). Initially, HCF-1 was identified as a necessary cellular
function for the interaction between VP16, Oct-1 and DNA (Katan er a1. 1990; Kristie et
al. 1990; Xiao et al. 1990; Stem et al. 1991). This function was mediated by a series of
related peptides, all expressed from one gene (Wilson et al. 1993).

HCF-1 was identified independently in a screen that selected for temperature
sensitive cell cycle defects in BHK cells (Goto et al. 1997). A serine to proline mutation
at residue 134 (Ser134Pro) of HCF (tsBN67) caused the cell to enter cell cycle arrest at
the non-permissive temperature (395°C). The HCF-1-VP16 interaction was found to be
sensitive to the Serl34Pro mutation and did not allow formation of the VIC complex
(Goto et al. 1997; Wilson et al. 1997). Serl34 lies within the kelch repeats and is

conserved between the repeats. In the proposed B-propeller, this residue falls within a

solvent exposed loop between two B-strands. The interaction surface of VP16 for HCF-1

15

was mapped by peptide competition and by mutagenesis (Shaw et al. 1995; Lai et al.
1997; Simmen et al. 1997). Two small motifs were identified, comprising aa 360-366
(REHAYS) and aa 385-3 87 (DDD). Mutations in these sites abolished VIC formation by
disrupting the interaction between VP16 and HCF-1 specifically but it did not appear to
affect the Oct-1 interaction. The cellular proteins Luman, Zhangfei (Zf) and HPIP were
later found to interact with HCF-1 through a similar motif (Lu et al. 1997; Lu et al. 1998;
Lu et al. 2000b; Mahajan et al. 2002). Luman and Zf are members of the basic leucine
zipper family, although their activities in the cell are not known. HPIP is a nuclear
shuttle factor that has a nuclear export signal and appears to shuttle HCF-1 between the
nucleus and the cytoplasm. The minimal consensus for the interaction between Luman,
Zf or HPIP and HCF-l was determined to be E/DHxY. This was termed the HCF-1
binding motif (HBM). Luman, Zf and HPIP also interact with HCF-1 in a manner that is
sensitive to mutations at residue 134. The three aspartates at aa 385-387 of VP16 are not
conserved in the Luman, Zf and HPIP. Other mutations in HCF-1 allow the protein to
maintain interaction with VP16 and Luman homologs, yet still cause cell cycle arrest
(Mahajan et al. 2000). This suggests that the loss of interaction between HCF-1 and
Luman is not required for cell cycle control. HCF-1 is present in most cell types with the
exception of brain tissue (Wilson et al. 1995a). In cells that express HCF-1, it is mostly
found to be nuclear with a notable exception of cells from trigeminal ganglia (Kristie et
al. 1999). When these cells are subjected to stress, similar to that which can re-activate
latent herpes genomes, HCF-1 rapidly changes cellular localization from the cytoplasm to
the nucleus. This suggests that the nuclear localization of HCF-l could be part of the

triggering mechanism for reactivation. Dawn Greensides has shown that VP16

l6

localization in the tsBN67 cell line is dependent on the localization of HCF-l (Greensides
2002). At permissive temperatures, when HCF-l is in the nucleus, VP16 is also found in
the nucleus. At non-permissive temperatures for tsBN67, both VP16 and HCF-1 were

found in the cytoplasm.

The VP16 Activation Domain

The activation domain of VP16 belongs to a group of activation domains known
as acidic activation domains (Hope et al. 1988). In VP16, the carboxyterminal activation
domain (aa 410-490) can be further divided into two subdomains known as the N-
subdomain (410-454) and the C-subdomain (455-490) (Regier et al. 1993). The VP16
activation domains were found to be very potent activators when fused to a heterologous
DNA binding domain (DBD) (Sadowski et al. 1988). Each of the two subdomains of this
activation domain is centered on a cluster of hydrophobic residues, one or more of which
are critical for transcriptional activation mediated by that domain (Cress er a1. 1991;
Sullivan et a1. 1998).

In addition to the carboxy terminal activation domain, some evidence suggests the
presence of a small activation domain in the very amino terminus of the protein. This
was discovered by sequence alignment to VP16 homologues from varicella zoster virus,
(Moriuchi et al. 1995) another herpes virus. Domain swapping experiments were
conducted in yeast and this domain, while fully competent to function in the context of a
DNA-binding domain, is not as potent as the carboxyterminal domain of VP16- It can be
removed from VP16, without affecting the activation potential of VP16 in transfection

assays (Triezenberg et al. 1988a). Yet, it is quite well conserved in the VP16 family and

17

whether it is a significant part of the VP16 activation function during any aspect of the

HSV-l lifecycle remains to be seen.

VP16 protein structure

The structure of the core domain of VP16 (aa 49-410) was determined by x-ray
crystallography (Liu et a1. 1999). This structure reveals a protein centered on a scaffold
made by two extended (at-helices, forming a large X. The remainder of the structural
elements, mainly shorter a-helices are arranged around this scaffolding in a structure that
resembles a chair. A stretch of DNA, adjacent to the response element, is proposed to lie
across the seat of this chair, with the other members of the VIC complex attached to the
side. Significantly, the domain of VP16 that contact both Oct-1 and HCF-1 is
unstructured in the crystal. Small angle X-ray scattering (SAXS) studies of full length
VP16 shows agreement with the crystal structure (Grossmann er al. 2001). It shows a
large central, kidney shaped molecule, with a wing on each side,. The dimensions of the
structure is in agreement with the crystal structure. The wings could correspond to the
activation domain and the VIC interaction domain, respectively.

The activation domain of VP16 was found to exist as a random coil in solution
(O'Hare er al. 1992). There was evidence that the domain did adOpt some degree of
helicity when interacting with TAFu32 (Shen et al. 1996a; Shen et al. 1996b; Uesugi et
al. 1997). This method of interaction led to a model termed the “induced fit”. In this
model, the VP16 activation domain adopts a structure only when interacting with its
target. The hypothetical strength of this mode of interaction is the ability to use a single

module to interact with a wide variety of targets.

18

Transcriptional Activation byVPlG

The prevalent model for the function of transcriptional activators like VP16 is that
they recognize a promoter element and then recruit other transcription factors to the
promoter. These factors in turn bring activities to the promoter that may be directly
involved in the generation of the RNA transcript or may facilitate formation of a
transcription complex by altering the promoter structure and environment. VP16 has
been shown to interact with factors from several of classes of transcription factors. These
can be broadly defined as general transcription factors, covalent histone modification
factors, such as histone acetyl transferases, and nucleosome remodeling complexes. In
addition to these interactions, there is emerging evidence for involvement of parts of the
proteasome and the ubiquitinylation pathway.

The general transcription factors act directly on the RNA polymerase, by
facilitating formation of the pre-initiation complex, initiation, promoter clearance or
elongation. In short, they are part of the physical mechanism of transcription. They are
required for accurate initiation by RNAP II in vitro (Orphanides et al. 1996). They may
act as recruitment devices or nucleation devices for the RNA-polymerase holoenzyme.
For a review of the activities of the general transcription factors in transcriptional
activation, see (Woychik et al. 2002).

Both the covalent histone modifiers and the nucleosome remodeling systems work
on the nucleosomal template. The covalent histone modifiers add or remove substituents
from histones. The initially discovered mechanism was acetylation of the histones, but
also phosphorylation, methylation and ubiquitinylation are part of the modifications that

take on histones. See (Berger 2002) for a review. Modification of histones affect gene

19

expression differentially, depending on the promoter context, even identical
modifications can have starkly different effects on different genes. The variety of
modifications that may occur at any given histone, as well as the gene specific effect of a
particular modification has led to the model of the histone code, where it is not the
individual modification per se, that regulates a promoter, but the combinatorial effect of
multiple modifications. This model seeks to explainihow all genes in an organism can
respond differently in response to stimuli, even though only a handful of modifications
takes place.

The last group of factors are the nucleosome remodeling complexes. These ATP-
dependent complexes are capable of rearranging the local nucleosome environment on a
given promoter in response to a transcription factor. The mechanism of the molecular
machines are not fully understood, although it has been suggested that they have a
helicase activity. For a review of the mechanisms of these factors, see (Narlikar et al.
2002)

It is clear that the activation of a particular gene does not solely rely on activation
through one of these mechanisms. In fact, all three groups may be recruited to an
activated promoter. The sequence by which this recruitment occurs appears to differ
between different promoters and is obscured by the fact that some complexes have
factors from more than one group, An example is the TAFs that are present in both the
TFIID as well as in HAT-containing complexes, such as the SAGA complex (Grant et a1.
1998). The ability of an activator to recruit factors from more than one of these groups of

factors may be key to its strength.

20

VP16 has been shown to interact with several of the general transcription factors.
These include TFIIA, TFIIB, TFIID and TFIIH. The prevalent model proposes that the
recruitment on one or more of these factors by VP16 serves to nucleate the assembly of a
transcriptional pro-initiation complex. This model is reviewed in (Woychik et al. 2002)
and (Ptashne et al. 1997). An example of how this could occur is the interaction of VP16
with TBP, the central factor of TFIID, is well established (Shen et al. 1996b; Nedyalkov
et al. 2003). This interaction appears to be mediated through a surface of TBP that is also
required for TBP dimerization and TBP-DNA interaction (Nishikawa et a1. 1997). This
surface can also interact with TAFuZSO (Liu et a1. 1998; Bagby et al. 2000). Several
interesting models have been created to resolve the problem of having multiple partners
interact through the same surface. Common to these models is that they suggests that the
interactions occur sequentially. VP16 could act both by recruiting TBP/TFIID to the
promoter and by disrupting either TBP-dimerization or TBP-TAFHZSO interaction, both
of which inhibit TBP-DNA interaction. The interaction between VP16 and TBP is
subsequently disrupted by the formation of a more favorable DNA-TBP complex, which
in turn forms the basis for the formation of the preinitiation complex. The events that
regulate the balance between binding affinities of the various interactions in TBP-
recruitment are under investigation (Liu et al. 1998; Bagby et a1. 2000; Pereira er al.
2001; Kobayashi et al. 2003). (See (Kobayashi et al. 2003) for an overview of several
models of the mechanism of activation by VP16 through TBP/TFIID.)
. The VP16 activation domain is a strong activator in yeast (Sadowski et a1. 1988).
The activation domain has been found to rely on certain factors for activation. Early on it

was found that VP16 was completely dependent on the ADA complex, some members of

21

which were Ada2, Ada3 and Gcn5 (Berger et al. 1992; Marcus er al. 1994; Barlev et al.
1995; Candau et al. 1997). It was clear that VP16 interacted directly with the complex,
and furthermore, not all activators were dependent on it. The discovery that Gcn5 was a
histone acetyl transferase (Brownell et al. 1996) added another layer of complexity to the
model, but also provided the model with an activity that could physically facilitate the
activation of transcription (Kuo et al. 1996; Kuo' et (231.1998).

VP16 has also been found to recruit chromatin remodeling factors. This
observation was initially considered a little curious, because the HSV-l genome was
reported not to be nucleosomal. It is possible that these factors have different functions

on naked DNA, or it is possible that HSV- DNA is nucleosomal.

Phosphorylation is a common mechanism of regulation in transcription.

The regulation of a function by affecting interaction between factors required for
a process is conceptionally quite simple. The function can be regulated by altering the
strength of the interaction between the factors and this can be achieved by modifying the
interaction surface in some manner. Covalent modification of one or both of the
interaction surfaces can positively or negatively affect the interaction. There are many
examples of this taking place in transcription. In the following examples, the
modification is phosphorylation, but other modifications, such as acetylation or
methylation, are also commonly found in transcriptional regulation.

An example of regulation of a transcription factor by phosphorylation is that of
p53. Here, the phosphorylation serves to release the transcription factor from a inhibitor,

MDM2. p53 is released from its inhibitor when p53 is phosphorylated at ser20. This

22

phosphorylation stabilizes the protein, allowing it to activate DNA damage response
genes (Shieh et al. 1997; Ashcroft et al. 1999; Unger et al. 1999a; Unger et al. 1999b).
In the case of NFKB, the transcription factor is released from its inhibitor by
phosphorylation of the inhibitor, IKB. Here, the IKB-NFKB complex is located in the
cytoplasm and not until IkB is phosphorylated does NFKB migrate to the nucleus (J ans et
al. 1996). These are both examples of how transcription factors can be activated when
phosphorylation abolishes interaction with a inhibitory factor.

The opposite effect can be observed in the interaction between CREB and the
CREB-binding protein (CBP). In this case, the phosphorylation of the kinase inducible
domain of CREB facilitates the interaction with the KIX domain of CBP. (Chrivia et al.
1993; Parker et al. 1996). The phosphorylation occurs directly in the interaction surface
between CREB and CBP, making the interaction favorable. The increased affinity allows
the CREB to recruit the CBP to the promoter. CBP in turn activates transcription through
its HAT activity.

It is also possible that phosphorylation can act as a signal for subsequent
modification. For instance, acetylation of p53 at Lys382 is stimulated by
phosphorylation at Ser46. The acetylation of the lysine then blocks the ubiquitinylation
of the same residue, which would otherwise mark p53 for proteasome-mediated

degradation.
Ubiquitinylation and the proteasome in transcription.

Recently, another aspect of transcriptional activation has emerged. Tansey and

colleagues noticed that activation domains typically look like degrons, the motifs that

23

signal degradation by the ubiquitin mediated pathway. They then discovered that Met30,
a yeast ubiquitin ligase, was required .for activation by the VP16 activation domain in
yeast (Salghetti et al. 2000; Salghetti et al. 2001). In the absence of Met30, not only was
VP16 unable to activate, it was also more stable. Furthermore, the addition of an in-
frame ubiquitin-moiety restored the activation by the VP16 activation domain in a Met30
null strain. Adding to this was the observation that'parts of the 19S proteasome particle
appeared to be required for activation by the yeast Gal4 activator (Ferdous et al. 2001;
Gonzalez er al. 2002). The striking requirement for the involvement of parts of the
ubiquitin and proteasome pathways for activation by certain activators in yeast may be

reflected in the action of these and similar activators in mammalian cells.

Virion host shut-off

Another protein that interacts with VP16 in vitro is the Virion host shut-off factor
(Vhs), the product of the HSV-l gene UL41 (Roizman et al. 1993). Vhs is not essential
for viability (Fenwick et al. 1990). Vhs is an RNase that specifically degrades mRNAs
(Zelus er al. 1996; Elgadi et al. 1999; Everly et al. 2002). It is packaged in the tegument
of the Virion and upon infection starts degrading both host and newly synthesized viral
mRNA (Smiley et al. 2001). The purpose of this factor appears to remove host cell
transcripts presumably to conserve cellular resources for the translation of viral proteins.
It also serves to specifically circumvent the host cells transcriptional response to the
pathogen (Matis et al. 2001). While vhs also degrades the viral mRNA, the expression
level of the IE and DE genes is so high that sufficient mRNA survives to provide proper

protein synthesis for the subsequent steps in the infection cascade. At late stages in

24

infection, however, the RNase activity in the cell is decreased (Lam er al. 1996). Vhs
function may need to be regulated to prevent depletion of viral transcripts late in
infection. Interestingly, the vhs activity in a virus deleted for VP16 is very robust even at
late times, (Lam et al. 1996) resulting in decreased yield of infectious particles. A double
mutant, where both VP16 and vhs are deleted has a less severe phenotype than either of
the deletion mutants by themselves. A model where VP16 plays a role in controlling vhs
could explain this double phenotype. The model that VP16 is involved in the control of
vhs is further supported by in vitro results that VP16 (1-344) is capable of interacting
with a small domain of vhs (aa 310-330) (Smibert er al. 1994; Schmelter et al. 1996;
Knez et al. 2003). Two point mutations have been identified in VP16 that abolish the
interaction between VP16 and vhs. These lie at lysine 343 and leucine344 (Knez et al.
2003). Vhs is a competitor for the formation of the VIC complex (Smibert et al. 1994).
While Oct-1 and vhs have distinct interaction surfaces on VP16, they appear to overlap.
This fits with a model that suggests that VP16 interacts with Oct-1 during IE to facilitate
IE gene expression. At this time, it has no effect on vhs function. Conversely, at late
time, VP16 no longer interacts with Oct-l, but it does interact with vhs and subsequently
inhibits its RNase activity. The determinants that alter the function of VP16 are not

known.

VP16 in egress and cellular localization:
There is some evidence that VP16 also affects the packaging and egress of the
maturing Virion. The interaction with another tegument protein, VP22, may be required

for this function. Tegument body formation and cellular co-localization of VP16 and

25

VP22 in infection relies on hydrophobic residues in the carboxyterminal AD of VP16 that
are also critical for the activation potential of VP16. The interaction also relies on
determinants in the putative amino terminal AD of VP16 (Elliott et al. 1995). VP16 is
found associated with the maturing capsid in the rat dorsal ganglion. (Miranda-Saksena
et al. 2002). This is not unexpected, as VP16 eventually will become packaged in the
mature Virion. There is also evidence that VPl6iis‘ functionally required for egress.
Mossman and colleagues show that HSV-l fails to package DNA in a VP16 deletion
strain. They also show that the phenotype is partially complemented by the deletion of
the vhs gene (Mossman et al. 2000). But in either case, immature virions accumulate in
and around the perinuclear space. Mature Virion or even partially formed Virion particles
do not appear in the cytoplasm per se. This suggests that VP16 is required for the
maturing Virion to traverse the nuclear membrane, but the mechanism has not been

established.

Cellular localization and function:

Like other viral factors, VP16 occupy distinct domains within the infected cell.
These domains change as the cell goes through the infectious cycle. Initially, newly
synthesized VP16 accumulate in distinct spots in the nucleus. VP16 then appears along
the nuclear periphery. The intranuclear spots grow bigger and merge as the infection
progresses. Late in infection, VP16 is spread through the nucleus as well as throughout
the cytoplasm (Greensides 2002). An interesting observation by O’Hare and colleagues
is that the localization of nuclear lamin B and the lamin B receptor changes dramatically

during infection (Scott et al. 2001). They argue that the lamin network, which serves as

26

a rigid nuclear scaffolding, is too finely meshed to allow the passage of newly assembled,
equally rigid viral capsids. The pattern of the redistributed lamins is surprisingly similar
to the distribution of VP16 for an extensive period of the infection, suggesting that VP16

could actively be involved in the egress of the capsids.

VP16 homologs

The core domain of VP16 is generally conserved between other herpesvirus
homologs. This corresponds to a 50-410 of HSV-1 VP16. The carboxy terminal
activation domain is partially conserved in some virus species, like pseudorabies virus,
equid herpes virus 1 and bovine herpes virus. 1. In other species, like the closely related
varicella zoster virus, the activation domain is entirely absent. Other domains that are
less conserved is the amino terminus and, surprisingly, small domains flanking the
residues required for interaction with Oct-1 and HCF-1. Even so, the VIC interaction

domains are strictly conserved in closely related species.

Mutant viral strains

While the mutagenesis of VP16 in the context of expression plasmids allows
focus on the transcriptional function of VP16, it excludes the effect that mutants may
have on the viral life cycle. A number of mutant HSV-l strains have been described and
their behavior in infection may help us understand the complex function of VP16 in
infection.

The VP16 deletion mutant, 8MA was mentioned earlier. It has had the entire

VP16 ORF removed and replaced with a LacZ gene. This viral mutant can only

27

proliferate at a high moi and in the presence of the differentiating agent
hexamethylenebisacetamide (HMBA) (Weinheimer et al. 1992). It does appear that the
virus is fully capable of replicating its genome. How HMBA facilitates the proliferation
of 8MA is not understood.

Truncations of the activation domain of VP16 affect the activation of the IE genes
(Smiley et al. 1997; Tal-Singer et al. 1999). Both the RPS and the V422 strains exhibit
high particle to pfu ratios. This indicates that the viruses are fully capable of generating
progeny virion, but that the virion do not easily give rise to a productive infection.
Furthermore, both RPS and V422 are capable of delivering their genomes to the nucleus
of the host cell, indicating that the proliferation deficiency lies in the initiation of the viral
gene expression cascade. Like 8MA, V422 proliferation improves in the presence of
HMBA. IE gene expression is also partially complemented (Smiley et al. 1997). The
In1814 strain contains an insertion of four amino acids at aa 379. This short insert
prevents VP16 from forming the VIC and there was a significant reduction in IE gene
expression during infection. The strain is also unable to generate plaques, indicating that
little if any progeny is generated (Ace et al. 1989). Curiously, the defects of in1814
could be partially overcome by infection at a high moi, by adding HMBA to the medium
or by overexpressing ICPO, (Jamieson et al. 1995) one of the IE gene products. The
activity of the HMBA could be repressed by other cytodifferentiating agents (McFarlane
et al. 1992; Preston er al. 1998). RP3, a strain that had the carboxy terminal subdomain
(11454-490) deleted was only mildly affected. The internal deletion of the amino terminal
subdomain of the activation domain (11412-453) had a an intermediate phenotype, both in

terms of growth and IE gene activation (Tal-Singer et a1. 1999).

28

Regulation of VP16:

Throughout infection with HSV-l, VP16 participates in different processes.
Activation of the IE genes, vhs interaction and possible inhibition, packaging and nuclear
egress of the maturing virion, possibly through interaction with VP22. In addition, a
population of VP16 molecules must be dissociated from all activities in the cell and
become packaged into the virion. There is evidence. that these activities can be mutually
exclusive. VP16 can not bind both vhs and Oct-l, and the virus must be programmed to
ensure that separate populations of VP16 are partitioned for those two firnctions. That
partition must be regulated through the life cycle of the virus. In a similar manner,
because VP22 interact with the AD of VP16, it is possible that VP22 can interfere with
the activation of VP16 responsive genes. That raises the question of whether regulation
of the interaction between VP16 and VP22 is necessary, and consequently of whether

phosphorylation could be involved in such a regulatory step.

Phosphorylation of VP16.

A way for the virus to regulate the function of VP16 is through phosphorylation.
VP16 was shown to be phosphorylated early on (Gibson et al. 1974; Lemaster et al.
1980; Meredith et al. 1991). These studies found that many HSV-l proteins were
phosphorylated, including VP16. VP16 was found to be phosphorylated both in the
virion and in infected cells. The next investigation was done in the laboratory of Peter
O’Hare. They expressed VP16 in HeLa cells, in the presence of [32P]-orthophosphate
and confirmed that VP16 became phosphorylated (O'Reilly 2! al. 1997). They then

proceeded to investigate the kinases for which bacterially expressed VP16 could serve as

29

a substrate. They focused on the aa 1-410 fragment of VP16. VP16 was discovered to be
a substrate for protein kinase A (PKA) and casein kinase 2 (CK2). A nuclear extract
from HeLa cells could also phosphorylate VP16. Lysylendopeptidase digestion
suggested that one phosphorylation site fell within the carboxyterminal 120 aa. In an
attempt to identify a phosphorylation site, they mutated a prominent CK2 site within this
region, Ser375. The alanine substitution completely abolished VIC formation and IE
gene activation by VP16. No reduction in the phosphorylation level of VP16 was
observed. It was concluded that if Ser375 was a phosphorylation site, there had to be
additional sites, which would explain why the loss of the serine did not abolish
phosphorylation. They surmised from a small change in migration of the mutated protein
as compared to the wild type, that the Ser375 site was in fact a phosphorylation site.
Meredith and colleagues isolated VP16 from virion and found it to be a substrate for
PKA, but not for CK2 (Morrison et al. 1998). They also saw that VP16 became
phosphorylated early in infection, when the cells were infected at a very high moi.

In this study the patterns of phosphorylation of the herpes simplex l (HSV l)
virion protein 16 (VP16) during infection were examined. It was described how it differs
from studies of VP16 in transfected cells. The significance of the discrepancy between
the two and the ramifications it has on the way we look at HSV-l infection in particular

and transcriptional regulation in general were addressed.

30

Chapter Two

VP16 is Phosphorvlated i3 the Infected Cell and in Virion

Introduction

The purpose of this project was to study the phosphorylation of VP16 in infection.
The goal was to identify any regulatory role the phosphorylation might serve in infection
by HSV-l. To examine the role of phosphorylation, it was necessary to determine if the
phosphorylation pattern of VP16 in infected cells was similar to the pattern previously
shown for VP16 expressed from a transfected plasmid (O'Reilly 8! al. 1997)

Phospho-VP16 can be detected both in infected cells and in purified virion
(Gibson et al. 1974; Lemaster er al. 1980; Wilcox et al. 1980; Morrison et al. 1998).
O’Hare and colleagues had identified one phosphorylation site, Ser375, in VP16 that was
expressed from a transfected plasmid in HeLa cells (O'Reilly er al. 1997). They also
reported that Ser375 was phosphorylatable in vitro by casein kinase 2 (CK2).
Substituting the serine at this site with alanine led to a loss of transcriptional activation by
VP16 and to loss of the ability to interact with the canonical response element in vitro.
Both of these effects were attributed to the loss of VP16 interaction with Oct-1 (Lai et al.
1997). The interaction with HCF-1 seemed to be maintained, albeit at a much reduced
affinity(Greaves et al. 1990; Lai et al. 1997). Meredith and colleagues had examined the
gross phosphorylation of a number of HSV-1 virion components, both in infected cell
extracts and in purified virions and had verified that VP16 is phosphorylated. They
demonstrated that detergent-purified heat-inactivated tegument-derived VP16 was readily

phosphorylated with PKA in vitro (Morrison et a1. 1998). In contrast to O’Hare and

31

colleagues, Meredith and colleagues saw no phosphorylation of VP16 by CK2 in vitro.
VP16 has several good consensus CK2 phosphorylation sites, but the lack of detectable
phosphorylation in the in vitra CK2 phosphorylation assay was not necessarily contrary
to the observations by the O’Hare group. If virion borne VP16 was already quantitatively
phosphorylated at the CK2 phosphorylation site, labeling of VP16 purified fi'om virion
would be blocked. This complication would not havearisen in the studies by O’Hare and

colleagues where bacterially expressed VP16 was used in the CK2 assay.

The search for phosphorylation sites in viva. Before addressing the regulatory
aspect of phosphorylation, it was first necessary to identify the phosphorylation sites
during infection, which may or may not be similar to that of VP16 expressed from a
transfected expression vector. The phosphorylation pattern of VP16 might be
significantly different in the context of an HSV-l infection than when expressed from a
plasmid, due to the presence of regulatory factors and kinases in the virus. The cellular
response to the infection could also cause VP16 to be modified differently.

In order to examine the phosphorylation of VP16, VP16 was purified from
infected cells and from virion. “The phosphorylation sites were examined using
phosphopeptide mapping, phosphoaminoacid analysis and mass spectrometry. Isolated
VP16 from wild type as well as mutant HSV-l strains was used in the experiment. These
mutants had either partially or completely truncated activation domains. Phosphorylation
was found at Serl8, Ser353 and Ser452. The likely phosphorylation of Ser411 was also
inferred. There was no evidence that Ser375 was phosphorylated. Furthermore, Ser375

was found not to be required for the in vitra phosphorylation of VP16 by CK2.

32

mate

Computer-aided prediction of phosphorylation sites in VP16. The first step in
identifying the sites of phosphorylation of VP16 was to see whether there were any
theoretically strong consensus sites of phosphorylation in the protein. The amino acid
sequence of VP16 was scanned for potential phosphorylation sites using a web-based
prediction server (Blom et a1. 1999). The Phosphobase phosphorylation prediction server
identifies consensus phosphorylation sites in a given polypeptide, based on known kinase
consensus sites. It also uses a database of previously characterized phosphoproteins to
generate an algorithm that compares the local primary structure of all serines threonines
and tyrosines in the query protein to the local primary structure of the phosphorylated
serines, threonines and tyrosines of the phosphoproteins in the database. The algorithm
then assigns each residue in the analyzed sequence a value between 0 and 1 according to
their predicted phosphorylation potential. A low number indicates a poor substrate, a
high number a good substrate.

The results of the prediction for VP16 suggested that a significant number of the
serines, threonines and tyrosines in the sequence might serve as phosphorylation sites in
VP16 (Figure 1). Panel A shows a graphical representation of the results of the
prediction. The sequence positions of the residues in the query protein are plotted on the
X-axis. The phosphorylation potential of each residue is indicated by the length of the
bar at that position. Serine residues are indicated by a blue bar, threonines by a green and
tyrosines by a red bar. The grey line indicates an arbitrary prediction cut-off. Residues

that fall below this line are dissimilar to the phosphoproteins in the database and are

33

NetPhos 2.0: Predicted Phosphorylation Sites in Sequence
Serine
Threonine

 

 

Tyrosine
Threshold

 

 

Phosphorylation Potential

 

 

0 - ‘ ‘ _
0 50 100 150 200 250 300 350 400 450 490
Sequence Position

B

1 MDLLVDELFA DMNADGASPP PPRPAGGPKN TPAAPPLYAT GRLSQAQLMP
51 SPPMPVPPAA LFNRLLDDLG FSAGPALCTM LDTWNEDLFS ALPTNADLYR
101 ECKFLSTLPS DVVEWGDAYV PERTQIDIRA HGDVAFPTLP ATRDGLGLYY
151 EALSRFFHAE LRAREESYRT VLANFCSALY RYLRASVRQL HRQAHMRGRD
201 RDLGEMLRAT IADRYYRETA RLARVLFLHL YLFLTREILW AAYAEQMMRP
251 DLFDCLCCDL ESWRQLAGLF QPFMFVNGAL TVRGVPIEAR RLRELNHIRE
301 HLNLPLVRSA ATEEPGAPLT TPPTLHGNQA RASGYFMVLI RAKLDSYSSF
351 TTSPSEAVMR EHAYSRARTK NNYGSTIEGL LDLPDDDAPE EAGLAAPRLS
401 FLPAGHTRRL STAPPTDVSL GDELHLDGED VAMAHADALD DFDLDMLGDG
451 DSPGPGFTPH DSAPYGALDM ADFEFEQMFT DALGIDEYGG

Figure 1: VP16 is a likely phosphorylation target. The VP16 protein sequence was exam-
ined using NetPhos, a web based phosphorylation site prediction server. A: The predict-
ed phosphorylation potential of serines, threonines and tyrosines of the VP16 protein are
plotted as bars as a function of their position in the VP16 protein. The potential is report-
ed as a number between 0 and 1. A number above 0.5 (indicated by the horizontal grey
line) reflects a likely phosphorylation site. B: The sequence of the VP16 protein. The
sites that were predicted by the NetPhos server to be likely phosphorylation substrates are
indicated in bold.

34

rarely found to be phosphorylated in viva. Residues with values at or above 0.5 are
predicted to be structurally similar to the phosphorylation sites in the database. As the
similarity increases the probability that the site in the query protein is phosphorylated
increases and the residue is given a higher score.

Panel B shows the sequence of VP16 where the sites that were predicted to be
potential phosphorylation sites are indicated in bold face type. 27 residues received a
score of 0.5 or better. Several sites were given scores very near 1.0, indicating a very
high likelihood of these sites being good kinase substrates. The predicted
phosphorylation sites included threonines, serines and tyrosines. While all of these sites
were theoretically likely phosphorylation sites, the likelihood that all of these sites were
in fact phosphorylated during infection was much lower. For instance, some of the sites
that had received a high score were not found to be phosphorylated by O’Hare and
colleagues. None of the predicted threonines and tyrosines were found to be
phosphorylated, nor were the two large endolysylpeptidase C fragments spanning residue
29-103 and 104-343. The server predicted eight sites in those two fragments alone.
Ser375 received a high score, but it was not the only potential site in the 120 aa
carboxyterminal fragment. There are six residues with scores above 0.5, two of which,
Ser4ll and Ser419, have scores higher than Ser375. Subsequently, it was to be expected
that most of the predicted sites in VP16 in infection would remain unphosphorylated.
The prediction did show that there were many potential sites of phosphorylation in VP16
and that a de nava search for phosphorylation during infection might reveal

phosphorylation events specific to infection.

35

Confirmation of VP16 phosphorylation in infected cells and in virion. In
order to start the identification of the phosphorylation sites of VP16 in infection, it was
necessary to confirm that the previously published experiments could be repeated with
the reagents available. It had been previously observed that VP16 isolated from virions
was phosphorylated and that phosphorylation of incoming VP16 occurred within 30
minutes of infection (Morrison et al. 1998). It wasialso found to be among the major
phosphoproteins expressed late in infection (Gibson et al. 1974).

To confirm that VP16 was phosphorylated in infection and in the virion, [32F]—
labeled VP16 was immunoprecipitated from infected HeLa cell lysates and from lysates
of purified virions. HeLa cells were infected with wild type HSV-l (KOS) at a
multiplicity of infection (moi) of 5. [32P]-orthophosphate was added at 1.5 hours post-
infection (hpi). Either the infected cells were lysed at 8 hpi or virion were harvested from
the cells at 14 hpi. VP16 was isolated from either sample by immunoprecipitation with
the monoclonal antibody LPl (McLean et al. 1982). The antibody recognizes an epitope
within the amino terminal end of VP16. Proteins were eluted from the
immunoprecipitation pellet at pH 12.5 and separated on a 10% SDS-PAGE gel. The
contents of the gel were transferred to a nitrocellulose membrane. The 32P-labelled
proteins were visualized by autoradiography. The results are shown in figure 2. In panel
A, a radiolabeled band of the expected mobility (apparent molecular weight 65 KDa) was
observed in the immunoprecipitation pellet both at 8 hours (8h) and in the virion. The
band is not observed in immunoprecipitation of lysate of non-infected cells from a

parallel 32P-labelling (mock). The radiolabeled species co-migrates with VP16, as

36

 

a)
‘5
3.» c:
s = 8 e e
G) v-a
E O > in I—
200 29.
97.4
18.4 'h
n
68 14.3-
43 6.2-
29
3-

 

 

 

 

 

Figure 2: VP16 is labeled in the carboxy terminal 80 amino acids. HeLa cells were
infected with KOS in the presence of [32P]-orthophosphate. Cell lysates were prepared
at 8 hpi and virion were isolated and lysed at 14 hpi. VP16 was immunoprecipitated
from either lysate with the monoclonal antibody LPl. A: The immunoprecipitated mate—
rial was separated on an 10% SDS-PAGE gel, transferred to a nitrocellulose membrane
and visualized by autoradiography. A labelled band was observed at 65 KDa in both
samples but not in a sample prepared from a mock infected lysate. Bands observed at
145 and 32 KDa were not identified. B: An aliquot of the immunoprecipitated VP16
from cell lysates were digested with lysylendopeptidase C (EKC) and trypsin (Tryp). The
digestion products were separated on a 16% tris/tricine gel, then transferred to nitrocellu-
lose membrane, The labeled bands were visualized by autoradiography. A band of 18
KDa was observed in the EKC sample and a band of 15 KDa was observed in the Tryp
sample. These bands correspond to the expected migration of the a 371-490 and aa 410-
490 carboxy terminal fragments of VP16, respectively.

37

determined by immunoblotting (See figure 4, panel A). These results confirmed that
VP16 isolated from either virion or from late in infection was phosphorylated.

The identities of the bands at 145 and 32 KDa seen in the immunoprecipitation
from the cell lysate are not known. These bands were also present in the virion sample,
albeit at a lower level. These bands were frequently but not consistently observed in
immunoprecipitates of VP16 in infected cell extracts. Because they were not present in
the mock sample, they may represent cellular proteins or viral proteins expressed in the
cell that specifically precipitated through interaction with VP16. But because their
presence in the immunoprecipitate was inconsistent between experiments, no effort was
made to determine their identity

There was a slight difference in the mobility of VP16 immunoprecipitated from
the cell lysate and VP16 purified from virion. This difference in migration between the
cellular and the virion-derived VP16 might have been caused by a difference in
phosphorylation of VP16. But, as subsequent efforts were to be focused on the
phosphorylation of VP16 at late times in infection and not in the virion, it was not
determined whether differences in phosphorylation or other modifications were the cause

of this difference in mobility.

Peptide mapping of in viva 32P-labelled VP16 isolated late in infection using
lysylendopeptidase C. The immunoprecipitation of 32P-labelled VP16 from infected
cells demonstrated that VP16 was phosphorylated, in accordance with the Phosphobase
predictions that indicate many potential sites of phosphorylation in VP16. The first

approach was to identify regions of VP16 that were phosphorylated. O’Reilly et al used

38

lysylendopeptidase C to map Ser375 as a phosphorylation site in HeLa cells. This
approach relied on the fact that VP16 when digested with lysylendopeptidase C generated
two large and three minor fiagments. Both the larger and at least one of the three minor
fragments could be resolved by SDS-PAGE. (Figure 4A)

To provide material for this mapping approach, HeLa cells were infected with
KOS at an MOI of 5-10. [32P]-orthophosphate was added to the infected cells at 1.5 hpi.
At 8 hpi, the cells were lysed and VP16 isolated by immunoprecipitation. A sample of
the [32P]-labeled protein eluted from the immunoprecipitation pellet was separated on a
10% SDS-PAGE gel. The position of the radiolabeled VP16 was determined by
autoradiography of the wet gel and the radioactive band at 65 KDa was excised. The
contents of the gel slice were digested with lysylendopeptidase C. The resulting products
of the digestion were separated on a 16.5% polyacrylamide gel in a tris/tricine buffer
system (Schagger et al. 1987). The separated peptides were transferred to nitrocellulose
and the positions of the radiolabeled species were visualized by autoradiography. Figure
2B shows a representative result of this experiment. Digestion with lysylendopeptidase
yields a single phosphorylated species migrating with an apparent molecular weight of 18
KDa with only faint evidence of additional radiolabeled species.

The four lysine residues in the VP16 protein sequence are located at aa 29, 103,
343 and 370. Complete digestion would result in five fragments of distinct sizes: 3.0
KDa (a 1-29), 8.0 KDa (aa 30—103), 28 KDa (aa 104-343), 3.1 KDa (aa 344-370) and 13
KDa (aa371-490). Assuming complete digestion of the protein, the labeled species in the
lysylendopetidase C digestion had to correspond to one of these fragments. While none

of them are close to the apparent molecular weight of the species on the gel, the 13.2

39

KDa carboxyterminal fiagment of VP16 is known to have lower mobility than predicted
from its molecular weight (O'Reilly et a1. 1997). The 18 KDa size of the labeled species
detected in the lysylendopeptidase digestion was consistent with the previously described
migration of the carboxyterminal fragment of VP16. The conclusion from the
experiment was that the labeled fragment corresponded to the carboxyterminal 120 aa of
VP16. This was also the conclusion drawn by O’Hare and colleagues when they mapped
VP16 expressed from a plasmid in HeLa cells.

The two smallest products of the lysylendopeptidase C digestions were nominally
larger than the smallest molecular weight marker which resolved on the gel. However,
the mobility of small peptides can vary significantly and it was not certain that these
fragments would have resolved on the gel. It was possible that the two smaller fragments

had failed to resolve on the gel and migrated with the dye front.

Peptide mapping of in viva [um-labeled VP16 isolated late in infection using
trypsin. Because the major labeled lysylendopeptidase C fragment contains multiple
putative phosphorylation sites (see figure 1) further efforts were needed to precisely
identify which of these residues were sites of phosphorylation. Using trypsin, it was
possible to digest the protein at the carboxyterminal side of both arginine and lysine
residues. If VP16 was digested with trypsin, most of the protein (aa1-410) would be
reduced to fragments smaller than 36 aa. However, the carboxyterminal 120 a of the
protein (371-490), which correspond to the proposed labeled lysylendopeptidase C

fragment, would yield only three fragments of 29a, 10 a and 80 aa, respectively. Of

40

these three fragments, the 80 a fragment would be expected to resolve on a 16%
tris/tricine gel.

To determine whether the carboxyterminal 80 aa fragment of VP16 was
phosphorylated late in infection, HeLa cells were infected with KOS at an MOI of 5-10.
At 1.5 hpi, 32P-orthophosphate was added to the infected cells and at 8hpi the cells were
lysed. VP16 was isolated from the lysate by immunoprecipitation. Protein was eluted
from the immunoprecipitation pellet and was separated on a 10% SDS-PAGE gel. The
position of the radiolabeled VP16 was determined by autoradiography of the wet gel.
The radioactive band at 65 KDa was excised and digested with trypsin. The products of
the digestion were separated on a 16% polyacrylamide gel in a tris/tricine buffer system.
The separated peptides were transferred to nitrocellulose and the positions of the
radiolabeled species were visualized by autoradiography.

The digestion of 32P-labelled VP16 isolated from HeLa cell infected with KOS
gave rise to a single phosphorylated species (Figure 2B). This peptide migrates at 15
KDa. While the mobility of this fragment is lower than any of the expected products of a
complete digestion of VP16 with trypsin, the carboxyterminal fragment was expected to
migrate slower than expected from the molecular weight. The conclusion drawn from
this experiment was that the 410-490 fiagment of VP16 was phosphorylated. It also
confirmed the identification of the labeled fragment from the lysylendopeptidase
digestion as being aa371-490.

The smaller fragments of the digestions may have been lost during transfer of the
peptides to the membrane. To address this issue, an alternative procedure of detecting the

labeled bands was used. After electrophoresis of the digestion products, the glycerol was

41

soaked out of the tris/tricine gel and the gel was then dried onto Whatrnan 3MM paper.
The dried gel was subjected to autoradiography. Again, the labeled species of higher
apparent molecular weight were detected in both digestions but there was still no
evidence of labeled species in the 3 KDa range. (Data not shown) However, even though
the 3 KDa marker resolved and was clearly discernible in the dried gel, there was a
concern that small fragments of the digestion might have been lost when glycerol was
soaked out of the gel. It was not possible to skip this step as the glycerol impeded the
drying of the gel and the presence of water in the gel would quench the emission of

radioactivity from any weakly labeled peptides.

Phosphorylation of VP16 truncation mutants in viva. The availability of a set
of truncation mutants of VP16 in the virus made it possible to further map the location of
the phosphorylation site. Two strains in particular would allow for the determination of
whether the activation domain was required for the phosphorylation of VP16. The
previous mapping of the full length VP16 suggested that phosphorylation took place
within the carboxyterminal 80 a of the protein. If the activation domain was the
predominant site of phosphorylation, deletion of the activation domain would diminish or
abolish the phosphorylation of the protein. The two strains that would be useful for this
approach were RP3 and RPS. These two HSV-l strains contain truncations within the
VP16 ORF (Figure 3). The carboxyterminal activation sub-domain of VP16 is deleted in
RP3 (A455-490) and the entire activation domain is deleted in RP5 (A413-490). Both of
these strains are viable without complementation, but RP5 is known to have a reduced

proliferation. An aspect of this phenotype is a 100 fold increased particle to pfu ratio.

42

Figure 3: Viral strains. KOS is a wild type strain and the background for the remainder
of the strains used in this study. The RP3 and RPS strains carry carboxyterminal
deletions of VP16, A454-490 and A413-490, respectively. DGl carries wild type VP16
with a carboxyterminal EGFP-tag. SJOZ, S01 1, S012 and 8013 were created for this
study. SJO2 carries an alanine substitution at Ser375. S011, 8012 and 8013 carry
carboxyterminal EGFP-tags and alanine, glutamate and threonine substitutions at Ser
411, respectively.

43

mmwm Z Enact 5:963. E Enact 800 WNM

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

i V V V. N 28
. . : Hem. . .
. AV . ....9.0.0.06406101919491.640401010.. . aom
. m: nvm” " u
. :Om
. 3 5m .
~on
" n . .memm . .
m T808433? me.
u a... u n u
. . _\ . Em
" e2. . .
//////////////////V//.4\\\\m _ Go

 

 

\\\F§§ mom

omN+ oav m; wwm 9» fl

 

 

HeLa cells were infected with KOS and RP3 at an MOI of S, and with RPS at an
MOI of 0.1. At 1.5 hpi, 32P-orthophospate was added and the infected cells were lysed at
8 hpi. Irnmunoprecipitation was performed as previously with the monoclonal antibody
LP1 and the precipitated proteins were eluted from the immunoprecipitation pellet at pH
12.5. The eluate was separated on a 10% SDS-PAGE gel and transferred to a
nitrocellulose membrane. The radiolabeled proteins were visualized by autoradiography.
Figure 4A shows a representative autoradiograrn. All three versions of VP16 give rise to
a phosphorylated species, corresponding to the expected apparent molecular weight of the
wild type or truncated proteins (65 KDa, 58 KDa and 52 KDa respectively.) Panel 3B
shows an immunoblot of the same membrane, in which the VP16 specific antiserum C8
was used to verify the position of the wild type and truncated VP16. The immunoblot
confirms that the radiolabeled species observed in the first panel correspond to VP16.
These results show that truncated VP16 from both RP3 and RPS was phosphorylated.
Therefore phosphorylation sites must be present within aa 1-412. Moreover, the
activation domain of VP16 is not necessary for phosphorylation of VP16. It could not be
determined whether the specific labeling differed between the three versions of VP16.
Therefore, the possibility that one or more phosphorylation sites of the full length protein

were removed in the truncation of RP3 or RPS could not be excluded.

Peptide mapping of the phosphorylation sites of VP16 truncation mutants
using lysylendopeptidase C. The previous experiments indicated that at least one site
was phosphorylated within the 80 carboxyterminal a of VP16, but did not resolve

whether any of the sites between a 1-28 and 344-410 were phosphorylated. This

45

Figure 4: VP16 is labeled outside of the activation domain. HeLa cells were infected
with KOS, RP3 and RPS at 5-10 moi in the presence of [32P]-orthophosphate. Cell
lysates were prepared from each infection at 8 hpi. VP16 was immunoprecipitated from
each lysate with the monoclonal antibody LP1. A: The immunoprecipitated material was
separated on an 10% SDS-PAGE gel, transferred to a nitrocellulose membrane and
visualized by autoradiography. A labeled band was observed at 65 KDa in the KOS
sample, at 60 KDa in the RP3 sample and at 53 KDa in the RPS sample. These bands
corresponded to the expected migration of the VP16 from each sample. Additional
labeled bands were similar in all three lanes. B: The identities of the labeled bands were
verified by immunoblotting. The nitrocellulose membrane was probed with the
polyclonal antiserum C8. In all three samples, the immunoblot identified species
corresponding to the migration of the labeled samples in the autoradiogram. A minor
band was observed at 150 KDa in both KOS and RP3. These bands did not comigrate
with a [32P]-labeled species. C: An aliquot of the immunoprecipitated VP16 from cell
lysates were digested with lysylendopeptidase C (EKC). The digestion products were
separated on a 16% tris/tricine gel, then transferred to nitrocellulose membrane, The
labeled bands were visualized by autoradiography. A band of 18 KDa was observed in
the EKC sample and a band of 15 KDa was observed in the Tryp sample. These bands
correspond to the expected migration of the a 371-490 and aa 410-490 carboxy terminal
fragments of VPl 6, respectively.

46

B Immuno-

 

 

 

 

 

 

 

 

 

 

 

 

 

[32mm blot
”3 m In m m In
E E E E E E
145- 145-"
98- 981
65--‘_ 65165ea
50- 50-
36- 36-
22- 22-
C EKC Tryp
“3 «a u) “3 on V)
E E E E E E

 

 

17 a
17 k -
16 £5 16

 

 

 

 

 

 

47

question was most easily approached by mapping the sites of phosphorylation within
VP16 derived from RP3 and RPS by proteolysis. Similarly to full length VP16, digestion
of either truncated proteins with lysylendopeptidase C would yield five fragments of
VP16. The four aminoterminal fragments would be the same in for all three versions of
VP16. The only fiagment that would vary between the three samples would be the
carboxyterminal fragment. In full length VP16, this fragment would be 120 a and was
seen in the previous experiments to migrate with the apparent molecular weight of 18
KDa. In the RP3 sample, the protein was truncated to aa 454. Correspondingly, the
expected carboxyterminal fragment of the digestion would be 84 amino acids. The
mobility of this fragment had not been determined previously, but was expected to
migrate faster than the 120 a fragment from the KOS sample. In RPS, VP16 is truncated
at aa 412. Subsequently, the carboxyterminal fragment resulting from the
lysylendopeptidase C digestion would only be 44 a. This fragment was expected to
resolve on the gel, yet migrate faster than both the corresponding fragment from the KOS
and the RP3 sample. Phosphorylation between a 412 and aa 454 would give rise to a
labeled band in the RP3 sample, but not in the RPS sample. Phosphorylation between
371 and 412 would result in labeled bands in both RP3 and RPS. In this case, additional
phosphorylation between a 412-454 could not be excluded. If no phosphorylation
occurred outside of aa 455-490, there would be no labeled band in either sample.
Regardless of whether any phosphorylation was detected in the mutants, phosphorylation
between aa 455-490 could not be excluded by this method.

HeLa cells were infected with KOS or RP3 at an MOI of 5, or with RPS at an

MOI of 0.1. At 1.5 hpi, [32P]-orthophosphate was added and the infected cells were lysed

48

at 8 hpi. Irnmunoprecipitation was performed as previously with LP1 and the
precipitated protein was eluted fiom the immunoprecipitation pellet at high pH. An
aliquot of the eluate from the immunoprecipitation pellet was resolved on a 10% SDS-
PAGE gel and the position of the labeled species was determined by autoradiography of
the wet gel. The radioactive bands corresponding to VP16 in the KOS, RP3 and RPS
samples were excised as described above and the labeled protein was subjected to
proteolysis in the gel by endolysylpeptidase C. The fragments resulting from the
digestion were separated on a 16.5% tris/tricine gel and visualized by autoradiography
(figure 4C).

As seen previously, digestion of the full length protein gave rise to a single
radiolabeled band of 17 KDa. The truncated VP16 from RP3 also gave rise to a single
radioactive band that migrated at 11 KDa. RPS gave rise to two bands, migrating at 6
and 5 KDa. The significance of the double band is unclear, although a band
corresponding in mobility to the upper band was also present in the KOS sample but not
in the RP3 sample. The lower band was unique to the RPS sample. The predominant
band in both the RP3 and the RPS samples migrated as would be predicted for the
carboxyterminal fragment of the lysylendopeptidase C digestion. This observation was in
agreement with the previous identification of the labeled peptide in the digestion of the
full length protein as the one spanning aa 410-490. Because both the carboxyterminal
fragments of RP3 and RPS were labeled, I conclude that VP16 was phosphorylated
between a 371-412. These results do not rule out additional phosphorylation sites

between 413-490.

49

Peptide mapping of the phosphorylation sites of VP16 truncation mutants
using trypsin. To determine whether any phosphorylation sites fall between a 410-454,
the mapping of the phosphorylation sites of VP16 by trypsin digestion was repeated on
VP16 derived from RP3 and RPS infection of HeLa cells. Like full length VP16,
digestion of RP3-derived VP16 would render the majority of the protein too small to be
detected on the gel. But like the full length protein, digestion of the carboxyterminal end
of VP16 would yield a peptide of 44 aa (aa 411-454). It was reasonable to expect this
peptide to resolve on the gel. A similar digestion of RPS was not expected to yield any
fragment of detectable size. If the RP3 peptide was labeled, it would indicate a
phosphorylation site between aa 411-454. If it was not, it would indicate a
phosphorylation site between 455-490.

HeLa cells were infected with KOS or RP3 at an MOI of S, or with RPS at an
MOI of 0.1. At 1.5 hpi, 32P-orthophosphate was added and the infected cells were lysed
at 8 hpi. Irnmunoprecipitation was performed as previously with LP1 and the
precipitated protein was eluted from the immunoprecipitation pellet at high pH. An
aliquot of the eluate fiom the immunoprecipitation pellet was resolved on a 10% SDS-
PAGE gel and the position of the labeled species was determined by autoradiography of
the wet gel. The radioactive bands corresponding to VP16 in the KOS, RP3 and RPS
samples were excised as described above and the labeled protein was subjected to
proteolysis in the gel by trypsin. A representative experiment is shown in figure 4C.

The digestion of full length VP16 once again gave rise to the expected 12 KDa
fragment. In this particular experiment a digestion product migrating at 14 KDa was

observed, suggesting incomplete digestion. This band most likely corresponds to the a

50

400-490 fragment of VP16. The RPS digestion gave little evidence of any unique bands.
The bands that were seen were common to either all three or to at least two of the
samples (SKDa band is present in both KOS and RPS.) The RP3 sample gave rise to a
band migrating at llKDa. This is likely to be a partial digestion product, similar in
migration to the carboxyterminal lysylendopeptidase C product of RP3 (aa 370-454).
There is also a band that migrates slightly faster than the SKDa band common to the KOS
and RPS samples.

The migration of the major species in the KOS sample as well as the absence of
bands in the RPS sample were expected. The result of the RP3 experiment is not as easy
to interpret. If the 4.5 KDa species did in fact correspond to the a 410-453 peptide, the
labeling of both the aa 371-454 and the a 371-412 fragments could be attributed to
phosphorylation at Ser411. Unfortunately, it is difficult to ignore the presence of bands
in both the KOS and RPS samples which have similar migration. If one argues that the
4.5 band is not unique to the RP3 sample, the conclusion must be that there is no
phosphorylation of RP3 between aa 411-454. This means that the phosphorylation of the
carboxyterminal lysylendopeptidase fragments of both RP3 and RPS is phosphorylated
aminoterminal to aa410 and that the carboxyterminal trypsin fragment of the wild type
would be phosphorylated carboxyterminal to aa454. Unfortunately, the data does not

favor one interpretation over the other.

Phosphoaminoacid analysis of VP16 from late in infection. The

phosphoamino acid composition of VP16 expressed from a plasmid in HeLa cells had

been shown only to consist of phosphoserine (O'Reilly et al. 1997). Given that the

51

regulatory controls of VP16 function imposed by the HSV-l gene expression cascade
would be different in infection, resulting in different phosphorylation patterns of VP16, it
would stand to reason that the type of phosphorylation site needed not be conserved,
either. It was possible that the labeled peptides discovered in the mapping attempts of
VP16 derived from infected cells represented phosphorylation of any of the three major
types of phosphorylation sites, namely serine, threonine or tyrosine. Excluding one or
more of these phosphoamino acids would narrow the range of possible phosphorylation
sites.

To identify the type of residues that are phosphorylated during infection, a
phosphoamino acid analysis was performed. An aliquot of each of the [32P]-labeled
proteins immunoprecipitated from HeLa cells infected with KOS, RP3 and RPS were
separated on a 10% SDS-PAGE gel and transferred to PVDF membrane. The position of
the radioactive band was determined by autoradiography and the bands corresponding to
the three VP16 species were excised from the membrane. The proteins on the membrane
were hydrolyzed in 6N hydrochloric acid at 100°C. Phosphoserine, phosphothreonine
and phosphotyrosine standards were added to the sample and the resulting amino acid
mixture was separated by one-dimensional thin layer electrophoresis at pH=2.5. The
radioactive amino acids were visualized by autoradiography (Figure 5) and the
phosphoamino acid standards were stained with ninhydrin. A single radioactive spot was
observed in each sample. In all three cases, this spot co-migrated with the phosphoserine
standard. The conclusion from the experiment was that serine was the only residue

phosphorylated in VP16 derived from late stages of infection of HeLa cells.

52

 

+

“”1 ’ ’11-‘51 If?)
pSer E... .7; I- s." V
‘1'» ’ _ a ‘

Thr ”3‘ k) (J
I a-

p 7% I, \‘ ’l ‘I
PTYT \ I, \ -- ’ T - I

 

 

 

Figure 5: VP16 is labelled on serines. HeLa cells were infected with KOS, RP3 or RPS
in the presence of [ 3-2P] -.orthophosphate VP16 was immunoprecipitated with the mono-
clonal antibody LPl. The immunoprecipitated material was separated on an SDS -PAGE
gel, transferred to a PVDF membrane and the labeled band at 65 KDA, corresponding to
VP16, was excised from the membrane. The protein was hydrolyzed in 6N HCl for 1
hour at 100°C. The samples were mixed with phosphoserine, -threonine and -tyrosine
standard and separated by thin layer electrophoresis on a cellulose plate at pH 2.5. The
phosphoamino acid standards were visualized by staining with ninhydrin (indicated by a
stippled outline) and the [3 2P]- labeled species were identified by autoradiography. The
only labeled species in each sample cornigrated with the phosphoserine standard. + and -
indicate the electrophoretic poles.

53

The experiment reduced the number of possible sites within the carboxyterminal
tryptic fragment of full length VP16 to only 4: Ser41 l, Ser419, Ser452 and Ser462. At
least one of these sites must be phosphorylated to give rise to the labeled tryptic fragment
in the peptide mapping of the full length protein. Phosphorylation at Thr412, Thr4l6,
Thr480, Tyr465 and Tyr488 can be excluded. In the potentially labeled but possibly not
detected minor fragments of either lysylendopeptidase C or trypsin digestion,
phosphorylation at Thr351, Thr352, Tyr364, Thr369, Thr376 or Thr407 was excluded. It
did not completely exclude phosphorylation of any of these fragments, as each
encompasses at least one serine. The potential phosphorylation sites that remain within
these fi'agrnents are Serl8, Ser346, Ser348, Ser349, Ser353, Ser355, Ser365, Ser375 and

Ser400.

Mass spectrometry of VP16 isolated late in infection. While the
phosphopeptide mapping and the phosphoamino acid analysis had excluded a number of
potential phosphorylation sites, thirteen serines remained as possible sites of
phosphorylation. Of these thirteen, phosphorylation had to occur at at least one site
among the sites at Ser 411, Ser419, Ser4S3 and Ser462. Whether any of the remaining
were phosphorylated was not known, but none could be rigorously excluded. To further
refine the mapping of the phosphorylation of VP16, a different approach was chosen.
Subjecting the protein to mass spectrometry would not only allow the detection of
phosphorylated fragments of a digestion, it would also allow positive identification of the

phosphorylation sites within the fragments.

54

Non-radiolabeled VP16 was purified from 109 HeLa cells infected with KOS by
immunoprecipitation. The protein was eluted from the immunoprecipitation pellet at
high pH and the resulting eluate was separated on a 10% SDS-PAGE gel. The gel was
stained with coomassie blue, and the gel was divided in two by cutting the lane of the
immunoprecipitated protein lengthwise into two unequal parts. Five prominent bands
migrating between 60 and 70 KDa were excised from both parts and the slices from the
smaller part (~l/5 of the loaded sample) were loaded back on a 10% SDS PAGE gel.
This gel was transferred to nitrocellulose and the slice corresponding to VP16 was
identified by immunoblotting with the polyclonal rabbit serum LA2-3. The slices fi'om
the other part of the gel (~4/5 of the sample) were washed in acetonitrile, dried and frozen
at -80°C. After determination of the band corresponding to VP16, this band was
submitted to the Harvard Microchemistry Facility for protease digestion and subsequent
LC-MS/MS. Separate samples were digested with trypsin and with aspartyl-
endopeptidase N to facilitate broader coverage of the protein.

A summary of the results from both digestions is shown in figure 6 and table 1.
The MS/MS returned data for 79% of the protein. The MS/MS positively demonstrated
phosphorylation at Serl8, Ser353 and Ser452. For these three sites, spectra were
observed that were consistent with the presence of both the phosphorylated and the
unphosphorylated species. Six serine residues were not covered by the MS. Of these six,
four fell in regions of the protein that were covered by the peptide map, but did not
appear to be phosphorylated (Ser72, Serl77, Ser186 and Ser262). The remaining two,
Ser 411 and Ser419 both fell within the [32P]-labeled carboxyterminal trypsin fragment

(aa4l 1-490). Their phosphorylation state could not be determined by either the peptide

55

51
101
151
201
251
301
351
401
451

MDLLVDELEA
SPPMPVPPAA
ECKFLSTLPS
EALSRFFHAE
RDLGEMLRAT
DLFDCLCCDL
HLNLPLVRSA
TTSPSEAVMR
FLPAGHTRRL
DSPGPGFTPH

DMDADGASPP
LFNRLLDDLG
DVVEWGDAYV
LRAREESYRT
IADRYYRETA
ESWRQLAGLF
ATEEPGAPLT
EHAYSRARTK
STAPPTDVSL
DSAPYGAL D M

PPRPAGGPKN
FSAGPALCTM
PERAQIDIRA
VLANFCSALY
RLARVLFLHL
QPFMFVNGAL
TPPTLHGNQA
NNYGSTIEGL
GDELHLDGED
ADFEFEQMFT

TPAAPPLYAT
LDTWNEDLFS
HGDVAFPTLP
RYLRASVRQL
YLFLTREILW
TVRGVPIEAR
RASGYEMVLI
LDLPDDDAPE
VBMAHADALD
DALGIDEYGG

GRLSQAQLMP
ALPTNADLYR
ATRDGLGLYY
HRQAHMRGRD
AAXAEQMMRP
RLRELNHIRE
RAKLDSYSSF
EAGLAAPRLS
DFDLDMLGDG

Figure 6: Peptide sequences identified by LC-MS/MS of VP16 isolated late in infection
from HeLa cells infected with HSV-l. The isolated protein was digested with either
endolysylpeptidase C or endoaspartylpeptidase C. Dark typeface indicates sequences of
peptides identified. Light typeface sequences were not observed.

56

TABLE 1. Phosphorylated and non-phosphorylated peptides detected by

MS/MS covering Serl8, Ser353, Ser375 and Ser452

 

 

Residue Coverage Representative spectra (peptide)"‘b
Serine 18 4 - 2 9 LVDELFADMZDADGASPPPPRPAGG PK
5-29 ‘VDELFADMDADGASPPPPRPAGGPK
5-29 'VDELFADMDADGASPPPPRPAGGPK.
6-29 DELFADMDADGASPPPPRPAGGPK
8-29 LFADMDADGASPPPPRPAGGPK
9-29 FADMDADGASPPPPRPAGGPK
10—29 ADMDADGASPPPPRPAGGPK
Serine 353 342—360 AKLDSYSSFTTSPSEAVMR
344-360 LDSYSSFTTSPSEAVMR
344-360 LDSYSSFTTSPSEAVMR
345-360 DSYSSFTTSPSEAVMR
348—360 SSFTTSPSEAVMR
351-359 TTSPSEAVM
351-360 TTSPSEAVMR
351-360 TTSPSEAVMR
351-362 TTSPSEAVMREH
351-364 TTSPSEAVMREHAY
353—360 SPSEAVMR
Serine 375 3 6 9 - 3 9 8 TKNNYGSTIEGLLDLPDDDAPEEAGLAAPR
369-401 TFNNYGSTIEGLLDLPDDDAPEEAGLAAPRLSF
371-398 NNYGSTIEGLLDLPDDDAPEEAGLAAPR
372-398 NYGSTIEGLLDLPDDDAPEEAGLAAPR
374-401 GSTIEGLLDLPDDDAPEEAGLAAPRLSF
Serine 452 4 4 5 - 4 6 0 DMLGDGDSPGPGFTPH
449-460 DGDSPGPGFTPH
449—460 DGDSPGPGFTPH

 

aBold-face S indicates phosphorylation of the serine. bAdditional spectra were
observed representing peptides of the same length as those shown here, but
carrying methioninesulfoxide, an artifact of sample treatment.

57

mapping, the MS or the combination of the two. Ser44, SerSl, Ser90, Ser106, Serl 10,
Se154, Ser167, Ser309, and Ser333 were found not to be phosphorylated by both the MS
and the peptide mapping. Ser346, Ser348, Ser349, Ser353 and Ser355 were covered by
the MS but not by the peptide mapping. The corresponding peptide (Ser344-Ser370) was
most likely too small to resolve on the tris/tricine gel. Of these, only Ser353 was found
to be phosphorylated.

The MS results were consistent with the mapping of the aa29-103 and the 104-
343 fragments of the lysylendopeptidase experiment. No spectra in the MS suggested
any phosphorylation of any sites within these two peptides. Furthermore, the MS did not
find evidence of phosphorylation of any threonine or tyrosine residues, in agreement with
the phosphoaminoacids analysis.

The detection of phosphorylation of Ser18 and Ser353 by mass spectrometry was
a strong indication that neither of the two smaller lysylendopeptidase C fragments
resolved on the tris/tricine gel. If either of the phosphorylation events at Ser18 and 353
represented major sites of phosphorylation it would have given rise to a signal, had the
corresponding fragment resolved on the gel. Subsequently, it was important to ascertain
whether any of the remaining sites within these fragments were in fact phosphorylated.
The experiment also invoked concerns about the detection of the 371-398 fragment of the
trypsin digestion.

No evidence of phosphorylation at Ser375 was found. All of the multiple spectra
covering this serine show it to be unphosphorylated. To ascertain with the highest
degree of confidence that this was the correct result targeted ion mass spectrometry

(TIMM) was done. The TIMM approach is a modification of the standard method of LC-

58

MS/MS, where peptides corresponding to the expected phosphorylated and non-
phosphorylated masses of a particular digestion are selected in the first dimension MS.
Only these fiagments are subjected to fragmentation and sequencing by the second
dimension MS. Because longer time is spent analyzing these peptides, the likelihood of
detecting a low-abundance species is greatly increased. This more sensitive technique

did not detect any phosphorylated fragments corresponding to phosphorylation at Ser3 75.

Acetylation of VP16. VP16, like a number of transcription factors, interact with
histone acetyl transferases a-IATs) when activating its target genes. The HATS were
initially identified as factors capable of acetylating histones. Recently, a number of these
have been shown to acetylate transcription factors as well. It was possible that VP16
could also serve as an acetylase substrate. To test whether VP16 isolated from infected
cells was acetylated, the MS/MS data set was scrutinized for evidence of acetylation.
VP16 encode four lysines, all of which were detected only as unmodified peptides. No
spectra were observed that would indicate that any of these sites were acetylated (Data
not shown). The conclusion fiom the experiment is that late in infection, VP16 was not

acetylated.

In vitra phosphorylation of the 375 CK2 site. Both the MS and the peptide
mapping show Ser375 of VP16 to be unphosphorylated late in infection. As mentioned
previously, the O’Hare lab found that residue to be phosphorylated (O'Reilly et al. 1997).
They also reported that bacterially expressed VP16 was a substrate for CK2 in vitro.

Because Ser375 is a CK2 consensus site, they tested whether mutations at Ser375

59

affected phosphorylation of VP16 in vitro. They found the Ser375Ala mutation to
abolish phosphorylation of VP16 in vitra and surmised that Ser375 was the primary site
of phosphorylation by CK2. In contrast, Meredith et al. unsuccessfully attempted to
phosphorylate VP16 purified fiom virion with CK2 in vitro. It is possible that the
inability of virion-derived to act as a CK2 substrate reflected a block of the CK2 sites of
VP16 in the virion, either by modification of the serine itself or of the consensus site. In
both instances it is peculiar that VP16 could not serve as a substrate. In addition to
Ser375, one other serine and five threonines of the VP16 core domain conform to the
CK2 consensus sequence (Ser353, Thr79, Thr83, Thr94, Thr124 and Thr210). It is a
little surprising that in either in vitra assay, whether Ser3 75 is mutated or not, none of the
remaining four consensus sites get phosphorylated. Especially when taking into account
that at least one of them (Ser353) falls in an unstructured domain that is surface-exposed
(Liu et al. 1999). To test whether Ser375 is the only kinase accessible CK2 site in VP16,
an in vitra kinase assay was done, using bacterially expressed VP16.

A GST-tagged version of the VP16 core domain (a 49-410) was expressed in E.
coli and affinity-purified using glutathione-sepharose. The core domain of VP16 was
eluted from the affinity column by cleaving the thrombin site located between the GST-
tag and VP16. The purity of the sample was examined by SDS-PAGE, followed by
Coomassie brilliant blue staining of the gel (Data not shown). Commercially available
CK2 was used to phosphorylate VP16 in the presence of y-[32P]-ATP. Labeling of VP16
was visualized by SDS-PAGE of the reaction mixture. An aliquot of each sample were
run on each of two separate gels. One gel was dried and subjected to autoradiography,

the other transferred to nitrocellulose and the presence of the VP16 core domain was

60

verified by immunoblotting with the polyclonal serum SOl-2. To verify that the VP16
was a substrate, the experiment was done with and without VP16. To verify that CK2
was the kinase, the experiment was done with and without CK2.

Figure 7A shows a representative result from these experiments. The two left
panels show the autoradiograms of the labeling reaction and the two right panels
represent the corresponding immunoblots. In the first panel, signals were detected in
both lanes. The band at 18 KDa as well as the fainter band at 40 KDa represent
autophosphorylation of the a and [3 subunits of the CK2 kinase. The sample in the first
contained both the kinase and VP16. In this lane, two unique labeled bands are present at
41 KDa and 12 KDa. These bands were surmised to represent the bacterially expressed
VP16. In the second panel, where no CK2 was added, none of the bands were detected.
This showed that CK2 was required for the phosphorylation of VP16. The immunoblot
of the samples in the two panels at the right verified that the 40 KDa band does in fact
correspond to VP16. The band reacted with the C8 antibody in the samples where VP16
was added, regardless of whether CK2 was present. The 12 KDa band did not react with
the antiserum and may be a degradation product or an abortive translation product of
VP16 that does not carry the predominant epitopes for the C8 antibody or it could be a
minor species fiom the bacterial extract that co-purified with VP16 and has a CK2
consensus signal. The conclusion of the experiment was that the VP16 core domain can
in fact serve as a CK2 substrate in vitro.

To determine whether Ser 375 was the major site of CK2 phosphorylation in
vitro, a number of mutations were introduced at this site. The site was altered to alanine,

proline or threonine. If the site was the major site of phosphorylation, the alanine and

61

A 32]) Immunoblot

(3&2 -+ -+ - - -r 4- - -
‘VP16 -+ - +- - 4— - -+ -
200-

97-
68-

43 1..

 

 

 

 

 

29'-
18-

14-

 

 

 

 

 

 

 

 

 

13 32P

 

 

Silver stain

 

 

WT SA SP ST WT SA SP ST

 

 

97- 97'

2g: . , 43'mw'mm'

29' 29-
18-
14- 14'

 

 

 

 

 

 

Figure 7: In vitro phosphorylation of the VP16 core domain by CK2 is independent of
Ser375. Wild type (wt) and Ser375 mutant (SA, SP, ST) of the VP16 core domains (49-
410) was expressed in E. coli. The purity of the protein was examined (Panel B. Silver

stain). The peptides were phosphorylated in vitro with CK2 in the presence of y—[32P]-
ATP. The labelled proteins were separated on a 10% SDS-PAGE gel and transferred to

nitro cellulose. A: The VP16 core domain is phosphorylated by CK2. The appearance

of the 40 KDa radiolabelled species, corresponding to VP16, is dependent on CK2. The

identity of the 41 KDa species was verified by immunoblotting with the polyclonal anti-

serum SOl-Z. The bands at 40 and 18 KDa correspond to autophosphorylation of the

CK2 a and B subunits. B: The substitution of Ser375 with alanine, proline or threonine

does not abolish labeling of VP16. All of the Ser375 mutants gave rise to a labelled

band corresponding to VP16. The variation in the intensity of the labeling was not

reproducible between experiments.

62

proline substitutions would be expected to abolish phosphorylation of VP16 in the in
vitro experiment. The threonine would most likely allow the kinase to phosphorylate
VP16.

The mutations were introduced into VP16 in the GST-core domain firsion by PCR
based site directed mutagenesis. The mutations were verified by sequencing and the
protein was expressed in E. coli and purified on a GST-sepharose matrix. The protein
was eluted by cleavage of the thrombin site that had been engineered between the GST-
tag and the VP16 protein. The expression of the mutant protein was verified by
coomassie brilliant blue staining. The mutant proteins were tested in the CK2 assay as
described above. The wild type and mutant proteins were subjected to phosphorylation
by CK2 in vitro in the presence of y-[32P]-ATP. Aliquots of each reaction were separated
by SDS-PAGE and one gel was dried and subjected to autoradiography, one gel was
stained with silver and one gel was transferred to nitrocellulose and VP16 was visualized
by immunoblotting with the polyclonal serum C8 (Data not shown). A representative
result of the autoradiogram and the silver staining is shown in figure 7B. In the
autoradiogram depicted in the left panel, each of the four samples tested gave rise to a
significant band migrating at the predicted size for the VP16 core domain. Additional
labeled species were observed at 14 KDa, 16 KDa and 25 KDa. The 25 KDa species was
an autophosphorylation product of the [3 subunit of CK2. The slower migration of the [3-
subunit migrated in this experiment compared to the previous may be attributed to either
to differences in the commercial CK2 source, the molecular weight markers or in gel
conditions. The fainter 41 KDa band that correspond to the a-subunit was occluded by

the VP16 in this experiment. The other two bands were described previously and do not

63

affect the outcome of this experiment. The major labeled species comigrated with the
purified VP16 shown in the silver stained gel. The variation in phosphorylation level was
not reproducible. No gross difference in phosphorylation appeared between the wild type
and the three mutants. The conclusion from this experiment was that because VP16 does
not depend on Ser375 to act as a substrate for CK2 in vitro, another CK2 site can serve as

a efficient acceptor for the CK2 kinase. .

Materials and methods

Cell lines and tissue culture

Four cell lines were used in this dissertation. HeLa and Vero cells were acquired
from the American Type Culture Collection. Vero 16-8 was a generous gift of Steven
Weinheimer. They were derived from Vero cells, into which a gene expressing VP16
(Weinheimer et al. 1992) has been stably transformed. BHK-MMTV-VP16, a Syrian
hamster kidney cell-derived cell line, also stably transformed to express VP16
(Hippenmeyer et al. 1993). The cells were grown in Dulbecco’s modified essential
medium (DMEM) (Gibco) supplemented with 3.7 g/L sodium bicarbonate and 10% fetal
calf serum (FCS) (Select fetal calf serum, Atlanta biologicals). Verol6-8 and BHK-
VP16 were grown in the presence of geneticin at 0.9 or 0.5 ug/mL, respectively, to
maintain the expression of the transgene. Cells were grown in a water-jacketed incubator
at 37°C and 10% C02. Passage was done by standard trypsinization methods.

In applications that required the seeding of a specific number of cells, the cells

were counted after trypsinization and resuspension in fresh growth medium. A small

64

aliquot of the cell suspension was counted on a standard Haemocytometer with a
chamber volume of 10‘4uL/mm2. The cell suspension was then diluted to the desired cell

density and the diluted suspension was distributed to the required tissue culture vessels.

Viral strains

Nine viral strains were used in this dissertation (figure 3). KOS is the wild-type
virus from which all of the other strains used were made. KOS was obtained from the
American type culture collection. RP3 and RPS contain truncations of the
carboxyterminal activation domain (AD) after amino acid 454 and 413, respectively (Tal-
Singer et a1. 1999). 8MA carries an substitution of the VP16 ORF with the LacZ-gene
(W einheimer et al. 1992). In DGl, the endogenous VP16 gene has been replaced with a
gene expressing a carboxyterminal EGFP-tagged wild type VP16 (Greensides 2002).
The strains SJ02, S011, 8012 and S013 were created for this study. SJ02 carries a
substitution of the codon for Ser375 with a codon for alanine. S011, 8012 and 8013 are
derived from DGl and encode EGFP-tagged VP16. Each carries a substitution of the
codon for Ser411 of VP16. S011 carries an alanine substitution, S012 carries a
glutamate substitution and S013 carries a threonine substitution of this codon.

Viral strains were propagated in Vero cells or Vero 16-8 cells. KOS, DGl, SJ 02
and RP3 stocks were grown in Vero cells. RPS, S011, S012 and $013 primary stocks
were grown in Vero 16-8 cells to complement the mutation in the virus strain. RPS,
S011, S012 and 8013 were grown in Vero cells when testing the phenotypes of these
strains in transcriptional assays and in growth curves. 8MA does not proliferate without

complementation and stocks were grown in Vero 16-8 cells.

65

To generate stocks of the viral strains, Vero or Vero 16-8 cells were seeded in 25
mL DMEM in T150 tissue culture flasks (Corning) at 5x106 cells per flask as described
above. After 24 hours, the cells were infected at a multiplicity of infection (moi) at 0.01.
The growth medium was removed fiom the cells and the cell layer was washed with
DMEM without serum. The viral inoculum was prepared by diluting an aliquot of viral
stock, containing 5x108 plaque forming units. This aliquot was diluted to a total volume
of 1.5 mL with DMEM. The inoculum was left on the cells for 1 hour at 37 °C. The
flasks were rocked every 15 minutes to ensure equal distribution of the inoculum and to
ensure that all cells remained immersed in the medium. After one hour, the inoculum
was removed by aspiration and the cells were washed once with DMEM. The cells were
then fed with 25 mL of DMEM supplemented with 2% FCS. The progression of the
infection was monitored and the cells were harvested when 80-90% of the cells exhibited
signs of cytopathic effect (CPE), such as balling-up and releasing from the tissue culture
vessel. Typically, KOS, RP3, RPS, DGl, SJ02, S011, S012 and 8013 stocks were
harvested after three days, RPS stocks were harvested at 4-5 days. The infected cells
were suspended in the grth medium and transferred to a 50 mL conical tube (Coming).
The cells were pelleted gently at ambient temperatures for 5 minutes at 200xg in a
Sorvall Techospin swinging bucket centrifuge (Sorvall) and approximately 15 mL of the
supernatant was removed and aliquoted in 0.5 or 1 mL aliquots in cryovials (Laboratory
Products Sale). This low titer stock was stored at -80 °C. The cell pellet was
resuspended in the remaining supernatant (approximately 9 mL) and transferred to a 25
mL tissue culture flask. The sample was sonicated in a sonicator (Heat Systems

Sonicator W-385) equipped with a cup born to disrupt the cells and release cell-

66

associated virus. The sonication was done discontinuously in 1 second intervals, at 100%
output and 100% load. The sonication was repeated 3 times for 30 seconds each time.
Between each sonication, the sample was allowed at least one minute for the temperature
to re-equilibrate. After sonication, the cellular debris was pelleted at 3000xg for 5
minutes in the Sorvall Techospin centrifuge at ambient temperature. This supernatant
was termed high titer stock and transferred to cryovials in 0.25 or 0.5 mL aliquots. The

stock was saved at -80 °C.

Plaque assays

Plaque assays were used to determine the titer of virus stocks and to determine the
titer of individual samples from single stage growth curves. Plaque assays were done on
Vero cells, except for RPS and 8MA, where Vero 16-8 cells were used. To do the plaque
assay, cells of the appropriate cell line were seeded in P60 tissue culture plates at 5x105
cells per plate in DMEM supplemented with 10% FCS and 0.9 uL/mL geneticin, if
required. 24 hours after seeding, the growth medium was removed by aspiration and the
cells were washed with 5 mL of DMEM. The virus samples were diluted in a tenfold
dilution series in triplicate to generate inoculi that were expected to contain 10 pfu/ 100
uL, 100 pfir/ 100 11L and 1000 pfu/ 100 11L. 100 11L of each inoculum was added to a P60
and the infection was allowed to proceed for one hour at 37°C. The plates were rocked
every 15 minutes to ensure even distribution of the inoculum. Afier one hour, the
inoculum was removed by aspiration and the cells were washed once with 5 mL of
DMEM. The cells were overlaid with DMEM supplemented with 5%FCS and melted

0.9% SeaPlaque agarose (FMC) and the SeaPlaque was allowed to congeal at ambient

67

temperature. The plates were incubated, until foci of infection (plaques) were clearly
visible with the naked eye, typically three days, four days when RPS was titered on Vero
cells. The Seaplaque overlay was removed and the cells were stained with methylene
blue (Sigma) in 70% isopropanol (J. T. Baker). The plaques were observed as well-
defined, circular voids in the blue-stained cell layer. The plaques were counted and the
titer of the original sample was calculated and recorded as pfu/mL. The standard

deviation of the triplicate samples was calculated to assess the accuracy of the final titer.

SDS-PAGE

Protein samples from infected cell and virion lysates, in vitro kinase assays,
immunoprecipitations, protease digestion and GST-purification of the bacterially
expressed VP16 core domain were separated by SDS-PAGE. When examining infected
cell and virion lysates, immunoprecipitated VP16, in vitro kinase assays and GST-
purified VP16 core domain, the samples were separated on a 10% PAGE gel, using the
laemmli buffer system (Gallagher 1999). Peptides derived from digestion of VP16 by
trypsin and endolysylpeptidase C were analyzed on a 16.5% gel, using a Tris/tricine

buffer system (Gallagher 1999).

Immunoblots
Immunoblots of full length wild type VP16, as well as point mutants and

truncated versions of VP16 were done using the rabbit polyclonal antisera LA2-3, C8 or
S01-2. LA2-3 was raised against a GAL4(1-147)-VP16(410-490) fusion protein and

recognizes the activation domain of VP16 (Sullivan et al. 1998) C8 is a rabbit polyclonal

68

antiserum. It was raised against full length VP16, derived from HSV-l virion
(Triezenberg et al. 1988a). SOl-2 was raised against a bacterially expressed VP16 core
fragment (aa49-410) (Ottosen). The proteins were electrophorectically transferred to a
0.45 pm nitrocellulose membrane (Protran BA85, Schleicher and Schuell) in western
transfer buffer (ZSmM Tris pH 8.3, 50 mM glycine, 20% methanol). The proteins were
transferred for 48 minutes at 400 mA in a Hoefer Transphor electrophoresis unit at 4°C.
After transfer, the transfer was assessed by reversible Ponceau S staining to verify even
transfer of all samples. The membrane was briefly incubated with a 0.1% solution of
Ponceau S in 5% acetic acid (BioRad) and the excess stain was washed out with
deionized water. The membrane was blocked for one hour at ambient temperature under
constant agitation with 10% non-fat dry milk in T-TBS (125 mM Tris pH 7.5, 1.5 M
sodium chloride , 0.1% Tween 20) to prevent non-specific interaction between the
antibody and the membrane. Approximately 1 mL of the blocking reagent and all
subsequent T-TBS washes were used per cm2 of the membrane. After removing the
blocking reagent, the membrane was washed three times with T-TBS for five minutes
each time with constant agitation to remove the excess blocking reagent. The primary
antibody was diluted in T-TBS supplemented with 10% calf serum (Gibco).
Approximately 0.2 mL of antibody solution was used per cm2 of membrane and the
antibody solution was left on the membrane for one hour under constant agitation. After
incubation, the antibody solution and the membrane was washed three times for five
minutes with T-TBS at ambient temperature under constant agitation. The secondary
antibody was HRP-conjugated Goat-anti-rabbit IgG (BioRad, 172-1013). In each

experiment, the secondary antibody was diluted at the same dilution as the primary

69

antibody in T-TBS supplemented with 10% calf serum. The membrane was incubated
for one hour with the secondary antibody solution under constant agitation. After the
incubation, the antibody solution was removed and the membrane was washed five times
for five minutes each time with T-TBS. The immunoblot was visualized with the Lumi—
Light chemiluminescence substrate (Roche). The chemiluminescence signal was
recorded on Kodak X-Omat AR film. Multiple exposures were done with each blot,
ranging from 5 seconds to 5 minutes. Holes were made through the edge of the
membrane into the film to align the fihn after development with the prestained molecular
weight markers on the membrane. The film was developed in an automated x-ray film

developer under standard conditions.

Mutagenesis

PCR-mediated site directed mutagenesis was used to generate all mutations in the
study. Complementing primers corresponding to the mutated codon and 9-12 flanking
nucleotides on each side were purchased (Table 2). PCR reactions were set up with 125
ng of each primer, 250 ng of parent plasmid, lefuTurbo PCR-buffer (Stratagene), 100
mM dNTP, 10% DMSO and 2.5 U of PfuTurbo DNA polymerase (Stratagene) in 100 11L.
Thermocycling was typically done with 1 minute at 94°C, 1 minute at 55°C and 8
minutes at 74°C. After 15 rounds of thermocycling, 1 U of Dpnl restriction enzyme
(New England Biolabs) was added to the reaction to remove the parental template DNA.
The resulting DNA was precipitated with 70% 2-propanol and redissolved in 10 11L of
Tris-HCl (pH=8.S). S uL of the sample was electroporated into DHSu cells, plated on

LB-plates, supplemented with 100 ug/mL ampicillin. Resistant clones were picked, re-

70

TABLE 2. Primers used for generation of point mutations in VP16

 

 

Mutation Primers
Ser375Ala 5 ' -CCCTCGATGGTAGCCCCGTAATT-3 '
5 ’ -AATTACGGGGCTACCATCGAGGG- 3 ’
Ser375Pro 5 ' —GCCCTCGATGGTAGGCCCGTAATT-3 ’
5 ' —AATTACGGGCCTACCATCGAGGGC—3 ’
Ser375Thr 5 ’ -GCCCTCGATGGTAGTCCCGTAATT-3 ’
5 ’ -AATTACGGGACTACCATCGAGGGC-3 ’
Ser4l lAla 5 ’ -CGCAGACTGGCCACGGCCCCC—3 ’
5 ’ —GGGGGCCGTGGCCAGTCTGCG-3 '
Ser411Glu 5 ’ -CGCAGACTGGAGACGGCCCCC-3 ’
5 ’ -GGGGGCCGTCTCCAGTCTGCG-3 ’
3341111,; 5 ’ —CGCAGACTGACGACGGCCCCC-3 ’

5 ’ -GGGGGCCGTCGTCAGTCTCGC- 3 ’

 

71

streaked and re-picked. The insertion of the correct mutation was verified by automated
sequencing. Ser37SAla were introduced into the VP16 gene of pKOS-VP16.2 (Tal-
Singer et al. 1999). Ser37SAla, Ser37SPro and Ser375Thr mutations were introduced
into the pGST-VP16 bacterial expression vector pGST-VP16 (49-412) (Liu et al. 1999).
Ser411Ala, Ser411Glu and Ser4llThr mutations were introduced into the pDRG29-1

plasmid (Greensides 2002).

CST-protein purification

The proteins were expressed in the E. coli strain BL21 DE3 Codon Plus. One
liter of LB was inoculated with the bacteria. The expression of the protein was induced
with 0.5 uM IPTG for 3 hours at 30°C. The cells were harvested by centrifirgation at
5000x g for 15 minutes in a Sorvall Superspeed with the SA rotor at 4°C.. The cell pellet
was resuspended on ice in 10 mL HEMGT-250 buffer (25 mM HEPES pH7.9, 0.1 mM
EDTA, 12.5 mM magnesium chloride, 10% glycerol, 0.1% Tween-20, 250 mM
potassium chloride, 1 mM DTT, 1xComplete Protease Inhibitor cocktail (Roche)) . The
cells were lysed in a french press. The lysate was cleared at 10,000xg in a SS32 rotor in
a Sorvall superspeed centrifuge at 4°C for 10 minutes. The GST-protein was bound to 1
mL of a 50% slurry of GSH-sepharose beads (Roche) for three hours at 4°C under
constant agitation. The beads were washed three times for 15 minutes each time in
HEMGT-250, two times in HEMGT-IOO (as HEMGT-ZSO, except only 100 mM
potassium chloride) and resuspended in 1 mL of thrombin cleavage buffer (20 mM
Tris-HCl, pH=8.S, 150 mM sodium chloride, 25 mM calcium chloride). The proteins

were eluted from the beads by proteolysis. 5U of biotinylated thrombin were added to

72

the slurry and the mixture was digested for 16 hours at 20°C. After digestion, 50 11L of a
50% slurry of streptavidin-sepharose beads were added to the slurry to remove the
thrombin. The sepharose beads were pelleted by centrifugation at SOOOxg in an
Eppendorf tabletop centrifuge. The supernatant was recovered, the protein concentration
was determined by Bradford analysis and stored at -80°C. The purity of the protein was

evaluated by SDS—PAGE followed by Coomassie staining.

Coomassie and silver staining.

Coomassie staining was done by fixing the protein gel in 40% methanol and 10%
acetic acid for 15 minutes. The fixing solution was replaced with the staining solution
(40% methanol, 10% acetic acid and 0.05% Coomassie brilliant blue R250 (Pierce)). The
gel was stained for 1 hour, then the stain was removed and the excess dye was washed
out with fixing solution, until background staining was no-longer discernible. Gels were
dried between two layers of cellophane, except when preparing LC-MS/MS samples.
Gels that were to be silver stained were also fixed in fixing solution. They were then
treated with 175 M DTT for 15 minutes and stained with 0.1% silver nitrate. The stain
was developed with 3% sodium carbonate/ 0.05% formaldehyde until the bands had
reached desired intensity. The reaction was stopped with fixing solution and the gel was

dried between sheets of cellophane.
Casein kinase 2 assay
200 ng of purified wild type or mutant VP16 core domain was incubated with 5

units of CK2 (New England Biolabs) for 30 minutes at 30°C in 1xCK2 reaction buffer

73

(20 mM Tris-HCl (pH 7.5), 50 mM potassium chloride, 10 mM magnesium chloride, 200
uM ATP and 1 uCi of y-[32P]-ATP). The labeled protein was separated on a 10% SDS-
PAGE gel. The gel was either transferred to nitrocellulose or coomassie stained and
dried between two pieces of cellophane. The [32P]-label was detected by

autoradiography.

Irnmunoprecipitation

[32P]-labeled VP16 was isolated from infected cells and from virion by
immunoprecipitation. 1x107 HeLa cells were infected with KOS, RP3 or RPS as
described above. Infections were typically done at an moi of 5-10, with the exception of
RPS, where infections were done at an moi of 0.1. After 1 hour, the DMEM was
replaced with phosphate free DMEM supplemented with 2% dialyzed FCS. After
another 30 minutes, the medium was replaced with fresh phosphate free DMEM
supplemented with 2% F CS. Then, 1 mCi of [32P]-orthophosphate was added per 1x107
cells. The infection was allowed to proceed for 8 hours, if isolating VP16 late in
infection or 14 hours, if isolating VP16 from virions. After 8 hours, the infected cells
were washed with PBSa, dislodged from the culture dish with a scraper, resuspended in 5
mL of PBSa and transferred to a 15 mL conical tube. The cells were pelleted gently at
ambient temperatures for 5 minutes at 200x g in a Sorvall Techospin swinging bucket
centrifuge (Sorvall). The supernatant was removed and the cells were disrupted by
addition of 500 11L of IP lysis buffer (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 0.1%
SDS, 1.0% sodium deoxycholate, 1.0% Triton X-100, 2 mM PMSF and 1xComplete

(Roche)). The cells were left on ice for 30 minutes and the tubes were then sonicated

74

directly in the cup horn to ensure complete disruption of the cells. The sonication was
done discontinuously in 1 second intervals, at 100% output and 100% load. The
sonication was repeated 3 times for 30 seconds each time. Between each sonication, the
sample was left on ice for at least one minute for the temperature to re-equilibrate. The
sample was transferred to a clean 1.5 mL eppendorf tube and 100 11L of a 10%
suspension of heat-killed, formalin-fixed staphylococcus aureus (SAC) was added to
remove factors that would bind non-specifically to the SAC during the IP. The sample
was incubated 3 hours at 4°C with constant agitation. The SAC was pelleted for 10
minutes at l6000x g at 4°C in an eppendorf microcentrifuge and the supernatant was
transferred to a new tube. 5 11L of the monoclonal antibody LP1 was added to the
supernatant and the sample was incubated for 1 hour at 4°C with constant agitation. 50
1.1L of SAC was then added and the incubation continued for another 2 hours. The SAC
was then pelleted at 200xg for 2 minutes in the microcentrifuge, the supernatant was
removed and the pellet resuspended in lysis buffer. This washing step was repeated for a
total of 3 times. After the last wash, the SAC were pelleted again and briefly washed in 1
mM phosphate buffer at pH 8. The SAC were pelleted again and 50 uL of 10 mM
phosphate buffer at pH 12.5 was added to elute the bound VP16. The sample was
incubated for 30 minutes, the SAC was pelleted and the supernatant was transferred to a
fresh tube, containing 20 11L of 100 mM phosphate buffer, pH 6.8, to neutralize the pH of

the sample.

Peptide mapping

75

In gel digestion: Irmnunoprecipitated [32P]-labeled VP16 isolated infected HeLa
cells 8 hpi was separated on a 10% SDS-PAGE gel. The migration of the labeled VP16
species was determined by autoradiography of the wet gel. The band was excised fiom
the gel, then washed twice in 0.1 M ammonium carbonate, 50% acetonitrile for 50
minutes at 30°C. The wash was removed and the gel slice was dried under a Nz-stream.
The gel was rehydrated with 4 uL of 0.1 M ammonium carbonate, pH 7.5, 0.02% Tween
20. 1 uL of digestion buffer with either 2 pg of trypsin or endolysylpeptidase C was
added immediately. The sample was digested overnight at 30°C. The protein was eluted
from the gel slice by gently sonicating the sample in a sonicating waterbath for five
minutes. The supernatant was recovered from the mixture and another 5 11L aliquot of
digestion buffer was added. The sonication was repeated and the supernatant pooled with
the first aliquot. The sample was separated on a 16.5% tris/tricine gel and then either
transferred to nitro cellulose or soaked in several washes of 10% methanol/7% acetic
acid, then dried onto Whatman 3MM paper. The labeled protein on either the membrane

or the dried gel was visualized by autoradiography.

Phosphoaminoacid analysis

Irnmunoprecipitated [32P]-labeled VP16 isolated infected HeLa cells 8 hpi was
separated on a 10% SDS-PAGE gel. The gel was transferred to a PVDF membrane and
the labeled species corresponding to VP16 was identified by autoradiography. The band
was excised from the membrane and washed in distilled water. 6N hydrochloric acid was
added to cover the membrane and the sample was digested for 1 hour at 100°C. The

hydrochloric acid was removed from the sample by evaporation under vacuum. The

76

dried samples were mixed with phosphoamino acid standards (2 pg each in 5 uL), then
spotted on a cellulose TLC plate. The samples were resolved by thin layer
electrophoresis was run at 1000V for 50 minutes at 0°C at pH=2.5 (2:1 mixture of pH 3.5
electrophoresis buffer (5% glacial acetic acid, 0.5% pyridine in 0.5 mM EDTA) and pH
1.9 electrophoresis buffer (2.2% formic acid, 7.8% glacial acetic acid). The plate was
dried and sprayed with ninhydrin (0.5% in acetone), then developed at 70°C, until the
standards were clearly visible. The radiolabeled amino acids were visualized by

autoradiography (V acratsis et al. 2002).

77

Discussion
Identification of phosphorylation sites in VP16 during late stages of infection

The goal of this thesis was to examine the function of phosphorylation of VP16 in
viral infection. An intermediate goal was to identify the phosphorylation sites of VP16 in
infected cells and compare that to what is known about the phosphorylation of VP16
expressed from a plasmid. I positively identified three phosphorylation sites by MS/MS.
These sites were Ser18, Ser353 and Ser452. I also inferred phosphorylation of Ser4l 1,
based on the peptide mapping and the absence of phosphorylation at Ser375 and Ser400,
detected by MS/MS. The in vitro phosphorylation of the VP16 core domain by CK2 was
examined. Both wild type VP16 and VP16 with mutations at Ser375 were tested. Both
the wild-type and the mutant VP16s were found to be a substrate for CK2. MS/MS did

not detect acetylation of VP16 late in infection.

VP16 in phosphorylated at several sites late in infection

Addressing the phosphorylation by MS/MS demonstrated that at least three sites
were phosphorylated. These sites were Ser18, Ser 353 and Ser452. The presence of a
site at Ser452 could explain the phosphorylation of the a 371-490, a 371-454 and aa
410-490 fragments, but could not account for the phosphorylation of the 371-412
fragment observed for RPS infection. Thus, at least one additional site must be present in
this fragment. There was MS data to show unphosphorylated species derived from Ser
375 and Ser 400. There was no data for the site at Ser411. Phosphorylation of Ser 411
would be consistent with the data from both the MS and the peptide mapping. There

were no contradictions between the mass spectrometry data and the peptide mapping.

78

The observations from both the MS and the peptide mapping must be understood to only
represent the majority or at least significant minority populations of the VP16 late in
infection. It is possible that minority populations have gone undetected. It is possible,
maybe even likely that the phosphorylation during other stages of infection is very
different from what is presented here. It must be taken into account that for at least one
of these serines (Ser375), the function that it has been implicated in takes place early in
infection. Maybe only the minority of protein required for this activity is modified and
subsequently not detected by these methods.

In vitro phosphorylation of the core domain of VP16 was shown not to depend on
Ser375. It was concluded that there is another CK2 site within VP16. While the in viva
phosphorylation site at Ser353 does conform to the CK2 consensus, no tests were
performed to show this to be the in viva phosphorylation site. The in vitro kinase data are
in disagreement with the published findings of O’Reilly et a1 (O'Reilly et al. 1997). They
present data that shows VP16 to be dependent on Ser375 for in vitro phosphorylation.
Here I show that the substitution of Ser375 with Ala or Pro does not abolish the in vitro
phosphorylation of VP16 by CK2. These results only show that Ser375 is not the sole
possible phosphorylation site in VP16 in vitro. To determine whether Ser375 is a CK2
site in vitro, it would be necessary to map the sites of phosphorylation of VP16 subjected
to CK2 phosphorylation, either by peptide mapping or by MS. At the very least, it would
be necessary to quantify the specific labeling of the wild type and the mutants in the in

vitro labeling assay.

79

Correlation between Phosphobase and the in viva phosphorylation site
mapping/redundancy of determination

0f the four identified phosphorylation sites, all were predicted to be
phosphorylation sites by the phosphobase prediction server. All of the 13 serines
predicted not to be phosphorylated by Phosphobase were determined not to be
phosphorylated in viva. 0f the 13 serines predicted to be phosphorylated, four were,
eight were not. The phosphorylation state of Ser419 could not be determined by either
peptide mapping or MS/MS . None of the threonines and tyrosines that were predicted to
be phosphorylated were phosphorylated. The summary of these results suggests that the
prediction server was quite reliable in predicting unphosphorylated sites under the
conditions tested. While most of the predicted phosphorylation sites were not found to be
phosphorylated under these conditions, it is possible that they are phosphorylated under

different conditions, such as earlier in infection or in different cell types.

Resolution of the minor peptides

VP16 derived from HeLa cells infected with the truncation mutants RP3 and RPS
were also phosphorylated, implying that at least one site of VP16 phosphorylation lies
between aa 1-412. Analysis of the phosphoamino acid complement of VP16 from either
of these three strains of HSV-1 clearly demonstrated that the only targets of
phosphorylation were serines. Also, peptide mapping of these strains clearly
demonstrated that no significant phosphorylation of VP16 occurred between aa 29-344.
The peptide mapping yielded evidence that phosphorylation took place in fragments

representing a 371-490, aa 371-453, aa 371-412 and aa 410-490. Some evidence also

80

suggested that the 410-454 fragment was phosphorylated. These experiments did not
determine whether this phosphorylation was the result of one or more phosphorylation
events, nor whether the 1-28 fragment was phosphorylated. The labeled fragment
detected in the trypsin digestion does not encompass the serine 375 described by O’Hare
and colleagues (O'Reilly et al. 1997). The fragment containing Ser375 (aa371-399)
would migrate at 3 KDa. No labeled species were seen at this molecular weight. But as
mentioned previously, it was not clear whether fragments of this size would resolve on
the gel. The issue of whether the small fragments of either trypsin or lysylendopeptidase
C digestions resolved on the gel was crucial for conclusions concerning phosphorylation
at several sites. The most critical fragments were the 1-28 and 344-370 fragments of the
lysylendopeptidase C digestion and the 371-399 fragment of the trypsin digestion. The
400-408 fragment of the trypsin digestion did encompass putative phosphorylation sites,
but it was clear that a fragment as small as this would not resolve. Furthermore, the
trypsin digestion would not give any information about the sites corresponding to the l-
28 and the 344-370 fragments of the lysylendopeptidase C digestion. These fragments
would be further digested and would clearly not be detectable on the gel. The molecular
weight marker expected to migrate at an apparent molecular weight of 2.9 KDa did
resolve on the Tris/tricine gel suggesting that the fragments of interest might resolve. But
as mentioned, small fragments are more likely to migrate aberrantly and it was possible
that the mobility of the marker was reduced or that one or more of the minor fragments
would migrate abnormally fast.

It was necessary to consider the ambiguous conclusions about the phosphorylation

status of the minor peptides in the lysylendopeptidase and trypsin before trying to draw

81

1.11

110111

611

V1

additional conclusions about specific phosphorylation sites. If it was hypothesized that
none of these minor peptides were in fact phosphorylated, the result of the
immunoprecipitation of the truncated VP16 suggested that all significant phosphorylation
sites of the 410-490 tryptic fragment of full length VP16 were present in a motif common
to the wild type and the two truncation mutants. The only common segment was a 410-
412 (LST(410.412)). Two of these were potential phosphorylation sites. This would not
exclude the possibility of additional sites in the activation domain, either in wild type
VP16 or in the truncated VP16 from RP3.

Another possibility was that phosphorylation of either of the two minor peptides
(aa 371-399 and aa 400-410) of the trypsin digestion was not detected. Phosphorylation
of any site within those peptides would be common to all three version of VP16. It could
not account for the phosphorylation of the carboxyterminal tryptic fragment of full length
VP16, but it would raise the possibility that RPS was labeled between 371 and 410. In
that case, it would not be clear whether VP16 was phosphorylated between 410 and 412.

Finally, it was apparent that phosphorylation of either of the two minor peptides
(a 1-29 and aa 344-3 70) resulting from the lysylendopeptidase C digestion went
undetected. The bulk of the phosphorylation of all versions of VP16 could be attributed
to phosphorylation in these peptides. The phosphorylation within the carboxyterminal

regions of VP] 6 might represent a small fraction of the overall phosphorylation of VP16.
Homologs and structural function of the identified phosphorylation sites

In order to gain insight in the function of these phosphorylation sites, one can

align the phosphorylated domains with the amino acid sequences of homologs of VP16

82

from (it-herpesviruses (Figure 8). This sequence alignment offers a different view of
these serines. While Ser375 is highly conserved throughout this group of proteins, Ser18,
Ser353 and Ser4ll are conserved only to a limited degree. Ser18 is conserved only in
between HSV-l and HSV-2, which already has an overall 90 % conservation. Serine 353
of HSV-1 aligns with threonines in EHV-l and -4. But despite being flanked by highly
conserved tracts, the putative phosphorylatable residue falls in a very poorly conserved
motif, even when compared with HSV-2. And although EHV-l and -4 both have a
conservation of the S/T-P motif, they are otherwise quite dissimilar. Serine 411 also only
has limited homologies in the related proteins. In this case, only HSV-2 and PRV have
conserved acceptors, although in these, the local environment is more conserved than in
the previous examples and may constitute a canonical phosphorylation site. Ser452 falls
in the carboxyterminal activation domain which is poorly conserved between the
homologs.

Interestingly, Ser353 of VP16 aligns well with serines in both Zhangfei and
Luman, if the HCF-1 binding motif (HBM) is used as a landmark (figure 9). The
sequence of the canonical HBM motif is E/D-H-x-Y and is conserved in both Luman and
Zhangfei. When aligned with the HBM of VP16, a conserved motif appears 7-8 a
aminoterminal to the first residue of the HBM. The sequence of this motif in VP16 is
PheThrThrSer and can be generalized from the alignment to (D-S/T-S/T-S where the last
S is invariant and (D denotes a hydrophobic residue. This motif is conserved between
mouse, rat and cow homologs of Zhangfei and Luman and with the HBM, the entire
motif becomes (D-S/T-S/T-S-x7-3-E/D-H-x-Y-s. The E/D-H-x-Y part of this motif has

been shown to be critical for interaction with HCF but the amino terminal residues of the

83

Figure 8: Phosphorylation site conservation in herpes homologs. A-D: Alignment of
identified phosphorylation sites in VP16 with VP16 homologs from other herpes viruses.
The phosphorylation sites of VP16 are indicated by an asterisk. VP16-2: HSV-2, VZV:
verocella zoster virus, PRV: pseudo rabies virus, EHV-l: equiid herpes virus 1, BHV -1:
bovine herpes virus 1, MDV: Marek’s disease virus.

84

PP. _ _DN
GG. . _DA
GGS_TNY
AAR_AEA
PPN_AF_F_
RRW.AAI
PPT_AAF_
PPD_GAA
PPT_GFF
PPS_CML
SSP.QVQ
AVH_SII
GGE_ASP
DDT.LAS
AAG_AAY
N_L_RPY
M_N_GRD
DDC_AAN
AAE.TDE
FFM_KFF
LL_.IYS
ED__RWM
4411464
1
2

6e 4:...

11VVVVV
.wwwmmmm

HHHHHHH
EEDDEEE
RRARRAK
MMR.ELH

VVV_PPV
ASH_AFH
EER_LLL

._____Q
SGP._PP
PEE_.TN
SSQ_RAP

TTP.GPP
TAHYGHH
FVKEAAS
SSVREDL
SSASASS
YYYYPYY
SSAAAAV
DDDEGEE
LLLMMML
KKKKKKK
AARIATS

339
345
301
344
376
361

VP16-2 337
WW

BHV-l
EHV-l

MN

VP16
WV

C

LL TQI

SS VAA

VV TLL

DD THT

TT ITAL
PI LAAT
PPRPPTS
AATASQI
TTLAART
_TF_VVT
SSGSRGA
LLCVVVV
RRTRTT.
RRPPAL.
TPPPAP.
HRRPPP.
__PTAV_
.QPPPPP
GGDDDDD
AAGCCGG
PSPPPPP
LLLLLLS

0933163
0994921
4333344

2

. 11
66 _ .
11VVVVV
PPZRHHD
VVVPBEM

m

LI...

441

VP16
VP16-2 442

DDPPPPS
PPPPPPS
LLA__.D
DDLPRDM
LLILLML
LLMLMML
GGAAAAA
EEEEEEE
IIIAPTV
TTSTTTT
SSSSTTT
GGGGGGG
YYYYYYY
NNNSNSN

___A_LP
NNRVIRP
KRNPAGS
TTESGGE
RRVPRIL
AGVPPRK
RRKRRKQ
SSACASV

2062794
66626900
3333333

2

. 11
66 . _
11VVVVV
.PPZRHHD
VVVPBEM

M

HBM

 

VP16 S Y S S P - S E A V M R E H A Y S R A R T
VP16-2 S Y S S! .E - G E S V M R E H A Y S R G R T
Luman s L N I- N - P C L v H H D H T Y S L P R E
Luman (B. taurus) v L N v D - P C L v Q H D H T Y S L S Q E
LZIP (M. musculus) V S R D4 -S - S S S I L H D H N Y S L P Q E
Zhangfei v G L R L F R o s P A G o H D Y A L P v G
Zhangfei v G L R L F R D s P A G D H D Y A L P v G
(R. norvegicus)
vzv A Y A v K H P Q E P - R H v R A D H P Y A K v v E
PRV A Y S R E Y --------- R D H T Y C R P P S
BHV-l A P A E A G G R - - - L A P E R E H S Y A R P R G
EHV—l A Y s D A H P A T P - L F P L A E H S Y S K R I G
EHv-4 A Y S D A H P A T P - L F P L A E H s Y s K R I G
MDV v Y S L s H P P N P Q L H v H K E H v H v Q K L E

 

Figure 9: Alignment of VP16 and the Luman family. A small motif aminoterminal to
the HBM is conserved between VP16 and Luman and its homologs, but not with the
viral homologs of VP16. The HBM consensus motif is E/DHxY. It is indicated with a

black box. The putative conserved aminoterminal motif has the consensus (D—S/TS/T-S

where (1) indicates a hydrophobic residue. The motif is indicated with a grey box and
labelled (DS3.

86

putative motif have never been tested for their role in HCF-1 interaction and whether this

motif is necessary for fimction remains to be shown.

Function of the identified phosphorylation sites.

None of the phosphorylation sites that were identified have been implicated in any
firnction of VP16. None of them are highly conserved within the herpes homologs and
none have been seen in functional mutagenic screens. But the sites may still offer clues
to their function.

Ser18 lies in a putative AD in the amino terminus of VP16. The domain is
conserved in the VZV homolog of VP16, 0rf10. This protein does not have the carboxy
terminal activation domain and relies solely on the amino terminal AD (Moriuchi et al.
1995). The aminoterminal activation domain of VZV 0rf10 can be functionally replaced
with the aminoterminal 25 a of VP16. In HSV-l, this domain is not required for
activation of IE promoters in cultured cells (Triezenberg et al. 1988a). The domain is
also dispensable for the formation of the VIC. However, the functional conservation of
the domain suggests that this domain might be involved in transcriptional activation
under different conditions. In this case, the phosphorylation at Ser18 could be involved
in the regulation of that function.

Ser353 lies adjacent to a region of VP16 required for the interaction of VP16 with
the Oct-1 and HCF-l. These proteins are required for the formation of the VIC, the DNA
binding complex of WI 6. None of the surveys of the VIC interaction domains has tested
mutations at Ser353 (Greaves et al. 1990; Lai et al. 1997). It is possible that the

phosphorylation of Ser353 affects the formation of the VIC.

87

The site is also adjacent to a region required for the interaction with the vhs
factor. The nearby Lys344 is a key determinant for interaction with vhs, but deletion
mutagenesis of the region clearly shows that Ser353 is dispensable for the interaction.
(Smibert et al. 1994; Smiley et al. 2001; Knez et al. 2003). Vhs is involved in the
degradation of mRNA early in infection and the interaction with VP16 may be involved
in the repression of the mRNase activity late in infection (Matis et al. 2001; Smiley et al.
2001; Taddeo et al. 2002). Because vhs and Oct-l are competitors for the interaction
with VP16, it is possible that the phosphorylation event at Ser353 regulates both
interaction by blocking the interaction of one of these factors, thereby favoring the other
interaction.

Ser411 falls in the disordered C-terrninus of the peptide used in the crystal
structure (Liu et al. 1999). It may simply have been disordered due to the proximity to
the carboxy terminus. Phosphorylation of this site is close to both the Oct-l and HCF-l
interaction sites and to the activation domain. This site could be involved in regulation of
either. The site marks the amino terminal boundary of what is considered the canonical
bipartite activation domain of VP16. The prevailing model proposes this segment of
VP16 to be an inert linker between the DNA binding core domain and the activation
domain of VP16. However, no thorough survey has been done on this putative linker and
again, the possibility remains that although not directly involved in the mechanism of
transcriptional activation, the site serves as a regulatory switch.

Ser452 lies between the two subdomains of the carboxyterminal activation
domain of VP16. Both of these subdomains have been examined in great detail

(Triezenberg et al. 1988a; Cress et al. 1991; Regier et a1. 1993; Shen et al. 1996a;

88

Sullivan et al. 1998) but Ser452 lies in a small motif between them that has not been
investigated. It is possible that the site is involved in regulation of the transcriptional
activation by these domains.

Ser18, Ser353 and Ser452 are all followed by a proline. The SerPro (SP) motif is
a target site for a group of kinases typified by the Jun kinase (JNK). The function of the
phosphorylation of SP sites has been associated with the binding of Pin1
prolylisomerases, which in turn can cause conformational changes in the protein and

subsequently may serve as a regulatory event.

Reports of phosphorylation of VP16.

The identification of phosphorylation sites in VP16 during infection and in the
virion is consistent with observations by several groups. The literature on the
phosphorylation of VP16 has shown that VP16 is phosphorylated in infection (Gibson et
al. 1974; Wilcox et al. 1980; Morrison et al. 1998). Morrison et al immunoprecipitated
VP16 from virion preparations and demonstrated that VP16 was phosphorylated
(Morrison et a1. 1998). I also found phosphorylated VP16 in the virion. They also saw
phosphorylation of VP16 at IE stages of infection. I was unable to see detect
phosphorylated VP16 this early in infection. However, they infected the cells at an MOI
of 50, providing five to ten fold higher amounts of VP16 as an immunoprecipitation
target than was used in my experiment.

The literature is in disagreement about the in vitro phosphorylation of VP16 by
CK2. There is evidence that bacterially derived VP16 is a CK2 substrate in vitro

(O'Reilly et al. 1997). This is countered by evidence that VP16 derived from virion is

89

not a CK2 substrate (Morrison et al. 1998). One might argue that this difference could be
explained by the prior modification of the CK2 sites of the virion-derived VP16. Both of
these studies find that VP16 is a PKA substrate. O’Reilly et al further report that the loss
of the phosphorylation site at Ser3 75 abolishes phosphorylation by CK2 and that the site
must be the only in vitro CK2 site in VP16. In my hands, I confirm that bacterially
expressed VP16 can serve as a substrate for CK2. However, in contrast to the findings of
O’Reilly et a1, (O'Reilly et al. 1997) the loss of Ser375 does not abolish phosphorylation
of VP16 by CK2 and therefore can not be the sole CK2 site in the protein. These results
are clearly contrasting and because of the similarity of the experimental procedures, they
are hard to reconcile. I believe my argument is bolstered by the fact that there are several
additional CK2 consensus sites in VP16. In the in vitro kinase assay, it is likely that at
least one of these sites would serve as a substrate.

An attempt of identifying the phosphorylation sites of VP16 in viva was done by
O’Reilly et al. The primary difference between their approach and mine was the source
of the material. I chose to examine VP16 derived from cells infected with HSV-l, while
they looked at VP16 that had been expressed from a plasmid in HeLa cells. The results
from both projects were to some extent similar, but there were also significant
differences. Both approaches employed the lysylendopeptidase mapping of VP16 to
identify the region of phosphorylation of VP16. In both cases, a single labeled fragment
was recovered from the digestion, corresponding to the aa371-49O fragment. The
phosphorylation of the 371-413 fragment of VP16 from infection with the RPS strain in

my experiment was mirrored by the observation of phosphorylation of the 371-411

90

fragment derived from a similar truncation mutant, expressed in HeLa cells. They also
saw phosphoserine as the only phosphorylated residue in VP16.

O’Reilly et al did not test the effect of the Ser375Ala mutation on
phosphorylation in viva. Rather, they examined the phosphorylation of Ser37SAla
mutants in in vitro kinase assays. They either used CK2 or HeLa nuclear extract as the
source of the kinase. They did not see a major effect of the alanine substitution on the
overall phosphorylation of VP16 by the nuclear extract. Furthermore, they also saw
phosphorylation of the 371-411 by the nuclear extract fragment of endolysylpeptidase C
digestion of both the wild type and the Ser375Ala mutation. Regardless, they took a
small difference in migration between the two species to indicate that Ser375 was in fact
a phosphorylation site. O’Reilly et al supported this by showing that VP16 can be
phosphorylated by CK2, but VP16-Ser37SAla can not. They proposed that the
phosphorylation of the fragment, despite the loss of Ser375, was due to additional
phosphorylation, presumably at either Ser400 or Ser411.

It is my contention that the phosphorylation data presented by O’Hare et al can be
interpreted in a manner not consistent with phosphorylation at Ser375. This
interpretation is independent of the source of the material. Their conclusions rely on two
results, the in vitro CK2 phosphorylation data that I have discussed above and the
assertion that a small difference in migration between two the lysylendopeptidase
fragments of VP16 can be attributed to the loss of a phosphorylation site. They make this
conclusion without addressing the migration difference experimentally. While this
difference in migration could be a phosphorylation effect, it could also simply be due to

an alteration in gel retention, for instance due to the loss of the hydroxyl group of Ser375.

91

They report that the specific labeling of the phosphoamino acids of the mutant VP16 is
reproducibly less than that of the wild type. This would be consistent with the loss of a
phosphorylation site, but they do not rigorously address the quantification of this result.
They also chose to downplay the major finding, that no significant loss of
phosphorylation of the Ser37SAla mutants by the nuclear extract is observed.

Regardless of the phosphorylation data, the results of the O’Hare group and others
(Greaves er al. 1990; Lai et al. 1997; O'Reilly et al. 1997) show clearly that Ser375 is
required for the interaction between VP16 and Oct-1 and for the activation of VP16-
responsive genes. Replacement of Ser375 with alanine, aspartate or glutamate (Greaves
et al. 1990; Lai et al. 1997; O'Reilly et al. 1997) or with proline (Greaves et al. 1990)
renders the protein unable to interact with Oct-1, unable to form the VIC and unable to
activate the IE promoters. Replacement with threonine does not diminish this interaction
and does not reduce the transcriptional potential of the protein (Greaves et al. 1990;
O'Reilly et al. 1997). Whether phosphorylation of Ser375 is required for this activity is
not clear. In fact, the formation of the VIC is possible with factors entirely expressed in
E. coli and without the use of a eukaryotic nuclear extract (Babb et a1. 2001). Assuming
that the factors are not phosphorylated in the bacterial expression host, it is reasonable to
conclude that phosphorylation of VP16 is not required for this interaction. Even so, the
phosphorylation of VP16 at Ser375 could increase the affinity of VP16 for Oct-1 and

support the formation of a stronger VIC.

92

Chapter Three

Mutational analysis of the potential phosphorylation sites in herpes simplex

infection at serine 375 and serine 411.

Introduction

The function of neither Ser375 nor Ser411 has been adequately described in
the context of viral infection. In the previous chapter, a number of phosphorylation
sites were identified. None of the sites have been examined in infection previously. The
function of these sites need to be determined. Ser375, which was expected to be
phosphorylated, was found not to be. This residue has been shown previously to have an
important role in the formation of the VIC complex in vitro and in transcription in viva,
when VP16 is expressed from a plasmid (Greaves et al. 1990; Lai et al. 1997). The
function of the residue has not been examined in infection, but if the function of VP16 in
infection mimics that of VP16 in transfection, it would be reasonable to suggest that
Ser375 is essential for the expression of the IE genes. Because Ser375 has been shown to
be important for interaction between 0ct1 and VP16, and thus for the expression of the
IE genes, we hypothesized that a virus with mutations at Ser375 would be deficient in IE
gene activation and thus in growth. A reduction or loss of proliferation had been
observed previously in other transcription-deficient versions of VP16. Both the RPS
strain used in this study and the in1814 strain, which contains a 4 amino acid insertion at
aa379, are deficient in IE gene activation as well as in growth. In the case of RPS, the
strain is viable, but the yield of infectious particles is reduced 100 fold, as compared to

KOS (Tal-Singer et al. 1999). Outside of a complementing cell line, in1814 replicates

93

poorly, although the proliferation is improved if infections are done at a high moi (Ace et
al. 1989). Based on the phenotypes of these two strains, one might expect that if a
mutation at 375 affects IE gene transcription, it might also affect the growth of the virus.
The phosphorylation of Ser411 site has not been examined previously. But it is
known that VP16 purified from virion is a PKA substrate in vitro. Ser411 lies in one of 6
serine consensus sites for PKA phosphorylation. But while the Phosphobase prediction
did suggest that this site was a good target for phosphorylation, its position in the linker
between the core domain and activation domain does not lend itself as a obvious site for
regulation. Notably, both the core domain and the activation domain is capable of
functioning without this particular domain. Just as Ser375, it is also most likely a surface
exposed residue. In the crystal structure of the core domain, adjacent residues are
located at the bottom of the “Chair”, although Ser4ll is not ordered in the structure and
the exact position is not known (Liu et al. 1999). The site is poorly conserved among the
herpes virus homologs (Figure 8). Any function of the region would therefore have to be
host-specific, either because it is not required in other animal models or because the
firnction is carried out in a different manner. Regardless of these facts, the site appeared
to be robustly phosphorylated and was in proximity of both the activation domain and the
VIC formation domain, both of which are regions that are putative targets of regulation.
This site was identified using the peptide mapping approach, even though the MS failed
to cover it. However, because the carboxyterminal lysylendopeptidase C fragment of
VP16 from RPS was labeled and the MS found no evidence of phosphorylation at the
only other two serines in the fragment, it was most likely that Ser411 is a phosphorylation

site.

94

The phosphorylation sites that were found at Ser18, Ser353 and Ser452 late in
infection are obvious choices for a mutagenic study. However, these sites were
discovered at such a late time in the course of this project that it was not feasible to
generate and test mutants of these sites in the virus. These sites should be examined at a

later date, as their phosphorylation may take part in regulating functions of VP16.

Determining the effect of substitution of Ser375 and Ser4ll on the proliferation and
transcriptional phenotype in infection.

To test the roles of Ser375 and Ser411 in viral infection, mutations of the
corresponding codons were introduced into the viral genome. An alanine substitution of
Ser375 had little or no effect on the phosphorylation level of VP16 and little or no effect
on the growth of the virus in tissue culture. The mutation had a significant effect on the
expression of the IE genes in infection when the transcriptional effect of VP16 was tested
directly. When other factors of the viral expression cascade were allowed to act on the
VP16-responsive genes, this defect was masked. To examine the function of
phosphorylation of Ser411, a panel of three virus strains were generated carrying alanine,
glutamate and threonine substitutions at Ser411, respectively. These mutations had no

apparent effect on the proliferation of the virus nor on the activation of the IE genes.

Results:
Function of Ser375 in viva.
To determine whether Ser375 is involved in the function of VP16 during infection

by HSV-l, a virus strain was generated in which Ser375 of VP16 was replaced with an

95

alanine. The strain was named SJ02 (Figure 3). The presence of the mutation in the
recombinant viral genome was confirmed by sequencing a PCR product spanning the
locus of interest amplified from the viral DNA. Southern blotting with a probe spanning
the open reading frame and parts of the 3’- and 5 ’-UTR demonstrated that VP16 had been
inserted into the VP16 locus of the VP16-null strain 8MA. The mutant protein was
properly expressed in infected cells, and could be immunoprecipitated and detected by
immunoblotting using an antibody directed against the activation domain (LA2) (See
figure 10B).

To determine whether the mutation affected the phosphorylation of VP16, the
gross level of phosphorylation of VP16 isolated from HeLa cells infected with SJ02 was
compared to that of VP16 from KOS-infected cells. This determination was done
similarly to the previous study of KOS, RP3 and RPS. VP16 was immunoprecipitated
from HeLa cells infected with SJ02 or with KOS 8 hours post-infection in the presence
of [32P]-labeled orthophosphate. The monoclonal antibody LP1 and heat-killed S. aureus
cells were used for the immunoprecipitation. The precipitated protein was eluted from
the antibody matrix at pH 12.5 and separated on a 10% SDS-PAGE gel. The proteins
were transferred to a nitrocellulose membrane and visualized by autoradiography. The
identity of the precipitated proteins was verified by immunoblotting, using the rabbit
polyclonal serum C8. The autoradiogram in figure 10A reveals phosphorylated bands at
65 KDa, corresponding to the expected size for VP16. This band corresponds to the
bands detected in the immunoblot, which shows that the phosphorylated species is VP16.
The experiment demonstrated that VP16 derived from SJ02 is phosphorylated.

Furthermore, there was no apparent alteration in the level of phosphorylation of SJ 02,

96

Figure 10: Loss of Ser375 does not affect the phosphorylation of VP16 late in infection.
HeLa cells were infected with KOS and SJ02 at 5-10 moi in the presence of [32P]-
orthophosphate. Cell lysates were prepared fi'om each infection at 8 hpi. VP16 was
immunoprecipitated from each lysate with the monoclonal antibody LP1. A: The
immunoprecipitated material was separated on an 10% SDS-PAGE gel, transferred to a
nitrocellulose membrane and visualized by autoradiography. A labeled band was
observed at 65 KDa in both the KOS and the SJ02 samples. These bands corresponded
to the expected migration of full length VP16. The 36 KDA band was observed
consistently, the identity of the band is not known. B: The identities of the labeled bands
in panel A were verified by immunoblotting of the same membrane. The nitrocellulose
membrane was probed with the polyclonal antiserum C8. In both samples, the
immunoblot identified species corresponding to the migration of the labeled samples in
the autoradiogram. Minor bands were observed at 150-160 KDa in both KOS and RP3.
These bands did not comigrate with a [32P]-labeled species. C: An aliquot of the
immunoprecipitated VP16 from cell lysates were digested with lysylendopeptidase C
(EKC). The digestion products were separated on a 16.5% tris/tricine gel, then
transferred to nitrocellulose membrane, The labeled bands were visualized by
autoradiography. A band of 18 KDa was observed in both the KOS and the SJ 02 sample
These bands correspond to the expected migration of the aa 371-490 carboxyterminal
fragments of VP16 observed previously. D: A sample of the [32P]-labeled
immunoprecipitate was separated on a 10% SDS-PAGE gel and transferred to PVDF
membrane. The labeled proteins were visualized by autoradiography and the 6SKDa
species was excised and the protein was hydrolyzed in 6N HCl for 1 hour at 100°C. The
samples were mixed with phosphoserine, -threonine and -tyrosine standard and separated
by thin layer chromatography on a cellulose plate at pH 2.5. The phosphoamino acid
standards were visualized by staining with ninhydrin (indicated by a stippled outline) and
the [32P]-labeled species were identified by autoradiography. The only labeled species
in each sample co-migrated with the phosphoserine standard. Numbers to the left in
panel A, B and C indicate the migration of a molecular weight marker.

97

 

 

 

 

 

 

 

 

 

 

148
98
65
36
D
m N m N
o O 0
an a S a
7"" '1 O pSer (~-
65 - "v r \n’
50_ «a - i‘, pThr/ I )
_ #3 \’ \"
36 w pTyr
17 :E&,_“ '
In"
1*: I“
6 _ . .
1.. I
4 -iv ‘
.. (’4

 

 

 

98

 

 

when compared to KOS. From this experiment we conclude that Ser375 is not necessary
for phosphorylation of VP16 and Ser375 is not a major site of phosphorylation at late
times in infection. This is in agreement with the findings in chapter 2, where no
phosphorylation of Ser375 was found.

In this experiment, an additional phosphorylated species co-precipitated with
VP16. This protein has an apparent molecular weights of 36 Kda. It most likely
correspond to the 36 KDa band observed in the immunoprecipitation of VP16 in chapter
2. The 145 KDa band that was observed then was not observed in this experiment. The
36 KDa species was not detected in the immunoblot, indicating that the labeled species

were not derived from VP16. It was not investigated further.

Phosphorylation of the EKC fragment of SJ02

The overall [32P]-labeling of the intact, length protein isolated from SJ 02 infected
HeLa cells did not seem to be affected by the mutation. However, results described in
chapter 2 demonstrated that some, but not all phosphorylation of VP16 occurred in the
carboxyterminal lysylendopeptidase fragment of VP16. If the sites at Ser18 and Ser353
that were identified in chapter two were phosphorylated much more abundantly, the loss
of a less abundantly phosphorylated site would not be detectable. Such a loss could
possible be detected if the signal from the other sites were removed. To determine
whether the Ser375Ala mutation caused a loss of phosphorylation in the 371-490
fragment, the lysylendopeptidase C digestion from chapter 2 was repeated on the SJ02

substrate.

99

As previously, HeLa cells were infected with KOS or SJ 02 at an MOI of 5. The
infection was done on the presence of [32P]-orthophosphate and the infected cells were
lysed at 8 hpi. VP16 or VP16-S37SA was immunoprecipitated from the lysates using the
monoclonal antibody LP1. An aliquot of the immunoprecipitated material was separated
on a 10 % SDS-PAGE gel. The labeled band corresponding to the predicted migration of
VP16 was located by autoradiography of the wet gel. These bands were excised and the
protein within the slice was digested with lysylendopeptidase. The resulting fragments
were separated on a 16.5% tris/tricine gel, then transferred to a nitrocellulose membrane
and the radiolabeled peptides were visualized by autoradiography. The digestion of both
the KOS and the SJ 02 samples gave rise to a band of an apparent molecular weight of 18
KDa (Fig 10C). This band was identical to the one observed in the experiments in
chapter 2. As in the labeling of the full length protein, the intensity of labeling of the
peptide was similar between the two samples. This experiment showed that both wt and
mutant VP16 recovered from late in infection are phosphorylated within the
carboxyterminal 120 amino acids. Assuming that the mutation did not cause gross
alteration in the pattern of phosphorylation, this implies that the phosphorylation sites in
the carboxyterminal fragment is conserved, and thus that Ser375 is not a major site of

phosphorylation late in infection of HeLa cells.

Phosphoaminoacid analysis of Ser375Ala.
It is possible that the loss of one phosphorylatable residue in a consensus
phosphorylation site could be compensated for by another. This is particularly a concern

with adjacent Ser/Ser or Set/T hr pairs. Here, both Ser375 and Thr376 represent

100

consensus CK2 sites. One might imagine that even though the kinase has a higher
affinity for Ser375, in the absence of serine at 375, it will phosphorylate Thr376.

To rule out the possibility that the kinase altered specificity to the adjacent
threonine of the CK2 site, or indeed that the Ser375 mutation caused any other threonine
or tyrosine to become phosphorylated, phosphoamino acid analysis was performed on the
VP16 derived from the SJ02 infection. An aliquot of the [32P]-labeled,
immunoprecipitated VP16 from KOS infection and SJ02 infection described in the
previous section was separated on a 10% SDS-PAGE gel and transferred to a PVDF
membrane. The radiolabeled band was visualized by autoradiography and then excised
from the membrane. The protein was hydrolyzed in 6N HCl at 100°C, and
phosphoserine, phosphothreonine and phosphotyrosine standards were added to each
sample. The resulting mixtures of amino acids were separated on a cellulose plate by 1D-
TLC at pH=2.5. The radioactive amino acids derived from the hydrolysis were
visualized by autoradiography and the standards were stained with ninhydrin. As shown
in figure 10D, only a single spot was detected in each sample. These spots corresponded
to the migration of the phosphoserine standard indicating that in either strain, serine
remains the only phosphorylated residue. The conclusion was that no compensatory
phosphorylation is taking place at the threonine adjacent to Ser375 or at any other
Threonine or Tyrosine.

These experiments confirm that the major phosphorylation site(s) of both KOS
and SJ02 lie within the last 120 a and that Ser375 is not a significant phosphorylation

site of VP16 late in infection. These experiments do not entirely rule out the possibility

101

that in the KOS infection, VP16 phosphorylated at Ser375 is present, either at other times

in infection or at low levels late in infection.

Growth kinetics of SJ02.

To examine the hypothesis that the loss of the phosphorylation site at Ser375 led
to a loss of activation of the IE genes and thereby to a loss of viability of the virus, the
single stage growth kinetics of SJ 02 was examined.

The single stage growth curve measures the rate of proliferation and the peak
yield of viral progeny in a single round of infection. Vero cells in p60 tissue culture
plates were infected with KOS or SJ 02 at an MOI of 10. In parallel, a set of cells were
infected with RPS at an MOI of 0.1. This was done to compensate for the 100 fold lower
particle/pfu ratio in RPS and ensured a similar particle/cell ratio in all three infections.
Triplicate samples of infected plates were harvested at 4, 8, 12, 16, 20 and 24 hpi,
respectively. The titer of each sample was determined and plotted as a function of time
after infection. Figure 11 shows that the grth kinetics of SJ02 are similar to those of
KOS and quite unlike those of RPS. At 4 hpi, infectious virus was present at comparable
levels in all three infections (1x104-4x104pfu/mL). Some of these may represent
inoculum virions that adhered to the cells but did not enter. The titers at 8 hpi are higher
than at 4 hpi for all three strains. This shows that the time of initial emergence of the new
virions is similar between the three strains. While the yield of infectious virus was
somewhat higher in the SJOZ infection than in the KOS and RPS infections, this
difference is marginal and does not represent a significant trend. At 12, 16, 20 and 24

hpi, the amount of KOS and SJ 02 progeny increased at similar rates, starting to taper off

102

at 5x107-1x108. The yield of virus from the RPS infection at these time points was 1.5 to
two orders of magnitude lower, reaching 106 pfu/mL at 24 hpi. The experiment shows
that the rate of proliferation and the yield of progeny virus in SJ 02 infection very closely
resemble those of KOS. In contrast, RPS proliferation and virus yields are both much
lower. The overall conclusion from these results is that SJ 02 proliferation in Vero cells
is indistinguishable from KOS and clearly distinct from those of RPS, which is deficient

in IE gene expression.

IE gene expression assay

The results of the growth curve seemed to contradict the expected phenotype of a
virus carrying the Ser375Ala mutation in VP16. The loss of promoter binding in vitro
and decrease in activation of the IE gene promoters by VP16-Ser375SA when expressed
by transfection, is well documented (Greaves et al. 1990; Lai et al. 1997; O'Reilly et a1.
1997). One possible explanation is that the Ser37SAla mutation does not disrupt IE gene
expression during infection as it does when VP16 is expressed from a plasmid.
Alternatively, it may be that the defects in the IE gene expression caused by this mutation
affect viral growth differentially than the defects caused by deletion of the activation
domain in RPS or the insertion in the HCF-1 interaction motif in the in1814 strain. To
resolve this conundrum, one of the basic assumption about VP16 action in early infection
had to be rejected. For instance, the transcription of the IE genes might in fact be
dispensable for viral proliferation. Alternatively, VP16 might be dispensable for IE gene

expression. Thirdly, the IE genes might require VP16 to be expressed but could do so

104

108 l

107 .

Titer/(pfu/mL)
S
O\

    

 

 

105 I +KOS
-I-RPS
+8102

104 - . . .

0 10 20 3O

Time/(Hours post-infection)

Figure 11: Loss of Ser375 does not affect the proliferation of HSV-1. Single stage
growth curves were performed on KOS, RPS and SJ02. HeLa cells were plated on p60
tissue culture plates (p603) at 2x105 cells per p60. 24 hours after seeding the cells, tripli-
cate samples for each time point were infected with KOS or SJ02 at an MOI of 10 or
RPS at an MOI of 0.1. Samples were harvested at 4 hour intervals for 24 hours. The titer
was established for each sample and the titer plotted as a function of time after infection.
The error bars represent the standard deviation of the triplicate samples.

103

independently of VIC formation. Finally, the VIC formation may be independent of
Ser375 in infection, even though it is clearly dependent in in vitro experiments.

To begin to differentiate between these four possibilities, IE gene activation
during infection with KOS, SJ 02 and RPS was tested using a CAT-reporter assay. In the
assay, reporter plasmids were used that express a CAT gene product under the
transcriptional control of the promoters from the HSV-l IE genes ICPO, ICP4, ICP22/47
or ICP27. The activation of these reporters was tested in HeLa cells during infection with
SJ02, KOS and RPS. These reporters were introduced individually into HeLa cells by
liposome mediated transfection. After 24 hours cell transfected by a given reporter
construct were pooled and plated in six-well plates. 48 hours after transfection, the cells
were infected with KOS or SJ02 at an MOI of 5-10, or RPS at an MOI of 005-01.
Mock infection was done in parallel. The infected cells were harvested at 2 hpi and
lysates were generated by disrupting the cells by multiple freeze/thaw cycles. The cell
lysates were then tested for CAT activity.

A representative set of results are shown in figure 12. Each panel of the figure
shows the CAT enzyme activity for one of the four reporters. As expected, the IE gene
expression of the IE genes in the KOS infection was 2-7 fold above the mock infection in
all four strains. Also as expected, infection with RPS did not activate any of the reporter
genes significantly above the mock infection. For SJ02, the activation varied between
comparable to that of KOS and to 50% of that of KOS. In no instance did the activation
of the IE genes fall to the level of RPS and mock infected cells. These results lead to the
conclusion that activation of the IE genes in SJ02 may be slightly reduced but remains

significantly more robust than that of RPS. In other words, the unexpected expression of

105

ICPO ICP4

A 1 A 0.6
3 14 3
.5 ' E 0.5
g 1.2 78,
is 1.0 a 0-4
E E
3 0.8 \e 0.3
\ a
,1; 0.6 I; 0.2
T3 0.4 g
0.1
12 0.2 '2
o 0.0 m m N x 0 0.0 m m N x
o g 0 g o g g 8
X (7) E X (D E
ICP22/47 4 5 ICP27
a 0.8 3 4'0
3 07 .s '
s ' E 3.5
E 0-6 E 30
2 o '
g 0-5 g 2.5
‘3" 0.4 E 2.0
g: 0.3 g 1.5
“g 0.2 8 1.0
E 0.1 g 0.5
0 0'0 m In N .1: 0'0 8 E S E
n. 0 0
g I! 3, g 3‘ m B s

Figure 12: Loss of Ser375 has a marginal effect on the activation of the IE genes during
infection. HeLa cells were transfected with pANS, pAN6, pAN7 or pICP27-CAT. These
reporters express the CAT-gene from the ICPO, ICP4, ICP22/47 or ICP27 promoters,
respectively. The cells were super infected in triplicate with KOS, RPS or SJ02 at an moi
of 10 or mock infected. Extracts were made from the infected cells 2 hpi. The extract
were tested for CAT activity, which is reported as pmole acetyl transferred per minute per
11g of total protein. The error bars represents the standard deviation between the tripli-
cates.

106

the IE genes in SJ02 would explain the wild type growth phenotype of the strain. But
while the expression of the IE genes in SJ 02 may explain the growth phenotype, it seems
incompatible with the transcriptional phenotype of the Ser375Ala mutation in transfected

VP16.

Transcription of IE genes by VP16-Ser375Ala requires protein translation.

To confirm that the expression of the IE-CAT reporters was in fact stimulated by
the incoming VP16 and not by proteins expressed subsequent to virion entry, it was
necessary to prevent the expression of proteins during infection. Other viral genes,
primarily ICPO, are capable of partially complementing loss of IE gene expression when
VP16 function is lost (Mossman et al. 1999). These secondary mechanisms may mask
the effects of the mutation of SI 02.

To block the translation of the IE proteins cycloheximide was added to the grth
medium of the host cells prior to infection. The IE gene expression assay was repeated in
the presence of cycloheximide using the same reporter construct and the same three
HSV-l strains. To permit the translation of the CAT reporter protein, the cycloheximide-
containing medium was removed after two hour of infection and replaced with medium
containing the RNA synthesis inhibitor actinomycin D. This allowed for the translation
of the accumulated reporter gene messages, while preventing any subsequent
transcription of the IE reporters. Lysates were prepared from the infected cells after two
hours incubation with actinomycin D. The lysates were assayed for CAT enzyme.

Representative results for this experiment are shown in figure 13.

107

ICPO ICP4

 

A A 1.2
E E
.E .E 1.0
E E
3.03 g 0.8
E, e 0.6
T E
E E 0.4
‘65 ‘6
cu m

0.2
'3: '2
0 0 0.0

3‘ «7) E x tn 5
ICP22/47 ICP27

A A 0.30
E E
.E .E 0.25
E E
g g 0.20
9; 9; 0.15
T E
g E 0.10
It '3
I— I— 0.05
< <
0 0 0.00

 

min
Q
So:

min
0.
Sn:

SJ02
Mock
SJ02
Mock

Figure 13: Loss of Ser375 severely diminishes IE gene activation in the absence of pro-
tein synthesis. HeLa cells were transfected with pANS, pAN6, pAN7 or pICP27-CAT.
These reporters express the CAT-gene from the ICPO, ICP4, ICP22/47 or ICP27 promot-
ers, respectively. 48 hours after transfection, cycloheximide was added to the growth
medium. After 30 minutes, the cells were superinfected in triplicate with KOS, RPS or
SJ02 or mock infected at 10 moi. 2 hpi, the cycloheximide was removed and replaced
with actinomycin D. Extracts were made from the infected cells 4 hpi. The extract were
tested for CAT activity, which is reported as pmole acetyl transferred per minute per 11g of
total protein. The error bars represents the standard deviation between the triplicates.

108

In this experiment, KOS activated the IE reporters 3-5 fold over mock infection.
This is similar to what is observed in the absence of cycloheximide. Likewise, the
activation domain truncation of RPS renders that virus unable to activate any of the
reporters here as it did in the absence of cycloheximide. But in contrast to the first set of
experiments, SJ02 was unable to activate the reporters, leaving it as or almost as
deficient as RPS for all four reporters.

The conclusion from this experiment was that VP16-Ser37SAla by itself was
unable to activate the IE genes. The combined conclusion from these two experiments
was that protein synthesis is required for IE gene expression during infection with SJ02.
This is most likely is due to stimulation of expression of the IE reporter constructs by the
products of the viral IE genes themselves

In summary, the investigation of the phosphorylation and function of Ser375 of
VP16 in infection with HSV-l showed that loss of the serine at 375 did not affect the
level of phosphorylation of VP16 late in infection. This led to the conclusion that Ser375
was not a significant site of phosphorylation when VP16 was examined late in infection.
This is consistent with the findings in chapter 2 which found no evidence of
phosphorylation of Ser375. The results also showed that the Ser375Ala mutation of
VP16 did not affect the grth of the virus in Vero cells. Furthermore, the expression of
the IE genes under standard conditions was largely unaffected. However, in the absence
of protein synthesis, a significant defect in IE gene expression was revealed. The
conclusions of these experiments are that the expression of the IE genes during infection
with SJ02 was stimulated by a factor expressed subsequent to infection. These

conclusions were in agreement with published findings that showed mutation of the

109

Ser375 led to a failure to form the VIC in vitro (Greaves et al. 1990; Lai et al. 1997;
O'Reilly et al. 1997). It also demonstrated that during infection the IE genes can be

activated by a mechanism that does not require this interaction.

Phosphorylation and function of Ser4ll.

Ser411 is one of the potential phosphorylation sites that were discovered in
chapter 2. This site was found by peptide mapping and appears to contain a significant
portion of the label. To test whether mutations of this site had any effect on VP16
function during infection a set of mutant viral strains were generated (Figure 3). These
strains, S011, S012, S013, carry alanine, glutamate or threonine substitutions,
respectively, at Ser411. S011 was examined to detect changes in phosphorylation levels
and phosphorylation residue preference. All three strains were examined for their growth

phenotypes and their ability to activate the IE genes during infection.

Generation of viral strains carrying point mutations at Ser411

The function of the potential phosphorylation site at Ser411 had not been
addressed previously and no correlation had been found to demonstrate a connection
between mutations in this region of VP16 and changes in the function of VP16. A set of
mutant strains of HSV-1 were generated to determine whether substitution of Ser4ll with
alanine, glutamate or threonine would interfere with the function of VP16 during
infection.

The viruses were to be generated by homologous recombination into 8MA, a

strain lacking the VP16 gene. This strain is not viable in the absence of a complementing

110

cell line. In the previous section, the SJ02 strain was also generated from 8MA by
recombination and candidates were selected by their ability to reproduce outside of a
complementing cell line. This method had been successful in generating the recombinant
virus but there was some concern about possible side effects of the selection method. It
was possible that by only selecting viable strains, a selection pressure for complementing
second site mutations was introduced. In effect, this selection would favor increased
viability due to both the introduction of the recombinant VP16 gene as well as any other
spurious mutation within the viral genome. The concern over the possibility of a second
site mutation prompted a change in the strategy for generation of the these new viruses.
In order to avoid the undesirable selection pressure, it was decided to screen all potential
recombinants for the presence of a marker while complementing any mutant phenotypes
caused by the mutation with wild type VP16. An enhanced green fluorescent protein
(EGFP) tag was added to VP16 to serve this purpose. A virus strain (DGl) carrying wild
type VP16 with a carboxyterminal EGFP-tag had been generated and characterized
previously (Greensides 2002). The EGFP-tag had no negative effect on the growth of
this virus. By introducing the mutations at Ser4ll into a version of VP16 that carried the
EGFP-tag, it would be possible to screen for recombination events in a complementing
cell line. This would greatly reduce the probability of inadvertently picking
recombinants with second site mutations in a viability screen.

The mutations at 411 were introduced into the VP16-EGFP plasmid construct
using a PCR based mutagenesis method. The plasmid was sequenced to verify the
presence of the mutations. The mutant gene along with flanking regions was excised

from the plasmid and introduced along with purified viral genomic DNA from 8MA into

111

BHK-MMTV-VP16 by liposome mediated transfection. The cell line is a derivative of a
baby hamster kidney cell line (BHK), constitutively expressing VP16. This cell line was
chosen for the recombination because it is very amenable to liposome-mediated
transfection. The supematants of these transfections were harvested and used as
inoculum for the infection of Vero 16-8 cells in agarose-containing growth medium.
Because BHK cells do not give rise to well-defined plaques, Vero 16-8 cells were used
for the infection. Infectious foci in Vero cells can be readily distinguished from non-
infected cells. This reduces the probability of picking plaques that contain progeny fiom
multiple, conjoined foci. Individual recombinant strains were recovered by screening for
EGFP-fluorescence. The infected vero 16-8 cells were examined under a fluorescence
microscope and solitary plaques that expressed the EGFP-protein were identified.
Recombinant HSV-l was recovered from these plaques and used to infect Vero 16-8
cells. Re—infection, identification and recovery of EGFP-expressing plaques was repeated
sequentially on the same initial plaque to ensure that the resulting strain was derived fiom
a single recombination event. After the purification, larger stocks of each recombinant
virus were grown in Vero 16-8 cells.

To verify that the desired mutation was indeed present in each strain, viral DNA
was purified from an aliquot of the virus stock. A fragment of the VP16 gene spanning
the mutation site was amplified by PCR and the PCR product was sequenced. The
expression of the VP16-EGFP firsion protein was verified by immunoblot. Aliquots of
the viral stock were separated on a 10% SDS-PAGE gel and the proteins were transferred
to a nitrocellulose membrane. The presence of VP16 was determined by immunoblot,

using the rabbit polyclonal serum LA2-3. Aliquots of the wild type KOS, the EFGP-

112

23
V

11!

di

b<

tagged wild type DGl and of 8MA were added as controls (figure 14). The KOS sample
gave rise to the expected 65 KDa band, corresponding to full length VP16. DGl gave
rise to a 110 KDA band. While this band migrates slower than would be expected, based
solely on the molecular weight of the protein, it was taken to correspond to full length,
EGFP-tagged VP16. The 8011, S012 and 8013 strains gave rise to both the 65 KDa
band and the 110 KDa band. The presence of both bands was expected because these
strains had been grown in the presence of wild type VP16 in Vero 16-8 cells. This
resulted in the packaging of both versions of the protein. The VP16 null mutant, 8MA
was also grown in 16-8 cell and therefore carried wt VP16 in the virion. As expected, the
65 KDa band was detected in this sample.

The virus stocks used for this experiment had dissimilar titers, giving rise to
different levels of both wt and EGFP-tagged VP16. The S013 sample does not appear to
have any wild type VP16. This is most likely an artifact of the low titer of the 8013
stock used in conjunction with the unequal loading in the virions of VP16 expressed by
the cell and the virus. The low titer of the S013 stock observed here is aberrant; the
strain ordinarily gives rise to titers similar to those of the 801 l, 8012 strains. (Data not
shown). Titers of all three stocks ranges from 5x108 to 2x109, both when generated in
complementing cell lines and in Vero cells, similar to wild type titers. The experiment
verifies that the recombinant virus strains express and package EGFP-tagged VP16.

It is significant that both tagged and untagged VP16 was detected in the virion of
both S011 and S012. This shows that the Ser411Ala and the Ser41 1G1u mutations do
not affect the packaging of VP16 into the virion. In fact, it appears that the virus

preferentially packages the mutant VP16. This is most likely to reflect a difference in the

113

KOS
DC]
8011
$012
8013
8MA

 

148-
E . ”~—
98-1 E3
.1
65 ‘ ' w, a
50- a "
36-

 

 

 

Figure 14. Aliquots of virus stock from KOS, DGl, 8011, S012,
8013 and 8MA were separated on a 10% SDS-PAGE gel. S011,
S012, S013 and 8MA were derived from Vero 16-8, a complementing
cell line. The proteins were transferred to nitrocellulose and the pres-
ence of VP16 proteins was detected by immunoblotting with the
polyclonal rabbit anserum LA2-3. The migration of molecular weight
markers is indicated in KDa in the left side of the figure

114

levels of the two proteins in the host cell. The expression of the wt VP16 gene from a
gene in the infected cell is likely lower that that of the EGFP-tagged gene expressed from

the VP16 locus in multiple copies of the viral genome.

Phosphorylation of VP16 lacking Ser4ll

To examine whether the Ser411Ala mutation in 8011 affected phosphorylation of
VP16, HeLa cells were infected with KOS, S011 and DG1 at an moi=5. [32P]-
orthophosphate was added at 1.5 hpi and the cells were lysed after 8 hours of infection.
VP16 was immunoprecipitated from the lysates, using the LP1 antibody and eluted from
the antibody at pH 12.5. The eluted protein was separated on a 10% SDS-PAGE gel and
then transferred to nitrocellulose. The radiolabeled species were detected by
autoradiography and their identities were verified by immunoblotting the membrane with
the polyclonal rabbit serum C8.

As expected, a 65 KDa, [32P]-labeled band was detected in the KOS sample.
Similarly, as expected, a 110 KDa labeled band was detected in the DG1 sample. This
demonstrated that the EGFP-tag did not abolish phosphorylation of VP16. The S011
sample also contained a 110 KDa [32P]-labeled species. This species was similar to the
one observed in the DG1 sample. This showed that the Ser411Ala mutation did not result
in the loss of phosphorylation of VP16. The immunoblot verified the identity of the
protein in all three samples. (Fig 15A) It also showed that the protein levels of VP16
were similar in DG1 and SJ 02. The lower level of VP16 recovered from in the DG1 and
8011 samples, as compared to the KOS sample could have been be due to a reduced

efficiency of transfer of the protein to nitrocellulose. This may be a function of the larger

115

size of the protein or it may be that EGFP imparts a lower affinity of VP16 for the
membrane. This experiment showed that the loss of Ser411 did not affect the gross level
of phosphorylation.

Additional labeled bands, including those at 36 Kda and 140 KDA were
conserved between the three samples but most are not detected in the immunoblot. One
exception was a labeled band of 80Kda which appeared in the DG1 and S011 samples.
It also appeared in the immunoblot of those samples but did not appear in the KOS
sample. This may have represented a premature termination of VP16 translation or a
degradation product. A similar band was never observed in KOS and presumably
represented an artifact caused by the EGFP-tag.

The conclusion from the experiment was that the level of phosphorylation of
VP16—Ser411Ala derived from 8011 was no different that that of the wild type EGFP-
tagged VP16. But as was discussed with SJ02, the possibility remained that the loss of

Ser411 caused an alteration in the pattern of phosphorylation remained.

Phosphoamino acid complement in S011

Like in the case of the Ser375 site, the conservation of the bulk label of the S011,
could be attributed to an alteration of kinase specificity. In the case of Ser411, the
nearest neighbor, Thr412 might serve as an alternative acceptor of the phosphate group.
To examine this possibility, the phosphoaminoacid complement in S011 was examined.
VP16 was immunoprecipitated from HeLa cells infected with KOS, S011 and DG1,
respectively, in the presence of 32P-orthophosphate. The precipitated protein was

separated on a 10% SDS-PAGE gel and transferred to a PVDF membrane. The position

116

of the labeled bands were determined by autoradiography of the membrane and the bands
were excised from the membrane. The protein in the band was hydrolyzed in 6N HCL at
100°C for l h. The resulting amino acid mixtures were spiked with phosphoserine, -
threonine and —tyrosine standards and loaded on a cellulose plate. The samples were
separated by 1D TLC at pH=2.5 and the radioactive residues were detected by
autoradiography. The standards were detected by ninhydrin staining and used as a guide
for the radioactive spots. Figure 15C shows that each sample only gave rise to a single
radioactive spot. The outlines of the ninhydrin-stained standards are indicated by the
stippled lines. In each sample [32P]-label was detected at the phosphoserine position, but
not at the phosphothreonine or —tyrosine. As in the immunoprecipitation, the relative
levels of [32P]-label in each sample may be due to the lower efficiency of transfer to the
PVDF membrane. The experiment shows that the only phosphorylated amino acid in
8011 is serine. This rules out compensatory phosphorylation at Thr412, as well as at any

other threonine or tyrosine site in VP16.

Growth kinetics of 8011, $012 and 8013

Although the loss of the serine at 411 does not seem to affect the total level of
phosphorylation of VP16 in infection, it was identified in chapter 2 as a site of
phosphorylation. This phosphorylation may indicate that it is a site of regulation of VP16
firnction. Because no data existed to point to a specific function that this phosphorylation
might regulate, the first step to find such a firnction was to determine whether the loss

affected the viability of any of the mutants.

117

Figure 15: Loss of Ser411 of VP16 does not affect the overall phosphorylation of VP16,
nor alter the phosphoamino acid preference. HeLa cells were infected with KOS, DG1
and SOllat 5-10 moi in the presence of [um-orthophosphate. Cell lysates were
prepared from each infection at 8 hpi. VP16 was immunoprecipitated from each lysate
with the monoclonal antibody LP1. A: The immunoprecipitated material was separated
on an 10% SDS-PAGE gel, transferred to a nitrocellulose membrane and visualized by
autoradiography. A labeled band was observed at 65 KDa in the KOS sample. A 100
KDa band was observed in both the DG1 and S011 sample. These sizes correspond to
the expected migration of full length VP16, without or with the EGFP-tag, respectively.
Additional [32P]-labeled bands in the DG1 and S011 samples were also present in the
KOS sample. B: The identities of the labeled bands in panel A were verified by
immunoblotting the same membrane. The nitrocellulose membrane was probed with the
polyclonal antiserum C8. In all samples, the immunoblot identified species
corresponding to the migration of the labeled samples in the autoradiogram. C: Samples
of the [32P]-labeled immunoprecipitates was separated on a 10% SDS-PAGE gel and
transferred to PVDF membrane. The labeled proteins were visualized by
autoradiography, a 6SKDa species was excised and the protein was hydrolyzed in 6N
HCl for 1 hour at 100 C. The samples were mixed with phosphoserine, -threonine and -
tyrosine standard and separated by thin layer chromatography on a cellulose plate at pH
2.5. The phosphoamino acid standards were visualized by staining with ninhydrin
(indicated by a stippled outline) and the [32P]-labeled species were identified by
autoradiography. The only labeled species in each sample co-migrated with the
phosphoserine standard. Numbers to the left in panel A, B indicate the migration of a
molecular weight marker.

118

 

148-

981
65"-

50‘

36‘

 

 

 

 

 

1484
98 -

65-!

50-

36-

 

 

 

 

 

 

 

 

 

 

119

 

 

 

 

As described previously, the usual approach to study major grth defects in
herpes is to examine the kinetics of proliferation. To this end, stocks of S011, S012 and
3013 were grown in non-complementing Vero cells. The titer of these stocks were
determined and found to be similar to those of wild-type virus. Each of these stocks as
well as wild type KOS were used to infect a set of p60 plates seeded with vero cells at an
MOI of 5. A subset of plates of infected cells were lysed at 4, 8, 12, 18 and 24 hpi and
the progeny virus was extracted fiom the cellular debris. The titer of each of these
supematants was determined by plaque assay and the results were plotted as a function of
time (figure 16). The growth of the mutants mimicked that of the wild type closely. The
infectious virus harvested at 4 hpi represented a mixture of remnants of the initial inoculi
and newly synthesized virion. At 8 hpi, the amount of virion harvested from all strains
had increased by an order of magnitude from that of 4 hpi. None of the three mutant
strains lagged behind in initial emergence of virion. At 12, 18 and 24 hpi, the titer of the
mutant strains remained very similar to that of KOS. This showed that the rate of virion
synthesis was the same and it demonstrated a maximal yield of virion at 18 hpi. This
shows that the mutations at Ser411 did not affect any vital function of VP16, required for
infection of HeLa cells. These three mutant strains exhibit wild type growth and kinetics.
They all differ from the activation domain deficient strain RPS. The conclusion was that
any function that is associated with the putative phosphorylation site at Ser4ll is not vital

for the proliferation of HSV-1 in lytic infection of cultured HeLa cells.

120

_e
O
00
I

 

 

 

107 -
3
e 106 -
E
E
a
g: 105 -
104 -
103 ' I I I
o 10 20 30

Time/(Hours post-infection)

Figure 16: Loss of Ser4ll does not affect the proliferation of HSV-1. Single stage
growth curves were performed on KOS, 8011, S012 and S013. HeLa cells were plated
on p60 tissue culture plates (p603) at 2x105 cells per p60. 24 hours after seeding the
cells, triplicate samples for each time point were infected with KOS, 8011, $012 or
$013 at an MOI of 10. Samples were harvested at 4 hour intervals for 24 hours. The
titer was established for each sample and the titer plotted as a function of time after infec-
tion. The error bars represent the standard deviation of the triplicate samples.

121

Transcriptional activation of the IE genes by 8011, 8012, 8013

The growth curve did not reveal any loss of function of VP16, associated with
Ser411. Consequently, the search for a function had to be approached in a different
manner. Because the closest functional domains to Ser411 are involved in DNA binding
and in transcriptional activation, it seemed reasonable to investigate whether there was
any defect in IE gene activation. To this end, the CAT-reporter assays were repeated for
the S011, S012 and S013 in the presence and in the absence of cycloheximide. The use
of cycloheximide would take address the possibility that if Ser411 was indeed involved in
the IE gene activation by VP16, it might also be bypassed by the same unknown function
that bypassed the Ser375Ala defect of SJ02 in infection. Again, reporter constructs
expressing the CAT gene under the control of the IE-gene promoter from ICPO, ICP4,
ICP22/47 and ICP27 were used. The reporters were transfected into HeLa cells. 24
hours after transfection, the transfected cells were plated in p60 plates. 48 hours after
transfection, the cells were infected with KOS, 801 1, S012 or S013 at an MOI of 5 or
RPS at an MOI of 0.05. As a negative control, a set of plates were treated with a mock
inoculum. When testing expression of the CAT reporters in the absence of
cycloheximide, the cells were harvested and lysed at 2 hour post infection and the cell
lysates were tested for CAT activity. Figure 17 show a representative result of this
experiment. As expected RPS was unable to activate the reporter genes above mock
(background) levels in any of the constructs. Activation of ICPO and ICP4 in KOS
infection was 3-5 fold over that of mock and RPS. In the light of the previous reports, the
very low level of activation by KOS (20-40%) over background in the ICP22/47 and

ICP27 experiments here is unexpected. The results are most likely due to over-

122

ICPO ICP4

E45 35
C
gas E
%30 E3
E25 e
E20 $2
31.5 g
81.0 .21
I20.5 00
00-00,... m... 83:91:13
:ru-0 0 O
3&8ggg ¥mm882
ICP22/47 ICP27
A9
“2.5 E
E g8
.5 E7
£20 E6
.9 o
21.5 E5
o. \4
z 3'
$10 2,553
'5 82
50.5 '21
6 0o
000110 01005: 8359936
:v-r- O o
95888; 3":01882

Figure 17: Loss of Ser4ll does not affect the transcriptional activation of the IE genes
during infection. HeLa cells were transfected with pANS, pAN6, pAN 7 or pICP27-CAT.
These reporters express the CAT-gene from the ICPO, ICP4, ICP22/47 or ICP27 promot-
ers, respectively. The cells were super infected in triplicate with KOS, RPS, S011, 8012
or $013 or mock infected. Extracts were made from the infected cells 2 hpi. The extract
were tested for CAT activity, which is reported as pmole acetyl transferred per minute per
ug of total protein. The error bars represents the standard deviation between the tripli-
cates.

123

transfection of the reporter, giving rise to a very high background expression level. For
the ICPO, ICP4 and ICP27 reporter plasmids, the expression levels induced by the mutant
strains are most similar to that of KOS and distinctly different from that of RPS and
mock. This indicates that the activation of these genes are unaffected by mutations at
Ser411. In ICP22/47, the low level of activation over background precludes any clear
conclusion. The overall results of the experiment was that there was no evidence of any
significant defects in IE gene expression in 801 1, 8012 and $013.

To rule out the possibility that an IE gene expression defect was masked by the
expression of a complementing factor after infection had begun, the experiment was
repeated in the presence of cycloheximide. As previously, HeLa cells were transfected
with one of the CAT reporters driven by the four IE gene promoters, ICPO, ICP4,
ICP22/47 and ICP27. The transfected cells were treated with cycloheximide a half hour
before infection and for the first two hour of infection. The medium was switched to
actinomycin-containing medium to allow the accumulated transcripts to be translated.
The cells were lysed two hours afier actinomycin addition and the extracts were assayed
for CAT activity. The experiment was repeated for all four of the reporters.
Representative results of these experiments are shown in figure 18. Again, both controls
gave the expected result. KOS activated the IE genes 2-7 fold over the mock infected
cells. The RPS did not activate the IE reporter genes above the Mock level. The mutant
strains activated the IE gene reporters well above the background, to levels similar to
those of KOS. The ICP22/47 and ICP27 constructs that did not give conclusive results in
the absence of cycloheximide showed in this experiment that there was no deficiency in

IE gene expression on these reporters. The variation in activity between the SO strains

124

ICPO ICP4

§6 374
E5 5
g g3
%4 TE:
53 e2
E Q
:22 .21
a *5
:1 g
50 O0
mmrwmx mmFme
xmasaz xmsaaz
ICP22/47 ICP27
A25 as
3'3 3
£20 E4
in E
O 0
E15 Es
e e
:10 .32
.2 g
":85 51
< <
O 0
°gg:e2a 083:32a
O O
xmaaaz xmaaaz

Figure 18: Loss of Ser411 did not affect IE gene activation in the absence of protein syn-
thesis. HeLa cells were transfected with pANS, pAN6, pAN7 or pICP27-CAT. These
reporters express the CAT-gene from the ICPO, ICP4, ICP22/47 or ICP27 promoters,
respectively. 48 hours after transfection, cycloheximide was added to the growth medi-
um. After 30 minutes, the cells were superinfected in triplicate with KOS, RPS, $011,
8012, or 8013 or mock infected at 10 moi. 2 hpi, the cycloheximide was removed and
replaced with actinomycin D. Extracts were made from the infected cells 4 hpi. The
extract were tested for CAT activity, which is reported as pmole acetyl transferred per
minute per ug of total protein. The error bars represents the standard deviation between

the triplicates.

125

were not conserved between the reporter constructs, nor were they conserved in replicates
of the same reporters (not shown). The variations are not due to inherent differences in
the potency of the IE gene activation between the strains. This experiment fails to
demonstrate the existence of a factor, expressed subsequent to infection, which is capable
of masking defects in IE gene expression caused by loss of the phosphorylation site at
Ser411.

In summary, VP16 derived from infection with $011 is phosphorylated at levels
comparable to those of KOS. The site or sites of phosphorylation are all serine sites.
There is no evidence of altered site specificity of the kinase to compensate. Furthermore,
viral strains carrying alanine, glutamate or threonine substitutions at the Ser411
phosphorylation site proliferate with kinetics similar to that of KOS. These strains also
do not demonstrate any deficiency in IE gene expression, regardless of whether

subsequent translation is allowed.

Methods and Materials

Cell culture, virus preparation, in vivo labeling experiments,
immunoprecipitation, SDS-PAGE, immunoblotting, peptide mapping and phosphoamino
acid analysis was performed as described in chapter 2. Plasmids for the generation of the

mutant virus strains were also generated as described in chapter 2, using the primers

described in Table 2.

Generation of viral strains.

126

 

HSV-l mutants were generated by co-transfection of HSV-1 genomic DNA from
the VP16-null strain 8MA with a linear fragment of the mutant VP16 gene with flanking
5’ and 3’ UTRs. Homologous recombination occurred between the UTRs in the linear
fragment and those of the 8MA DNA. The VP16 fiagment was excised from pKOS-
SB75A, pSOB65.4, pSOB65.5 or pSOB65.6 by digestion with EcoRI and PstI. Vero cells
were transfected with 5 ug of pKOS-S375A insert and 5 pg of viral DNA by calcium
phosphate transfection (Graham et al. 1973). Afier transfection, the cells were overlaid
with DMEM supplemented with 5% F CS and 0.9% SeaPlaque agarose and incubated for
3-4 days. Plaques were picked and re-purified by two additional rounds of plaque
purification. A stock was generated from the last round of plaque purification (SJ 02).

BHK-MMTV-VP16 cells were transfected with 1 ug of 8MA genomic DNA and
1 ug of either pSO365.4, pSO365.5 or pSO365.6 insert using Lipofectamine (Stratagene)
in DMEM by standard protocols. The transfection was fed with DMEM supplemented
with 2% FCS and 0.5 ug/mL Geneticin after 3 hours and the transfection was allowed to
proceed for three days. The supernatant was harvested and used to infect Vero 16-8 cells
as described for plaque assays in chapter 2. Plaques were screened for fluorescence using
a Meridian Insight confocal laser scanning microscope, using a 20xobjective. Plaques
were screened and purified three times and stocks were generated in Vero 16-8 cells.
(801 1, 8012 and 8013).

For all four strains, a small aliquot of DNA was extracted from 500 uL of a high
titer stock. The stock was incubated for 2 hours with 1 mg of proteinase K and 1% SDS

at 37°C. Protein was removed by phenol/chloroform extraction and subsequent 2-

propanol precipitation. The DNA pellet was redissolved in 10 uL of 10 mM Tris-HCL

127

pH=8.0. A fragment, spanning the mutation, was amplified by PCR and tested for loss of
the Ach site at the codon for aa 375 or 411, respectively. The PCR fragment was then

submitted for automated sequencing to confirm the presence of the correct mutation.

Viral DNA preparation

5x106 Vero 16-8 cells were infected with 8MA at an moi of 0.01 as described in
chapter 2. After three days, when 80% CPE was observed, the infected cells were
harvested and sonicated as described in Chapter 2. The cellular debris was removed by
centrifugation and the supernatant was overlaid on a 20% glycerol in PBSa. The sample
was centrifuged for 90 minutes at 28 krpm in a sw28 swinging bucket rotor. The
supernatant was gently removed and the virion pellet was lysed in virion lysis buffer (100
mM sodium chloride, 10 mM EDTA, 50 mM Tris-HCl, pH 8.0, and 0.5% lauryl
sarcosinate.) Proteinase K was added to a final concentration of 0.2 mg/mL and the
sample was incubated at 37°C for 2 hours. After incubation, equal volumes of cesium
trifluoroacetate (CsTFA, Pharmacia) and sample were mixed and 25 uL of ethidium
bromide was added (lOmg/mL). The samples were centrifuged for 6 hours at 65 krpm in
a vti90 rotor at 15°C. The purified DNA was removed, the ethidium bromide was
removed by n-pentanol extraction and the CsTF A was removed by dialysis against TE
(10 mM Tris-HCI pH 8.0, 1 mM EDTA). DNA concentration was determined by

absorbance at 260 nm.

Chloramphenieol transferase assay

128

pAN 5, pAN6, pAN 7 or pICP27-CAT plasmids were introduced into HeLa cells
using Lipofectamine and the Plus reagent (Stratagene). 24 hours afier transfection,
transfected cells were pooled to ensure consistent transfection efficiency and seeded at
2x 105 cells/well in six well plates. 24 hours after seeding the cells were prepared for
infection. When infecting in the absence of cycloheximide, the cells were washed in
DMEM and infected at an moi of 5 or 0.1, for RPS. Infections were fed after 1 hour with
DMEM supplemented with 2% FCS. The cells were washed and overlaid with PBSa
2hpi. The cells were dislodged from the culture dish with a disposable scraper and
transferred to a microcentrifuge tube. The cells were pelleted at 200x g for 2 minutes in
an eppendorf microcentrifuge. The supernatant was removed and the cells resuspended
in 100 uL of 0.25 M TriS'HCI. Cells were disrupted by three repeated cycles of freezing
and thawing. Each cycle consisted of 5 minutes in a dry ice ethanol bath followed by five
minutes at 37°C. Extracts were tested for CAT activity by continuous organic extraction
as described (Nielsen et al. 1989). 20 uL of sample was mixed with 20 uL of reaction
buffer (5mM CAM, SOOug/mL BSA in 250 mM Tris-HCl (pH=7.8)) or 20 uL of
(SOOug/mL BSA in 250 mM T ris~HCl (pH=7.8)). Total volume was brought up to 80 uL
with ddeO. Reactions were started with 20 uL of [3H]-acetyl-COA (20 uCi/mL; 150
,uM) in 75 uM HCl. Samples were counted twice at either 30 minute or 45 minute
intervals. Activity was normalized to total protein concentration, as determined by
Bradford assay (BioRad)

To assay the activity of VP16 in the absence of protein translation, cycloheximide
was added to the culture medium 30 minutes prior to infection. Infection was maintained

in cycloheximide for another 2 hours, at which time the cycloheximide was washed out

129

and replaced with Actinomycin D. To allow translation of the CAT enzyme, the
infection continued for another 2 hours in the presence of Actinomycin D. The cells were

then harvested and assayed as above.

Growth curves

Stocks of KOS, RPS, SJ02, 801 1, $012 and 8013 were prepared and titered on
Vero or Verol6-8 cells as described in Chapter 2. HeLa cells were plated on p60 tissue
culture plates (p605) at 2x105 cells per p60. 24 hours after splitting the cells, they were
infected with KOS, SJ 02 or RPS at an MOI of 10. Samples were harvested at 4 to 6 hour
intervals for 24 hours. The samples were harvested by removing the cells with a
disposable scraper and transferring the suspended cells and growth medium to a 15 mL
conical tube. The cells were sonicated as described in chapter 2, to disrupt the cells and
release the virions from the cellular debris. The insoluble cell debris was pelleted by
centrifugation at 3000x g in a Sorvall Techospin centrifuge for 5 minutes at ambient
temperatures. The titer of each sample was determined by plaque assay as described in

chapter 2.

130

Discussion:
Function of Ser375 and Ser4ll in VP16 during infection.

In this chapter, Ser375 and Ser411 were tested for their function in infection.
Virus strains which carried mutations at one or the other of these sites were generated.
These strains were tested for growth phenotypes and for transcriptional phenotypes. A
Ser375Ala mutation was found to have no effect on the growth of the virus, yet the
transcription of the IE genes was significantly reduced when the IE genes were activated
solely by VP16. Under conditions where other factors could aid in the activation of the
IE genes, their expression was near wild-type. Mutations at Ser4ll were not found to
have any effect on the growth of the virus, nor did these strains exhibit any defect in IE
gene expression, whether examining only the VP16-derived activation or total IE gene
activation.

Changing either serine to an alanine did not lead to the loss of phosphorylation of
VP16. The loss of the serine did not affect the labeling of the carboxyterminal
lysylendopeptidase C fragment either. The conclusion was that neither of these sites are
major phosphorylation sites late in infection.

Because SJ02 grew with wild type kinetics, the hypothesis that VIC-derived
transcription was required for the efficient proliferation of HSV-1 in infection had to be
rejected. The hypothesis that Ser375 is the only in vitro CK2 phosphorylation site in the
core domain of VP16 was also rejected. The hypothesis that phosphorylation of Ser411
served as a regulatory signal for transcriptional activation by VP16 or any other function

required for viral proliferation in HeLa cells was also rejected.

131

 

Activation of the IE genes in infection by SJ02 overcomes the transcriptional
deficiencies of the Ser375Ala mutation.

The investigation of the function of Ser375 of VP16 in the virus was prompted by
the potential regulatory role that phosphorylation could have if it was a phosphorylation
site. Furthermore, the site was known to be required for VIC formation in vitro and for
IE gene activation in vivo . When testing the function of the Ser375Ala mutation under
conditions that most stringently tested VP16 activation during infection, the protein was
unable to activate the IE genes. This result was in accordance with the observation of
previous studies of this particular mutation in vivo and in vitro (Greaves et al. 1990; Lai
et al. 1997). Both Greaves et al. and Lai and Herr had discovered this mutation to disrupt
the formation of the VIC complex and also prevents activation fi'om an IE gene promoter.
Furthermore, substitution with acidic residues or proline also disrupted the interaction.
The conservative substitution with threonine did conserve the interaction and the
activation potential (Greaves et al. 1990; O'Reilly et al. 1997). Lai et al also determined
that the critical interaction affected by the mutation was that between Oct-1 and VP16
(Lai et al. 1997). None of the experiments addressed whether the function of VP16 relied
on the phosphorylation of Ser375 or whether it was the serine per se that was required for
the interaction with Oct-1.

In contrast to those experiments, when protein synthesis was allowed during
infection, IE gene activation in the SJ02 was nearly indistinguishable fi'om that of the
wild type virus. Furthermore, this mutant virus strain was completely unaffected in its
growth, replicating at the same speed as the wild type KOS. Because it was clear that the

Ser375Ala mutation would not allow the VIC-dependent activation of the IE genes, the

132

IE genes had to be expressed by another mechanism. Here, two models to achieve this

result will be presented.

The threshold model

Factors that have been shown to facilitate activation of the IE genes include the
IE-genes themselves, especially ICPO and ICP4. It is possible that the majority of IE
gene expression seen in SJ02 infection is due to activation by these factors. In this
model, it is proposed that VP16-S375A, even though it is a much weaker activator that
the wild type, it is still capable of activating the [E genes at a low level. When translated,
these IE gene transcripts then trigger the subsequent transcription cascade. This would
mean that the threshold level for triggering this cascade would lie in the marginal

difference that is observed between RPS and SJ 02 in figure 13.

The TAATGARAT is not the only significant response element through which VP16
acts

An alternative model might suggest that the S37SA mutants inability to bind the
IE promoters in vitro is not mirrored in vivo. One could imagine additional factors that
facilitate interaction between VP16 and conserved elements in the IE promoters, apart
from the TAATGARAT element. This is not an unlikely scenario, as deletion of the GA-
element in the IE promoters decreases VP16-mediated activation (Triezenberg et al.
1988b). It is also know that HCF-1 interact with both Spl and GABP, as well as with
VP16 (Gunther et al. 2000; Vogel et al. 2000). This suggests that VP16 could be

recruited to the IE promoters through either the Spl sites or GA-elements. Both of these

133

 

interactions rely on surfaces of HCF-l distinct from the B-propeller where VP16
interacts. This might suggest that HCF-1 can interact with both VP16 and either of these
factors, recruiting VP16 to the promoter. In light of the effect of cycloheximide on IE
gene expression by SJ02, the action of such a factor would be prevented by either the
lack of protein synthesis or possibly by another effect of cycloheximide treatment. This
could mean that the factor is a de novo synthesized protein whose expression is triggered
by the infection. Both GABP and Spl are widely expressed genes.

Ser411 was found in chapter 2 to be a likely phosphorylation site. While there
was no direct correlation between the site and a function of VP16, the phosphorylation
itself made it a likely candidate for regulation. There was no evidence that the mutations
at Ser4ll affected the activation of the IE genes in any way. Furthermore, there was no
evidence that these mutations affected the growth of the virus. It was concluded that
phosphorylation of Ser411 had no function in infection of HeLa cells.

The growth and transcription phenotypes are a little surprising if it is assumed that the
bulk of IE gene expression comes from the VIC. Here, healthy growth and abundant
activation of the IE genes was observed in a virus expressing a VP16 that is unable to

bind the TAATGARAT in vitro.

Caveats

The identification of phosphorylation sites within VP16 were tempered by the
following facts. The peptide mapping provided an incomplete picture of the
phosphorylation of VP16. The method could only reliably identify phosphorylation

betWeen aa30-343 and 371-490. It did not address whether phosphorylation occurred

134

between aa1-29 or 344-370. Furthermore , while it did exclude phosphorylation of the
serines between a 30-343, it could not distinguish between the serines in the 371-490
fragment. There was positive evidence for phosphorylation of the 371-412 fragment.
This site included serines at Ser375, Ser400 and Ser411. There was also positive
evidence of both the 410-454 and the 410-490 fragments. These encompass the serines at
411, 419, 452 and , for the latter fragment, 462.

The phosphorylation sites that were identified by MS may not fully represent the
sites in the protein. Six serines were not represented within the peptides identified by
MS. Phosphorylated peptides representing three of the remaining twenty serines were
found, but the lack of phosphorylated peptides representing the remaining seventeen sites
can not be taken as rigorous evidence that no phosphorylation occurred there. It is
possible that the ionization of a specific peptide in the MS can be negatively affected by
the presence of a phosphogroup and therefore is underrepresented or absent in the MS.

The peptide mapping and the MS does not conflict about the phosphorylation of
any residues. None of the sites within the 30-370 fragment that were identified by MS
were found to be phosphorylated. Similarly, the labeled fragments encompassing 371-
490, 371-454, 371-412, 410-490 and 410-454 were all found to have at least one serine
residue that was either identified as a phosphorylation site by MS or was not found by the
MS. However, both methods may fail to detect low abundance phosphorylated species.
This means that if a small sub-population of VP16 in the cell is phosphorylated as a

method of tagging it for a specific action, it is possible that neither method would be able

to detect it.

135

The immunoprecipitation of VP16 from late in infection of HeLa cells is a “snap-
shot” of the function of VP16. If the phosphorylation of VP16 is a regulatory function,
the phosphorylatable residues must exist both in a phosphorylated and an
unphosphorylated state. This means that as VP16 changes function from early to late in
infection or in the virion or in reactivation from latency, the entire population of VP16
might change its phosphorylation state. The isolation of VP16 from a specific stage of
infection will only address the phosphorylation state of VP16 when performing the
activities required by VP16 at that stage. This means that phosphorylation could be quite
different under different conditions and could include both serines that were found not to
be phosphorylated in this study as well as both threonine and tyrosine residues.

The phosphorylation sites that were detected in the MS must be understood to be
separated events. One can not draw any conclusion about whether these modifications
occur in the same VP16 molecule or not. All of these modification might be present in a
single VP16 molecule, but if the phosphorylation serves as a regulatory switch, a
particular molecule is more likely to only have a subset of these modifications.

The immunoprecipitation of VP16 from infected cells is intended to isolate a
representation of all of the VP16 populations in that particular stage of infection. It is
however possible that the immunoprecipitation unintentionally fractionates the VP16
population. For instance, if a specific phosphorylation event causes VP16 to associate
with an insoluble part of the cell lysate, the population of VP16 which carries this
modification would be lost in the generation of the infected cell extract. Another cause
might be that phosphorylation of the epitope recognized by the precipitating monoclonal

antibody could interfere with the precipitation of a population of VP16, if it was modified

136

at this specific site. The fractionation of VP16 by events such as these are unlikely to
completely remove VP16 from the immunoprecipitate, but they might cause an
underrepresentation of a particular sub-population of VP16, causing it to go undetected in
either the MS or the peptide mapping.

Specifically for SJ02, there is a concern in the selection of the mutant strain.
Homologous recombination, selection and generation of 8102 stock took place in the
absence of complementation by a wt VP16. If the mutation of SJ02 at Ser375 does in
fact generate an inviable or weakened virus, the lack of complementation introduces a
selective pressure which favors strains that have acquired second-site compensatory
mutations. Such mutations might hypothetically mask any deficiencies in the function of
SJ02. This concern is tempered to some extent by the fact that in the absence of protein
synthesis, SJ02 [B gene expression is similar to that of a Ser375 mutant of VP16
expressed from a plasmid. That means that none of the components that the virus carries
into the host cell are capable of compensating for the deficiencies of VP16 3758A. It still
leaves the possibility that some factor, expressed afier infection, is capable of
compensating for the mutant VP16. It is not possible to determine whether this
compensation is due to a random mutation even or whether the function is present in the
wild type virus. The most reliable method to address this issue is to generate additional
strains, carrying the $375A mutation in VP16, but screening for the recombinants in the
presence of a complementing wild type VP16. This would require a reliable method for
screening the mutants. One such method has been developed and used in the generation

of the SO series, but this has caveats associated with it, too.

137

In the generation of the SO strains, the major issue is the effects of adding a bulky
domain to the activation domain of VP16. While there has been no report that EGFP is
phosphorylated, it is a distinct possibility that the protein is a substrate for kinases. It
might explain why there is no discernible loss of phosphorylation in the SO series. If the
bulk of phosphorylation takes place within the EGFP, it would further mask small
changes in phospholevels. Especially if these changes are already masked by the
presence of additional phosphorylation sites within VP16 proper. It does not appear that
there are gross changes in the level of phosphorylation, when comparing KOS and DG1.
Yet, no real quantitative data is available to show this.

Furthermore, while fusing large peptides to the activation domain of VP16 does
not seem to affect its firnction (McKnight et al. 1994) it may be that the EGFP itself
complements the defects of the Ser411 mutants. At first glance, this suggestion seems
unlikely, but one could imagine a number of ways that this could be the case. First, if the
EGF P acts as a cryptic activation domain, it might complement loss of function of one or
both of the VP16 ADS. Or the added strength might complement for lower affinity for
the promoter response element. It is also possible that the EGFP packaged in the virion
affects the kinetics of virion entry, intracellular transport of VP16 or virion packaging
and egress. VP16 takes part in all of these events and changes in the activity due to the
mutations that are introduced may be ameliorated by the presence of EGFP.

The viability of all of the strains in this dissertation were tested under conditions
of single stage proliferation. Infection in this type of experiment is done at an MOI of S-
10. For RPS and to some extent for in18l4, a high MOI has been shown to reduce the

apparent viability phenotype. The difference in the proliferation at high and low MOI of

138

these strains may be due to the expression of viral genes from multiple genomes. The
mounting of a successful infectious event most likely relies on the presence of a threshold
level of certain factors, whether this level is achieved through high levels of expression
from a few templates or low levels of expression from a large number of templates. If
testing 8.102, $011, $012 and 8013 under low MOI conditions, these strains might
display a similar low MOI phenotype. The single stage growth curve also does not
address whether the mutations affect re-infection of neighboring cells. If the mutations
affect the ability of VP16 to quench the stress response in the primary host cell,
neighboring cells may be able to mount a defense, possibly rendering them less
susceptible to infection. In a single stage infection, all cells are infected simultaneously.

The growth of the mutant strains generated here were tested only in Vero cells.
These cells are used in this assay because of their inherent susceptibility to infection with
HSV-l. But the mutations at Ser3 75 or Ser411 could give rise to growth phenotypes that
are either host-specific, either due to the presence or absence of a specific factor in the
host cell or simply because a slow-growth phenotype inherent to these mutations are
masked by the overall susceptibility of the host cell.

The transcription of the IE genes was tested in infection using a set of reporter
genes. This method may not represent the effect of the mutations on the endogenous IE
genes for several reasons. First, the environment of the IE promoters of the transfected
reporter is clearly different from that of the viral genome. The reporter plasmids are
introduced into the cell in the context of transfection reagent that by themselves may
affect the expression of the genes. The plasmids may also be sequestered in a different

nuclear compartment than the viral genome. The templates could also associate with

139

different factors, such as histones or polyarnines, which would affect the expression of
the genes encoded in these templates. Secondly, by looking at the expression from a
plasmid reporter, artifactual misexpression of the viral genes could be masked. This
could be due to mutations acquired in addition to those at Ser375 or Ser411 of VP16. For
instance, a constitutive up-mutation in the ICPO promoter could render transcription from
that promoter independent of VP16 activation. The expression of the ICPO gene would in
turn cause the activation of the rest of the IE genes and subsequently trigger the viral
gene expression cascade. In the presence of cycloheximide, because the up-mutation is
not present in the reporter plasmid expression of the ICPO gene would not be detected in
the reporter assay. On the other hand, if the IE genes of the viral genome were tested
directly, this defect would be much easier to identify. The misexpression would also be a
result of cycloheximide treatment or the transfection reagent.

As for the proliferation phenotype, some transcriptional activation defects can be
masked in infections at high MOI. Because the transcriptional activation assays in this
dissertation were at high MOI, it is possible that similar defects in SJ02 and the SO
series went undiscovered. The hypothesis is that the optimal expression of the [E genes
can be achieved by high expression from a few templates or by lower expression from
more templates. Therefore, the transcriptional assays that were done at high MOI may
have missed transcription phenotypes of this particular type.

Finally, the observation that expression of the IE reporters in the transcription
assay of SJ02 might be explained by a decrease in the protein stability of VP16-
Ser375Ala, as compared to wild type VP16. To explain the wild type phenotype of SJ 02

in the absence of cycloheximide, one would have to argue that either the protein is

140

directly destabilized by the drug or that the mutant protein is susceptible to degradation
only in the absence of protein expression.

The findings of this dissertation are internally consistent. The mapping of the
phosphorylation sites of VP16 by the peptide mapping and by MS did not show any
conflict between assignment of phosphorylation sites. The exclusive identification of
serines as phosphorylation sites in the MS was supported by the phosphoamino acid
analysis. The mutagenesis of VP16 at either Ser375 or Ser4ll did not lead to loss of
phosphorylation of bulk VP16, consistent with the finding of multiple phosphorylation
sites discovered in the MS. In particular, there was no evidence that the loss of Ser375
led to any appreciable loss of phosphorylation in the peptide mapping. This finding was
in agreement with the MS, which found Ser375 to be unphosphorylated.

The loss of Ser375 did not cause an appreciable defect in proliferation with SJ 02.
Yet, a defect was discovered in IE gene activation by that strain, when translation was
blocked during infection. The apparent disagreement between those two observations
was resolved by the observation that when translation was not blocked, the expression of
the IE genes were only modestly affected in SJ02 infection. These three observations
were consistent with a model where IE gene expression in HSV-l can be activated by
VP16 alone, if VP16 is capable of interacting with Oct-1. If VP16 can not interact with
Oct-1, the IE genes can be activated through a mechanism that requires de novo synthesis
of protein upon infection. If protein expression is allowed, the expression of the IE genes
may be required for the proliferation of the virus, but the expression of the [E genes are
not dependent on the binding of VP16 to Oct-l. That means that the expression of the IE

genes, when protein translation is allowed, is consistent with the grth phenotype. The

141

absence of expression of the IE genes in the absence of protein synthesis is also
consistent with the model, if the translation-independent mechanism of IE gene activation

relies on the binding of VP] 6 to Oct-1

 

142

Chapter Four

Discussion and Future Directions

Hypotheses

Hypothesis: The Ser375 dependent formation of the VIC does not require
phosphorylation of VP16.

In retrospect, there is no good evidence to suggest that the interaction between
VP16 and Oct-1 relies on phosphorylation. In my research, I have found no evidence to
suggest that VP16 is phosphorylated at Ser375 late in infection and the evidence
presented by O’Reilly et al is not convincing (O'Reilly et al. 1997). Furthermore, Babb
et a1 demonstrate convincingly that the formation of the VIC occurs in the absence of
phosphorylation of VP16 (Babb et al. 2001). While it can not be ruled out that
phosphorylation of VP16 would further enhance this interaction, it is interesting that the
interaction between Oct-1 and VP16 is disrupted by replacing Ser375 with acidic residues
(O'Reilly et al. 1997). Although it is not universally true that substitution of a
phosphoresidue can be functionally replaced by an acidic residue, it does add support to
the hypothesis that the mechanism of interaction does not simply rely on the presence of
a negative charge.

Therefore, I hypothesize that if a) there is no evidence of phosphorylation of
VP16 at Ser375 in infection or by the kinases in a nuclear extract, b) if the formation of
the VIC does not require phosphorylation of any kind and c) if replacement of the serine
with neither aspartate nor glutamate conserves the transcriptional activity of VP16, then

phosphorylation of VP16 at Ser375 is not required for activation. because VP16 isolated

143

late in infection still is not phosphorylated at Ser375, it is unlikely that phosphorylation

of Ser375 serves to prevent the interaction between VP16 and Oct-l.

Hypothesis: Ser400 is a major site of phosphorylation.

In this dissertation, I have demonstrated that Ser375 is very unlikely to be a
phosphorylation site. I have also demonstrated that while mapping phosphorylation sites
of VP16 can not rule out Ser411, there is no functional evidence to show that Ser4ll
need be a phosphorylation site. Furthermore, the site is not conserved between the
homologs of VP16 (Figure 9). However, it is clear that the 371-413 fragment of RPS is
phosphorylated at a serine. The only remaining serine in the fragment is Ser400. While
the MS data did not find evidence that this site is phosphorylated, it is only represented
by a few spectra and may have been missed by the MS. And in contrast to Ser4l l, the
putative phosphorylation at Ser4OO is conserved between HSV—1, HSV-1, VZV, EHV -1
and BHV—l. Based on these facts and observations, I hypothesize that the primary

phosphorylation site of the 371-413 fragment is Ser400.

Hypothesis: VP16 requires HCF-1 to activate IE genes during infection by a
mechanism other than the VIC complex.

In my research, I discovered that mutating VP16 at Ser375 severely diminishes
the activation of the IE genes by VP16. However, the same mutation does not affect
viability of 8102 in infection, nor does it abolish the activation of the IE genes through
another mechanism. It is clear that the loss of the activation domain of RPS and V422, a

similar truncation mutant, has a significant effect on the activation of the IE genes

144

(Smiley et al. 1997; Tal-Singer et al. 1999) as well as proliferation. It is also clear that
the in1814 insertion, which most strongly affects the interaction with HCF-1, abolishes
activation of the IE genes (Ace et al. 1989).

I hypothesize that the interaction with HCF-1 is required for the activation of the
IE genes through VP16. This requirement can be divided into two aspects: Nuclear
localization and direct activation. The formation of the VIC relies on the presence of all
three members of the complex in the nucleus. It has been suggested that the interaction
of VP16 and HCF-l is required for the nuclear localization of VP16. (Mahajan et al.
2002). Therefore, any disruption of this interaction will initially prevent activation of the
IE genes due to the retention of VP16 in the cytoplasm. However, if the interaction is
conserved, and facilitated the nuclear localization of VP16, the disruption of the VIC
through abolishing the interaction with Oct-1 by the Ser375Ala mutation is not sufficient
to prevent the expression of the IE genes. RPS and V422 can not activate the IE genes,
even though they presumably make it to the nucleus, therefore the VP16 AD must be
required for activation by both the VIC dependent and the VIC independent pathway.

In this hypothesis, I do not distinguish between a model in which the interaction
with HCF-1 is only required for VIC-independent activation due to its role in nuclear

localization of VP16 or whether it has a direct role in activation of the IE genes.

Hypothesis: The luman family of proteins may be phosphorylated at a position
homologous to Ser353 of VP16.
VP16 uses host factors for DNA binding. The only cellular proteins with a

similar fimction to that of the core domain of VP16 are those of the Luman family of

14S

transcription factors. These factors has a very limited sequence similarity, except for the
4 aa HBM. These factors interact with HCF—l in a similar manner to VP16 through
HBM (E/D-H-x-Y) . The HBM is also conserved in the (Jr-herpesviruses. However, the
phosphorylation site proximal to this site, Ser353 does not have a good homology in any
of the herpes viruses. Even HSV-2, which is otherwise very conserved and does have a
conserved serine at this site, is quite divergent in this motif.

Curiously, if the HBM motif of the cellular HBM-proteins is used as an anchor,
Ser353, which is 7 amino acids aminoterminal to the HBM motif, aligns with an invariant
serine in Luman and Zf, as well as their homologues in B. taurus, M. musculus and R.
norwegicus. Furthermore, the three preceding amino acids (FTT) are either identical or
conservative substitutions. I propose that the interaction surface that VP16 mimics to
interact with HCF-1 includes this motif. Because this site was found to be
phosphorylated in VP16, it is also possible that it is phosphorylated in the cellular
homologs. That means that it could be a regulatory site for interaction with HCF-l, not
only for VP16, but also for the cellular factors. In this model, I favor the proposal that
the interaction between VP16 and HCF-1 is disrupted, rather than facilitated by
phosphorylation. This suggestion is based on the fact that VIC can form from bacterially
expressed factors, which are likely to be unmodified. This idea is also in agreement with
the observation that the site is phosphorylated late in infection. At this time, prevention
of interaction with HCF-1 could be required to ensure the availability of VP16 for other
functions. This model is supported, in the observation that late in infection, VP16 is
separated into two populations, one that co-localizes with HCF-l in the nucleus of the

infected cell and one that does not (LaBoissiere et al. 2000).

146

The phosphorylation sites at Ser18, Ser353 and Ser452 may be involved in
regulating other VP16 functions.

In addition to the specific hypotheses that I have presented above, there are still
more possible functions for the phosphorylation of VP16. These other functions of VP16
that may change during infection with HSV-l include the putative repression of the vhs,
the interaction with VP22 and the role in DNA encapsidation and nuclear egress.
Regulation of these processes may also be achieved through phosphorylation of VP16. I
did not address the involvement of the mutant versions of VP16 in any of these activities
in my project, but it may be possible to develop specific hypotheses about each of these
functions and the putative role of phosphorylation in their regulation. These models can
be can be generalized to the role of phosphorylation in regulation of interaction between
partners. Phosphorylation of one or more of the discovered phosphorylation sites could

serve to either facilitate or prevent interaction between VP16 and these factors.

Future research: Addressing the hypotheses experimentally

The reagents that have been generated in the course of this dissertation are
obvious starting points for the continuation of this project. Also, the determination of
additional phosphorylation sites would prompt the generation of additional mutant
strains, carrying knock-outs at this site. Finally, the continuing investigation of
phosphorylation of VP16 during other times in infection would also assist in

understanding the complete role of VP16 in infection.

147

Investigate the SJ OZ phenotype

The first set of these experiments involves the continuing examination of the
phosphorylation sites at Ser375 and 411 during infection. Especially the Ser375 strain
promises to add to our knowledge of VP16-mediated activation. If one hypothesizes,
based on the GABP and Spl interactions, (Gunther et al. 2000; Vogel et al. 2000) that
VP16 gets recruited to the promoter during infection, it should be possible to detect it
there by chromatin immunoprecipitation. Depending on whether VP16 was found at the
promoter or not, it would be possible to attribute the effect to a deficiency in promoter
binding or transactivation. Secondly, if VP16 is found at the promoter, it would be
interesting to see whether it remains there in the presence of cycloheximide. If not, it
would suggest that in the absence of the VIC, VP16 relies either on a newly synthesized
protein factor, or more generally, on a factor which changes character in the presence of
cycloheximide. The more general interpretation must be kept in mind, because
cycloheximide may affect the cell in ways other than blocking protein synthesis.

Another aspect that should be addressed is the potential low multiplicity effect of
the mutation. In1814 and RPS both have demonstrated greater problems in proliferation
at lower multiplicities of infection. It is possible that SJ 02 has similar problems at lower
moi, but to a lesser extent. SJ 02 may be fully rescued at high moi, in contrast to RPS and
inl 814, where the phenotype is only partially complemented. But if SJ 02 were to exhibit
a mutant phenotype, similar to those of in1814 and RPS, even if it is less severe, it would

suggest that the mutation in all three work through the same mechanism.

Ser18, Ser343 and Ser452 phosphorylation

148

TJ

 

The obvious next step in regards to the phosphorylation of these sites is
mutagenesis. Because there is a wide varieties of effects these sites might have on the
virus, it would be prudent, to remain in the context of the virus. It would be possible to
use the EGFP-tag screen to identify these mutants, leaving the possibility of generating
them in complementing cell lines. The continued research into these sites will be driven

by the phenotypes that are observed as a result of the mutations.

Other approaches

An intriguing possibility is that VP16 might be recruited to the IE promoters
through Spl or GABP sites. This mechanism could be investigated outside of the virus.
The interactions could be observed in vitro or by transfection. This method would
facilitate a greater range of mutations that can be introduced and tested for their
relevance. The interactions could be examined in transfection, both by examining the
recruitment of VP16 to the promoter, for instance by chromatin immunoprecipitation but

also by standard promoter expression assays.

Luman homology

Finally, the homology with the Luman family could be tested. If the hypothesis
holds true, I would expect both Luman and Zf to be phosphorylated. I would also expect
alanine substitutions at that site to affect the function of Luman and Zf, as well as of
VP16. This line of investigation would be started in the same manner as this project was,
simply by verifying the phosphorylation of the protein and then proceeding to map it by

MS.

149

LIST OF REFERENCES

Ace, C. I., T. A. McKee, J. M. Ryan, J. M. Cameron and C. M. Preston (1989).
“Construction and characterization of a herpes simplex virus type 1 mutant unable
to transinduce immediate-early gene expression.” J Virol 63: 2260-9.

Adams, J ., R. Kelso and L. Cooley (2000). “The kelch repeat superfamily of proteins:
propellers of cell function.” Trends Cell Biol 10: 17-24.

Andrew, C. D., J. Warwicker, G. R. Jones and A. J. Doig (2002). “Effect of
phosphorylation on alpha-helix stability as a function of position.” Biochemistry
41: 1897-905.

Amosti, D. N., C. M. Preston, M. Hagrnann, W. Schaffner, R. G. Hope, G. Laughlan
and B. F. Luisi (1993). “Specific transcriptional activation in vitro by the herpes
simplex virus protein VP16.” Nucleic Acids Res 21: 5570-6.

Ashcroft, M., M. H. Kubbutat and K. H. Vousden (1999). “Regulation of p53 function
and stability by phosphorylation.” Mol Cell Biol 19: 1751-8.

Babb, R., C. C. Huang, D. J. Aufiero and W. Herr (2001). “DNA recognition by the
herpes simplex virus transactivator VP16: a novel DNA-binding structure.” _M_o_l
Cell Biol 21: 4700-12.

Bagby, S., T. K. Mal, D. Liu, E. Raddatz, Y. Nakatani and M. lkura (2000). “TFIIA-
TAF regulatory interplay: NMR evidence for overlapping binding sites on TBP.”
FEBS Lett 468: 149-54.

Barlev, N. A., R. Candau, L. Wang, P. Darpino, N. Silverrnan and S. L. Berger
(1995). “Characterization of physical interactions of the putative transcriptional
adaptor, ADA2, with acidic activation domains and TATA-binding protein.” J
Biol Chem 270: 19337-44.

Baru, M., M. Shlissel and H. Manor (1991). “The yeast GAL4 protein transactivates the
polyomavirus origin of DNA replication in mouse cells.” J Virol 65: 3496-503.

Bauer, D. and R. Tarnpe (2002). “Herpes viral proteins blocking the transporter
associated with antigen processing TAP--from genes to function and structure.”
Curr Top Microbiol hnmunol 269: 87-99.

Bennett-Cook, E. R. and J. A. Hassell (1991). “Activation of polyomavirus DNA

replication by yeast GAL4 is dependent on its transcriptional activation domains.”
Embo J 10: 959-69.

150

Berger, S. L. (2002). “Histone modifications in transcriptional regulation.” Current
Opinion in Genetics & Development 12: 142—148.

Berger, S. L., B. Pina, N. Silverman, G. A. Marcus, J. Agapite, J. L. Regier, S. J.
Triezenberg and L. Guarente (1992). “Genetic isolation of ADA2: a potential
transcriptional adaptor required for function of certain acidic activation domains.”
_erl 70: 251-65.

Bhella, D., F. J. Rixon and D. J. Dargan (2000). “Cryomicroscopy of human
cytomegalovirus virions reveals more densely packed genomic DNA than in
herpes simplex virus type 1.” J Mol Biol 295: 155-61.

Blom, N., S. Gammeltofi and S. Brunak (1999). “Sequence and structure-based
prediction of eukaryotic protein phosphorylation sites.” J Mol Biol 294: 1351-62.

Boutell, C. and R. Everett (2003). “The HSV-l regulatory protein ICPO interacts with
and ubiquitinates p53.” J Biol Chem Epub ahead of print.

Bowman, B. R., M. L. Baker, F. J. Rixon, W. Chiu and F. A. Quiocho (2003).
“Structure of the herpesvirus major capsid protein.” Embo J 22: 757-65.

Bresnahan, W. A. and T. Shenk (2000). “A subset of viral transcripts packaged within
human cytomegalovirus particles.” Science 288: 2373-6.

Brownell, J. E., J. Zhou, T. Ranalli, R. Kobayashi, D. G. Edmondson, S. Y. Roth
and C. D. Allis (1996). “Tetrahymena histone acetyltransferase A: a homolog to
yeast GcnSp linking histone acetylation to gene activation.” Cill 84: 843-51.

Burton, E. A., C. S. Hong and J. C. Glorioso (2003). “The Stable 2.0-Kilobase Intron
of the Herpes Simplex Virus Type 1 Latency-Associated Transcript Does Not

Function as an Antisense Repressor of ICPO in Nonneuronal Cells.” J Virol 77:
3516-30.

Campbell, M. E., J. W. Palfreyman and C. M. Preston (1984). “Identification of
herpes simplex virus DNA sequences which encode a trans-acting polypeptide
responsible for stimulation of immediate early transcription.” J Mol Biol 180: 1-
19.

Candau, R., J. X. Zhou, C. D. Allis and S. L. Berger (1997). “Histone
acetyltransferase activity and interaction with ADA2 are critical for GCNS
function in vivo.” Embo J 16: 555-65.

Chen, 1. H., K. S. Sciabica and R. M. Sandri-Goldin (2002). “ICP27 interacts with the

RNA export factor Aly/REF to direct herpes simplex virus type 1 intronless
mRNAs to the TAP export pathway.” J Virol 76: 12877-89.

151

Chi, J. H. and D. W. Wilson (2000). “ATP-Dependent localization of the herpes
simplex virus capsid protein VP26 to sites of procapsid maturation.” J Virol 74:
1468-76.

 

Chiu, W. and F. J. Rixon (2002). “High resolution structural studies of complex
icosahedral viruses: a brief overview.” Virus Res 82: 9-17.

Chrivia, J. C., R. P. Kwok, N. Lamb, M. Hagiwara, M. R. Montminy and R. H.
Goodman (1993). “Phosphorylated CREB binds specifically to the nuclear
protein CBP.” Nature 365: 855-9.

Cress, W. D. and S. J. Triezenberg (1991). “Critical structural elements of the VP16
transcriptional activation domain.” Science 251: 87-90.

Desai, P., J. C. Akpa and S. Person (2003). “Residues of VP26 of herpes simplex virus
type 1 that are required for its interaction with capsids.” J Virol 77 : 391-404.

Desai, P., N. A. DeLuca and S. Person (1998). “Herpes simplex virus type 1 VP26 is
not essential for replication in cell culture but influences production of infectious
virus in the nervous system of infected mice.” Virology 247: 115-24.

Desai, P. J. (2000). “A null mutation in the UL36 gene of herpes simplex virus type 1
results in accumulation of unenveloped DNA-filled capsids in the cytoplasm of
infected cells.” J Virol 74: 11608-18.

Deshmane, S. L. and N. W. Fraser (1989). “During latency, herpes simplex virus type
1 DNA is associated with nucleosomes in a chromatin structure.” J Virol 63: 943-
7.

Doig, A. J. (2002). “Recent advances in helix-coil theory.” Biophys Chem 101-102:
281-93.

Doig, A. J. and R. L. Baldwin (1995). “N- and C-capping preferences for all 20 amino
acids in alpha-helical peptides.” Protein Sci 4: 1325-36.

Dolter, K. E., R. Ramaswamy and T. C. Holland (1994). “Syncytial mutations in the

herpes simplex virus type 1 gK (UL53) gene occur in two distinct domains.” J
Virol 68: 8277-81.

Elgadi, M. M., C. B. Hayes and J. R. Smiley (1999). “The herpes simplex virus vhs
protein induces endoribonucleolytic cleavage of target RNAs in cell extracts.” J
Virol 73: 7153-64.

Elliott, G., G. Mouzakitis and P. O'Hare (1995). “VP16 interacts via its activation

domain with VP22, a tegument protein of herpes simplex virus, and is relocated to
a novel macromolecular assembly in coexpressing cells.” J Virol 69: 7932-41.

152

Elliott, G. and P. O'Hare (1997). “Intercellular trafficking and protein delivery by a
herpesvirus structural protein.” §e_ll 88: 223-33.

Everett, R. D. (2000). “ICPO, a regulator of herpes simplex virus during lytic and latent
infection.” Bioessays 22: 761-70.

Everett, R. D. and G. G. Maul (1994). “HSV-l IE protein VmwllO causes
redistribution of PML.” Embo J 13: 5062-9.

Everett, R. D., G. Sourvinos and A. Orr (2003). “Recruitment of Herpes Simplex Virus
Type 1 Transcriptional Regulatory Protein ICP4 into Foci Juxtaposed to ND10 in
Live, Infected Cells.” J Virol 77: 3680-9.

Everly, D. N., Jr., P. Feng, I. S. Mian and G. S. Read (2002). “mRNA degradation by
the virion host shutoff (Vhs) protein of herpes simplex virus: genetic and
biochemical evidence that Vhs is a nuclease.” J Virol 76: 8560-71.

Fenwick, M. L. and R. D. Everett (1990). “Inactivation of the shutoff gene (UL41) of
herpes simplex virus types 1 and 2.” J Gen Virol 71: 2961-7.

Ferdous, A., F. Gonzalez, L. Sun, T. Kodadek and S. A. Johnston (2001). “The 19S
regulatory particle of the proteasome is required for efficient transcription
elongation by RNA polymerase 11.” Mol Cell 7: 981-91.

Gaffney, D. F., J. McLauchlan, J. L. Whitton and J. B. Clements (1985). “A modular
system for the assay of transcription regulatory signals: the sequence
TAATGARAT is required for herpes simplex virus immediate early gene
activation.” Nucleic Acids Res 13: 7847-63.

Gallagher, S. R. (1999). One-dimensional SDS gel electrophoresis of proteins. Current
Protocols in Current Biology. F. M. Ausubel. 2.

Gandhi, M. and R. M. Greenblatt (2002). “Human herpesvirus 8, Kaposi's sarcoma,
and associated conditions.” Clin Lab Med 22: 883-910.

Gibson, W. and B. Roizman (1974). “Proteins specified by herpes simplex virus.
Staining and radiolabeling properties of B capsid and virion proteins in
polyacrylamide gels.” J Virol 13: 155-65.

Gonzalez, F ., A. Delahodde, T. Kodadek and S. A. Johnston (2002). “Recruitment of a
198 proteasome subcomplex to an activated promoter.” Science 296: 548-50.

Goto, H., S. Motomura, A. C. Wilson, R. N. Freiman, Y. Nakabeppu, K. Fukushima,
M. Fujishima, W. Herr and T. Nishimoto (1997). “A single-point mutation in

153

HCF causes temperature-sensitive cell-cycle arrest and disrupts VP16 function.”
Genes Dev 11: 726-37.

Graham, F. L. and A. J. van der Eb (1973). “A new technique for the assay of
infectivity of human adenovirus 5 DNA.” Virology 52: 456-67.

Grant, P. A., D. Schieltz, M. G. Pray-Grant, D. J. Steger, J. C. Reese, J. R. Yates,
3rd and J. L. Workman (1998). “A subset of TAF(II)s are integral components
of the SAGA complex required for nucleosome acetylation and transcriptional
stimulation.” _C__e_ll 94: 45-53.

Greaves, R. F. and P. O'Hare (1990). “Structural requirements in the herpes simplex
virus type 1 transactivator Vmw65 for interaction with the cellular octarner-
binding protein and target TAATGARAT sequences.” J Virol 64: 2716-24.

Greensides, D. R. (2002). VP16 localization during herpes simplex-1 infection. Cell
and Molecular Biology. East Lansing, Michigan State University: 209.

Greijer, A. E., C. A. Dekkers and J. M. Middeldorp (2000). “Human cytomegalovirus
virions differentially incorporate viral and host cell RNA during the assembly
process.” J Virol 74: 9078-82.

Grossmann, J. G., A. J. Sharff, P. O'Hare and B. Luisi (2001). “Molecular shapes of
transcription factors TFIIB and VP16 in solution: implications for recognition.”
Biochemistry 40: 6267-74.

Gu, B., R. Rivera-Gonzalez, C. A. Smith and N. A. DeLuca (1993). “Herpes simplex
virus infected cell polypeptide 4 preferentially represses Spl-activated over basal
transcription from its own promoter.” Proc Natl Afld Sci U S A 90: 9528-32.

Gunther, M., M. Laithier and O. Brison (2000). “A set of proteins interacting with

transcription factor Spl identified in a two-hybrid screening.” Mol Cell Biochem
210: 131-42.

Halford, W. P., C. D. Kemp, J. A. Isler, D. J. Davido and P. A. Schaffer (2001).
“ICPO, ICP4, or VP16 expressed from adenovirus vectors induces reactivation of
latent herpes simplex virus type 1 in primary cultures of latently infected
trigeminal ganglion cells.” J Virol 75: 6143-53.

Harper, E. T. and G. D. Rose (1993). “Helix stop signals in proteins and peptides: the
capping box.” Biochemistry 32: 7605-9.

Heine, J. W., R. W. Honess, E. Cassai and B. Roizman (1974). “Proteins specified by

herpes simplex virus. XII. The virion polypeptides of type 1 strains.” J Virol 14:
640-51.

154

Herr, W. (1998). “The herpes simplex virus VP16-induced complex: mechanisms of
combinatorial transcriptional regulation.” Cold Spring Harb Symp Quant Biol 63:
599-607.

Hippenmeyer, P. and M. Highkin (1993). “High level, stable production of recombinant
proteins in mammalian cell culture using the herpesvirus VP16 transactivator.”
Biotechnology (N E 11: 1037-41.

Homa, F. L. and J. C. Brown (1997). “Capsid assembly and DNA packaging in herpes
simplex virus.” Rev Med Virol 7: 107-122.

Hope, I. A., S. Mahadevan and K. Struhl (1988). “Structural and fiinctional
characterization of the short acidic transcriptional activation region of yeast
GCN4 protein.” Nature 333: 635-40.

Hukkanen, V., E. Broberg, A. Salmi and J. P. Eralinna (2002). “Cytokines in
experimental herpes simplex virus infection.” Int Rev Irnmunol 21: 355-71.

Hutchinson, L., F. L. Graham, W. Cai, C. Debroy, S. Person and D. C. Johnson
(1993). “Herpes simplex virus (HSV) glycoproteins B and K inhibit cell fusion
induced by HSV syncytial mutants.” Virology 196: 514-31.

Jamieson, D. R., L. H. Robinson, J. I. Daksis, M. J. Nicholl and C. M. Preston
(1995). “Quiescent viral genomes in human fibroblasts after infection with herpes
simplex virus type 1 Vmw65 mutants.” J Gen Virol 76: 1417-31.

Jans, D. A. and S. Hubner (1996). “Regulation of protein transport to the nucleus:
central role of phosphorylation.” thsiol Rev 76: 651-85.

Jones, K. A. and R. Tjian (1985). “Spl binds to promoter sequences and activates
herpes simplex virus 'immediate-early’ gene transcription in vitro.” Nature 317:
179-82.

Katan, M., A. Haigh, C. P. Verrijzer, P. C. van der Vliet and P. O'Hare (1990).
“Characterization of a cellular factor which interacts functionally with Oct-l in

the assembly of a multicomponent transcription complex.” Nucleic Acids Res 18:
6871-80.

Kennedy, P. G. and A. Chaudhuri (2002). “Herpes simplex encephalitis.” J Neurol
Neurosurg Psychiafl 73: 237-8.

Khodarev, N. N., S. J. Advani, N. Gupta, B. Roizman and R. R. Weichselbaum
(1999). “Accumulation of specific RNAs encoding transcriptional factors and
stress response proteins against a background of severe depletion of cellular
RNAs in cells infected with herpes simplex virus 1.” Proc Natl gag Sci U S A
96: 12062-7.

155

 

Knez, J ., P. T. Bilan and J. P. Capone (2003). “A Single Amino Acid Substitution in
Herpes Simplex Virus Type 1 VP16 Inhibits Binding to the Virion Host Shutoff
Protein and Is Incompatible with Virus Growth.” J Virol 77 : 2892-902.

Kobayashi, A., T. Miyake, M. Kawaichi and T. Kokubo (2003). “Mutations in the
histone fold domain of the TAP 12 gene show synthetic lethality with the TAFl
gene lacking the TAP N-terrninal domain (TAND) by different mechanisms from
those in the SPT15 gene encoding the TATA box-binding protein (TBP).”
Nucleic Acidis Res 31: 1261-74.

Kristie, T. M., J. H. LeBowitz and P. A. Sharp (1989). “The octamer-binding proteins
form multi-protein—-DNA complexes with the HSV alpha TIF regulatory protein.”
Embo J 8: 4229-38.

Kristie, T. M. and B. Roizman (1984). “Separation of sequences defining basal
expression from those conferring alpha gene recognition within the regulatory
domains of herpes simplex virus 1 alpha genes.” Proc Natl Acad Sci U S A 81:
4065-9.

Kristie, T. M. and B. Roizman (1986). “DNA-binding site of major regulatory protein
alpha 4 specifically associated with promoter-regulatory domains of alpha genes
of herpes simplex virus type 1.” Proc Natl Acad Sci U S A 83: 4700-4.

Kristie, T. M. and P. A. Sharp (1990). “Interactions of the Oct-1 POU subdomains
with specific DNA sequences and with the HSV alpha-trans-activator protein.”
Genes Dev 4: 2383-96.

Kristie, T. M., J. L. Vogel and A. E. Sears (1999). “Nuclear localization of the C1
factor (host cell factor) in sensory neurons correlates with reactivation of herpes
simplex virus from latency.” Proc Natl Acad Sci U S A 96: 1229-33.

Kuo, M. H., J. E. Brownell, R. E. Sobel, T. A. Ranalli, R. G. Cook, D. G.
Edmondson, S. Y. Roth and C. D. Allis (1996). “Transcription-linked
acetylation by GcnSp of histones H3 and H4 at specific lysines.” Nature 383: 269-
72.

Kuo, M. H., J. Zhou, P. Jambeck, M. E. Churchill and C. D. Allis (1998). “Histone
acetyltransferase activity of yeast GcnSp is required for the activation of target
genes in vivo.” Genes Dev 12: 627-39.

La Boissiere, S., T. Hughes and P. O'Hare (1999). “HCF-dependent nuclear import of
VP16.” Embo J 18: 480-9.

LaBoissiere, S. and P. O'Hare (2000). “Analysis of HCF, the cellular cofactor of VP16,
in herpes simplex virus- infected cells.” J Virol 74: 99-109.

156

 

Lai, J. S., M. A. Cleary and W. Herr (1992). “A single amino acid exchange transfers
VP16-induced positive control from the Oct-1 to the Oct-2 homeo domain.”
Genes Dev 6: 2058-65.

Lai, J. S. and W. Herr (1997). “Interdigitated residues within a small region of VP16
interact with Oct- 1, HCF, and DNA.” Mol Cell Biol 17: 3937-46.

Lam, Q., C. A. Smibert, K. E. Koop, C. Lavery, J. P. Capone, S. P. Weinheimer and
J. R. Smiley (1996). “Herpes simplex virus VP16 rescues viral mRNA from
destruction by the virion host shutoff function.” Embo J 15: 2575-81.

Leib, D. A. (2002). “Counteraction of interferon-induced antiviral responses by herpes
simplex viruses.” Curr Top Microbiol Irnmun01269: 171-85.

Leinbach, S. S. and W. C. Summers (1980). “The structure of herpes simplex virus
type 1 DNA as probed by micrococcal nuclease digestion.” J Gen Virol 51: 45-59.

Lemaster, S. and B. Roizman ( 1980). “Herpes simplex virus phosphoproteins. II.
Characterization of the virion protein kinase and of the polypeptides
phosphorylated in the virion.” J Virol 35: 798-811.

Lentine, A. F. and S. L. Bachenheimer (1990). “Intracellular organization of herpes
simplex virus type 1 DNA assayed by staphylococcal nuclease sensitivity.” Virus
M 16: 275-92.

Li, R. and M. R. Botchan (1993). “The acidic transcriptional activation domains of
VP16 and p53 bind the cellular replication protein A and stimulate in vitro BPV-l
DNA replication.” @ 73: 1207-21.

Li, T., Y. Jin, A. K. Vershon and C. Wolberger (1998). “Crystal structure of the
MATaI/MATalphaZ homeodomain heterodimer in complex with DNA
containing an A-tract.” Nucleic Acids Res 26: S707-18.

Li, T., M. R. Stark, A. D. Johnson and C. Wolberger (1995). “Crystal structure of the
MATal/MAT alpha 2 homeodomain heterodimer bound to DNA.” Science 270:
262-9.

Liesegang, T. J. (2001). “Herpes simplex virus epidemiology and ocular importance.”
Cornea 20: l-13.

Liu, D., R. Ishima, K. I. Tong, S. Bagby, T. Kokubo, D. R. Muhandiram, L. E. Kay,
Y. Nakatani and M. lkura (1998). “Solution structure of a TBP-TAF(II)230

complex: protein mimicry of the minor groove surface of the T ATA box unwound
by TBP.” §_e_l_l 94: 573-83.

157

Liu, Y., W. Gong, C. C. Huang, W. Herr and X. Cheng (1999). “Crystal structure of
the conserved core of the herpes simplex virus transcriptional regulatory protein
VP16.” Genes Dev 13: 1692-703.

Lium, E. K., C. A. Panagiotidis, X. Wen and S. Silverstein (1996). “Repression of the
alpha0 gene by ICP4 during a productive herpes simplex virus infection.” J Virol
70: 3488-96.

Long, M. C., V. Leong, P. A. Schaffer, C. A. Spencer and S. A. Rice (1999).
“ICP22 and the UL13 protein kinase are both required for herpes simplex virus-
induced modification of the large subunit of RNA polymerase II.” J Virol 73:
5593-604.

Lu, R. and V. Misra (2000a). “Potential role for luman, the cellular homologue of
herpes simplex virus VP16 (alpha gene trans-inducing factor), in herpesvirus
latency.” J Virol 74: 934-43.

Lu, R. and V. Misra (2000b). “Zhangfei: a second cellular protein interacts with herpes
simplex virus accessory factor HCF in a manner similar to Luman and VP16.”
Nucleic Acids Res 28: 2446-54.

Lu, R., P. Yang, P. O'Hare and V. Misra (1997). “Luman, a new member of the
CREB/ATP family, binds to herpes simplex virus VP16-associated host cellular
factor.” Mol Cell Biol 17: 51 17-26.

Lu, R., P. Yang, S. Padmakumar and V. Misra (1998). “The herpesvirus transactivator
VP16 mimics a human basic domain leucine zipper protein, luman, in its
interaction with HCF.” J Virol 72: 6291-7.

 

Mahajan, S. S., M. M. Little, R. Vazquez and A. C. Wilson (2002). “Interaction of
HCF-1 with a cellular nuclear export factor.” J Biol Chem 277: 44292-9.

Mahajan, S. S. and A. C. Wilson (2000). “Mutations in host cell factor 1 separate its
role in cell proliferation from recruitment of VP16 and LZIP.” Mol Cell Biol 20:
919-28.

Marcus, G. A., N. Silverman, S. L. Berger, J. Horiuchi and L. Guarente (1994).
“Functional similarity and physical association between GCNS and ADA2:
putative transcriptional adaptors.” Embo J 13: 4807-15.

Matis, J. and M. Kudelova (2001). “Early shutoff of host protein synthesis in cells
infected with herpes simplex viruses.” Acta Virol 45: 269-77.

Maul, G. G., H. H. Guldner and J. G. Spivack (1993). “Modification of discrete

nuclear domains induced by herpes simplex virus type 1 immediate early gene 1
product (ICPO).” J Gen Virol 74: 2679-90.

158

 

McFarlane, M., J. I. Daksis and C. M. Preston (1992). “Hexamethylene bisacetamide
stimulates herpes simplex virus immediate early gene expression in the absence of
trans-induction by Vmw65.” J Gen Virol 73: 285-92.

McKnight, J. L., M. Doerr and Y. Zhang (1994). “An 85-kilodalton herpes simplex
virus type 1 alpha trans-induction factor (VP16)-VP13/ 14 firsion protein retains
the transactivation and structural properties of the wild-type molecule during virus
infection.” J Virol 68: 1750-7.

 

McKnight, J. L., T. M. Kristie and B. Roizman (1987a). “Binding of the virion protein
mediating alpha gene induction in herpes simplex virus l-infected cells to its cis
site requires cellular proteins.” Proc Natl Acad Sci U S A 84: 7061-5.

McKnight, J. L., P. E. Pellett, F. J. Jenkins and B. Roizman (1987b).
“Characterization and nucleotide sequence of two herpes simplex virus 1 genes
whose products modulate alpha-trans-inducing factor-dependent activation of
alpha genes.” J Virol 61: 992-1001.

 

McLean, C., A. Buckmaster, D. Hancock, A. Buchan, A. Fuller and A. Minson
(1982). “Monoclonal antibodies to three non-glycosylated antigens of herpes
simplex virus type 2.” J Gen Virol 63: 297-305.

McMahan, L. and P. A. Schaffer (1990). “The repressing and enhancing firnctions of
the herpes simplex virus regulatory protein ICP27 map to C-terrninal regions and

are required to modulate viral gene expression very early in infection.” J Virol 64:
3471-85.

McNabb, D. S. and R. J. Courtney (1992). “Characterization of the large tegument
protein (ICP1/2) of herpes simplex virus type 1.” Virology 190: 221-32.

Meredith, D. M., J. A. Lindsay, I. W. Halliburton and G. R. Whittaker(l991). “Post-
translational modification of the tegument proteins (VP13 and VP14) of herpes

simplex virus type 1 by glycosylation and phosphorylation.” J Gen Virol 72:
2771-5.

Mettenleiter, T. C. (2000). “Aujeszky's disease (pseudorabies) virus: the virus and
molecular pathogenesis--state of the art, June 1999.” Vet Res 31: 99-115.

Miranda-Saksena, M., R. A. Boadle, P. Armati and A. L. Cunningham (2002). “In rat
dorsal root ganglion neurons, herpes simplex virus type 1 tegument forms in the
cytoplasm of the cell body.” J Virol 76: 9934-51.

Moriuchi, H., M. Moriuchi, R. Pichyangkura, S. J. Triezenberg, S. E. Straus and J. 1.
Cohen (1995). “Hydrophobic cluster analysis predicts an amino-terminal domain

159

 

of varicella-zoster virus open reading frame 10 required for transcriptional
activation.” Proc Natl/Wad Sci U S A 92: 9333-7.

Morrison, E. E., Y. F. Wang and D. M. Meredith (1998). “Phosphorylation of
structural components promotes dissociation of the herpes simplex virus type 1
tegument.” J Virol 72: 7108-14.

Mossman, K. L., P. F. Macgregor, J. J. Rozmus, A. B. Goryachev, A. M. Edwards
and J. R. Smiley (2001). “Herpes simplex virus triggers and then disarms a host
antiviral response.” J Virol 75: 750-8.

Mossman, K. L., R. Sherbume, C. Lavery, J. Duncan and J. R. Smiley (2000).
“Evidence that herpes simplex virus VP16 is required for viral egress downstream
of the initial envelopment event.” J Virol 74: 6287-99.

Mossman, K. L. and J. R. Smiley (1999). “Truncation of the C-terrninal acidic
transcriptional activation domain of herpes simplex virus VP16 renders
expression of the immediate-early genes almost entirely dependent on ICPO.” J
lir_ol 73: 9726-33.

Narlikar, G. J ., H. Y. Fan and R. E. Kingston (2002). “Cooperation between
complexes that regulate chromatin structure and transcription.” Gill 108: 475-87.

Nedyalkov, Y. A. and S. J. Triezenberg (2003). in press.

Nguyen-Huynh, A. T. and P. A. Schaffer (1998). “Cellular transcription factors
enhance herpes simplex virus type 1 oriS- dependent DNA replication.” J Virol
72: 3635-45.

Nielsen, D. A., T. C. Chang and D. J. Shapiro (1989). “A highly sensitive, mixed-
phase assay for chloramphenicol acetyltransferase activity in transfected cells.”
Anal Biochem 179: 19-23.

Nishikawa, J., T. Kokubo, M. Horikoshi, R. G. Roeder and Y. Nakatani (1997).
“Drosophila TAF(II)230 and the transcriptional activator VP16 bind
competitively to the TATA box-binding domain of the TATA box-binding
protein.” Proc Natl AcaflSci U S A 94: 85-90.

Ogle, W. O. and B. Roizman (1999). “Functional anatomy of herpes simplex virus 1
overlapping genes encoding infected-cell protein 22 and US1.5 protein.” J Virol
73: 4305-15.

O'Hare, P. and G. Williams (1992). “Structural studies of the acidic transactivation

domain of the Vmw65 protein of herpes simplex virus using 1H NMR.”
Biochemistry 31: 4150-6.

160

 

 

Ojala, P. M., B. Sodeik, M. W. Ebersold, U. Kutay and A. Helenius (2000). “Herpes
simplex virus type 1 entry into host cells: reconstitution of capsid binding and
uncoating at the nuclear pore complex in vitro.” Mol Cell Biol 20: 4922-31.

O'Reilly, D., O. Hanscombe and P. O'Hare (1997). “A single serine residue at position
375 of VP16 is critical for complex assembly with Oct-1 and HCF and is a target
of phosphorylation by casein kinase II.” Embo J 16: 2420-30.

Orphanides, G., T. Lagrange and D. Reinberg (1996). “The general transcription factors
of RNA polymerase II.” Genes Dev 10: 2657-83.

Ottosen, S. Personal communication.

Parker, D., K. Ferreri, T. Nakajima, V. J. LaMorte, R. Evans, S. C. Koerber, C.
Hoeger and M. R. Montminy (1996). “Phosphorylation of CREB at Ser-133
induces complex formation with CREB- binding protein via a direct mechanism.”
Mol Cell Biol 16: 694-703.

Pathakamuri, J. A. and D. A. Theilmann (2002). “The acidic activation domain of the
baculovirus transactivator IE1 contains a virus-specific domain essential for DNA
replication.” J Virol 76: 5598-604.

Payne, L. N. and K. Venugopal (2000). “Neoplastic diseases: Marek's disease, avian
leukosis and reticuloendotheliosis.” Rev Sci Tech 19: 544-64.

Pereira, L. A., J. A. van der Knaap, V. van den Boom, F. A. van den Heuvel and H.
T. Timmers (2001). “TAF(II)170 interacts with the concave surface of TATA-
binding protein to inhibit its DNA binding activity.” Mol Cell Biol 21: 7523-34.

Pomerantz, J. L., T. M. Kristie and P. A. Sharp (1992). “Recognition of the surface of
a homeo domain protein.” Genes Dev 6: 2047-57.

Poppers, J ., M. Mulvey, D. Khoo and I. Mohr (2000). “Inhibition of PKR activation by
the proline-rich RNA binding domain of the herpes simplex virus type 1 Usll
protein.” J Virol 74: 11215-21.

Preston, C. M. and M. McFarlane (1998). “Cytodifferentiating agents affect the
replication of herpes simplex virus type 1 in the absence of functional VP16.”
Virology 249: 418-26.

Preston, V. G., J. W. Palfreyman and B. M. Dutia (1984). “Identification of a herpes

simplex virus type 1 polypeptide which is a component of the virus-induced
ribonucleotide reductase.” J Gen Virol 65: 1457-66.

161

 

 

 

Prod'hon, C., I. Machuca, H. Berthomme, A. Epstein and B. Jacquemont (1996).
“Characterization of regulatory functions of the HSV-1 immediate-early protein
ICP22.” Virology 226: 393-402.

Ptashne, M. and A. Gann (1997). “Transcriptional activation by recruitment.” Nature
386: 569-77.

Purves, F. C., W. O. Ogle and B. Roizman (1993). “Processing of the herpes simplex
virus regulatory protein alpha 22 mediated by the UL13 protein kinase determines
the accumulation of a subset of alpha and gamma mRNAs and proteins in infected
cells.” Proc Neg] Afld Sci U S A 90: 6701-5.

Regier, J. L., F. Shen and S. J. Triezenberg (1993). “Pattern of aromatic and
hydrophobic amino acids critical for one of two subdomains of the VP16
transcriptional activator.” Proc Natl Acad Sci U S A 90: 883-7.

Rice, S. A., M. C. Long, V. Lam, P. A. Schaffer and C. A. Spencer(1995). “Herpes
simplex virus immediate-early protein ICP22 is required for viral modification of
host RNA polymerase II and establishment of the normal viral transcription
program.” J Virol 69: 5550-9.

Roizman, B., R. Whitley and C. Lopez (1993). Human Herpesviraga. New York,
Raven Press.

Roller, R. J ., L. L. Monk, D. Stuart and B. Roizman (1996). “Structure and function
in the herpes simplex virus 1 RNA-binding protein U(s)1]: mapping of the
domain required for ribosomal and nucleolar association and RNA binding in
vitro.” J Virol 70: 2842-51.

Sadowski, 1., J. Ma, S. Triezenberg and M. Ptashne (1988). “GAL4-VP16 is an
unusually potent transcriptional activator.” Nature 335: 563-4.

Salghetti, S. E., A. A. Caudy, J. G. Chenoweth and W. P. Tansey (2001).
“Regulation of transcriptional activation domain function by ubiquitin.” Science
293: 1651-3.

Salghetti, S. E., M. Muratani, H. Wijnen, B. Futcher and W. P. Tansey (2000).
“Functional overlap of sequences that activate transcription and signal ubiquitin-
mediated proteolysis.” Proc Natl Acad Sci U S A 97: 3118-23.

Sandri-Goldin, R. M. (1994). “Properties of an HSV-1 regulatory protein that appears
to impair host cell splicing.” Infect Agents Dis 3: 59-67.

Schaerer-Uthurralt, N., M. Erard, K. Kindbeiter, J. J. Madjar and J. J. Diaz (1998).
“Distinct domains in herpes simplex virus type 1 US11 protein mediate post-
transcriptional transactivation of human T-lymphotropic virus type I envelope

162

 

glycoprotein gene expression and specific binding to the Rex responsive
element.” J Gen Virol 79: 1593-602.

Schagger, H. and G. v. J agow (1987). “Tricine-sodium dodecyl sulfate-polyacrylanride
gel electrophoresis for the separation of proteins in the range from 1 to 100 KDa.”
Anal. Biochem. 166: 368-379.

Sclnnelter, J., J. Knez, J. R. Smiley and J. P. Capone (1996). “Identification and
characterization of a small modular domain in the herpes simplex virus host
shutoff protein sufficient for interaction with VP16.” J Virol 70: 2124-31.

 

Sciabica, K. S., Q. J. Dai and R. M. Sandri-Goldin (2003). “ICP27 interacts with
SRPKl to mediate HSV splicing inhibition by altering SR protein
phosphorylation.” Embo J 22: 1608-19.

Sciortino, M. T., M. Suzuki, B. Taddeo and B. Roizman (2001). “RNAs extracted
from herpes simplex virus 1 virions: apparent selectivity of viral but not cellular
RNAs packaged in virions.” J Virol 75: 8105-16.

Sciortino, M. T., B. Taddeo, A. P. Poon, A. Mastino and B. Roizman (2002). “Of the
three tegument proteins that package mRNA in herpes simplex virions, one

(VP22) transports the mRNA to uninfected cells for expression prior to viral
infection.” Proc Natl Agd Sci U S A 99: 8318-23.

Scott, E. S. and P. O'Hare (2001). “Fate of the inner nuclear membrane protein lamin B
receptor and nuclear lamins in herpes simplex virus type 1 infection.” J Virol 75:
8818-30.

Sears, A. E., I. W. Halliburton, B. Meignier, S. Silver and B. Roizman (1985).
“Herpes simplex virus 1 mutant deleted in the alpha 22 gene: growth and gene

expression in permissive and restrictive cells and establishment of latency in
mice.” J Virol 55: 338-46.

Shaw, P., J. Knez and J. P. Capone (1995). “Amino acid substitutions in the herpes
simplex virus transactivator VP16 uncouple direct protein-protein interaction and

DNA binding from complex assembly and transactivation.” J Biol Chem 270:
29030-7.

Shen, F., S. J. Triezenberg, P. Hensley, D. Porter and J. R. Knutson (1996a).
“Critical amino acids in the transcriptional activation domain of the herpesvirus

protein VP16 are solvent-exposed in highly mobile protein segments. An intrinsic
fluorescence study.” J Biol Chem 271: 4819-26.

Shen, F., S. J. Triezenberg, P. Hensley, D. Porter and J. R. Knutson (1996b).
“Transcriptional activation domain of the herpesvirus protein VP16 becomes

163

conforrnationally constrained upon interaction with basal transcription factors.” 1
Biol Chem 271: 4827—37.

Shieh, S. Y., M. Ikeda, Y. Taya and C. Prives (1997). “DNA damage-induced
phosphorylation of p53 alleviates inhibition by MDM2.” _QQLI 91: 325-34.

Simmen, K. A., A. Newell, M. Robinson, J. S. Mills, G. Canning, R. Handa, K.
Parkes, N. Borkakoti and R. Jupp (1997). “Protein interactions in the herpes
simplex virus type 1 VP16-induced complex: VP16 peptide inhibition and
mutational analysis of host cell factor requirements.” J Virol 71: 3886-94.

Smibert, C. A., B. Popova, P. Xiao, J. P. Capone and J. R. Smiley (1994). “Herpes
simplex virus VP16 forms a complex with the virion host shutoff protein vhs.” ._J_
Virol 68: 2339-46.

Smiley, J. R. and J. Duncan (1997). “Truncation of the C-terrninal acidic
transcriptional activation domain of herpes simplex virus VP16 produces a
phenotype similar to that of the in1814 linker insertion mutation.” J Virol 71:
6191-3.

Smiley, J. R., M. M. Elgadi and H. A. Saffran (2001). “Herpes simplex virus vhs
protein.” Methods Enzymol 342: 440-51.

Smith, J. S. and N. J. Robinson (2002). “Age-specific prevalence of infection with
herpes simplex virus types 2 and l: a global review.” J Infect Dis 186 Suppl 1:
S3-28.

Sodeik, B., M. W. Ebersold and A. Helenius (1997). “Microtubule-mediated transport
of incoming herpes simplex virus 1 capsids to the nucleus.” J Cell Biol 136: 1007-
21.

Stark, M. R., D. Escher and A. D. Johnson (1999). “A trans-acting peptide activates
the yeast a1 repressor by raising its DNA-binding affinity.” Embo J 18: 1621-9.

Steiner, 1., J. G. Spivack, S. L. Deshmane, C. 1. Ace, C. M. Preston and N. W.
Fraser (1990). “A herpes simplex virus type 1 mutant containing a
nontransinducing Vmw65 protein establishes latent infection in vivo in the

absence of viral replication and reactivates efficiently from explanted trigeminal
ganglia.” J Virol 64: 1630-8.

Stern, S. and W. Herr (1991). “The herpes simplex virus trans-activator VP16
recognizes the Oct-1 homeo domain: evidence for a homeo domain recognition
subdomain.” Genes Dev 5: 2555-66.

164

 

 

Stern, S., M. Tanaka and W. Herr (1989). “The Oct-1 homoeodomain directs formation
of a multiprotein-DNA complex with the HSV transactivator VP16.” Nature 341:
624-30.

Su, Y. H., M. J. Moxley, A. K. Ng, J. Lin, R. Jordan, N. W. Fraser and T. M.
Block (2002). “Stability and circularization of herpes simplex virus type 1
genomes in quiescently infected PC12 cultures.” J Gen Virol 83: 2943-50.

Sullivan, S. M., P. J. Horn, V. A. Olson, A. H. Koop, W. Niu, R. H. Ebright and S.
J. Triezenberg (1998). “Mutational analysis of a transcriptional activation region
of the VP16 protein of herpes simplex virus.” Nucleic Acids Res 26: 4487-96.

Suszkiw, J. (2001). “Cracking the Code of Marek's Disease.” Agricultural Researc_h
Magazine 49: 16-17.

Suzuki, N., W. Peter, T. Ciesiolka, P. Gruss and H. R. Scholer (1993). “Mouse Oct-1
contains a composite homeodomain of human Oct-1 and Oct-2.” Nucleic Acids
Rfi 21: 245-52.

Taddeo, B., A. Esclatine and B. Roizman (2002). “The patterns of accumulation of
cellular RNAs in cells infected with a wild-type and a mutant herpes simplex
virus 1 lacking the virion host shutoff gene.” Proc NaLAcad Sci U S A 99:
17031-6.

Tal-Singer, R., R. Pichyangkura, E. Chung, T. M. Lasner, B. P. Randazzo, J. Q.
Trojanowski, N. W. Fraser and S. J. Triezenberg (1999). “The transcriptional
activation domain of VP16 is required for efficient infection and establishment of
latency by HSV-1 in the murine peripheral and central nervous systems.”
Virology 259: 20-33.

Thomas, S. K., G. Gough, D. S. Latchman and R. S. Coffin (1999). “Herpes simplex
virus latency-associated transcript encodes a protein which greatly enhances virus
growth, can compensate for deficiencies in immediate-early gene expression, and
is likely to function during reactivation from virus latency.” J Virol 73: 6618-25.

 

Thomas, S. K., C. E. Lilley, D. S. Latchman and R. S. Coffin (2002). “A protein
encoded by the herpes simplex virus (HSV) type 1 2-kilobase latency-associated
transcript is phosphorylated, localized to the nucleus, and overcomes the

repression of expression from exogenous promoters when inserted into the
quiescent HSV genome.” J Virol 76: 4056-67.

Triezenberg, S. J., R. C. Kingsbury and S. L. McKnight (1988a). “Functional

dissection of VP16, the trans-activator of herpes simplex virus immediate early
gene expression.” Genes Dev 2: 718-29.

165

 

Triezenberg, S. J., K. L. LaMarco and S. L. McKnight (1988b). “Evidence of DNA:
protein interactions that mediate HSV-1 immediate early gene activation by
VP16.” Genes Dev 2: 730-42.

Uesugi, M., O. Nyanguile, H. Lu, A. J. Levine and G. L. Verdine (1997). “Induced
alpha helix in the VP16 activation domain upon binding to a human TAF.”
Science 277: 1310-3.

Unger, T., T. Juven-Gershon, E. Moallem, M. Berger, R. Vogt Sionov, G. Lozano, M.
Oren and Y. Haupt (1999a). “Critical role for Ser20 of human p53 in the
negative regulation of p53 by Mdm2.” Embo J 18: 1805-14.

Unger, T., R. V. Sionov, E. Moallem, C. L. Yee, P. M. Howley, M. Oren and Y.
Haupt (1999b). “Mutations in serines 15 and 20 of human p53 impair its
apoptotic activity.” Oncogene 18: 3205-12.

Vacratsis, P. O., B. R. Phinney, D. A. Gage and K. A. Gallo (2002). “Identification
of in Vivo Phosphorylation Sites of MLK3 by Mass Spectrometry and
Phosphopeptide Mapping.” Biochemistry 41: 5613 -5624.

van Genderen, I. L., R. Brandimarti, M. R. Torrisi, G. Campadelli and G. van Meer
(1994). “The phospholipid composition of extracellular herpes simplex virions
differs from that of host cell nuclei.” Virology 200: 831-6.

Van Sant, C., R. Hagglund, P. Lopez and B. Roizman (2001). “The infected cell
protein 0 of herpes simplex virus 1 dynamically interacts with proteasomes, binds
and activates the cdc34 E2 ubiquitin- conjugating enzyme, and possesses in vitro
E3 ubiquitin ligase activity.” Proc Natl Acad Sci U S A 98: 8815-20.

Vogel, J. L. and T. M. Kristie (2000). “The novel coactivator C1 (HCF) coordinates
multiprotein enhancer formation and mediates transcription activation by GABP.”
Embo J 19: 683-90.

Warner, S. C., P. Desai and S. Person (2000). “Second-site mutations encoding
residues 34 and 78 of the major capsid protein (VPS) of herpes simplex virus type
1 are important for overcoming a blocked maturation cleavage site of the capsid
scaffold proteins.” Virology 278: 217-26.

Weinheimer, S. P., B. A. Boyd, S. K. Durham, J. L. Resnick and D. R. O'Boyle,
2nd (1992). “Deletion of the VP16 open reading frame of herpes simplex virus
type 1.” J Virol 66: 258-69.

Werstuck, G. and J. P. Capone (1989). “Mutational analysis of the herpes simplex
virus trans-inducing factor Vmw65.” Gene 75: 213-24.

166

 

White, C. A., N. D. Stow, A. H. Patel, M. Hughes and V. G. Preston (2003).
“Herpes simplex virus type 1 portal protein UL6 interacts with the putative
tenninase subunits UL15 and UL28.” J Virol 77: 6351-8.

Wilcox, K. W., A. Kohn, E. Sklyanskaya and B. Roizman (1980). “Herpes simplex
virus phosphoproteins. I. Phosphate cycles on and off some viral polypeptides
and can alter their affinity for DNA.” J Virol 33: 167-82.

Wilson, A. C., R. N. Freiman, H. Goto, T. Nishimoto and W. Herr (1997). “VP16
targets an amino—terminal domain of HCF involved in cell cycle progression.”
M01 Cell Biol 17: 6139-46.

Wilson, A. C., K. LaMarco, M. G. Peterson and W. Herr (1993). “The VP16
accessory protein HCF is a family of polypeptides processed from a large
precursor protein.” Qgfl 74: 115-25.

Wilson, A. C., J. E. Parrish, H. F. Massa, D. L. Nelson, B. J. Trask and W. Herr
(1995a). “The gene encoding the VP16-accessory protein HCF (HCF C1) resides
in human Xq28 and is highly expressed in fetal tissues and the adult kidney.”
Genomics 25: 462-8.

Wilson, A. C., M. G. Peterson and W. Herr (1995b). “The HCF repeat is an unusual
proteolytic cleavage signal.” Genes Dev 9: 2445-58.

Woychik, N. A. and M. Hampsey (2002). “The RNA polymerase II machinery:
structure illuminates function.” Qafl 108: 453-63.

Wysocka, J. and W. Herr (2003a). “The herpes simplex virus VP16-induced complex:
the makings of a regulatory switch.” Trends Biochem Sci 28: 294-304.

Wysocka, J., M. P. Myers, C. D. Laherty, R. N. Eisenman and W. Herr (2003b).
“Human Sin3 deacetylase and trithorax-related Setl/Ash2 histone H3-K4

methyltransferase are tethered together selectively by the cell- proliferation factor
HCF-1.” Genes Dev 17: 896-911.

Xiao, P. and J. P. Capone (1990). “A cellular factor binds to the herpes simplex virus
type 1 transactivator Vmw65 and is required for Vmw65-dependent protein-DNA
complex assembly with Oct-1.” M01 Cell Biol 10: 4974-7.

Zelus, B. D., R. S. Stewart and J. Ross (1996). “The virion host shutoff protein of
herpes simplex virus type 1: messenger ribonucleolytic activity in vitro.” J Virol
70: 2411-9.

Zhang, Y., D. A. Sirko and J. L. McKnight (1991). “Role of herpes simplex virus type

1 UL46 and UL47 in alpha TIP-mediated transcriptional induction:
characterization of three viral deletion mutants.” J Virol 65: 829-41.

167

Zhou, Z. H., D. H. Chen, J. Jakana, F. J. Rixon and W. Chiu (1999). “Visualization
of tegument-capsid interactions and DNA in intact herpes simplex virus type 1
virions.” J Virol 73: 3210-8.

168