Citation
Cloning and characterization of Epstein-Barr virus glycoprotein gp85

Material Information

Title:
Cloning and characterization of Epstein-Barr virus glycoprotein gp85
Creator:
Yaswen, Linda Ruth
Publication Date:
Language:
English
Physical Description:
viii, 137 leaves : illustrations ; 29 cm.

Subjects

Subjects / Keywords:
Research ( mesh )
Herpesvirus 4, Human ( mesh )
Viral Envelope Proteins -- genetics ( mesh )
Viral Envelope Proteins -- chemistry ( mesh )
Membrane Proteins -- genetics ( mesh )
Membrane Proteins -- chemistry ( mesh )
Glycoproteins -- genetics ( mesh )
Glycoproteins -- chemistry ( mesh )
Molecular Sequence Data ( mesh )
Amino Acid Sequence ( mesh )
Department of Pathology and Laboratory Medicine thesis Ph.D ( mesh )
Dissertations, Academic -- College of Medicine -- Department of Pathology and Laboratory Medicine -- UF ( mesh )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )
Academic theses. ( lcgft )
Academic theses ( fast )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 1993.
Bibliography:
Includes bibliographical references (leaves 121-136).
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Linda Ruth Yaswen.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
029752060 ( ALEPH )
51986277 ( OCLC )

Downloads

This item has the following downloads:


Full Text
CLONING AND CHARACTERIZATION
OF EPSTEIN-BARR VIRUS GLYCOPROTEIN gp85
By
LINDA RUTH YASWEN
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA


DEDICATION
This dissertation is dedicated to my parents who inspire me to reach for the stars.
11


ACKNOWLEDGMENTS
Great appreciation and thanks for all those who have helped complete this work.
Thanks to my committee members, Drs. Maureen Goodenow, Edward Wakeland and
Ammon Peck who have been extremely encouraging and helpful. A very special thanks
to the co-chair of my committee, Dr. Edward Stephens, who has been both my mentor
and friend. My deepest gratitude to the person who has made this all possible, the chair
of my committee, Dr. Lindsey Hutt-Fletcher. She has been an inspirational teacher.
Many thanks to Susan who has made the mountains seem like small hills; to Paula,
Steven, and Minh-Thanh who have been such incredible friends throughout the years; to
my parents, sister and brother for their wonderful senses of humor and enormous
support, and to John who has endured the long hours with me and become such a big
part of my world.
iii


CONTENTS
CHAPTER 1
INTRODUCTION 1
General Characteristics of Epstein-Barr Virus 1
Disease Associations of EBV 2
Virus Replication in Vitro 4
Early Events in Viral Infection 5
CHAPTER 2
CLONING OF gp85 18
Introduction 18
Materials and Methods 21
DNA sequencing 21
Construction of shuttle vectors 22
Blunt ending of DNA 24
Colony blot hybridization 24
Cell lines 25
Generation of recombinants 26
Virus 27
DNA dot blot analysis 27
Preparation of DNA from colonies 28
Large scale preparation of DNA 29
Results 30
Discussion 46
CHAPTER 3
EXPRESSION AND BIOCHEMICAL CHARACTERIZATION
OF RECOMBINANT gp85 48
Introduction 48
Materials and Methods 50
Cells and virus 50
Antibody purification 51
IV


Enzyme-linked immunosorbant assay 52
Analysis of vaccinia proteins 53
Electrophoresis of proteins 54
Enzyme and glycosylation inhibitors 55
Silver staining 56
Results 56
Discussion 65
CHAPTER 4
COMPARISON OF NATIVE AND RECOMBINANT gp85 67
Introduction 67
Materials and Methods 69
Cells and viral induction 69
Radiolabeling of native gp85 in Akata cells 69
Extrinsic labeling of cells 70
Antibodies 71
Immunofluorescence 71
Results 72
Discussion 85
CHAPTER 5
ASSOCIATION OF THE EBV BKRF2 GENE PRODUCT
WITH NATIVE gp85 87
Introduction 87
Materials and Methods 88
Cells and virus 88
Antibodies 89
Immunofluorescence 90
Boiling analysis 91
Sucrose gradient centrifugation 91
Coinfection of AHV-1 and vaccinia virus 92
Results 92
Discussion 104
v


CHAPTER 6
CONSTRUCTION OF BACTERIAL FUSION PROTEINS 106
Introduction 106
Materials and Methods 107
Polymerase chain reaction 107
Western blot analysis 107
Purification of bacterial fusion proteins 108
Results 109
Discussion 116
CHAPTER 7
CONCLUSION 117
vi


Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
CLONING AND CHARACTERIZATION OF EPSTEIN-BARR VIRUS
GLYCOPROTEIN gp85
By
Linda Ruth Yaswen
May 1993
Chairman: Lindsey Hutt-Fletcher
Major Department: Pathology and Laboratory Medicine
The Epstein-Barr virus (EBV) glycoprotein gp85 is the product of the BXLF2
gene and is found in the viral envelope as well as in the plasma membrane of EBV
infected cells. Although it plays a critical role in the penetration of virus through the B
lymphocyte membrane, only very small amounts of it are made in EBV infected cells.
To facilitate the structural and functional analysis of gp85, recombinant vaccinia viruses
were constructed to overexpress the glycoprotein. Recombinant gp85 was recognized by
a polyclonal antibody made to a peptide derived from the predicted amino acid sequence
of the BXLF2 gene. However, it was not recognized by either of two monoclonal
antibodies made to the native molecule. In contrast to native gp85, the recombinant
protein contained no sugars that were resistant to endoglycosidase H and its synthesis
was unaffected by inhibitors that influence processing in the Golgi complex.
Furthermore, the recombinant protein was not detected at the cell surface by indirect
vii


immunofluorescence assays nor was it accessible to labeling with radioactive iodine.
These data indicate that the recombinant molecule, as opposed to native gp85, is not
transported to the cell surface.
When native proteins were labeled metabolically with [3H]glucosamine or
extrinsically with radioactive iodine, two glycoproteins with apparent Mrs of 25,000 and
42,000, were found to associate with gp85. Homologues of gp85 in other herpesviruses
require additional viral proteins for processing and transport of recombinant molecules.
Within the EBV genome, the BKRF2 open reading frame has been identified as the
positional homologue of these chaperones. Antibodies made to a peptide derived from
the predicted BKRF2 sequence, immunoprecipitated glycoproteins with mobilities of
85,000 and 25,000 daltons. Gradient centrifugation and immunoprecipitation of proteins
from EBV infected cells confirmed that the 25,000 dalton protein cosedimented with
gp85. Furthermore, the anti-BKRF2 antibody was shown to react with the same 25,000
dalton protein that was isolated in complexed form with native gp85. These data suggest
that the BKRF2 gene product is specifically recognized by the anti-BKRF2 antibody and
that this protein associates with gp85 during processing. It is possible that gp85 is
biologically active only if this association takes place.
viii


CHAPTER 1
INTRODUCTION
General Characteristics of Epstein-Barr Virus
Epstein-Barr virus (EBV) is an enveloped DNA virus classified with the
herpesviridae. It contains a linear double-stranded DNA molecule of approximately
172,000 base pairs (Beisel et al., 1985). The genome consists of five large regions of
unique DNA domains, U1-U5, which are separated by four regions of internal repeats,
11-14. The genome is flanked on both ends by tandem direct repeats (Cameron et al.,
1987; Daumbaugh et al., 1982; Given et al., 1979). Epstein-Barr virus is a member of
the gamma herpesvirus subfamily, and is designated as the human herpesvirus type 4
(Roizman, B., 1982). The diameter of the mature virion is approximately 150 to 180 nm.
The viral envelope is acquired as the virus buds through the host cell; however, the
proteins within the envelope are virally derived (Spear, P. G., 1980). At least five virally
encoded proteins have been detected within the envelope of EBV. The plasma
membrane of productively infected cells contain these proteins as well. The molecular
masses of these proteins are approximately 300-350, 200-250, 140, 55-78 and 85
kilodaltons (Edson and Thorley-Lawson, 1981; Thorley-Lawson and Edson 1979;
Thorley-Lawson, D.A., 1988; Mackett et al., 1990). The proteins gp350/300,
gp250/200, gp55/78 and gp85 are glycosylated molecules, while the fifth, pi40, is not
(Balachandran et al., 1986). Within the envelope of the virus lies an icosahedral
1


2
nucleocapsid which is surrounded by an amorphous tegument. The nucleocapsid
contains the double-stranded DNA molecules in association with core proteins.
Disease Associations of EBV
Gamma herpesviruses have a very limited host range. All members of the group
infect lymphoblastoid cells in vivo and in vitro. EBV is the only member of the group
that infects humans and it was originally identified as having tropism for human B
lymphocytes. Other members of the group include Marek's disease virus of chickens,
herpes teles and herpes samiri virus of new world monkeys and murine herpes 68
(Efstathiou et al., 1990). These viruses infect T cells (Fleckenstein and Desrosiers, 1982;
Nonoyama, 1982). The host range of EBV in vitro is restricted to B lymphocytes of
humans and new world primates. EBV establishes latency in these cells and
immortalizes them. Latently infected cells usually contain more than one copy of the
complete EBV genome. The genome can be integrated, but is most often found as a
covalently closed circular episome (Lindahl et al., 1976). Recombination events between
terminally repeated DNA sequences allow the linear genome to circularize (Dambaugh et
al., 1980). The stable covalently closed circle (CCC) is maintained as an autonomously
replicating episome.
Epithelial cells have been identified as a second target for EBV infection (Sixbey
et al., 1987; Sixbey et al., 1984). Thus, the virus can persist in both the epithelial and B
cells (Thorley-Lawson, 1988). The epithelial cell is permissive for replication and is
thought to be the source of virus that is shed in the oropharynx. Entry of EBV into the
oropharynx and subsequent replication at that site represents the initial step in


3
pathogenesis of any primary EBV infection. Viral replication within the pharyngeal
epithelium (Sixbey et al., 1984; Sixbey et al., 1983) allows for the infection and
subsequent transformation of B lymphocytes. This transformation event enables the
cells to undergo rapid proliferation (Rickinson, 1987; Svedmer et al., 1984).
Approximately ninety-five percent of the adult population worldwide is infected with
EBV. Primary infection with EBV normally occurs during childhood and is clinically
silent. Elowever, if primary infection is delayed until young adulthood, it leads to
infectious mononucleosis (IM) in approximately fifty percent of cases (Niederman et al.,
1970).
EBV has also been associated with several proliferative disorders of both
lymphoid and epithelial origin. Initially, the association of EBV with nasopharyngeal
carcinoma (NPC) was based on the observation that high antibody titers to EBV were
found in all NPC patients (de-The and Zeng, 1987). In all cases of NPC, the EBV DNA
is found in the CCC episomal form (Thorley-Lawson, 1988). EBV infection is
hypothesized to be an important factor in NPC tumor progression. NPC is one of the
most common tumors in Southern China. Since this represents such a large population
of the world, NPC must be considered a significant world health problem. EBV is also
associated with the development of a second tumor called Burkitf s lymphoma. The viral
DNA is found in the majority of endemic Burkitf s lymphomas which is the most
common children's tumor seen in Africa (Burkitt, 1987). In addition to EBV, these
tumors contain characteristic chromosomal translocations. The c-myc region from
chromosome 8 is juxtaposed with either the immunoglobin heavy chain region located


4
on chromosome 14, the lambda light chain genes of chromosome 22 or the kappa light
chain genes of chromosome 2 (Lenoir, 1987). The unrestricted proliferation of B cells
caused by EBV infection may act synergistically with the immunosuppression caused by
holoendemic malarial infections in this region. These two factors acting in cooperation
with one another may increase the emergence of malignant cell clones.
EBV is also hypothesized to contribute to the development of post transplant
disorders. Immunosuppressive drugs such as cyclosporine A can disrupt homeostatic
EBV infections within post transplant patients. This may allow for the development of
lymphoproliferation (Cleary et al., 1986). The presence of EBV in AIDS patients causes
an increased risk of both malignant lymphoma development as well as development of
oral hairy leukaplakia (OHL). OHL is an EBV-induced proliferative lesion of the lateral
tongue epithelium.
Virus Replication in Vitro
Cultures of EBV-infected lymphocytes vary in their permissiveness for viral
replication: most cultures are nonpermissive, but replication does occur in a small
fraction of cells in some cultures. The nonpermissiveness of EBV infection has made it
difficult to study virus replication and also limits the amounts of purified virus available
for studying the components of mature virus particles. Clones of infected lymphocytes
that are more permissive of virus replication have been selected (Miller and Lipman,
1973) and have facilitated studies of the virus replication cycle and biochemical
analyses. Two isolates of EBV have been extensively studied, B95-8 and P3HR1. The
B95-8 strain is produced by a cell line derived from a clone of marmoset lymphocytes.


5
These lymphocytes had been infected with virus obtained from a patient with infectious
mononucleosis. The B95-8 viral DNA has been completely sequenced (Baer et al.,
1984) and has been the prototype used for gene mapping. The P3HR1 cell line is a clone
of the Jijoye Burkitt tumor derived cell line (Hinuma et al., 1967). The P3HR1 cells are
more permissive than the parent clone for virus replication and the virus produced by
P3HR1 cells lacks the ability to growth-transform noninfected B lymphocytes (Miller et
al., 1974).
Early Events in Viral Infection
Attachment, penetration and uncoating represent the earliest events occurring in a
viral replication cycle. Initially, the virus attaches via a membrane protein to a specific
cellular receptor. This specific interaction in part determines the type of cell that a virus
can infect. Other functions of membrane proteins include viral fusion and penetration.
The penetration event allows for liberation of the nucleocapsid into the cytoplasm of the
virally infected host cell. The ability of EBV to infect B lymphocytes is initiated by
attachment of the virus to the cellular CR2 receptor. CR2 is also known as CD21. The
physiologic ligand of the CR2 glycoprotein is the C3d fragment of complement
(Fingeroth et al., 1984; Nemerow et al., 1985). Epithelial cells in the oro- and
nasopharyngeal epithelium also express a receptor for EBV attachment. This receptor
may be a similar or related molecule to CR2 (Sixby et al., 1987; Young et al., 1989). A
loss of EBV cellular receptors occurs as both epithelial cells and B lymphocytes
differentiate. Attachment of EBV to CR2 is mediated by at least one viral membrane
glycoprotein. This glycoprotein, gp350/300 (Fingeroth et al., 1984; Nemerow et al.,


6
1985), is encoded by the BLLF1 open reading frame of EBV. BLLF1 also encodes for a
smaller membrane glycoprotein called gp220. The two glycoproteins represent products
of a differentially spliced RNA transcript (Beisel et al., 1985). While it is not clear
whether gp220 is also involved in the viral attachment event, it has been suggested that
the two glycoproteins bind more effectively to different forms of the EBV cellular
receptor.
Membrane fusion occurs as two lipid bilayers adhere and join to one another.
Endocytosis, exocytosis, cell division and cell fusion are only some of the cellular events
which involve this fundamental event. Within the cell, fusion is important for
intracellular communication. Here fusion allows for the transport vesicles to bud from
certain cellular organelles and fuse with others. The mechanism of membrane fusion is
still under investigation. Due to a relatively simple membrane composition, enveloped
viruses provide an excellent means of probing this event. Enveloped viruses enter cells
by fusing with the lipid bilayer of the cell membrane (Lonberg-Holm and Philipson,
1974; White et al., 1983). Two pathways of viral entry are commonly utilized. Some
viruses, such as Sendai virus (Scheid and Choppin, 1976), deposit their nucleocapsids
directly into the cytoplasm by fusing with the cellular plasma membrane at physiologic
pH. The alternative route utilizes receptor-mediated endocytosis followed by fusion with
the vesicle membrane. Macromolecules are taken into cells by this same process. This
pathway is initiated by binding of a ligand to a cell surface receptor. Subsequently, the
membrane invaginates to form a vesicle (Goldstein et al., 1979). For some viruses,
particular regions of the plasma membrane have been identified as the sites for


receptor-mediated endocytosis (Goldstein et al., 1979). These regions are called coated
pits. The protein clathrin is a major component within these pits and is thought to
participate in the early stages of endocytosis (Doxsey et al., 1987). Receptors and
receptor-ligand complexes are thought to be concentrated within coated pits at sites of
internalization (Pearse, 1975). Conditions that trigger fusion of the viral envelope with
the vesicle membrane are provided as the endosomal environment becomes acidified.
This process of receptor-mediated endocytosis and vesicle membrane fusion is utilized
for viral entry by Semliki Forest virus (SFV) (Helenius et al., 1980; Marsh and Helenius,
1980), influenza A (Matlin et al., 1981; White and Helenius, 1983; White et al., 1981)
and vesicular stomatitis virus (VSV) (White et al., 1983).
The major entry mechanism for human immunodeficiency virus (HIV) is reported
to be fusion with the plasma membrane at the cell surface. McClure et al. have reported
that the entry of HIV-1 occurs by a pH-independent mechanism (McClure et al., 1988).
However, these data are in disagreement with the conclusion of Maddon et al. (1986) and
Pauza and Price (1988) who proposed that HIV entry into T lymphoblastoid cells
occurred via an endocytic entry pathway. Therefore, the possibility exists that virus
could enter by both pathways in a pH independent manner.
Studies utilizing electron microscopy and immunoelectron microscopy have
reported that EBV fuses directly at the plasma membrane of the Raji lymphoblastoid cell
line (Nemerow and Cooper, 1984; Seigneurin et al., 1977). EBV nucleocapsids were
found in the cytoplasm directly beneath the cellular plasma membrane. Virus was never
found to be bound to clathrin-coated areas of the plasma membrane, nor was the virus


8
observed in endocytic vesicles. However, studies using normal B lymphocytes revealed
transfer of membrane bound virus into membrane vesicles. These vesicles were distinct
in size and appearance from clathrin-coated vesicles and after 30 minutes, very few
virus particles remained in the vesicles. Miller and Hutt-Fletcher (1992) compared the
penetration events of EBV in epithelial and B lymphocytes. The penetration of EBV into
nonpolarized suspensions of epithelial cells was reported to occur by fusion at the cell
surface while it appeared that EBV fuses with normal B cells after being endocytosed.
Furthermore, both of these events appeared to be pH independent. EBV fusion events
were monitored using the pH insensitive fluorophore octadecyl rhodamine B chloride
(R18) and the pH dependent 5(N-octadecanoyl) aminofluorescein (AF). The AF looses
emission intensity when the pH falls below 7.4. Fusion was detected in both cell types
using R18 labeled EBV. However, penetration with AF labeled EBV could be detected
only in epithelial cells and not B cells unless they were first pretreated with drugs that
raise endosomal pH.
Fusion of two lipid bilayers is energetically unfavorable. This is due to a strong
repulsion that occurs between the two membranes as they approach one another. Energy
is required to displace water molecules from the hydrophilic surfaces. This hydration
force plus electrostatic forces between the two bilayers, prevents fusion events from
occurring spontaneously. Van der Waal forces can provide a weak attraction between
the membranes at distances of 20-30 angstroms. This enables the early events of
aggregation and attachment to occur; however, the combined repulsive forces soon
become overwhelming. Therefore, it seems likely that specialized fusogenic proteins


9
catalyze membrane fusion. Cellular fusogenic proteins have not been clearly identified,
whereas viral fusogenic proteins have been demonstrated. Viral fusion is mediated by a
specific viral membrane fusion protein. Hydrophobic regions within the viral fusion
protein are believed to destabilize the bilayer structure of the target membrane (Roizman,
1982). Several viruses have well characterized fusion proteins including Sendai, Semliki
Forest, influenza, vesicular stomatitis virus, and human immunodeficiency virus (HIV).
The fusion proteins of these viruses are all glycoproteins which span the lipid bilayer and
have the majority of their mass exposed externally.
Two proteins are contained within the envelope of Sendai virus which is a
member of paramyxoviridae. The hemagglutinin-neuraminidase (HN) protein is
responsible for attachment of the virus to cell surface sialic acid residues. The fusion (F)
protein initiates viral penetration at the plasma membrane (Hsu et al., 1981; Scheid and
Choppin, 1974; Scheid and Choppin, 1976). The F protein consists of two
sulfhydryl-linked glycopeptides called FI and F2. These two glycopeptides result from
proteolytic cleavage of the inactive precursor F0. The proteolytic cleavage occurs by a
host cell protease enzyme (Hsu et al., 1982). Virus produced by cells that lack a suitable
protease for F protein activation are noninfectious (Hsu et al., 1982). The F2 protein
corresponds to the amino terminus of F0, and the protein is anchored in the bilayer
through FI. The amino terminus of FI, resulting after cleavage of F0, has been found to
be unusually hydrophobic (Gething et al., 1978). It has been suggested that this
hydrophobic domain might be involved in the fusion event. Furthermore, the amino acid


10
sequences in this region are highly conserved among paramyxoviruses (Scheid et al.,
1978).
The orthomyxoviruses also have two types of spike glycoproteins. One of the
glycoproteins is neuraminidase (NA). The other, hemagglutinin (HA), has the capability
to bind to cell surface sialic acid residues and to catalyze fusion (Choppin and Compans,
1975; White et al., 1982). Unlike paramyxoviruses, othomyxoviruses are endocytosed
and fuse with the endocytic vesicle. The HA consists of two disulphide linked
glycopeptide chains, HA1 and HA2, resulting from proteolytic cleavage of a precursor
glycoprotein HA^ (Lazarowitz and Choppin, 1975; White et al., 1983). The cleavage is
irrelevant to adsorption, but is a prerequisite for infectivity (White et al., 1983). The
cleavage generates a new amino terminus on HA2 which is hydrophobic and highly
conserved in different influenza strains and somewhat homologous with the amino
terminus of FI. The HA molecule in its neutral form is a trimer and the hydrophobic
fusion peptide in each monomer is unexposed until the low pH of the endocytic vesicle
causes partial dissociation of the HA trimer. It is thought that this dissociation exposes
the fusion peptide which can insert into the target bilayer ( Dorns et al., 1985; Schlegel et
al., 1982) and initiate endosomal membrane fusion.
The envelope spikes of Semliki Forest virus (SFV), a togavirus, consist of a three
glycopeptide complex El, E2 and E3. El and E2 are transmembrane glycoproteins. E3
is non-covalently associated with E2 and is external to the bilayer. SFV does not fuse
with the plasma membrane at physiologic pH (Helenius et al., 1980). Virions are
endocytosed and a drop in pH within the endocytic vesicle activates membrane fusion


11
(Marsh et al., 1983). Lysosomotropic agents which elevate endosomal pH inhibit SFV
penetration (Helenius et al., 1982). SFV can fuse directly with the plasma membrane in
vitro at low pH (White et al., 1980). The SFV spike glycoproteins have been shown to
be fusogenic in the absence of other virus components (Marsh et al., 1983). It has been
suggested that the role of peptide El is directly linked to the fusion event. Both SFV and
Sindbis, another togavirus, have El proteins containing a hydrophobic peptide segment
located close to the amino terminus and this segment has an external position in the virus
membrane (Garoff et al., 1980; White et al., 1983). Since El and E2 occur as a complex,
E2 may also participate in the fusion reaction. The role of E3 is not clear; it is a small
peripheral glycopeptide and there is no homologue in Sindbis virus (Welch and Sefton,
1979).
Vesicular stomatitis virus (VSV), a rhabdovirus, has only one envelope
glycoprotein (the G-protein). The G-protein has a hydrophobic region near the carboxyl
terminus forming the membrane spanning domain. A small hydrophilic sequence at the
carboxyl terminus is in contact with the cytoplasm. The larger amino terminal domain,
containing the oligosaccharide chains, is exposed to the exterior of the cell (Rose and
Gallione, 1981; Rose et al., 1980). Fusion activity occurs at the plasma membrane when
VSV is attached to the surface of cells and placed in a low pH medium (Blumenthal et
al., 1987; Matlin et al., 1982). Eukaryotic cells expressing the cloned G-protein gene
fuse at low but not at neutral pH (Rose and Gallione, 1981). These data indicate that the
G-protein is both necessary and sufficient for fusion activity (Reidel et al., 1984).


12
The DNA-containing herpesviruses are considerably more complex than any of
the RNA-containing viruses. The best studied herpesvirus is herpes simplex virus
(HSV). HSV has an envelope that contains at least seven glycoproteins which have been
characterized and sequenced (Bzik et al., 1984; Frink et ah, 1983; Gompels and Minson,
1986; McGeoch et ah, 1985; Pellet et ah, 1985; Watson et ah, 1982). Three of the
glycoproteins, namely gB, gD and gH, induce antibodies capable of neutralizing HSV
infectivity in the absence of complement. All three glycoproteins have also been
implicated in virus penetration (Fuller and Spear, 1985; Gompels and Minson, 1986;
Sarmiento et ah, 1979). Evidence implicating gB in the viral penetration event comes
from studies of temperature sensitive HSV-1 mutants. These mutants fail to process
precursor gB molecules to mature forms at nonpermissive temperatures. Though the
virions produced are noninfectious, they can still bind to cells and block superinfection
with wild type HSV-1. Furthermore, the block in viral penetration is overcome when the
temperature sensitive mutants and cell complexes are treated with the membrane fusing
agent, polyethylene glycol (Little et ah, 1981; Sarmiento et ah, 1979).
Neutralizing anti-gD monoclonal antibodies have been shown to block HSV
infection by preventing virus cell fusion at the plasma membrane (Fuller and Spear,
1987). Antibodies to this glycoprotein also block HSV-induced cell-cell fusion, a
process which may be similar to the virus-cell fusion required for entry (Noble et ah,
1983). In addition, it has also been reported that deletion mutants of gD can bind to cells
but not block superinfection with wild-type HSV (Johnson and Ligas, 1988). The
glycoprotein gH is present in the viral envelope at ten-fold lower concentrations than gD


13
(Richman et al., 1986). Despite this fact, antibodies against gH demonstrate neutralizing
activity comparable to that of antibodies against gD (Cranage et al., 1988). Fuller and
Spear (1987) demonstrated that anti-gH monoclonal antibodies block the fusion of the
virus with the target cell membrane. These monoclonal antibodies do not block viral
attachment. Furthermore, monoclonal antibodies to gH are able to abolish syncytium
formation in those cells infected with syncytial HSV-1 strains (Gompels and Minson,
1986). A temperature sensitive mutant of HSV was made that contained a substitution in
the gH glycoprotein (Desai et al., 1988). The extracellular virus particles produced were
devoid of gH and noninfectious. Recently, it was reported that replacement of deleted
gH coding sequences with the E. coli lacZ gene resulted in mutant virus that could not
enter cells. In contrast to the gD- mutant, however, this gH- virus was able to block
superinfection with wild-type HSV (Forrester et al., 1991). The authors suggest that the
gH functions after gD in the viral entry process. Taken together, these findings indicate
that gH-1 is involved in cellular fusion events. Thus, the three glycoproteins gB, gD and
gH are likely to either induce or influence the fusion process. This fusion event occurs in
a pH-independent manner at the surface of the cell. There is no evidence to suggest that
gB, gD and gH act as a single functional heteropolymer. Homodimers of gB extracted
from virions of infected cells are not associated with other glycoproteins
(Claesson-Welsh and Spear, 1986). Furthermore, gB and gD have been shown to form
morphologically distinct structures in the virion envelope (Stannard et al., 1987).
The viral envelope protein that mediates the fusion event between EBV and the
host cell has not been demonstrated conclusively. However, the EBV envelope


14
glycoprotein, gp85, does have characteristics of a fusion molecule. Monoclonal
antibodies to gp85 neutralize virus have provided indirect evidence suggesting that the
glycoprotein may play an active role in virus penetration through the cell membrane
(Miller and Hutt-Fletcher, 1988). The antibody, F-2-1, failed to inhibit binding of EBV
to its receptor, but interfered with virus fusion as measured with the self-quenching
fluorophore octadecyl rhodamine B chloride (R18). This fluorescent amphiphile,
diffuses readily into biologic membranes and exhibits self-quenching properties at high
concentrations. Such high concentrations can be reached in the viral envelope.
However, a measurable relief of self-quenching and fluorescence is detected when virus
and cell membranes fuse. This is due to R18 molecules moving within the fused
membranes, creating lower concentrations of the fluorescent molecule. The relief of
self-quenching was inhibited by the neutralizing monoclonal antibody F-2-1; however,
inhibition was not seen when a non-neutralizing antibody to the same molecule was used
(Miller and Hutt-Fletcher, 1988).
Further evidence to support the hypothesis that gp85 functions as a fusion
protein, is provided by experiments in which EBV virion proteins including or depleted
of gp85 were incorporated into lipid vesicles to form virosomes. R18 labeled virosomes
that were made with nondepleted protein were shown to behave in a manner similar to
that of R18-labeled virus in the experiments described above. The virosomes bound to
receptor positive, but not to receptor negative cells. Fusion was demonstrated with Raji
cells, but not with receptor positive, fusion incompetent Molt 4 cells. Furthermore,
monoclonal antibodies that inhibited either binding or fusion of virus, inhibited binding


15
and fusion of virosomes. Lastly, virus competed with virosomes for attachment to cells.
Those virosomes made from viral proteins depleted of gp85 remained capable of binding
to receptor positive cells, but failed to fuse.
Neutralizing sera obtained from patients in the acute phase of mononucleosis
demonstrate a predominant antibody response against the virally encoded gp85
(Qualtiere and Pearson, 1979). As well as mediating viral neutralization, monoclonal
antibodies to gp85 also demonstrate complement dependent cytolysis of EBV infected
cells (Strnad et al., 1982). Although gp85 is thought to be a minor component of the
virion (Douglas Oba, Thesis), it is an essential glycoprotein. It is an important target of
the host immune response directed against EBV. Glycoprotein gp85 maps to the BXLF2
open reading frame of EBV DNA (Heineman et al., 1988; Oba and Hutt-Fletcher, 1988).
Analysis of the BXLF2 sequences predicts that gp85 contains five N-linked
glycosylation sites, a potential amino terminal signal sequence and a carboxy terminal
anchor sequence (Figure 1-1). The sequence also includes a stretch of 16 extremely
apolar amino acids that could be a fusion domain (Oba and Hutt-Fletcher, 1988) are also
detected within BXLF2 sequences.
The gp85 glycoprotein has been identified as the EBV gH, which has
homologues in each of the three herpesviruses subfamilies (Gompels and Minson 1986;
McGeoch and Davison, 1986; Keller et al., 1987; Cranage et al., 1988; Gompels et al.,
1988; Pachl et al., 1989; Nicholson et al., 1990; Josephs et al., 1991; Klupp and
Mettenleiter, 1991; Meyer et al., 1991). In at least four of these herpesviruses, gH has
been shown to play a role in virus penetration (Gompels and Minson, 1986; Fuller et al.,


1989; Forrester et al, 1992; Peeters et al., 1992). The goal of the present work was to
analyze in greater depth, the structure and function of the EBV gH molecule called gp85.


17
1 20 40
MQLLCVFCLVLLWEVGAASLSEVKLHLDIEGHASHYTIPWTELMAKVPG
60 80
LSPEALWREAEVTEDLASMLNRYKLIYKTSGTLGIALAEPVDIPAVSEGS
100 120 140
MQVDASKVHPGVISGLNSPACMLSAPLEKQLFYYIGTMLPNTRPHSYVF
160 180
YQLRCHLSYVALLSINGDKFQYTGAMTSKFLMGTYKRVTEKGDEHVLSL
200 220 240
VFGKTKDLPDLRGPFSYPSLTSAQSGDYSLVIVTTFVHYANFHNYFVPNL
260 280
KDMPSRAVTMTAASYARYVLQKLVLLEMKGGCREPELDTETLTTMFEVS
300 320 340
VAFFKVGHAVGETGNGCVDLRWLAKSFFELTVLKDIIGICYGATVKGMQ
360 380
SYGLERLAAMLMATVKMEELGHLTTEKQEYALRLATVGYPKAGVYSGLI
400 420 440
ggatsvllsaynrhplfqplhtvmretlfigshvvlrelrlEIvttqgpn
460 480
LALYQLLSTALCSALEIGEVLRGLALGTESGLFSPCYLSLRFDLTRDKLLS
500
520
V-1
540
MAPQEATLDQAAVSNAVDGFLGRLSLEREDRDAWHLPAYKCVDRLDKV
560 580
lmiipliIBvtfiissdrevrgsalyeasttylssslflspvimnkcsqgava
600 620 640
geprqipkiqEJftrtqkscifcgfallsydekegletttyitsqevqnsil
660 680
SSNYFDFDNLHVHYLLLTTNGTVMEIAGLYEERAHVVLAIILYFIAFALGIF
700
LVHKIVMFFL
Figure 1-1. Predicted amino acid sequence of the BXLF2 open reading frame. Solid
underlines indicate hydrophobic regions predicted by the program of Klein and
colleagues to have a high probability of being membrane spanning regions; residues in
black boxes indicate potential N-glycosylation sites; region designated as V-1 indicates
residues synthesized.


CHAPTER 2
CLONING OF gp85
Introduction
The EBV glycoprotein gp85 is produced in small amounts by the virus. Oba
reported that less than five picograms of the glycoprotein could be purified from a
culture of approximately 109 EBV-producing cells (Oba, D. E. Ph.D. Thesis, University
of Florida, 1988). This low yield impeded further structural and functional analysis of
the molecule. To surmount the problem, it was decided to overexpress the protein in a
eukaryotic expression system that would allow production of glycosylated molecules.
All fusion proteins studied to date have been shown to be glycosylated and authentic
glycosylation of at least one of these, the envelope protein of the human
immunodeficiency virus, is known to be critical to its biologic activity (Gruters et al.,
1987; Matthews et al., 1987; Walker et al., 1987). The eukaryotic expression system
chosen makes use of vaccinia virus, a member of the orthopoxvirus family. Several
features of this virus make it an excellent vehicle for expression of foreign genes. First,
it has a very large double-stranded DNA genome of 187 kilobase pairs (Geshelin, P. and
K. I. Berns, 1974) packaged in a capsid which is capable of accommodating large
segments of additional nucleic acid. Second, the vaccinia genome includes many genes
that influence virulence in vivo, but which are not essential for growth in tissue culture
and which can readily be replaced by recombination events. Third, unlike many viruses,
18


19
vaccinia is capable of replication in a wide variety of cells from several species.
Recombinant vaccinia viruses have been used by many workers to express faithful copies
of glycoproteins from a variety of other RNA and DNA viruses.
Vaccinia virus replicates in the cytoplasm of a cell and makes use of its own
virally encoded DNA and RNA polymerases. Packaged within the virus core is a
complete transcriptional enzyme system that is essential for infectivity. This
transcriptional enzyme system is necessary since transcription can only be initiated by
the vaccinia RNA polymerase and not by eukaryotic RNA polymerase II. In order to
express heterologous DNA in a recombinant vaccinia virus it is first necessary to
construct a chimeric gene that contains a vaccinia promoter fused to the protein coding
sequences of the foreign gene. This is accomplished using a plasmid vector. The
engineered vector must contain the transcriptional start site of the vaccinia promoter
upstream of the translational initiation codon of the foreign gene. After assembly within
the plasmid vector, the expression cassettes are inserted into the vaccinia genome by
homologous recombination. This is accomplished in vivo by transfection of plasmid
DNA into vaccinia virus infected cells (Mackett et al., 1984). This step is necessary
because the large size of vaccinia DNA makes it impractical to construct recombinant
molecules in vitro.
The insertion of the BXLF2 gene into vaccinia virus was accomplished using two
insertion vectors, pSCl 1 and pEBl (Figure 2-1). The pSCl 1 vector contains two
vaccinia promoters, one of which is the early/late promoter, p7.5. This promoter is
upstream of a unique Smal restriction endonuclease cleavage site into which the foreign


20
gene to be expressed is cloned. In the opposite orientation, the late promoter pi 1
controls expression of a p-galactosidase gene. The pEBl insertion vector is identical in
sequence to pSCl 1, except that the p7.5 promoter has been replaced by a second pi 1
promoter.
To facilitate homologous recombination the Sma\ site and the p-glactosidase gene
are flanked by sequences from the vaccinia thymidine kinase (TK) gene. These
sequences direct the insertion of intervening DNA into the nonessential region of the
virus genome. The vectors also contain a bacterial origin of DNA replication and a gene
conferring ampicillin resistance which are important to amplification and selection of the
plasmids in bacteria, but which are lost upon recombination with vaccinia. Recombinant
TK viruses are amplified by growth in TK' osteosarcoma cells in the presence of
5-bromo-2-deoxyuridine (5-BUdR). Phosphorylation of this nucleoside analog results in
its lethal incorporation into the DNA of those cells that have been infected with a TK+
wild type vaccinia virus. The resulting TK amplified viral stocks are plaqued in
monolayers of African Green Monkey kidney cells and plaques derived from
recombinant virus which contains the P-galactosidase gene are identified by overlaying
with agarose containing the chromogenic substrate 5-bromo-4-chloro-beta-
galactopyranoside (X-gal). This chapter describes the construction and selection of
recombinant vaccinia viruses containing the EBV sequence corresponding to the BXLF2
open reading frame.


21
Materials and Methods
DNA sequencing
Cesium chloride (CsCl) purified DNA was sequenced with the Sequenase kit
(Sequenase kit, Product No. 70700, United States Biochemical Corporation), which
uses a modified form of T7 DNA polymerase and the principles of the DNA
chain-termination sequencing procedure described by Tabor and Richardson. The T3
promoter sequence (5ATTAACCCTCACTAAAG3) was the primer for the reaction and
the nucleotides used were radiolabeled with [35S]dATP. Briefly, the DNA was denatured
in 0.2 M NaOH for five minutes at 25 C, neutralized by adding 0.4 volumes of 5M
ammonium acetate (pH 7.5) and ethanol precipitated. Two micrograms of precipitated
DNA were added to 50 ng of primer and 2 pi of 5x sequencing buffer (200 mM
Tris-HCl, pH 7.5, 100 mM MgCl2 and 250 mM NaCl) and the volume was adjusted
with dH20 to 10 pi. The solution was vortexed, heated to 65 C for two minutes and
held at 25 C for 30 minutes to allow annealing of DNA. Enzyme was added in 6 mM
DTT, containing 20 pCi of [a-35S]dATP, and unlabeled nucleotides (dGTP, dCTP and
dTTP), incubated at 25 C for five minutes and then aliquotted into each of four
termination solutions containing all the deoxynucleotide triphosphates, a
dideoxynucleoside triphosphate and 50 mM NaCl for further incubation for five minutes
at 37 C. The reaction was stopped by addition of 95% formamide, 20 mM EDTA.
Bromophenol Blue 0.05% and 0.05% xylene cyanol were added and samples held on ice
until electrophoresis in 6% acrylamide cross-linked with bis-acrylamide.


22
Construction of shuttle vectors
Three plasmids were used in the construction of vaccinia virus transfer vectors.
These were pSCl 1 (Chakrabarti et al., 1985) pEBl (Perez et. al., 1992; Stephens et. al.,
1992) and Bluescript pBSKS (+) (Stratagene) into which BXLF2 cDNA had been cloned
(pBKSKS-BXLF2; a gift of Dr. Elliott Kieff). The first vaccinia virus transfer vector to
be made was pSCl l-BXLF2a. This recombinant vector was made by digesting one
microgram of CsCl purified pBSKS-BXLF2 with four units of the restriction
endonucleases Smal and Hindi at 37 C overnight. The Smal Hindi fragment was
recovered by electroelution into a dialysis bag from a 1% agarose gel containing
ethidium bromide. The fragment was concentrated by ethanol precipitation for 30
minutes with two volumes of ice cold ethanol and 250 mM sodium acetate (pH 5.2) at
-20 C followed centrifugation at 10,000 x g for ten minutes. The supernatant was
discarded, the pellet was dried by vacuum desiccation for five minutes and the DNA was
resuspended in 25 pi of sterile distilled water (dH20). DNA was quantitated by visual
inspection after electrophoresis in 1% agarose in TAE buffer (50X: 242 gm Tris base,
57.1 ml glacial acetic acid, 37.2 gm Naj EDTA-2H20, pH 8.5). Three hundred
nanograms of the Smal Hindi fragment and 200 ng of Smal digested pSCl 1 were
added to a ligation reaction containing six units of T4 DNA ligase and 10 pi of 2X ligase
buffer (100 mM Tris-HCl, pH 7.5, 20 mM MgCl2, and 20 mM DTT). The total volume
was adjusted to 20 pi with dH20. The ligation reaction was incubated for 24 hours at 16
C. Ten microliters of this reaction was used to transform 300 pi of Escherichia coli
JM109 cells made competent by treatment with CaCl2. Bacteria and DNA were


23
incubated for 30 minutes at 0 C and then heat-shocked for two minutes at 37 C. One
ml of Luria-Bertani (LB) medium was added to the transformed cells which were
incubated for one hour at 37 C and then used to inoculate LB plates containing 60 pg
per ml ampicillin. The cultures were plated at volumes of 25, 50, 100, 150 and 200 pi
and incubated overnight at 37 C. Colonies containing BXLF2 DNA was identified by
colony hybridization.
The second vaccinia virus transfer vector to be made was pEB 1-BXLF2. Once
again the pBSKS-BXLF2 cDNA was digested with the restriction endonucleases Smal
and Hindi. The Smal Hindi fragment was ligated into Smal digested pEBl vector by
the previously described method.
The third vaccinia virus transfer vector to be made was pSCl l-BXLF2b which
was obtained using polymerase chain reaction (PCR) techniques (Saiki et al 1985; Saiki
et al., 1988) with the pBSKS-BXLF2 cDNA being used as the template for the reaction.
The amplified DNA contained only the open reading frame of the BXLF2 gene plus six
base pairs corresponding to Xhol sites added to the 5' ends of the EBV DNA sequence
(Figure 2-1). The primer for the forward reaction was
5'-CTCGAGATGCAGTTGCTCTGTGTTTTTTGC-3', while the primer for the reverse
reaction was 5 '-CTCGAGAAGGAAAAACATAACAATCTTGTG-3'. The DNA was
amplified in a DNA Thermal Cycler (Coy Temp Cycler Model 50) by 10 minutes
denaturation at 92 C followed by 25 cycles of denaturation at 92 C for one minute,
annealing at 55 C for three minutes and extension at 72 C for three minutes. Following
the last cycle, the reaction was held at 4 C. Each PCR reaction consisted of 10 pi of lOx


24
buffer (100 mM Tris-HCl, pH 8.3, 500 mM KC1, 15 mM MgCl2 and 0.01% gelatin;
Perkin-Elmer Cetus), 16 pi of 200 pM dCTP, dATP, dGTP and dTTP, 10 pi of each
primer at a concentration of 20 uM, 2.5 units of Amplitaq DNA polymerase
(Perkin-Elmer Cetus) and 500 ng of template DNA. The total volume was adjusted to
100 pi with dH20 and overlayed with 100 pi of mineral oil. The product of the reaction
was ethanol precipitated and stored in dH20 at 4 C.
Blunt ending of DNA
Two hundred nanograms of amplified DNA was blunt ended using 20 units of
DNA polymerase Klenow fragment in buffer (50 mM Tris-HCl, pH 7.6 with 10 mM
MgCl2) containing 2 mM of each of the nucleotides dATP, dGTP, dCTP and dTTP. The
total reaction volume was 30 pi. The mixture was incubated at 25 C for 30 minutes.
The blunt ended DNA product was purified using spin tubes (Millipore Inc.), ethanol
precipitated and resuspended in 20 pi of dH20.
Colony blot hybridization
Transformed Escherichia coli strain JM109 colonies were plated onto LB plates
containing ampicillin (Maniatis et al., 1989) and incubated for 15 hours at 37 C. One
hundred of the resulting colonies were streaked onto master plates, reincubated for a
further 15 hours at 37 C and then transferred to nitrocellulose filters (Micron
Separations Inc.). Filters were placed sequentially, colony side up, for 5 minutes each on
3MM Whatman filter paper saturated with 10% sodium dodecyl sulfate (SDS), with 0.5
M NaOH and 1.5 M NaCl, with 1.5 M NaCl and 0.5 M Tris HC1 (pH 8.0.) and with TE
buffer (10 mM Tris HC1, 1.0 mM EDTA, pH 8.0). The nitrocellulose was dried for a


25
minimum of 30 minutes at room temperature and then baked under vacuum at 80 C for
two hours. The baked filters were prehybridized for 24 hours at 45 C in a solution
containing 50% formamide, 0.5% SDS, 6x SSC (20X SSC: 3M NaCl, 0.3 M Na3
citrate-2H20, pH 7.0), 0.4% Ficoll 400, 0.4% polyvinylpyrrolidone, 0.4% bovine serum
albumin and 100 pg per ml of denatured and sheared herring sperm DNA. The
prehybridization solution was discarded and replaced with hybridization solution
containing 50% formamide, 20% SDS, 4x SSC, 0.4% Ficoll 400, 0.4%
polyvinylpyrrolidone, 0.4% bovine serum albumin, 100 ug per ml of denatured and
sheared herring sperm DNA and 1 to 15 ng of 32P labeled probe (specific activity of 107
cpm/pg) per ml of hybridization reaction. Filters were hybridized for 24 hours at 42 C
and washed for 15 minutes at 25 C in 2x SSC and 0.1% SDS and for 30 minutes at 60
C in a 0.1% SSC and 0.5% SDS. This last wash was repeated until the radioactivity
associated with the filter and measured with a hand held monitor was unevenly
distributed.
Cell lines
Two African Green Monkey Kidney cell lines were used, CV-1 and Vero. A
human osteosarcoma cell line, TK-143, was used to obtain TK- recombinant vaccinia
virus. All three cell lines were grown in Dulbecco modified Eagle medium (DMEM;
Gibco) containing 5% heat inactivated fetal bovine serum (FBS) and 100 IU per ml
penicillin and 100 pg per ml of streptomycin.


26
Generation of recombinants
All three recombinant shuttle vectors were used to make recombinant vaccinia
virus (Mackett et al., 1984). One million CV-1 cells were washed once in DMEM and
then infected with 500 pi wild-type, thymidine kinase positive (TK+) vaccinia virus
strain IHD-J at a multiplicity of infection (MOI) of 0.1 to 0.05 plaque forming units
(PFU). The infected cells were incubated at 37 C for two hours. At two hours post
infection, the viral inoculum was removed and cells were washed twice with one ml of
2x Hepes buffered saline (0.14 M NaCl, 5 mM KC1, 1 mM Naj HP04,0.1% Dextrose,
20 mM Hepes) and incubated at room temperature for 30 minutes with 1 ml recombinant
plasmid DNA that had been purified over CsCl and precipitated with 2.5 M CaCl2. Two
and a half hours later, 2.5 ml DMEM supplemented with serum and antibiotics were
added and the cells were reincubated at 37 C for 48 hours. The cells were then
harvested by scraping into spent culture medium and pelleted by centrifugation at 500 x
g for ten minutes. The pelleted cells were resuspended in 1 ml of DMEM and sonicated
for 1 minute in a sonic bath. Serial dilutions of the disrupted cells were used to infect
106 TK- 143 osteosarcoma cells. After 1 hour the viral inoculum was removed and
replaced with 1 ml of DMEM containing 25 pg per ml of 5-bromo-2-deoxyuridine.
Once viral plaques were visible, a 2% agarose overlay containing 400 pg per ml of the
chromogenic substrate 5-bromo-4-B-D-galactopyranoside (X-gal) was added. Blue
plaques were aspirated using sterile pasteur pipettes. The individual plaque lifts were
placed into 1 ml of DMEM and were either sonicated for 1 minute or frozen and thawed
three times. The plaque purification procedure was repeated a minimum of five times.


27
Virus
Virus stocks were grown either in Vero or CV-1 cells. Confluent cells were
infected at an MOI of 0.01 to 0.03 for 1 hour at 37 C. The total volume of infection
media was 500 pi for one well of a six well plate (9.4 cm2) and 5 mis for a T-150 flask
(150 cm2). At one hour post infection, 2 mis of DMEM containing 2% FBS were added
to one well of a six well plate and 12 mis were added to a T-150 flask. Infected cells
were harvested at 48 hours post infection and pelleted at 666 x g for 10 minutes at 4 C.
Pelleted cells were resuspended in 1 ml of serum free DMEM per 3 x 106 cells. The
resuspended pellets were frozen and thawed three times and centrifuged at 4 C for 10
minutes at 500 x g. The supernatant was stored at -85 C and titrated. Serial dilutions of
the virus stock was used to infect Vero or CV-1 cells. After several days of growth, the
medium was removed and the cells were stained with neutral red (10 mg/ml in dH20).
The viral plaques appeared as areas of reduced staining of approximately 1 mm in
diameter. The viral titer was then calculated by counting the plaques within the wells
and multiplying by the dilution factor.
DNA dot blot analysis
One million CV-1 cells were infected with pSCl l-BXLF2a or a recombinant
vaccinia virus expressing the BILF2 open reading frame (Mackett et al, 1990). The cells
were infected at an MOI of 10 and at 36 hours post infection washed twice with DMEM
before harvesting with a rubber policeman. One fifth of the harvested cells were
deposited onto a nitrocellulose filter and dried at room temperature for 1 hour. The
nitrocellulose was wetted with 100 mM NaCl and 50 mM Tris-HCl (pH 7.5) and then


28
placed sequentially on 3MM Whatman filter paper saturated with 0.5 M NaOH, 1.5 M
NaCl with 1.5 M NaCl, 0.5 Tris-HCl (pH 8.0) and with 2x SSC. The filter was dried at
room temperature for 30 minutes and before baked at 80 C for 2 hours. The filter was
prehybridized and hybridized as described above.
An alternative approach (monolayer blot analysis) was used to test for the
presence of BXLF2 DNA in the pEBl-BXLF2 vector. Nitrocellulose filters were placed
directly onto vaccinia virus infected CV-1 cell monolayers that had been infected with a
MOI of 10 for 36 hours. The complete nitrocellulose filter was moistened by the
monolayer of infected cells. A piece of 3MM Whatman filter paper saturated with 2x
SSC was placed on top of the wetted nitrocellulose. Pressure was applied to the filter
paper for 3 minutes. The nitrocellulose filter was then removed from the plate, dried at
room temperature and treated as before.
Preparation of DNA from colonies
Small scale preparation of DNA (miniprep) was performed according to a
procedure described by Zhou and colleagues (1990). Briefly, 1.5 ml of an overnight
culture was centrifuged at 10,000 x g for 1 minute and the supernatant was gently
decanted. The pellet was vortexed vigorously until completely resuspended in residual
fluid. Three hundred microliters of TENS buffer (TE buffer containing 0.1 N NaOH and
0.5% SDS) were added, vortexed for 5 seconds and 150 pi of 3.0 M sodium acetate (pH
5.2) were added. The mixture was again vortexed for 5 seconds and centrifuged for 2
minutes at 10,000 x g. The supernatant was transferred to a tube containing 0.9 ml of
ethanol that had been precooled to -20 C, mixed and centrifuged for two minutes. The


29
pellet was washed twice with 1 ml of 70% ethanol, dried under vacuum for ten minutes
and resuspended in 50 pi of dH20. Five microliters of resuspended DNA was analyzed
for purity and quantity by electrophoresis through a 1% agarose gel.
Large scale preparation of DNA
DNA was prepared in large quantities (maxiprep) from bacteria according to
published methods (Sambrook et al., 1989). A 500 ml overnight culture was centrifuged
at 4,000 x g for 10 minutes. Pelleted bacteria were resuspended in 12 ml of plasmid lysis
buffer, frozen at -85 C for 20 minutes, thawed and transferred to a Pyrex Erlemeyer
flask. One milliliter of freshly prepared lysozyme (Sigma) at a final concentration of 30
mg per ml was added to the flask. The flask was gently swirled over an open flame until
the solution began to boil. It was then immediately immersed in 300 ml of boiling water
for 40 seconds. The lysed bacteria were transferred into a 45 ml Nalgene tube and
chilled on ice for five minutes. Three milliliters of TE was added to the chilled solution
which was mixed well and centrifuged at 25,000 RPM (Beckman SW41) for 30 minutes
at 4C. The supernatant was decanted into a fresh 45 ml Nalgene tube containing an
equal volume of isopropanol, frozen at -85 C for 15 minutes, thawed and centrifuged
9,000 RPM for 20 minutes at 4 C. The supernatant was discarded and the pellet was
dried by inverting the tube onto filter paper. The dried pellet was dissolved in 4.25 ml
TE, and CsCl was added to a final concentration of 1.0 gm per ml. Ethidium bromide
(600 pg per ml) was added, and the solution was placed into 2 polyallomer Quick-Seal
centrifuge tubes (Beckman). The samples were centrifuged overnight in a vertical rotor
(VTi 65.2) at 45,000 RPM at 20 C. Two bands of DNA were visible after


30
centrifugation. An 18 gauge needle was inserted into the top of the tube to release
pressure and the lower band of supercoiled plasmid DNA was collected with a 21 gauge
needle and five ml syringe. The plasmid DNA was extracted by mixing with equal
volumes of water saturated butanol and discarding the upper layer until ethidium
bromide (pink color) could no longer be detected. After the extraction, DNA was
purified from residual CsCl by dilution with two volumes of dHzO followed by the
addition of six volumes of ethanol. The mixture was chilled at -85C for 20 minutes and
precipitated DNA recovered by centrifugation at 9,000 RPM for 20 minutes at 4C. The
supernatant was gently decanted and 400 pi of TNE buffer was used to resuspend the
DNA in the base of the tube. The TNE and DNA were transferred to an eppendorf tube
to which 1 ml of ice-cold ethanol was added. The tube was chilled at -85 C for 20
minutes and centrifugated at 10,000 x g for 10 minutes. The pelleted DNA was
resuspended in 100 pi of TE buffer.
Results
Sequencing of the pBSKS-BXLF2 which contained the BXLF2 cDNA cloned
into the EcoR I site of pBSKS confirmed that the BXLF2 cDNA had been oriented for
transcription under control of the T7 promotor (Figure 2-2). The plasmid was digested
with Smal and Hindi, two restriction endonuclease sites which flank the Eco R1 site, and
electrophoresed in 1% agarose in parallel with Hindlll digested Lamba phage DNA. A
band corresponding to the predicted 2.3 kb digestion fragment was detected (Figure 2-3).
Similar analysis of pSCl 1 digested with Smal revealed the 7.7 kb fragment that would be
expected from linearized plasmid (Figure 2-4). The Smal Hindi BXLF2 fragment was


31
ligated with the Smal digested pSCl 1 vector and used to transform E. coli strain JM109.
Colony blot hybridization of transformed bacteria with [32P] labeled Sma\ Hindi
BXLF2 indicated that thirteen of the two hundred colonies contained BXLF2 DNA
(Figure 2-5). Minipreps of DNA were made from two of these positive colonies.
Orientation of the insert within the pSCl 1 vector was analyzed with BamHI. Figure 2-6
depicts the size of digested fragments which would be expected after digestion of DNA
cloned in either orientation. When BXLF2 is in the correct orientation for transcription
under control of the pSCl 1 p7.5 promoter, fragments of 5.95, 3.2, 0.45 and 0.4 are
expected; in the opposite orientation, fragments of 4.45, 3.2, 1.9 and 0.45 should be
produced. Fragments obtained from two positive clones digested with BamHI indicated
that one (clone 45) was in the correct orientation for expression under control of the p7.5
promotor, and one (clone 25) was in the incorrect orientation (Figure 2-7).
To confirm these results, the digested DNA was electrophoresed in an agarose
gel, transferred to nitrocellulose using the Southern technique and hybridized with the
[32P] labeled Smal Hie II BXLF2 fragment (Figure 2-8). As expected, the probe
hybridized to the 4.45 and 1.9 kb fragments of clone 25 which contained BXLF2 in the
incorrect orientation (lanes 1 and 4). It hybridized to the 5.95 and 0.4 kb fragments of
clone 45 which contained BXLF2 in the correct orientation (lanes 2 and 5). Microgram
quantities of plasmid DNA were grown and purified from clone 45. This purified DNA
was used to transfect CV-1 cells infected with wild type vaccinia virus. Recombinant
virus was amplified in TK' cells in the presence BUdR, and virus expressing
P-galactosidase was plaque purified repeatedly in CV-1 overlayed with X-Gal. Virus


32
that produced greater than 90% blue plaques was obtained after 5-6 rounds of plaque
purification and was amplified as VVpSCl l-BXLF2a. DNA dot blots of this stock
confirmed that it still contained the BXLF2 sequence (Figure 2-9). The second vaccinia
virus recombinant, VVpEBl-BXLF2 was generated by inserting the same BXLF2
fragment, by the same protocol as used for VVpSCl l-BXLF2a, into the pEB-1 vector
linearized with Sma I (Figure 2-10). A DNA monolayer blot analysis was performed to
show that the BXLF2 sequence was contained within the vaccinia virus genome (Figure
2-11).
The third vaccinia virus recombinant VVpSCII-BXLF2b was made with BXLF2
DNA that had been amplified by PCR technology (Figure 2-12). The PCR product
(Figure 2-13) was blunt-ended and ligated into Smal digested pSCl 1 and Sma I digested
pEBl. The ligated DNA was used to transform E. coli and colony blot hybridizations
were done on resulting ampicillin resistant colonies. The probe was generated by
radiolabeling BXLF2 PCR products with [32P], DNA from two out of 100 pSC 11 clones
hybridized with labeled probe. None of the DNA from 100 pEBl clones hybridized with
the radiolabeled probe. DNA from both positive pSCl 1 clones was amplified in E. coli
and purified on CsCl gradients. The recombinant plasmids were tested for the
orientation of BXLF2 DNA within pSCl 1 using BamHI. The clone containing BXLF2
in the correct orientation was used to transfect CV-1 cells infected with wild type
vaccinia virus. Recombinant viruses were plaque purified and amplified.


33
UNIQUE
Smal
ATCTCGAGGATCCCCGGG
p11 p 11
FIGURE 2-1. Schematic diagrams of the vaccinia virus recombination vectors used in
the studies. pEBl contains pi 1 (vaccinia virus late promoter), Lac Z (beta galactosidase
from E.coli.), TK (thymidine kinase gene from vaccinia virus), and Ampr (ampicillin
resistance gene). The pSCl 1 vector contains both a pi 1 and p7.5 (vaccinia virus
early/late promoter region) as well as the Lac Z, TK, and Ampr genes.


34
FIGURE 2-2. Diagram of the pBSKS-BXLF2 plasmid with restriction sites contained
within the plasmid Multiple Cloning Site (MCS) and BXLF2 sequencing information
obtained.


35
1 2
23.1 kb-
9.42 kb-
6.56 kb-
4.37 kb-
2.32 kb-
2.03 kb-
FIGURE 2-3. 1% agarose gel containing ethidium bromide. Lane 1 contains the DNA
standard, Hindlll Lambda digest with resulting fragments of 23.1, 9.42, 6.56, 4.37,
2.32 and 2.03 kb. Lane 2 contains the electroeluted Smal Hie II digestion fragment of
pBSKS-BXLF2.


36
1 2
23.1 kb-
9.42 kb-
6.56 kb-
4.37 kb-
2.32 kb-
2.03 kb-
FIGURE 2-4. 1% agarose gel containing ethidium bromide. Lambda DNA standards
are in lane 1 and Sma\ digested pSCl 1 vector is in lane 2.


37
FIGURE 2-5. Colony blot of E. coli that was transformed with Sma I digested pSCl 1
and Sma I Hie II digested pBSKS-BXLF2. The colonies were hybridized with 32P
labeled Sma I Hie II digestion fragments of pBSKS-BXLF2.


38
BamHI
BamHI
FIGURE 2-6. Diagram of BamHI digested pSCl 1-BXLF2. Digestion fragments of 5.95,
3.2, 0.45 and 0.4 kB are seen when the BXLF2 sequence is ligated in the correct
orientation (A). Digestion fragments of 4.45, 3.2, 1.9 and 0.45 kB are seen when the
BXLF2 sequence is ligated in the incorrect orientation (B).


39
1 2 3
9.42 kb-
6.56 kb-
4.37 kb-
2.32 kb-
2.03 kb-
0.5 kb-
FIGURE 2-7. 1% agarose gel containing ethidium bromide. DNA BamHI digestion
patterns of pSCl 1 plasmid alone (lane 3), and of two clones positive by colony blot
hybridization. Lanes 1 (clone 25) and 2 (clone 45) contain the digestion patterns of
DNA from two positive clones.


40
23.1 kb-
9.42 kb-
6.56 kb-
4.37 kb-
2.32 kb-
2.03 kb-
0.5 kb-
FIGURE 2-8. Southern blot of BamHl digested pSCl 1 (lanes 3 and 6), pSCl 1-BXLF2
clone #25 (lanes 1 and 4) and pSCl 1-BXLF2 clone #45 (lanes 2 and 5). The blot was
hybridized with 32P labeled Sma I Hie II digestion fragments of pBSKS-BXLF2 DNA.


41
Sample # 1
#
Sample #2
FIGURE 2-9. Dot blot in which sample #1 contained DNA from uninfected CV-1 cells,
sample #2 contained DNA from CV-1 cells infected with a vaccinia virus recombinant
expressing the EBV membrane glycoprotein, gp55/75 and sample #3 contained DNA
from CV-1 cells infected with VVpSCl l-BXLF2a. The blot was hybridized with [32P]
labeled Sma I Hie II digestion fragments of pBSKS-BXLF2.


42
1 2
kb
kb
kb
kb
kb
kb
FIGURE 2-10. 1% agarose gel containing ethidium bromide. Lane 1 contains Sma I
linearized pEBl vector. Lane 2 contains Hindlll digested Lambda DNA standard.


43
FIGURE 2-11. DNA blot of CV-1 monolayers. Sample #1 contained DNA lifts from
approximately 1 x 106 uninfected CV-1 cells. Sample #2 contained DNA lifts from
CV-1 cells infected with VVpEBl-BXLF2. The blot was hybridized with 32P labeled
Sma I Hie II digestion fragments of pBSKS-BXLF2.


44
Primer B
Primer A: 5 CTCG AG AT GC AGTT GCT CT GT GTTTTTT GC 3
Primer B: 3 CTCGAGAAGGAAAAACATAACAATCTTGTG 3
FIGURE 2-12. Diagram of DNA sequences achieved by using PCR. These sequences
were subsequently used to make the third vaccinia virus shuttle vector.


45
1
2
-23.1
kb
-9.42
kb
-6.56
kb
-4.37
kb
-2.32
kb
-2.03
kb
FIGURE 2-13. 1% agarose gel containing ethidium bromide. Lane 1 contains 10 pi of
PCR product with BXLF2 DNA. Lane 2 contains Hindlll digested Lambda DNA.


46
Discussion
The pSCl 1 insertion vector was the first vector available to our laboratory, thus,
the first vaccinia virus recombinant (VVpSCl l-BXLF2a) was made using pSCl 1. The
pEBl insertion vector became available shortly after the generation of this vaccinia virus
recombinant. The substitution of the late pi 1 promoter for the early/late p7.5 promoter
was described as expressing greater amounts of the inserted foreign gene (personal
communication, Dr. Edward Stephens). However, preliminary analyses of proteins
expressed by VVpSCl l-BXLF2a and VVpEBl-BXLF2 did not reveal any glycoprotein
with the characteristics of gp85. It was hypothesized that the presence of 226 base pairs
(bp) upstream of the BXLF2 start site that had been cloned from the cDNA into the
vaccinia virus genomes were interfering with translation. To allay these concerns, we
utilized PCR technology to generate a third recombinant vaccinia virus clone
(VVpSCl l-BXLF2b). Primers were designed such that these upstream 226 bp were
deleted.
Restriction endonuclease analysis of recombinant shuttle vector DNA indicated
that the BXLF2 sequence was of the correct size and orientation. Furthermore, data
from dot blots and monoblots revealed that the BXLF2 sequence was contained within
the vaccinia virus genome. At this point in our studies we decided that northern blot
analysis of the vaccinia virus mRNAs would not yield useful information because of
reports in the literature describing the mRNA's transcribed from late genes as being both
long and heterogeneous in size. For this reason, we decided to continue our analysis of


47
recombinant gp85 at the protein level. The vaccinia virus expression system has been
used to make properly processed and transported proteins that contain the appropriate
posttranslational modifications (Moss and Earl, 1989). The phosphorylation, N- and
O-glycosylation, myristylation, cleavage and assembly of proteins made using the
vaccinia virus expression system has been reported to occur in an apparently faithful
manner.


CHAPTER 3
EXPRESSION AND BIOCHEMICAL CHARACTERIZATION OF
RECOMBINANT gp85
Introduction
The BXLF2 open reading frame is predicted to encode for a polypeptide with a
Mr of 78,000 ( Baer et al., 1984). As previously described, computer analysis of the
BXLF2 sequence revealed three highly hydrophobic regions (Klein et al., 1985). One of
these regions is located at the amino terminus from amino acids 3-19. This region may
represent a signal sequence of the protein. A second hydrophobic region is at the
carboxy terminus, and spans from amino acids 680-696. This region may serve as a
membrane anchor. The third hydrophobic domain stretches from amino acids 538-554
and may either be a second anchor sequence or possibly a fusion peptide similar to that
found in fusion proteins of ortho and paramyxoviruses (White et al., 1983). Analysis of
the gp85 open reading frame has also revealed five potential N-linked glycosylation sites
(Baeretal., 1984)
The addition of asparagine or N-linked sugars to cell and virus glycoproteins
begins as the precursor oligosaccharide Glc3Man9GlcNAc2 and is transferred to the
nascent polypeptide chain of the glycoprotein from the carrier lipid, dolichol phosphate.
The transfer occurs on the luminal side of the endoplasmic reticulum (Komfeld and
Kornfeld, 1985). This first step is followed by a number of processing and trimming
48


49
reactions which begin in the endoplasmic reticulum and continue as the protein is
transported through the Golgi apparatus to its final destination (Figure 3-1). Soon after
transfer of the sugar from dolichol phosphate, three glucose residues may be removed by
the activity of two membrane bound glucosidases, a-glucosidase I and a-glucosidase II.
Glucosidase I removes the terminal alpha 1,2 glucose and glucosidase II removes the
next two alpha 1,3 glucose residues to produce an oligosaccharide chain with the
structure Man9GlcNAc2. This Man9GlcNAc2 molecule is referred to as the "core"
region. Deoxynojirimycin (DNM) inhibits the activity of glucosidase I, thus preventing
the removal of all three terminal glucose residues. If formed, Man9GlcNAc2 can be
processed by two distinct pathways. Failure to remove mannose residues can result in
production of the high mannose oligosaccharide Man8GlcNAc2 or Man5GlcNAc2. In
contrast, the removal of four mannose residues by mannosidase I, an enzyme found in
the Golgi, paves the way for the addition of complex N-acetylglucosamine residues to
Man5GlcNAc2. Furthermore, the molecule is now sensitive to removal of two more
mannose residues with mannosidase II. The action of mannosidase II is inhibited by
swainsonine. Finally, addition of peripheral sugar residues such as N-acetylglucosamine,
galactose, N-acetylneuraminic acid, fucose and sialic acid occurs by specific
glycosyltransferases. The addition of N-acetylglucosamine to Man5GlcNAc2 renders the
glycoprotein resistant to removal of sugar side chains by endoglycosidase H (endo H),
although they remain sensitive to the action of endoglycosidase F. Since this step occurs
early in processing in the Golgi the relative sensitivity of the oligosaccharides on a


50
glycoprotein to digestion with endo H and endo F is commonly used to monitor
glycoprotein transport.
The transport of the amino-linked oligosaccharide from its site of synthesis in the
endoplasmic reticulum through the cis, medial and trans Golgi stacks can be inhibited by
ionophores. Concentration of ionophores that inhibit transport and hence terminal
glycosylation reactions do not affect the initial incorporation of amino acids into protein.
The ionophore monensin is thought to deplete Ca2+ levels, rendering the Golgi vesicles
unable to fuse. Thus, vesicular transport of molecules between the endoplasmic
reticulum and the Golgi is perturbed (Uchida et al., 1979). Newly synthesized proteins
accumulate within intracellular vacuoles and post-translational modifications are
blocked.
This chapter describes the expression and biosynthesis of the EBV recombinant
gp85 molecule.
Materials and Methods
Cells and virus
CV-1 cells were grown in DMEM supplemented with 5% FBS, 100 IU of
penicillin and 100 pg of streptomycin per ml. Stocks of VVpSCl l-BXLF2a,
VVpEBl-BXLF2 and VVpSCl l-BXLF2b were prepared as described in Chapter 2.
Antibodies
Two monoclonal antibodies and one rabbit anti-peptide antibody were used. The
monoclonal F-2-1 (Strnad et al., 1982) and monoclonal E1D1 (Balachandran, 1987) both


51
recognize native gp85. The anti-peptide antibody (anti-BX), was made to a synthetic
peptide corresponding to residues 518 to 528 of the EB V BXLF2 ORF (Oba and
Hutt-Fletcher, 1988). This sequence is located amino terminal to the internal
hydrophobic region. The peptide, cys-ser-leu-glu-arg-glu-asp-arg-asp-ala-
trp-his-leu-pro-ala-tyr-lys, (V-l) was synthesized by the Protein Chemistry Core Facility
of the University of Florida. The amino terminal cysteine residue was added to the EBV
sequence to facilitate coupling to keyhole limpet hemocyanin (Liu et al., 1979).
Rabbits were immunized initially with a subcutaneous injection containing one
mg of peptide conjugated to keyhole limpet hemocyanin and emulsified in Freunds
complete adjuvant. The rabbits were subsequently immunized at biweekly intervals,
with the same amounts of conjugated peptide emulsified in Freunds incomplete adjuvant.
After approximately two biweekly injections, the rabbits were bled from the central ear
artery. Prior to bleeding, the rabbits had been anesthetized for approximately 30 minutes
with 0.1 ml per kg of Innovarvet (fentanyl 0.04% and droperidol 2%). The rabbit was
then routinely immunized and bled between every two injections.
Antibody purification
All antibodies were purified by chromatography on protein A that had been
coupled to Sepharose CL-4B (Sigma). Approximately 100 mis of concentrated
hybridoma culture supernatant or approximately 25 mis of serum were filtered through a
0.45 urn filter and added to a bed of protein-A sepharose. The 20 ml bed volume was
washed with approximately 80 mis of phosphate buffered saline (PBS) at 4 C. The
antibody was adsorbed to the column and subsequently eluted with 0.1 M acetic acid in


52
PBS. The pH of the eluate was immediately equilibrated with 1.5 M Tris-HCl (pH 8.8)
and the purified antibody was dialyzed against 16 liters of PBS at 4C. Dialyzed
antibody was concentrated by dialysis against polyethylene glycol (Sigma) to one tenth
of the volume of the serum or culture supernatant from which it was purified.
Anti-BX was further purified by affinity chromatography on peptide that had
been coupled to cyanogen-bromide activated Sepharose CL-4B (Oba and Hutt-Fletcher,
1988). BXLF2 peptide (7 mg/ml in PBS) was added to cyanogen bromide activated
Sepharose (CNBr-Sepharose; Sigma, St. Louis, MO) in 0.2 M carbonate/bicarbonate
buffer (pH 9.6) and mixed for a minimum of four hours at 4C. Coupled gel was then
pelleted by centrifugation and the supernatant was stored for analysis of both protein
concentration (Lowry et al., 1951) and coupling efficiency. The residual active groups
on the Sepharose were blocked with ethanolamine (1M, pH 8.0) before washing beads
extensively with PBS. Subsequent peptide columns were made using Affi-Gel-15
(Bio-Rad Laboratories) instead of CNBr-Sepharose. The use of Affigel 15 dispenses
with cyanogen bromide activation steps. Affigel spontaneously forms stable covalent
bonds with the primary amines of proteins that have isoelectric points below 6.5.
Columns were made with bed volumes of approximately five mis.
Enzyme-linked immunosorbant assay
Reactivity of rabbit antisera with the V-l peptide was measured in ELISA assays
(Oba, D. E. Ph.D. Thesis, University of Florida, 1988). Fifty microliters of peptide at a
concentration of 6 jug/ml in carbonate bicarbonate buffer (pH 9.6) were added to each
well of a 96-well microtiter plate and incubated overnight at 37 C (Voller et al., 1976).


53
Plates were subsequently incubated with 5% skimmed milk in PBS-Tween 20 to block
binding surfaces that remained on the polystyrene plate after antigen absorption.
Blocked plates were then sequentially incubated with dilutions of rabbit antibody, with
goat anti-rabbit antibody conjugated to horseradish peroxidase (Organon Teknika Corp.,
Durham, N.C.) and with the chromogenic substrate orthophenylenediamine (Sigma).
The bound enzyme converts the substrate to a colored product at rates proportional to the
amount of primary antibody present in each well. The final enzyme reaction was stopped
with 2 N H2S04. Color change was analyzed at 492 nm using a spectrophotometric
ELISA plate reader. Plates were washed between each incubation by flooding the wells
three times with 0.05% Tween 20 in 0.85% NaCl. The Tween-20 solution reduces
nonspecific interactions by removing unbound or weakly adsorbed molecules.
Analysis of vaccinia proteins
Six well plates containing 106 CV-1 per well were infected with wild type
vaccinia virus, with VVpSCl l-BXLF2b or with VVpEBl-BXLF2 at an MOI of 10 at
37C. At one hour post infection, the viral inoculum was removed and replaced with 1
ml of methionine deficient medium. Two hours later cells were labeled for 20 hours with
100 jaCi [35S] methionine. Supernatant medium was replaced with
radioimmunoprecipitation buffer (RIPA; 0.05 M Tris-HCl pH 7.2, 0.15 M NaCl, 1.0%
deoxycholate, 1.0% Triton X-100, 0.1% sodium dodecyl sulfate, 100 U aprotinin per ml
and 0.1 mM phenylmethylsulfonylfluoride) and cells were held on ice for thirty minutes.
At thirty minutes, the solubilized cells were transferred to eppendorf tubes and vortexed
vigorously for five minutes. The vortexed samples were centrifuged at 10,000 x g for


54
fifteen minutes and the supernatant was immunoprecipitated for 20 hours at 4C with
100 pg of antibody and protein A-agarose beads. Immunoprecipitated complexes were
washed 3 times with RIPA. The proteins were dissociated by boiling for 10 minutes in
sample buffer (0.625 M Tris (pH 8.8), 1% SDS, 10% glycerol, and 1%
2-mercaptoethanol) and analyzed by gel electrophoresis after a five pi aliquot of each
sample was counted to determine the amount of radiolabeled protein immuno
precipitated.
Electrophoresis of proteins
Immunoprecipitated samples were analyzed by SDS-polyacrylamide gel
electrophoresis (SDS-PAGE; Laemmli, 1970). Polyacrylamide gel separating buffer was
made with 0.65 M Tris and 7.5, 9.0 or 12.5% acrylamide, 0.1% SDS, 0.1% glycerol,
0.83% ammonium persulfate, and 0.28% N,N'diallyltartardiamide (DATD) or
bis-acrylamide (BIS). The separating gel was overlayed with 1-butanol and allowed to
polymerize for a minimum of five hours. After polymerization, the gel was rinsed with
distilled water and a stacking gel was added. The stacking gel was made with 0.65 M
Tris (pH 6.8), 4% acrylamide, 0.1% SDS, 0.05% glycerol, 0.83% ammonium persulfate,
and 0.28% DATD and allowed to polymerize for a minimum of 45 minutes before
samples were loaded. The gel was electrophoresed for approximately six hours at a
constant current of 26 mAmps or overnight at a constant voltage of 50 volts. Gels were
stained with 0.1% Coomassie blue stain for ten minutes and destained with 7% methanol
and 7% acetic acid. Gels containing proteins labeled with l2:>I were directly dried on
filter paper. Gels containing [35S] or [3H] labeled proteins were either infused for 30


55
minutes in two changes of dimethyl sulfoxide (DMSO), for 3 hours in 22%
2,5-diphenyloxazole (PPO) in DMSO and for 1 hour with water or they were infused
with Amplify (Amersham) for 30 minutes. All dried gels were exposed to XAR film at
-85C for fluorography (Bonner and Laskey, 1974).
Enzyme and glvcosvlation inhibitors
For both digestion with endoglycosidase H (endo H; Boehringer Mannheim
Corp) and for digestion with endoglycosidase F (endo F; Boehringer Mannheim Corp.)
immunoprecipitated protein-antibody-bead complexes were washed twice with 1 ml of
RIPA. For endo H digestion, the pelleted complexes were resuspended in 0.01 M sodium
citrate (pH 5.5), and 0.01 U of enzyme in a total volume of 100 pi for 20 hours at 37C.
The enzyme reaction was stopped with the addition of 2x SDS-PAGE sample buffer
containing 2-mercaptoethanol. For digestion with endo F, samples were washed using
the same procedure as used for endo H digestion and resuspended in 0.1 M sodium
acetate (pH 5.0) containing 50 mM EDTA, 1% NP-40, 0.1% SDS, 1%
2-mercaptoethanol and 2 U of endo F for 20 hours at 37C. The reaction was stopped
after 20 hours with the addition of 100 pi of 2x SDS-PAGE sample buffer without
2-mercaptoethanol. Both endo H and endo F digested samples were boiled for ten
minutes and analyzed by SDS-PAGE gel under reducing conditions. Monensin (Sigma
Chemical Co.) and stocks prepared in methanol. Swainsonine and 1-deoxynojirimycin
(DNM) were obtained from Boehringer-Mannheim Biochemicals. The swainsonine and
DNM were freshly prepared in DMEM without methionine and filter sterilized just prior
to their use. Swainsonine was used at 300 ng per ml, monensin was used at


56
concentrations of 0.001 to 100 pM and DNM was used at a 1 mM concentration.
Inhibitors were added to the initial infection medium. At 3 hours post infection, the viral
inoculum containing inhibitors was removed. New inhibitors were added along with 100
pCi of [35S] methionine. The radiolabeled, infected cells were incubated for 20 hours at
37C and harvested and immunoprecipitated as previously described.
Silver staining
Proteins were visualized in gels by silver staining according to a modification of
standard protocols (Heideshoven and Dernick, 1985; Merril et al., 1979). Briefly, gels
were fixed with 30% ethanol and 10% acetic acid for one hour followed by two ten
minute washes in 10% ethanol. The gels were then reduced in 0.5% Farmer's solution
(2.27 mM sodium thiosulfate, and 60.7 mM potassium ferricyanide) for approximately
15 seconds or until the gel appeared slightly yellow. The gel was then immediately
washed four times with distilled water for ten minutes, soaked for 15 minutes in 0.1%
silver nitrate and washed in running distilled water for 20 seconds. The silver
impregnated gel was developed with 2.5% sodium bicarbonate, and 0.02% formaldehyde
until the background of the gel began to change color. The reaction was stopped by
washing in distilled water.
Results
Expression of gp85 by recombinant vaccinia virus was first examined in CV-1
cells (Figure 3-2). The anti-peptide antibody, anti-BX, was used to immunoprecipitate
proteins from mock infected (lane 1), wild type vaccinia virus infected (lane 3),


57
VVpSCl l-BXLF2b infected (lane 2) or VVpEBl-BXLF2 (lane 4) infected. Bands with
an Mr of 85,000 were immunoprecipitated from cells infected with either construct, but
not from uninfected cells or cells infected with wild type vaccinia virus. To determine
the quantity of recombinant protein expressed, cell lysates immunoprecipitated with
anti-BX were electrophoresed in 7.5 % acrylamide cross-linked with BIS-acrylamide and
silver stained (Figure 3-3). Comparison of the intensity of staining of gp85 and known
amounts of molecular weight markers run in parallel indicated that 1.6 x 105 recombinant
infected cells expressed over 500 ng of the recombinant protein. However, further
analysis revealed that although recombinant gp85 made by VVpEBl-BXLF2 or
VVpSCl l-BXLF2b was immunoprecipitated by anti-peptide antibody anti-BX, it was
not immunoprecipitated by either of the monoclonal antibodies F-2-1 or E1D1 that
recognize native gp85 made in B lymphocytes (Figure 3-4).
To examine whether the recombinant protein was properly processed within the
endoplasmic reticulum, CV-1 cells were infected with either wild type vaccinia virus or
VVpEBl-BXLF2 in the presence of the amino linked glycosylation inhibitor, DNM.
Treatment with DNM increased the apparent Mr by 2,500. (Figure 3-5, lane 5). This
result is consistent with the inhibition of the action of trimming enzymes glycosidase I
and II within the endoplasmic reticulum. The addition of complex sugars which are
resistant to digestion with endo H, but not endo F, occurs only if glycoproteins are
transported out of the endoplasmic reticulum into the Golgi apparatus. To determine if
recombinant gp85 was being properly transported through the Golgi apparatus, a
comparison was made of the sensitivity of the molecules to digestion by the enzymes


58
endo H and F. Both endo H and F reduced the mass of the recombinant from 85,000 to
72,000 (Figure 3-6). Thus, there appeared to be a block in transport from the
endoplasmic reticulum through the Golgi. To test this hypothesis further, the processing
of recombinant gp85 was analyzed in cells treated with swainsonine and monensin,
which would be expected to inhibit processing steps within the Golgi (Figure 3-5). The
mobility of the recombinant molecule was not altered by either of these drugs.


Glycoprotein Processing
Nascent Glycoprotein
Cleaving Enzyme
Protein-Asn-A-A
Glucosidase 1
Protein-Asn-A-A
f
Protein-Asn-A-A
Glucosidase II
Glucosidase II
High Mannose Type Protein-Asn-A-A
Protein-Asn-A-A
-UDP-GlcNAc
Protein-Asn-A-A
Mannosidase I
(a-1,2-Gnkages)
Mannosidase II
Protein-Asn-A-A
-UOP-GicNAc
UDP-Gal
CMP-SA
AGal SA
Complex Type Pro;ein-Asn-.i-A
X A Gol SA
Asn
SA
Gal
Inhibitor
of Cleavage
1 -Deoxynojirinrvfdn
1 -Oeoxynojirimycin
1 -Oeoxynojirimycin
Swainsonine
N-acetylglucosamine
Glucose
Mannose
Asparagine
Sialic Ac:d
Galactose
FIGURE 3-1. Diagram of where inhibitors of trimming enzymes are active


60
FIGURE 3-2. SDS-PAGE analysis of recombinant EBV glycoprotein gp85. CV-1 cells
were mock infected (lane 1) or infected at an MOI of 10 with wild type vaccinia virus
strain IHD-J (lane 3), vaccinia virus construct VVpSCl l-BXLF2b (lane 2) or vaccinia
virus construct VVpEBl-BXLF2 (lane 4). Infected cells were radiolabeled with 75 pCi
of [35S]methionine at three hours post infection and harvested at 24 hours post infection.
Radiolabeled samples were immunoprecipitated with anti-BX and analyzed by
SDS-PAGE under reducing conditions. Results were visualized by standard
fluorographic techniques.


61
1 2 3
Figure 3-3. Silver stain of recombinant gp85. Molecular weight markers were run in
lane 1. CV-1 cells were infected with VVpEBl-BXLF2 (lane 2) or with wild type
vaccinia virus (lane 3). Cell lysates were immunoprecipitated with anti-BX and analyzed
on a 9% SDS-PAGE gel under reducing conditions.


62
97.5 Kd-
69.0 Kd-
45.0 Kd-
30.0 Kd-
FIGURE 3-4. SDS-PAGE analysis of recombinant gp85. CV-1 cells were infected with
the vaccinia construct VVpEBl-BXLF2 in lanes 1, 3 and 5, while CV-1 cells in lanes 2,
4 and 6 were infected with wild type vaccinia virus. Infected cells were radiolabeled
with [35 S] methionine and immunoprecipitated with anti-BX (lanes 1 and 2), with
monoclonal F-2-1 (lanes 3 and 4) or with monoclonal E1D1 (lanes 5 and 6). Samples
were analyzed under reducing conditions by SDS-PAGE and visualized by standard
fluorographic techniques.


63
FIGURE 3-5. Synthesis of recombinant gp85 in the presence of swainsonine, monensin
or DNM. CV-1 cells were infected with either VVpEBl-BXLF2 vaccina virus construct
(lanes 1 through 6), or wild type vaccinia virus strain IHDJ (lane 7). The infected cells
were radiolabeled with 75 pCi of [35S] methionine for 20 hours in methionine deficient
medium and 0.001 pM monensin (lane 1), 0.1 pM monensin (lane 2), 100 pM monensin
(lane 3), 300 ng per ml of swainsonine (lane 4), 1 mM DNM (lane 5) or left untreated
(lanes 6 and 7). The radiolabeled cell lysates were immunoprecipitated with anti-BX
antibody. Samples were analyzed under reducing conditions by SDS-PAGE and
visualized by standard fluorographic techniques.


64
FIGURE 3-6. Effects of endoglycosidases on recombinant gp85. CV-1 cells were
infected with either wild type vaccinia virus strain IHD-J (lanes 1 through 4), the
VVpSCl l-BXLF2b construct (lanes 5 through 8), or the VVpEBl-BXLF2 construct
(lanes 9 through 12). The infected cells were radiolabeled with 75 pCi of [35S]
methionine for 20 hours. The radiolabeled cell lysates were immunoprecipitated with
anti-BX and treated with endo F (lanes 3, 7 and 11), with endo F buffers (lanes 4, 8, and
12) with endo H (lanes 2, 6 and 10) or with endo H buffers alone (lanes 1, 5, and 9). All
samples were analyzed under reducing conditions by SDS-PAGE and visualized by
standard fluorographic techniques.


65
Discussion
We initially chose to express the BXLF2 gene in vaccinia virus both because we
hoped that this vector might express larger amounts of protein and because we would be
able to use such a vector to express protein in lymphoblastoid cell lines. The vaccinia
constructs that we made fulfilled our expectations as to quantities of protein expressed.
Furthermore, the protein had properties consistent with its predicted structure as the
apparent Mr of the molecule was compatible with the use of all five N-linked
glycosylation sites. However, we were able to detect recombinant protein only with the
antibody made to a peptide derived from the BXLF2 sequence (anti-BX) and not with
either of two monoclonal antibodies made to native protein. This suggested that the
recombinant was antigenically different than the native molecule. We decided to
examine whether this difference was caused by differential glycoprotein processing of
the two molecules.
The enzyme inhibitor, DNM, inhibits the removal of three terminal glucose
residues from the oligosaccharide side chains transferred contranslationally from the
carrier dolichol to the nascent peptide chain. The increased mobility of recombinant
gp85 immunoprecipitated from cells treated with DNM suggested that this first step in
glycoprotein processing, which occurs in the endoplasmic reticulum, was occurring
normally. The recombinant protein was, however, equally sensitive to digestion with the
endoglyclosidases H and F both of which reduced the apparent Mr of the molecule to
72,000, a mobility consistent with that predicted for the non-glycosylated core protein
(Edson and Thorley-Lawson, 1983). This suggests that addition of N-acetylglucosamine


66
to the Man5GlcNAc2 had not occurred in the Golgi. The results of swainsonine and
monensin treatment also suggest that the recombinant gp85 undergoes none of the
processing reactions known to take place in the Golgi. The data are consistent with the
hypothesis that recombinant gp85 molecule is blocked in transport from the endoplasmic
reticulum through the Golgi stacks.
Further examination of this hypothesis required more precise comparison of the
processing of recombinant and native gp85.


CHAPTER 4
COMPARISON OF NATIVE AND RECOMBINANT gp85
Introduction
The native gp85 molecule is partially sensitive to digestion with endoglycosidase
H suggesting that it carries both complex and high-mannose type oligosaccharides
(Edson and Thorley-Lawson, 1983). Endoglycosidase H digestion of gp85 from virus
producing cells revealed five species ranging in apparent Mr from 86,000 to 71,000,
while a protein with an apparent Mr of 69,000 was produced in cells treated with
tunicamycin (Stmad et al, 1983), an inhibitor which interferes with the transfer of
oligosaccharides to dolichol phosphate (Heifetz et al., 1979). This is consistent with the
use of all five potential N-linked glycosylation sites in the gp85 sequence and suggests
that the molecule carries no O-linked sugars.
Since EBV preferentially establishes latent infections as opposed to productive
infections in lymphocytes, virus yields are often low. Approximately 5% of cells in EBV
containing lymphoblastoid cell lines will spontaneously produce virus. Treatment of
cells carrying the P3HR1 or the B95-8 strains of virus with the tumor promoter
12-O-tetradecanoyl phorbol-13-acetate (TPA) (Zur Hausen et al., 1978), increases the
number of virus producing cells over a period of several days. However, a third strain of
virus, Akata, (Takada and Ono, 1984) has been derived from an EBV genome-positive
67


68
Burkitt lymphoma cell line in which virus replication can be induced by addition of
anti-human immunoglobulin G. Induction of virus replication by this technique is much
faster and occurs in a larger percentage of cells. A detectable increase is seen in the
number of cells positive for the EBV specific early antigen (EA) (Henle et al, 1970) by
three hours after addition of anti-immunoglobulin. By six hours of induction many of
the cells test positive for EA using indirect immunofluorescence assays. Takada and
Ono (1989) demonstrated that at nine hours post anti-immunoglobulin treatment, 40% of
Akata cells contain viral capsid antigen (VCA) (Henle and Henle, 1966). The percentage
of VCA positive cells increase until a plateau of 60% is reached at 12 hours. In contrast,
after 24 hours of observation, fewer than 0.5% of nontreated cells contained EA and
VCA.
The almost synchronous induction of EBV replication in Akata cells makes them
ideal for comparative studies of processing of native and recombinant gp85. This
chapter describes a detailed analysis of expression of recombinant and native gp85 in
Akata cells. The drug Brefeldin A (BFA) was included in the study. This antiviral
antibiotic inhibits post-translational processing of glycoproteins at a later stage than
monensin (Collins and Mottet, 1991). It blocks anterograde transport from the
endoplasmic reticulum, but, unlike monensin, it allows retrograde transport to continue.
This has the effect of causing a backflow of Golgi cisternae into the endoplasmic
reticulum or an intermediate endoplasmic reticulum-Golgi compartment (Collins and
Mottet, 1991; Dorns et al., 1989; Lippincott-Schwartz et al., 1990).


69
Materials and Methods
Cells and viral induction
Three lymphoblastoid cell lines were used, BJAB, an EBV negative Burkitt's
lymphoma line, B95-8, an EBV producing marmoset cell line ( Baer et al., 1984) and
Akata, an EBV producing Burkitt's lymphoma cell line (Takada and Ono, 1989; gift of
Dr. John Sixbey, St Jude's Childrens Research Hospital). Cells were grown in RPM1
1640 supplemented with 10% FBS, 100 units of penicillin and 100 pg of streptomycin
per ml and diluted biweekly. Virus replication occurred spontaneously in B95-8 cells
and was induced in Akata cells. Akata cells with a viability greater than 85% as judged
by trypan blue exclusion were centrifuged at 400 x g for five minutes at 4C. The
supernatant was discarded and resuspended in fresh medium at a concentration of 106
viable cells per ml. Latent EBV was induced into the lytic cycle by the addition of 100
pg per ml of anti-human immunoglobulin G, F(ab')2 fragment (Cappel, Organon Teknika
Corporation).
Radiolabeling of native gp85 in Akata cells
At various times after induction, the Akata cells were resuspended in either
methionine free RPMI or RPMI containing 1/10 concentration of glucose. Two hours
later 100 pCi [35S] methionine or 100 pCi [3H] glucosamine were added per ml. Cells
were either kept in radioactive medium until harvested or washed and resuspended in
radioactive-free medium containing 100 x methionine. Cells were harvested by
centrifugation and solubilized in RIPA as previously described for recombinant infected


70
CV-1 cells. Immunoprecipitated proteins were digested with endoglycosidases as
previously described. The inhibitors, DNM (ImM) and swainsonine (300 ng per ml)
were added to Akata cells at six hours post induction. Inhibitors of glycoprotein
transport, monensin and BFA, were added to suspensions of Akata cells at six hours post
induction. Final concentrations of ImM to 10 pM monensin and 2 pg per ml of BFA
were used in these experiments.
Extrinsic labeling of cells
One million Akata cells were induced with anti-immunoglobulin for 24 hours.
One million BJAB cells were infected at a MOI of 10 with either wild type vaccinia virus
strain 1HD-J or the VVpSCl l-BXLF2b construct. At either one hour post infection, or
24 hours post induction cells were centrifuged at 400 x g for five minutes, and
resuspended in 1 ml of serum free RPM1. Seven hours later the cells were centrifuged
and resuspended in 1/10 glucose media. The samples were subsequently incubated for
two hours at 37lC. Those samples that were to be labeled with 1251 were washed twice
and resuspended in 300 pi ice cold PBS. The 300 pi samples were transferred to tubes
containing one Iodobead (Pierce Chemical Company) and incubated at 25 C for 30
minutes. The Iodobead had previously been incubated with 300 pCi of 125I for five
minutes. After the incubation period, the cells were washed six times with 15 mis of
PBS. Cells were solubilized as previously described in 1 ml of RIPA for
immunoprecipitation and analysis by SDS-PAGE. Efficiency of labeling was determined
by comparing total radioactivity in 10 pi samples with the amount of radioactivity
incorporated into material precipitated by 5% trichloroacetic acid.


71
Antibodies
Two monoclonal antibodies were used, 72A1, which reacts with gp350/220
(Hoffman et al., 1980) and F-2-1, which reacts with gp85 (Strnad et al., 1982). One
polyclonal antibody, anti-BX, was used which reacts with gp85 (Oba and Hutt-Fletcher,
1988). All the antibodies were purified by chromatography on protein A-agarose and
anti-BX antibody was further purified on peptide coupled to Affigel as previously
described.
Immunofluorescence
For cytoplasmic staining, slides bearing cells that had been air-dried and fixed for
10 minutes in acetone at -20 C were incubated with antibody in a humidified
atmosphere at 37C for 35 minutes. Rabbit anti-peptide antibody, anti-BX, was used at
a concentration of 720 pg per ml, and monoclonal antibody 72A1 was used at a
concentration of 10 pg per ml. Cells were washed three times with PBS and reincubated
for 35 minutes at 37 C with the appropriate dilution of fluorescein
isothiocyante-conjugated goat anti-rabbit or rabbit anti-mouse immunoglobulin. The
cells were washed three times and mounted in a solution containing 50% PBS and 50%
glycerol.
For surface staining, 2 x 106 cells were fixed briefly with ice cold 0.1%
paraformadehyde in PBS. The fixed cells were reacted sequentially with rabbit
anti-peptide antibody and fluorescein conjugated goat anti-rabbit serum. Cells were
washed three times between incubations by centrifugation in ice cold PBS and three
times before being mounted on slides for examination with a fluorescence microscope.


72
Results
We initially wanted to compare the expression and biosynthesis of [35S]
methionine labeled recombinant gp85 with that of [35S] methionine labeled native gp85.
We had successfully radiolabeled the recombinant with [35S] methionine as described in
Chapter 3. However, we had not been able to achieve such high levels of [35S]
methionine incorporation into the native molecule. In contrast, using the Akata cell line
which we received from Dr. Sixbey enabled us to accomplish this goal (data not shown).
Before using the Akata cell line to analyze expression and biosynthesis of native gp85,
we needed to demonstrate that uninduced Akata cells would not also produce gp85. To
accomplish this, we compared the gp85 synthesized in uninduced Akata cells with that
produced in the induced Akata cells (Figure 4-1). Both samples were radiolabeled with
[35S] methionine and immunoprecipitated with the monoclonal antibody F-2-1. As
expected, F-2-1 immunoprecipitated a protein with an apparent Mr of 85,000 from
induced Akata cell lysates. No similar size species was detected in the uninduced Akata
sample.
We began analyzing the biosynthesis of native gp85 by determining the kinetics
of its synthesis. Synthesis of gp85 was first detected at 6.5 hours post induction (Figure
4-2, lane 2) and high levels of labeled protein were seen by eleven hours (lane 6 8).
Native gp85 appeared as a dimer in these experiments. The dimer formation was seen
particularly well in lysates of cells that had been labeled for 15 minutes at 8 hours after
induction (lane 5).


73
To examine processing of the native protein within the endoplasmic reticulum,
induced Akata cells were labeled in the presence of the glucosidase I inhibitor, DNM.
As expected, treatment with DNM increased the apparent Mr of gp85 by approximately
2.5 K (Figure 4-3, lane 2). We next wanted to determine whether native gp85 contained
any endo H resistant oligosaccharides which would be indicative of transport from the
endoplasmic reticulum to the Golgi apparatus. A comparison was made of the ability of
endo H and F to remove sugars from the protein (Figure 4-4). Repeated analysis of
native gp85 revealed a clear difference between the mobility of the endo H and the endo
F digested protein. The lysates that had been digested with endo F precipitated protein
species with a Mr of 74,000, while those digested with endo H contained
immunoprecipitated proteins of 79,000. Since lysates of virus producing cells contain
glycoproteins at different stages in processing and might have biased the results of this
experiment, we also examined the susceptibility of mature gp85 present in the virion
(Figure 4-5). No significant differences were seen in the digestion patterns of protein
immunoprecipitated from mature virions or from infected cells.
We next analyzed biosynthesis of native gp85 in the presence of swainsonine,
which inhibits mannosidase II (Figure 4-3). Native gp85 appeared less heterogeneous
when made in the presence of swainsonine. Inhibitors of glycoprotein movement, BFA
and monensin, were also used to examine gp85 processing within the Golgi (Figure 4-3,
lanes 4 and 5). Treatment of cells with BFA resulted in a glycoprotein species with a
slightly higher Mr. Glycoprotein gp85 made in cells treated with monensin, however,
was similar in size to the molecule made in untreated cells. These experiments were


74
repeated with higher concentrations of monensin (Figure 4-6). Increasing the
concentration of monensin, reduced the amount of protein made, but did not alter its
size.
The studies to this point clearly indicated that the native gp85 produced in Akata
cells and recombinant gp85 synthesized in CV-1 cells differed in both antigenicity and
processing. Glycosylation and processing of proteins can vary from cell type to cell type
(Rademacher et al., 1988). It thus remained possible that the differences between the
recombinant and the native proteins might be attributable to the fact that they were being
made in different cells. To examine this possibility further, we compared expression of
native gp85 in induced Akata cells with that of recombinant gp85 produced in the BJAB
cells, which, like Akata, are B cells derived from a Burkitt's lymphoma; however, unlike
Akata, do not contain the EBV genome.
We first compared the antigenicity of gp85 in each cell. Anti-BX antibody
immunoprecipitated a glycoprotein with an apparent Mr of 85,000 from both infected and
induced cells (Figure 4-7). In contrast, antibody F-2-1 immunoprecipitated a
glycoprotein with an apparent Mrof 85,000 only from induced Akata cells. We next
compared the expression of native and recombinant gp85 on the cell surface. We
analyzed the cell surface localization of the molecules using two methods. The first
approach was to determine the accessibility of the glycoproteins to labeling at the cell
surface with radioactive iodine, while the second approach was to stain the molecules
with fluorescent antibodies. Anti-BX antibody immunoprecipitated 12>I labeled gp85
from induced Akata cells, but not from BJAB cells infected with recombinant or wild


75
type vaccinia virus (Figure 4-8, panel A). This was despite the fact that in a parallel
control experiment, anti-BX immunoprecipitated the recombinant protein from BJAB
cells that have been labeled metabolically with [3H] glucosamine (panel B).
Immunofluorescence staining confirmed the failure of transport of recombinant gp85.
Anti-BX antibody stained gp85 in both acetone and paraformaldehyde fixed EBV
expressing B95-8 cells (Figure 4-9). However, despite the high percentage of acetone
fixed cells that stained with anti-BX antibody, no surface staining was detectable. In
addition, a different intracellular staining pattern was evident between cells producing
the native or the recombinant molecule. While a homogeneous pattern of fluorescence
was seen throughout the Akata cell, fluorescence was concentrated at perinuclear regions
of the BJAB cell infected with recombinant vaccinia.


76
1 2
-97.5 Kd
-69.0 Kd
-45.0 Kd
FIGURE 4-1. Comparison of gp85 synthesis in uninduced (lane 1) and in induced
Akata cells (lane 2). One million Akata cells were radiolabeled with 100 pCi of
[35S]methionine and resulting cell lysates were immunoprecipitated with antibody F-2-1.
Samples were run on an SDS-PAGE gel under reducing conditions and visualized by
standard fluorographic techniques.


77
FIGURE 4-2. Determination of optimum labeling of gp85. One million Akata cells
were labeled with 100 pCi of [,5S] methionine for 1 hour (lanes 1, 4 and 6), 30 minutes
(lanes 2 and 7) or 15 minutes (lanes 3, 5 and 8) at either six hours post induction (lanes 1
through 3), eight hours post induction (lanes 4 and 5) or ten hours post induction (lanes 6
through 8). All samples were incubated for and additional 20 hours after labeling and
immunoprecipitated with monoclonal F-2-1. Cell lysates were analyzed on SDS-PAGE
gels under reducing conditions and visualized by standard fluorographic techniques.


78
1 2 3 4 5
jgfKMji
-97.5 Kd
-69.0 Kd
-45.0 Kd
FIGURE 4-3. Analysis of native gp85 expression in the presence of 1 mM DNM (lane
2), 300 ng per ml of swainsonine (lane 3), 2 pg per ml of BFA (lane 4), 0.001 pM
monensin (lane 5), or mock treated (lane 1). Akata cells were induced for six hours with
anti-immunoglobulin prior to being radiolabeled with 100 pCi [35S] methionine for 20
hours and immunoprecipitated with monoclonal antibody F-2-1. All samples were
analyzed under reducing conditions by SDS-PAGE and visualized by standard
chromatographic techniques.


79
FIGURE 4-4. Effects of endoglycosidases on native gp85. Akata cells were induced for
six hours with anti-immunoglobulin prior to being radiolabeled with 100 uCi of [35S]
methionine for 20 hours. The radiolabeled cell lysates were immunoprecipitated with
F-2-1 and incubated overnight with enzyme buffer (lane 1), endoglycosidase H (lane 2)
or endoglycosidase F (lane 3). All samples were analyzed under reducing conditions by
SDS-PAGE and visualized by standard fluorographic techniques.


80
FIGURE 4-5. Endo H and F digestion pattern of proteins immunoprecipitated from
mature virions released into the culture supernatant (lanes 1 through 3) and endo H and F
digestion pattern of proteins immunoprecipitated from the infected cell (lanes 4 through
6). All samples were radiolabeled with 100 pCi of [35S] methionine and
immunoprecipitated with monoclonal antibody F-2-1. Immunoprecipitated samples were
either mock treated (lanes 1 and 4), treated with endo H (lanes 2 and 5) or treated with
endo F (lanes 3 and 6). All samples were analyzed under reducing conditions by
SDS-PAGE and visualized by standard chromatographic techniques.


81
12 3 4
07.5 Kd
69.0 Kd
15.0 Kd
FIGURE 4-6. Treatment of induced Akata cells with varying concentrations of
monensin. Samples were mock treated (lane 1) or treated with 0.001 pM monensin (lane
2), 0.01 pM monensin (lane 3), or 1.0 pM monensin (lane 4). Akata cells were induced
for six hours with anti-immunoglobulin prior to being radiolabeled with 100 pCi [35S]
methionine for 20 hours. All samples were analyzed under reducing conditions by
SDS-PAGE and visualized by standard chromatographic techniques.


82
FIGURE 4-7. SDS-PAGE analysis of recombinant and native gp85. BJAB cells were
infected with VVpEBl-BXLF2 (lanes 1 and 2). Akata cells were induced with
anti-immunoglobulin (lanes 3 and 4). At 6 hours post infection or post induction, cells
were radiolabeled with 100 pCi of [3H] glucosamine for 20 hours and EBV proteins
immunoprecipitated with antibody F-2-1 (lanes 1 and 4) or anti-BX antibody (lanes 2
and 3). The immunoprecipitated proteins were analyzed under reducing conditions in
7.5% acrylamide cross-linked with BIS. Proteins were visualized by standard
fluorographic techniques.


83
FIGURE 4-8. SDS-PAGE analysis of recombinant and native gp85. Panel A. Akata cells
were induced with anti-immunoglobulin (lane 1). BJAB cells were infected with
VVpSCl l-BXLF2b (lane 2) or with wild type vaccinia virus (lane 3). Cells were labeled
extrinsically with 1251 at 24 hours post induction or 10 hours post infection, lysed and
immunoprecipitated with anti-BX antibody. Panel B. BJAB cells were infected with
VVpSCl l-BXLF2b (lane 1) or with wild type vaccinia virus (lane 2). Cells were
metabolically labeled for ten hours post infection with [3H] glucosamine lysed and
immunoprecipitated with anti-BX antibody. The immunoprecipitated proteins were
analyzed under reducing conditions in a 7.5% acrylamide gel cross-linked with BIS and
proteins were visualized by standard fluorographic techniques.


84
FIGURE 4-9. Indirect immunofluorescence staining of native gp85 and recombinant
gp85. Cells were fixed with acetone (panels 1, 3 and 5), or with paraformaldehyde
(panels 2, 4 and 6), and incubated sequentially with anti-BX antibody and
fluorescein-conjugated goat anti-rabbit antibody. B95-8 cells (panels 1 and 2),
pSCl l-BXLF2b infected BJAB cells (panels 3, 4 and 6) or uninfected BJAB cells (panel
5). Panel 6 is a bright field view of panel 4.


85
Discussion
The availability of the Akata cell line facilitated analysis of the differences in
processing of native and recombinant gp85. Both native and recombinant proteins were
sensitive to the action of DNM which inhibits cleavage of terminal glucose residues on
oligosaccharide side chains in the endoplasmic reticulum. However, there were clear
differences in the sensitivity of the two proteins to inhibitors and enzymes that are used
to track movement and processing of glycoproteins beyond this first step. Our data
confirm previous observations with other strains of EBV (Edson and Thorley-Lawson,
1983) which suggested that mature gp85 carries some sugars that are sensitive and some
sugars that are resistant to digestion with endo H. However, none of the sugars on the
recombinant protein were resistant to endo H resistant. Since the processing of sugars
from high mannose, endo H sensitive type to the complex endo H resistant type, occurs
in the Golgi apparatus, this suggests that the recombinant molecule is blocked in
transport into this compartment. The alternative explanation for the absence of endo H
resistant species is that the recombinant molecule is folded in such a way that the sugar
residues are inaccessible to digestion. However, absence of recombinant gp85 on the
surface of infected cells argues in favor of a block in transport. While the effects of
swainsonine, monensin and BFA were not dramatic, they were consistent with the fact
that even the native protein apparently carries some high mannose, unprocessed sugar.
However, native gp85 and not recombinant gp85, migrated as a more discrete species in
the presence of these drugs, implying that differential processing of side chains is part of


86
the normal biosynthetic pathway of the protein in the Golgi and was not part of the
processing pathway of recombinant gp85.
Reevaluation of expression of recombinant gp85 in the lymphoid cell line, BJAB,
suggested that the aberrant processing was intrinsic to the recombinant molecule and was
not a function of the cell in which it was expressed. Heineman and coworkers (1988)
reported that gp85 is poorly expressed by retroviral vectors and probably not correctly
transported under these circumstances. Several groups have described expression of the
herpes simplex virus type 1 (HSV-1) and the human cytomegalovirus (HCMV) gH
proteins which are the homologues of EBV gp85. In both cases, gH is incompletely
processed and transported. Gompels and colleagues (1989) suggested that a second virus
protein might be required as a "chaperone". At this point it seemed likely that EBV
gp85 might share this requirement.


CHAPTER 5
ASSOCIATION OF THE EBV BKRF2 GENE PRODUCT
WITH NATIVE gp85
Introduction
Hutchinson and colleagues (1992) recently confirmed the speculation that an
additional protein was necessary for authentic processing and transport of HSV-1
glycoprotein, gH. These workers made antibodies to a peptide corresponding to the
predicted sequence of the HSV-1 UL1 gene product and immunoprecipitated two
differentially processed forms of a glycoprotein, designated gL, which complexed with
gH in the endoplasmic reticulum. The HSV-1 gL is predicted to have a Mr of 25,000
and contain a single site for the attachment of amino linked oligosaccarides (McGeoch et
al., 1988). The predicted sequence contains no methionine residues other than that at the
translational start site. A single hydrophobic sequence that might function as a signal
peptide is found at the amino terminus and since it was not possible to label gL with [35S]
methionine it was suggested that the signal sequence might be cleaved. Glycoproteins
gH and gL form a noncovalently associated heterodimer that is relatively stable in high
salt. Experiments done with cells coinfected with recombinant vaccinia viruses
expressing gH and gL demonstrated that the two glycoproteins are mutually dependent
on one another for processing and transport. Similar results were obtained by Kaye and
coworkers (1992) who found that correct processing of the HCMV gH was dependent on
co-expression of the product of the HCMV UL115 gene.
87


88
The HCMV UL115 gene product has no sequential homology with HSV-1 gL,
but the CMV gene is a positional homologue of HSV-1 UL1. Similar positional
homologues of the HSV-1 UL1 have been identified in varicella zoster virus and EBV
(Figure 5-1). The EBV homologue is the BKRF2 gene (Baer et al., 1984: Davison and
Taylor, 1987: McGeoch et al., 1988). The BKRF2 open reading frame (Figure 5-2) is
predicted to encode a protein of 137 amino acids with a calculated molecular weight of
15,000 (Baer et al., 1984). The predicted sequence includes three potential amino linked
glycosylation sites. The protein has a hydrophobic domain at the amino terminus, which
has characteristics of a signal peptide and like HSV-1 gL, no potential membrane anchor
sequence at the carboxyl terminus. Similar to the HSV gL, the BKRF2 open reading
frame contains only one methionine residue at position one. The BKRF2 gene product is
also predicted to contain five cysteine and four tyrosine residues. In this chapter, we
describe the search for chaperones of gp85 and demonstrate that the BKRF2 gene
product associates with gp85 in the virus producing cell.
Materials and Methods
Cells and virus
The lymphoblastoid cell lines B95-8 and Akata were used in addition to bovine
embryonic lung (BEL) cells. These cells were grown in DMEM supplemented with 5%
FBS, 100 IU of penicillin and 100 jag of streptomycin per ml. Stocks of AHV-1 (strain
1982: Zoological Society of San Diego, San Diego, Calif.) were prepared by first
washing BEL cells one time with DMEM that did not contain serum. These washed cells


89
were subsequently infected at a MOI of 5. The viral inoculum was removed at two hours
post infection, overlaid with DMEM supplemented with 2% FBS and incubated for 72
hours at 37C. Cells were harvested, frozen and thawed twice and virus stocks stored at
-85 C. EBV was obtained from Akata cells which were resuspended at a concentration
of 4 x 106 cells per ml and induced with 100 pg anti-human immunoglobulin G per ml
for 48 hours. Spent culture medium was clarified by centrifugation at 4,000 x g for 10
minutes, 100 pg bacitracin per ml was added, the virus was pelleted by centrifugation at
20,000 x g for 90 minutes and resuspended in 1/125 of the original volume of RPMI
1640 with 100 pg bacitracin per ml. EBV was labeled extrinsically with 125I after
pelleting from 4 mis of concentrated culture supernatant. Pelleted virus was resuspended
in 0.5 ml PBS and labeled with 0.5 mCi of l25I by use of Iodobeads.
Antibodies
One polyclonal rabbit anti-peptide antibody, anti-BK, was used in addition to the
previously described polyclonal antibody anti-BX and the monoclonal antibodies F-2-1,
and 72A1. A third monoclonal antibody, 12B5, recognizes a complex of envelope
proteins of Alcelaphine herpesvirus 1 (AHV-1) (Adams and Hutt-Fletcher, 1990).
Anti-BK antibody was generated against a synthetic peptide corresponding to residues
125 to 137 of the BKRF2 open reading frame. The peptide was synthesized by the
Protein Chemistry Core Facility of the University of Florida and a cysteine residue was
added to the amino terminus of the sequence to facilitate coupling with keyhole limpet
hemocyanin (Liu et al., 1979). The rabbit was immunized subcutaneously with 1 mg of
the conjugated BKFR2 peptide-KLH complex that had been emulsified in Freund's


90
complete adjuvant and subsequently at biweekly intervals with conjugated peptide
emulsified in Freund's Incomplete Adjuvant. The antibody was purified by
chromatography on protein A-Sepharose as previously described in chapter 3.
Immunofluorescence
For cytoplasmic staining, indirect immunofluorescence was performed on acetone
fixed Akata cells that had been induced with anti-human immunoglobulin for 24 hours.
Slides bearing air-dried acetone fixed cells and primary antibody were incubated in a
humidified atmosphere at 37C for 35 minutes. Rabbit anti-peptide antibody, anti-BK,
or preimmune rabbit antibody were used as the primary antibodies at a concentration of
7.4 pg per ml. The fixed cells were subsequently washed three times with PBS and
reincubated for 35 minutes with the appropriate dilution of fluorescein
isothiocyante-conjugated goat anti-rabbit serum. After incubation with the primary
antibody for 35 minutes the cells were washed three times in PBS and mounted in a
solution containing PBS and glycerol at a 1:1 ratio.
For surface staining, cells were fixed briefly in suspension with ice cold 0.1%
paraformaldehyde in phosphate buffered saline. The fixed cells were reacted in
suspension with either rabbit anti-peptide antibody, anti-BK, or preimmune rabbit
antibody at a concentration of 7.4 pg per ml. The fixed cells and primary antibody were
incubated for 35 minutes at 37C. Fluorescein conjugated goat anti-rabbit serum was
subsequently added, in proper dilution, to washed cells. The cells were washed by
centrifugation at 400 x g three times between incubations and three times before cells


91
were mounted in phosphate buffered saline for examination with a fluorescence
microscope.
Boiling analysis
Three samples of 2 x 106 Akata cells were induced with anti-immunoglobulin G
for six hours, incubated in medium containing 1/10 concentration of glucose for 2 hours
and radiolabeled with 100 (j.Ci of [3H] glucosamine for 20 hours. Cells were solubilized
in RIPA and immunoprecipitated with 120 pg of monoclonal antibody F-2-1 for 20
hours at 4C. One of the three samples was held at 4 C. The remaining two were boiled
for 5 minutes cooled to room temperature and reprecipitated overnight at 4 C with either
100 pg of anti-BX or 100 pg of anti-BK antibody. All three samples were then analyzed
by SDS-PAGE in 12% acrylamide cross-linked with DATD.
Sucrose gradient centrifugation
Five million induced Akata cells were radiolabeled with 500 pCi of [3H]
glucosamine for 20 hours. The cells were harvested as previously described and lysed
with Lysing buffer (50 mM Tris-HCl, pH 7.4, 100 mM NaCl, 1.0% Triton X-100, 0.5%
deoxycholate, 100 units per ml of Aprotinin and ImM PMSF) for 1 hour before
vortexing and centrifugation. The lysates were loaded onto continuous 5-25% sucrose
gradients and centrifuged for 18 hours at 175,000 x g at 18C. One ml fractions of the
gradient were then immunoprecipitated with 100 pg of anti-BKRF2 antibody and
protein-A agarose for 20 hours at 4C. Samples were processed, as previously described
for electrophoretic analysis on 12% acrylamide gels cross-linked with DATD.


92
Coinfection of AHV-1 and vaccinia virus
Thirty million BEL cells were infected with AHV-1 1982 (strain P2) at an MOI
of 5. At two hours post infection, 10 mis of DMEM with 10% FBS was added to each
T-75 flask. AHV-1 infected cells were then incubated at 37C for 24 hours and
superinfected with VVpSCl l-BXLF2b, VVpEBl-BXLF2 or wild type vaccinia virus
strain IHD-J. Two hours later, medium was replaced with five mis of methionine
deficient medium. After reincubation for one hour 50 pCi per ml of [35S] methionine
was added for 19 hours. The cells were harvested by centrifugation, washed with
DMEM and solubilized in 2 mis of RIPA. The solubilized cells were sonicated on ice for
one minute and centrifuged at 100,000 x g for 1 hour at 4C. Five hundred microliters
of the supernatant was preabsorbed with protein A-Agarose for 24 hours at 4 C and then
immunoprecipitated overnight with 100 pg of antibody and fresh protein A-Agarose.
Results
Previous studies with the EBV gp85 homologues in HSV-1 and HCMV had
suggested that an additional virus encoded protein is required for their intracellular
transport. Mutants of EBV that lack expression of gp85 are not available. In an attempt
to provide a putative virus chaperone protein, we therefore coinfected cells with
recombinant vaccinia virus and with another gamma herpes virus Alcelaphine herpes
virus 1. Expression of AHV-1 glycoproteins was confirmed by immunoprecipitation of
dually infected cells with antibody 12B5. The gp 115 complex recognized by this
antibody was clearly visible in analyses of samples coinfected with VVpSCl l-BXLF2b,


Full Text
CLONING AND CHARACTERIZATION
OF EPSTEIN-BARR VIRUS GLYCOPROTEIN gp85
By
LINDA RUTH YASWEN
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA

DEDICATION
This dissertation is dedicated to my parents who inspire me to reach for the stars.
11

ACKNOWLEDGMENTS
Great appreciation and thanks for all those who have helped complete this work.
Thanks to my committee members, Drs. Maureen Goodenow, Edward Wakeland and
Ammon Peck who have been extremely encouraging and helpful. A very special thanks
to the co-chair of my committee, Dr. Edward Stephens, who has been both my mentor
and friend. My deepest gratitude to the person who has made this all possible, the chair
of my committee, Dr. Lindsey Hutt-Fletcher. She has been an inspirational teacher.
Many thanks to Susan who has made the mountains seem like small hills; to Paula,
Steven, and Minh-Thanh who have been such incredible friends throughout the years; to
my parents, sister and brother for their wonderful senses of humor and enormous
support, and to John who has endured the long hours with me and become such a big
part of my world.
iii

CONTENTS
CHAPTER 1
INTRODUCTION 1
General Characteristics of Epstein-Barr Virus 1
Disease Associations of EBV 2
Virus Replication in Vitro 4
Early Events in Viral Infection 5
CHAPTER 2
CLONING OF gp85 18
Introduction 18
Materials and Methods 21
DNA sequencing 21
Construction of shuttle vectors 22
Blunt ending of DNA 24
Colony blot hybridization 24
Cell lines 25
Generation of recombinants 26
Virus 27
DNA dot blot analysis 27
Preparation of DNA from colonies 28
Large scale preparation of DNA 29
Results 30
Discussion 46
CHAPTER 3
EXPRESSION AND BIOCHEMICAL CHARACTERIZATION
OF RECOMBINANT gp85 48
Introduction 48
Materials and Methods 50
Cells and virus 50
Antibody purification 51
IV

Enzyme-linked immunosorbant assay 52
Analysis of vaccinia proteins 53
Electrophoresis of proteins 54
Enzyme and glycosylation inhibitors 55
Silver staining 56
Results 56
Discussion 65
CHAPTER 4
COMPARISON OF NATIVE AND RECOMBINANT gp85 67
Introduction 67
Materials and Methods 69
Cells and viral induction 69
Radiolabeling of native gp85 in Akata cells 69
Extrinsic labeling of cells 70
Antibodies 71
Immunofluorescence 71
Results 72
Discussion 85
CHAPTER 5
ASSOCIATION OF THE EBV BKRF2 GENE PRODUCT
WITH NATIVE gp85 87
Introduction 87
Materials and Methods 88
Cells and virus 88
Antibodies 89
Immunofluorescence 90
Boiling analysis 91
Sucrose gradient centrifugation 91
Coinfection of AHV-1 and vaccinia virus 92
Results 92
Discussion 104
v

CHAPTER 6
CONSTRUCTION OF BACTERIAL FUSION PROTEINS 106
Introduction 106
Materials and Methods 107
Polymerase chain reaction 107
Western blot analysis 107
Purification of bacterial fusion proteins 108
Results 109
Discussion 116
CHAPTER 7
CONCLUSION 117
vi

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
CLONING AND CHARACTERIZATION OF EPSTEIN-BARR VIRUS
GLYCOPROTEIN gp85
By
Linda Ruth Yaswen
May 1993
Chairman: Lindsey Hutt-Fletcher
Major Department: Pathology and Laboratory Medicine
The Epstein-Barr virus (EBV) glycoprotein gp85 is the product of the BXLF2
gene and is found in the viral envelope as well as in the plasma membrane of EBV
infected cells. Although it plays a critical role in the penetration of virus through the B
lymphocyte membrane, only very small amounts of it are made in EBV infected cells.
To facilitate the structural and functional analysis of gp85, recombinant vaccinia viruses
were constructed to overexpress the glycoprotein. Recombinant gp85 was recognized by
a polyclonal antibody made to a peptide derived from the predicted amino acid sequence
of the BXLF2 gene. However, it was not recognized by either of two monoclonal
antibodies made to the native molecule. In contrast to native gp85, the recombinant
protein contained no sugars that were resistant to endoglycosidase H and its synthesis
was unaffected by inhibitors that influence processing in the Golgi complex.
Furthermore, the recombinant protein was not detected at the cell surface by indirect
vii

immunofluorescence assays nor was it accessible to labeling with radioactive iodine.
These data indicate that the recombinant molecule, as opposed to native gp85, is not
transported to the cell surface.
When native proteins were labeled metabolically with [3H]glucosamine or
extrinsically with radioactive iodine, two glycoproteins with apparent Mrs of 25,000 and
42,000, were found to associate with gp85. Homologues of gp85 in other herpesviruses
require additional viral proteins for processing and transport of recombinant molecules.
Within the EBV genome, the BKRF2 open reading frame has been identified as the
positional homologue of these chaperones. Antibodies made to a peptide derived from
the predicted BKRF2 sequence, immunoprecipitated glycoproteins with mobilities of
85,000 and 25,000 daltons. Gradient centrifugation and immunoprecipitation of proteins
from EBV infected cells confirmed that the 25,000 dalton protein cosedimented with
gp85. Furthermore, the anti-BKRF2 antibody was shown to react with the same 25,000
dalton protein that was isolated in complexed form with native gp85. These data suggest
that the BKRF2 gene product is specifically recognized by the anti-BKRF2 antibody and
that this protein associates with gp85 during processing. It is possible that gp85 is
biologically active only if this association takes place.
viii

CHAPTER 1
INTRODUCTION
General Characteristics of Epstein-Barr Virus
Epstein-Barr virus (EBV) is an enveloped DNA virus classified with the
herpesviridae. It contains a linear double-stranded DNA molecule of approximately
172,000 base pairs (Beisel et al., 1985). The genome consists of five large regions of
unique DNA domains, U1-U5, which are separated by four regions of internal repeats,
11-14. The genome is flanked on both ends by tandem direct repeats (Cameron et al.,
1987; Daumbaugh et al., 1982; Given et al., 1979). Epstein-Barr virus is a member of
the gamma herpesvirus subfamily, and is designated as the human herpesvirus type 4
(Roizman, B., 1982). The diameter of the mature virion is approximately 150 to 180 nm.
The viral envelope is acquired as the virus buds through the host cell; however, the
proteins within the envelope are virally derived (Spear, P. G., 1980). At least five virally
encoded proteins have been detected within the envelope of EBV. The plasma
membrane of productively infected cells contain these proteins as well. The molecular
masses of these proteins are approximately 300-350, 200-250, 140, 55-78 and 85
kilodaltons (Edson and Thorley-Lawson, 1981; Thorley-Lawson and Edson 1979;
Thorley-Lawson, D.A., 1988; Mackett et al., 1990). The proteins gp350/300,
gp250/200, gp55/78 and gp85 are glycosylated molecules, while the fifth, pi40, is not
(Balachandran et al., 1986). Within the envelope of the virus lies an icosahedral
1

2
nucleocapsid which is surrounded by an amorphous tegument. The nucleocapsid
contains the double-stranded DNA molecules in association with core proteins.
Disease Associations of EBV
Gamma herpesviruses have a very limited host range. All members of the group
infect lymphoblastoid cells in vivo and in vitro. EBV is the only member of the group
that infects humans and it was originally identified as having tropism for human B
lymphocytes. Other members of the group include Marek's disease virus of chickens,
herpes teles and herpes samiri virus of new world monkeys and murine herpes 68
(Efstathiou et al., 1990). These viruses infect T cells (Fleckenstein and Desrosiers, 1982;
Nonoyama, 1982). The host range of EBV in vitro is restricted to B lymphocytes of
humans and new world primates. EBV establishes latency in these cells and
immortalizes them. Latently infected cells usually contain more than one copy of the
complete EBV genome. The genome can be integrated, but is most often found as a
covalently closed circular episome (Lindahl et al., 1976). Recombination events between
terminally repeated DNA sequences allow the linear genome to circularize (Dambaugh et
al., 1980). The stable covalently closed circle (CCC) is maintained as an autonomously
replicating episome.
Epithelial cells have been identified as a second target for EBV infection (Sixbey
et al., 1987; Sixbey et al., 1984). Thus, the virus can persist in both the epithelial and B
cells (Thorley-Lawson, 1988). The epithelial cell is permissive for replication and is
thought to be the source of virus that is shed in the oropharynx. Entry of EBV into the
oropharynx and subsequent replication at that site represents the initial step in

3
pathogenesis of any primary EBV infection. Viral replication within the pharyngeal
epithelium (Sixbey et al., 1984; Sixbey et al., 1983) allows for the infection and
subsequent transformation of B lymphocytes. This transformation event enables the
cells to undergo rapid proliferation (Rickinson, 1987; Svedmer et al., 1984).
Approximately ninety-five percent of the adult population worldwide is infected with
EBV. Primary infection with EBV normally occurs during childhood and is clinically
silent. Elowever, if primary infection is delayed until young adulthood, it leads to
infectious mononucleosis (IM) in approximately fifty percent of cases (Niederman et al.,
1970).
EBV has also been associated with several proliferative disorders of both
lymphoid and epithelial origin. Initially, the association of EBV with nasopharyngeal
carcinoma (NPC) was based on the observation that high antibody titers to EBV were
found in all NPC patients (de-The and Zeng, 1987). In all cases of NPC, the EBV DNA
is found in the CCC episomal form (Thorley-Lawson, 1988). EBV infection is
hypothesized to be an important factor in NPC tumor progression. NPC is one of the
most common tumors in Southern China. Since this represents such a large population
of the world, NPC must be considered a significant world health problem. EBV is also
associated with the development of a second tumor called Burkitf s lymphoma. The viral
DNA is found in the majority of endemic Burkitf s lymphomas which is the most
common children's tumor seen in Africa (Burkitt, 1987). In addition to EBV, these
tumors contain characteristic chromosomal translocations. The c-myc region from
chromosome 8 is juxtaposed with either the immunoglobin heavy chain region located

4
on chromosome 14, the lambda light chain genes of chromosome 22 or the kappa light
chain genes of chromosome 2 (Lenoir, 1987). The unrestricted proliferation of B cells
caused by EBV infection may act synergistically with the immunosuppression caused by
holoendemic malarial infections in this region. These two factors acting in cooperation
with one another may increase the emergence of malignant cell clones.
EBV is also hypothesized to contribute to the development of post transplant
disorders. Immunosuppressive drugs such as cyclosporine A can disrupt homeostatic
EBV infections within post transplant patients. This may allow for the development of
lymphoproliferation (Cleary et al., 1986). The presence of EBV in AIDS patients causes
an increased risk of both malignant lymphoma development as well as development of
oral hairy leukaplakia (OHL). OHL is an EBV-induced proliferative lesion of the lateral
tongue epithelium.
Virus Replication in Vitro
Cultures of EBV-infected lymphocytes vary in their permissiveness for viral
replication: most cultures are nonpermissive, but replication does occur in a small
fraction of cells in some cultures. The nonpermissiveness of EBV infection has made it
difficult to study virus replication and also limits the amounts of purified virus available
for studying the components of mature virus particles. Clones of infected lymphocytes
that are more permissive of virus replication have been selected (Miller and Lipman,
1973) and have facilitated studies of the virus replication cycle and biochemical
analyses. Two isolates of EBV have been extensively studied, B95-8 and P3HR1. The
B95-8 strain is produced by a cell line derived from a clone of marmoset lymphocytes.

5
These lymphocytes had been infected with virus obtained from a patient with infectious
mononucleosis. The B95-8 viral DNA has been completely sequenced (Baer et al.,
1984) and has been the prototype used for gene mapping. The P3HR1 cell line is a clone
of the Jijoye Burkitt tumor derived cell line (Hinuma et al., 1967). The P3HR1 cells are
more permissive than the parent clone for virus replication and the virus produced by
P3HR1 cells lacks the ability to growth-transform noninfected B lymphocytes (Miller et
al., 1974).
Early Events in Viral Infection
Attachment, penetration and uncoating represent the earliest events occurring in a
viral replication cycle. Initially, the virus attaches via a membrane protein to a specific
cellular receptor. This specific interaction in part determines the type of cell that a virus
can infect. Other functions of membrane proteins include viral fusion and penetration.
The penetration event allows for liberation of the nucleocapsid into the cytoplasm of the
virally infected host cell. The ability of EBV to infect B lymphocytes is initiated by
attachment of the virus to the cellular CR2 receptor. CR2 is also known as CD21. The
physiologic ligand of the CR2 glycoprotein is the C3d fragment of complement
(Fingeroth et al., 1984; Nemerow et al., 1985). Epithelial cells in the oro- and
nasopharyngeal epithelium also express a receptor for EBV attachment. This receptor
may be a similar or related molecule to CR2 (Sixby et al., 1987; Young et al., 1989). A
loss of EBV cellular receptors occurs as both epithelial cells and B lymphocytes
differentiate. Attachment of EBV to CR2 is mediated by at least one viral membrane
glycoprotein. This glycoprotein, gp350/300 (Fingeroth et al., 1984; Nemerow et al.,

6
1985), is encoded by the BLLF1 open reading frame of EBV. BLLF1 also encodes for a
smaller membrane glycoprotein called gp220. The two glycoproteins represent products
of a differentially spliced RNA transcript (Beisel et al., 1985). While it is not clear
whether gp220 is also involved in the viral attachment event, it has been suggested that
the two glycoproteins bind more effectively to different forms of the EBV cellular
receptor.
Membrane fusion occurs as two lipid bilayers adhere and join to one another.
Endocytosis, exocytosis, cell division and cell fusion are only some of the cellular events
which involve this fundamental event. Within the cell, fusion is important for
intracellular communication. Here fusion allows for the transport vesicles to bud from
certain cellular organelles and fuse with others. The mechanism of membrane fusion is
still under investigation. Due to a relatively simple membrane composition, enveloped
viruses provide an excellent means of probing this event. Enveloped viruses enter cells
by fusing with the lipid bilayer of the cell membrane (Lonberg-Holm and Philipson,
1974; White et al., 1983). Two pathways of viral entry are commonly utilized. Some
viruses, such as Sendai virus (Scheid and Choppin, 1976), deposit their nucleocapsids
directly into the cytoplasm by fusing with the cellular plasma membrane at physiologic
pH. The alternative route utilizes receptor-mediated endocytosis followed by fusion with
the vesicle membrane. Macromolecules are taken into cells by this same process. This
pathway is initiated by binding of a ligand to a cell surface receptor. Subsequently, the
membrane invaginates to form a vesicle (Goldstein et al., 1979). For some viruses,
particular regions of the plasma membrane have been identified as the sites for

receptor-mediated endocytosis (Goldstein et al., 1979). These regions are called coated
pits. The protein clathrin is a major component within these pits and is thought to
participate in the early stages of endocytosis (Doxsey et al., 1987). Receptors and
receptor-ligand complexes are thought to be concentrated within coated pits at sites of
internalization (Pearse, 1975). Conditions that trigger fusion of the viral envelope with
the vesicle membrane are provided as the endosomal environment becomes acidified.
This process of receptor-mediated endocytosis and vesicle membrane fusion is utilized
for viral entry by Semliki Forest virus (SFV) (Helenius et al., 1980; Marsh and Helenius,
1980), influenza A (Matlin et al., 1981; White and Helenius, 1983; White et al., 1981)
and vesicular stomatitis virus (VSV) (White et al., 1983).
The major entry mechanism for human immunodeficiency virus (HIV) is reported
to be fusion with the plasma membrane at the cell surface. McClure et al. have reported
that the entry of HIV-1 occurs by a pH-independent mechanism (McClure et al., 1988).
However, these data are in disagreement with the conclusion of Maddon et al. (1986) and
Pauza and Price (1988) who proposed that HIV entry into T lymphoblastoid cells
occurred via an endocytic entry pathway. Therefore, the possibility exists that virus
could enter by both pathways in a pH independent manner.
Studies utilizing electron microscopy and immunoelectron microscopy have
reported that EBV fuses directly at the plasma membrane of the Raji lymphoblastoid cell
line (Nemerow and Cooper, 1984; Seigneurin et al., 1977). EBV nucleocapsids were
found in the cytoplasm directly beneath the cellular plasma membrane. Virus was never
found to be bound to clathrin-coated areas of the plasma membrane, nor was the virus

8
observed in endocytic vesicles. However, studies using normal B lymphocytes revealed
transfer of membrane bound virus into membrane vesicles. These vesicles were distinct
in size and appearance from clathrin-coated vesicles and after 30 minutes, very few
virus particles remained in the vesicles. Miller and Hutt-Fletcher (1992) compared the
penetration events of EBV in epithelial and B lymphocytes. The penetration of EBV into
nonpolarized suspensions of epithelial cells was reported to occur by fusion at the cell
surface while it appeared that EBV fuses with normal B cells after being endocytosed.
Furthermore, both of these events appeared to be pH independent. EBV fusion events
were monitored using the pH insensitive fluorophore octadecyl rhodamine B chloride
(R18) and the pH dependent 5(N-octadecanoyl) aminofluorescein (AF). The AF looses
emission intensity when the pH falls below 7.4. Fusion was detected in both cell types
using R18 labeled EBV. However, penetration with AF labeled EBV could be detected
only in epithelial cells and not B cells unless they were first pretreated with drugs that
raise endosomal pH.
Fusion of two lipid bilayers is energetically unfavorable. This is due to a strong
repulsion that occurs between the two membranes as they approach one another. Energy
is required to displace water molecules from the hydrophilic surfaces. This hydration
force plus electrostatic forces between the two bilayers, prevents fusion events from
occurring spontaneously. Van der Waal forces can provide a weak attraction between
the membranes at distances of 20-30 angstroms. This enables the early events of
aggregation and attachment to occur; however, the combined repulsive forces soon
become overwhelming. Therefore, it seems likely that specialized fusogenic proteins

9
catalyze membrane fusion. Cellular fusogenic proteins have not been clearly identified,
whereas viral fusogenic proteins have been demonstrated. Viral fusion is mediated by a
specific viral membrane fusion protein. Hydrophobic regions within the viral fusion
protein are believed to destabilize the bilayer structure of the target membrane (Roizman,
1982). Several viruses have well characterized fusion proteins including Sendai, Semliki
Forest, influenza, vesicular stomatitis virus, and human immunodeficiency virus (HIV).
The fusion proteins of these viruses are all glycoproteins which span the lipid bilayer and
have the majority of their mass exposed externally.
Two proteins are contained within the envelope of Sendai virus which is a
member of paramyxoviridae. The hemagglutinin-neuraminidase (HN) protein is
responsible for attachment of the virus to cell surface sialic acid residues. The fusion (F)
protein initiates viral penetration at the plasma membrane (Hsu et al., 1981; Scheid and
Choppin, 1974; Scheid and Choppin, 1976). The F protein consists of two
sulfhydryl-linked glycopeptides called FI and F2. These two glycopeptides result from
proteolytic cleavage of the inactive precursor F0. The proteolytic cleavage occurs by a
host cell protease enzyme (Hsu et al., 1982). Virus produced by cells that lack a suitable
protease for F protein activation are noninfectious (Hsu et al., 1982). The F2 protein
corresponds to the amino terminus of F0, and the protein is anchored in the bilayer
through FI. The amino terminus of FI, resulting after cleavage of F0, has been found to
be unusually hydrophobic (Gething et al., 1978). It has been suggested that this
hydrophobic domain might be involved in the fusion event. Furthermore, the amino acid

10
sequences in this region are highly conserved among paramyxoviruses (Scheid et al.,
1978).
The orthomyxoviruses also have two types of spike glycoproteins. One of the
glycoproteins is neuraminidase (NA). The other, hemagglutinin (HA), has the capability
to bind to cell surface sialic acid residues and to catalyze fusion (Choppin and Compans,
1975; White et al., 1982). Unlike paramyxoviruses, othomyxoviruses are endocytosed
and fuse with the endocytic vesicle. The HA consists of two disulphide linked
glycopeptide chains, HA1 and HA2, resulting from proteolytic cleavage of a precursor
glycoprotein HA^ (Lazarowitz and Choppin, 1975; White et al., 1983). The cleavage is
irrelevant to adsorption, but is a prerequisite for infectivity (White et al., 1983). The
cleavage generates a new amino terminus on HA2 which is hydrophobic and highly
conserved in different influenza strains and somewhat homologous with the amino
terminus of FI. The HA molecule in its neutral form is a trimer and the hydrophobic
fusion peptide in each monomer is unexposed until the low pH of the endocytic vesicle
causes partial dissociation of the HA trimer. It is thought that this dissociation exposes
the fusion peptide which can insert into the target bilayer ( Dorns et al., 1985; Schlegel et
al., 1982) and initiate endosomal membrane fusion.
The envelope spikes of Semliki Forest virus (SFV), a togavirus, consist of a three
glycopeptide complex El, E2 and E3. El and E2 are transmembrane glycoproteins. E3
is non-covalently associated with E2 and is external to the bilayer. SFV does not fuse
with the plasma membrane at physiologic pH (Helenius et al., 1980). Virions are
endocytosed and a drop in pH within the endocytic vesicle activates membrane fusion

11
(Marsh et al., 1983). Lysosomotropic agents which elevate endosomal pH inhibit SFV
penetration (Helenius et al., 1982). SFV can fuse directly with the plasma membrane in
vitro at low pH (White et al., 1980). The SFV spike glycoproteins have been shown to
be fusogenic in the absence of other virus components (Marsh et al., 1983). It has been
suggested that the role of peptide El is directly linked to the fusion event. Both SFV and
Sindbis, another togavirus, have El proteins containing a hydrophobic peptide segment
located close to the amino terminus and this segment has an external position in the virus
membrane (Garoff et al., 1980; White et al., 1983). Since El and E2 occur as a complex,
E2 may also participate in the fusion reaction. The role of E3 is not clear; it is a small
peripheral glycopeptide and there is no homologue in Sindbis virus (Welch and Sefton,
1979).
Vesicular stomatitis virus (VSV), a rhabdovirus, has only one envelope
glycoprotein (the G-protein). The G-protein has a hydrophobic region near the carboxyl
terminus forming the membrane spanning domain. A small hydrophilic sequence at the
carboxyl terminus is in contact with the cytoplasm. The larger amino terminal domain,
containing the oligosaccharide chains, is exposed to the exterior of the cell (Rose and
Gallione, 1981; Rose et al., 1980). Fusion activity occurs at the plasma membrane when
VSV is attached to the surface of cells and placed in a low pH medium (Blumenthal et
al., 1987; Matlin et al., 1982). Eukaryotic cells expressing the cloned G-protein gene
fuse at low but not at neutral pH (Rose and Gallione, 1981). These data indicate that the
G-protein is both necessary and sufficient for fusion activity (Reidel et al., 1984).

12
The DNA-containing herpesviruses are considerably more complex than any of
the RNA-containing viruses. The best studied herpesvirus is herpes simplex virus
(HSV). HSV has an envelope that contains at least seven glycoproteins which have been
characterized and sequenced (Bzik et al., 1984; Frink et ah, 1983; Gompels and Minson,
1986; McGeoch et ah, 1985; Pellet et ah, 1985; Watson et ah, 1982). Three of the
glycoproteins, namely gB, gD and gH, induce antibodies capable of neutralizing HSV
infectivity in the absence of complement. All three glycoproteins have also been
implicated in virus penetration (Fuller and Spear, 1985; Gompels and Minson, 1986;
Sarmiento et ah, 1979). Evidence implicating gB in the viral penetration event comes
from studies of temperature sensitive HSV-1 mutants. These mutants fail to process
precursor gB molecules to mature forms at nonpermissive temperatures. Though the
virions produced are noninfectious, they can still bind to cells and block superinfection
with wild type HSV-1. Furthermore, the block in viral penetration is overcome when the
temperature sensitive mutants and cell complexes are treated with the membrane fusing
agent, polyethylene glycol (Little et ah, 1981; Sarmiento et ah, 1979).
Neutralizing anti-gD monoclonal antibodies have been shown to block HSV
infection by preventing virus cell fusion at the plasma membrane (Fuller and Spear,
1987). Antibodies to this glycoprotein also block HSV-induced cell-cell fusion, a
process which may be similar to the virus-cell fusion required for entry (Noble et ah,
1983). In addition, it has also been reported that deletion mutants of gD can bind to cells
but not block superinfection with wild-type HSV (Johnson and Ligas, 1988). The
glycoprotein gH is present in the viral envelope at ten-fold lower concentrations than gD

13
(Richman et al., 1986). Despite this fact, antibodies against gH demonstrate neutralizing
activity comparable to that of antibodies against gD (Cranage et al., 1988). Fuller and
Spear (1987) demonstrated that anti-gH monoclonal antibodies block the fusion of the
virus with the target cell membrane. These monoclonal antibodies do not block viral
attachment. Furthermore, monoclonal antibodies to gH are able to abolish syncytium
formation in those cells infected with syncytial HSV-1 strains (Gompels and Minson,
1986). A temperature sensitive mutant of HSV was made that contained a substitution in
the gH glycoprotein (Desai et al., 1988). The extracellular virus particles produced were
devoid of gH and noninfectious. Recently, it was reported that replacement of deleted
gH coding sequences with the E. coli lacZ gene resulted in mutant virus that could not
enter cells. In contrast to the gD- mutant, however, this gH- virus was able to block
superinfection with wild-type HSV (Forrester et al., 1991). The authors suggest that the
gH functions after gD in the viral entry process. Taken together, these findings indicate
that gH-1 is involved in cellular fusion events. Thus, the three glycoproteins gB, gD and
gH are likely to either induce or influence the fusion process. This fusion event occurs in
a pH-independent manner at the surface of the cell. There is no evidence to suggest that
gB, gD and gH act as a single functional heteropolymer. Homodimers of gB extracted
from virions of infected cells are not associated with other glycoproteins
(Claesson-Welsh and Spear, 1986). Furthermore, gB and gD have been shown to form
morphologically distinct structures in the virion envelope (Stannard et al., 1987).
The viral envelope protein that mediates the fusion event between EBV and the
host cell has not been demonstrated conclusively. However, the EBV envelope

14
glycoprotein, gp85, does have characteristics of a fusion molecule. Monoclonal
antibodies to gp85 neutralize virus have provided indirect evidence suggesting that the
glycoprotein may play an active role in virus penetration through the cell membrane
(Miller and Hutt-Fletcher, 1988). The antibody, F-2-1, failed to inhibit binding of EBV
to its receptor, but interfered with virus fusion as measured with the self-quenching
fluorophore octadecyl rhodamine B chloride (R18). This fluorescent amphiphile,
diffuses readily into biologic membranes and exhibits self-quenching properties at high
concentrations. Such high concentrations can be reached in the viral envelope.
However, a measurable relief of self-quenching and fluorescence is detected when virus
and cell membranes fuse. This is due to R18 molecules moving within the fused
membranes, creating lower concentrations of the fluorescent molecule. The relief of
self-quenching was inhibited by the neutralizing monoclonal antibody F-2-1; however,
inhibition was not seen when a non-neutralizing antibody to the same molecule was used
(Miller and Hutt-Fletcher, 1988).
Further evidence to support the hypothesis that gp85 functions as a fusion
protein, is provided by experiments in which EBV virion proteins including or depleted
of gp85 were incorporated into lipid vesicles to form virosomes. R18 labeled virosomes
that were made with nondepleted protein were shown to behave in a manner similar to
that of R18-labeled virus in the experiments described above. The virosomes bound to
receptor positive, but not to receptor negative cells. Fusion was demonstrated with Raji
cells, but not with receptor positive, fusion incompetent Molt 4 cells. Furthermore,
monoclonal antibodies that inhibited either binding or fusion of virus, inhibited binding

15
and fusion of virosomes. Lastly, virus competed with virosomes for attachment to cells.
Those virosomes made from viral proteins depleted of gp85 remained capable of binding
to receptor positive cells, but failed to fuse.
Neutralizing sera obtained from patients in the acute phase of mononucleosis
demonstrate a predominant antibody response against the virally encoded gp85
(Qualtiere and Pearson, 1979). As well as mediating viral neutralization, monoclonal
antibodies to gp85 also demonstrate complement dependent cytolysis of EBV infected
cells (Strnad et al., 1982). Although gp85 is thought to be a minor component of the
virion (Douglas Oba, Thesis), it is an essential glycoprotein. It is an important target of
the host immune response directed against EBV. Glycoprotein gp85 maps to the BXLF2
open reading frame of EBV DNA (Heineman et al., 1988; Oba and Hutt-Fletcher, 1988).
Analysis of the BXLF2 sequences predicts that gp85 contains five N-linked
glycosylation sites, a potential amino terminal signal sequence and a carboxy terminal
anchor sequence (Figure 1-1). The sequence also includes a stretch of 16 extremely
apolar amino acids that could be a fusion domain (Oba and Hutt-Fletcher, 1988) are also
detected within BXLF2 sequences.
The gp85 glycoprotein has been identified as the EBV gH, which has
homologues in each of the three herpesviruses subfamilies (Gompels and Minson 1986;
McGeoch and Davison, 1986; Keller et al., 1987; Cranage et al., 1988; Gompels et al.,
1988; Pachl et al., 1989; Nicholson et al., 1990; Josephs et al., 1991; Klupp and
Mettenleiter, 1991; Meyer et al., 1991). In at least four of these herpesviruses, gH has
been shown to play a role in virus penetration (Gompels and Minson, 1986; Fuller et al.,

1989; Forrester et al, 1992; Peeters et al., 1992). The goal of the present work was to
analyze in greater depth, the structure and function of the EBV gH molecule called gp85.

17
1 20 40
MQLLCVFCLVLLWEVGAASLSEVKLHLDIEGHASHYTIPWTELMAKVPG
60 80
LSPEALWREAEVTEDLASMLNRYKLIYKTSGTLGIALAEPVDIPAVSEGS
100 120 140
MQVDASKVHPGVISGLNSPACMLSAPLEKQLFYYIGTMLPNTRPHSYVF
160 180
YQLRCHLSYVALLSINGDKFQYTGAMTSKFLMGTYKRVTEKGDEHVLSL
200 220 240
VFGKTKDLPDLRGPFSYPSLTSAQSGDYSLVIVTTFVHYANFHNYFVPNL
260 280
KDMPSRAVTMTAASYARYVLQKLVLLEMKGGCREPELDTETLTTMFEVS
300 320 340
VAFFKVGHAVGETGNGCVDLRWLAKSFFELTVLKDIIGICYGATVKGMQ
360 380
SYGLERLAAMLMATVKMEELGHLTTEKQEYALRLATVGYPKAGVYSGLI
400 420 440
ggatsvllsaynrhplfqplhtvmretlfigshvvlrelrlEIvttqgpn
460 480
LALYQLLSTALCSALEIGEVLRGLALGTESGLFSPCYLSLRFDLTRDKLLS
500
520
V-1
540
MAPQEATLDQAAVSNAVDGFLGRLSLEREDRDAWHLPAYKCVDRLDKV
560 580
lmiipliIBvtfiissdrevrgsalyeasttylssslflspvimnkcsqgava
600 620 640
geprqipkiqEJftrtqkscifcgfallsydekegletttyitsqevqnsil
660 680
SSNYFDFDNLHVHYLLLTTNGTVMEIAGLYEERAHVVLAIILYFIAFALGIF
700
LVHKIVMFFL
Figure 1-1. Predicted amino acid sequence of the BXLF2 open reading frame. Solid
underlines indicate hydrophobic regions predicted by the program of Klein and
colleagues to have a high probability of being membrane spanning regions; residues in
black boxes indicate potential N-glycosylation sites; region designated as V-1 indicates
residues synthesized.

CHAPTER 2
CLONING OF gp85
Introduction
The EBV glycoprotein gp85 is produced in small amounts by the virus. Oba
reported that less than five picograms of the glycoprotein could be purified from a
culture of approximately 109 EBV-producing cells (Oba, D. E. Ph.D. Thesis, University
of Florida, 1988). This low yield impeded further structural and functional analysis of
the molecule. To surmount the problem, it was decided to overexpress the protein in a
eukaryotic expression system that would allow production of glycosylated molecules.
All fusion proteins studied to date have been shown to be glycosylated and authentic
glycosylation of at least one of these, the envelope protein of the human
immunodeficiency virus, is known to be critical to its biologic activity (Gruters et al.,
1987; Matthews et al., 1987; Walker et al., 1987). The eukaryotic expression system
chosen makes use of vaccinia virus, a member of the orthopoxvirus family. Several
features of this virus make it an excellent vehicle for expression of foreign genes. First,
it has a very large double-stranded DNA genome of 187 kilobase pairs (Geshelin, P. and
K. I. Berns, 1974) packaged in a capsid which is capable of accommodating large
segments of additional nucleic acid. Second, the vaccinia genome includes many genes
that influence virulence in vivo, but which are not essential for growth in tissue culture
and which can readily be replaced by recombination events. Third, unlike many viruses,
18

19
vaccinia is capable of replication in a wide variety of cells from several species.
Recombinant vaccinia viruses have been used by many workers to express faithful copies
of glycoproteins from a variety of other RNA and DNA viruses.
Vaccinia virus replicates in the cytoplasm of a cell and makes use of its own
virally encoded DNA and RNA polymerases. Packaged within the virus core is a
complete transcriptional enzyme system that is essential for infectivity. This
transcriptional enzyme system is necessary since transcription can only be initiated by
the vaccinia RNA polymerase and not by eukaryotic RNA polymerase II. In order to
express heterologous DNA in a recombinant vaccinia virus it is first necessary to
construct a chimeric gene that contains a vaccinia promoter fused to the protein coding
sequences of the foreign gene. This is accomplished using a plasmid vector. The
engineered vector must contain the transcriptional start site of the vaccinia promoter
upstream of the translational initiation codon of the foreign gene. After assembly within
the plasmid vector, the expression cassettes are inserted into the vaccinia genome by
homologous recombination. This is accomplished in vivo by transfection of plasmid
DNA into vaccinia virus infected cells (Mackett et al., 1984). This step is necessary
because the large size of vaccinia DNA makes it impractical to construct recombinant
molecules in vitro.
The insertion of the BXLF2 gene into vaccinia virus was accomplished using two
insertion vectors, pSCl 1 and pEBl (Figure 2-1). The pSCl 1 vector contains two
vaccinia promoters, one of which is the early/late promoter, p7.5. This promoter is
upstream of a unique Smal restriction endonuclease cleavage site into which the foreign

20
gene to be expressed is cloned. In the opposite orientation, the late promoter pi 1
controls expression of a p-galactosidase gene. The pEBl insertion vector is identical in
sequence to pSCl 1, except that the p7.5 promoter has been replaced by a second pi 1
promoter.
To facilitate homologous recombination the Sma\ site and the p-glactosidase gene
are flanked by sequences from the vaccinia thymidine kinase (TK) gene. These
sequences direct the insertion of intervening DNA into the nonessential region of the
virus genome. The vectors also contain a bacterial origin of DNA replication and a gene
conferring ampicillin resistance which are important to amplification and selection of the
plasmids in bacteria, but which are lost upon recombination with vaccinia. Recombinant
TK viruses are amplified by growth in TK' osteosarcoma cells in the presence of
5-bromo-2-deoxyuridine (5-BUdR). Phosphorylation of this nucleoside analog results in
its lethal incorporation into the DNA of those cells that have been infected with a TK+
wild type vaccinia virus. The resulting TK amplified viral stocks are plaqued in
monolayers of African Green Monkey kidney cells and plaques derived from
recombinant virus which contains the P-galactosidase gene are identified by overlaying
with agarose containing the chromogenic substrate 5-bromo-4-chloro-beta-
galactopyranoside (X-gal). This chapter describes the construction and selection of
recombinant vaccinia viruses containing the EBV sequence corresponding to the BXLF2
open reading frame.

21
Materials and Methods
DNA sequencing
Cesium chloride (CsCl) purified DNA was sequenced with the Sequenase kit
(Sequenase kit, Product No. 70700, United States Biochemical Corporation), which
uses a modified form of T7 DNA polymerase and the principles of the DNA
chain-termination sequencing procedure described by Tabor and Richardson. The T3
promoter sequence (5ATTAACCCTCACTAAAG3) was the primer for the reaction and
the nucleotides used were radiolabeled with [35S]dATP. Briefly, the DNA was denatured
in 0.2 M NaOH for five minutes at 25 C, neutralized by adding 0.4 volumes of 5M
ammonium acetate (pH 7.5) and ethanol precipitated. Two micrograms of precipitated
DNA were added to 50 ng of primer and 2 pi of 5x sequencing buffer (200 mM
Tris-HCl, pH 7.5, 100 mM MgCl2 and 250 mM NaCl) and the volume was adjusted
with dH20 to 10 pi. The solution was vortexed, heated to 65 C for two minutes and
held at 25 C for 30 minutes to allow annealing of DNA. Enzyme was added in 6 mM
DTT, containing 20 pCi of [a-35S]dATP, and unlabeled nucleotides (dGTP, dCTP and
dTTP), incubated at 25 C for five minutes and then aliquotted into each of four
termination solutions containing all the deoxynucleotide triphosphates, a
dideoxynucleoside triphosphate and 50 mM NaCl for further incubation for five minutes
at 37 C. The reaction was stopped by addition of 95% formamide, 20 mM EDTA.
Bromophenol Blue 0.05% and 0.05% xylene cyanol were added and samples held on ice
until electrophoresis in 6% acrylamide cross-linked with bis-acrylamide.

22
Construction of shuttle vectors
Three plasmids were used in the construction of vaccinia virus transfer vectors.
These were pSCl 1 (Chakrabarti et al., 1985) pEBl (Perez et. al., 1992; Stephens et. al.,
1992) and Bluescript pBSKS (+) (Stratagene) into which BXLF2 cDNA had been cloned
(pBKSKS-BXLF2; a gift of Dr. Elliott Kieff). The first vaccinia virus transfer vector to
be made was pSCl l-BXLF2a. This recombinant vector was made by digesting one
microgram of CsCl purified pBSKS-BXLF2 with four units of the restriction
endonucleases Smal and Hindi at 37 C overnight. The Smal Hindi fragment was
recovered by electroelution into a dialysis bag from a 1% agarose gel containing
ethidium bromide. The fragment was concentrated by ethanol precipitation for 30
minutes with two volumes of ice cold ethanol and 250 mM sodium acetate (pH 5.2) at
-20 C followed centrifugation at 10,000 x g for ten minutes. The supernatant was
discarded, the pellet was dried by vacuum desiccation for five minutes and the DNA was
resuspended in 25 pi of sterile distilled water (dH20). DNA was quantitated by visual
inspection after electrophoresis in 1% agarose in TAE buffer (50X: 242 gm Tris base,
57.1 ml glacial acetic acid, 37.2 gm Naj EDTA-2H20, pH 8.5). Three hundred
nanograms of the Smal Hindi fragment and 200 ng of Smal digested pSCl 1 were
added to a ligation reaction containing six units of T4 DNA ligase and 10 pi of 2X ligase
buffer (100 mM Tris-HCl, pH 7.5, 20 mM MgCl2, and 20 mM DTT). The total volume
was adjusted to 20 pi with dH20. The ligation reaction was incubated for 24 hours at 16
C. Ten microliters of this reaction was used to transform 300 pi of Escherichia coli
JM109 cells made competent by treatment with CaCl2. Bacteria and DNA were

23
incubated for 30 minutes at 0 C and then heat-shocked for two minutes at 37 C. One
ml of Luria-Bertani (LB) medium was added to the transformed cells which were
incubated for one hour at 37 C and then used to inoculate LB plates containing 60 pg
per ml ampicillin. The cultures were plated at volumes of 25, 50, 100, 150 and 200 pi
and incubated overnight at 37 C. Colonies containing BXLF2 DNA was identified by
colony hybridization.
The second vaccinia virus transfer vector to be made was pEB 1-BXLF2. Once
again the pBSKS-BXLF2 cDNA was digested with the restriction endonucleases Smal
and Hindi. The Smal Hindi fragment was ligated into Smal digested pEBl vector by
the previously described method.
The third vaccinia virus transfer vector to be made was pSCl l-BXLF2b which
was obtained using polymerase chain reaction (PCR) techniques (Saiki et al 1985; Saiki
et al., 1988) with the pBSKS-BXLF2 cDNA being used as the template for the reaction.
The amplified DNA contained only the open reading frame of the BXLF2 gene plus six
base pairs corresponding to Xhol sites added to the 5' ends of the EBV DNA sequence
(Figure 2-1). The primer for the forward reaction was
5'-CTCGAGATGCAGTTGCTCTGTGTTTTTTGC-3', while the primer for the reverse
reaction was 5 '-CTCGAGAAGGAAAAACATAACAATCTTGTG-3'. The DNA was
amplified in a DNA Thermal Cycler (Coy Temp Cycler Model 50) by 10 minutes
denaturation at 92 C followed by 25 cycles of denaturation at 92 C for one minute,
annealing at 55 C for three minutes and extension at 72 C for three minutes. Following
the last cycle, the reaction was held at 4 C. Each PCR reaction consisted of 10 pi of lOx

24
buffer (100 mM Tris-HCl, pH 8.3, 500 mM KC1, 15 mM MgCl2 and 0.01% gelatin;
Perkin-Elmer Cetus), 16 pi of 200 pM dCTP, dATP, dGTP and dTTP, 10 pi of each
primer at a concentration of 20 uM, 2.5 units of Amplitaq DNA polymerase
(Perkin-Elmer Cetus) and 500 ng of template DNA. The total volume was adjusted to
100 pi with dH20 and overlayed with 100 pi of mineral oil. The product of the reaction
was ethanol precipitated and stored in dH20 at 4 C.
Blunt ending of DNA
Two hundred nanograms of amplified DNA was blunt ended using 20 units of
DNA polymerase Klenow fragment in buffer (50 mM Tris-HCl, pH 7.6 with 10 mM
MgCl2) containing 2 mM of each of the nucleotides dATP, dGTP, dCTP and dTTP. The
total reaction volume was 30 pi. The mixture was incubated at 25 C for 30 minutes.
The blunt ended DNA product was purified using spin tubes (Millipore Inc.), ethanol
precipitated and resuspended in 20 pi of dH20.
Colony blot hybridization
Transformed Escherichia coli strain JM109 colonies were plated onto LB plates
containing ampicillin (Maniatis et al., 1989) and incubated for 15 hours at 37 C. One
hundred of the resulting colonies were streaked onto master plates, reincubated for a
further 15 hours at 37 C and then transferred to nitrocellulose filters (Micron
Separations Inc.). Filters were placed sequentially, colony side up, for 5 minutes each on
3MM Whatman filter paper saturated with 10% sodium dodecyl sulfate (SDS), with 0.5
M NaOH and 1.5 M NaCl, with 1.5 M NaCl and 0.5 M Tris HC1 (pH 8.0.) and with TE
buffer (10 mM Tris HC1, 1.0 mM EDTA, pH 8.0). The nitrocellulose was dried for a

25
minimum of 30 minutes at room temperature and then baked under vacuum at 80 C for
two hours. The baked filters were prehybridized for 24 hours at 45 C in a solution
containing 50% formamide, 0.5% SDS, 6x SSC (20X SSC: 3M NaCl, 0.3 M Na3
citrate-2H20, pH 7.0), 0.4% Ficoll 400, 0.4% polyvinylpyrrolidone, 0.4% bovine serum
albumin and 100 pg per ml of denatured and sheared herring sperm DNA. The
prehybridization solution was discarded and replaced with hybridization solution
containing 50% formamide, 20% SDS, 4x SSC, 0.4% Ficoll 400, 0.4%
polyvinylpyrrolidone, 0.4% bovine serum albumin, 100 ug per ml of denatured and
sheared herring sperm DNA and 1 to 15 ng of 32P labeled probe (specific activity of 107
cpm/pg) per ml of hybridization reaction. Filters were hybridized for 24 hours at 42 C
and washed for 15 minutes at 25 C in 2x SSC and 0.1% SDS and for 30 minutes at 60
C in a 0.1% SSC and 0.5% SDS. This last wash was repeated until the radioactivity
associated with the filter and measured with a hand held monitor was unevenly
distributed.
Cell lines
Two African Green Monkey Kidney cell lines were used, CV-1 and Vero. A
human osteosarcoma cell line, TK-143, was used to obtain TK- recombinant vaccinia
virus. All three cell lines were grown in Dulbecco modified Eagle medium (DMEM;
Gibco) containing 5% heat inactivated fetal bovine serum (FBS) and 100 IU per ml
penicillin and 100 pg per ml of streptomycin.

26
Generation of recombinants
All three recombinant shuttle vectors were used to make recombinant vaccinia
virus (Mackett et al., 1984). One million CV-1 cells were washed once in DMEM and
then infected with 500 pi wild-type, thymidine kinase positive (TK+) vaccinia virus
strain IHD-J at a multiplicity of infection (MOI) of 0.1 to 0.05 plaque forming units
(PFU). The infected cells were incubated at 37 C for two hours. At two hours post
infection, the viral inoculum was removed and cells were washed twice with one ml of
2x Hepes buffered saline (0.14 M NaCl, 5 mM KC1, 1 mM Naj HP04,0.1% Dextrose,
20 mM Hepes) and incubated at room temperature for 30 minutes with 1 ml recombinant
plasmid DNA that had been purified over CsCl and precipitated with 2.5 M CaCl2. Two
and a half hours later, 2.5 ml DMEM supplemented with serum and antibiotics were
added and the cells were reincubated at 37 C for 48 hours. The cells were then
harvested by scraping into spent culture medium and pelleted by centrifugation at 500 x
g for ten minutes. The pelleted cells were resuspended in 1 ml of DMEM and sonicated
for 1 minute in a sonic bath. Serial dilutions of the disrupted cells were used to infect
106 TK- 143 osteosarcoma cells. After 1 hour the viral inoculum was removed and
replaced with 1 ml of DMEM containing 25 pg per ml of 5-bromo-2-deoxyuridine.
Once viral plaques were visible, a 2% agarose overlay containing 400 pg per ml of the
chromogenic substrate 5-bromo-4-B-D-galactopyranoside (X-gal) was added. Blue
plaques were aspirated using sterile pasteur pipettes. The individual plaque lifts were
placed into 1 ml of DMEM and were either sonicated for 1 minute or frozen and thawed
three times. The plaque purification procedure was repeated a minimum of five times.

27
Virus
Virus stocks were grown either in Vero or CV-1 cells. Confluent cells were
infected at an MOI of 0.01 to 0.03 for 1 hour at 37 C. The total volume of infection
media was 500 pi for one well of a six well plate (9.4 cm2) and 5 mis for a T-150 flask
(150 cm2). At one hour post infection, 2 mis of DMEM containing 2% FBS were added
to one well of a six well plate and 12 mis were added to a T-150 flask. Infected cells
were harvested at 48 hours post infection and pelleted at 666 x g for 10 minutes at 4 C.
Pelleted cells were resuspended in 1 ml of serum free DMEM per 3 x 106 cells. The
resuspended pellets were frozen and thawed three times and centrifuged at 4 C for 10
minutes at 500 x g. The supernatant was stored at -85 C and titrated. Serial dilutions of
the virus stock was used to infect Vero or CV-1 cells. After several days of growth, the
medium was removed and the cells were stained with neutral red (10 mg/ml in dH20).
The viral plaques appeared as areas of reduced staining of approximately 1 mm in
diameter. The viral titer was then calculated by counting the plaques within the wells
and multiplying by the dilution factor.
DNA dot blot analysis
One million CV-1 cells were infected with pSCl l-BXLF2a or a recombinant
vaccinia virus expressing the BILF2 open reading frame (Mackett et al, 1990). The cells
were infected at an MOI of 10 and at 36 hours post infection washed twice with DMEM
before harvesting with a rubber policeman. One fifth of the harvested cells were
deposited onto a nitrocellulose filter and dried at room temperature for 1 hour. The
nitrocellulose was wetted with 100 mM NaCl and 50 mM Tris-HCl (pH 7.5) and then

28
placed sequentially on 3MM Whatman filter paper saturated with 0.5 M NaOH, 1.5 M
NaCl with 1.5 M NaCl, 0.5 Tris-HCl (pH 8.0) and with 2x SSC. The filter was dried at
room temperature for 30 minutes and before baked at 80 C for 2 hours. The filter was
prehybridized and hybridized as described above.
An alternative approach (monolayer blot analysis) was used to test for the
presence of BXLF2 DNA in the pEBl-BXLF2 vector. Nitrocellulose filters were placed
directly onto vaccinia virus infected CV-1 cell monolayers that had been infected with a
MOI of 10 for 36 hours. The complete nitrocellulose filter was moistened by the
monolayer of infected cells. A piece of 3MM Whatman filter paper saturated with 2x
SSC was placed on top of the wetted nitrocellulose. Pressure was applied to the filter
paper for 3 minutes. The nitrocellulose filter was then removed from the plate, dried at
room temperature and treated as before.
Preparation of DNA from colonies
Small scale preparation of DNA (miniprep) was performed according to a
procedure described by Zhou and colleagues (1990). Briefly, 1.5 ml of an overnight
culture was centrifuged at 10,000 x g for 1 minute and the supernatant was gently
decanted. The pellet was vortexed vigorously until completely resuspended in residual
fluid. Three hundred microliters of TENS buffer (TE buffer containing 0.1 N NaOH and
0.5% SDS) were added, vortexed for 5 seconds and 150 pi of 3.0 M sodium acetate (pH
5.2) were added. The mixture was again vortexed for 5 seconds and centrifuged for 2
minutes at 10,000 x g. The supernatant was transferred to a tube containing 0.9 ml of
ethanol that had been precooled to -20 C, mixed and centrifuged for two minutes. The

29
pellet was washed twice with 1 ml of 70% ethanol, dried under vacuum for ten minutes
and resuspended in 50 pi of dH20. Five microliters of resuspended DNA was analyzed
for purity and quantity by electrophoresis through a 1% agarose gel.
Large scale preparation of DNA
DNA was prepared in large quantities (maxiprep) from bacteria according to
published methods (Sambrook et al., 1989). A 500 ml overnight culture was centrifuged
at 4,000 x g for 10 minutes. Pelleted bacteria were resuspended in 12 ml of plasmid lysis
buffer, frozen at -85 C for 20 minutes, thawed and transferred to a Pyrex Erlemeyer
flask. One milliliter of freshly prepared lysozyme (Sigma) at a final concentration of 30
mg per ml was added to the flask. The flask was gently swirled over an open flame until
the solution began to boil. It was then immediately immersed in 300 ml of boiling water
for 40 seconds. The lysed bacteria were transferred into a 45 ml Nalgene tube and
chilled on ice for five minutes. Three milliliters of TE was added to the chilled solution
which was mixed well and centrifuged at 25,000 RPM (Beckman SW41) for 30 minutes
at 4C. The supernatant was decanted into a fresh 45 ml Nalgene tube containing an
equal volume of isopropanol, frozen at -85 C for 15 minutes, thawed and centrifuged
9,000 RPM for 20 minutes at 4 C. The supernatant was discarded and the pellet was
dried by inverting the tube onto filter paper. The dried pellet was dissolved in 4.25 ml
TE, and CsCl was added to a final concentration of 1.0 gm per ml. Ethidium bromide
(600 pg per ml) was added, and the solution was placed into 2 polyallomer Quick-Seal
centrifuge tubes (Beckman). The samples were centrifuged overnight in a vertical rotor
(VTi 65.2) at 45,000 RPM at 20 C. Two bands of DNA were visible after

30
centrifugation. An 18 gauge needle was inserted into the top of the tube to release
pressure and the lower band of supercoiled plasmid DNA was collected with a 21 gauge
needle and five ml syringe. The plasmid DNA was extracted by mixing with equal
volumes of water saturated butanol and discarding the upper layer until ethidium
bromide (pink color) could no longer be detected. After the extraction, DNA was
purified from residual CsCl by dilution with two volumes of dHzO followed by the
addition of six volumes of ethanol. The mixture was chilled at -85C for 20 minutes and
precipitated DNA recovered by centrifugation at 9,000 RPM for 20 minutes at 4C. The
supernatant was gently decanted and 400 pi of TNE buffer was used to resuspend the
DNA in the base of the tube. The TNE and DNA were transferred to an eppendorf tube
to which 1 ml of ice-cold ethanol was added. The tube was chilled at -85 C for 20
minutes and centrifugated at 10,000 x g for 10 minutes. The pelleted DNA was
resuspended in 100 pi of TE buffer.
Results
Sequencing of the pBSKS-BXLF2 which contained the BXLF2 cDNA cloned
into the EcoR I site of pBSKS confirmed that the BXLF2 cDNA had been oriented for
transcription under control of the T7 promotor (Figure 2-2). The plasmid was digested
with Smal and Hindi, two restriction endonuclease sites which flank the Eco R1 site, and
electrophoresed in 1% agarose in parallel with Hindlll digested Lamba phage DNA. A
band corresponding to the predicted 2.3 kb digestion fragment was detected (Figure 2-3).
Similar analysis of pSCl 1 digested with Smal revealed the 7.7 kb fragment that would be
expected from linearized plasmid (Figure 2-4). The Smal Hindi BXLF2 fragment was

31
ligated with the Smal digested pSCl 1 vector and used to transform E. coli strain JM109.
Colony blot hybridization of transformed bacteria with [32P] labeled Sma\ Hindi
BXLF2 indicated that thirteen of the two hundred colonies contained BXLF2 DNA
(Figure 2-5). Minipreps of DNA were made from two of these positive colonies.
Orientation of the insert within the pSCl 1 vector was analyzed with BamHI. Figure 2-6
depicts the size of digested fragments which would be expected after digestion of DNA
cloned in either orientation. When BXLF2 is in the correct orientation for transcription
under control of the pSCl 1 p7.5 promoter, fragments of 5.95, 3.2, 0.45 and 0.4 are
expected; in the opposite orientation, fragments of 4.45, 3.2, 1.9 and 0.45 should be
produced. Fragments obtained from two positive clones digested with BamHI indicated
that one (clone 45) was in the correct orientation for expression under control of the p7.5
promotor, and one (clone 25) was in the incorrect orientation (Figure 2-7).
To confirm these results, the digested DNA was electrophoresed in an agarose
gel, transferred to nitrocellulose using the Southern technique and hybridized with the
[32P] labeled Smal Hie II BXLF2 fragment (Figure 2-8). As expected, the probe
hybridized to the 4.45 and 1.9 kb fragments of clone 25 which contained BXLF2 in the
incorrect orientation (lanes 1 and 4). It hybridized to the 5.95 and 0.4 kb fragments of
clone 45 which contained BXLF2 in the correct orientation (lanes 2 and 5). Microgram
quantities of plasmid DNA were grown and purified from clone 45. This purified DNA
was used to transfect CV-1 cells infected with wild type vaccinia virus. Recombinant
virus was amplified in TK' cells in the presence BUdR, and virus expressing
P-galactosidase was plaque purified repeatedly in CV-1 overlayed with X-Gal. Virus

32
that produced greater than 90% blue plaques was obtained after 5-6 rounds of plaque
purification and was amplified as VVpSCl l-BXLF2a. DNA dot blots of this stock
confirmed that it still contained the BXLF2 sequence (Figure 2-9). The second vaccinia
virus recombinant, VVpEBl-BXLF2 was generated by inserting the same BXLF2
fragment, by the same protocol as used for VVpSCl l-BXLF2a, into the pEB-1 vector
linearized with Sma I (Figure 2-10). A DNA monolayer blot analysis was performed to
show that the BXLF2 sequence was contained within the vaccinia virus genome (Figure
2-11).
The third vaccinia virus recombinant VVpSCII-BXLF2b was made with BXLF2
DNA that had been amplified by PCR technology (Figure 2-12). The PCR product
(Figure 2-13) was blunt-ended and ligated into Smal digested pSCl 1 and Sma I digested
pEBl. The ligated DNA was used to transform E. coli and colony blot hybridizations
were done on resulting ampicillin resistant colonies. The probe was generated by
radiolabeling BXLF2 PCR products with [32P], DNA from two out of 100 pSC 11 clones
hybridized with labeled probe. None of the DNA from 100 pEBl clones hybridized with
the radiolabeled probe. DNA from both positive pSCl 1 clones was amplified in E. coli
and purified on CsCl gradients. The recombinant plasmids were tested for the
orientation of BXLF2 DNA within pSCl 1 using BamHI. The clone containing BXLF2
in the correct orientation was used to transfect CV-1 cells infected with wild type
vaccinia virus. Recombinant viruses were plaque purified and amplified.

33
UNIQUE
Smal
ATCTCGAGGATCCCCGGG
p11 p 11
FIGURE 2-1. Schematic diagrams of the vaccinia virus recombination vectors used in
the studies. pEBl contains pi 1 (vaccinia virus late promoter), Lac Z (beta galactosidase
from E.coli.), TK (thymidine kinase gene from vaccinia virus), and Ampr (ampicillin
resistance gene). The pSCl 1 vector contains both a pi 1 and p7.5 (vaccinia virus
early/late promoter region) as well as the Lac Z, TK, and Ampr genes.

34
FIGURE 2-2. Diagram of the pBSKS-BXLF2 plasmid with restriction sites contained
within the plasmid Multiple Cloning Site (MCS) and BXLF2 sequencing information
obtained.

35
1 2
23.1 kb-
9.42 kb-
6.56 kb-
4.37 kb-
2.32 kb-
2.03 kb-
FIGURE 2-3. 1% agarose gel containing ethidium bromide. Lane 1 contains the DNA
standard, Hindlll Lambda digest with resulting fragments of 23.1, 9.42, 6.56, 4.37,
2.32 and 2.03 kb. Lane 2 contains the electroeluted Smal Hie II digestion fragment of
pBSKS-BXLF2.

36
1 2
23.1 kb-
9.42 kb-
6.56 kb-
4.37 kb-
2.32 kb-
2.03 kb-
FIGURE 2-4. 1% agarose gel containing ethidium bromide. Lambda DNA standards
are in lane 1 and Sma\ digested pSCl 1 vector is in lane 2.

37
FIGURE 2-5. Colony blot of E. coli that was transformed with Sma I digested pSCl 1
and Sma I Hie II digested pBSKS-BXLF2. The colonies were hybridized with 32P
labeled Sma I Hie II digestion fragments of pBSKS-BXLF2.

38
BamHI
BamHI
FIGURE 2-6. Diagram of BamHI digested pSCl 1-BXLF2. Digestion fragments of 5.95,
3.2, 0.45 and 0.4 kB are seen when the BXLF2 sequence is ligated in the correct
orientation (A). Digestion fragments of 4.45, 3.2, 1.9 and 0.45 kB are seen when the
BXLF2 sequence is ligated in the incorrect orientation (B).

39
1 2 3
9.42 kb-
6.56 kb-
4.37 kb-
2.32 kb-
2.03 kb-
0.5 kb-
FIGURE 2-7. 1% agarose gel containing ethidium bromide. DNA BamHI digestion
patterns of pSCl 1 plasmid alone (lane 3), and of two clones positive by colony blot
hybridization. Lanes 1 (clone 25) and 2 (clone 45) contain the digestion patterns of
DNA from two positive clones.

40
23.1 kb-
9.42 kb-
6.56 kb-
4.37 kb-
2.32 kb-
2.03 kb-
0.5 kb-
FIGURE 2-8. Southern blot of BamHl digested pSCl 1 (lanes 3 and 6), pSCl 1-BXLF2
clone #25 (lanes 1 and 4) and pSCl 1-BXLF2 clone #45 (lanes 2 and 5). The blot was
hybridized with 32P labeled Sma I Hie II digestion fragments of pBSKS-BXLF2 DNA.

41
Sample # 1
#
Sample #2
FIGURE 2-9. Dot blot in which sample #1 contained DNA from uninfected CV-1 cells,
sample #2 contained DNA from CV-1 cells infected with a vaccinia virus recombinant
expressing the EBV membrane glycoprotein, gp55/75 and sample #3 contained DNA
from CV-1 cells infected with VVpSCl l-BXLF2a. The blot was hybridized with [32P]
labeled Sma I Hie II digestion fragments of pBSKS-BXLF2.

42
1 2
kb
kb
kb
kb
kb
kb
FIGURE 2-10. 1% agarose gel containing ethidium bromide. Lane 1 contains Sma I
linearized pEBl vector. Lane 2 contains Hindlll digested Lambda DNA standard.

43
FIGURE 2-11. DNA blot of CV-1 monolayers. Sample #1 contained DNA lifts from
approximately 1 x 106 uninfected CV-1 cells. Sample #2 contained DNA lifts from
CV-1 cells infected with VVpEBl-BXLF2. The blot was hybridized with 32P labeled
Sma I Hie II digestion fragments of pBSKS-BXLF2.

44
Primer B
Primer A: 5 CTCG AG AT GC AGTT GCT CT GT GTTTTTT GC 3
Primer B: 3 CTCGAGAAGGAAAAACATAACAATCTTGTG 3
FIGURE 2-12. Diagram of DNA sequences achieved by using PCR. These sequences
were subsequently used to make the third vaccinia virus shuttle vector.

45
1
2
-23.1
kb
-9.42
kb
-6.56
kb
-4.37
kb
-2.32
kb
-2.03
kb
FIGURE 2-13. 1% agarose gel containing ethidium bromide. Lane 1 contains 10 pi of
PCR product with BXLF2 DNA. Lane 2 contains Hindlll digested Lambda DNA.

46
Discussion
The pSCl 1 insertion vector was the first vector available to our laboratory, thus,
the first vaccinia virus recombinant (VVpSCl l-BXLF2a) was made using pSCl 1. The
pEBl insertion vector became available shortly after the generation of this vaccinia virus
recombinant. The substitution of the late pi 1 promoter for the early/late p7.5 promoter
was described as expressing greater amounts of the inserted foreign gene (personal
communication, Dr. Edward Stephens). However, preliminary analyses of proteins
expressed by VVpSCl l-BXLF2a and VVpEBl-BXLF2 did not reveal any glycoprotein
with the characteristics of gp85. It was hypothesized that the presence of 226 base pairs
(bp) upstream of the BXLF2 start site that had been cloned from the cDNA into the
vaccinia virus genomes were interfering with translation. To allay these concerns, we
utilized PCR technology to generate a third recombinant vaccinia virus clone
(VVpSCl l-BXLF2b). Primers were designed such that these upstream 226 bp were
deleted.
Restriction endonuclease analysis of recombinant shuttle vector DNA indicated
that the BXLF2 sequence was of the correct size and orientation. Furthermore, data
from dot blots and monoblots revealed that the BXLF2 sequence was contained within
the vaccinia virus genome. At this point in our studies we decided that northern blot
analysis of the vaccinia virus mRNAs would not yield useful information because of
reports in the literature describing the mRNA's transcribed from late genes as being both
long and heterogeneous in size. For this reason, we decided to continue our analysis of

47
recombinant gp85 at the protein level. The vaccinia virus expression system has been
used to make properly processed and transported proteins that contain the appropriate
posttranslational modifications (Moss and Earl, 1989). The phosphorylation, N- and
O-glycosylation, myristylation, cleavage and assembly of proteins made using the
vaccinia virus expression system has been reported to occur in an apparently faithful
manner.

CHAPTER 3
EXPRESSION AND BIOCHEMICAL CHARACTERIZATION OF
RECOMBINANT gp85
Introduction
The BXLF2 open reading frame is predicted to encode for a polypeptide with a
Mr of 78,000 ( Baer et al., 1984). As previously described, computer analysis of the
BXLF2 sequence revealed three highly hydrophobic regions (Klein et al., 1985). One of
these regions is located at the amino terminus from amino acids 3-19. This region may
represent a signal sequence of the protein. A second hydrophobic region is at the
carboxy terminus, and spans from amino acids 680-696. This region may serve as a
membrane anchor. The third hydrophobic domain stretches from amino acids 538-554
and may either be a second anchor sequence or possibly a fusion peptide similar to that
found in fusion proteins of ortho and paramyxoviruses (White et al., 1983). Analysis of
the gp85 open reading frame has also revealed five potential N-linked glycosylation sites
(Baeretal., 1984)
The addition of asparagine or N-linked sugars to cell and virus glycoproteins
begins as the precursor oligosaccharide Glc3Man9GlcNAc2 and is transferred to the
nascent polypeptide chain of the glycoprotein from the carrier lipid, dolichol phosphate.
The transfer occurs on the luminal side of the endoplasmic reticulum (Komfeld and
Kornfeld, 1985). This first step is followed by a number of processing and trimming
48

49
reactions which begin in the endoplasmic reticulum and continue as the protein is
transported through the Golgi apparatus to its final destination (Figure 3-1). Soon after
transfer of the sugar from dolichol phosphate, three glucose residues may be removed by
the activity of two membrane bound glucosidases, a-glucosidase I and a-glucosidase II.
Glucosidase I removes the terminal alpha 1,2 glucose and glucosidase II removes the
next two alpha 1,3 glucose residues to produce an oligosaccharide chain with the
structure Man9GlcNAc2. This Man9GlcNAc2 molecule is referred to as the "core"
region. Deoxynojirimycin (DNM) inhibits the activity of glucosidase I, thus preventing
the removal of all three terminal glucose residues. If formed, Man9GlcNAc2 can be
processed by two distinct pathways. Failure to remove mannose residues can result in
production of the high mannose oligosaccharide Man8GlcNAc2 or Man5GlcNAc2. In
contrast, the removal of four mannose residues by mannosidase I, an enzyme found in
the Golgi, paves the way for the addition of complex N-acetylglucosamine residues to
Man5GlcNAc2. Furthermore, the molecule is now sensitive to removal of two more
mannose residues with mannosidase II. The action of mannosidase II is inhibited by
swainsonine. Finally, addition of peripheral sugar residues such as N-acetylglucosamine,
galactose, N-acetylneuraminic acid, fucose and sialic acid occurs by specific
glycosyltransferases. The addition of N-acetylglucosamine to Man5GlcNAc2 renders the
glycoprotein resistant to removal of sugar side chains by endoglycosidase H (endo H),
although they remain sensitive to the action of endoglycosidase F. Since this step occurs
early in processing in the Golgi the relative sensitivity of the oligosaccharides on a

50
glycoprotein to digestion with endo H and endo F is commonly used to monitor
glycoprotein transport.
The transport of the amino-linked oligosaccharide from its site of synthesis in the
endoplasmic reticulum through the cis, medial and trans Golgi stacks can be inhibited by
ionophores. Concentration of ionophores that inhibit transport and hence terminal
glycosylation reactions do not affect the initial incorporation of amino acids into protein.
The ionophore monensin is thought to deplete Ca2+ levels, rendering the Golgi vesicles
unable to fuse. Thus, vesicular transport of molecules between the endoplasmic
reticulum and the Golgi is perturbed (Uchida et al., 1979). Newly synthesized proteins
accumulate within intracellular vacuoles and post-translational modifications are
blocked.
This chapter describes the expression and biosynthesis of the EBV recombinant
gp85 molecule.
Materials and Methods
Cells and virus
CV-1 cells were grown in DMEM supplemented with 5% FBS, 100 IU of
penicillin and 100 pg of streptomycin per ml. Stocks of VVpSCl l-BXLF2a,
VVpEBl-BXLF2 and VVpSCl l-BXLF2b were prepared as described in Chapter 2.
Antibodies
Two monoclonal antibodies and one rabbit anti-peptide antibody were used. The
monoclonal F-2-1 (Strnad et al., 1982) and monoclonal E1D1 (Balachandran, 1987) both

51
recognize native gp85. The anti-peptide antibody (anti-BX), was made to a synthetic
peptide corresponding to residues 518 to 528 of the EB V BXLF2 ORF (Oba and
Hutt-Fletcher, 1988). This sequence is located amino terminal to the internal
hydrophobic region. The peptide, cys-ser-leu-glu-arg-glu-asp-arg-asp-ala-
trp-his-leu-pro-ala-tyr-lys, (V-l) was synthesized by the Protein Chemistry Core Facility
of the University of Florida. The amino terminal cysteine residue was added to the EBV
sequence to facilitate coupling to keyhole limpet hemocyanin (Liu et al., 1979).
Rabbits were immunized initially with a subcutaneous injection containing one
mg of peptide conjugated to keyhole limpet hemocyanin and emulsified in Freunds
complete adjuvant. The rabbits were subsequently immunized at biweekly intervals,
with the same amounts of conjugated peptide emulsified in Freunds incomplete adjuvant.
After approximately two biweekly injections, the rabbits were bled from the central ear
artery. Prior to bleeding, the rabbits had been anesthetized for approximately 30 minutes
with 0.1 ml per kg of Innovarvet (fentanyl 0.04% and droperidol 2%). The rabbit was
then routinely immunized and bled between every two injections.
Antibody purification
All antibodies were purified by chromatography on protein A that had been
coupled to Sepharose CL-4B (Sigma). Approximately 100 mis of concentrated
hybridoma culture supernatant or approximately 25 mis of serum were filtered through a
0.45 urn filter and added to a bed of protein-A sepharose. The 20 ml bed volume was
washed with approximately 80 mis of phosphate buffered saline (PBS) at 4 C. The
antibody was adsorbed to the column and subsequently eluted with 0.1 M acetic acid in

52
PBS. The pH of the eluate was immediately equilibrated with 1.5 M Tris-HCl (pH 8.8)
and the purified antibody was dialyzed against 16 liters of PBS at 4C. Dialyzed
antibody was concentrated by dialysis against polyethylene glycol (Sigma) to one tenth
of the volume of the serum or culture supernatant from which it was purified.
Anti-BX was further purified by affinity chromatography on peptide that had
been coupled to cyanogen-bromide activated Sepharose CL-4B (Oba and Hutt-Fletcher,
1988). BXLF2 peptide (7 mg/ml in PBS) was added to cyanogen bromide activated
Sepharose (CNBr-Sepharose; Sigma, St. Louis, MO) in 0.2 M carbonate/bicarbonate
buffer (pH 9.6) and mixed for a minimum of four hours at 4C. Coupled gel was then
pelleted by centrifugation and the supernatant was stored for analysis of both protein
concentration (Lowry et al., 1951) and coupling efficiency. The residual active groups
on the Sepharose were blocked with ethanolamine (1M, pH 8.0) before washing beads
extensively with PBS. Subsequent peptide columns were made using Affi-Gel-15
(Bio-Rad Laboratories) instead of CNBr-Sepharose. The use of Affigel 15 dispenses
with cyanogen bromide activation steps. Affigel spontaneously forms stable covalent
bonds with the primary amines of proteins that have isoelectric points below 6.5.
Columns were made with bed volumes of approximately five mis.
Enzyme-linked immunosorbant assay
Reactivity of rabbit antisera with the V-l peptide was measured in ELISA assays
(Oba, D. E. Ph.D. Thesis, University of Florida, 1988). Fifty microliters of peptide at a
concentration of 6 jug/ml in carbonate bicarbonate buffer (pH 9.6) were added to each
well of a 96-well microtiter plate and incubated overnight at 37 C (Voller et al., 1976).

53
Plates were subsequently incubated with 5% skimmed milk in PBS-Tween 20 to block
binding surfaces that remained on the polystyrene plate after antigen absorption.
Blocked plates were then sequentially incubated with dilutions of rabbit antibody, with
goat anti-rabbit antibody conjugated to horseradish peroxidase (Organon Teknika Corp.,
Durham, N.C.) and with the chromogenic substrate orthophenylenediamine (Sigma).
The bound enzyme converts the substrate to a colored product at rates proportional to the
amount of primary antibody present in each well. The final enzyme reaction was stopped
with 2 N H2S04. Color change was analyzed at 492 nm using a spectrophotometric
ELISA plate reader. Plates were washed between each incubation by flooding the wells
three times with 0.05% Tween 20 in 0.85% NaCl. The Tween-20 solution reduces
nonspecific interactions by removing unbound or weakly adsorbed molecules.
Analysis of vaccinia proteins
Six well plates containing 106 CV-1 per well were infected with wild type
vaccinia virus, with VVpSCl l-BXLF2b or with VVpEBl-BXLF2 at an MOI of 10 at
37C. At one hour post infection, the viral inoculum was removed and replaced with 1
ml of methionine deficient medium. Two hours later cells were labeled for 20 hours with
100 jaCi [35S] methionine. Supernatant medium was replaced with
radioimmunoprecipitation buffer (RIPA; 0.05 M Tris-HCl pH 7.2, 0.15 M NaCl, 1.0%
deoxycholate, 1.0% Triton X-100, 0.1% sodium dodecyl sulfate, 100 U aprotinin per ml
and 0.1 mM phenylmethylsulfonylfluoride) and cells were held on ice for thirty minutes.
At thirty minutes, the solubilized cells were transferred to eppendorf tubes and vortexed
vigorously for five minutes. The vortexed samples were centrifuged at 10,000 x g for

54
fifteen minutes and the supernatant was immunoprecipitated for 20 hours at 4C with
100 pg of antibody and protein A-agarose beads. Immunoprecipitated complexes were
washed 3 times with RIPA. The proteins were dissociated by boiling for 10 minutes in
sample buffer (0.625 M Tris (pH 8.8), 1% SDS, 10% glycerol, and 1%
2-mercaptoethanol) and analyzed by gel electrophoresis after a five pi aliquot of each
sample was counted to determine the amount of radiolabeled protein immuno
precipitated.
Electrophoresis of proteins
Immunoprecipitated samples were analyzed by SDS-polyacrylamide gel
electrophoresis (SDS-PAGE; Laemmli, 1970). Polyacrylamide gel separating buffer was
made with 0.65 M Tris and 7.5, 9.0 or 12.5% acrylamide, 0.1% SDS, 0.1% glycerol,
0.83% ammonium persulfate, and 0.28% N,N'diallyltartardiamide (DATD) or
bis-acrylamide (BIS). The separating gel was overlayed with 1-butanol and allowed to
polymerize for a minimum of five hours. After polymerization, the gel was rinsed with
distilled water and a stacking gel was added. The stacking gel was made with 0.65 M
Tris (pH 6.8), 4% acrylamide, 0.1% SDS, 0.05% glycerol, 0.83% ammonium persulfate,
and 0.28% DATD and allowed to polymerize for a minimum of 45 minutes before
samples were loaded. The gel was electrophoresed for approximately six hours at a
constant current of 26 mAmps or overnight at a constant voltage of 50 volts. Gels were
stained with 0.1% Coomassie blue stain for ten minutes and destained with 7% methanol
and 7% acetic acid. Gels containing proteins labeled with l2:>I were directly dried on
filter paper. Gels containing [35S] or [3H] labeled proteins were either infused for 30

55
minutes in two changes of dimethyl sulfoxide (DMSO), for 3 hours in 22%
2,5-diphenyloxazole (PPO) in DMSO and for 1 hour with water or they were infused
with Amplify (Amersham) for 30 minutes. All dried gels were exposed to XAR film at
-85C for fluorography (Bonner and Laskey, 1974).
Enzyme and glvcosvlation inhibitors
For both digestion with endoglycosidase H (endo H; Boehringer Mannheim
Corp) and for digestion with endoglycosidase F (endo F; Boehringer Mannheim Corp.)
immunoprecipitated protein-antibody-bead complexes were washed twice with 1 ml of
RIPA. For endo H digestion, the pelleted complexes were resuspended in 0.01 M sodium
citrate (pH 5.5), and 0.01 U of enzyme in a total volume of 100 pi for 20 hours at 37C.
The enzyme reaction was stopped with the addition of 2x SDS-PAGE sample buffer
containing 2-mercaptoethanol. For digestion with endo F, samples were washed using
the same procedure as used for endo H digestion and resuspended in 0.1 M sodium
acetate (pH 5.0) containing 50 mM EDTA, 1% NP-40, 0.1% SDS, 1%
2-mercaptoethanol and 2 U of endo F for 20 hours at 37C. The reaction was stopped
after 20 hours with the addition of 100 pi of 2x SDS-PAGE sample buffer without
2-mercaptoethanol. Both endo H and endo F digested samples were boiled for ten
minutes and analyzed by SDS-PAGE gel under reducing conditions. Monensin (Sigma
Chemical Co.) and stocks prepared in methanol. Swainsonine and 1-deoxynojirimycin
(DNM) were obtained from Boehringer-Mannheim Biochemicals. The swainsonine and
DNM were freshly prepared in DMEM without methionine and filter sterilized just prior
to their use. Swainsonine was used at 300 ng per ml, monensin was used at

56
concentrations of 0.001 to 100 pM and DNM was used at a 1 mM concentration.
Inhibitors were added to the initial infection medium. At 3 hours post infection, the viral
inoculum containing inhibitors was removed. New inhibitors were added along with 100
pCi of [35S] methionine. The radiolabeled, infected cells were incubated for 20 hours at
37C and harvested and immunoprecipitated as previously described.
Silver staining
Proteins were visualized in gels by silver staining according to a modification of
standard protocols (Heideshoven and Dernick, 1985; Merril et al., 1979). Briefly, gels
were fixed with 30% ethanol and 10% acetic acid for one hour followed by two ten
minute washes in 10% ethanol. The gels were then reduced in 0.5% Farmer's solution
(2.27 mM sodium thiosulfate, and 60.7 mM potassium ferricyanide) for approximately
15 seconds or until the gel appeared slightly yellow. The gel was then immediately
washed four times with distilled water for ten minutes, soaked for 15 minutes in 0.1%
silver nitrate and washed in running distilled water for 20 seconds. The silver
impregnated gel was developed with 2.5% sodium bicarbonate, and 0.02% formaldehyde
until the background of the gel began to change color. The reaction was stopped by
washing in distilled water.
Results
Expression of gp85 by recombinant vaccinia virus was first examined in CV-1
cells (Figure 3-2). The anti-peptide antibody, anti-BX, was used to immunoprecipitate
proteins from mock infected (lane 1), wild type vaccinia virus infected (lane 3),

57
VVpSCl l-BXLF2b infected (lane 2) or VVpEBl-BXLF2 (lane 4) infected. Bands with
an Mr of 85,000 were immunoprecipitated from cells infected with either construct, but
not from uninfected cells or cells infected with wild type vaccinia virus. To determine
the quantity of recombinant protein expressed, cell lysates immunoprecipitated with
anti-BX were electrophoresed in 7.5 % acrylamide cross-linked with BIS-acrylamide and
silver stained (Figure 3-3). Comparison of the intensity of staining of gp85 and known
amounts of molecular weight markers run in parallel indicated that 1.6 x 105 recombinant
infected cells expressed over 500 ng of the recombinant protein. However, further
analysis revealed that although recombinant gp85 made by VVpEBl-BXLF2 or
VVpSCl l-BXLF2b was immunoprecipitated by anti-peptide antibody anti-BX, it was
not immunoprecipitated by either of the monoclonal antibodies F-2-1 or E1D1 that
recognize native gp85 made in B lymphocytes (Figure 3-4).
To examine whether the recombinant protein was properly processed within the
endoplasmic reticulum, CV-1 cells were infected with either wild type vaccinia virus or
VVpEBl-BXLF2 in the presence of the amino linked glycosylation inhibitor, DNM.
Treatment with DNM increased the apparent Mr by 2,500. (Figure 3-5, lane 5). This
result is consistent with the inhibition of the action of trimming enzymes glycosidase I
and II within the endoplasmic reticulum. The addition of complex sugars which are
resistant to digestion with endo H, but not endo F, occurs only if glycoproteins are
transported out of the endoplasmic reticulum into the Golgi apparatus. To determine if
recombinant gp85 was being properly transported through the Golgi apparatus, a
comparison was made of the sensitivity of the molecules to digestion by the enzymes

58
endo H and F. Both endo H and F reduced the mass of the recombinant from 85,000 to
72,000 (Figure 3-6). Thus, there appeared to be a block in transport from the
endoplasmic reticulum through the Golgi. To test this hypothesis further, the processing
of recombinant gp85 was analyzed in cells treated with swainsonine and monensin,
which would be expected to inhibit processing steps within the Golgi (Figure 3-5). The
mobility of the recombinant molecule was not altered by either of these drugs.

Glycoprotein Processing
Nascent Glycoprotein
Cleaving Enzyme
Protein-Asn-A-A
Glucosidase 1
Protein-Asn-A-A
f
Protein-Asn-A-A
Glucosidase II
Glucosidase II
High Mannose Type Protein-Asn-A-A
Protein-Asn-A-A
-UDP-GlcNAc
Protein-Asn-A-A
Mannosidase I
(a-1,2-Gnkages)
Mannosidase II
Protein-Asn-A-A
-UOP-GicNAc
UDP-Gal
CMP-SA
AGal SA
Complex Type Pro;ein-Asn-.i-A
X A Gol SA
Asn
SA
Gal
Inhibitor
of Cleavage
1 -Deoxynojirinrvfdn
1 -Oeoxynojirimycin
1 -Oeoxynojirimycin
Swainsonine
N-acetylglucosamine
Glucose
Mannose
Asparagine
Sialic Ac:d
Galactose
FIGURE 3-1. Diagram of where inhibitors of trimming enzymes are active

60
FIGURE 3-2. SDS-PAGE analysis of recombinant EBV glycoprotein gp85. CV-1 cells
were mock infected (lane 1) or infected at an MOI of 10 with wild type vaccinia virus
strain IHD-J (lane 3), vaccinia virus construct VVpSCl l-BXLF2b (lane 2) or vaccinia
virus construct VVpEBl-BXLF2 (lane 4). Infected cells were radiolabeled with 75 pCi
of [35S]methionine at three hours post infection and harvested at 24 hours post infection.
Radiolabeled samples were immunoprecipitated with anti-BX and analyzed by
SDS-PAGE under reducing conditions. Results were visualized by standard
fluorographic techniques.

61
1 2 3
Figure 3-3. Silver stain of recombinant gp85. Molecular weight markers were run in
lane 1. CV-1 cells were infected with VVpEBl-BXLF2 (lane 2) or with wild type
vaccinia virus (lane 3). Cell lysates were immunoprecipitated with anti-BX and analyzed
on a 9% SDS-PAGE gel under reducing conditions.

62
97.5 Kd-
69.0 Kd-
45.0 Kd-
30.0 Kd-
FIGURE 3-4. SDS-PAGE analysis of recombinant gp85. CV-1 cells were infected with
the vaccinia construct VVpEBl-BXLF2 in lanes 1, 3 and 5, while CV-1 cells in lanes 2,
4 and 6 were infected with wild type vaccinia virus. Infected cells were radiolabeled
with [35 S] methionine and immunoprecipitated with anti-BX (lanes 1 and 2), with
monoclonal F-2-1 (lanes 3 and 4) or with monoclonal E1D1 (lanes 5 and 6). Samples
were analyzed under reducing conditions by SDS-PAGE and visualized by standard
fluorographic techniques.

63
FIGURE 3-5. Synthesis of recombinant gp85 in the presence of swainsonine, monensin
or DNM. CV-1 cells were infected with either VVpEBl-BXLF2 vaccina virus construct
(lanes 1 through 6), or wild type vaccinia virus strain IHDJ (lane 7). The infected cells
were radiolabeled with 75 pCi of [35S] methionine for 20 hours in methionine deficient
medium and 0.001 pM monensin (lane 1), 0.1 pM monensin (lane 2), 100 pM monensin
(lane 3), 300 ng per ml of swainsonine (lane 4), 1 mM DNM (lane 5) or left untreated
(lanes 6 and 7). The radiolabeled cell lysates were immunoprecipitated with anti-BX
antibody. Samples were analyzed under reducing conditions by SDS-PAGE and
visualized by standard fluorographic techniques.

64
FIGURE 3-6. Effects of endoglycosidases on recombinant gp85. CV-1 cells were
infected with either wild type vaccinia virus strain IHD-J (lanes 1 through 4), the
VVpSCl l-BXLF2b construct (lanes 5 through 8), or the VVpEBl-BXLF2 construct
(lanes 9 through 12). The infected cells were radiolabeled with 75 pCi of [35S]
methionine for 20 hours. The radiolabeled cell lysates were immunoprecipitated with
anti-BX and treated with endo F (lanes 3, 7 and 11), with endo F buffers (lanes 4, 8, and
12) with endo H (lanes 2, 6 and 10) or with endo H buffers alone (lanes 1, 5, and 9). All
samples were analyzed under reducing conditions by SDS-PAGE and visualized by
standard fluorographic techniques.

65
Discussion
We initially chose to express the BXLF2 gene in vaccinia virus both because we
hoped that this vector might express larger amounts of protein and because we would be
able to use such a vector to express protein in lymphoblastoid cell lines. The vaccinia
constructs that we made fulfilled our expectations as to quantities of protein expressed.
Furthermore, the protein had properties consistent with its predicted structure as the
apparent Mr of the molecule was compatible with the use of all five N-linked
glycosylation sites. However, we were able to detect recombinant protein only with the
antibody made to a peptide derived from the BXLF2 sequence (anti-BX) and not with
either of two monoclonal antibodies made to native protein. This suggested that the
recombinant was antigenically different than the native molecule. We decided to
examine whether this difference was caused by differential glycoprotein processing of
the two molecules.
The enzyme inhibitor, DNM, inhibits the removal of three terminal glucose
residues from the oligosaccharide side chains transferred contranslationally from the
carrier dolichol to the nascent peptide chain. The increased mobility of recombinant
gp85 immunoprecipitated from cells treated with DNM suggested that this first step in
glycoprotein processing, which occurs in the endoplasmic reticulum, was occurring
normally. The recombinant protein was, however, equally sensitive to digestion with the
endoglyclosidases H and F both of which reduced the apparent Mr of the molecule to
72,000, a mobility consistent with that predicted for the non-glycosylated core protein
(Edson and Thorley-Lawson, 1983). This suggests that addition of N-acetylglucosamine

66
to the Man5GlcNAc2 had not occurred in the Golgi. The results of swainsonine and
monensin treatment also suggest that the recombinant gp85 undergoes none of the
processing reactions known to take place in the Golgi. The data are consistent with the
hypothesis that recombinant gp85 molecule is blocked in transport from the endoplasmic
reticulum through the Golgi stacks.
Further examination of this hypothesis required more precise comparison of the
processing of recombinant and native gp85.

CHAPTER 4
COMPARISON OF NATIVE AND RECOMBINANT gp85
Introduction
The native gp85 molecule is partially sensitive to digestion with endoglycosidase
H suggesting that it carries both complex and high-mannose type oligosaccharides
(Edson and Thorley-Lawson, 1983). Endoglycosidase H digestion of gp85 from virus
producing cells revealed five species ranging in apparent Mr from 86,000 to 71,000,
while a protein with an apparent Mr of 69,000 was produced in cells treated with
tunicamycin (Stmad et al, 1983), an inhibitor which interferes with the transfer of
oligosaccharides to dolichol phosphate (Heifetz et al., 1979). This is consistent with the
use of all five potential N-linked glycosylation sites in the gp85 sequence and suggests
that the molecule carries no O-linked sugars.
Since EBV preferentially establishes latent infections as opposed to productive
infections in lymphocytes, virus yields are often low. Approximately 5% of cells in EBV
containing lymphoblastoid cell lines will spontaneously produce virus. Treatment of
cells carrying the P3HR1 or the B95-8 strains of virus with the tumor promoter
12-O-tetradecanoyl phorbol-13-acetate (TPA) (Zur Hausen et al., 1978), increases the
number of virus producing cells over a period of several days. However, a third strain of
virus, Akata, (Takada and Ono, 1984) has been derived from an EBV genome-positive
67

68
Burkitt lymphoma cell line in which virus replication can be induced by addition of
anti-human immunoglobulin G. Induction of virus replication by this technique is much
faster and occurs in a larger percentage of cells. A detectable increase is seen in the
number of cells positive for the EBV specific early antigen (EA) (Henle et al, 1970) by
three hours after addition of anti-immunoglobulin. By six hours of induction many of
the cells test positive for EA using indirect immunofluorescence assays. Takada and
Ono (1989) demonstrated that at nine hours post anti-immunoglobulin treatment, 40% of
Akata cells contain viral capsid antigen (VCA) (Henle and Henle, 1966). The percentage
of VCA positive cells increase until a plateau of 60% is reached at 12 hours. In contrast,
after 24 hours of observation, fewer than 0.5% of nontreated cells contained EA and
VCA.
The almost synchronous induction of EBV replication in Akata cells makes them
ideal for comparative studies of processing of native and recombinant gp85. This
chapter describes a detailed analysis of expression of recombinant and native gp85 in
Akata cells. The drug Brefeldin A (BFA) was included in the study. This antiviral
antibiotic inhibits post-translational processing of glycoproteins at a later stage than
monensin (Collins and Mottet, 1991). It blocks anterograde transport from the
endoplasmic reticulum, but, unlike monensin, it allows retrograde transport to continue.
This has the effect of causing a backflow of Golgi cisternae into the endoplasmic
reticulum or an intermediate endoplasmic reticulum-Golgi compartment (Collins and
Mottet, 1991; Dorns et al., 1989; Lippincott-Schwartz et al., 1990).

69
Materials and Methods
Cells and viral induction
Three lymphoblastoid cell lines were used, BJAB, an EBV negative Burkitt's
lymphoma line, B95-8, an EBV producing marmoset cell line ( Baer et al., 1984) and
Akata, an EBV producing Burkitt's lymphoma cell line (Takada and Ono, 1989; gift of
Dr. John Sixbey, St Jude's Childrens Research Hospital). Cells were grown in RPM1
1640 supplemented with 10% FBS, 100 units of penicillin and 100 pg of streptomycin
per ml and diluted biweekly. Virus replication occurred spontaneously in B95-8 cells
and was induced in Akata cells. Akata cells with a viability greater than 85% as judged
by trypan blue exclusion were centrifuged at 400 x g for five minutes at 4C. The
supernatant was discarded and resuspended in fresh medium at a concentration of 106
viable cells per ml. Latent EBV was induced into the lytic cycle by the addition of 100
pg per ml of anti-human immunoglobulin G, F(ab')2 fragment (Cappel, Organon Teknika
Corporation).
Radiolabeling of native gp85 in Akata cells
At various times after induction, the Akata cells were resuspended in either
methionine free RPMI or RPMI containing 1/10 concentration of glucose. Two hours
later 100 pCi [35S] methionine or 100 pCi [3H] glucosamine were added per ml. Cells
were either kept in radioactive medium until harvested or washed and resuspended in
radioactive-free medium containing 100 x methionine. Cells were harvested by
centrifugation and solubilized in RIPA as previously described for recombinant infected

70
CV-1 cells. Immunoprecipitated proteins were digested with endoglycosidases as
previously described. The inhibitors, DNM (ImM) and swainsonine (300 ng per ml)
were added to Akata cells at six hours post induction. Inhibitors of glycoprotein
transport, monensin and BFA, were added to suspensions of Akata cells at six hours post
induction. Final concentrations of ImM to 10 pM monensin and 2 pg per ml of BFA
were used in these experiments.
Extrinsic labeling of cells
One million Akata cells were induced with anti-immunoglobulin for 24 hours.
One million BJAB cells were infected at a MOI of 10 with either wild type vaccinia virus
strain 1HD-J or the VVpSCl l-BXLF2b construct. At either one hour post infection, or
24 hours post induction cells were centrifuged at 400 x g for five minutes, and
resuspended in 1 ml of serum free RPM1. Seven hours later the cells were centrifuged
and resuspended in 1/10 glucose media. The samples were subsequently incubated for
two hours at 37lC. Those samples that were to be labeled with 1251 were washed twice
and resuspended in 300 pi ice cold PBS. The 300 pi samples were transferred to tubes
containing one Iodobead (Pierce Chemical Company) and incubated at 25 C for 30
minutes. The Iodobead had previously been incubated with 300 pCi of 125I for five
minutes. After the incubation period, the cells were washed six times with 15 mis of
PBS. Cells were solubilized as previously described in 1 ml of RIPA for
immunoprecipitation and analysis by SDS-PAGE. Efficiency of labeling was determined
by comparing total radioactivity in 10 pi samples with the amount of radioactivity
incorporated into material precipitated by 5% trichloroacetic acid.

71
Antibodies
Two monoclonal antibodies were used, 72A1, which reacts with gp350/220
(Hoffman et al., 1980) and F-2-1, which reacts with gp85 (Strnad et al., 1982). One
polyclonal antibody, anti-BX, was used which reacts with gp85 (Oba and Hutt-Fletcher,
1988). All the antibodies were purified by chromatography on protein A-agarose and
anti-BX antibody was further purified on peptide coupled to Affigel as previously
described.
Immunofluorescence
For cytoplasmic staining, slides bearing cells that had been air-dried and fixed for
10 minutes in acetone at -20 C were incubated with antibody in a humidified
atmosphere at 37C for 35 minutes. Rabbit anti-peptide antibody, anti-BX, was used at
a concentration of 720 pg per ml, and monoclonal antibody 72A1 was used at a
concentration of 10 pg per ml. Cells were washed three times with PBS and reincubated
for 35 minutes at 37 C with the appropriate dilution of fluorescein
isothiocyante-conjugated goat anti-rabbit or rabbit anti-mouse immunoglobulin. The
cells were washed three times and mounted in a solution containing 50% PBS and 50%
glycerol.
For surface staining, 2 x 106 cells were fixed briefly with ice cold 0.1%
paraformadehyde in PBS. The fixed cells were reacted sequentially with rabbit
anti-peptide antibody and fluorescein conjugated goat anti-rabbit serum. Cells were
washed three times between incubations by centrifugation in ice cold PBS and three
times before being mounted on slides for examination with a fluorescence microscope.

72
Results
We initially wanted to compare the expression and biosynthesis of [35S]
methionine labeled recombinant gp85 with that of [35S] methionine labeled native gp85.
We had successfully radiolabeled the recombinant with [35S] methionine as described in
Chapter 3. However, we had not been able to achieve such high levels of [35S]
methionine incorporation into the native molecule. In contrast, using the Akata cell line
which we received from Dr. Sixbey enabled us to accomplish this goal (data not shown).
Before using the Akata cell line to analyze expression and biosynthesis of native gp85,
we needed to demonstrate that uninduced Akata cells would not also produce gp85. To
accomplish this, we compared the gp85 synthesized in uninduced Akata cells with that
produced in the induced Akata cells (Figure 4-1). Both samples were radiolabeled with
[35S] methionine and immunoprecipitated with the monoclonal antibody F-2-1. As
expected, F-2-1 immunoprecipitated a protein with an apparent Mr of 85,000 from
induced Akata cell lysates. No similar size species was detected in the uninduced Akata
sample.
We began analyzing the biosynthesis of native gp85 by determining the kinetics
of its synthesis. Synthesis of gp85 was first detected at 6.5 hours post induction (Figure
4-2, lane 2) and high levels of labeled protein were seen by eleven hours (lane 6 8).
Native gp85 appeared as a dimer in these experiments. The dimer formation was seen
particularly well in lysates of cells that had been labeled for 15 minutes at 8 hours after
induction (lane 5).

73
To examine processing of the native protein within the endoplasmic reticulum,
induced Akata cells were labeled in the presence of the glucosidase I inhibitor, DNM.
As expected, treatment with DNM increased the apparent Mr of gp85 by approximately
2.5 K (Figure 4-3, lane 2). We next wanted to determine whether native gp85 contained
any endo H resistant oligosaccharides which would be indicative of transport from the
endoplasmic reticulum to the Golgi apparatus. A comparison was made of the ability of
endo H and F to remove sugars from the protein (Figure 4-4). Repeated analysis of
native gp85 revealed a clear difference between the mobility of the endo H and the endo
F digested protein. The lysates that had been digested with endo F precipitated protein
species with a Mr of 74,000, while those digested with endo H contained
immunoprecipitated proteins of 79,000. Since lysates of virus producing cells contain
glycoproteins at different stages in processing and might have biased the results of this
experiment, we also examined the susceptibility of mature gp85 present in the virion
(Figure 4-5). No significant differences were seen in the digestion patterns of protein
immunoprecipitated from mature virions or from infected cells.
We next analyzed biosynthesis of native gp85 in the presence of swainsonine,
which inhibits mannosidase II (Figure 4-3). Native gp85 appeared less heterogeneous
when made in the presence of swainsonine. Inhibitors of glycoprotein movement, BFA
and monensin, were also used to examine gp85 processing within the Golgi (Figure 4-3,
lanes 4 and 5). Treatment of cells with BFA resulted in a glycoprotein species with a
slightly higher Mr. Glycoprotein gp85 made in cells treated with monensin, however,
was similar in size to the molecule made in untreated cells. These experiments were

74
repeated with higher concentrations of monensin (Figure 4-6). Increasing the
concentration of monensin, reduced the amount of protein made, but did not alter its
size.
The studies to this point clearly indicated that the native gp85 produced in Akata
cells and recombinant gp85 synthesized in CV-1 cells differed in both antigenicity and
processing. Glycosylation and processing of proteins can vary from cell type to cell type
(Rademacher et al., 1988). It thus remained possible that the differences between the
recombinant and the native proteins might be attributable to the fact that they were being
made in different cells. To examine this possibility further, we compared expression of
native gp85 in induced Akata cells with that of recombinant gp85 produced in the BJAB
cells, which, like Akata, are B cells derived from a Burkitt's lymphoma; however, unlike
Akata, do not contain the EBV genome.
We first compared the antigenicity of gp85 in each cell. Anti-BX antibody
immunoprecipitated a glycoprotein with an apparent Mr of 85,000 from both infected and
induced cells (Figure 4-7). In contrast, antibody F-2-1 immunoprecipitated a
glycoprotein with an apparent Mrof 85,000 only from induced Akata cells. We next
compared the expression of native and recombinant gp85 on the cell surface. We
analyzed the cell surface localization of the molecules using two methods. The first
approach was to determine the accessibility of the glycoproteins to labeling at the cell
surface with radioactive iodine, while the second approach was to stain the molecules
with fluorescent antibodies. Anti-BX antibody immunoprecipitated 12>I labeled gp85
from induced Akata cells, but not from BJAB cells infected with recombinant or wild

75
type vaccinia virus (Figure 4-8, panel A). This was despite the fact that in a parallel
control experiment, anti-BX immunoprecipitated the recombinant protein from BJAB
cells that have been labeled metabolically with [3H] glucosamine (panel B).
Immunofluorescence staining confirmed the failure of transport of recombinant gp85.
Anti-BX antibody stained gp85 in both acetone and paraformaldehyde fixed EBV
expressing B95-8 cells (Figure 4-9). However, despite the high percentage of acetone
fixed cells that stained with anti-BX antibody, no surface staining was detectable. In
addition, a different intracellular staining pattern was evident between cells producing
the native or the recombinant molecule. While a homogeneous pattern of fluorescence
was seen throughout the Akata cell, fluorescence was concentrated at perinuclear regions
of the BJAB cell infected with recombinant vaccinia.

76
1 2
-97.5 Kd
-69.0 Kd
-45.0 Kd
FIGURE 4-1. Comparison of gp85 synthesis in uninduced (lane 1) and in induced
Akata cells (lane 2). One million Akata cells were radiolabeled with 100 pCi of
[35S]methionine and resulting cell lysates were immunoprecipitated with antibody F-2-1.
Samples were run on an SDS-PAGE gel under reducing conditions and visualized by
standard fluorographic techniques.

77
FIGURE 4-2. Determination of optimum labeling of gp85. One million Akata cells
were labeled with 100 pCi of [,5S] methionine for 1 hour (lanes 1, 4 and 6), 30 minutes
(lanes 2 and 7) or 15 minutes (lanes 3, 5 and 8) at either six hours post induction (lanes 1
through 3), eight hours post induction (lanes 4 and 5) or ten hours post induction (lanes 6
through 8). All samples were incubated for and additional 20 hours after labeling and
immunoprecipitated with monoclonal F-2-1. Cell lysates were analyzed on SDS-PAGE
gels under reducing conditions and visualized by standard fluorographic techniques.

78
1 2 3 4 5
jgfKMji
-97.5 Kd
-69.0 Kd
-45.0 Kd
FIGURE 4-3. Analysis of native gp85 expression in the presence of 1 mM DNM (lane
2), 300 ng per ml of swainsonine (lane 3), 2 pg per ml of BFA (lane 4), 0.001 pM
monensin (lane 5), or mock treated (lane 1). Akata cells were induced for six hours with
anti-immunoglobulin prior to being radiolabeled with 100 pCi [35S] methionine for 20
hours and immunoprecipitated with monoclonal antibody F-2-1. All samples were
analyzed under reducing conditions by SDS-PAGE and visualized by standard
chromatographic techniques.

79
FIGURE 4-4. Effects of endoglycosidases on native gp85. Akata cells were induced for
six hours with anti-immunoglobulin prior to being radiolabeled with 100 uCi of [35S]
methionine for 20 hours. The radiolabeled cell lysates were immunoprecipitated with
F-2-1 and incubated overnight with enzyme buffer (lane 1), endoglycosidase H (lane 2)
or endoglycosidase F (lane 3). All samples were analyzed under reducing conditions by
SDS-PAGE and visualized by standard fluorographic techniques.

80
FIGURE 4-5. Endo H and F digestion pattern of proteins immunoprecipitated from
mature virions released into the culture supernatant (lanes 1 through 3) and endo H and F
digestion pattern of proteins immunoprecipitated from the infected cell (lanes 4 through
6). All samples were radiolabeled with 100 pCi of [35S] methionine and
immunoprecipitated with monoclonal antibody F-2-1. Immunoprecipitated samples were
either mock treated (lanes 1 and 4), treated with endo H (lanes 2 and 5) or treated with
endo F (lanes 3 and 6). All samples were analyzed under reducing conditions by
SDS-PAGE and visualized by standard chromatographic techniques.

81
12 3 4
07.5 Kd
69.0 Kd
15.0 Kd
FIGURE 4-6. Treatment of induced Akata cells with varying concentrations of
monensin. Samples were mock treated (lane 1) or treated with 0.001 pM monensin (lane
2), 0.01 pM monensin (lane 3), or 1.0 pM monensin (lane 4). Akata cells were induced
for six hours with anti-immunoglobulin prior to being radiolabeled with 100 pCi [35S]
methionine for 20 hours. All samples were analyzed under reducing conditions by
SDS-PAGE and visualized by standard chromatographic techniques.

82
FIGURE 4-7. SDS-PAGE analysis of recombinant and native gp85. BJAB cells were
infected with VVpEBl-BXLF2 (lanes 1 and 2). Akata cells were induced with
anti-immunoglobulin (lanes 3 and 4). At 6 hours post infection or post induction, cells
were radiolabeled with 100 pCi of [3H] glucosamine for 20 hours and EBV proteins
immunoprecipitated with antibody F-2-1 (lanes 1 and 4) or anti-BX antibody (lanes 2
and 3). The immunoprecipitated proteins were analyzed under reducing conditions in
7.5% acrylamide cross-linked with BIS. Proteins were visualized by standard
fluorographic techniques.

83
FIGURE 4-8. SDS-PAGE analysis of recombinant and native gp85. Panel A. Akata cells
were induced with anti-immunoglobulin (lane 1). BJAB cells were infected with
VVpSCl l-BXLF2b (lane 2) or with wild type vaccinia virus (lane 3). Cells were labeled
extrinsically with 1251 at 24 hours post induction or 10 hours post infection, lysed and
immunoprecipitated with anti-BX antibody. Panel B. BJAB cells were infected with
VVpSCl l-BXLF2b (lane 1) or with wild type vaccinia virus (lane 2). Cells were
metabolically labeled for ten hours post infection with [3H] glucosamine lysed and
immunoprecipitated with anti-BX antibody. The immunoprecipitated proteins were
analyzed under reducing conditions in a 7.5% acrylamide gel cross-linked with BIS and
proteins were visualized by standard fluorographic techniques.

84
FIGURE 4-9. Indirect immunofluorescence staining of native gp85 and recombinant
gp85. Cells were fixed with acetone (panels 1, 3 and 5), or with paraformaldehyde
(panels 2, 4 and 6), and incubated sequentially with anti-BX antibody and
fluorescein-conjugated goat anti-rabbit antibody. B95-8 cells (panels 1 and 2),
pSCl l-BXLF2b infected BJAB cells (panels 3, 4 and 6) or uninfected BJAB cells (panel
5). Panel 6 is a bright field view of panel 4.

85
Discussion
The availability of the Akata cell line facilitated analysis of the differences in
processing of native and recombinant gp85. Both native and recombinant proteins were
sensitive to the action of DNM which inhibits cleavage of terminal glucose residues on
oligosaccharide side chains in the endoplasmic reticulum. However, there were clear
differences in the sensitivity of the two proteins to inhibitors and enzymes that are used
to track movement and processing of glycoproteins beyond this first step. Our data
confirm previous observations with other strains of EBV (Edson and Thorley-Lawson,
1983) which suggested that mature gp85 carries some sugars that are sensitive and some
sugars that are resistant to digestion with endo H. However, none of the sugars on the
recombinant protein were resistant to endo H resistant. Since the processing of sugars
from high mannose, endo H sensitive type to the complex endo H resistant type, occurs
in the Golgi apparatus, this suggests that the recombinant molecule is blocked in
transport into this compartment. The alternative explanation for the absence of endo H
resistant species is that the recombinant molecule is folded in such a way that the sugar
residues are inaccessible to digestion. However, absence of recombinant gp85 on the
surface of infected cells argues in favor of a block in transport. While the effects of
swainsonine, monensin and BFA were not dramatic, they were consistent with the fact
that even the native protein apparently carries some high mannose, unprocessed sugar.
However, native gp85 and not recombinant gp85, migrated as a more discrete species in
the presence of these drugs, implying that differential processing of side chains is part of

86
the normal biosynthetic pathway of the protein in the Golgi and was not part of the
processing pathway of recombinant gp85.
Reevaluation of expression of recombinant gp85 in the lymphoid cell line, BJAB,
suggested that the aberrant processing was intrinsic to the recombinant molecule and was
not a function of the cell in which it was expressed. Heineman and coworkers (1988)
reported that gp85 is poorly expressed by retroviral vectors and probably not correctly
transported under these circumstances. Several groups have described expression of the
herpes simplex virus type 1 (HSV-1) and the human cytomegalovirus (HCMV) gH
proteins which are the homologues of EBV gp85. In both cases, gH is incompletely
processed and transported. Gompels and colleagues (1989) suggested that a second virus
protein might be required as a "chaperone". At this point it seemed likely that EBV
gp85 might share this requirement.

CHAPTER 5
ASSOCIATION OF THE EBV BKRF2 GENE PRODUCT
WITH NATIVE gp85
Introduction
Hutchinson and colleagues (1992) recently confirmed the speculation that an
additional protein was necessary for authentic processing and transport of HSV-1
glycoprotein, gH. These workers made antibodies to a peptide corresponding to the
predicted sequence of the HSV-1 UL1 gene product and immunoprecipitated two
differentially processed forms of a glycoprotein, designated gL, which complexed with
gH in the endoplasmic reticulum. The HSV-1 gL is predicted to have a Mr of 25,000
and contain a single site for the attachment of amino linked oligosaccarides (McGeoch et
al., 1988). The predicted sequence contains no methionine residues other than that at the
translational start site. A single hydrophobic sequence that might function as a signal
peptide is found at the amino terminus and since it was not possible to label gL with [35S]
methionine it was suggested that the signal sequence might be cleaved. Glycoproteins
gH and gL form a noncovalently associated heterodimer that is relatively stable in high
salt. Experiments done with cells coinfected with recombinant vaccinia viruses
expressing gH and gL demonstrated that the two glycoproteins are mutually dependent
on one another for processing and transport. Similar results were obtained by Kaye and
coworkers (1992) who found that correct processing of the HCMV gH was dependent on
co-expression of the product of the HCMV UL115 gene.
87

88
The HCMV UL115 gene product has no sequential homology with HSV-1 gL,
but the CMV gene is a positional homologue of HSV-1 UL1. Similar positional
homologues of the HSV-1 UL1 have been identified in varicella zoster virus and EBV
(Figure 5-1). The EBV homologue is the BKRF2 gene (Baer et al., 1984: Davison and
Taylor, 1987: McGeoch et al., 1988). The BKRF2 open reading frame (Figure 5-2) is
predicted to encode a protein of 137 amino acids with a calculated molecular weight of
15,000 (Baer et al., 1984). The predicted sequence includes three potential amino linked
glycosylation sites. The protein has a hydrophobic domain at the amino terminus, which
has characteristics of a signal peptide and like HSV-1 gL, no potential membrane anchor
sequence at the carboxyl terminus. Similar to the HSV gL, the BKRF2 open reading
frame contains only one methionine residue at position one. The BKRF2 gene product is
also predicted to contain five cysteine and four tyrosine residues. In this chapter, we
describe the search for chaperones of gp85 and demonstrate that the BKRF2 gene
product associates with gp85 in the virus producing cell.
Materials and Methods
Cells and virus
The lymphoblastoid cell lines B95-8 and Akata were used in addition to bovine
embryonic lung (BEL) cells. These cells were grown in DMEM supplemented with 5%
FBS, 100 IU of penicillin and 100 jag of streptomycin per ml. Stocks of AHV-1 (strain
1982: Zoological Society of San Diego, San Diego, Calif.) were prepared by first
washing BEL cells one time with DMEM that did not contain serum. These washed cells

89
were subsequently infected at a MOI of 5. The viral inoculum was removed at two hours
post infection, overlaid with DMEM supplemented with 2% FBS and incubated for 72
hours at 37C. Cells were harvested, frozen and thawed twice and virus stocks stored at
-85 C. EBV was obtained from Akata cells which were resuspended at a concentration
of 4 x 106 cells per ml and induced with 100 pg anti-human immunoglobulin G per ml
for 48 hours. Spent culture medium was clarified by centrifugation at 4,000 x g for 10
minutes, 100 pg bacitracin per ml was added, the virus was pelleted by centrifugation at
20,000 x g for 90 minutes and resuspended in 1/125 of the original volume of RPMI
1640 with 100 pg bacitracin per ml. EBV was labeled extrinsically with 125I after
pelleting from 4 mis of concentrated culture supernatant. Pelleted virus was resuspended
in 0.5 ml PBS and labeled with 0.5 mCi of l25I by use of Iodobeads.
Antibodies
One polyclonal rabbit anti-peptide antibody, anti-BK, was used in addition to the
previously described polyclonal antibody anti-BX and the monoclonal antibodies F-2-1,
and 72A1. A third monoclonal antibody, 12B5, recognizes a complex of envelope
proteins of Alcelaphine herpesvirus 1 (AHV-1) (Adams and Hutt-Fletcher, 1990).
Anti-BK antibody was generated against a synthetic peptide corresponding to residues
125 to 137 of the BKRF2 open reading frame. The peptide was synthesized by the
Protein Chemistry Core Facility of the University of Florida and a cysteine residue was
added to the amino terminus of the sequence to facilitate coupling with keyhole limpet
hemocyanin (Liu et al., 1979). The rabbit was immunized subcutaneously with 1 mg of
the conjugated BKFR2 peptide-KLH complex that had been emulsified in Freund's

90
complete adjuvant and subsequently at biweekly intervals with conjugated peptide
emulsified in Freund's Incomplete Adjuvant. The antibody was purified by
chromatography on protein A-Sepharose as previously described in chapter 3.
Immunofluorescence
For cytoplasmic staining, indirect immunofluorescence was performed on acetone
fixed Akata cells that had been induced with anti-human immunoglobulin for 24 hours.
Slides bearing air-dried acetone fixed cells and primary antibody were incubated in a
humidified atmosphere at 37C for 35 minutes. Rabbit anti-peptide antibody, anti-BK,
or preimmune rabbit antibody were used as the primary antibodies at a concentration of
7.4 pg per ml. The fixed cells were subsequently washed three times with PBS and
reincubated for 35 minutes with the appropriate dilution of fluorescein
isothiocyante-conjugated goat anti-rabbit serum. After incubation with the primary
antibody for 35 minutes the cells were washed three times in PBS and mounted in a
solution containing PBS and glycerol at a 1:1 ratio.
For surface staining, cells were fixed briefly in suspension with ice cold 0.1%
paraformaldehyde in phosphate buffered saline. The fixed cells were reacted in
suspension with either rabbit anti-peptide antibody, anti-BK, or preimmune rabbit
antibody at a concentration of 7.4 pg per ml. The fixed cells and primary antibody were
incubated for 35 minutes at 37C. Fluorescein conjugated goat anti-rabbit serum was
subsequently added, in proper dilution, to washed cells. The cells were washed by
centrifugation at 400 x g three times between incubations and three times before cells

91
were mounted in phosphate buffered saline for examination with a fluorescence
microscope.
Boiling analysis
Three samples of 2 x 106 Akata cells were induced with anti-immunoglobulin G
for six hours, incubated in medium containing 1/10 concentration of glucose for 2 hours
and radiolabeled with 100 (j.Ci of [3H] glucosamine for 20 hours. Cells were solubilized
in RIPA and immunoprecipitated with 120 pg of monoclonal antibody F-2-1 for 20
hours at 4C. One of the three samples was held at 4 C. The remaining two were boiled
for 5 minutes cooled to room temperature and reprecipitated overnight at 4 C with either
100 pg of anti-BX or 100 pg of anti-BK antibody. All three samples were then analyzed
by SDS-PAGE in 12% acrylamide cross-linked with DATD.
Sucrose gradient centrifugation
Five million induced Akata cells were radiolabeled with 500 pCi of [3H]
glucosamine for 20 hours. The cells were harvested as previously described and lysed
with Lysing buffer (50 mM Tris-HCl, pH 7.4, 100 mM NaCl, 1.0% Triton X-100, 0.5%
deoxycholate, 100 units per ml of Aprotinin and ImM PMSF) for 1 hour before
vortexing and centrifugation. The lysates were loaded onto continuous 5-25% sucrose
gradients and centrifuged for 18 hours at 175,000 x g at 18C. One ml fractions of the
gradient were then immunoprecipitated with 100 pg of anti-BKRF2 antibody and
protein-A agarose for 20 hours at 4C. Samples were processed, as previously described
for electrophoretic analysis on 12% acrylamide gels cross-linked with DATD.

92
Coinfection of AHV-1 and vaccinia virus
Thirty million BEL cells were infected with AHV-1 1982 (strain P2) at an MOI
of 5. At two hours post infection, 10 mis of DMEM with 10% FBS was added to each
T-75 flask. AHV-1 infected cells were then incubated at 37C for 24 hours and
superinfected with VVpSCl l-BXLF2b, VVpEBl-BXLF2 or wild type vaccinia virus
strain IHD-J. Two hours later, medium was replaced with five mis of methionine
deficient medium. After reincubation for one hour 50 pCi per ml of [35S] methionine
was added for 19 hours. The cells were harvested by centrifugation, washed with
DMEM and solubilized in 2 mis of RIPA. The solubilized cells were sonicated on ice for
one minute and centrifuged at 100,000 x g for 1 hour at 4C. Five hundred microliters
of the supernatant was preabsorbed with protein A-Agarose for 24 hours at 4 C and then
immunoprecipitated overnight with 100 pg of antibody and fresh protein A-Agarose.
Results
Previous studies with the EBV gp85 homologues in HSV-1 and HCMV had
suggested that an additional virus encoded protein is required for their intracellular
transport. Mutants of EBV that lack expression of gp85 are not available. In an attempt
to provide a putative virus chaperone protein, we therefore coinfected cells with
recombinant vaccinia virus and with another gamma herpes virus Alcelaphine herpes
virus 1. Expression of AHV-1 glycoproteins was confirmed by immunoprecipitation of
dually infected cells with antibody 12B5. The gp 115 complex recognized by this
antibody was clearly visible in analyses of samples coinfected with VVpSCl l-BXLF2b,

93
VVpEBl-BXLF2 or wild type vaccinia virus (Figure 5-3). A band with an apparent Mr
of 85,000 was seen in coinfected samples infected with AHV-1 and either
VVpSCl l-BXLF2b or VVpEBl-BXLF2 and immunoprecipitated with anti-BX.
However, a protein with an Mr of 85,000 was not immunoprecipitated by antibodies
F-2-1 and E1D1 from cells coinfected with AHV-1 and either VVpSCl l-BXLF2b or
VVpEBl-BXLF2.
While this work was in progress it was demonstrated that the HSV UL1 gene
product, gL, forms a heterodimer with the HSV gH, facilitates transport and processing
of gH and restores the reactivity of recombinant expressed protein with monoclonal
antibodies directed against conformational epitopes. Further, it has been shown that gH
and gL are mutually dependent for transport to the cell surface. Repeated analyses in
high percentage acrylamide gels of gp85 immunoprecipitated by antibody F-2-1 from
virion particles labeled with l25I frequently revealed two coprecipitating proteins with
apparent Mr of 42,000 and 25,000. These species were also seen in similar analyses of
proteins immunoprecipitated from [3H] glucosamine labeled cells (Figure 5-4). Although
there is no protein with significant sequence homology to gL in the EBV genome, the
BKRF2 gene has been identified as a positional homologue. To test the hypothesis that a
product of the BKRF2 gene associated with gp85, we made an antibody to a peptide
corresponding to the thirteen carboxyterminal residues of the putative BKRF2 sequence.
This antibody was used to immunoprecipitate proteins from induced Akata cells that had
been labeled with [3H] glucosamine (Figure 5-5). Preimmune rabbit antibody
immunoprecipitated no glycoproteins, but anti-BK antibody and antibody F-2-1

94
immunoprecipitated glycoproteins with Mr of 85,000 and 25,000. It remained possible,
however, either that the anti-BK antibody cross reacted with gp85 or it coincidentally
reacted with glycoproteins of similar sizes to gp85 and its associated glycoprotein. To
examine this possibility further, proteins were first immunoprecipitated from [3H]
glucosamine-labeled Akata cells by antibody F-2-1 and then analyzed either by
SDS-PAGE without further treatment or boiled to dissociate antigen-antibody complexes
and reprecipitated with anti-BK antibody or with anti-BX antibody before electophoresis.
Proteins with Mr of 85,000 and 25,000 were immunoprecipitated by antibody F-2-1
(Figure 5-6). The 85,000 dalton species was reprecipitated from the dissociated complex
by the anti-BX antibody and the 25,000 daltons species was reprecipitated by the
anti-BK antibody. These results indicated that the glycoprotein immunoprecipitated by
anti-BK antibody was the same as that present in the gp85 complex immunoprecipitated
by antibody F-2-1 and that the anti-BK antibody did not cross react with gp85.
To confirm that gp85 and the BKRF2 gene product form a heterodimer, Akata
cells were induced for six hours and labeled for 20 hours with [3H] glucosamine. Cells
were solubilized in RIPA, layered on 5-25% sucrose gradients and centrifuged at
175,000 x g for 18 hours. Ten 1 ml fractions were collected and each was
immunoprecipitated with anti-BK antibody (Figure 5-7). A glycoprotein of 25,000 was
immunoprecipitated from fractions 3 and 4 at the top of the gradient. Two glycoproteins
of 25,000 and 85,000 were immunoprecipitated from fractions 5, 6, 7 and 8. No
immunoprecipitated fraction contained solely gp85. A higher molecular weight species

95
of approximately 110,000 was immunoprecipitated from fraction 9 at the bottom of the
gradient.
Finally, the reactivity of anti-BK antibody with EBV producing cells was
examined in indirect immunofluorescence assays. Anti-BK antibody stained
intracellular protein in induced Akata cells that had been fixed with acetone, but did not
stain proteins in uninduced cells (Figure 5-8). It also stained the surfaces of B95-8 that
had been fixed with paraformaldehyde. Positive surface fluorescence was seen in those
cells stained with anti-BK antibody. No similar staining with preimmune rabbit
antibody was detected.

96
TFT1"
BBRF2
/
/
/

BBRF1
BBLF4
/
/
/
*18' 1
T 7T
/
T
T
EBV
"no kb
*^"BKRF3
BKRF4 BKRF2
i *
/
/
/
/
HSV-1
UL7-
UL6
i i r
/
^.UL2
UL3
9 kb
UL1
UL5
UL4
FIGURE 5-1. Location of a positional homologue of the HSV-1 UL1 gene in the EBV
genome. The arrangement of HSV-1 genes was compared with the arrangement of
known homologues in EBV (McGeoch et al., 1988; Chee et al., 1990). Conserved genes
are connected by dashed lines and the 'positional homologue' of HSV-1 UL1 is marked
with an asterisk.

97
MET ARG ALA VAL GLY VAL PHE LEU ALA ILE CYS LEU VAL THR ILE
PHE VAL LEU PRO THR TRP GLY ASN TRP ALA TYR PRO CYS CYS HIS VAL
THR GLN LEU ARG ALA GLN HIS LEU LEU ALA LEU GLU ASN ILE SER ASP
ILE TYR LEU VAL SER ASN GLN THR CYS ASP GLY PHE SER LEU ALA SER
LEU ASN SER PRO LYS ASN GLY SER ASN GLN LEU VAL ILE SER ARG CYS
ALA ASN GLY LEU ASN VAL VAL SER PHE PHE ILE SER ILE LEU LYS ARG
SER SER SER ALA LEU THR GLY HIS LEU ARG GLU LEU LEU THR THR LEU
GLU THR LEU TYR GLY SER PHE SER VAL GLU ASP LEU PHE GLY ALA ASN
LEU ASN ARG TYR ALA TRP HIS ARG GLY GLY
FIGURE 5-2. Predicted amino acid sequence of the protein potentially encoded by the
EBV BKRF2 open reading frame. Single underlined region at the carboxy terminus
indicates the synthesized residues. Single underlined and bold printing indicate potential
N-glycosylation sites. Bold printing at the amino terminus indicates the hydrophobic
region predicted by the program of Klein and colleagues.

98
1 2 3 4 5 6 7 8 9 10 11 12
FIGURE 5-3. Coinfection experiment with AHV-1. Samples were infected with AHV-1 and
either wild type vaccinia virus strain IHD-J (lanes 1, 4, 7 and 10), or VVpSCl l-BXLF2b (lanes
2, 5, 8 and 11) or VVpEBl-BXLF2 (lanes 3, 6, 9 and 12). Infected cells were radiolabeled with
100 pCi of [3S S] methionine and cell lysates were immunoprecipitated with monoclonal E1D1
(lanes 1 through 3), monoclonal F-2-1 (lanes 4 through 6), anti-BX (lanes 7 through 9) or
monoclonal 12B5. Samples were analyzed under reducing conditions by SDS-PAGE and
visualized by standard fluorographic techniques.

99
FIGURE 5-4. SDS-PAGE analysis of proteins immunoprecipitated by antibody F-2-1
from Akata cells induced by anti-immunoglobulin and labeled for 20 hours at 6 hours
post-induction with [3H] glucosamine (lane 1) or EBV labeled extrinsically with 125I (lane
2). Immunoprecipitated proteins were analyzed under reducing conditions in 12%
acrylamide cross-linked with DATD and proteins were visualized by standard
fluorographic techniques.

100
¡M^L *43
-31
- 22
FIGURE 5-5. Antibody F-2-1, anti-BK antibody or anti-BK preimmune sera was used to
immunoprecipitate proteins from induced Akata cells that were labeled with 100 pCi of
[3H] glucosamine. Akata cells were analyzed under reducing conditions in a 12%
acrylamide gel cross-linked with DATD. Proteins were visualized by standard
fluorographic techniques.

101
1 2 3
-97
-66
-43
-31
-22
FIGURE 5-6. Antibody F-2-1 was used to immunoprecipitate [3H] glucosamine labeled,
induced Akata cell lysates. Immunoprecipitated proteins were then liberated from the
antibody-bead complex by boiling and re-immunoprecipitated with either antibody F-2-1
(lane 1), anti-BK (lane 2) or anti-BX (lane 3). Akata cells were analyzed under reducing
conditions in a 12% acrylamide gel cross-linked with DATD. Proteins were visualized
by standard fluorographic techniques.

102
FIGURE 5-7. Immunoprecipitations were performed on induced [3H] glucosamine
labeled Akata cell lysates (lane 1), or on sucrose gradient fractions of induced [3H]
glucosamine labeled Akata cell lysates (lanes 2-10). Cell lysates were
immunoprecipitated with anti-BK antibody (lanes 2 through 9) or monoclonal antibody
F-2-1 (lane 1). Akata cells were analyzed under reducing conditions in a 12%
acrylamide gel cross-linked with DATD. Proteins were visualized by standard
fluorographic techniques

103
1
2
FIGURE 5-8. Indirect immunofluorescence staining with anti-BK antibody. Cells were
fixed with acetone (panels 1 and 2), or with paraformaldehyde (panels 3 and 4) and
incubated sequentially with anti-BK antibody (panels 1 and 3) or with pre-immune rabbit
antibody (panels 2 and 4) and fluorescein-conjugated goat anti-rabbit antibody.

104
Discussion
Like its counterparts in other herpesviruses, glycoprotein gp85 cannot be
expressed in isolation as an authentically processed protein. It was first suggested that a
"chaperone" was necessary for the proper processing and transport of the homologues of
g85 in two other human herpesviruses. It was hypothesized that a viral or cellular factor
might provide this function. Alcelaphine herpesvirus 1 (AHV-1) was used to test
whether another gamma herpesvirus might provide the appropriate environment for
synthesis of authentic gp85. This virus was chosen for two reasons. First, it is in the
same subfamily as EBV, and secondly, it contains a protein that is cross reactive with
native gp85 (unpublished). Glycoproteins of AHV-1 could clearly be expressed in cells
coinfected with recombinant vaccinia virus. However, their expression had no effect on
the aberrant antigenicity of gp85.
It was subsequently shown that products of the HSV-1 UL1 gene and the HCMV
UL115 gene were sufficient for complete processing and transport of HSV-1 and HCMV
gH. The EBV BKRF2 gene has been identified as the positional homologue of UL1 and
UL115. Anti-BK antibody, made to a peptide derived from the predicted BKRF2
sequence, immunoprecipitated a glycoprotein with properties consistent with the
predicted structure of the BKRF2 gene product. The ability to label the protein with [3H]
glucosamine, coupled with the apparent Mr of the molecule suggested that the protein
utilizes all three predicted N-linked glycosylation sites. Staining of paraformaldehyde
fixed B95-8 cells with anti-BK suggests that the glycoprotein was transported through
the endoplasmic reticulum to the cell surface. Of even greater interest, however, was the

105
observation that the anti-BKRF2 antibody and antibody F-2-1 immunoprecipitated
proteins that had identical mobilities. The ability of anti-BK antibody to recognize a
protein of 25,000, which had been dissociated from the complex recognized antibody
F-2-1 confirmed that the glycoproteins recognized by the two antibodies were one and
the same. The analysis of glycoproteins co-sedimenting in sucrose provided further
evidence that anti-BK recognizes a specific EBV glycoprotein that forms a heterodimer
with gp85.
No clue is yet provided as to the identity of the third glycoprotein (gp42)
immunoprecipitated by antibody F-2-1. The relative amounts of gp85 and this molecule
are not constant. One possible explanation is that it is a differentially processed form of
the BKRF2 gene product which is not recognized by the anti-BK antibody. It would then
perhaps be analogous to the larger of the two forms of the HSV-1 gL which are derived
by differential processing of the UL1 gene product (Hutchinson et al., 1992). This would
imply that two types of complexes between gp85 and the BKRF2 gene product are made,
one with the larger (gp42) and one with the smaller (gp25) form. Alternatively, gp42
might be a novel species that cross-reacts with antibody F-2-1, or a proteolytic product of
gp85 which does not include the sequence recognized by anti-BX. Further clarification
of these points may be provided when the BKRF2 gene product is coexpressed as a
recombinant protein with gp85.

CHAPTER 6
CONSTRUCTION OF BACTERIAL FUSION PROTEINS
Introduction
The EBV gp85 homologue, gH, has been shown to be a protein conserved in all
herpesviruses so far sequenced. All the gH homologues have been implicated in virus
penetration. When known sequences of herpesvirus gH proteins are aligned by a
glycosylation site near the carboxyterminus, several of the cysteine residues are in similar
positions, otherwise there is relatively little sequence homology between these proteins
with the exception of an 11 amino acid sequence (Figure 6-1). This sequence has quite
extraordinary homology among the different viruses. There is presumably some
significant reason why the region has been so well conserved and is therefore an
important target for mutagenesis and production of antibodies. To help elucidate the
functional significance of this region we have synthesized a bacterial fusion protein to be
used as an antigen to generate antibodies in New Zealand White rabbits.
We cloned the BXLF2 sequence that flanks this highly homologous region into a
pSEM expression vector (Knapps et al., 1990). The pSEM vector contains the amino
terminal 375 amino acids of a P-galactosidase gene. The BXLF2 sequence was cloned
within the multiple cloning site (MSC) and 3' to the P-galactosidase carrier sequence
(Figure 6-2). Transcription of the hybrid gene is driven by a lac promoter and is
106

107
terminated by transcriptional terminators positioned downstream of the polylinker
region. Translational stop codons in all three reading frames are located downstream of
the ///Will site within the MCS. This chapter describes both the construction and
selection of recombinant pSEM-1 vectors containing the region of gH high homology
within the BXLF2 sequence, as well as the subsequent purification procedures used to
obtain expressed protein.
Materials and Methods
Polymerase chain reaction
DNA encompassing 302 base pairs of the BXLF2 open reading frame were
amplified by PCR techniques as previously described. The primer for the forward
reaction was 5'-TATGGATCCGGCCCAACCTTGCCCTATACC-3', while the primer for
the reverse reaction was 5'-ATAAAGCTTTTTGTCGAGCCTGTCCACGCA-3'. DNA
from the previously described pBSKS-BXLF2 vector was used as template for the PCR
reactions.
Western blot analysis
Radiolabeling and immunoprecipitations were carried out as described
previously. For Western Blots the immunoprecipitated proteins were electrophoresed in
9% acrylamide cross-linked with DATD and then electrotransferred onto nitrocellulose
membranes (0.45 um pore size; Schleicher and Schuell, Inc., Keene, NH; Towbin et al.,
1979). The transferred sheets were treated for three hours with blocking buffer (10 mM

108
Tris-HCl, pH 7.2, 0.15 M NaCl, 5% skimmed milk, and 0.05% sodium azide) and
reacted for three hours with blocking buffer containing mouse anti-p-galactosidase
antibody at 400 ug per ml (Sigma). They were then washed five times with wash buffer
(10 mM Tris-HCL, pH 7.2, 0.15 M NaCl, 0.3% Tween 20) for 10 minutes each, followed
by one overnight wash. The washed sheets were reacted with alkaline
phosphatase-conjugated goat anti-mouse antibodies for 3 hours and the bound
anti-mouse antibodies detected by reaction with substrate
(5-bromo-4-chloro-3-indophophate and nitroblue tetrazolium; Sigma).
Purification of bacterial fusion proteins
A 200 ml culture of transformed E. coli was pelleted at 4,000 x g for 10 minutes.
The pellet was resuspended in 8 mis of lysis buffer (50 mM Tris-HCl, pH 8.0, 25%
sucrose and 1 mM EDTA) containing 20 mg of freshly prepared lysozyme (Sigma). The
sample was placed on ice for 30 minutes, sonicated for three minutes and placed into a
tube containing 10 mM MgCl2, 1 mM MnCl2 and 10 pg of DNase 1. This sample was
placed on ice for 30 minutes and then added to detergent buffer containing 0.2 M NaCl,
1% deoxycholate, 1% NP40 Triton X-100, 20 mM Tris-HCl, pH 7.5, and 2 mM EDTA.
The sample underwent centrifugation at 15,000 RPM for 30 minutes and the pellet was
resuspended in 20 ml of 0.5% Triton X-100 and 1 mM EDTA. This triton extraction
was then repeated after the sample had been centrifuged at 10,000 RPM for 20 minutes.
The resulting pellet was resuspended in 10 mis of dH20 and 10 pi was analyzed in a 9%
acrylamide gel cross-linked with DATD.

Results
A polymerase chain reaction (PCR) was run in order to amplify the segment of
DNA containing the gH high homology region of BXLF2 DNA. Ten microliters of the
PCR product was analyzed in an agarose gel parallel to a lane of markers of known size.
A PCR band corresponding to the predicted 302 base pairs was detected (Figure 6-3).
The primers had been chosen such that cutting the amplified DNA with Bam\\\ and
HindlU would produce a fragment that could be directionally cloned in frame into
pSEM-1. However, the restriction sites were very close to the termini of the DNA. In
order to ensure complete digestion of the amplified DNA, it was first cloned into the
Smal site of pGEM-3Zf(-) after being blunt ended. The ligation was used to transform
competent E. coli strain XL1B cells. Ampicillin resistant colonies were obtained and
minipreps of DNA were made. Colony blot hybridization of transformed bacteria with
[32 P] labeled Sma \-Hinc II BXLF2 indicated which colonies contained BXLF2 DNA.
Microgram quantities of plasmid DNA were grown from a positive colony. This was
done to generate sufficient quantities of plasmid to allow digestion with Bam HI and
Hind III. This Bam HI Hind III BXLF2 fragment was ligated into the Bam HI and Hind
III digested pSEM-1 vector and used to transform E. coli. Western blot analysis was
conducted on lysates of resulting colonies. An apparent Mr of 48,000 was detected in the
sample containing unfused (3-galactosidase protein (lane 4), while a protein with an
apparent Mr of 60,000 was seen in those samples containing the (3-galactosidase-
high homology fusion protein (Figure 6-4). Microgram quantities of purified plasmid
DNA was purified from one colony that expressed the fusion protein used to transform

110
E. coli. Fifteen milligrams of purified bacterial fusion protein was obtained from the
transformed bacteria. Samples of this purified fusion protein was analyzed on a 9%
actylamide gel that was cross-linked with DATD and stained with coomassie. A major
band with an apparent Mr of 60,000 was clearly visible in the gel (Figure 6-5).

Ill
706 Amino
r Acids
BXLF2
FIGURE 6-1. Region of gp85 to be expressed as a bacterial fusion protein (residues 441
through 540) and region of high homology among the herpesvirus gH homologues
(magnified area) herpesvirus saimiri (HVS), herpes simplex virus (HSV), varicella-zoster
(VZV), cytomegalovirus (CMV), encelaphine herpesvirus 4 (EHV-4), bovine
herpesvirus-1 (BHV-1), pseudorabies virus (PrV), human herpesvirus 6 (HHV6).

112
EcoRJ SstI Kpril Smal BamHI Xbal Sail PstI SphI Hindi!I
GAT^GGG CAT GGA ATT CCA GCT CGG TAC CCG GGG ATC CTC TAG 7T CGA CCT GCA GGC ATG CAA GCT TGG CTC
I
FIGURE 6-2. Restriction map of the pSEM vector. The vector contains 3730 bp and an
inducible lac promoter. Downstream from the promoter is a truncated form of the E. coli
lacZ gene (lacZ'), encoding the 375 amino acids of p-Galactosidase. The vector also
contains a polylinker region downstream from the lacZ' gene.

113
1
2
23.1
kB-
9.42
kB-
6.56
kB-
4.37
kB-
2.32
kB-
2.03
kB-
0.5 kB
FIGURE 6-3. 1% agarose gel containing ethidium bromide. Lambda DNA standards
are seen lane 1 and 10 jul of resulting PCR product are in lane 2.

114
FIGURE 6-4. Western blot analysis of whole cell lysates. Samples from colonies
expressing fusion protein (lanes 1 through 3) or from a colony expressing unfused
p-galactosidase protein (lane 4). Proteins were separated by SDS-PAGE and blotted
onto nitrocellulose.

115
FIGURE 6-5. Coomassie stain of purified bacterial fusion protein in lane 2 or of
molecular weight markers in lane 1. Proteins were run on a 9% SDS-PAGE gel that had
been cross linked with DATD.

116
Discussion
Amino acids corresponding to residues 441 through 540 of EBV gp85 were
expressed as a bacterial fusion protein. Substantial amounts of purified bacterial fusion
protein were produced using the pSEM bacterial expression system. Furthermore,
purified bacterial fusion protein was easily obtained due to the fusion protein being
located predominantly in the insoluble fraction of the cell lysate. The resulting purified
fusion protein had properties consistent with its predicted structure. The apparent Mr of
60,000 was compatible with the calculated 475 amino acids of the fusion protein. Three
hundred and seventy five amino acids (Mr of 47,000) were derived from the truncated
P-galactosidase protein and 100 amino acids from the high homology region within
BXLF2. Western blot analysis using anti-beta-galactosidase antibody as the primary
antibody, enabled us to easily identify those colonies which were expressing fusion
protein. This fusion protein is currently being used as antigen to generate antibodies in
New Zealand White rabbits.
Antisera from the rabbit immunized with purified fusion protein has been
purified on protein-A Agarose. Recent experiments have indicated that native EBV
gp85 is recognized by the purified rabbit anti-high homology region antibody. Once
higher titers of this antibody are achieved, EBV neutralization assays will be conducted.
Such studies would reveal whether antibodies generated to this region of gp85 can
inhibit viral infectivity. This would suggest an important role for this domain in the
EBV fusion event and once recombinant gp85 expression is optimized, mutagenesis
studies of this area could be initiated.

CHAPTER 7
CONCLUSION
At least two glycoproteins, the homologues of HSV glycoproteins gB and gH, are
conserved among all subgroups of herpesviruses (Cranage et al., 1986; Gompels and
Minson, 1986; McGeoch and Davison, 1986; Davison and Taylor, 1987 Cranage et al.,
1988). Both are thought to be involved in penetration of HSV through target cell
membranes and both are essential for replication of HSV in tissue culture (Sarmiento et
al., 1979; Little et al., 1981; DeLuca et al., 1982; Gompels and Minson, 1986; Cai et al.,
1987; Cai et al., 1988; Desai et al., 1988; Fuller et al., 1989; Highlander et al., 1989).
However, of the two EBV homologues, gpl 10 (the gB equivalent) and gp85 (the EBV
gH), only gp85 has been shown to have properties similar to those of its prototype.
Although gpl 10 is found in relatively large amounts in infected cells (Gong et al., 1987)
and has significant sequence and predicted secondary structure homology with HSV gB
(Pellet et al., 1985), only a more extensively glycosylated form of the molecule, gpl 25, is
found in the virion and all available evidence suggests that gp 125 is not accessible on the
virus surface (Kishishita et al., 1984; Emini et al., 1987; Gong and Kieff, 1990). In
contrast, although the sequence homology between gp85 and other members of the gH
family of glycoproteins is not extensive and is limited to two regions clustered around
four conserved cysteine residues in the carboxyterminal halves of the proteins (Klupp
and Mettenleiter, 1991), at least one of its functions appears to be conserved. The
117

118
glycoprotein is found on the cell surface, antibodies to it can neutralize infectivity
(Qualtiere and Pearson, 1979; Strnad et al., 1982) and it appears to be involved in cell
fusion (Miller and Hutt-Fletcher, 1988; Haddad and Hutt-Fletcher, 1989). Results
presented here suggest that these similarities extend to a requirement for complexing
with at least one other virus glycoprotein. Gompels and colleagues (1989) have
suggested that this associated protein may function as a molecular 'chaperone'.
A set of proteins called 'chaperones' have evolved to ensure that polypeptides will
fold and be transported properly. It has been shown that the probability a given
polypeptide will fold properly increases as the relative protein concentrations decrease.
This is attributed to low concentrations of protein limiting the amount of
inter-polypeptide aggregation that can occur. Furthermore, the probability of proper
folding increases as temperature decreases. This is due to a weakening of hydrophobic
interactions at lower temperatures. However, the cytosol subjects the growing
polypeptide chain to relatively high protein concentrations and temperatures. As the
polypeptide emerges from the ribosome, it is exposed to premature interactions with
other intra- or inter-polypeptide domains which often leads to misfolding and
aggregation (Jaenicke, R., 1991; Mitraki et al., 1991.). It has been proposed that one
mode of chaperone action is to actively unfold badly aggregated or misfolded proteins to
a conformation from which they could refold spontaneously. It is hypothesized that
improperly folded proteins are identified by their large stretches of exposed backbone as
opposed to hydrophobic patches. Two theories have been suggested for the molecular
mechanism underlying the unfolding process. The first suggested mechanism is that the

119
chaperone repeatedly binds and dissociates from the chain until no exposed areas of
backbone are found. The second model is that the protein passes through a chaperone
binding cleft and this allows the linearized chain to refold spontaneously. The term
'molecular chaperone' was first used by Laskey and colleagues to describe the role of
nucleoplasmin in nucleosome assembly. The term was subsequently expanded after
investigations with bacteriophage X (Ellis and Hemmingsen, 1989) led to the discovery
that mutant groE genes blocked Lambda growth. Growth was blocked due to improper
assembly of a bacteriophage-coded protein called the head-tail connector (Friedman et
al., 1984; Georgopoulos et al., 1990; Ang et al., 1991). Furthermore, mutant dnaK, dnaJ
and grpE genes were found to block E. coli replication. It was determined that their role
was to disassemble a helicase complex at the origin of X replication. It soon became
evident that the DnaK, DnaJ, GrpE, GroEL and GroES proteins are conserved among
both prokaryotes and eukaryotes and also belong to the heat shock class of proteins. It
was previously thought that these were all specialized proteins, but it has since been
determined that some of the members of this group have a more generalized
housekeeping role of polypeptide folding, unfolding, transport and dissaggregation
(Jaenicke, R. 1991).
The HSV-1 UL1 gene product, gL, is required for the proper folding and
transport of the HSV-1 gH. However, this gL molecule is reciprocally dependent on the
expression of gH for its proper processing and transport (Roop et al., 1993).
Furthermore, it appears that gL and gH may remain as a functional complex as
investigators have been unable to detect large amounts of nondissociated gL. How the

120
gL molecule will fit into the newly developing concept of molecular chaperones will be
quite interesting.
The coincidence of positional homology with HSV-1 UL1 and HCMV UL115
and apparent association with gH, strongly suggests that BKRF2 is the functional
homologue of gL. The BKRF2 gene is currently being cloned into vaccinia virus to test
whether coexpression of both recombinant gp85 with the recombinant BKRF2 gene
product will produce antigenically normal, properly processed and transported
recombinant gp85.

REFERENCES
Adams, S. W. and L. M. Hutt-Fletcher. 1990. Characterization of Envelope
Proteins of Alcelaphine Herpesvirus 1. J. Virol. 64:3382-3390.
Ang, D., K. Liberek, D. Skowyra, M. Zylicz and C. Georgopoulos. Biological role
and regulation of the universally conserved heat shock proteins. 1991. J. Biol. Chem.
266:24233-24236.
Baer, R., A. T. Bankier, M. D. Biggin, P. L. Deininger, P. J. Farrell, T.
J.Gibson, G. Hatfull, G. S. Hudson, S. C. Satchwell, C. Seguin, P. S. Tuffnell, and B. G.
Barrell. 1984. DNA sequence and expression of the B95-8 Epstein-Barr virus genome.
Nature (London) 310:207-211.
Balachandran, N., D. E. Oba, and L.M. Hutt-Fletcher. 1987. Antigenic
cross-reactions amongst HSV type 1 and 2, EBV, and CMV. J. Virol. 60:1125-1135.
Beisel, C., J. Tanner, T. Matsuo, D. Thorley-Lawson, F. Kexdeg, and E. Kieff.
1985. Two major outer envelope glycoproteins of EBV are encoded by the same gene.
J. Virol. 54:665-674.
Blumenthal, R., A. Bali-Puri, A. Walker, D. Covell, and O. Eidelman. 1987. pH
dependent fusion of vesicular stomatitis virus with vero cells. J. Biol. Chem.
262:13614-13619.
Bonner, W. M. and R. A. Laskey. 1974. A film detection method tor tritium-
labeled proteins and nucleic acids in polyacrylamide gels. Eur. J. Biochem. 40:83-88.
Burkitt, D.P. 1987. Discovering Burkitt's lymphoma. In: P.H. Levine, D.V.
Ablashi, M. Nonoyama, G.R. Pearson, and R. Glaser (ed.), Epstein-Barr virus and
human disease, p.xxi-xxxi. Humana Press, Clifton, New Jersey.
Bzik, D. J., B. A. Fox, N. A. DeLuca, and S. Person. 1984. Nucleotide sequence
specifying the glycoprotein gB of herpes simplex virus type 1. Virology. 133:301-314.
Cai, W., S. Person, S. Warner, J. Zhou and N. A. DeLuca. 1987. Linker
insertion, nonsense, and restriction site deletion mutants of the gB glycoprotein of herpes
simplex virus type 1. J. Virol. 61: 714-721.
121

122
Cai, W., B. Gu and S. Person. 1988. Role of glycoprotein B in viral entry and
cell fusion. J. Virol. 62:2596-2604.
Cameron, K. R., T. Stamminger, M. Craxton, W. Bodemer, R., W. Honess, and
B. Fleckenstein. 1987. The 160,000 Mr virion protein encoded at the right end of the
herpesvirus saimiri genome is homologous to the 140,000 Mr membrane antigen encoded
at the left end of the Epstein-Barr virus genome. J. Virol. 61:2063-2070.
Chakrabarti, S., K.Brechling and B. Moss. 1985. Vaccinia virus expression
vector: Coexpression of p-Gal actos i dase provides visual screening of recombinant virus
plaques. Mol. Cell. Biol. 5:3403-3409.
Chee, M. S., A. T. Bankier, S. Beck, R. Bohni, C. M. Brown, R. Cesny T.
Horsnell, C. A. Hutchinson, T. Kouzarides, J. A. Martignitti, S. C. Satchwell, P.
Tomlinson, K. M. Weston and B. G. Barrell. 1990. An analysis of the protein coding
content of the sequence of human cytomegalovirus strain AD 169. Current Topics in
Microbiology and Immunology 154:125-169.
Choppin, P. W. and R. W. Compans. 1975. The structure of influenza virus. In:
E. Kilbourne (ed.), The influenza virus and influenza, p. 15-51. Academic Press, New
York.
Claesson-Welsh, L. and P. G. Spear. 1986. Oligomerization of herpes simplex
virus glycoprotein B. J. Virol. 60:803-806.
Cleary, M. L., R. F. Dorfman, and J. Sklar. 1986. Failure in immunologic control
of the virus infection: post transplant lymphomas. In: M. Epstein and B. Achong (ed.),
The EBV: Recent Advances, p.163-181. Heinemann Pub., London.
Collins, P. L. and G. Mottet. 1991. Post-translational processing and
oligomerization of the fusion glycoprotein of human respiratory syncytial virus. J. Virol.
72:3095-3101.
Cranage, M.P., T. Kouzarides, A. T. Bankier, S. Satchwell, K. Weston, P.
Tomlinson, B. Barrell, H. Hart, S. E. Bell, A. C. Minson and G. L. Smith. 1986.
Identification of the human cytomegalovirus glycoprotein B gene and induction of
neutralizing antibodies via its expression in a recombinant vaccinia virus. EMBO J.
5:3057-3063.
Cranage, M. P., G. L. Smith, S. E. Bell, H. Hart, C. Brown, A. T. Bankier, P.
Tomlinson, B.G. Barrell, T. C. Minson. 1988. Identification and expression of a human
cytomegalovirus glycoprotein with homology to the EBV BXLF-2 product,
varicella-zoster virus gpIII, and herpes simplex type 1 glycoprotein H. J. of Virol.
62:1416-1422.

123
Dambaugh, T. R., C. Beisel, M. Hummel, W. King, S. Fennewald, A. Cheung, M.
Heller, N. Raab-Traub, and E. Kieff. 1980. Epstein-Barr virus (B95-8) DNA VII.
Molecular cloning and detailed mapping. Proc. Natl. Acad. Sci.USA 77:2999-3003.
Daumbaugh, T.R. and E. Kieff. 1982. Identification and nucleotide sequences of
two similar tandem repeats in Epstein-Barr virus DNA. J. Virol. 44:823-833.
Davison, A. J. and P. Taylor. 1987. Genetic relations between varicella-zoster
virus and Epstein-Barr virus. J. Gen. Virol. 68:1067-1079.
DeLuca, N.A., D.J. Bzik, V.C. Bond, S. Person, and W. Snipes. 1982.
Nucleotide sequences of herpes simplex virus type 1 affecting virus entry, cell fusion and
production of glycoprotein B. Virology 122:411-423.
Desai, P. J., P. A. Schaffer, and A. C. Minson. 1988. Excretion of non-infectious
particles lacking glycoprotein H by a temperature sensitive mutant of herpes simplex
virus type I: evidence that gH is essential for virion infectivity. J. Gen. Virol.
69:1147-1156.
de-The, G. and Y. Zeng. 1987. Population screening for EBV markers: toward
improvement of nasopharyngeal carcinoma control. In: M.A. Epstein and B.G. Achong
(ed.), The EBV: Recent Advances, p. 236-249. John Wiley and Sons, New York.
Dorns, R.W., A. Helenius, and J. White. 1985. Membrane fusion activity of the
influenza virus hemagglutinin: the low pH-induced conformational change. J. Biol.
Chem. 260:2923-2931.
Dorns, R. W., G. Russ and J. W. Yewdell. 1989. Brefeldin A redistributes
resident and itinerant Golgi proteins to the endoplasmic reticulum. J. Cell Biol.
109:61-72.
Doxsey, S. J., F. M. Brodsky, G. S. Blank, and A. Helenius. 1987. Inhibition of
endocytosis by anti-clathrin antibodies. Cell 50:453-463.
Edson, C. M. and D.A. Thorley-Lawson. 1981. Epstein-Barr virus membrane
antigens: characterization, distribution, and strain differences. J. Virol. 39: 172-182.
Edson, C. M. and D. A.Thorley-Lawson. 1983. Synthesis and processing of the
three major envelope glycoproteins of Epstein-Barr virus. J. Virol. 46:547-556.
Efstathiou, S., Y. M. Ho and A. C. Minson. 1990. Cloning and molecular
characterization of the murine herpesvirus 68 genome. J. Virol. 71:1355-1364.

124
Elbein, A. D. 1987. Inhibitors of the biosynthesis and processing of N-Linked
oligosaccharide chains. Ann. Rev. Biochem. 56:497-534.
Ellis, R.J. and S. M. Hemmingsen. Molecular chaperones: proteins essential for
the biogenesis of some macromolecular structures. 1989. Trends Biochem. Sci.
14:339-342.
Emini, E. A., J. Luka, M. E. Armstrong, P. M. Keller, R. W. Ellis and G. R.
Pearson. 1987. Identification of an Epstein-Barr virus glycoprotein which is
antigenetically homologous to the varicella-zoster virus glycoprotein II and the herpes
simplex virus glycoprotein B. Virology 157:552-555.
Fingeroth, J.D., J. J. Weiss, T. F. Tedder, J. L. Strominger, P. A. Biro, and D. T.
Fearon. 1984. EBV receptor of human B lymphocytes is the C3d receptor. Proc. Natl.
Acad.Sci.USA 81:4510-4514.
Fleckenstein, B. and R.C. Desrosiers. 1982. Herpesvirus saimiri and Herpesvirus
teles, In: B. Roizman (ed.), The Herpesviruses, vol. 1. p. 253. Plenum Press, New York.
Forrester, A. J., V. Sullivan, A. Simmons, B. A. Blacklaws, G. L. Smith, A. A.
Nash and A. Minson. 1991. Induction of protective immunity with antibody to herpes
simplex virus type 1 glycoprotein H (gH) and analysis of the immune response to gH
expressed in recombinant vaccinia virus. J. Gen. Virol. 72:369-375.
Frink, R. J., R. Eisenberg, G. Cohen, and E. K. Wagner. 1983. Detailed analysis
of the portion of the herpes simplex virus type 1 genome encoding glycoprotein C. J.
Virol. 45:634.
Friedman, D.I., E.R. Olson, C. Georgopoulos, K. Tilly, I. Herskowitz and F.
Banuett. 1984. Interactions of bacteriophage and host macromolecules in the growth of
bacteriophage. Microbiol. Rev.48:299-325.
Fuller, A. O. and P. G. Spear. 1985. Specificities of monoclonal and polyclonal
antibodies that inhibit adsorption of herpes simplex virus to cells and lack of inhibition
by potent neutralizing antibodies. J. Virol. 55:475-482.
Fuller, A. O. and P. G. Spear. 1987. Anti-glycoprotein D antibodies that permit
adsorption but block infection by herpes simplex virus 1 prevent virion-cell fusion at the
cell surface. Proc. Natl. Acad. Sci. USA 84:5454-5458.
Fuller, A. O., R. Santos and P. Spear. 1989. Neutralizing antibodies specific for
glycoprotein gH of herpes simplex virus permit viral attachment to cells but prevent
penetration. J. Virol. 63:3435-3443.

125
Garoff, H., A. M. Frischauf, K. Simon, H. Lehrach, and H. Delius. 1980.
Nucleotide sequence of cDNA coding for Semlili Forest virus membrane glycoproteins.
Nature (London) 288:236-241.
Geshelin, P. and K. I. Berns. 1974. Characterization of the naturally occurring
crosslinks in vaccinia virus DNA. J. Mol. Biol. 88:785-796.
Gething, M. J., J. M. White, and M. D. Waterfield. 1978. Purification of the
fusion protein of Sendai virus : analysis of the NH2-terminal sequence generated during
precursor activation. Proc. Natl. Acad. Sci. USA 75:2737-2740.
Forrester, A., H. Farrell, G. Wilkinson, J. Kaye, N. Davis-Poynter and T. Minson.
1992. Construction and properties of a mutant herpes simplex virus type 1 with
glycoprotein H coding sequences deleted. J. Virol. 66:341-348.
Georgopoulos, C., D. Ang, K. Liberek and M. Zylicz. 1990. In: Stress proteins in
biology and medicine, Morimoto, R., A. Tissieres and C. Georgopoulos eds. pp.
191-221, Cold Spring Harbor Laboratory Press.
Given, D., D. Yee, K. Griem, and E. Kieff. 1979. DNA of Epstein-Barr virus
direct repeats of the ends of Epstein-Barr virus DNA. J. Virol. 30:852-862.
Goldstein, J. L., R. G. Anderson, and M. S. Brown. 1979. Coated pits, coated
vesicles, and receptor-mediated endocytosis. Nature (London) 279:230-247.
Gompels, U. and A. Minson. 1986. The properties and sequence of glycoprotein
H of herpes simplex virus type 1. Virology 153:230-247.
Gompels, U. A., M. Craxton and R. Honess. 1988. Conservation of glycoprotein
H (gH) in herpesviruses: nucleotide sequence of the gH gene from herpesvirus saimir. J.
Gen. Virol. 69:2819-2829.
Gompels, U.A. and A. C. Minson. 1989. Antigenic properties and cellular
localization of herpes simplex virus glycoprotein H synthesized in a mammalian cell
expression system. J. Virol. 63:4744-4755.
Gong, M., T. Ooka, T. Matsuo, and E. Kieff. 1987. Epstein-Barr virus
glycoprotein homologous to herpes simplex virus gB. J. Virol. 61:499-508.
Gong, M., and E. Kieff. 1990. Intracellular trafficking of two major Epstein-Barr
virus glycoprotein, gp350/220 and gpl 10. J. Virol. 64:1507-1516.

126
Gruters, R. A., J. J. Neefjes, M. Tersmette, R. E. Y. de Goede, A. Tulp, H. G.
Huisman, F. Miedema, and H. L. Ploegh. 1987. Interference with HIV-induced
syncytium formation and viral infectivity by inhibitors of trimming glucosidase. Nature
(London) 330:74-77.
Haddad, R. S., and L. M. Hutt-Fletcher. 1989. Depletion of glycoprotein gp85
from virosomes made with Epstein-Barr virus proteins abolishes their ability to fuse with
virus receptor-bearing cells. J. Virol. 63:4998-5005.
Heifetz, A., R. W. Keenan and A. D. Elbein. 1979. Mechanism of action of
tunicamycin on the UDP-GlcNAc:Dolichyl-phosphate GlcNAc-1-phosphate transferase.
Biochemistry 18:2186-2192.
Heikeshoven, J. and R. Dernick. 1985. Simplified method for silver staining of
proteins in polyacrylamide gels and the mechanism of silver staining. Electrophoresis
6:103-112.
Heineman, T., M. Gong, J. Sample, and E. Kieff. 1988. Identification of the
Epstein-Barr virus gp85 gene. J. F/ro/. 62:1101-1107.
Helenius, A., J. Kartenbeck, K. Simons, and E. Fries. 1980. On the entry of
Semliki Forest Virus into BHK-2 cells. J. Cell Biol. 84:404-420.
Helenius, A., M. Marsh, and J. White. 1982. Inhibition of Semliki Forest virus
penetration by lysosomotropic weak bases. J. Gen. Virol. 58:47-61.
Henle, W., G. Henle, B. A. Aajac, G. Pearson, R. Waubke, and M. Scriba. 1970.
Differential reactivity of human serums with early antigens induced by Epstein-Barr
virus. Science 169:188-190.
Henle, G., and W. Henle. 1966. Immunofluorescence in cells derived from
Burkitt lymphoma. J. Bacteriol. 91:1248-1256.
Heston, L., M. R. Abson, N. Brown, and G. Miller. 1982. New Epstein-Barr virus
variants from cellular subclones of P3J-HR-1 Burkitt lymphoma. Nature (London)
295:160-163.
Highlander, S. L., W. Cai, S. Person, M. Levine, and J. C. Glorioso. 1988.
Monoclonal antibodies define a domain on herpes simplex virus glycoprotein B involved
in virus penetration. J. Virol. 62:1881-1888.
Hinuma, Y., M. Kohn, J. Yamaguchi, and J. T. Grace. 1967. Replication of
herpes-type virus in a Burkitt lymphoma cell line. J. Virol. 1:1203-1206.

127
Hoffman, G. J., S. G. Lazarowitz and S. D. Hayward. 1980. Monoclonal
antibody against a 250,000 dalton glycoprotein of Epstein-Barr virus identifies a
membrane antigen and a neutralizing antigen. Proc. Natl. Acad. Sci. USA.
77:2979-2983.
Hsu, M., A. Scheid, and P. W. Choppin. 1981. Activation of Sendai F protein
involves a conformational change with exposure of a new hydrophobic region. J. Biol.
Chem. 256:3557-3563.
Hsu, M., A. Scheid, and P. W. Choppin. 1982. Enhancement of
membrane-fusing activity of Sendai virus by exposure of the virus to basic pH is
correlated with a conformational change in the fusion protein. Proc. Natl. Acad.
Sci.USA 79:5862-5866.
Hutchinson, L., H. Browne, V. Wargent, N. Davis-Poynter, S. Primorac, K.
Goldsmith, A. C. Minson and D. C. Johnson. 1992. A novel herpes simplex virus
glycoprotein, gL, forms a complex with glycoprotein H (gH) and affects normal folding
and surface expression of gH. J. Virol. 66:2240-2250.
Johnson, D.C. and M.W. Ligas. 1988. Herpes simplex viruses lacking
glycoprotein D are unable to inhibit virus penetration: quantitative evidence for
virus-specific cell surface receptors. J. F/>o/.62:4605-4612.
Jaenicke, R. 1991. Protein folding: local structures, domains, subunits and
assemblies. Biochemistry 30:3147-3161.
Josephs, S., D. Ablashi, S. Salahuddin, L. Jagodzinski, F. Wong-Staal, and R.
Gallo. 1991. Identification of the human herpesvirus 6 glycoprotein H and putative large
tegument protein genes. J. Virol. 65:5597-5604.
Kaye, J. F., U. A. Gompels and A. C. Minson. 1992. Glycoprotein H of human
cytomegalovirus (HCMV) forms a stable complex with the HCMV UL115 gene product.
J. Gen. Virol. 73:2693-2698.
Keller, P. M., A.J Davison, R. S. Lowe, M. W. Reimon, and R. W. Ellis. 1987.
Identification and sequence of the gene encoding gpIII, a major glycoprotein of
varicella-zoster virus. Virology. 157:526-533.
Kishishita, M., J. Luka, B. Vroman, J. F. Poduslo, G. R. Pearson. 1984.
Production of monoclonal antibody to a late intracellular Epstein-Barr virus induced
antigen. Virology 133:363-375.
Klein, P., M. Kanehisa, and C. DeLisi. 1985. The detection and classification of
membrane-spanning proteins. Biochem. Biophys. Acta 815:468-476.

128
Klupp, B. G., and T. C. Mettenleiter. 1991. Sequence and expression of the
glycoprotein gH gene of pseudorabies virus. Virology 182:732-741.
Knapps S., M. Broker and E. Amann. 1990. pSEM vectors: high level expression
of antigenic determinants and protein domains. Biotechniques 8:280.
Kornfeld, R. and S. Kornfeld. 1985. Assembly of asparagine linked
oligosaccharides. Annual Rev. Biochem. 54:631-663.
Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the
head of bacteriophage T4. Nature (London) 227:680-685.
Laskey, R.A. and W.C. Earnshaw. 1980. Nucleosome Assembly. Nature
286:763-767.
Lazarowitz, S., and P. W. Choppin. 1975. Enhancement of the infectivity of
influenza A and B viruses by proteolytic cleavage of the haemagglutinin polypeptide.
Virology 68:440-454.
Lenoir, G.M. 1987. Role of the virus, chromosomal translocations, and cellular
oncogenes in the aetiology of Burkitt's lymphoma. In: M.A. Epstein and B.G.Achong
(ed.), The EBV:Recent advances, p. 184-205. John Wiley and Sons, Inc., New York.
Lindahl, T., A. Adams, G. Bjursell, G. Bornkamm, G. Kaska-Pierich, and U.
Jehn. 1976. Covalently closed circular duplex DNA of Epstein-Barr virus in a human
lymphoblastoid cell line. J. Mol. Biol. 102:511-530.
Lippincott-Schwartz, J., J. G. Donaldson, A. Schweizer, E. G. Berger, H.-P.
Hauri, L. C. Yuan and R. D. Klusner. 1990. Microtubule-dependent retrograde transport
of proteins into the ER in the presence of brefeldin A suggests an ER recycling pathway.
Cell 60:821-836.
Little, S. P., J. T. Jofre, R. J. Courtney, and P. A. Schaffer. 1981. A
virion-associated glycoprotein essential for infectivity of herpes simplex virus type 1.
Virology 115:149-158.
Lonberg-Holm, K., and L. Philipson. 1974. Early interaction between animal
viruses and cells. Monographs in virology 9:1-149.
Lowry, O., N. Rosebrough, A. Farr and R. Randall. 1951. Protein measurement
with the Folin phenol reagent. J. Biol Chem. 193:265-275.

129
Lu, K.T., M. Zinnecker, T. Hamaska, D. H. Katz. 1979. New procedures for
preparation and isolation of conjugates of proteins and a synthetic copolymer of D-amino
acids and immunochemical characterization of such conjugates. Biochemistry
18:690-697.
Mackett, M., G. L. Smith and B. Moss. 1984. A general method for the
production and selection of infectious Vaccinia virus recombinants expressing foreign
genes. J. Virol. 49:857-864.
Mackett, M., M. J. Conway, J. R. Arrand, R. S. Haddad, and L. M.
Hutt-Fletcher. 1990. Characterization and expression of a glycoprotein encoded by the
Epstein-Barr virus BamHI.l fragmant. J. Virol. 64:2545-2552.
Maddon, P., A. Dalgeish, J. McDougal, P. Chapman, R. Weiss, and R. Axel.
1986. The T4 gene encodes the AIDS virus receptor and is exposed in the immune
system and the brain. Cell 47:333-348.
Maniatis, T, E. F. Fritsch and J. Sambrook. 1989. In: Molecular cloning: A
laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, N. Y.
Marsh, M., and A. Helenius. 1980. Adsorptive endocytosis of Semliki Forest
virus. J. Mol. Biol. 142:439-454.
Marsh M., E. Bolzau, J. White and A. Helenius. 1983. Interaction of Semliki
Forest virus spike glycoprotein rosettes and vesicles with cultured cells. J. Cell Biol.
96:455-461.
Matlin, S., H. Reggio, A. Helenius, and K. Simons. 1981. Infectious entry
pathway of influenza virus in a canine kidney cell line. J. Cell Biol. 91:601-613.
Matlin, S., H. Reggio, A. Helenius, and K. Simons. 1982. Pathways of vesicular
stomatitis virus entry leading to infection. J. Mol. Biol. 156:609-631.
Matthews, T. J., K. J. Weinhold, H. K. Lyerly, A. J. Langlois, H. Wigzell, and D.
P. Bolognesi. 1987. Interaction between the human T-cell lymphotropic virus type IIIB
envelope glycoprotein gpl20 and the surface antigen CD4: role of carbohydrate in
binding and cell fusion. Proc. Natl. Acad. Sci. USA 84:5424-5428.
Meyer, A., E. Petrovskis, W. Duffus, D. Thomsan, and L. Post. 1991. Cloning
and sequence of an infectious bovine rhinotracheitis (BHV-1) gene homologous to
glycoprotein H of herpes simplex virus. Biochim. Biophys. Acta 1090:267-269.
McClure, M.O., M. Marsh and R. A Weiss. 1988. Human immunodeficiency
virus infection of CD4-bearing cells occurs by a pH-independent mechanism. The
EMBOJ. 7:513-518.

130
McGeoch, D.J., A. Dolan, S. Donald, F. J. Rixon. 1985. Sequence determination
and genetic content of the short unique region in the genome of herpes simplex 1. J.
Mol. Biol. 181:1-13.
McGeoch, D. J. and A. J. Davison. 1986. DNA sequence of the herpes simplex
type 1 gene encoding glycoprotein gH and identification of homologues in the genomes
of varicella zoster virus and EBV. Nucl. Acids Res. 14:4281-4292.
McGeoch, D. J., M. A. Dalrymple, A. J. Davidson A. Dolan, M. C. Frame, D.
McNab, L. J. Perry, J. E. Scott and P. Taylor. 1988. The complete DNA sequence of the
long unique region in the genome of herpes simplex virus type 1. J. Gen. Virol.
96:1531-1574.
Merril, C. R., R. C. Switzer and M. L van Keuren. 1979. Trace polypeptides in
cellular extracts and human body fluids detected by two-dimensional electrophoresis and
a highly sensitive silver stain. Proc. Natl. Acad. Sci USA 76:4335-4338.
Miller, G., T. Shope, H. Lisio, D. Stitt and M. Lipman. 1972. Epstein-Barr virus:
Transformation, cytopathic changes, and viral antigens in squirrel monkey and marmoset
leukocytes. Proc. Natl. Acad. Sci. U.S.A. 69:383-387.
Miller, G., and M. Lipman, 1973. Comparison of the yield of infectious virus
from clones of human and simian lymphoblastoid lines transformed by Epstein-Barr
virus. J. Exp. Med. 138:1398-1412.
Miller, G., J. Robinsin, D. Heston, and M. Lipman. 1974. Differences between
laboratory strains of Epstein-Barr virus based on immortalization, abortive infection and
interference. Proc. Natl. Acad. Sci. USA 71:4006-4010.
Miller, N. and L. M. Hutt-Fletcher. 1988. A monoclonal antibody to
glycoprotein gp85 inhibits fusion but not attachment of Epstein-Barr Virus. J. Virol.
62:2366-2372.
Mitraki, A., B. Fane, C. Haase-Pettingell, J. Sturtevant, J. King. 1991. Global
suppression of protein folding defects and inclusion body formation. Science 253:54-58.
Moss, B. and P. L. Earl. 1989. Expression of proteins in mammalian cells using
vaccinia viral vectors, In: Current Protocols in Molecular Biology. John Wiley and
Sons. 16.15.1-3
Nemerow, G. R. and N. R. Cooper. 1984. Early events in the infection of human
B lymphocytes by EBV: the internalization process. Virology 132:186-198.

131
Nemerow, G. R., R. Wolfert, M. McNaughton, and N. R. Cooper. 1985.
Identification and characterization of the EBV receptor on human B lymphocytes and its
relationship to the C3d complement receptor CR2. J. Virol. 55:347-351.
Nicolson, L., A. Cullinane, D. Onions. 1990. The nucleotide sequence of an
equine herpesvirus 4 gene homologue of the herpes simplex virus type 1 glycoprotein H
gene. J. Gen. Virol. 71:1793-1800.
Niederman, J. C., A. S. Evans, L. Subrahmanyan, R. C. McCollum. 1970.
Prevalence, incidence, and persistence of EBV antibody in young adults. New Engl. J.
Med. 282:361-365.
Noble, A. G., G. Lee, and P. G. Spear. 1983. Anti-gD monoclonal antibodies
inhibit cell fusion induced by herpes simplex virus type 1. Virology 129:218-224.
Nonoyama, M. 1982. The molecular biology of Marek's disease herpesvirus. In:
B. Roizman (ed.), The herpesviruses, vol. 1. p. 333. Plenum Press, New York.
Oba, D. E. and L. M. Hutt-Fletcher. 1988. Induction of antibodies to the EBV
glycoprotein gp85 with a synthetic peptide corresponding to a sequence in the BXLF-2
open reading frame. J. Virol. 62:1108-1114.
Pachl, C., W. S. Probert, K. M. Hermsen, F. R. Masiarz, L. Rasmussen, T. C.
Merigan, and R. R. Spaete. 1989. The human cytomegalovirus strain Towne
glycoprotein H gene encodes glycoprotein p86. Virology 169:418-426.
Papahadjopoulos, D. 1978. Calcium-induced phase changes and fusion in natural
and model membranes. In: Membrane Fusion, Cell Surface Reviews, vol. 5 (ed.), G.
Poste and G. Nicholson, p. 765-790. Elsevier, North-Holland: Biomedical Press.
Pauza, C. D., and T. M. Price. 1988. Human Immunodeficiency virus infection of
T cells and monocytes proceeds via receptor-mediated endocytosis. J. Cell Biol.
107:959-968.
Pearse, B. 1975. Clathrin: a unique protein associated with intracellular transfer
of membrane by coated vesicles. Proc. Natl. Acad. Sci. USA 73:1255-1259.
Pellet, P. E., K. G. Kousoulas, L. Pereira, and B. Roizman. 1985. The anatomy of
the herpes simplex virus gB gene: primary sequence and predicted protein structure of
the wild type and of monoclonal antibody resistant mutants. J. Virol. 53:243-253.
Perez, L., M. McDonnell and E. B. Stephens. 1992. The transmembrane
glycoprotein of human immunodeficiency virus type 1 induces syncytium formation in
the absence of the receptor binding glycoprotein. J. Virol. 66:4134-4143.

132
Peeters, B., N. De Wind, R. Broer, A. Gieldens, and R. Moormann. 1992.
Glycoprotein H of pseudorabies virus is essential for entry and cell-to-cell spread of the
virus. J. Virol. 66:3888-3892.
Potts, W. M. and V. M. Vogt. 1987. A simple method for immunoaffinity
purification of nondenatured avian sarcoma and leukemia virus gag-containing proteins.
Virology 160:494-497.
Rademacher, T. W., R. B. Parekh and R. A. Dwek. 1988. Glycobiology. Ann.
Rev. Biochem. 57:785-838.
Reidel, H., C. Kondor-Koch, and H. Garoff. 1984. Cell surface expression of
fusogenic vesicular stomatitis virus G protein from cloned cDNA. Eur. Mol. Biol. Org.
J. 3:1477.
Richman, D. D., A. Buckmaster, S. Bell, C. Hodgman, and A. C. Minsonl. 1986.
Identification of a new glycoprotein of herpes simplex virus type 1 and genetic mapping
of the gene that codes for it. J. Virol. 57:647-655.
Rickinson, A.B. 1987. Cellular immunological responses to the virus infection.
In: M.A. Epstein and B.G. Achong (ed.), The Epstein-Barr Virus: Recent Advances, p.
77-125. John Wilet and Sons, Inc., New York.
Roizman, B. 1982. The family herpesviridae: general description, taxonomy and
classification. In: The herpesviruses, vol. 1., p. 1-23. B. Roizman (ed.), Plenum Press,
New York.
Roop, C., L. Hutchinson, and D. C. Johnson. 1993. A mutant herpes simplex
virus type 1 unable to express glycoprotein L cannot enter cells and its particles lack
glycoprotein H. J. Virol. 67:2285-2297.
Rose, J. K M. J. Welch, B. M. Sefton, F. S. Esch, and N. C. Ling. 1980.
Vesicular stomatitis virus glycoprotein is anchored in the viral membrane by a
hydrophobic domain near the COOH terminus. Proc. Natl. Acad. Sci. USA
77:3884-3888.
Rose, J.K. and C.J. Gallione. 1981. Nucleotide sequence of the mRNA encoding
the vesicular stomatitis virus G and M proteins determined from cDNA clones
containing the complete coding regions. J. Virol. 39:519-528.
Qualtiere, L. F., G. R. Pearson. 1979. EBV induced membrane antigens:
immunochemical characterization of Triton X-100 solubilized viral membrane antigens
from EBV-superinfected Raji cells. Int. J. Cancer 23:808-817.

133
Saiki, R. K., S. Scharf, F. Faloona, K. B. Mullin, G. T. Horn, H. A. Erlich and N.
Arnheim. 1985. Enzymatic amplification of beta-globin genomic sequences and
restriction analysis for diagnosis of sickle cell anemia. Science 230:1350-1354.
Saiki, R. K., D. M. Gelfand, S. Stoffel, S. J. Scharf, R. Higuchi, G. T. Horn, K.
B. Mullis and H. A. Erlich. 1988. Primer-directed enzymatic amplification of DNA with
a thermostable DNA polymerase. Science 239: 487-491.
Sambrood, Fritsch and Maniatis. 1989. Molecular Cloning: A Laboratory
Manual second edition. Cold Spring Harbor Laboratory Press.
Sarmiento, M., M. Haffey and P. Spear. 1979. Temperature sensitive mutant of
herpes simplex-1 gB adsorbs but does not penetrate. J. Virol. 29:1149-1158.
Scheid, A. and P. Choppin. 1974. Identification of biological activities of
paramyxovirus glycoproteins. Activation of cell fusion, hemolysis, and infectivity by
proteolytic cleavage of an inactive precursor protein of Sendai virus. Virology
57:475-490.
Scheid, A. and P. Choppin. 1976. Mutants of Sendai virus. Virology 9:265.
Scheid, A., M. C. Graves, S. M. Silver, and P. W. Choppin. 1978. Studies on the
structure and function of paramyxovirus glycoproteins. In: B.W.J. Mahy and R.D. Barry
(ed.), Negative strand viruses and the host cell. p. 181-193. Academic Press, London.
Schlegel, R., R. B. Dickson, M. C. Willingham and I. H. Pastan. 1982.
Amantadine and dansylcadaverine inhibit vesicular stomatitis virus uptake and
receptor-mediated endocytosis of alpha 2-macroglobulin. Proc. Natl. Acad. Sci.
79:2292-2295.
Seigneurin, J. M., M. Vuillaume, G. Lenour, and G. de-The. 1977. Replication of
EBV: ultrastructural and immunofluorescent studies of P3HR-1 superinfected Raji cells.
J. Virol. 24:835-845.
Sixbey, J. W., D. S. Davis, L. S. Young, L. M. Hutt-Fletcher, T. F. Tedder and A.
B. Rickinson. 1987. Human epithelial cell expression of an Epstein-Barr virus receptor.
J. Gen. Virol. 68:805-811.
Sixbey, J. W., J. G. Nedrud, N. Raab-Traub, R. A. Hanes, and J. S. Pagano. 1984
Epstein-Barr virus replication in oropharyngeal epithelial cells. New Engl. J. Med.
310:1225-1230.
Sixbey, J. W., E. H. Vesterinen, J. G. Nedrud, N. Raab-Traub, L. A. Walton and
J. S. Pagano. 1983. Replication of Epstein-Barr virus in human epithelial cells in vitro.
Nature (London) 306:480-483.

134
Spear, P.G. 1980. Herpesviruses. In: H.A. Blough and J.M. Tiffany (ed.), Cell
membranes and virus envelopes, vol. 2. p. 709. Academic Press, London.
Stannard, L. M., A. O. Fuller, and P. G. Spear. 1987. Herpes simplex virus
glycoprotein associated with different morphological entities projecting from the virion
envelope. J. Gen. Virol. 68:715-725.
Stein, B. S., S. D. Gowda, J. D. Lifson, R. C. Penhallow, K. G. Bensch, E. G.
Engleman. 1987. pH-independent HIV entry into CD4-positive T cells via virus fusion
to the plasma membrane. Cell 49:656-668.
Stephens, E. B., E. Butfiloski and E. Monck,. 1992. Analysis of the amino
terminal presequence of the feline immunodeficiency virus glycoprotein:effect of
deletions on the extracellular transport of gp95. Virology 190:569-578.
Stmad, B.C., T. Schuster, R. Klein, R. F. Hopkins, T. Witmer, R. H. Neubauer,
and H. Rabin. 1982. Production and characterization of monoclonal antibodies against
the EBV membrane antigen. J. Virol. 41:258-264.
Stmad, B.C., M.R. Adams, and H. Rabin. 1983. Glycosylation pathways of two
major Epstein-Barr virus membrane antigens. Virology 127:168-176.
Svedmer, E., I. Emberg, J. Seeley, O. Weiland, G. Masucci, K. Tsukuda, R.
Szigeti, M. G, Masucci, H. Blomgren, W. Berthold, W. Henle, and G. Klein. 1984.
Virologic, immunologic, and clinical observations on a patient during the incubation,
acute, and convalescent phases of infectious mononucleosis. Clin. Immunol.
Immunopathol. 30:437-450.
Tabor S.and C. C. Richardson. 1987. DNA sequence analysis with a modified
bacteriophage T7 DNA polymerase. Proc. Natl. AcadSci. USA 84:4767-71.
Takada, K. 1984. Cross-linking of cell surface immunoglobulin induces
Epstein-Barr virus in Burkitt lymphoma lines. Int. J. Cancer 33:27-32.
Takada, K. and Y. Ono. 1989. Synchronous and Sequential Activation of
Latently Infected Epstein-Barr Virus Genomes. J. Virol. 63:445-49.
Thorley-Lawson, D. A. 1988. Basic Virological Aspects of Epstein-Barr Virus
Infection. Seminars in Hematology 25:247-260.
Thorley-Lawson, D. A. and C. M. Edson. 1979. Polypeptides of the Epstein-Barr
virus membrane antigen complex. J. Virol. 32:458-467.

135
Towbin, H. T., Staehelin, and J Gordon. 1979. Electrophoretic transfer of
proteins from polyacrylamide gels to nitrocellulose sheets: Procedure and some
application. Proc. Natl. Acad. Sci. USA 76:4350-4354.
Uchida, N., H. Smilowitz and M. L. Tanzer. 1979. Proc. Natl. Acad. Sci. USA
76:1868-72.
Voller, A., D. Bidwell, and A. Bartlett. 1976. Microplate enzyme immunoassays
for the immunodiagnosis of virus infections, P 506-512. In: N.R. Rose and H. Friedman
(ed.) Manual of Clinical Immunology. American Society for Microbiology,
Washington,D.C.
Walker, B. D., M. Kowalski, W. C. Goh, K. Kozarsky, M. Krieger, C. Rosen, L.
Rohrschneider, W. A. Haseltine and J. Sodroski. 1987. Inhibition of human
immunodeficiency virus syncytium formation and virus replication by castanospermine.
Proc. Natl. Acad. Sci. USA 84:8120-8124.
Walter, G. 1986. Production and use of antibodies against synthetic peptides. J
Immunol. Methods 88: 149-161.
Watson, R. J., J. H. Weis, J. S. Salstrom, and L. W. Enquist. 1982. Herpes
simplex virus type-1 glycoprotein D gene: nucleotide sequence and expression in
Escherichia coli. Science 218:381-384.
Welch, W. J. and B.M. Sefton. 1979. Two small virus-specific polypeptides are
produced during infection with Sinbis virus. J. Virol. 29:1186-1195.
White, J., A. Helenius, and J. Kartenbeck. 1982. Membrane fusion ability of
influenza virus. EMBO J. 1:217-222.
White, J., J. Kartenbeck, A. Helenius. 1980. Fusion of Semliki Forest Virus with
the plasma membrane can be induced by low pH. J. Cell Biol. 87:264-272.
White, J., M. Kielian, and A. Helenius. 1983. Membrane fusion proteins of
enveloped animal viruses. Q. Rev. Biophys. 16:151-195.
White, J., Matlin, and A. Helenius. 1981. Cell fusion by Semliki Forest virus,
influenza, and vesicular stomatitis virus. J. Cell Biol. 89:674-679.
Young L. S., C. W. Dawson, K.W. Brown and A. B. Rickinson. 1989.
Identification of a human epithelial cell surface protein sharing an epitope with the
C3d/Epstein-Barr virus receptor molecule of B lymphocytes. Int. J. Cancer 43:786.
Zhou, C., Y. Yang and A. Y. Jong. 1990. Mini-Prep in Ten Minutes.
Biotechniques 8.

136
Zur Hausen, H., F. J. O'Neill and V. K. Freese. 1978. Persisting oncogenic
herpesviruses induced by tumor promoter TPA. Nature (London) 272:373-375.

BIOGRAPHICAL SKETCH
The candidate, Linda Ruth Yaswen, graduated from Commack High School
South in 1979. She earned a Bachelor of Science degree majoring in medical technology
from the State University of New York at Stony Brook in 1984. In 1984 through
February 1985, she worked as a medical technologist in the Department of Histology at
the State University Hospital at Stony Brook New York. In March of 1985 through July
1986 she was the evening supervisor of the Clinical Chemisty laboratory at the same
hospital. In August of 1986 through July of 1987 she was a research assistant in the
laboratory of Dr. William Mancini. This research was conducted on the development of
anti-cancer drugs at the University of Michigan, Department of Internal Medicine. Her
graduate work at the University of Florida began in 1987 as a student in the Department
of Pathology and Laboratory Medicine. She became a Ph.D. candidate in 1989 in the
same department. Her research was conducted under the excellent guidance of Dr.
Lindsey Hutt-Fletcher in the Department of Comparative and Experimental Pathology.
137

I certify that 1 have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Lindsey
Lindsey M. Hutt-Fletcher, Chair
Professor of Pathology and Laboratory
Medicine
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Assistant Professor of Veterinary Medicine
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Madreen Goodeno^-
Assistant Professor of Pathology and
Laboratory Medicine

I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Ammon P§
Associate Professor of Pathology and
Laboratory Medicine
I certify that 1 have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Edward Wakeland
Professor of Pathology and Laboratory
Medicine