Citation
The gamma-glutamyl transpeptidase enzyme is associated with lymphocyte activation proteins

Material Information

Title:
The gamma-glutamyl transpeptidase enzyme is associated with lymphocyte activation proteins
Creator:
Fletcher, Dana Rene
Place of Publication:
Denver, CO
Publisher:
University of Colorado Denver
Publication Date:
Language:
English
Physical Description:
x, 66 leaves : illustrations (some color) ; 29 cm

Subjects

Subjects / Keywords:
Clinical enzymology ( lcsh )
Autoimmune diseases -- Research ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 61-66).
Thesis:
Submitted in partial fulfillment of the requirements for the degree, Master of Arts, Biology
General Note:
Department of Integrative Biology
Statement of Responsibility:
by Dana Rene Fletcher.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
37899871 ( OCLC )
ocm37899871
Classification:
LD1190.L45 1997m .F54 ( lcc )

Downloads

This item has the following downloads:


Full Text
THE GAMMA-GLUTAMYL TRANSPEPTIDASE ENZYME IS ASSOCIATED
WITH LYMPHOCYTE ACTIVATION PROTEINS
by
Dana Rene Fletcher
B.S., Pennsylvania State University, 1995
A thesis submitted to the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Master of Arts
Biology
1997
AL


This thesis for the Master of Arts
degree by
Dana Rene Fletcher
has been approved
by
Teresa Audesirk


Fletcher, Dana Rene (M.A., Biology)
The Gamma-Glutamyl Transpeptidase Enzyme is Associated With Lymphocyte
Activation Proteins
Thesis directed by Associate Professor Bradley J. Stith
A monoclonal antibody has been developed during the search for an inhibitor of B
cell activation through the complement receptor type 2. This antibody recognizes
the gamma-glutamyl transpeptidase enzyme that is involved in the cycling of the
antioxidant tripeptide glutathione. This project has employed immunofluorescent
staining techniques to look at characteristics of the gamma-glutamyl transpeptidase
enzyme on peripheral blood lymphocytes and a tumor cell line. By flow
cytometry, gamma-glutamyl transpeptidase is present on a population of peripheral
blood B lymphocytes. It is also found in higher amounts on the surfaces of
activated/memory phenotypes of T cytotoxic cells and T helper cells than on
unactivated/naive phenotypes of those T cells. Confocal Microscopy confirms
results of other methodologies which show physical associations between gamma-
glutamyl transpeptidase and members of the complement-induced B cell signal
transduction complex. The presence of gamma-glutamyl transpeptidase on
activated phenotypes of T cells and its association with members of a protein
complex involved with B cell activation supports the hypothesis that this enzyme is
involved with some aspect of lymphocyte activation.
This abstract accurately represents the content of the candidates thesis. I
recommend its publication.
ABSTRACT
Signed
hi


DEDICATION
I would like to dedicate this thesis
to all those who have smiled at the glistening snow;
to those who have found themselves in a stormy sea and swam;
and to those who have heard the whisper of reason and listened.


ACKNOWLEDGEMENTS
My thanks to the UCHSC Department of Rheumatology for their technical and
financial support and to the Arthritis Foundation grant which funded this project. I
would like to thank Michael Holers, who has allowed me to use the research Ive
conducted while working in his lab, for this thesis.
Special thanks to my committee members Bradley Stith, Ellen Levy, and Teresa
Audesirk for their technical advice and dedication to their students.
I would like to thank Timothy Nichols who has been instrumental in my research
education as well as a champion for my success.
In addition, I would like to thank my family and friends for their continued support
during my life and during this Masters program.


CONTENTS
List of Figures......................................................... viii
List of Tables............................................................. x
Chapter
1. Introduction....................................................... 1
1.1 Perspective............................................................ 1
1.2 Relevance of Project................................................. 2
1.3 The Immune System: A Brief Review..................................... 3
1.4 Technical Background................................................. 10
1.4.1 A Model for a Signal Transducing Complex in B Cell Activation.... 10
1.4.2 Complement Receptor Type II......................................... 11
1.4.3 An Amplification Molecule: CD19..................................... 15
1.4.4 The Tetraspan Family................................................ 17
1.4.5 Leu-13.............................................................. 19
1.4.6 Interactions Between the CR2 Signalling Complex Molecules........... 20
1.4.7 Events Following Activation Through the CR2 Signalling Complex .... 21
1.4.8 Development of An Antibody which Inhibits Homotypic Adhesion..... 25
VI


CONTENTS (continued)
1.4.9 Gamma-Glutamyl Transpeptidase is the Target of the mAb 3A8.... 28
1.4.10 Gamma-Glutamyl Transpeptidase: Structure and Function........... 31
1.5 Hypotheses......................................................... 33
2. Methods............................................................ 35
2.1 Expression of Gamma-Glutamyl Transpeptidase on Peripheral Blood
Lymphocytes............................................................ 35
2.2 Association of Gamma-Glutamyl Transpeptidase with the CR2 Signal
Transduction Complex................................................... 37
3. Results............................................................ 40
3.1 Expression of GGT on Human Lymphocytes.................... 40
3.2 Association of GGT with CR2 Signalling Complex Members............. 47
4. Discussion......................................................... 53
Bibliography .......................................................... 61
vii


LIST OF FIGURES
1.1 The B Cell Immune Response.............................................. 5
1.2 The T Cell Primary Response............................................. 7
1.3 The Complement Cascade: Classical and Alternative Pathways.............. 9
1.4 A Model for Signal Amplification in B Cells............................ 11
1.5 A Model of the Proposed CR2 Structure.................................. 13
1.6 A Model of the Proposed Structure of CD19.............................. 16
1.7 A Model of the Proposed Structure of the TMSF4 Members................. 18
1.8 Associations of the Molecules of the CR2 Signalling Complex............ 20
1.9 An in vitro Model for the Activation of B Cells ....................... 23
1.10 Inhibition of Homotypic Adhesion by mAb 3A8........................ 27
1.11 The mAb 3A8 Recognizes a Two Protein Complex........................ 29
1.12 GGT Activity In Immunoprecipitates.................................. 30
1.13 A Model for the Structure of GGT................................... 31
1.14 The Gamma-Glutamyl Cycle........................................... 33
3.1 An Example of a Size Exclusion Scatter Plot Using FACS Analysis..... 41
3.2 GGT Expression On B Cells........................................... 42
3.3 GGTs Presence on Subsets of T Cells................................ 45
viii


LIST OF FIGURES (continued)
3.4 Merged Confocal Images of CR2 and GGT............................. 48
3.5 Merged Confocal Images of CD 19 and GGT............................ 49
3.6 Merged Confocal Images of CD81 and GGT............................ 50
3.7 Merged Confocal Images of CD82 and GGT............................ 51
3.8 Merged Confocal Images of CD53 and GGT............................ 52
IX


LIST OF TABLES
3.1 Percent of T Cell Subsets Which Are GGT Positive
X


1. Introduction
1.1 Perspective
In the journey from the primordial soup, the Earth has evolved complex
ecological systems where an organisms survival depends on its ability to play roles
of both aggressive predator and elusive prey. All creatures, from the virus to the
human, have developed intricate mechanisms of attack and protection. Many times
multiple mechanisms serve as checks and balances within one organism. For
example, proteins have been found to have more than one function, and one
function can have more than one protein controlling it (1).
These complex arrangements of overlapping systems make characterization
of our species difficult to complete. Each year more discrete pieces of the puzzle
of life are put into place. We can benefit greatly from this information. Not only
can we better understand ourselves and our heritage, but we can understand what
happens when one or more of these mechanisms goes awry. Even our complex
protective mechanisms can make mistakes and allow altered cells to wreak havoc
within us. The mechanisms by which our bodies launch attacks against foreign
invaders and monitor the health of the billions of cells of which we are composed
l


are called the Immune System. This paper deals with a subset of this system which
has been studied extensively for its role in diseases where the immune system
launches an inappropriate attack on the body it is supposed to protect.
1.2 Relevance of Project
Autoimmune diseases, a wide variety of illnesses which include rheumatoid
arthritis and systemic lupus erythematosus, present complications that result from
inappropriate activation of ones own immune system by normal cells (2). In the
near future, treatments for these diseases may include molecules which block
inappropriate stimulation of different aspects of the immune system. This project
began with the goal of identifying potential targets for such a treatment. Our lab
studies both structural and functional aspects of the complement system of the
immune system. Through our research of complement, we developed an antibody
which inhibits normal immune function. This antibody recognizes a cell surface
enzyme that is involved with the cycling of an antioxidant tripeptide. Previously,
this enzyme and the process of activating the immune system had not been linked.
The goal of this project was to clarify the association between this enzyme and the
activation of immune cells.
2


1.3 The Immune System: A Brief Review
The immune system is a complex network which acts to protect the body.
This research focuses on activation of the immune response through a mechanism
that incorporates the three major components of the immune system: B cells, T
cells, and the complement cascade.
B cells are a type of white blood cell called lymphocytes that are responsible
for the production of proteins called antibodies. Antibodies circulate in the blood
and recognize particles that need to be removed. These particles, regardless of
their origin, are called antigens. Commonly, they are thought of as invading
bacteria and viruses, but often these antigens are parts of our own altered cells and
occasionally are parts of our normal cells. In addition to secreting antibodies, B
cells have molecules attached to their surfaces which recognize antigens. These
molecules are called antigen receptors or B cell receptors. When a B cell comes in
contact with an antigen, the antigen will bind to the antigen receptor on the B cell.
The B cell then internalizes the antigen and the receptor, enzymatically degrades the
antigen, and distributes small parts of the antigen on its surface. Immune cells
which distribute foreign particles on their surfaces like this are called Antigen-
Presenting Cells (APCs). The antigen pieces on APCs are associated with a
molecule which is unique to each person and allows the immune system to identify
self. After B cells bind to the antigen which its antigen receptor specifically
3


recognizes, the B cell becomes activated. Sometimes they have help becoming
activated by other cells or proteins of the immune system. Those mechanisms will
be discussed in the following paragraphs.
Once activated, the B cells differentiate into one of two types of cells (see
Figure 1.1). The first resulting cell type is the antibody-producer called a plasma
cell. These cells are short-lived secretion machines which have been estimated to
produce 2000 molecules of antibody per second (3). Plasma cells produce
antibodies which recognize the specific antigen to which the original B cell was
exposed. This initial recognition and antibody production is called a primary
immune response because the antigen had not been seen by the immune system
before. The second type of cell are memory B cells. These long-lived cells are
programmed to remember the specific antigen with which it has come in contact so
that if the memory cell sees the same antigen again, the response to the invasion can
be carried out much more quickly.
T cells are lymphocytes like B cells. These immune cells fall into two main
subtypes which can be identified by unique cell surface protein markers, designated
as CD numbers. A CD number is a name given to a molecule by an
international conference which confers a cluster of differentiation (CD) number to
molecules under stringent guidelines (3).
4


Naive
B Cell
Primary
Response
Secondary
Response
1st contact
with, antigen
4
Figure 1.1. The B cell immune response. When activated By antigen,
a £ cell will-produce plasma cells and memory cells during die primary
response. In die secondary response, memory cells can differentiate more
quickly into plasma cells and more memory cells.
This system of naming arose because, independently of one another, numerous
laboratories began engineering antibodies which recognized different portions
(called epitopes) of various molecules. Each laboratory had its own name for any
given protein. The CD numbering system allows for consistent naming of these
proteins even if the antibodies used to detect those molecules are different.
5


One subset of T cells is called T cytotoxic cells which can be identified by
the cell surface marker CD8. Active T cytotoxic cells directly kill non-immune
cells which are infected or damaged and which have antigen particles expressed
on their surfaces. The T cytotoxic cell binds to the altered cell and disrupts its
membrane, killing it.
The other main subtype of T cells is the T helper cell which can be
identified by the surface protein CD4. These cells are aptly named because they
help to activate the other types of immune cells. They recognize and bind to the
antigen which is on the surface of APCs (see Figure 1.2).
Once this binding occurs, the T helper cells become activated and begin to
divide and produce small proteins (cytokines) which circulate in the blood to
activate B cells, T cytotoxic cells, and other T helper cells. These cytokines also
augment the stimulation of the T helper cell which produced them. Both the
production of cytokines and actual binding to the antigen-presenting cell (B cell)
help to activate the B cell response. Activated T cells of both types also produce
memory T cells which function similarly to memory B cells.
6



naive
T helper cell ( 1 h ft* APC
Activated T htlptr cell..^
produces cytokines ( Tl F=f ^
T helper cells, T cytotoxic
cells, and B cells ' ^
y/f
cytokines stimulate other / l \
.-(S
cr
Bccfl
i
o

T cytotoxic cell
vilK altered body cell
activated T helper cells
produce memory T cells
Figure 1.2. T cell activation pathway during flic primary response. The
memory T cells produced wdl respond quickly to a second presentation
of antigen.
By looking at cell surface markers expressed on T cells, we can distinguish
between activated and non-activated phenotypes. The cell surface marker for
unactivated T cells is CD45RA. The cell surface marker for activated / memory T
cells is CD45RO. These two cell surface markers are different forms of a protein
called CD45. The CD45 protein is a membrane-spanning molecule which has
protein tyrosine phosphatase activity (PTP).
7


While B cells and T cells are functioning to recognize antigens, the third
component of the immune system joins the fray. This pathway is the called the
complement cascade. Websters definition of complement is that which
completes or perfects. The complement cascade acts via two distinct pathways to
augment the cellular mechanisms by which antigen is removed from the body: the
classical pathway and the alternative pathway. In the classical pathway, an
antibody which has already been produced from a plasma cell tags an antigen on a
damaged or infected cell. Complement proteins bind to the antibody / antigen
complex on the cell and induce the series of proteolytic cleavages which activate
other complement proteins to produce the complement cascade. This sequential
protein cleavage pathway leads to ultimate lysis of the antibody-defined target cell.
Like the classical pathway, the alternative pathway ultimately can lead to
lysis of an altered cell. Unlike the classical pathway, the alternative pathway
activates and recruits B cells to help the body recognize weak antigens. The
alternative pathway begins with complement proteins binding to antigen instead of
antibody-tagged antigen. The resulting complement protein cascade produces
fragments which circulate in the blood or bind to cells. The complement fragments
bound directly to the antigen on cells induce B cell / T cell activity known as
cellular activation events. Since explicit antibodies are not produced until cellular
activation occurs, the alternative pathway is very important in the early
8


identification of low level antigen. It is also important in the removal of pathogens
whose surfaces are not recognized easily as foreign by antigen-presenting cells
(3,4). Figure 1.3 demonstrates the complement cascade.
Classical Pathway
1
complement proteins
hind to antibodies
Alternative Pathway
I
complement proteins
Iwnd tD antigens
and start a cascade of
protein reactions
\
a protein complex
is assembled which
Iritis die cell
and start a cascade of
protein reactions
canses activation
Of imme cells
Figure 1.3. The Complement Cascade: classical and alternative pathways.
The overlapping interactions of the Complement cascade, B cells, and T
cells are an excellent example of the type of checks and balances found in nature.
9


This research deals with activation of the immune response through the
convergence of these three immune pathways. This convergence of pathways
produces a synergistic activation of B cells that results in a more efficient immune
system.
1.4 Technical Background
1.4.1 A Model for a Signal Transducing Complex in B Cell Activation
The molecular mechanisms by which B cells become activated are quite
complicated. Recently, four to six integral membrane proteins have been found to
associate non-covalently on the surface of B cells and to have related functions.
This multimolecular protein complex mediates cell growth, motility, and cell-cell
interaction. Four components of this complex have been identified by monoclonal
antibodies purified antibodies which are produced to recognize a specific peptide
sequence, or epitope, of a molecule. This group of proteins is collectively called the
CR2 signal transduction complex (5,6,7,8,9).
The basic model is that a complement fragment will bind to a cell or clump
of debris and antibodies, and become an immune complex, which then binds to a
complement receptor on a B cell (see Figure 1.4). The immune complex bound to
the complement receptor is physically close to an antigen receptor. The antigen
receptor will bind to the localized immune complex. These two binding events
10


transduce signals through the CR2 signalling complex which result in a synergistic
effect called cross-linking. Chapters 1.4.2 through 1.4.6 will discuss the properties
of each protein and how they relate to one another.
Figure 1.4. A model for signal amplification in B cells. (1) A fragment of a complement
protein binds to its receptor (2) on the B cell (Ch, 1.4.2). (3) The antigen receptor of
the B cell binds to an antigen on an immune complex. Then growth signals are
modulated by the other members of the signalling complex (4; Ch. 1.4.3. 5; Ch. 1.4.4,
and 6; Ch. 1.4.5) and transduced into the cell.
1.4.2 Complement Receptor Type 2
Complement Receptor Type 2 (see number 2 in Figure 1.4) is a member of
the Regulators of Complement Activation family (RCA) which is a group of
li


proteins responsible for controlling cell activation by Complement cleavage
products. This group includes Complement Receptor Type 1 (CD35), membrane
cofactor protein (MCP, CD46), decay accelerating factor (DAF, CD55), C4b-
binding protein, and factor H in addition to Complement Receptor Type 2.
Alternate names for this protein are CR2 or CD21.
CR2 is an integral membrane protein found mainly on mature B cells and
follicular dendritic cells (FDC), but also is found on cells of the thymus,
endothelial cells, and subsets of T helper and T cytotoxic lymphocytes (10).
Extracellularly, this 145 kDa glycoprotein consists of 15 or 16 domains called
SCRs or short consensus repeats (see Figure 1.5). These areas are composed of 60
to 70 amino acid sequences of which eleven have potential N-glycosylation sites.
N-glycosylation sites are amino acid residues which are bound to oligosaccharide
side chains and can be recognition sites. Sometimes they are associated with cell-
cell interactions (11). All of the SCRs have similar amino acid sequences.
12


^ taos wmuwrasTisw iftiwwwisftHJwwijCT wwts tss mb
COOK
Figure l.S. A model of the proposed CR2 secondary structure.
Note the repealing motifs called short consensus repeats (SCRs).
Red areas indicate disulfide bonds and green areas indicate
potential N glycosylation sites.
Each one has between 20-40% homology with the other SCRs of CR2. Conserved
in each SCR are four cysteines and one tryptophan. These areas are structurally
significant within each domain as they confer the same secondary structure among
the SCRs. A short transmembrane section of CR2 connects the SCRs to a 34
amino acid long C-terminal cytoplasmic domain (3,9,12,13,14).
CR2 is the receptor for several proteins (also called ligands). Each of these
ligands binds to an SCR region which varies with the ligand. The primary
physiological ligand for CR2 is a proteolytically cleaved fragment produced during
the Complement cascade (see number 1 in Figure 1.4). This fragment is called
C3dg (3,15).
13


C3dg is formed during a regulatory process of the Complement cascade.
The Alternative and Classical pathways of the Complement cascade converge when
they reach a protein called C3. C3 is broken down to produce C3b which is
membrane bound. This protein is required in both pathways to initiate the
formation of a membrane attack complex which kills the cell. The production of
this attack complex is blocked when CR1, Factor H, or MCP ( all members of the
RCA family) binds to C3b. This allows C3b cleavage into two new fragments.
One remains bound to the membrane and is further cleaved to produce C3dg
(membrane bound) (4). The result of these cleavage reactions is that direct lysis of
the target cell has been averted by blocking complement. However, C3dg remains
on the surface of the cell and binds to the CR2 receptor of immune cells. Binding
of CR2 activates pathways in immune cells and tags the cell for destruction so that
complete clearance of the antigen can occur (3,4,16,17). The regulation of the
Complement cascade by the production of C3dg, and its consequent binding to CR2
is of great significance because it allows regulation without sacrificing efficiency.
Not only is CR2 needed for signalling initiated by Complement, but it is
needed for T helper cells to interact with the self-recognition complex found on B
cells. By themselves, some antigens do not elicit a very strong response from B
cells. In order to initiate a robust response to these antigens, the B cells must
interact directly with T helper cell. These are called T-dependent antigens (3).
14


Mice which are genetically CR2 deficient produce lower amounts of antibodies in
response to the model T-cell-dependent antigen source, sheep red blood cells, than
their CR2 sufficient counterparts. Since antibody production relies on B
lymphocytes, these studies reflect the importance of CR2 in the activation of those
cells by T helper cells (18,19).
1.4.3 An Amplification Molecule: CD 19
Commonly used as a cell surface marker for B cells, another member of the
complex is referred to as CD19 (see number 4 in Figure 1.4). CD19 is a 95 kDa
integral membrane protein which is expressed only on B cells. It is an excellent
marker for B cells as it is found from early in their development to plasma cell
differentiation. The extracellular portion of CD19 is approximately 273 amino
acid residues long and has five potential N-glycosylation sites ( see Figure 1.6).
15


Figure 1.6. A model for toe proposed secondary structure of CD19.
Hie red dots indicate disulfide bonds and the green areas indicate
potential Nglycosylation sites.
CD19 of human, mouse, and guinea pig is highly conserved (79% similar between
human and mouse) in its 240 amino acid cytoplasmic domain (3). Some localized
regions of the cytoplasmic tail have a strong net negative charge. Of the tails
amino acids, 19% have acidic R groups whereas only 9.5% have basic R groups.
The preponderance of acidic R groups in the cytoplasmic tail domain may explain
the function of CD19 as a signal transduction molecule. Signalling through CD19
appears to be very complex. Binding the extracellular portion of multiple CD 19
molecules with antibodies (cross-linking) can result in growth and proliferation or
in inhibition depending on the type and amount of antibodies used. Activation of
protein tyrosine kinases (PTK), increases in the level of phospholipase C (PLC) and
16


increased intracellular concentrations of free calcium ions ([Ca2+]i) which result
from cross-linking CD 19 with antibodies suggest that CD 19 has a role in
proliferation and growth in B cells. However, binding of this CD19 with anti-CD19
antibodies alone inhibits the [Ca2+] increase produced by mitogen stimulation
(3,5,9,21). Mitogens are chemicals which induce cell division. When CD19 was
crosslinked with the antigen receptor of the B cell, only 0.03% of the antigen
receptor needed to be bound to cause the cells to proliferate (20). This result
suggests that CD 19 acts in a very subtle, condition-dependent manner to amplify
weak signals coming through the B cell antigen receptor (6).
1.4.4 The Tetraspan Family
The CR2 signal transduction complex contains at least one of the members
of this family of proteins (see number 5 in Figure 1.4). The first protein classified
in this family and found to be associated with the signalling complex was a 26 kDa
protein named TAPA-1 for Target of an Anti-Proliferative Antibody. It was
characterized by Oren et al. in 1990 (22). It has been given the cluster designation
of CD81. In addition to CD81, two other members of this family of proteins may
be associated with the CR2 signalling complex. These two are named CD82 and
CD53 (7,8,23,24).
17


Each of the members of this protein family has a proposed structure which
crosses the membrane four times and thus are referred to also as Transmembrane
Super Family 4 (TMSF4) members. The proposed TMSF4 model (see Figure
1.7) is comprised of four hydrophobic regions which are transmembrane domains,
two large loop-like extracellular domains, and cytoplasmic N- and C- termini
(7,22,25,26). The amino acid sequence is 91% conserved between human and
mouse CD81. Unlike other members of the tetraspan family, there are no N-
glycosylation sites on its extracellular domains (9,27).
figure 1.7. A cartoon of the model proposed for the TMSF4
members inctndmg CD81, CD82, CD53, and others. There are
no potential N-gLycosylation sites on CD81. Red dots indicate
sulfur containing residues which may conidtute to conformation.
18


The TMSF4 superfamily proteins have structural similarities to a family of
gated ion channels found in the nervous system which includes receptors for the
neurotransmitters glutamate and acetylcholine. These similarities suggest that the
two groups may have related biochemistries, although no direct evidence has been
presented yet (26,28).
The TMSF4 family of proteins has a rather widespread tissue distribution.
CD81 is found ubiquitously in B cells but its expression is increased significantly
following activation. Antibodies to TAPA-1 have been shown to inhibit cell
proliferation and be involved with activation (9,27). There is evidence that CD81
is the molecule of the complex which is responsible for cell-cell interactions,
specifically the clumping seen during the activation of B cells (6).
1.4.5 Leu-13
The final member of the CR2 signalling complex is an integral membrane protein
whose primary amino acid structure has just recently been determined (see number
6 in Figure 1.4) (29,30). This protein is 16-17 kDa and is called Leu-13. A model
for its secondary structure has not been proposed at this time. It is found on some
white blood cells, endothelial cells and a specialized portion of the placenta. It
physically associates with CD81 in some cell lines (31). Like CD81, antibodies to
Leu-13 promote clumping (see Ch. 1.4.7) and inhibit proliferation. A cytokine
19


called Interferon-oc (IFN-a) causes an increase in the amount of Leu-13 on the
cell surface (32,33). It has been suggested that the complex formed by Leu-13 and
CD81 can associate with the complex of CR2 and CD 19 to mediate cell adhesion
events (6,9).
1.4.6 Interactions between the CR2 Signalling Complex Molecules
Figure 1.8 depicts the associative relationships between the members of this
complex. The circles represent the population of the corresponding molecules.
Overlapping areas indicate which molecules have the strongest physical associations
with one another.
logon; 1.8. Representation of file stronger associations Tie tween members of the
CE2 Signalling Conqdex. Other associations ate not shown.
20


The diagram does not represent actual amounts of the molecules nor weak
associations between the molecules of the complex. CR2 and CD 19 are non-
covalently associated on the B cell surface. CD81 associates with both Leu-13 and
CD19, but only directly with CR2 in some cell lines. Potential associations with
proteins outside of this signal transduction complex are indicated by white areas but
are not defined for the purpose of this paper.
This macromolecular complex is comprised of molecules which also can be
found alone or in complexes with different proteins. Under certain conditions in B
cells, these molecules come together into an association which regulates the
activation of those cells. This small population of cells provides an extremely
important source of immune activation because sometimes an antigen is found in
low concentrations or is not recognized very well by the immune system. In these
situations, the CR2 signalling complex acts to lower the threshold of B cell
activation so that even small stimuli can produce an antibody response (6).
1.4.7 Events Following Activation Through the CR2 Signalling Complex
In the body, blood is constantly cycled through secondary lymphoid tissues.
These tissues, which include the spleen and lymph nodes, collect free antigens and
antigen-antibody complexes from the blood. These immune complexes bind to
follicular dendritic cells (FDCs) which are antigen-presenting cells in the secondary
21


lymphoid tissues. Antigens stay on the FDCs for long periods of time (sometimes
years) before they are degraded. As blood is cycling through the lymph tissues, B
cells which have not been exposed to an antigen bind to the immune complexes via
the model described previously (CR2-C3dg and antigen-antigen receptor). This
initiates signalling events like ones through the CR2 signalling complex which
activates the B cells. Each B cell begins to divide and quickly produces a colony of
cells. This newly formed mass of densely aggregated cells is called the B cell
germinal center. In each germinal center there are only 2 or 3 original B cells
which produce the large colonies (oligoclonal expansion). Memory T helper cells
which are found in the germinal centers are thought to help segregate the
proliferating B cells into types which have high-affinity or low-affinity receptors
for antigen. The high-affinity B cells proliferate into memory cells and plasma
cells. The low-affinity B cells undergo programmed cell death (apoptosis), thus
removing weakly interacting cells from the immunologically active population (3).
For experimental purposes an in vitro assay has been developed which
mimics in vivo cell activation responses of the type that are involved in germinal
center B cell proliferation. When immortalized B cell lines which express the CR2
signalling complex are incubated with immune complexes (immune complexes can
be simulated by covalently linking purified C3dg molecules), a germinal center-like
phenomenon occurs. Within an hour, the cells will change from their normal
22


dispersed appearance to a densely clumped state (see Figure 1.9). This model for
clumping of same-type cells is named homotypic adhesion.
Figure 1.9. An in vilro model for the activation of B cells. Polymerized C3dg
is bound to CR2 on a tumor cell line. Within an hour (he cells dump together
via an unknown mechanism. This phenomenon is called homotypic adhesion.
It is used widely to study B cell activation because it mimics the oligoclonal
populations of the B cell germinal centers (3,34). The clumping activity can be
produced using a wide panel of antibodies to CD81 and Leu-13. A limited number
of proteins that bind to CR2 have been found to induce the clumping model,
homotypic adhesion, in culture. These proteins are a subset of proteins which
recognize the four SCRs at the amino-terminal end of the molecule and include
polymerized C3dg and a monoclonal antibody (mAb or purified antibodies which
23


recognize only one epitope) called OKB7. Not only do these proteins produce the
adhesion response in immortalized cell lines, but also in primary cell
populations(15,35).
Our lab (37) has done extensive work to characterize CR2 and the
mechanisms by which it influences cell-cell adhesion. The Balm-1 human B
lymphoma cell line has been used in these endeavors because these cells express the
members of the CR2 signalling complex and because their background rate of
spontaneous adhesion is low. Some cell lines exist naturally in a highly clumped
state which makes it difficult to determine when a positive adhesion reaction has
occurred. In the Balm-1 cell line any ligand-induced adhesion reaction is easy to
identify. This cell line undergoes homotypic adhesion in response to the same
molecules as do other human B lymphoma cell lines and is not unique in its
response. We have shown that the CR2 ligands which cause clumping act
independently of other known adhesion molecules, such as selectins, integrins,
LFA-1, and ICAM. Evidence to support this conclusion includes the following:
1) CR2- induced adhesion can be induced in an integrin-deficient cell line, 2) the
adhesion does not require divalent cations to undergo CR2-induced adhesion as do
other known adhesion molecules, and 3) mAbs to LFA-1, VLA-4, and ICAM-1
do not inhibit CR2-induced adhesion. These data support other reports of LFA-1
24


and divalent cation independence of homotypic adhesion through the CR2 signalling
complex (31,33,36).
1.4.8 Development of an Antibody which Inhibits Homotypic Adhesion
The above observations led to the search for a mAb which would block
homotypic adhesion that was induced through the binding of C3dg or mAb OKB7
to CR2. The strategy adopted was to make mAbs which would inhibit the clumping
of the cells in culture when OKB7 was present.
The cell line which was used in the homotypic adhesion-producing assay,
the Balm-1 cell, were injected into mice (as antigen). When these mice produced
an immunological response, their spleen cells were fused with mouse myeloma
cells. The resultant hybrid cells or hybridomas were immortalized antibody-
producing cells. Grown in culture, the new cells produced antibodies which could
be screened for their ability to block the clumping of Balm-1 cells when C3dg was
present.
From over 5000 hybridomas produced, three mAb were identified which
blocked the adhesion reaction. Two of the mAbs were found to recognize CR2.
This finding internally validates the method by which the antibodies were
developed. The third mAb (3A8) recognized an unknown protein. Figure 1.10
demonstrates the ability of mAb 3A8 to block the homotypic adhesion reaction
25


which was induced by mAb 0KB7. The mechanism for the mAb 3A8 inhibition of
the adhesion reaction could not be explained by toxicity to cells nor by inhibition of
C3dg/mAb OKB7 binding because it did neither of these things.
26


anti-CR2
(mAb 0KB7)
anti-CR2
(mAb 0KB7)
+
mAb 3A8
Figure 1.10. Inbibition of Homotypic Adhesion by mAb 3A8. Balm-1 cells were induced to
undergo CR2-induced homotypic adhesion with an anti-CR2 antibody (mAb OKB7) in the
presence of mAb 3A8 or a negative control (anti-CD46). Cells were observed after 4 hours at
37C. Cells in the right photo illustrate mAb3A8-induced inhibition of the mAb OKB7-
induced homotypic adhesion reaction.
27


1.4.9 Gamma-Glutamyl Transpeptidase is the Target of the Third Antibody
The third antibody, mAb 3A8, immunoprecipitates two proteins which are
unique to it. Immunoprecipitation is a technique in which cell membranes are
disrupted using detergents. The resultant pieces of the cell are incubated with
monoclonal antibodies. These mAbs bind to their target proteins and can be
separated away from the unbound solubilized protein. Then the bound target
proteins can be identified using electrophoretic techniques. The mAb 3A8
immunoprecipitated two proteins represented by bands at the 68 kDa and 27 kDa
molecular weight range. These size bands are seen in immunoprecipitation and
electrophoretic analysis of each type of cell to which the mAb 3A8 binds. Further
separation of the two bands by two-dimensional gel purification allowed for N-
terminal sequencing of the antibody target protein (see Figure 1.11). Amino acid
sequence homology was identical to the first 14 residues of an ectoenzyme which is
involved in glutathione cycling (see Figure 1.14), y-glutamyl transpeptidase (GGT).
The following studies were conducted to confirm that the sequenced protein bands
were in fact GGT.
28


Acidic Basic Acidic Basic
ISOTYPE CONTROL (UPC-10) 3A8
Figurel.il. The mAb 3A8 recognizes a two protein complex. Solubilized membrane
proteins were immunoprecipitated with either UPC-10 (isotype control) or m3A8 and
analyzed by two dimensional gel electrophoresis. The two proteins are identified by the
boxes in the mAb 3A8 gel.
GGT enzyme activity can be measured by a change in substrate from colorless to
yellow in the presence of the enzyme GGT. In addition to the amino acid sequence
identification, the target of mAb 3A8 was confirmed to be GGT by
immunoprecipitating Balm-1 cells with mAb 3A8 and measuring loss of GGT
enzyme activity in the supernatants. The mAb 3A8 removed most of the
enzymatic activity from the supernatants compared to other mAbs. Figure 1.12
shows the units of GGT activity left in the supernatants after monoclonal antibody-
29


Remaining
bound protein was removed. These enzymatic data support the sequencing results
which identify the antibody target as GGT. We concluded that we had identified
the GGT enzyme (37).
mAb used
Figure 1.12. GGT activity left in solubilized protein after immunoprecipitadon with
andbody. Balm-1 cells were solubilized and immunoprecipitated with mAbs. The antibodies
were removed and the remaining GGT activity was measured.
30


1.4.10 Gamma-Glutamyl Transpeptidase: Structure and Function
GGT is an integral membrane protein which is composed of two subunits
that result from one precursor protein. Meister proposed the structural model
shown in Figure 1.13 (38). The precursor protein is cleaved proteolytically and
the subunits are assembled non-covalently. The active site is found on the smaller
of the subunits but enzymatic function is dependent upon the association of the two
subunits.
Figure 1.13. A model for the structure of gamma-ghdamyl transpeptidase
(GGT). The active site is on the small subunit, butbofli subunits are needed,
for die enzyme to function. The green areas are potential glycosylatimi sites.
GGT has been studied primarily for its role in transport of nutrient cysteine
into the cell, but is also the ectoenzyme required in the pathway which recycles the
tripeptide glutathione (GSH) into the cell. GSH is an important molecule in
Active site
N
31


various cellular mechanisms including free radical scavenging, nutrient cysteine
source, leukotriene protection, and detoxification reactions (38,39,40). Very little
glutathione can enter the cell whole and thus, must be broken down into
components to be transported through the cell membrane (39,41,42). Figure 1.14
shows the glutathione pathway which is called the gamma-glutamyl cycle. This
cycle is the major pathway for cells to recycle glutathione. GGT (a) cleaves GSH
into cysteinyl-glycine and a glutamyl group which is subsequently transferred to
water or to another peptide. A dipeptidase (b) cleaves cysteinyl-glycine either at
the surface of the cell or within the cytoplasm. It is not fully known if the
dipeptidase is found at the cells surface, intracellularly, or both (43,44).
Individual amino acids or the dipeptides then enter the cell by unknown transport
mechanisms and are resynthesized into GSH (c-f). The rate limiting step in the
intracellular synthesis of GSH is the synthesis of y-glutamyl-cysteine by y-
glutamyl-cysteine synthetase which is inhibited by excess glutathione (45).
32


gamma- glutamyl
cysteine synthase
Figure 1.14. The gamma-gliUamyl cycle. Onlaflticme is cleaved outside die cell aid
resynthesized once its component are inftacdbilar.
1.5 Hypothesis
Since binding of GGT with an anti-GGT antibody (mAb 3A8) can inhibit
CR2-induced homotypic adhesion in the Balm-1 B lymphoma cell line, and because
CR2 is associated with the CR2 signal transduction complex in the alternative
complement pathway, it can be hypothesized that GGT has a role in complement-
induced B cell activation.
The association of complement-induced activation and GGT was first
observed in immortalized cell lines. We wanted to explore GGT levels on
lymphocytes taken directly from normal human blood. The first hypothesis states
33


that GGT will be expressed on peripheral blood mononuclear cells, or PBMC
(which include lymphocytes).
The second hypothesis is that GGT is associated with members of the CR2
signalling complex. Although the action of the anti-GGT mAb is related to CR2
directly, CR2 does not act alone in the activation of B cells. We wanted to
determine if GGT is associated with any of the members of the CR2 signal
transduction complex.
34


2. Methods
2.1 Expression of GGT on Peripheral Blood Lymphocytes
Preparation of Lymphocytes: Blood was donated from normal human
donors (2 male). The blood was heparinized and diluted 2X with phosphate
buffered saline (PBS). It was then carefully layered over ficoll-hypaque (Sigma,
St.. Louis, MO) to separate the mononuclear cells from the other cell types. The
cells were centrifuged for 25 minutes at 1500 RPM without braking. The
mononuclear layer (immediately above the ficoll-hypaque) was removed and
washed 2 times with PBS at 2000 RPM for 5 minutes (4C). No further
separations were done to purify the lymphocytes. The cells were counted and
resuspended to 5 x 106 cells /ml in FACS buffer (PBS/1% Bovine Serum Albumin
(BSA)/.01% Sodium Azide) (Sigma). Samples of 100jj.1 each were placed in 75 x
100 polypropylene tubes so each was estimated to contain 5 x 10s cells.
Immunofluorescence Staining: The cells were blocked (to reduce non-
specific staining) with 50 |il of 10% heat-deactivated human serum in PBS for 10
minutes on ice. Either 3A8 or the isotype control UPC-10 (Sigma) was added as a
primary antibody at a concentration of l|ig/sample. These antibodies were
incubated with the cells for 20 minutes on ice. The cells were washed 2 times
35


as above in FACS buffer. Biotinylated sheep F(ab)2 anti-mouse IgG (Cappel,
West Chester, PA) was added as the secondary antibody, incubated, and washed as
before. The tertiary staining step included the addition of Avidin-Per-cp (Caltag,
San Francisco, CA) to fluorescently label the biotin-coated secondary antibody +
primary antibody complex, and FITC- (fluoroscein isothiocyanate) or PE- (phyco-
erythrin) conjugated antibodies to the following molecules: CD4 ( Coulter,
Miami, FL, FITC-conj.), CD8 (Coulter, FITC-conj.), CD45RA (Coulter, PE-
conj.), CD45RO (Coulter, PE-conj.), CD19 (Coulter, PE-conj.), GGT
(unconjugated), and UPC-10 (isotype control). These additions were incubated and
washed as above. After the final wash the cells were fixed in .4 ml of 1 %
paraformaldehyde (Sigma).
Fluorescence-Activated-Cell-Sorting (FACS): Analysis was conducted at
the UCHSC Core Facility on a Coulter Profiler Flow Cytometer. The analysis was
done on a population of cells with the appropriate size and density for the
lymphocyte population. This population was assessed for positive staining of either
GGT or UPC-10 with Per-cp. The positive population also was assessed for
positive FITC and Phyco-erythrin staining. Positivity was determined by setting
the +/- quadrants on the mean channel fluorescence of the lymphocyte population
of the negative control (UPC-10). Positive FITC and/or PE staining of the GGT
population was determined in the same manner. Comparison of the various
36


intensities of the markers was conducted qualitatively and by percentage of +/+
cells. These +/+ percentages between the CD45RA population and the CD45RO
population for each donor were compared using the Students paired T test.
2.2 Association of GGT with the CR2 Signal Transduction Complex
Co-Capping / Immunofluorescense Staining : Co-capping experiments are
used to identify co-localized membrane proteins to determine their presence in
association complexes. Balm-1 cells which had >90% viability as measured by
Trypan Blue (Sigma) staining were washed 2 times in 37C PBS + 2% BSA
(capping buffer) to remove the culture media. They were resuspended at 1 x 107
cells / ml in capping buffer. Samples of 1 x 10 6 cells were placed in non-adjacent
wells of a 96 well V-bottom plate. A primary antibody was added at 1 |ig per
sample. The primary antibody was either the mAb 3A8, mAb CD46 (CD46 is an
RCA family member which is not a member of the CR2 signalling complex)
(Immunotech, Marseille, France), or the isotype control (UPC-10) which is the
same antibody subtype as mAb 3A8 but has no binding site on the cell. These
antibodies were incubated with the cells for 30 minutes at 37C in the dark so that
capping of associated membrane proteins could occur. The unbound antibodies
were washed out 1 time with 37C capping buffer. The bound antibodies were
37


simultaneously cross-linked and fluorescently labelled with 1 |j,g per sample of
sheep anti-mouse IgG antibody which was covalently linked to the TEXAS RED
fluorescent molecule (Molecular Probes, Eugene, OR). This antibody was
incubated with the samples for 60 minutes at 37C in the dark. Following this
incubation the cells were washed 3 times with ice-cold PBS / 2% BSA / 0.02%
Sodium azide to wash out the unbound antibody and to halt any further movement
of membrane proteins. Next, mAbs to the members of the CR2 signal transduction
complex were added to detect which members co-capped with GGT. These
included anti-CD 19 (mAb HD37, DAKO,Glostrup, Denmark), anti-CD82 (clone
50F11, Pharmingen, San Diego, CA), anti-CD53 (cloneH129,Pharmingen), anti-
CD81 (mAb 5A6, donated by Shoshona Levy, Stanford, CA), or anti-CR2 (mAb
HB5). The anti-CD19, anti-CD82, and anti-CD53 antibodies were directly
conjugated to FITC, but the anti-CD81 and anti-CR2 antibodies were biotinylated.
The cells were incubated with lpg of one of the listed mAbs on ice for 20 minutes
in the dark. The unbound antibody was washed out 3 times as before. The samples
which had fluorescently coupled mAbs added were fixed with 4%
paraformaldehyde and stored in the dark at 4C. The samples which had
biotinylated mAbs added were incubated a final time in the dark, on ice for 20
minutes with streptavidin-FITC (Southern Biotechnology Associates, Inc.,
38


Birmingham, AB). Upon completion of the last incubation they were washed,
fixed, and stored like the other samples.
Mounting Cells for Confocal Microscopy: An anti-fade media was used to
mount the cells on glass slides. This media was made fresh for each experiment out
of 1 ml of Fluoromount (Southern Biotechnology Associates) and lmg of p-
phenylene diamine (Sigma). The media was mixed on a rotator at room
temperature for three hours. One drop of the media (3-4 pi) was placed on the
glass microscope slide and mixed with 5-7 pi of the cell suspension. A coverslip
was placed over the mixture and allowed to dry in the dark.
Observation by Confocal Microscopy: A BioRad Confocal Microscope was
used for this analysis. Background settings were established using the samples
which were stained with the isotype control (UPC-10). The level of signal was
adjusted manually for consistency throughout all samples. Digital photographs
were produced on a Mitsubishi Video Printer. Molecules stained with FITC were
observed as green and molecules stained with TEXAS RED were observed as
red. Each fluorochrome was observed using the appropriate wavelength filter and
then the files were merged to observe areas of colocalization which appeared
yellow. No statistical analyses were done using this method.
39


3. Results
3.1 Expression of GGT on Human Lymphocytes
The human peripheral blood was purified to include all the mononuclear
cells. Our interest was in the lymphocyte population of these cells. Using FACS
analysis, PBMC populations are distinguished based on size and strength of
fluorescent cell surface markers. The strength of the fluorescence can be used to
infer the relative amounts of each surface marker. To limit the group of cells
which were examined for fluorescence, cells falling outside the lymphocyte size
range were excluded on the size vs. density scatter plot. Density in this case refers
to the amount of granularity of the cell. Figure 3.1 indicates the sizes of different
types of cells found in peripheral blood and where they typically fall on this type of
scatter plot. Polymorphonuclear cells are highly granulated cells which include
eosinophils, basophils, and neutrophils. Monocytes are cells which differentiate
into phagocytic cells called macrophages. The computer generated gating
enables the analysis to focus only on the cell types of interest. In populations which
are not highly purified there will be contamination of other cell types in the size
range of interest. Included in the lymphocyte size range are other cell types such
40


polymorphonuclear cells
INCREASING
DENSITY
------------
INCREASING SIZE
monocytes
lynqjhocytes
led blood cells and debris
figure 3.1 Size exclusion. FACS scalier plot Pram, peripheral blood, cell will
be segregated doe to their size and density. The cells of interest for fids research
are the lymphocytes. Note that there is overlap between regions.
as natural killer cells and monocytes. FACS analysis can exclude these cell types
on the basis of fluorescence.
Figure 3.2 is a scatter plot of the cells of appropriate size which also are
fluorescing in two colors. This figure is a representative sample of the
experimental data. The first fluorescent molecule is tagging CD19, the B cell
surface marker. The second fluorescent molecule is tagging GGT. The lines
breaking the scatter plot into quadrants are placed according to the background
staining found on the isotype control (UPC-10) sample. All of the cells which are
recorded to the left and below the quadrant lines are negative for the marker on that
axis.
41


AMOUNT OF B CELL
MARKER (CD 19)
B cells
without
GGT
B cells
with
GGT
non-B cells
without
GGT
non-B cells
with GGT
AMOUNT OF GGT
Figure 3.2. GGT expression on B cells. Of the lymphocyte sized population, these cells were
analyzed for CD19 (FITC fluorescence) and GGT (Per-cp fluorescence). As indicated,
quadrant number 2 contains cells which are B cells and have GGT on their surfaces.
42


The quadrants can be labeled as follows: 1) upper left quadrant=B cells which
have no GGT on their surface (+/-); 2) upper right quadrant=B cells which have
high amounts of GGT on their surface (+/+); 3) lower left quadrant = non-B
cells which have no GGT on their surface (-/-); 4) lower right quadrant = non-B
cells which have high amounts of GGT on their surface (-/+). These data
demonstrate that GGT is expressed on a population of peripheral blood B cells.
This population represents approximately 2% of the total gated lymphocytes, but
approximately 25 % of the total B cell population as defined by the B cell marker
CD19.
In addition to B cells, T cells were analyzed. Three color staining of T cells
allowed identification of four subsets of T cells. T helper (CD4+) and T cytotoxic
(CD8+) cells were identified with FITC stain. These populations were further
defined by PE stain to determine activated (CD45RO+) or unactivated (CD45RA+)
phenotype. All cells were stained with Per-cp stain to label GGT. So, with three
types of fluorescent molecules, the amount of GGT can be determined on four
subtypes of T cells: unactivated T helper cells, activated/memory T helper cells,
unactivated/naive T cytotoxic cells, activated/memory T cytotoxic cells. Figure 3.3
represents these staining patterns. Like the B cell scatter plots, the upper right
quadrant contains the cells that are positive for the second color marker (activated
or non-activated) and the cells that have GGT on their surfaces (+/+ quadrant).
43


Table 3.1 summarizes the T cell data. The data is expressed as percent of
analyzed cells which are in quadrant 2 (+/+). The data from these donors
suggests that expression of GGT on activated /memory T cells was significantly
greater than on unactivated/ naive T cells. In order to demonstrate this the
unactivated T cells were compared to the activated T cells for each donor using a
paired Students T test. For both groups of T cells, T cytotoxic cells and T helper
cells, the difference between the unactivated cells and the activated cells was
significant. The resultant p values for these analyses were p< 0.01 for T
cytotoxic cells and p< 0.05 for T helper cells.
44


T CYTOTOXIC CELLS
(CD8+)
T HELPER CELLS
(CD4+)
Q
W
H
£
i
H
O
<
<
Pi
n
Q
U
1 2
+/- +/+



Si ite 4
-im ife'- ./+
AMOUNT OF GGT
1 2
Q +/- +/+
S
H o
<2 Pi
& 3
H Q
u B 'Wx\ 'rr?' 4
< ST*** -/+
Q
W
H
£
hH
H
U
o
Pi
Q
U
l 2
+/- +/+
t i'.tii'' i
4
-/+
AMOUNT OF GGT
AMOUNT OF GGT
Figure 3.3. GGTs presence on subsets of T cells. Four subsets of T cells are represented in
these scatter plots: unactivated T cytotoxic cells, activated T cytotoxic cells, unactivated T
helper cells, and activated T helper cells. Quadrant 2 represents the cells of those subsets
which have GGT present.
45


(A)
GGT+ T CYTOTOXIC CELLS (CD8+)
EXPERIMENT UNACTIVATED ACTIVATED
NUMBER (CD45RA+) (CD45RO+)
1 6.9 14.3
2 0.5 3.8
3 2.0 8.3
4 5.5 8.4
5 3.2 11.0
(B)
GGT+ T HELPER CELLS (CD4+)
EXPERIMENT UNACTIVATED ACTIVATED
NUMBER (CD45RA+) (CD45RO+)
1 4.0 19.9
2 0.3 8.4
3 1.5 16.4
4 1.3 10.8
5 2.3 19.6
Table 3.1. Percent of T cell subsets which are GGT positive (quadrant 2).
(A) shows the percent of all T cytotoxic cells which are either unactivated or activated and
have GGT on their surfaces, (B) shows the percent of all T helper cells which are either
unactivated or activated and have GGT on their surfaces. The cell markers used for
identification are given in parentheses.
46


3.2 Association of GGT with CR2 Signalling Complex Members Individual
cells which were representative of each sample were photographed. Each photo
represents the merge of two single color confocal images. The red areas indicate
those portions of the cell which have either the negative control (CD46) protein or
GGT stained. The molecule CD46 is related to CR2 (RCA family), but it does not
associate with the CR2 signalling complex. The green areas indicate those portions
of the cell which have the CR2 signalling complex member stained (CR2, CD19,
CD81, CD82, or CD53). Leu-13 was not studied due to limited supply of the anti-
Leu-13 monoclonal antibody. Areas where the two molecules are colocalized are
indicated by yellow. Yellow co-capped areas of the cells were clearly visible in the
samples which were stained with GGT and the CR2 signalling complex members
(CR2, CD19, CD81, CD82, or CD53). The control samples (CD46 and CR2
signalling complex members) for each of those combinations had little or no yellow
co-capped areas on the cells. These data suggest that GGT is associated with CR2,
CD19, CD81, CD82, and CD53 in varying degrees on the surface of Balm-1 cells.
47


B
Figure 3.4 (a) Merged confocal images of a Balm-1 cell stained with anti-CD46 + Texas
Red ( in red) and anti-CR2 (mAh HB5) + FITC ( in green) are the negative control, (b)
Another Balm-1 cell prepared the same way as (a) but stained with anti-GGT (mAb 3A8) +
Texas Red (in red) and anti-CR2 + FITC (in green). The colocalized areas are shown in
yellow.
48


Figure 3.5. (a) Merged confocal images of a Balm-1 cell stained with anti-CD46 + Texas
Red ( in red) and anti-CD19 (mAb HD37) + FITC ( in green) are the negative control, (b)
Another Balm-1 cell prepared the same way as (a) but stained with anti-GGT (mAb 3A8) +
Texas Red (in red) and anti-CD 19 + FITC (in green). The colocalized areas are shown in
yellow.
49


Figure 3.6. (a) Merged confocal images of a Balm-1 cell stained with anti-CD46 + Texas
Red ( in red) and anti-CD81 (mAh 5A6) + FITC ( in green) are the negative control, (b)
Another Balm-1 cell prepared the same way as (a) but stained with anti-GGT (mAb 3A8) +
Texas Red (in red) and anti-CD81 + FITC (in green). The colocalized areas are shown in
yellow.
50


Figure 3.7. (a) Merged confocal images of a Balm-1 cell stained with anti-CD46 + Texas
Red ( in red) and anti-CD82 + FITC ( in green) are the negative control, (b) Another Balm-
1 cell prepared the same way as (a) but stained with anti-GGT (mAb 3A8) + Texas Red (in
red) and anti-CD82 + FITC (in green). The colocalized areas are shown in yellow.
31


B
*
11
Figure 3.8. (a) Merged confocal images of a Balm-1 cell stained with anti-CD46 + Texas
Red ( in red) and anti-CD53 + FITC ( in green) are the negative control, (b) Another Balm-1
cell prepared the same way as (a) but stained with anti-GGT (mAb 3A8) + Texas Red (in
red) and anti-CD53 + FITC (in green). The colocalized areas are shown in yellow.
52


4.0 Discussion
In these studies we have further clarified the relationship between the GGT
enzyme and the CR2 signalling complex. We have found that GGT is expressed on
peripheral blood lymphocytes in addition to immortalized cell lines and that it does
colocalize on the surface of the cell with the protein members of the CR2 signalling
complex.
The initial observations of GGT expression on lymphocytes were made
using immortalized cell lines. Therefore, experiments were conducted to
demonstrate that GGT is expressed on normal human blood lymphocytes and
which subsets of lymphocytes. B cells were analyzed because the model used to
produce the anti-GGT antibody in this lab was a B cell activation model. As Fig
3.2 shows, there is a population of B cells (25% of marked B cells) which do have
a significant amount of GGT on their surfaces. The CR2 signalling complex is
thought to modulate very small signals to insure B cell activation (6).
The observation that the population of B cells which express high levels of
GGT is rather small is not contradictory to what was expected. The cells used to
produce mAb 3A8 were a mature B cell line which was activated through CR2. It
is possible that the population of B cells which expresses high levels of GGT is a
population which has been activated and is proliferating and secreting antibody.
53


Other studies have shown that B cells typically have low levels of GGT enzyme
activity and lower GGT amounts expressed on their surfaces than do other types of
peripheral blood cells like natural killer cells and monocytes(46). Monocytes are
thought to help lymphocytes regulate their glutathione levels by secreting sulfur-
containing thiols (47). The cells which have been shown to have higher activities
and surface expression of GGT enzyme are those which also are proliferating and
secreting material (46).
In the future, an attempt to dissect out the effect of activation on GGT
expression in peripheral B cells is needed. Surface markers which adequately
distinguish activated B cells from naive B cells have not been identified. To look at
an activated population of B cells, the cells can be cultured with a known mitogen
(a molecule which stimulates cell growth) such as LPS (lipopolysaccaride) or SAC
(Staph aureas Cowan Type A) for several days. After the addition of an
appropriate mitogen, resting B cells become activated and proliferate. Following
stimulation, they are stained for the amount of GGT on their surface. The
differences of GGT staining between day 0 and day 3 can be used as evidence for
changes due to activation. These differences would be exaggerated due to
activation of the entire B cell population. Comparison of mitogen-stimulated B
cells with untreated B cells in culture would account for tissue culture effects as
well as baseline GGT levels in circulating PBMC.
54


Because we wanted to examine GGT expression on lymphocytes in general,
we also looked at T cells. T cells have well-documented surface markers which
distinguish between unactivated/ naive types and activated/memory types. Using
three color flow cytometric analysis, we looked at the unactivated and activated
phenotypes of both T cytotoxic cells and T helper cells (Figure 3.3 and Table 3.1).
We did not see any distinct differences between the amount of GGT on the surface
of T helper cells and T cytotoxic cells. This agrees with data presented by Tager
(48). However, we did see a significant increase in the amount of GGT present on
activated cell types over the non-activated cell types in both T helper cells and T
cytotoxic cells. Interestingly, new studies have shown that the unactivated/naive
cells of advanced AIDS patients die disproportionately and that these cells have a
lowered amount of glutathione intracellularly (49). It is possible that a lower
expression of GGT may result in death due to the inability of these cells to recycle
glutathione. In addition, Staal et al. has shown that a portion of the HIV viral
gene can be regulated by controlling intracellular glutathione levels. Lowered
glutathione levels in the cell activate the transcription factor NF-kB which allows
the virus to be replicated (50). So a cell which has a higher level of intracellular
glutathione may be able to withstand the virus longer. Neither the GGT enzyme
activity nor expression on the surface of these diseased cells has been studied.
Also, tumor cells which have increased glutathione levels may be able to withstand
55


oxidative death due to anti-cancer therapies better than their low glutathione level
counterparts (51). Tager et al. suggests that since some drug resistant-lymphomas
and myelomas have increased expression of the GGT enzyme on their surfaces,
these tumor cells might have a higher survival rate when treated with cysteine-
depleting chemotherapy drugs (48). B and T cells which have higher amounts of
GGT present on their surfaces or an increased level of intracellular glutathione may
be able to withstand higher oxidative stresses as well.
Although the anti-GGT antibody (3A8) was produced using an activated B
cell model, it was produced through CR2 activation specifically. In addition to
being associated with the CR2 signalling complex of B cells, CR2 is found
associated with other B cell proteins, by itself in the membrane of B cells, and on
other types of cells. For example, it is found on subsets of T helper cells and T
cytotoxic cells. Although the lymphoid cells, B and T cells, serve different
functions, they may have redundant surface proteins that act in a similar fashion on
the two cell types. These cell types also may undergo some of the same metabolic
changes while becoming activated, so the presence of similar mechanisms is likely.
The development of the anti-GGT antibody using a B cell activation model and the
finding that activated T cell subsets have increased levels of GGT support the
hypothesis that GGT plays a role in immune cell activation.
56


The Confocal Microscopy experiments conducted for this paper were used
as a third method of establishing GGTs association with the members of the CR2
signalling complex. The two methods used previously by our lab (37) establish
molecular and functional associations between GGT and the complex by co-
immunoprecipitation studies. Results of these experiments were similar in that the
TMSF4 members showed the highest amount of GGT activity precipitating with
them. These methods did not reveal an association between GGT and CD 19 nor
CR2. The problem with immunoprecipitation assays is that different detergents
maintain different types of structural associations. So an association which is real
may not be maintained by the available detergents. In our immunoprecipitation
experiments with CR2 and CD19, no association with GGT has been observed.
This lack of association is not consistent with the manner of development of the
anti-GGT antibody. The Confocal Microscopy experiments were done in the
absence of detergents, so that any association between CR2 and GGT could be
observed. The results of the Confocal experiments show an association between the
members of the CR2 signalling complex and GGT. In support of the co-
immunoprecipitation studies, GGT is associated most strongly with the members of
the TMSF4 family (CD81, CD82, and CD53). Additionally, both CR2 and CD19
can be seen to colocalize with the enzyme GGT using this method. The strong
association of GGT with the TMSF4 members in all three techniques may result
57


from two phenomena: (1) the antibodies to the TMSF4 members may bind strongly
to their epitopes on their respective proteins in highly accessible areas; or (2) the
expression of the TMSF4 members is so great that any association with GGT is
easily observed. Conversely, the antibodies to CR2 and CD19 may not bind as
strongly if they lack easily accessible epitopes, or their expression may not be as
ubiquitous as that of the TMSF4 members, making them harder to observe by these
techniques. The anti-GGT antibody (mAb 3A8) may also lack strong binding, an
accessible epitope, or GGT is not expressed in great numbers on these cells.
Because mAb 3A8 inhibits the clumping seen in the homotypic adhesion reaction
and can be seen using flow cytometry, the third reason is most likely. The flow
cytometry data presented here suggests that GGT is highly expressed in only a
small population of cells so it is possible that some physical associations may be
below the detection limits of the immunopreciptation assays used.
In accordance with the evidence for each of these proteins existence with
and without the other members of the complex, the confocal data indicate areas of
co-localization and areas of independently stained proteins. However, the relative
amount of co-localization between members of the signalling complex is greater
than the amount of co-localization of those proteins and the control protein CD46.
These Confocal images support the hypothesis that GGT is associated with the CR2
signalling complex. Associations between GGT and CR2, CD19, CD81, CD82,
58


and CD53 have not been shown previously. GGT has been related to growth
induction in a myeloid cell line. When hematopoietic growth factors are incubated
with the KG-1 myeloid cell line, significant increases in GGT enzyme activity are
observed. These increases support a role for GGT in growth and proliferation of
myeloid cells(52). Since the CR2 signalling complex is involved with growth and
proliferation in B cells, the association of GGT with its members supports the
hypothesis that the GGT enzyme plays a role in lymphocyte growth events (5,6,9).
However, it is not known at this time whether GGTs involvement is a cause or an
effect of the cellular events which promote growth in immune cells.
The function of GGT in the CR2 signalling complex has not yet been
established. It may be that GGT becomes associated with the complex in response
to increased amino acid need during growth. It may be that an altered level of
glutathione is needed in the cell in response to oxidizing activation events. But,
GGT may in fact act within the CR2 signalling complex to further modulate the
activation by low-level antigens within the body. If this is the case, it would be a
novel function for GGT.
In the future, the involvement of GGT with the CR2 signalling complex
will be dissected out by using cellular techniques such as the ones described in this
thesis. In addition, a strain of mouse has been bred to lack the gene for GGT (53).
Studies using these mice will indicate if there is any divergence from the normal
59


development of B and T cells. Such a divergence may point toward specific times
during the development of an immune response where GGT is most important. The
time at which GGT is needed may indicate its involvement with the CR2 signalling
complex.
Whether the function of GGT is to cause activation events or is an effect of
activation events further demonstrates the ability of our bodies to evolve a limited
number of proteins for a variety of functions. These findings support the idea that
not only do the cells of the immune system overlap in function, but individual
proteins of the body may perform functions which act upon a variety of systems.
These many overlapping functions can make mistakes. Portions of the
immune system can become stimulated by damaged cell proteins which in some
cases can lead to serious overcompensation by the immune system against healthy
cells in the body. This is what happens in autoimmune disorders. By understanding
the mechanisms by which our cells and systems work together, we can learn how to
restore and maintain harmony among cells which have lost the ability to work
together.
60


BIBLIOGRAPHY
(1) Modrich, P. Mismach Repair, Genetic Stability, and Cancer. Science 1994;
266:1959-1960.
(2) Crowley, LV. Introduction to Human disease, third edition. 1992. Jones and
Bartlett Publishers, Inc. Boston, MA. 820pps.
(3) Paul WE. Fundamental Immunology, third Edition. 1993, Raven Press,NY.
1490pps.
(4) Kuby J. Immunology, second edition. 1994. W H Freeman and Co, NY. 660
pps.
(5) Bradbury LE, Kansas GS, Levy S, Evans RL, Tedder TF. The CD19/CD21
signal transducing complex of human B lymphocytes includes the target of
antiproliferative antibody-1 and Leu-13 molecules. J Immunol 1992; 149: 2841-
2850.
(6) Matsumoto AK, Martin DR, Carter RH, Klickstein LB, Ahearn JM, Fearon
DT. Functional dissection of the CD21/CD19/TAPA-1/Leu-13 complex of B
lymphocytes. J Exp Med 1993; 178:1407-1417.
(7) Olweus J, Lund-Johansen F, Horejsi V. CD53, a protein with four
transmembrane spanning domains, mediates signal transduction in human
monocytes and B cells. 1993 J Immunol; 151: 707-714.
(8) Rasmussen AM, Blomhoff HK, Stokke T, Horejsi V, Smeland EB. Cross-
linking of CD53 promotes activation of resting human B cells. 1994 J Immunol;
153: 4997-5007.
(9) Tedder TF, Zhou L-J, Engel P. The CD19/CD21 signal transduction complex
of B lymphocytes. Immunol Today 1994; 15: 437-442.
(10) Fischer E, Delibrias C, Kazatchkine MD. Expression of CR2 (the C3dg/EBV
receptor, CD21) on normal human peripheral blood T lymphocytes. J Immunol;
146 : 865-869.
61


(11) Voet D, Voet J. Biochemistry. 1990. John Wiley and Sons, NY. 1223pps.
(12) Molina H, Perkins SJ, Guthridge J, Gorka J, Kinoshita T, Holers VM.
Characterization of a complement receptor 2 (CR2, CD21) ligand binding site for
C3: an initial model of ligand interaction with two linked SCR modules.
J Immunol 1995; 154: 5426-5435.
(13) Barlow PN, Baron N, Norman DG, Day AJ, Willis AC, Sim RB, Campbell
ID. Secondary Structure of a complement control protein module by two-
dimensional H NMR. Biochem 1991; 30 : 997-1004.
(14) Perkins SJ, Haris, PI, Sim RB, Chapman D. A study of the structure of
Human Complement Component Factor H By Fourier transform infrared
spectroscopy and secondary structure averaging methods. Biochem 1988; 27:
4004-4012
(15) Carell JC, Myones BL, Frazier B, Holers VM. Structural requirements for
C3d,g/Epstein-Barr virus receptor (CR2/CD21) ligand binding, internalization,
and viral infection. J Biol Chem 1990; 265:12293-12299.
(16) Bohnsack JF, Cooper NR. CR2 ligands modulate human B cell activation.
J Immunol 1988; 141:2569-2576.
(17) Luxembourg AT, Cooper NR. Modulation of signalling via the B cell antigen
receptor by CD21, the receptor for C3dg and EBV. J Immunol 1994; 4448-4457.
(18) Croix AD, Ahearn MJ, Rosengard MA, Han S, Kelsoe G, Minghe M,
Carroll MC. Antibody Response to a T-dependent Antigen Requires B Cell
Expression of Complement Receptors. J Exp Med 1996; 183:1857-1854.
(19) Molina H, Holers VM, Li B, Fang Y-F, Mariathasan D, Goellner J, Strauss-
Schoenberger J, Karr RW, Chaplin D. Markedly impaired humoral immune
response in mice deficient in complement receptors 1 and 2. Proc Natl Acad Sci
1996; 93:3357-3361.
(20) Carter RH, Fearon DT. CD 19: lowering the threshold for antigen receptor
stimulation of B lymphocytes. Science 1992; 256: 105-107.
62


(21) Pesando JM, Bouchard LS, McMaster BE. CD19 is functionally and
physically associated with surface immunoglobulin. J Exp Med 1989; 170: 2159-
2164.
(22) Oren R. Takahashi S, Doss C, Levy R, Levy S. TAPA-1, the target of an
antiproliferative antibody, defines a new family of transmembrane proteins. Mol
Cell Biol 1990; 10:4007-4015.
(23) Mannion BA, Berditchevski F, Kraeft S, Chen LB, Hemler ME.
Transmembrane-4 Superfamily Proteins CD81 (TAPA-1), CD82, CD63, and CD53
Specifically Associate with Integrin a4bl (CD49d/CD29). J Immunol
1996;157:2039-2047.
(24) Szollosi J, Horejsi V, Bene L, Angelisova P, Damjanovich S.
Supramolecular complexes of MHC classl, MHC class II, CD20, and Tetraspan
molecules (CD53, CD81, CD82) at the surface of a B cell line JY. J Immun,
1996; 157: 2939-2946.
(25) Schick MR, Hguyen VQ, Levy S. Anti-TAPA-l antibodies induce protein
tyrosine phosphorylation that is prevented by increasing intracellular thiol levels.
J Immunol 1993; 151: 1918-1925.
(26) Wright MD, Tomlinson MG. The ins and outs of the transmembrane 4
superfamily. Immunol Today; 15:588-594.
(27) Levy S, Nguyen VQ, Andria ML, Takahashi S. Structure and membrane
topology of TAPA-1. J Biol Chem 1991;266:14597-14602.
(28) Wo ZG, Oswald, RE. Unraveling the modular design of glutamated-gated ion
channels. Trends Neuroscience 1995; 18:161-168.
(29) Deblandre GA, Marinx OP, Evans SS, Majjaj S, Leo O, Caput D, Huez GA,
Wathelet MG. Expression cloning of an interferon-inducible 17 k-Da membrane
protein implicated in the control of cell growth. JBC 1995; 270 (40) :23860-23866.
(30) Reid LE, Drasnett AH, Gilbert CS, Porter ACG, Gewert DR, Stark GR,
Kerr IM. A single DNA response element can confer inducibility by both alpha-
and gamma- interferons. PNAS 1989; 86: 840-844.
63


(31) Takahashi S, Doss C, Levy S, Levy R. TAPA-1, the target of an
antiproliferative antibody, is associated on the cell surface with the Leu-13
antigen. J Immunol 1990; 145:2207-2213.
(32) Deblandre G, Marinx OP, Evans S, Majjaj S, Leo, O, Caput D, Huez GA,
Wathelet MG. Expression Cloning of Interferon-inducible 17-kDa membrane
protein implicated in the control of cell growth. J Biol Chem 1995; 270:23860-
23866.
(33) Evans SS, Collea RP, Leasure JA, Lee DB. IFN-a induces homotypic
adhesion and Leu-13 expression in human B lymphoid cells, J Immunol
1993;150:736-747.
(34) Ramirez R, Carracedo J, Mooney N, Charron D. HLA class-II-mediated
homotypic aggregation: involvement of a protein tyrosine kinase and protein kinase
c. Human Immunology 1992; 34:115-125.
(35) Bjork P, Elenstrom-Magnusson C, Rosen A, Severinson E, Paulie S. CD23
and CD21 function as adhesion molecules in homotypic aggregation of human B
lymphocytes. Eur J Immunol 1993; 23: 1771-1775.
(36) Kansas GS, Tedder TF. Transmembrane signals generated through MHC
class II, CD 19, CD20, CD39, and CD40 antigens induce LFA-1 dependent and
independent adhesion in human B cells through a tyrosine kinase-dependent
pathway. J Immunol 1991; 147:4094-4102.
(37) Nichols T, Guthridge J, Fletcher D, Holers, VM. A role for gamma-
glutamyl transpeptidase in CD19/CD21 complex function. Manuscript in progress,
1997.
(38) Tate S, Meister A. g-Glutamyl transpeptidase: catalytic, structural and
functional aspects. Mol Cell Biochem 1981;39:357-368.
(39) Meister A. Transport and metabolism of glutathione and g-glutamyl amino
acids. Group meeting of the peptide and protein grouuup of the Biochemical
Society and the Royal Society of Chemistry, held at the University of Durham on
30 March 1983, and organized and edited by J. W. Payne.
64


(40) Monks TJ, Anders MW, Dekant W, Stevens JL, Lau SS, and Bladeren PJ.
Glutathione conjugate mediated toxicities. Toxicology and Applied Pharmacology
1990; 106: 1-19.
(41) Kannan R, Yi JR, Tang D, Zlokovic BV, Kaplowitz N. Identification of a
novel, sodium dependent, reduced glutathione transporter in the rat lens epithelium.
Invest Opthalmol Vis Sci. 1996; 37: 2269-2275.
(42) Sze G, Kaplowitz N, Ookhtens M, Lu SC. Bidirectional membrane trasport
of intact glutathione in Hep G2 cells. Am. J. Physiol. 1993;265: G1128-G1134).
(43) Habib GM, Barrios R, Shi ZZ, Lieberman MW. Four distinct membrane-
bound dipeptidase RNAs are differentially expressed and show discordant
regulation with g-glutamyl transpeptidase. J Biol. Chem. 1996; 271:16273-16280.
(44) Adachi H, Kubota I, Okamura N, Iwata H, Tsujimoto M, Nakazato H,
Nishihara T, Noguchi T. Purification and characterization of human microsomal
dipeptidase. J. Biochem. 1989; 105 : 957-961.
(45) Deneke S, Fanburg BL. Regulation of cellular glutathione. Am J Physiol
1989; 257:L163-L173.
(46) Grisk O, Kuster U, Ansorge S. The activity of g-glutamyl transpeptidase (g-
GT) in populations of mononuclear cells from human perepheral blood. Biol.
Chem. Hoppe-Seyler 1993; 374: 287-290.
(47) Gmuder H, Eck HP, Benninghoff B, Roth S, Droge. Macrophages regulate
intracellular glutathione levels of lymphocytes. Evidence for an immunoregulatory
role of cysteine. Cell Immunol 1990; 260: 32-46.
(48) Tager M, Ittenson A, Franke A, Frey A, Gassen HG, Ansorge S. g-Glutamyl
transpeptidase-cellular expression in populations of normal human nononuclear cells
and patients suffering form leukemias. Ann. Hematol. 1995;70: 237-242.
(49) Aukrust P, Svardal AM, Muller F, Lunden B, Nordoy I, Froland SS.
Markedly disturbed glutathione redox status in CD45RA(-I-) cells in human
65


immunodeficiency virus type 1 infection is associated with selective depletion of
this lymphocyte subset. Blood 1996: 88(7) : 2626-2623.
(50) Staal FJT, Roederer M, Herzenberg LA, Herzenberg LA. Intracellular thiols
regulate activation of nuclear factor kB and transcription of human
immunodeficiency virus. PNAS 1990; 87: 9943-9947.
(51) HaniganMH, Ricketts WA. Extracellular glutathione is a source of cysteine
for cells that express gamma-glutamyl transpeptidase. Biochemistry 1993 ; 32 :
6302-6306.
(52) Miller AM, Sandler E, Kobb M, Eastgate J, Zucali J. Hematopoeeietic
growth factor induction of gamma-glutamyl transferase in the KG-1 myeloid cell
line. Environmental Hematology 1993; 21 :9-15.
(53) Lieberman MW, Wiseman AL, Shi ZZ, Carter BZ, Barrios R, Ou CN,
Chevez-Barrios P, Wang Y, Habib GM, Goodman JC, Huang SL, Lebovitz RM,
Matzuk MM. Growth retardation and cysteine deficiency in gamma-glutamyl
transpeptidase-deficient mice. PNAS 1996; 91 : 7923-7926.
66


Full Text

PAGE 1

THE GAMMA-GLUT AMYL TRANSPEPTIDASE ENZYME IS ASSOCIATED WITH LYMPHOCYTE ACTIVATION PROTEINS by Dana Rene Fletcher B.S., Pennsylvania State University, 1995 A thesis submitted to the University of Colorado at Denver in partial fulfillment of the requirements for the degree of Master of Arts Biology 1997

PAGE 2

This thesis for the Master of Arts degree by Dana Rene Fletcher has been approved by Teresa Audesirk 5-9-97 Date

PAGE 3

Fletcher, Dana Rene (M.A., Biology) The Garnma-Glutamyl Transpeptidase Enzyme is Associated With Lymphocyte Activation Proteins Thesis directed by Associate Professor Bradley J. Stith ABSTRACT A monoclonal antibody has been developed during the search for an inhibitor of B cell activation through the complement receptor type 2. This antibody recognizes the garnma-glutamyl transpeptidase enzyme that is involved in the cycling of the antioxidant tripeptide glutathione. This project has employed immunofluorescent staining techniques to look at characteristics of the garnma-glutamyl transpeptidase enzyme on peripheral blood lymphocytes and a tumor cell line. By flow cytometry, garnma-glutamyl transpeptidase is present on a population of peripheral blood B lymphocytes. It is also found in higher amounts on the surfaces of activated/memory phenotypes of T cytotoxic cells and T helper cells than on unactivated/naive phenotypes of those T cells. Confocal Microscopy confirms results of other methodologies which show physical associations between garnma glutamyl transpeptidase and members of the complement-induced B cell signal transduction complex. The presence of gamma-glutamyl transpeptidase on activated phenotypes ofT cells and its association with members of a protein complex involved with B cell activation supports the hypothesis that this enzyme is involved with some aspect of lymphocyte activation. This abstract accurately represents the content of the candidate's thesis. I recommend its publication. iii

PAGE 4

DEDICATION I would like to dedicate this thesis to all those who have smiled at the glistening snow; to those who have found themselves in a stormy sea and swam; and to those who have heard the whisper of reason and listened.

PAGE 5

ACKNOWLEDGEMENTS My thanks to the UCHSC Department of Rheumatology for their technical and financial support and to the Arthritis Foundation grant which funded this project. I would like to thank Michael Holers, who has allowed me to use the research I've conducted while working in his lab, for this thesis. Special thanks to my committee members Bradley Stith, Ellen Levy, and Teresa Audesirk for their technical advice and dedication to their students. I would like to thank Timothy Nichols who has been instrumental in my research education as well as a champion for my success. In addition, I would like to thank my family and friends for their continued support during my life and during this Master's program.

PAGE 6

CONTENTS List of Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viii List of Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x Chapter 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.1 Perspective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.2 Relevance of Project.............................................................. 2 1.3 The Immune System: A Brief Review......................................... 3 1.4 Technical Background ......................... ............................ ... ... 10 1.4.1 A Model for a Signal Transducing Complex in B Cell Activation. . . . 10 1.4.2 Complement Receptor Type II................................................ 11 1.4.3 An Amplification Molecule: CD19. ... .............. ..... ........... ........ 15 1.4.4 The Tetraspan Family .............. .... ....... ....... .. .. ... .. . .. .... .. ........ 17 1.4.5 Leu-13 ............................................................................. 19 1.4.6 Interactions Between the CR2 Signalling Complex Molecules . . . . . . 20 1.4.7 Events Following Activation Through the CR2 Signalling Complex.... 21 1.4.8 Development of An Antibody which Inhibits Homotypic Adhesion..... 25 vi

PAGE 7

CONTENTS (continued) 1.4.9 Gamma-Glutamyl Transpeptidase is the Target of the mAb 3A8 ........ 28 1.4.10 Gamma-Glutamyl Transpeptidase: Structure and Function ............. 31 1.5 Hypotheses ..... .......... .... .. ......... ..... ..... ... .. ......... ..... ... ........... 33 2. Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 2.1 Expression of Gamma-Glutamyl Transpeptidase on Peripheral Blood Lymphocytes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 2.2 Association of Gamma-Glutamyl Transpeptidase with the CR2 Signal Transduction Complex . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 3. Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 3.1 Expression of GGT on Human Lymphocytes . . . . . . . . . . . . . . . . . 40 3.2 Association of GGT with CR2 Signalling Complex Members .............. 47 4. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 Bibliography ... -... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 vii

PAGE 8

LIST OF FIGURES 1.1 The B Cell Immune Response . . . . .. . . . . . . . . . . . . . . . . . . . . . 5 1.2 The T Cell Primary Response.................................................... 7 1.3 The Complement Cascade: Classical and Alternative Pathways .... ....... 9 1.4 A Model for Signal Amplification in B Cells .......... ..... .......... ... ...... 11 1.5 A Model of the Proposed CR2 Structure . . .. .. .. .. .. . . . . . .. .. .. .. 13 1.6 A Model of the Proposed Structure of CD19 .................................. 16 1. 7 A Model of the Proposed Structure of the TMSF4 Members . . . . . . . 18 1.8 Associations of the Molecules of the CR2 Signalling Complex ..... ........ 20 1. 9 An in vitro Model for the Activation of B Cells .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 23 1.10 Inhibition of Homotypic Adhesion by mAb 3A8 . . . . . . . . . . . . . . 27 1.11 The mAb 3A8 Recognizes a Two Protein Complex .. .. .. .. .. .. .. .. .. .. .. .. 29 1.12 GGT Activity In Immunoprecipitates .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 30 1.13 A Model for the Structure of GGT .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 31 1.14 The Gamma-Glutamyl Cycle .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 33 3.1 An Example of a Size Exclusion Scatter Plot Using FACS Analysis..... 41 3.2 GGT Expression On B Cells..................................................... 42 3.3 GGT's Presence on Subsets ofT Cells......................................... 45 viii

PAGE 9

LIST OF FIGURES (continued) 3.4 Merged Confocal Images of CR2 and GGT .................................... 48 3.5 Merged Confocal Images of CD19 and GGT .................................. 49 3.6 Merged Confocal Images of CD81 and GGT .................................. 50 3. 7 Merged Confocal Images of CD82 and GGT .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 51 3.8 Merged Confocal Images of CD53 and GGT .................................. 52 ix

PAGE 10

LIST OF TABLES 3.1 Percent ofT Cell Subsets Which Are GGT Positive .. . .. . .. . . . . . . 46 X

PAGE 11

1. Introduction 1.1 Perspective In the journey from the primordial soup, the Earth has evolved complex ecological systems where an organisms' survival depends on its ability to play roles of both aggressive predator and elusive prey. All creatures, from the virus to the human, have developed intricate mechanisms of attack and protection. Many times multiple mechanisms serve as checks and balances within one organism. For example, proteins have been found to have more than one function, and one function can have more than one protein controlling it (1). These complex arrangements of overlapping systems make characterization of our species difficult to complete. Each year more discrete pieces of the puzzle of life are put into place. We can benefit greatly from this information. Not only can we better understand ourselves and our heritage, but we can understand what happens when one or more of these mechanisms goes awry. Even our complex protective mechanisms can make mistakes and allow altered cells to wreak havoc within us. The mechanisms by which our bodies launch attacks against foreign invaders and monitor the health of the billions of cells of which we are composed

PAGE 12

are called the Immune System. This paper deals with a subset of this system which has been studied extensively for its role in diseases where the immune system launches an inappropriate attack on the body it is supposed to protect. 1.2 Relevance of Project Autoimmune diseases, a wide variety of illnesses which include rheumatoid arthritis and systemic lupus erythematosus, present complications that result from inappropriate activation of one's own immune system by normal cells (2). In the near future, treatments for these diseases may include molecules which block inappropriate stimulation of different aspects of the immune system. This project began with the goal of identifying potential targets for such a treatment. Our lab studies both structural and functional aspects of the complement system of the immune system. Through our research of complement, we developed an antibody which inhibits normal immune function. This antibody recognizes a cell surface enzyme that is involved with the cycling of an antioxidant tripeptide. Previously, this enzyme and the process of activating the immune system had not been linked. The goal of this project was to clarify the association between this enzyme and the activation of immune cells. 2

PAGE 13

1.3 The Immune System: A Brief Review The immune system is a complex network which acts to protect the body. This research focuses on activation of the immune response through a mechanism that incorporates the three major components of the immune system: B cells, T cells, and the complement cascade. B cells are a type of white blood cell called lymphocytes that are responsible for the production of proteins called antibodies. Antibodies circulate in the blood and recognize particles that need to be removed. These particles, regardless of their origin, are called antigens. Commonly, they are thought of as invading bacteria and viruses, but often these antigens are parts of our own altered cells and occasionally are parts of our normal cells. In addition to secreting antibodies, B cells have molecules attached to their surfaces which recognize antigens. These molecules are called antigen receptors or B cell receptors. When a B cell comes in contact with an antigen, the antigen will bind to the antigen receptor on the B cell. The B cell then internalizes the antigen and the receptor, enzymatically degrades the antigen, and distributes small parts of the antigen on its surface. Immune cells which distribute foreign particles on their surfaces like this are called AntigenPresenting Cells (APCs). The antigen pieces on APCs are associated with a molecule which is unique to each person and allows the immune system to identify self. After B cells bind to the antigen which its antigen receptor specifically 3

PAGE 14

recognizes, the B cell becomes activated. Sometimes they have help becoming activated by other cells or proteins of the immune system. Those mechanisms will be discussed in the following paragraphs. Once activated, the 8 cells differentiate into one of two types of cells (see Figure 1.1). The first resulting cell type is the antibody-producer called a plasma cell. These cells are short-lived secretion machines which have been estimated to produce 2000 molecules of antibody per second (3). Plasma cells produce antibodies which recognize the specific antigen to which the original 8 cell was exposed. This initial recognition and antibody production is called a primary immune response because the antigen had not been seen by the immune system before. The second type of cell are memory 8 cells. These long-lived cells are programmed to remember the specific antigen with which it has come in contact so that if the memory cell sees the same antigen again, the response to the invasion can be carried out much more quickly. T cells are lymphocytes like B cells. These immune cells fall into two main subtypes which can be identified by unique cell surface protein markers, designated as "CD" numbers. A "CD" number is a name given to a molecule by an international conference which confers a cluster of differentiation (CD) number to molecules under stringent guidelines (3). 4

PAGE 15

Primary Response Naive BCeii Acli.DRd B ccn Secondary[ Response Figure 1.1. The B ceD. :i:tnnmne :response. When activated by an1Jgen. a .B cell Wlll. produce -plasma cells and :memocy cells during the primacy response. In the secondary response, memory cells can differeruiate more quicldy in1D plasma cells and :more memo.ry cells. This system of naming arose because, independently of one another, numerous laboratories began engineering antibodies which recognized different portions (called epitopes) of various molecules. Each laboratory had its own name for any given protein. The CD numbering system allows for consistent naming of these proteins even if the antibodies used to detect those molecules are different. 5

PAGE 16

One subset ofT cells is called T cytotoxic cells which can be identified by the cell surface marker CDS. Active T cytotoxic cells directly kill non-immune cells which are infected or damaged and which have antigen particles expressed on their surfaces. The T cytotoxic cell binds to the altered cell and disrupts its membrane, killing it. The other main subtype ofT cells is the T helper cell which can be identified by the surface protein CD4. These cells are aptly named because they help to activate the other types of immune cells. They recognize and bind to the antigen which is on the surface of APCs (see Figure 1.2). Once this binding occurs, the T helper cells become activated and begin to divide and produce small proteins (cytokines) which circulate in the blood to activate B cells, T cytotoxic cells, and other T helper cells. These cytokines also augment the stimulation of the T helper cell which produced them. Both the production of cytokines and actual binding to the antigen-presenting cell (B cell) help to activate the B cell response. Activated T cells of both types also produce memory T cells which function similarly to memory B cells. 6

PAGE 17

T=ccn0Q Acllftkd T lldpcr produces cym:t:iJH:s \..._) \ cells, aa4B cc:lls "'G _____ -+ APC l T cy1o1Dxic cell BccD 0 ::u;li.n.k:d T hclpcr cdls produce :memory T ccDs liDs aUcred body cdl F:ipcc 1.2. T cdl. acli.n.timt padtway cturiwa du: primary n:spome. "l1K: :memory T ccDs ]II'Oduced will rcspcm4 qaiddy 1D a seccmd. prescwDtioa ofaJdilea.. -By looking at cell surface markers expressed on T cells, we can distinguish between activated and non-activated phenotypes. The cell surface marker for unactivated T cells is CD45RA. The cell surface marker for activated I memory T cells is CD45RO. These two cell surface markers are different forms of a protein called CD45. The CD45 protein is a membrane-spanning molecule which has protein tyrosine phosphatase activity (PTP). 7

PAGE 18

While B cells and T cells are functioning to recognize antigens, the third component of the immune system joins the fray. This pathway is the called the complement cascade. Webster's definition of complement is "that which completes or perfects". The complement cascade acts via two distinct pathways to augment the cellular mechanisms by which antigen is removed from the body: the classical pathway and the alternative pathway. In the classical pathway, an antibody which has already been produced from a plasma cell tags an antigen on a damaged or infected cell. Complement proteins bind to the antibody I antigen complex on the cell and induce the series of proteolytic cleavages which activate other complement proteins to produce the complement cascade. This sequential protein cleavage pathway leads to ultimate lysis of the antibody-defined target cell. Like the classical pathway, the alternative pathway ultimately can lead to lysis of an altered cell. Unlike the classical pathway, the alternative pathway activates and recruits B cells to help the body recognize weak antigens. The alternative pathway begins with complement proteins binding to antigen instead of antibody-tagged antigen. The resulting complement protein cascade produces fragments which circulate in the blood or bind to cells. The complement fragments bound directly to the antigen on cells induce B cell I T cell activity known as cellular activation events. Since explicit antibodies are not produced until cellular activation occurs, the alternative pathway is very important in the early 8

PAGE 19

identification of low level antigen. It is also important in the removal of pathogens whose surfaces are not recognized easily as foreign by antigen-presenting cells (3,4). Figure 1.3 demonstrates the complement cascade. CJassical Pathway 1 cop1clllCBI proll:im bind. ID amiJodies u4 smt a cascade of pr011:in reaclimls .-.:.:_ ... ... _.,._ a prokia. compkx is asscmbkd whid tills du: ccD A11r:mative Pathway 1 u4 s1Ut a cascade of pi'Ok:ill reactions /1 cnses activatio:a. of iagw: ccDs Fiprc 1. 3. Dt: CowplCJDc:Dl Cascade: classical ad altanati:vc pathways. The overlapping interactions of the Complement cascade, B cells, and T cells are an excellent example of the type of checks and balances found in nature. 9

PAGE 20

This research deals with activation of the immune response through the convergence of these three immune pathways. This convergence of pathways produces a synergistic activation of B cells that results in a more efficient immune system. 1.4 Technical Background 1.4.1 A Model for a Signal Transducing Complex in B Cell Activation The molecular mechanisms by which B cells become activated are quite complicated. Recently, four to six integral membrane proteins have been found to associate non-covalently on the surface of B cells and to have related functions. This multimolecular protein complex mediates cell growth, motility, and cell-cell interaction. Four components of this complex have been identified by monoclonal antibodies purified antibodies which are produced to recognize a specific peptide sequence, or epitope, of a molecule. This group of proteins is collectively called the CR2 signal transduction complex (5,6,7,8,9). The basic model is that a complement fragment will bind to a cell or clump of debris and antibodies, and become an immune complex, which then binds to a complement receptor on a B cell (see Figure 1.4). The immune complex bound to the complement receptor is physically close to an antigen receptor. The antigen receptor will bind to the localized immune complex. These two binding events 10

PAGE 21

transduce signals through the CR2 signalling complex which result in a synergistic effect called cross-linking. Chapters 1.4.2 through 1.4.6 will discuss the properties of each protein and how they relate to one another. wifh bound compkrne:nt ftagme:nt '"' 3 cdl Wlh B ceH wi1b Anfisen an4 Siana1 Transduction C Ollllikx Figure 1.4. A model for signal amplification in B cells. (1) A fragment of a complement protein binds to its receptor (2) on the B ceU (Ch. 1.4. 2). (3) The antigen receptor of the B ceU binds to an antigen on an immune complex. Then growth signals are modulated by the other members of the signalliog complex (4; Ch. 1.4.3, 5; Ch. 1.4.4, and 6; Ch. 1.4.5) and transduced into the cell. 1.4.2 Complement Receptor Type 2 Complement Receptor Type 2 (see number 2 in Figure 1.4) is a member of the Regulators of Complement Activation family (RCA) which is a group of II

PAGE 22

proteins responsible for controlling cell activation by Complement cleavage products. This group includes Complement Receptor Type 1 (CD35), membrane cofactor protein (MCP, CD46), decay accelerating factor (DAF, CD55), C4b binding protein, and factor H in addition to Complement Receptor Type 2. Alternate names for this protein are CR2 or CD21. CR2 is an integral membrane protein found mainly on mature B cells and follicular dendritic cells (FDC), but also is found on cells of the thymus, endothelial cells, and subsets ofT helper and T cytotoxic lymphocytes (10). Extracellularly, this 145 kDa glycoprotein consists of 15 or 16 domains called SCRs or short consensus repeats (see Figure 1.5). These areas are composed of 60 to 70 amino acid sequences of which eleven have potential N-glycosylation sites. N-glycosylation sites are amino acid residues which are bound to oligosaccharide side chains and can be recognition sites. Sometimes they are associated with cell cell interactions (11). All of the SCRs have similar amino acid sequences. 12

PAGE 23

FigtD:e 1.5. A JDDdel of the proposed CR2 secondary slructure. 1lte repea1ing motifS calkd sbort consensm repea1S (SCRs). Red areas d:isulfide bond'i and green areas poaDal N-glycosyla1ion siks. Each one has between 20-40% homology with the other SCRs of CR2. Conserved in each SCR are four cysteines and one tryptophan. These areas are structurally significant within each domain as they confer the same secondary structure among the SCRs. A short transmembrane section of CR2 connects the SCRs to a 34 amino acid long C-terminal cytoplasmic domain (3,9, 12, 13, 14). CR2 is the receptor for several proteins (also called ligands). Each of these ligands binds to an SCR region which varies with the ligand. The primary physiological ligand for CR2 is a proteolytically cleaved fragment produced during the Complement cascade (see number 1 in Figure 1.4). This fragment is called C3dg (3, 15). 13

PAGE 24

C3dg is fonned during a regulatory process of the Complement cascade. The Alternative and Classical pathways of the Complement cascade converge when they reach a protein called C3. C3 is broken down to produce C3b which is membrane bound. This protein is required in both pathways to initiate the formation of a membrane attack complex which kills the cell. The production of this attack complex is blocked when CR1, Factor H, or MCP ( all members of the RCA family) binds to C3b. This allows C3b cleavage into two new fragments. One remains bound to the membrane and is further cleaved to produce C3dg (membrane bound) (4). The result of these cleavage reactions is that direct lysis of the target cell has been averted by blocking complement. However, C3dg remains on the surface of the cell and binds to the CR2 receptor of immune cells. Binding of CR2 activates pathways in immune cells and tags the cell for destruction so that complete clearance of the antigen can occur (3,4, 16, 17). The regulation of the Complement cascade by the production of C3dg, and its consequent binding to CR2 is of great significance because it allows regulation without sacrificing efficiency. Not only is CR2 needed for signalling initiated by Complement, but it is needed for T helper cells to interact with the self-recognition complex found on B cells. By themselves, some antigens do not elicit a very strong response from B cells. In order to initiate a robust response to these antigens, the B cells must interact directly with T helper cell. These are called T-dependent antigens (3). 14

PAGE 25

Mice which are genetically CR2 deficient produce lower amounts of antibodies in response to the model T -cell-dependent antigen source, sheep red blood cells, than their CR2 sufficient counterparts. Since antibody production relies on B lymphocytes, these studies reflect the importance of CR2 in the activation of those cells by T helper cells (18, 19). 1.4.3 An Amplification Molecule: CD19 Commonly used as a cell surface marker for B cells, another member of the complex is referred to as CD19 (see number 4 in Figure 1.4). CD19 is a 95 kDa integral membrane protein which is expressed only on B cells. It is an excellent marker for B cells as it is found from early in their development to plasma cell differentiation. The extracellular portion of CD19 is approximately 273 amino acid residues long and has five potential N-glycosylation sites (see Figure 1.6). 15

PAGE 26

Figure 1.6. A moddfor lhe proposed secondary sb.'Ucture of CD19. Thf: red dols :indicat: ctisulfide borul'i and the green areas :indicare potcnlial. N-glycosy:Jation silts. CD19 of human, mouse, and guinea pig is highly conserved (79% similar between human and mouse) in its 240 amino acid cytoplasmic domain (3). Some localized regions of the cytoplasmic tail have a strong net negative charge. Of the tail's amino acids, 19% have acidic R groups whereas only 9.5% have basic R groups. The preponderance of acidic R groups in the cytoplasmic tail domain may explain the function of CD 19 as a signal transduction molecule. Signalling through CD 19 appears to be very complex. Binding the extracellular portion of multiple CD 19 molecules with antibodies (cross-linking) can result in growth and proliferation or in inhibition depending on the type and amount of antibodies used. Activation of protein tyrosine kinases (PTK), increases in the level of phospholipase C (PLC) and 16

PAGE 27

increased intracellular concentrations of free calcium ions ([Ca2+]i) which result from cross-linking CD19 with antibodies suggest that CD19 has a role in proliferation and growth in B cells. However, binding of this CD19 with anti-CD19 antibodies alone inhibits the [Ca2+] increase produced by mitogen stimulation (3,5,9,21). Mitogens are chemicals which induce cell division. When CD19 was crosslinked with the antigen receptor of the B cell, only 0.03% of the antigen receptor needed to be bound to cause the cells to proliferate (20). This result suggests that CD19 acts in a very subtle, condition-dependent manner to amplify weak signals coming through the B cell antigen receptor (6). 1.4.4 The Tetraspan Family The CR2 signal transduction complex contains at least one of the members of this family of proteins (see number 5 in Figure 1.4). The first protein classified in this family and found to be associated with the signalling complex was a 26 kDa protein named TAPA...:l for Target of an Anti-Proliferative Antibody. It was characterized by Oren et al. in 1990 (22). It has been given the cluster designation of CD81. In addition to CD81, two other members of this family of proteins may be associated with the CR2 signalling complex. These two are named CD82 and CD53 (7 ,8,23,24). 17

PAGE 28

Each of the members of this protein family has a proposed structure which crosses the membrane four times and thus are referred to also as Transmembrane Super Family 4 (TMSF4) members. The proposed TMSF4 model (see Figure 1. 7) is comprised of four hydrophobic regions which are transmembrane domains, two large loop-like extracellular domains, and cytoplasmic Nand Ctermini (7,22,25,26). The amino acid sequence is 91% conserved between human and mouse CD81. Unlike other members of the tetraspan family, there are no N-glycosylation sites on its extracellular domains (9,27). NH2 COooOM Figure 1. 7. A cartDon of file JDDdel proposed for 1be 'Jl\..ISF4 members induding CD81, CD82, CDS3, and others. Tht:re are no poknDal.N-gtycosy1ation sites on CD81. Red indicl1c sulfur conJaiaing residues wlrich may condrule m confODDation. 18

PAGE 29

The TMSF4 superfamily proteins have structural similarities to a family of gated ion channels found in the nervous system which includes receptors for the neurotransmitters glutamate and acetylcholine. These similarities suggest that the two groups may have related biochemistries, although no direct evidence has been presented yet (26,28). The TMSF4 family of proteins has a rather widespread tissue distribution. CD81 is found ubiquitously in B cells but its expression is increased significantly following activation. Antibodies to TAPA-1 have been shown to inhibit cell proliferation and be involved with activation (9,27). There is evidence that CD81 is the molecule of the complex which is responsible for cell-cell interactions, specifically the clumping seen during the activation of B cells (6). 1.4.5 Leu-13 The final member of the CR2 signalling complex is an integral membrane protein whose primary amino acid structure has just recently been determined (see number 6 in Figure 1.4) (29,30). This protein is 16-17 kDa and is called Leu-13. A model for its secondary structure has not been proposed at this time. It is found on some white blood cells, endothelial cells and a specialized portion of the placenta. It physically associates with CD81 in some cell lines (31). Like CD81, antibodies to Leu-13 promote clumping (see Ch. 1.4.7) and inhibit proliferation. A cytokine 19

PAGE 30

called Interferon-a. (IFN-a.) causes an increase in the amount of Leu-13 on the cell surface (32,33). It has been suggested that the complex formed by Leu-13 and CD81 can associate with the complex of CR2 and CD19 to mediate cell adhesion events (6,9). 1.4.6 Interactions between the CR2 Signalling Complex Molecules Figure 1.8 depicts the associative relationships between the members of this complex. The circles represent the population of the corresponding molecules. Overlapping areas indicate which molecules have the strongest physical associations with one another. Associations of the Molecules of the CR2 Signalling Complex C1t2 Signalting Comp1ex r---1 01ha L___j Associalions Figure 1.8. ReprescruatWn of 1he stronger associa1ions betwec:n :members of 1he CR2 S:ignalling Complex. 01ha associations are mt shoWJL 20

PAGE 31

The diagram does not represent actual amounts of the molecules nor weak associations between the molecules of the complex. CR2 and CD19 are non covalently associated on the B cell surface. CD81 associates with both Leu-13 and CD19, but only directly with CR2 in some cell lines. Potential associations with proteins outside of this signal transduction complex are indicated by white areas but are not defined for the purpose of this paper. This macromolecular complex is comprised of molecules which also can be found alone or in complexes with different proteins. Under certain conditions in B cells, these molecules come together into an association which regulates the activation of those cells. This small population of cells provides an extremely important source of immune activation because sometimes an antigen is found in low concentrations or is not recognized very well by the immune system. In these situations, the CR2 signalling complex acts to lower the threshold of B cell activation so that even small stimuli can produce an antibody response (6). 1.4. 7 Events Following Activation Through the CR2 Signalling Complex In the body, blood is constantly cycled through secondary lymphoid tissues. These tissues, which include the spleen and lymph nodes, collect free antigens and antigen-antibody complexes from the blood. These immune complexes bind to follicular dendritic cells (FDCs) which are antigen-presenting cells in the secondary 21

PAGE 32

lymphoid tissues. Antigens stay on the FDCs for long periods of time (sometimes years) before they are degraded. As blood is cycling through the lymph tissues, B cells which have not been exposed to an antigen bind to the immune complexes via the model described previously (CR2-C3dg and antigen-antigen receptor). This initiates signalling events like ones through the CR2 signalling complex which activates the B cells. Each B cell begins to divide and quickly produces a colony of cells. This newly formed mass of densely aggregated cells is called the B cell germinal center. In each germinal center there are only 2 or 3 original B cells which produce the large colonies ( oligoclonal expansion). Memory T helper cells which are found in the germinal centers are thought to help segregate the proliferating B cells into types which have high-affinity or low-affinity receptors for antigen. The high-affinity B cells proliferate into memory cells and plasma cells. The low-affinity B cells undergo programmed cell death (apoptosis), thus removing weakly interacting cells from the immunologically active population (3). For experimental purposes an in vitro assay has been developed which mimics in vivo cell activation responses of the type that are involved in germinal center B cell proliferation. When immortalized B cell lines which express the CR2 signalling complex are incubated with immune complexes (immune complexes can be simulated by covalently linking purified C3dg molecules), a germinal center-like phenomenon occurs. Within an hour, the cells will change from their normal 22

PAGE 33

dispersed appearance to a densely clumped state (see Figure 1.9). This model for clumping of same-type cells is named homotypic adhesion. Figure 1. 9. An in vitro mocJel. for 1bc acti.va!io:n of B cells. Poly.maized C3dg is bol.Did 1D CRl on a IDmor cdl:tine. Vvilbin 1111. hour lbe cells dump IDgelhe:r via an llllknDwn. mecbanimt. This pbcrmntrum is caiicdlmnotypic atthesion. It is used widely to study B cell activation because it mimics the oligoclonal populations of the B cell germinal centers (3,34). The clumping activity can be produced using a wide panel of antibodies to CD81 and Leu-13. A limited number of proteins that bind to CR2 have been found to induce the clumping model, homotypic adhesion, in culture. These proteins are a subset of proteins which recognize the four SCRs at the amino-terminal end of the molecule and include polymerized C3dg and a monoclonal antibody (mAb or purified antibodies which 23

PAGE 34

recognize only one epitope) called OKB7. Not only do these proteins produce the adhesion response in immortalized cell lines, but also in primary cell populations( 15, 35). Our lab (37) has done extensive work to characterize CR2 and the mechanisms by which it influences cell-cell adhesion. The Balm-1 human B lymphoma cell line has been used in these endeavors because these cells express the members of the CR2 signalling complex and because their background rate of spontaneous adhesion is low. Some cell lines exist naturally in a highly clumped state which makes it difficult to determine when a positive adhesion reaction has occurred. In the Balm-1 cell line any ligand-induced adhesion reaction is easy to identify. This cell line undergoes homotypic adhesion in response to the same molecules as do other human B lymphoma cell lines and is not unique in its response. We have shown that the CR2 ligands which cause clumping act independently of other known adhesion molecules, such as selectins, integrins, LFA-1, and ICAM. Evidence to support this conclusion includes the following: 1) CR2induced adhesion can be induced in an integrin-deficient cell line, 2) the adhesion does not require divalent cations to undergo CR2-induced adhesion as do other known adhesion molecules, and 3) mAbs to LFA-1, VLA-4, and ICAM-1 do not inhibit CR2-induced adhesion. These data support other reports of LFA-1 24

PAGE 35

and divalent cation independence of homotypic adhesion through the CR2 signalling complex (31,33,36). 1.4.8 Development of an Antibody which Inhibits Homotypic Adhesion The above observations led to the search for a mAb which would block homotypic adhesion that was induced through the binding of C3dg or mAb OKB7 to CR2. The strategy adopted was to make mAbs which would inhibit the clumping of the cells in culture when OKB7 was present. The cell line which was used in the homotypic adhesion-producing assay, the Balm-1 cell, were injected into mice (as antigen). When these mice produced an immunological response, their spleen cells were fused with mouse myeloma cells. The resultant hybrid cells or hybridomas were immortalized antibody producing cells. Grown in culture, the new cells produced antibodies which could be screened for their ability to block the clumping of Balm-1 cells when C3dg was present. From over 5000 hybridomas produced, three mAb were identified which blocked the adhesion reaction. Two of the mAbs were found to recognize CR2. This finding internally validates the method by which the antibodies were developed. The third mAb (3A8) recognized an unknown protein. Figure 1.10 demonstrates the ability of mAb 3A8 to block the homotypic adhesion reaction 25

PAGE 36

which was induced by mAb OKB7. The mechanism for the mAb 3A8 inhibition of the adhesion reaction could not be explained by toxicity to cells nor by inhibition of C3dg/mAb OKB7 binding because it did neither of these things. 26

PAGE 37

anti-CR2 (mAb OKB7) anti-CR2 (mAb OKB7) + mAb3A8 Figure 1.10. Inhibition of Homotypic Adhesion by mAb 3A8. Balm-1 cells were induced to undergo CR2-induced homotypic adhesion with an anti-CR2 antibody (mAb OKB7) in the presence of mAb 3A8 or a negative control (anti-CD46). Cells were observed after 4 hours at 37C. Cells in the right photo illustrate mAb3A8-induced inhibition of the mAb OKB7induced homotypic adhesion reaction. 27

PAGE 38

1.4.9 Gamma-Glutamyl Transpeptidase is the Target of the Third Antibody The third antibody, mAb 3A8, immunoprecipitates two proteins which are unique to it. Immunoprecipitation is a technique in which cell membranes are disrupted using detergents. The resultant pieces of the cell are incubated with monoclonal antibodies. These mAbs bind to their target proteins and can be separated away from the unbound solubilized protein. Then the bound target proteins can be identified using electrophoretic techniques. The mAb 3A8 immunoprecipitated two proteins represented by bands at the 68 kDa and 27 kDa molecular weight range. These size bands are seen in immunoprecipitation and electrophoretic analysis of each type of cell to which the mAb 3A8 binds. Further separation of the two bands by two-dimensional gel purification allowed for N terminal sequencing of the antibody target protein (see Figure 1.11). Amino acid sequence homology was identical to the first 14 residues of an ectoenzyme which is involved in glutathione cycling (see Figure 1.14), y-glutamyl transpeptidase (GGT). The following studies were conducted to confirm that the sequenced protein bands were in fact GGT. 28

PAGE 39

684631-21-14-684631-21-14-Acidic Basic Acidic ISOTYPE CONTROL (UPC-10) Basic 3A8 Figure 1.11. The m.Ab 3A8 recognizes a two protein complex. Solubilized membrane proteins were immunoprecipitated with either UPC-10 (isotype control) or m3A8 and analyzed by two dimensional gel electrophoresis. The two proteins are identified by the boxes in the m.Ab 3A8 gel. GGT enzyme activity can be measured by a change in substrate from colorless to yellow in the presence of the enzyme GGT. In addition to the amino acid sequence identification, the target of mAb 3A8 was confirmed to be GGT by immunoprecipitating Balm-1 cells with mAb 3A8 and measuring loss of GGT enzyme activity in the supernatants. The mAb 3A8 removed most of the enzymatic activity from the supernatants compared to other mAbs. Figure 1.12 shows the units of GGT activity left in the supernatants after monoclonal antibody-29

PAGE 40

bound protein was removed. These enzymatic data support the sequencing results which identify the antibody target as GGT. We concluded that we had identified the GGT enzyme (37). 0.07 0.06 -=-.--=-;-=--Ul c1i 0.05 OJ) '"'4 Iii '"'4 ::::1 0.04 c-.t'j s 0.03 Q) 0.02 r:::: ::::1 O.D1 0.00 mAbused Figure 1.12. GGT activity left in solubilized protein after immunoprecipitation with antibody. Balm-1 cells were solubilized and immunoprecipitated with mAbs. The antibodies were removed and the remaining GGT activity was measured. 30

PAGE 41

1.4.10 Gamma-Glutamyl Transpeptidase: Structure and Function GGT is an integral membrane protein which is composed of two subunits that result from one precursor protein. Meister proposed the structural model shown in Figure 1.13 (38). The precursor protein is cleaved proteolytically and the subunits are assembled non-covalently. The active site is found on the smaller of the subunits but enzymatic function is dependent upon the association of the two subunits. Active sin: N Figure 1.13. A model. for the structure of gamma-gtutamyl tra:nspeptidase (GGT). The active s:i1e is on the s:maD. Slibunit, but both sobunils are needed for the enzyme to fonction. The green areas are porenDal glycosy1a1ion. sires. GGT has been studied primarily for its role in transport of nutrient cysteine into the cell, but is also the ectoenzyme required in the pathway which recycles the tripeptide glutathione (GSH) into the cell. GSH is an important molecule in 31

PAGE 42

various cellular mechanisms including free radical scavenging, nutrient cysteine source, leukotriene protection, and detoxification reactions (38,39,40). Very little glutathione can enter the cell whole and thus, must be broken down into components to be transported through the cell membrane (39,41,42). Figure 1.14 shows the glutathione pathway which is called the gamma-glutamyl cycle. This cycle is the major pathway for cells to recycle glutathione. GGT (a) cleaves GSH into cysteinyl-glycine and a glutamyl group which is subsequently transferred to water or to another peptide. A dipeptidase (b) cleaves cysteinyl-glycine either at the surface of the cell or within the cytoplasm. It is not fully known if the dipeptidase is found at the cell's surface, intracellularly, or both (43,44). Individual amino acids or the dipeptides then enter the cell by unknown transport mechanisms and are resynthesized into GSH (c-f). The rate limiting step in the intracellular synthesis of GSH is the synthesis of y-glutamyl-cysteine by y glutamyl-cysteine synthetase which is inhibited by excess glutathione (45). 32

PAGE 43

'GSH l (a)/ t'c OlJt GG'f m I GSH I (-1hione. synttme) (f) 1.5 Hypothesis lcys-glyl lcys I I glolamyt-a.a. I

I gluJamyl-a.a. I (c) I S-oxoprolinc I ... (d) ( S-oxOJ.l(Olinase) Since binding of GGT with an anti-GGT antibody (mAb 3A8) can inhibit CR2-induced homotypic adhesion in the Balm-1 B lymphoma cell line, and because CR2 is associated with the CR2 signal transduction complex in the alternative complement pathway, it can be hypothesized that GGT has a role in complementinduced B cell activation. The association of complement-induced activation and GGT was first observed in immortalized cell lines. We wanted to explore GGT levels on lymphocytes taken directly from normal human blood. The first hypothesis states 33

PAGE 44

that GGT will be expressed on peripheral blood mononuclear cells, or PBMC (which include lymphocytes). The second hypothesis is that GGT is associated with members of the CR2 signalling complex. Although the action of the anti-GGT mAb is related to CR2 directly, CR2 does not act alone in the activation of B cells. We wanted to determine if GGT is associated with any of the members of the CR2 signal transduction complex. 34

PAGE 45

2. Methods 2.1 Expression of GGT on Peripheral Blood Lymphocytes Preparation of Lymphocytes: Blood was donated from normal human donors (2 male). The blood was heparinized and diluted 2X with phosphate buffered saline (PBS). It was then carefully layered over ficoll-hypaque (Sigma, St.. Louis, MO) to separate the mononuclear cells from the other cell types. The cells were centrifuged for 25 minutes at 1500 RPM without braking. The mononuclear layer (immediately above the ficoll-hypaque) was removed and washed 2 times with PBS at 2000 RPM for 5 minutes (4C). No further separations were done to purify the lymphocytes. The cells were counted and resuspended to 5 x 106cells /ml in FACS buffer (PBS/1% Bovine Serum Albumin (BSA)/.01% Sodium Azide) (Sigma). Samples of 100f..tl each were placed in 75 x 100 polypropylene tubes so each was estimated to contain 5 x 105 cells. Immunofluorescence Staining: The cells were blocked (to reduce non specific staining) with 50 J.Ll of 10% heat-deactivated human serum in PBS for 10 minutes on ice. Either 3A8 or the isotype control UPC-10 (Sigma) was added as a primary antibody at a concentration of 1J.Lg/sample. These antibodies were incubated with the cells for 20 minutes on ice. The cells were washed 2 times 35

PAGE 46

as above in FACS buffer. Biotinylated sheep F(ab')2 anti-mouse IgG (Cappel, West Chester, PA) was added as the secondary antibody, incubated, and washed as before. The tertiary staining step included the addition of Avidin-Per-cp (Caltag, San Francisco, CA) to fluorescently label the biotin-coated secondary antibody + primary antibody complex, and FITC(fluoroscein isothiocyanate) or PE(phyco erythrin) conjugated antibodies to the following molecules: CD4 ( Coulter, Miami, FL, FITC-conj.), CDS (Coulter, FITC-conj.), CD45RA (Coulter, PE conj.), CD45RO (Coulter, PE-conj.), CD19 (Coulter, PE-conj.), GGT (unconjugated), and UPC-10 (isotype control). These additions were incubated and washed as above. After the final wash the cells were fixed in .4 ml of 1% paraformaldehyde (Sigma). Fluorescence-Activated-Cell-Sorting (FACS): Analysis was conducted at the UCHSC Core Facility on a Coulter Profiler Flow Cytometer. The analysis was done on a population of cells with the appropriate size and density for the lymphocyte population. This population was assessed for positive staining of either GGT or UPC-10 with Per-cp. The positive population also was assessed for positive FITC and Phyco-erythrin staining. Positivity was determined by setting the +Iquadrants on the mean channel fluorescence of the lymphocyte population of the negative control (UPC-10). Positive FITC and/or PE staining of the GGT population was determined in the same manner. Comparison of the various 36

PAGE 47

intensities of the markers was conducted qualitatively and by percentage of +I+ cells. These +I+ percentages between the CD45RA population and the CD45RO population for each donor were compared using the Student's paired T test. 2.2 Association of GGT with the CR2 Signal Transduction Complex Co-Capping I Immunofluorescense Staining : Co-capping experiments are used to identify co-localized membrane proteins to determine their presence in association complexes. Balm-1 cells which had > 90% viability as measured by Trypan Blue (Sigma) staining were washed 2 times in 37C PBS + 2% BSA (capping buffer) to remove the culture media. They were resuspended at 1 x 107 cells I ml in capping buffer. Samples of 1 x 10 6 cells were placed in non-adjacent wells of a 96 well V -bottom plate. A primary antibody was added at 1 J..lg per sample. The primary antibody was either the mAb 3A8, mAb CD46 (CD46 is an RCA family member which is not a member of the CR2 signalling complex) (lmmunotech, Marseille, France), or the isotype control (UPC-10) which is the same antibody subtype as mAb 3A8 but has no binding site on the cell. These antibodies were incubated with the cells for 30 minutes at 37C in the dark so that capping of associated membrane proteins could occur. The unbound antibodies were washed out 1 time with 37C capping buffer. The bound antibodies were 37

PAGE 48

simultaneously cross-linked and fluorescently labelled with 1 J..Lg per sample of sheep anti-mouse IgG antibody which was covalently linked to the TEXAS RED fluorescent molecule (Molecular Probes, Eugene, OR). This antibody was incubated with the samples for 60 minutes at 37C in the dark. Following this incubation the cells were washed 3 times with ice-cold PBS I 2% BSA I 0.02% Sodium azide to wash out the unbound antibody and to halt any further movement of membrane proteins. Next, mAbs to the members of the CR2 signal transduction complex were added to detect which members co-capped with GGT. These included anti-CD19 (mAb HD37, DAKO,Glostrup, Denmark), anti-CD82 (clone 50Fll, Pharmingen, San Diego, CA), anti-CD53 (cloneH129,Pharmingen), anti CD81 (mAb 5A6, donated by Shoshona Levy, Stanford, CA), or anti-CR2 (mAb HBS). The anti-CD19, anti-CD82, and anti-CD53 antibodies were directly conjugated to FITC, butthe anti-CD81 and anti-CR2 antibodies were biotinylated. The cells were incubated with lJ..Lg of one of the listed mAbs on ice for 20 minutes in the dark. The unbound antibody was washed out 3 times as before. The samples which had fluorescently coupled mAbs added were fixed with 4% paraformaldehyde and stored in the dark at 4C. The samples which had biotinylated mAbs added were incubated a final time in the dark, on ice for 20 minutes with streptavidin-FITC (Southern Biotechnology Associates, Inc., 38

PAGE 49

Birmingham, AB). Upon completion of the last incubation they were washed, fixed, and stored like the other samples. Mounting Cells for Confocal Microscopy: An anti-fade media was used to mount the cells on glass slides. This media was made fresh for each experiment out of 1 ml of Fluoromount (Southern Biotechnology Associates) and 1mg of p phenylene diamine (Sigma). The media was mixed on a rotator at room temperature for three hours. One drop of the media (3-4 was placed on the glass microscope slide and mixed with 5-7 of the cell suspension. A coverslip was placed over the mixture and allowed to dry in the dark. Observation by Confocal Microscopy: A BioRad Confocal Microscope was used for this analysis. Background settings were established using the samples which were stained with the isotype control (UPC-10). The level of signal was adjusted manually for consistency throughout all samples. Digital photographs were produced on a Mitsubishi Video Printer. Molecules stained with FITC were observed as green and molecules stained with TEXAS REDTM were observed as red. Each fluorochrome was observed using the appropriate wavelength filter and then the files were merged to observe areas of colocalization which appeared yellow. No statistical analyses were done using this method. 39

PAGE 50

3. Results 3.1 Expression of GGT on Human Lymphocytes The human peripheral blood was purified to include all the mononuclear cells. Our interest was in the lymphocyte population of these cells. Using FACS analysis, PBMC populations are distinguished based on size and strength of fluorescent cell surface markers. The strength of the fluorescence can be used to infer the relative amounts of each surface marker. To limit the group of cells which were examined for fluorescence, cells falling outside the lymphocyte size range were excluded on the size vs. density scatter plot. Density in this case refers to the amount of granularity of the cell. Figure 3.1 indicates the sizes of different types of cells found in peripheral blood and where they typically fall on this type of scatter plot. Polymorphonuclear cells are highly granulated cells which include eosinophils, basophils, and neutrophils. Monocytes are cells which differentiate into phagocytic cells called macrophages. The computer generated "gating" enables the analysis to focus only on the cell types of interest. In populations which are not highly purified there will be contamination of other cell types in the size range of interest. Included in the lymphocyte size range are other cell types such 40

PAGE 51

INCREASING DENSITY INCREASING SIZE Figure 3.1 Size exclosio:n.FACS scalrerplot Fromperi;pbernlblood, cell will be segregated due to 1heir size and densily. The cells of :inrerest for fhis research are 1he l,ympbocytcs. Note lhat lhere is overlap between. :regions. as natural killer cells and monocytes. FACS analysis can exclude these cell types on the basis of fluorescence. Figure 3.2 is a scatter plot of the cells of appropriate size which also are fluorescing in two colors. This figure is a representative sample of the experimental data. The first fluorescent molecule is tagging CD19, the B cell surface marker. The second fluorescent molecule is tagging GGT. The lines breaking the scatter plot into quadrants are placed according to the background staining found on the isotype control (UPC-10) sample. All of the cells which are recorded to the left and below the quadrant lines are negative for the marker on that axis. 41

PAGE 52

B cells without GGT 1 +/-non-B cells without GGT B cells with GGT 2 +I+ 4 AMOUNTOFGGT non-B cells withGGT Figure 3.2. GGT expression on B cells. Of the lymphocyte sized population, these cells were analyzed for CD19 (FITC fluorescence) and GGT (Per-cp fluorescence). As indicated, quadrant number 2 contains cells which are B cells and have GGT on their surfaces. 42

PAGE 53

The quadrants can be labeled as follows: 1) upper left quadrant=B cells which have no GGT on their surface ( +/-); 2) upper right quadrant=B cells which have high amounts of GGT on their surface (+I+); 3) lower left quadrant = nonB cells which have no GGT on their surface(-/-); 4) lower right quadrant = non-B cells which have high amounts of GGT on their surface ( -/ +). These data demonstrate that GGT is expressed on a population of peripheral blood B cells. This population represents approximately 2% of the total gated lymphocytes, but approximately 25% of the total B cell population as defined by the B cell marker CD19. In addition to B cells, T cells were analyzed. Three color staining ofT cells allowed identification of four subsets of T cells. T helper (CD4 +) and T cytotoxic (CD8+) cells were identified with FITC stain. These populations were further defined by PE stain to determine activated (CD45RO+) or unactivated (CD45RA +) phenotype. All cells were stained with Per-cp stain to label GGT. So, with three types of fluorescent molecules, the amount of GGT can be determined on four subtypes ofT cells: unactivated T helper cells, activated/memory T helper cells, unactivated/naive T cytotoxic cells, activated/memory T cytotoxic cells. Figure 3.3 represents these staining patterns. Like the B cell scatter plots, the upper right quadrant contains the cells that are positive for the second color marker (activated or non-activated) and the cells that have GGT on their surfaces (+I+ quadrant). 43

PAGE 54

Table 3.1 summarizes the T cell data. The data is expressed as percent of analyzed cells which are in quadrant 2 (+I+). The data from these donors suggests that expression of GGT on activated /memory T cells was significantly greater than on unactivated/ naive T cells. In order to demonstrate this the unactivated T cells were compared to the activated T cells for each donor using a paired Student's T test. For both groups ofT cells, T cytotoxic cells and T helper cells, the difference between the unactivated cells and the activated cells was significant. The resultant p values for these analyses were p < 0.01 forT cytotoxic cells and p < 0.05 forT helper cells. 44

PAGE 55

T CYTOTOXIC CELLS (CDS+) 1 +I-2 +I+ 4 -I AMOUNT OF GGT 1 +I-2 +I+ 4 -/+ AMOUNT OF GGT T HELPER CELLS (CD4+) 1 +I,_ ..:' 2 +I+ AMOUNT OF GGT 1 +Ii . _; .; t;ili' 2 +I+ 4 -/+ AMOUNT OF GGT Figure 3.3. GGT's presence on subsets ofT cells. Four subsets ofT cells are represented in these scatter plots: unactivated T cytotoxic cells, activated T cytotoxic cells, unactivated T helper cells, and activated T helper cells. Quadrant 2 represents the cells of those subsets which have GGT present. 45

PAGE 56

(A) GGT+ T CYTOTOXIC CELLS (CDS+) EXPERIMENT UNACTIVATED ACTIVATED NUMBER (CD45RA+) (CD45RO+) 1 6.9 14.3 2 0.5 3.8 3 2.0 8.3 4 5.5 8.4 5 3.2 11.0 (B) GGT+ T HELPER CELLS (CD4+) EXPERIMENT UNACTIVATED ACTIVATED NUMBER (CD45RA+) (CD45RO+) 1 4.0 19.9 2 0.3 8.4 3 1.5 16.4 4 1.3 10.8 5 2.3 19.6 Table 3.1. Percent ofT cell subsets which are GGT positive (quadrant 2). (A) shows the percent of all T cytotoxic cells which are either unactivated or activated and have GGT on their surfaces, (B) shows the percent of all T helper cells which are either unactivated or activated and have GGT on their surfaces. The cell markers used for identification are given in parentheses. 46

PAGE 57

3.2 Association of GGT with CR2 Signalling Complex Members Individual cells which were representative of each sample were photographed. Each photo represents the merge of two single color confocal images. The red areas indicate those portions of the cell which have either the negative control (CD46) protein or GGT stained. The molecule CD46 is related to CR2 (RCA family), but it does not associate with the CR2 signalling complex. The green areas indicate those portions of the cell which have the CR2 signalling complex member stained (CR2, CD19, CD81, CD82, or CD53). Leu-13 was not studied due to limited supply of the anti Leu-13 monoclonal antibody. Areas where the two molecules are colocalized are indicated by yellow. Yellow co-capped areas of the cells were clearly visible in the samples which were stained with GGT and the CR2 signalling complex members (CR2, CD19, CD81, CD82, or CD53). The control samples (CD46 and CR2 signalling complex members) for each of those combinations had little or no yellow co-capped areas on the cells. These data suggest that GGT is associated with CR2, CD19, CD81, CD82, and CD53 in varying degrees on the surface of Balm-1 cells. 47

PAGE 58

A 8 Figure 3.4 (a) Merged confocal images of a BalmI cell stained with anti-CD46 + Texas Red ( in red) and anti-CR2 (mAb HB5) + FITC ( in green) are the negative control, (b) Another BalmI cell prepared the same way as (a) but stained with anti-GGT (mAb 3A8) + Texas Red (in red) and anti-CR2 + FITC (in green). The colocalized areas are shown in yellow.

PAGE 59

A 8 Figure 3.5. (a) Merged confocal images of a Balm-1 cell stained with anti-CD46 + Texas Red ( in red) and anti-CD 19 (mAb HD37) + FITC ( in green) are the negative control, (h) Another Balm-1 cell prepared the same way as (a) but stained with anti-GGT (mAb 3A8) + Texas Red (in red) and anti-CD19 + FITC (in green). The colocalized areas are shown in yellow.

PAGE 60

A 8 Figure 3.6. (a) Merged confocal images of a BalmI cell stained with anti-CD46 + Tex.as Red (in red) and anti-CD81 (mAb 5A6) + FITC (in green) are the negative control, (b) Another BalmI cell prepared the same way as (a) but stained with anti-GGT (mAb 3A8) + Texas Red (in red) and anti-CD81 + FITC (in green). The colocalized areas are shown in yellow. 50

PAGE 61

A 8 Figure 3.7. (a) Merged confocal images of a Balm-1 cell stained with anti-CD46 +Texas Red ( in red) and anti-CD82 + FITC ( in green) are the negative control, (h) Another Balm1 cell prepared the same way as (a) but stained with anti-GGT (mAb 3A8) + Texas Red (in red) and anti-CD82 + FITC (in green). The colocalized areas are shown in yellow. 51

PAGE 62

A B Figure 3.8. (a) Merged confocal images of a BalmI cell stained with anti-CD46 + Texas Red (in red) and anti-CD53 + FITC (in green) are the negative control, (b) Another Balm-! cell prepared the same way as (a) but stained with anti-GGT (mAb 3A8) + Texas Red (in red) and anti-CD53 + FITC (in green). The colocalized areas are shown in yellow. 52

PAGE 63

4.0 Discussion In these studies we have further clarified the relationship between the GGT enzyme and the CR2 signalling complex. We have found that GGT is expressed on peripheral blood lymphocytes in addition to immortalized cell lines and that it does colocalize on the surface of the cell with the protein members of the CR2 signalling complex. The initial observations of GGT expression on lymphocytes were made using immortalized cell lines. Therefore, experiments were conducted to demonstrate that GGT is expressed on normal human blood lymphocytes and which subsets of lymphocytes. B cells were analyzed because the model used to produce the anti-GGT antibody in this lab was a B cell activation model. As Fig 3.2 shows, there is a population of B cells (25% of marked B cells) which do have a significant amount of GGT on their surfaces. The CR2 signalling complex is thought to modulate very small signals to insure B cell activation ( 6). The observation that the population of B cells which express high levels of GGT is rather small is not contradictory to what was expected. The cells used to produce mAb 3A8 were a mature B cell line which was activated through CR2. It is possible that the population of B cells which expresses high levels of GGT is a population which has been activated and is proliferating and secreting antibody. 53

PAGE 64

Other studies have shown that B cells typically have low levels of GGT enzyme activity and lower GGT amounts expressed on their surfaces than do other types of peripheral blood cells like natural killer cells and monocytes( 46). Monocytes are thought to help lymphocytes regulate their glutathione levels by secreting sulfur containing thiols (47). The cells which have been shown to have higher activities and surface expression of GGT enzyme are those which also are proliferating and secreting material ( 46). In the future, an attempt to dissect out the effect of activation on GGT expression in peripheral B cells is needed. Surface markers which adequately distinguish activated B cells from naive B cells have not been identified. To look at an activated population of B cells, the cells can be cultured with a known mitogen (a molecule which stimulates cell growth) such as LPS (lipopolysaccaride) or SAC (Staph aureas Cowan Type A) for several days. After the addition of an appropriate mitogen, resting B cells become activated and proliferate. Following stimulation, they are stained for the amount of GGT on their surface. The differences of GGT staining between day 0 and day 3 can be used as evidence for changes due to activation. These differences would be exaggerated due to activation of the entire B cell population. Comparison of mitogen-stimulated B cells with untreated B cells in culture would account for tissue culture effects as well as baseline GGT levels in circulating PBMC. 54

PAGE 65

Because we wanted to examine GGT expression on lymphocytes in general, we also looked at T cells. T cells have well-documented surface markers which distinguish between unactivated/ naive types and activated/memory types. Using three color flow cytometric analysis, we looked at the unactivated and activated phenotypes of both T cytotoxic cells and T helper cells (Figure 3.3 and Table 3.1). We did not see any distinct differences between the amount of GGT on the surface of T helper cells and T cytotoxic cells. This agrees with data presented by Tager (48). However, we did see a significant increase in the amount of GGT present on activated cell types over the non-activated cell types in both T helper cells and T cytotoxic cells. Interestingly, new studies have shown that the unactivated/naive cells of advanced AIDS patients die disproportionately and that these cells have a lowered amount of glutathione intracellularly ( 49). It is possible that a lower expression of GGT may result in death due to the inability of these cells to recycle glutathione. In addition, Staal et al. has shown that a portion of the HIV viral gene can be regulated by controlling intracellular glutathione levels. Lowered glutathione levels in the cell activate the transcription factor NF-KB which allows the virus to be replicated (50). So a cell which has a higher level of intracellular glutathione may be able to withstand the virus longer. Neither the GGT enzyme activity nor expression on the surface of these diseased cells has been studied. Also, tumor cells which have increased glutathione levels may be able to withstand 55 --------------------------

PAGE 66

oxidative death due to anti-cancer therapies better than their low glutathione level counterparts (51). Tager et al. suggests that since some drug resistant-lymphomas and myelomas have increased expression of the GGT enzyme on their surfaces, these tumor cells might have a higher survival rate when treated with cysteine depleting chemotherapy drugs (48). BandT cells which have higher amounts of GGT present on their surfaces or an increased level of intracellular glutathione may be able to withstand higher oxidative stresses as well. Although the anti-GGT antibody (3A8) was produced using an activated B cell model, it was produced through CR2 activation specifically. In addition to being associated with the CR2 signalling complex of B cells, CR2 is found associated with other B cell proteins, by itself in the membrane of B cells, and on other types of cells. For example, it is found on subsets of T helper cells and T cytotoxic cells. Although the lymphoid cells, B and T cells, serve different functions, they may have redundant surface proteins that act in a similar fashion on the two cell types. These cell types also may undergo some of the same metabolic changes while becoming activated, so the presence of similar mechanisms is likely. The development of the anti-GGT antibody using a B cell activation model and the finding that activated T cell subsets have increased levels of GGT support the hypothesis that GGT plays a role in immune cell activation. 56

PAGE 67

The Confocal Microscopy experiments conducted for this paper were used as a third method of establishing GGT's association with the members of the CR2 signalling complex. The two methods used previously by our lab (37) establish molecular and functional associations between GGT and the complex by co immunoprecipitation studies. Results of these experiments were similar in that the TMSF4 members showed the highest amount of GGT activity precipitating with them. These methods did not reveal an association between GGT and CD19 nor CR2. The problem with immunoprecipitation assays is that different detergents maintain different types of structural associations. So an association which is real may not be maintained by the available detergents. In our immunoprecipitation experiments with CR2 and CD 19, no association with GGT has been observed. This lack of association is not consistent with the manner of development of the anti-GGT antibody. The Confocal Microscopy experiments were done in the absence of detergents, so that any association between CR2 and GGT could be observed. The results of the Confocal experiments show an association between the members of the CR2 signalling complex and GGT. In support of the co immunoprecipitation studies, GGT is associated most strongly with the members of the TMSF4 family (CD81, CD82, and CD53). Additionally, both CR2 and CD19 can be seen to co localize with _the enzyme GGT using this method. The strong association of GGT with the TMSF4 members in all three techniques may result 57

PAGE 68

from two phenomena: (1) the antibodies to the TMSF4 members may bind strongly to their epitopes on their respective proteins in highly accessible areas; or (2) the expression of the TMSF4 members is so great that any association with GGT is easily observed. Conversely, the antibodies to CR2 and CD19 may not bind as strongly if they lack easily accessible epitopes, or their expression may not be as ubiquitous as that of the TMSF4 members, making them harder to observe by these techniques. The anti-GGT antibody (mAb 3A8) may also lack strong binding, an accessible epitope, or GGT is not expressed in great numbers on these cells. Because mAb 3A8 inhibits the clumping seen in the homotypic adhesion reaction and can be seen using flow cytometry, the third reason is most likely. The flow cytometry data presented here suggests that GGT is highly expressed in only a small population of cells so it is possible that some physical associations may be below the detection limits of the immunopreciptation assays used. In accordance with the evidence for each of these proteins' existence with and without the other members of the complex, the confocal data indicate areas of co-localization and areas of independently stained proteins. However, the relative amount of co-localization between members of the signalling complex is greater than the amount of co-localization of those proteins and the control protein CD46. These Confocal images support the hypothesis that GGT is associated with the CR2 signalling complex. Associations between GGT and CR2, CD19, CD81, CD82, 58

PAGE 69

and CD53 have not been shown previously. GGT has been related to growth induction in a myeloid cell line. When hematopoietic growth factors are incubated with the KG-1 myeloid cell line, significant increases in GGT enzyme activity are observed. These increases support a role for GGT in growth and proliferation of myeloid cells(52). Since the CR2 signalling complex is involved with growth and proliferation in B cells, the association of GGT with its members supports the hypothesis that the GGT enzyme plays a role in lymphocyte growth events (5,6,9). However, it is not known at this time whether GGT' s involvement is a cause or an effect of the cellular events which promote growth in immune cells. The function of GGT in the CR2 signalling complex has not yet been established. It may be that GGT becomes associated with the complex in response to increased amino acid need during growth. It may be that an altered level of glutathione is needed in the cell in response to oxidizing activation events. But, GGT may in fact act within the CR2 signalling complex to further modulate the activation by low-level antigens within the body. If this is the case, it would be a novel function for GGT. In the future, the involvement of GGT with the CR2 signalling complex will be dissected out by using cellular techniques such as the ones described in this thesis. In addition, a strain of mouse has been bred to lack the gene for GGT (53). Studies using these mice will indicate if there is any divergence from the normal 59

PAGE 70

development of B and T cells. Such a divergence may point toward specific times during the development of an immune response where GGT is most important. The time at which GGT is needed may indicate its involvement with the CR2 signalling complex. Whether the function of GGT is to cause activation events or is an effect of activation events further demonstrates the ability of our bodies to evolve a limited number of proteins for a variety of functions. These findings support the idea that not only do the cells of the immune system overlap in function, but individual proteins of the body may perform functions which act upon a variety of systems. These many overlapping functions can make mistakes. Portions of the immune system can become stimulated by damaged cell proteins which in some cases can lead to serious overcompensation by the immune system against healthy cells in the body. This is what happens in autoimmune disorders. By understanding the mechanisms by which our cells and systems work together, we can learn how to restore and maintain harmony among cells which have lost the ability to work together. 60

PAGE 71

BIBLIOGRAPHY (1) Modrich, P. Mismach Repair, Genetic Stability, and Cancer. Science 1994; 266:1959-1960. (2) Crowley, LV. Introduction to Human disease. third edition. 1992. Jones and Bartlett Publishers, Inc. Boston, MA. 820pps. (3) Paul WE. Fundamentllf Immunology. third Edition.1993, Raven Press,NY. 1490pps. (4) Kuby J. Immunology. second edition. 1994. W H Freeman and Co, NY. 660 pps. (5) Bradbury LE, Kansas GS, Levy S, Evans RL, Tedder TF. The CD19/CD21 signal transducing complex of human B lymphocytes includes the target of antiproliferative antibody-1 and Leu-13 molecules. J Immunol 1992; 149: 28412850. (6) Matsumoto AK, Martin DR, Carter RH, Klickstein LB, Ahearn JM, Fearon DT. Functional dissection of the CD21/CD19/TAPA-1/Leu-13 complex of B lymphocytes. J Exp Med 1993; 178:1407-1417. (7) Olweus J, Lund-Johansen F, Horejsi V. CD53, a protein with four transmembrane spanning domains, mediates signal transduction in human monocytes and B cells. 1993 J Immunol; 151: 707-714. (8) Rasmussen AM, Blomhoff HK, Stokke T, Horejsi V, Smeland EB. Cross linking of CD53 promotes activation of resting human B cells. 1994 J Immunol; 153: 4997-5007. (9) Tedder TF, Zhou L-J, Engel P. The CD19/CD21 signal transduction complex of B lymphocytes. Immunol Today 1994; 15: 437-442. (10) Fischer E, Delibrias C, Kazatchkine MD. Expression of CR2 (the C3dg/EBV receptor, CD21) on normal human peripheral blood T lymphocytes. J lmmunol; 146 : 865-869. 61

PAGE 72

(11) Voet D, Voet J. Biochemistry. 1990. John Wiley and Sons, NY. 1223pps. (12) Molina H, Perkins SJ, Guthridge J, Gorka J, Kinoshita T, Holers VM. Characterization of a complement receptor 2 (CR2, CD21) ligand binding site for C3: an initial model of ligand interaction with two linked SCR modules. J Immunol 1995; 154: 5426-5435. (13) Barlow PN, Baron N, Norman DG, Day AJ, Willis AC, Sim RB, Campbell ID. Secondary Structure of a complement control protein module by two dimensional H NMR. Biochem 1991; 30: 997-1004. (14) Perkins SJ, Haris, PI, Sim RB, Chapman D. A study of the structure of Human Complement Component Factor H By Fourier transform infrared spectroscopy and secondary structure averaging methods. Biochem 1988; 27: 4004-4012 ( 15) Carell JC, Myones BL, Frazier B, Holers VM. Structural requirements for C3d,g/Epstein-Barr virus receptor (CR2/CD21) ligand binding, internalization, and viral infection. J Bioi Chern 1990; 265:12293-12299. (16) Bohnsack JF, Cooper NR. CR2 ligands modulate human B cell activation. J lmmunol 1988; 141 :2569-2576. ( 17) Luxembourg AT, Cooper NR. Modulation of signalling via the B cell antigen receptor by CD21, the receptor for C3dg and EBV. J lmmunol 1994; 4448-4457. (18) Croix AD, Ahearn MJ, Rosengard MA, HanS, Kelsoe G, Minghe M, Carroll MC. Antibody Response to aT-dependent Antigen Requires B Cell Expression of Complement Receptors. J Exp Med 1996; 183:1857-1854. (19) Molina H, Holers VM, LiB, Fang Y-F, Mariathasan D, Goellner J, Strauss Schoenberger J, Karr RW, Chaplin D. Markedly impaired humoral immune response in mice deficient in complement receptors 1 and 2. Proc Nat! Acad Sci 1996; 93:3357-3361. (20) Carter RH, Fearon DT. CD 19: lowering the threshold for antigen receptor stimulation of B lymphocytes. Science 1992; 256: 105-107. 62

PAGE 73

(21) Pesando JM, Bouchard LS, McMaster BE. CD19 is functionally and physically associated with surface immunoglobulin. J Exp Med 1989; 170: 21592164. (22) Oren R. Takahashi S, Doss C, Levy R, Levy S. TAPA-1, the target of an antiproliferative antibody, defines a new family of transmembrane proteins. Mol Cell Biol 1990; 10:4007-4015. (23) Mannion BA, Berditchevski F, Kraeft S, Chen LB, Hemler ME. Transmembrane-4 Superfamily Proteins CD81 (TAPA-1), CD82, CD63, and CD53 Specifically Associate with Integrin a4b1 (CD49d/CD29). J Immunol 1996; 157:2039-2047. (24) Szollosi J, Horejsi V, Bene L, Angelisova P, Damjanovich S. Supramolecular complexes of MHC class1, MHC class II, CD20, and Tetraspan molecules (CD53, CD81, CD82) at the surface of a B cell line JY. J Immun, 1996; 157: 2939-2946. (25) Schick MR, Hguyen VQ, Levy S. Anti-TAPA-1 antibodies induce protein tyrosine phosphorylation that is prevented by increasing intracellular thiol levels. J Immunol 1993; 151: 1918-1925. (26) Wright MD, Tomlinson MG. The ins and outs of the transmembrane 4 superfamily. Immunol Today; 15:588-594. (27) Levy S, Nguyen VQ, Andria ML, Takahashi S. Structure and membrane topology of TAPA-1. J Biol Chern 1991;266: 14597-14602. (28) Wo ZG, Oswald, RE. Unraveling the modular design of glutarnated-gated ion channels. Trends Neuroscience 1995; 18:161-168. (29) Deblandre GA, Marinx OP, Evans SS, Majjaj S, Leo 0, Caput D, Huez GA, Wathelet MG. Expression cloning of an interferon-inducible 17 k-Da membrane protein implicated in the control of cell growth. JBC 1995; 270 (40) :23860-23866. (30) Reid LE, Drasnett AH, Gilbert CS, Porter ACG, Gewert DR, Stark GR, Kerr IM. A single DNA response element can confer inducibility by both alpha and gammainterferons. PNAS 1989; 86: 840-844. 63

PAGE 74

(31) Takahashi S, Doss C, Levy S, Levy R. TAPA-I, the target of an antiproliferative antibody, is associated on the cell surface with the Leu-13 antigen. 1 lmmunol 1990; 145:2207-2213. (32) Deblandre G, Marinx OP, Evans S, Majjaj S, Leo, 0, Caput D, Huez GA, Wathelet MG. Expression Cloning of Interferon-inducible 17-kDa membrane protein implicated in the control of cell growth. 1 Bioi Chern 1995; 270:2386023866. (33) Evans SS, Collea RP, Leasure 1A, Lee DB. IFN-a induces homotypic adhesion and Leu-13 expression in human B lymphoid cells, 1 Immunol 1993; 150:736-747. (34) Ramirez R, Carracedo 1, Mooney N, Charron D. HLA class-11-mediated homotypic aggregation: involvement of a protein tyrosine kinase and protein kinase c. Human Immunology 1992; 34:115-125. (35) Bjork P, Elenstrom-Magnusson C, Rosen A, Severinson E, Paulie S. CD23 and CD21 function as adhesion molecules in homotypic aggregation of human B lymphocytes. Eur 1 Immunol 1993; 23: 1771-1775. (36) Kansas GS, Tedder TF. Transmembrane signals generated through MHC class II, CD19, CD20, CD39, and CD40 antigens induce LFA-1 dependent and independent adhesion in human B cells through a tyrosine kinase-dependent pathway. 1 Immunol 1991; 147:4094-4102. (37) Nichols T, Guthridge 1, Fletcher D, Holers, VM. A role for gamma glutamyl transpeptidase in CD19/CD21 complex function. Manuscript in progress, 1997. (38) Tate S, Meister A. g-Glutamyl transpeptidase: catalytic, structural and functional aspects. Mol Cell Biochem 1981 ;39:357-368. (39) Meister A. Transport and metabolism of glutathione and g-glutamyl amino acids. Group meeting of the peptide and protein grouuup of the Biochemical Society and the Royal Society of Chemistry, held at the University of Durham on 30 March 1983, and organized and edited by 1. W. Payne. 64

PAGE 75

(40) Monks TJ, Anders MW, Dekant W, Stevens JL, Lau SS, and Bladeren PJ. Glutathione conjugate mediated toxicities. Toxicology and Applied Pharmacology 1990; 106: 1-19. (41) Kannan R, Yi JR, Tang D, Zlokovic BV, Kaplowitz N. Identification of a novel, sodium dependent, reduced glutathione transporter in the rat lens epithelium. Invest Opthalmol Vis Sci. 1996; 37: 2269-2275. (42) Sze G, Kaplowitz N, Ookhtens M, Lu SC. Bidirectional membrane trasport of intact glutathione in Hep G2 cells. Am. J. Physiol. 1993;265: G 1128-G 1134). ( 43) Habib GM, Barrios R, Shi ZZ, Lieberman MW. Four distinct membrane bound dipeptidase RNAs are differentially expressed and show discordant regulation with g-glutamyl transpeptidase. J Bioi. Chern. 1996; 271:16273-16280. (44) Adachi H, Kubota I, Okamura N, Iwata H, Tsujimoto M, Nakazato H, Nishihara T, Noguchi T. Purification and characterization of human microsomal dipeptidase. J. Biochem. 1989;105: 957-961. (45) Deneke S, Fanburg BL. Regulation of cellular glutathione. Am J Physiol 1989; 257:L163-L173. (46) Grisk 0, Kuster U, Ansorge S. The activity of g-glutamyl transpeptidase (g GT) in populations of mononuclear cells from human perepheral blood. Bioi. Chern. Hoppe-Seyler 1993; 374: 287-290. (47) Gmuder H, Eck HP, Benninghoff B, RothS, Droge. Macrophages regulate intracellular glutathione levels of lymphocytes. Evidence for an immunoregulatory role of cysteine. Cell Immunol 1990; 260: 32-46. (48) Tager M, Ittenson A, Franke A, Frey A, Gassen HG, Ansorge S. g-Glutamyl transpeptidase-cellular expression in populations of normal human nononuclear cells and patients suffering form leukemias. Ann. Hematol. 1995;70: 237-242. (49) Aukrust P, Svardal AM, Muller F, Lunden B, Nordoy I, Froland SS. Markedly disturbed glutathione redox status in CD45RA( +) cells in human 65

PAGE 76

immunodeficiency virus type 1 infection is associated with selective depletion of this lymphocyte subset. Blood 1996: 88(7) : 2626-2623. (50) Staal FJT, Roederer M, Herzenberg LA, Herzenberg LA. Intracellular thiols regulate activation of nuclear factor kB and transcription of human immunodeficiency virus. PNAS 1990; 87: 9943-9947. (51) HaniganMH, Ricketts WA. Extracellular glutathione is a source of cysteine for cells that express gamma-glutamyl transpeptidase. Biochemistry 1993; 32 : 6302-6306. (52) Miller AM, Sandler E, Kobb M, Eastgate J, Zucali J. Hematopoeeietic growth factor induction of gamma-glutamyl transferase in the KG-1 myeloid cell line. Environmental Hematology 1993; 21 :9-15. (53) Lieberman MW, Wiseman AL, Shi ZZ, Carter BZ, Barrios R, Ou CN, Chevez-Barrios P, Wang Y, Habib GM, Goodman JC, Huang SL, Lebovitz RM, Matzuk MM. Growth retardation and cysteine deficiency in gamma-glutamyl transpeptidase-deficient mice. PNAS 1996; 91 : 7923-7926. 66