University of Massachusetts Medical School eScholarship@UMMS
GSBS Dissertations and Theses Graduate School of Biomedical Sciences
2008-01-10
CD4 T Cell-Mediated Lysis and Polyclonal Activation of B Cells During Lymphocytic Choriomeningitis Virus Infection: A Dissertation
Evan Robert Jellison University of Massachusetts Medical School
Let us know how access to this document benefits ou.y Follow this and additional works at: https://escholarship.umassmed.edu/gsbs_diss
Part of the Biological Factors Commons, Cells Commons, Hemic and Immune Systems Commons, and the Viruses Commons
Repository Citation Jellison ER. (2008). CD4 T Cell-Mediated Lysis and Polyclonal Activation of B Cells During Lymphocytic Choriomeningitis Virus Infection: A Dissertation. GSBS Dissertations and Theses. https://doi.org/ 10.13028/4g3d-jc06. Retrieved from https://escholarship.umassmed.edu/gsbs_diss/349
This material is brought to you by eScholarship@UMMS. It has been accepted for inclusion in GSBS Dissertations and Theses by an authorized administrator of eScholarship@UMMS. For more information, please contact [email protected].
CD4 T CELL-MEDIATED LYSIS AND
POLYCLONAL ACTIVATION OF B CELLS DURING
LYMPHOCYTIC CHORIOMENINGITIS VIRUS INFECTION
A dissertation presented
By
EVAN ROBERT JELLISON
Submitted to the Faculty of the University of Massachusetts,
Graduate School of Biomedical Sciences, Worcester in
partial fulfillment of the requirements for the degree of:
DOCTOR OF PHILOSOPHY
January 10, 2008
IMMUNOLOGY AND VIROLOGY ii
COPYRIGHT NOTICE
Parts of this dissertation have appeared in the following publications:
Jellison ER, Kim SK, Welsh RM. (2005). Cutting edge: MHC class II-restricted killing in vivo during viral infection. J. Immunol. 174(2):614-8.
Jellison ER, Guay HM, Szomolanyi-Tsuda E, Welsh RM. (2007). Dynamics and magnitude of virus-induced polyclonal B cell activation mediated by
BCR-independent presentation of viral antigen. Eur. J. Immunol. 37(1):119-28. iii
CD4 T CELL-MEDIATED KILLING AND POLYCLONAL ACTIVATION OF B CELLS DURING LYMPHOCYTIC CHORIOMENINGITIS VIRUS INFECTION A Dissertation Presented By Evan Robert Jellison Approved as to style and content by:
Lawrence J. Stern, Ph.D., Chair of Committee
Rachel M. Gerstein, Ph.D., Member of Committee
Edmund J. Gosselin, Ph.D., Member of Committee
Timothy F. Kowalik, Ph.D., Member of Committee
Liisa K. Selin, M.D. Ph.D., Member of Committee
Raymond M. Welsh, Ph.D., Thesis Advisor
Anthony Carruthers, Ph.D.,
Dean of the Graduate School of Biomedical Sciences Program in Immunology andVirology January 10, 2008 iv
ACKNOWLEDGEMENTS
I dedicate this work to my family. If not for their constant support and
encouragement, I would never have sought this degree. I especially want to recognize my mother who passed away during my time in graduate school. Prior to her passing, her
constant encouragement always set me on the right course. I also cannot forget my father,
who has been a diligent listener during my time in graduate school and has successfully
embraced the role of “the encourager” that was left to him.
I also wish to recognize the members of the Welsh, Selin, and Szomolanyi-
Tsuda labs both past and present. Not only did their work provide the foundation for my
own work, but their personalities and friendships will not soon be forgotten. They helped
create an environment of intellectual stimulation where scientific achievement is almost
guaranteed. Thank you all for making me want to come to the lab each and every day. I
especially want to thank S.K. Kim, Heath Guay, and Eva Szomolanyi-Tsuda for their
specific contributions to my thesis work. Their help pushed at least one aspect of this
thesis forward in some way. I was also fortunate to work with several very talented
technicians who were never too busy to lend a hand. To Keisha Mathurin, Tatyana
Andreyeva, Carey Zammitti, Keith Daniels, and Cara West, thank you.
Of course, no graduate student can achieve success without an outstanding
mentor. I was fortunate and privileged to have Ray Welsh as my advisor in graduate
school. His scientific curiosity is infectious. His attention to detail made me more
professional. His openness made it easy to discuss anything with him. His regard for v
honesty and integrity has been and will be instilled in me forever. Thank you Ray for everything you have done for me.
Finally, to my wife Hillary, you will always be my lady. Thank you for propping me up whenever I needed it. I can’t imagine sharing my life with anyone else.
vi
ABSTRACT
CD4 T cells and B cells are cells associated with the adaptive immune system.
The adaptive immune system is designed to mount a rapid antigen-specific response to pathogens by way of clonal expansions of T and B cells bearing discrete antigen-specific
receptors. During viral infection, interactions between CD4 T cells and B cells occur in a
dynamic process, where B cells that bind to the virus internalize and degrade virus
particles. The B cells then present viral antigens to virus-specific CD4 T cells that
activate the B cells and cause them to proliferate and differentiate into virus-specific
antibody-secreting cells. Yet, non-specific hypergammaglobulinemia and the production
of self-reactive antibodies occur during many viral infections, and studies have suggested that viral antigen-presenting B cells may become polyclonally activated by CD4 T cells in vivo in the absence of viral engagement of the B cell receptor. This presumed polyclonal B cell activation associated with virus infection is of great medical interest because it may be involved in the initiation of autoimmunity or contribute to the long- term maintenance of B cell memory.
In order to directly examine the interactions that occur between T cells and B cells, I asked what would happen to a polyclonal population of B cells that are presenting
viral antigens, if they were transferred into virus-infected hosts. I performed these studies
in mice using the well-characterized lymphocytic choriomeningitis virus (LCMV) model
of infection. I found that the transferred population of antigen-presenting B cells had two fates. Some antigen-expressing B cells were killed in vivo by CD4 T cells in the first day vii
after transfer into LCMV-infected hosts. However, B cells that survived the cytotoxicity underwent a dynamic polyclonal activation manifested by proliferation, changes in phenotype, and antibody production.
The specific elimination of antigen-presenting B cells following adoptive transfer into LCMV-infected hosts is the first evidence that MHC class II-restricted killing can occur in vivo during viral infection. This killing was specific, because only cells expressing specific viral peptides were eliminated, and they were only eliminated in
LCMV-infected mice. In addition to peptide specificity, killing was restricted to MHC class II high cells that expressed the B cell markers B220 and CD19. Mice depleted of
CD4 T cells prior to adoptive transfer did not eliminate virus-specific targets, suggesting that CD4 T cells are required for this killing. I found that CD4 T cell-dependent cytotoxicity cannot be solely explained by one mechanism, but Fas-FasL interactions and perforin are mechanisms used to induce lysis.
Polyclonal B cell activation, hypothesized to be the cause of virus-induced hypergammaglobulinemia, has never been formally described in vivo. Based on previous studies of virus-induced hypergammaglobulinemia, which showed that CD4 T cells were required and that hypergammaglobulinemia was more likely to occur when virus grows to high titer in vivo, it was proposed that the B cells responsible for hypergammaglobulinemia may be expressing viral antigens to virus-specific CD4 T cells
in vivo. CD4 T cells would then activate the B cells. However, because the antibodies
produced during hypergammaglobulinemia are predominantly not virus-specific, non-
virus-specific B cells must be presenting viral antigens in vivo. viii
In my studies, the adoptively transferred B cells that survived the MHC class II-
restricted cytotoxicity became polyclonally activated in LCMV-infected mice. Most of
the surviving naïve B cells presenting class II MHC peptides underwent an extensive
differentiation process involving both proliferation and secretion of antibodies. Both
events required CD4 cells and CD40/CD40L interactions to occur but B cell division did
not require MyD88-dependent signaling, type I interferon signaling, or interferon γ signaling within B cells. No division or activation of B cells was detected at all in virus- infected hosts in the absence of cognate CD4 T cells and class II antigen. B cells taken from immunologically tolerant donor LCMV carrier mice with high LCMV antigen load became activated following adoptive transfer into LCMV-infected hosts, suggesting that
B cells can present sufficient antigen for this process during a viral infection. A transgenic population of B cells presenting viral antigens was also stimulated to undergo polyclonal activation in LCMV-infected mice. Due to the high proportion of B cells stimulated by virus infection and the fact that transgenic B cells can be activated in this manner, I conclude that virus-induced polyclonal B cell activation is independent of B cell receptor specificity. This approach, therefore, formally demonstrates and quantifies a virus-induced polyclonal proliferation and differentiation of B cells which can occur in a
B cell receptor-independent manner.
By examining the fate of antigen-presenting B cells following adoptive transfer into LCMV-infected mice, I have been able to observe dynamic interactions between virus-specific CD4 T cells and B cells during viral infection. Adoptive transfer of antigen-presenting B cells results in CD4 T cell-mediated killing and polyclonal ix
activation of B cells during LCMV infection. Studies showing requirements for CD4 T cells or MHC class II to control viral infections must now take MHC class II-restricted cytotoxicity into account. Polyclonal B cell activation after viral infection has the potential to enhance the maintenance of B cell memory or lead to the onset of autoimmune disease.
x
TABLE OF CONTENTS
COPYRIGHT NOTICE...... ii
ACKNOWLEDGEMENTS...... iv
ABSTRACT…………………………………………………………….………………..vi
TABLE OF CONTENTS...... x
LIST OF TABLES...... xiv
LIST OF FIGURES...... xv
ABBREVIATIONS…………………………………………………………………...xvii
CHAPTER I: INTRODUCTION……………………………………………………….1
A. THE LCMV MODEL OF VIRAL INFECTION………………………………….…..2
B. T CELL SUBSETS……………………………………………….……………….....11
C. CYTOTOXIC CD4 T CELLS…………………………….…………………………15
D. THE CD4 T CELL RESPONSE TO LCMV…………………….…………………..17
E. B CELLS AND THEIR ACTIVATION………………….……………………...…..21
F. THE B CELL RESPONSE TO LCMV…………………….……………...………...28
G. HYPERGAMMAGLOBULINEMIA AND VIRUS-INDUCED POLYCLONAL
ACTIVATION OF B CELLS……………….……………………………………….32
H. THESIS OBJECTIVES……………………………………………………………...49
CHAPTER II: MATERIALS AND METHODS……………………………………..51
A. MICE…………………………………………………………………………………51 xi
B. VIRUSES, PROTEIN ANTIGEN, AND MOUSE INOCULATION………………52
C. PREPARATION OF CELLS………………………………………………………..52
D. ANTIBODIES AND REAGENTS…………………………………………………..53
E. FLOW CYTOMETRY………………………………………………………...... 54
F. INTRACELLULAR CYTOKINE STAIN…………………………………………..54
G. IN VIVO CYTOTOXICITY ASSAY…………………………………………….…55
H. PLAQUE ASSAY……………………………………………………………………56
I. ADOPTIVE TRANSFER OF ANTIGEN-LOADED CELLS………………………56
J. ANTIBODY AND ANTIGEN-SPECIFIC ELISA………………………………….57
K. ANTIBODY AND ANTIGEN-SPECIFIC ELISPOT………………………...... 58
L. STATISTICAL ANALYSIS………………………………………………………...59
CHAPTER III: MHC CLASS II-RESTRICTED CYTOTOXICITY IN VIVO
FOLLOWING LCMV INFECTION...... 60
A. ELIMINATION OF LCMV MHC CLASS II PEPTIDE-COATED TARGET CELLS
IN INFECTED ANIMALS………………………………………...………………...61
B. CD4 CELL-DEPENDENT CYTOTOXICITY………………..…………………….64
C. THE ROLE OF FAS-FASL INTERACTIONS IN MHC CLASS II-RESTRICTED
KILLING………………………………………………………………...…………..66
D. THE ROLE OF TNF-TNFR INTERACTIONS IN MHC CLASS II-RESTRICTED
KILLING………………………………………………………………………….....67
E. THE ROLE OF PERFORIN IN MHC CLASS II-RESTRICTED KILLING……....70 xii
F. THE ROLE OF CD40-CD40L INTERACTIONS IN MHC CLASS II-RESTRICTED
KILLING…………………………………………………………………………….71
G. THE ROLE OF INTERFERON GAMMA IN MHC CLASS II-RESTRICTED
KILLING………………………………………………………………..…………...73
H. CHAPTER SUMMARY ………………………………..…………………………..75
CHAPTER IV: THE POLYCLONAL ACTIVATION OF B CELLS FOLLOWING
LCMV INFECTION……………………………………………………………….77
A. PROLIFERATION OF LCMV CLASS II EPITOPE-EXPRESSING B CELLS IN
VIVO IN LCMV-INFECTED ANIMALS………………….………………………78
B. CHARACTERIZATION OF PHENOTYPIC CHANGES IN POLYCLONALLY
ACTIVATED B CELLS……………………………………….……………………83
C. CD4 CELL-DEPENDENT PROLIFERATION………………….…………………88
D. THE ROLE OF CD40-CD40L INTERACTIONS IN POLYCLONAL B CELL
ACTIVATION…………………………………………………….………………...93
E. THE ROLE OF MYD88-DEPENDENT SIGNALING IN POLYCLONAL B CELL
ACTIVATION…………………………………………………………………….…93
F. THE ROLE OF TYPE I INTERFERON SIGNALING IN POLYCLONAL B CELL
ACTIVATION……………………………………………………………………….95
G. THE ROLE OF INTERFERON-GAMMA SIGNALING IN POLYCLONAL B
CELL ACTIVATION………………………………………………………………..97
H. POLYCLONAL ACTIVATION OF B CELLS FROM LCMV CARRIER MICE…99
I. ANTIBODY PRODUCTION BY POLYCLONALLY ACTIVATED B CELLS…103 xiii
J. POLYCLONAL ACTIVATION OF A TRANSGENIC B CELL
POPULATION……………………………………………………………………..108
K. SUMMARY OF CHAPTER FOUR………………………………………………..114
CHAPTER V: DISCUSSION...... 116
A. THE CYTOTOXIC CD4 T CELL RESPONSE TO LCMV……………………….117
B. MECHANISMS BY WHICH CD4 T CELLS KILL THEIR TARGETS………….120
C. ANALYSIS OF THE B CELL POPULATION THAT UNDERGOES A
POLYCLONAL ACTIVATION AFTER LCMV INFECTION…………………...126
D. THE ANTIBODIES PRODUCED DURING VIRUS-INDUCED POLYCLONAL
ACTIVATION…………………………………………………………………..…129
E. IMPLICATIONS OF VIRUS-INDUCED POLYCLONAL B CELL
ACTIVATION……………………………………………………………………..131
F. CONCLUSIONS…………………………………………………………………...137
CHAPTER VI: REFERENCES……………………...………………………….…...139 xiv
LIST OF TABLES
Table Page
III-1 Specific killing of LCMV peptide-coated target population in vivo…….63
III-2 MHC class II-restricted killing in the absence of CD40 or IFN-γ
signaling………………………………………………………………….72
IV-1 Changes in B cell number as a result of virus-induced polyclonal B cell
activation…………………………………………………………………80 xv
LIST OF FIGURES
Figure Page
III-1 MHC class II-restricted killing shown by in vivo cytotoxicity assay……62
III-2 Depletion of CD4+ T cells by treatment with GK1.5 ablates MHC class II
restricted killing………………………………………………………….65
III-3 IFN-γ production by gld or C57BL/6 mice in response to GP61
stimulation………………………………………………………………..68
III-4 The surviving GP61-coated target B cells have elevated Fas
expression………………………………………………………………..69
III-5 The surviving GP61-coated CD40KO B cell targets fail to upregulate Fas
expression in LCMV-infected hosts……………………………………..74
IV-1 Polyclonal activation of B cells coated with LCMV-specific class II
peptides…………………………………………………………………..79
IV-2 Polyclonally activated B cells display an activated phenotype………….84
IV-3 Dynamic changes of Ig expression occur in polyclonally activated B cells
as they divide…………………………………………………………….86
IV-4 LCMV-induced polyclonal activation results in Fas upregulation on B
cells………………………………………………………………………87
IV-5 B cells of three distinct phenotypes in the spleen undergo LCMV-induced
polyclonal activation……………………………………………………..89 xvi
IV-6 CD4 depletion by mAb GK1.5 ablates polyclonal B cell activation in
LCMV-infected mice…………………………………………………….90
IV-7 Polyclonal activation of LCMV peptide-coated B cells can occur during
the memory state…………………………………………………………92
IV-8 Polyclonal B cell activation requires CD40-CD40L interactions………..94
IV-9 Polyclonal B cell activation does not require MyD88-dependent
signaling………………………………………………………………….96
IV-10 Type I interferon signaling in B cells is not required for LCMV-induced
polyclonal B cell activation……………………………………………...98
IV-11 Interferon-γ receptor signaling in B cells is not required for LCMV-
induced polyclonal B cell activation……………………………………100
IV-12 B cells taken from donor mice experiencing high antigen loads proliferate
upon transfer into LCMV-infected host mice………………………….102
IV-13 Ig allotype mismatched IgHa B cells coated with LCMV-specific peptides
secrete Ab in LCMV-infected hosts following adoptive transfer………105
IV-14 Ig allotype mismatched IgHa B cells coated with LCMV-specific peptides
secrete class-switched Ab in LCMV-infected hosts following adoptive
transfer………………………………………………………………….106
IV-15 Virus-induced proliferation of BCR transgenic B cells coated with viral
antigens in LCMV-infected mice……………………………………….111
IV-16 Transgenic B cells coated with LCMV-specific peptides secrete antibody
in LCMV-infected hosts……………………………………...………...112 xvii
ABBREVIATIONS
Ab antibody
AEC 3-amino-9-ethylcarbazole
APC antigen presenting cell
ASC antibody secreting cell
BCIP 5-bromo-4-chloro-3'-indolyphosphate p-toluidine Salt
BCR B cell receptor
BHK baby hamster kidney cells
CD complementarity determinant
CFSE 5(6) carboxyfluorescein diacetate succinimidyl ester
CMV cytomegalovirus
CTL cytotoxic T lymphocyte
ELISA enzyme-linked immunosorbent assay
ELISPOT enzyme-linked immunospot assay
FACS florescence activate cell sorting
FCS fetal calf serum
FDC follicular dendritic cell
GC germinal center
GP glycoprotein
HA hemagluttinin
HCV hepatitis C virus xviii
HIV human immunodeficiency virus
HRP horse radish peroxidase i.c. intracranial i.p. intraperitoneal i.v. intravenous
IFN interferon
IFNR interferon receptor
IFN-γ interferon gamma
Ig immunoglobulin
IL interleukin
KO knock out
LCMV lymphocytic choriomeningitis virus
LDV lactate dehydrogenase elevating virus mAb monoclonal antibody
MCMV murine cytomegalovirus
MEM minimum essential medium
MFI mean fluorescence intensity
MHC major histocompatibility complex
MZ marginal zone
NBT nitro-blue tetrazolium chloride
NP nucleoprotein
PBS phosphate buffered saline xix
SHM somatic hypermutation
TCR T cell receptor
TLR toll-like receptor
TD T cell dependent
TI T cell independent
Th helper T cell
TNF tumor necrosis factor
Treg regulatory T cell
VSV vesicular stomatitis virus
WT wild-type 1
CHAPTER I
INTRODUCTION
This thesis will examine some of the interactions that occur between CD4 T cells
and B cells during a viral infection. To accomplish this, I performed several adoptive
transfer studies, where B cells were coated with viral antigens and transferred into virus-
infected hosts. During the course of my thesis, I will show that the transferred population
of antigen-presenting B cells has two fates. CD4 T cells can eliminate B cells in virus-
infected mice. This major histocompatibility complex (MHC) class II-restricted killing is the first evidence of CD4 T cell-dependent cytotoxicity in vivo during viral infection. I also show that those B cells that survive killing, somewhat paradoxically, become polyclonally activated by CD4 T cells. Polyclonal B cell activation was implicated to be the cause of virus-induced hypergammaglobulinemia that is seen after various human and murine virus infections. However, my demonstration of virus-induced polyclonal B cell activation is the first, and through the approach described herein, I am able to make
specific conclusions about virus-induced hypergammaglobulinemia and polyclonal B cell activation. To examine these processes in detail, I used the well-characterized lymphocytic choriomeningitis virus (LCMV) model of virus infection in mice where the kinetics of the immune response have been studied extensively, the antigenic MHC class
II epitopes are defined, and where hypergammaglobulinemia has been previously observed. 2
A. The LCMV model of viral infection. LCMV infection of mice is perhaps the best studied model of viral infection. Studies of mice infected with LCMV have allowed researchers to define many aspects of the mammalian immune system. These definitions have translated across other models of infection and have lead to a deeper understanding of our own immune response. Concepts such as immune tolerance, virus-induced immunopathology, immune complex disease, T cell recognition of antigen-MHC complexes, T cell education in the thymus, NK cell activation, virus-induced cellular dysfunction, immune deficiency, autoimmunity, immunologic memory, and the cellular effector mechanisms used to control the viral infections have all been founded based on studies performed in LCMV-infected mice (Welsh, 2000). Two Nobel prizes have been awarded based on discoveries made with the LCMV system. The first was awarded to
Burnet based on his theories of immune tolerance (Burnet, 1991). The second was awarded to Doherty and Zinkernagel for the discovery MHC restriction, the process by which T cells recognized virus-infected target cells through MHC molecules
(Zinkernagel and Doherty, 1974).
Although LCMV is a natural pathogen of mice, humans can be infected. LCMV infection clinically presents in humans initially as a viremia with fever and myalgia.
Other symptoms include headache, photophobia, and vomiting. Concurrent with these symptoms are leukopenia and thrombocytopenia. This period is followed by the onset of the cellular immune response with swelling of the meninges and lymphocyte infiltration of the central nervous system (Buchmeier et al., 2001). 3
LCMV was the first arenavirus to be isolated. It has an ambisense RNA genome with two gene segments, a long segment (L) and a short segment (S). The genome encodes four viral proteins: GP, the viral glycoprotein; NP, the viral nucleoprotein; L, the viral polymerase; and z, a RING finger protein that may act as a scaffold for the viral glycoprotein (Buchmeier et al., 2001). The GP protein is a precursor that is cleaved into two separate proteins before becoming part of the virion (Wright et al., 1990). The interaction between each of these proteins and the cellular immune response has been studied. For instance, in C57BL/6 mice, the GP, NP, and L protein epitopes presented by
MHC class I molecules have been identified, and their subsequent stimulations of CD8 T cells have been studied (Whitton et al., 1988; van der Most et al., 1998; Masopust et al.,
2007; Kotturi et al., 2007). The GP and NP epitopes presented by MHC class II molecules in C57BL/6 mice and their subsequent stimulation of CD4 T cells have also
been characterized (Oxenius et al., 1995; Varga and Welsh, 1998a). For B cells, GP is the
primary target of the neutralizing antibody (Ab) response, although non-neutralizing
responses against other proteins are also detected (Wright et al., 1989; Battegay et al.,
1993). The details of some of these cellular immune responses will be outlined in later
sections.
LCMV may be one of the most useful models of viral infection because it has the
ability to generate different patterns of disease and different immune responses,
depending on the route of inoculation, the strain of virus used, the strain of mouse
infected, and the age of the infected mouse. Three laboratory strains of LCMV were
originally isolated in the 1930’s: Armstrong, Traub, and WE. Armstrong, named after the 4
person who originally isolated the virus, was isolated from a monkey experiencing severe
lymphocytic choriomeningitis, hence the virus’ name. Traub, a serologically identical
strain also named after the investigator who originally isolated it, was found in an
infected colony of mice at Rockefeller University. WE is a strain isolated from a patient
“W.E.” who contracted LCMV from the Traub animal colony and subsequently died.
Some other strains that are used in research were derived from these original three.
Docile, a highly replicating strain of LCMV isolated from mouse liver was derived from
WE (Pfau et al., 1982). Clone 13, another highly disseminating strain of LCMV, has a
broader cellular tropism and better ability to grow in infected cells than its parent strain,
Armstrong (Ahmed et al., 1984). Each strain and the different immune response
generated by it has been used to study the various outcomes of viral infection. As an
example of this, there are differences in hypergammaglobulinemia that occur when
infecting mice with Armstrong, WE, Docile, and clone 13. These differences and the
phenomenon of hypergammaglobulinemia will be discussed in section G.
LCMV infection in mice results in either a long-term persistent infection, an acute
infection followed by sterilizing immunity, or lethal meningoencephalitis (Welsh, 2000).
LCMV does not kill the infected host. Instead, the immune response to LCMV causes the
pathology associated with the virus. As mentioned above, the route of infection and dose
of infection can change the outcome of viral pathogenesis and the immune response. If
the virus is given peripherally at any site, low to moderate doses of virus result in the
generation of a potent cytotoxic T lymphocyte (CTL) response, which clears the virus,
and leads to lifelong immunity. However, virus given intra-cranially at these same doses 5
leads to a severe acute meningoencephalitis, with CD8 T cell infiltrates entering the brain and meninges. These infiltrates lead to the eventual death of the infected animal.
Infection with high doses of virus or infection with moderate doses of highly disseminating strains of the virus, such as Clone 13 and Docile, results in a completely different outcome than acute infection. Here, both peripheral and intracranial (i.c.) infections of mice results in persistent viral infections. A CTL response is generated in these mice but rapidly declines, and a phenomenon now known as clonal exhaustion occurs which was originally defined as a “high dose immune paralysis”(Hotchin, 1962).
Clonal exhaustion is a state in which T cells die or become dysfunctional following prolonged exposure to antigen (Moskophidis et al., 1993). Several mechanisms have been proposed to account for clonal exhaustion, including the loss of functional antigen presenting cells (APC) and the absence of CD4 T cell help (Althage et al., 1992; Borrow et al., 1991; Borrow et al., 1995; Battegay et al., 1994; Matloubian et al., 1994). Both result in less effective CD8 T cell priming and enhancement of clonal exhaustion. Some of the receptors and molecular requirements in CD8 T cells for clonal exhaustion are beginning to be identified (Barber et al., 2006; Wherry et al., 2007). For instance, the over-expression of the CD28 family member PD1 on exhausted CD8 T cells has been observed, and blockade of PD1 signaling in CD8 T cells results in the restoration of function in clonally exhausted CD8 T cells (Barber et al., 2006). Others have found that the cytokine IL-10 is involved in the exhaustion of CD8 T cells during LCMV clone 13 infection and that blockade of IL-10 or IL-10R results in the restoration of CD8 T cell function and subsequent viral clearance (Brooks et al., 2006; Ejrnaes et al., 2006). 6
The chronic LCMV infection that can occur after high doses of infection or as a result of infection with highly disseminating strains of LCMV is due to the clonal exhaustion of T cells which prevents the control of virus infection. However, clonal exhaustion is not the only means to generate a chronic LCMV infection. Age also plays a role in the outcome of LCMV infection. Mice infected with LCMV in utero or less than
24 hours after birth do not develop a protective T cell response to LCMV. These mice enter into a “carrier state” with a persisting life-long viremia. These observations led to the concept of immunological tolerance proposed by Burnet, who stated that the immune system will recognize all antigens as self during its development and will not make a response to them. Congenitally-infected LCMV carrier mice can, however, generate antiviral antibodies that form immune complexes and mediate circulatory system disease
(Oldstone and Dixon, 1967). Thus, tolerance in congenitally infected LCMV carrier mice is restricted to T cells and not B cells.
The dynamics of the immune response to acute LCMV infection have been well characterized. The immune response can be divided into sections according to time after infection: the early response, which obviously occurs in the first few days after viral infection; the induction phase, which corresponds to the proliferation of lymphocytes in response to the virus; the apoptosis phase, which occurs at the height of the immune response; the immune silencing phase, which is part of the apoptosis phase and corresponds to the reduction in lymphocyte number; and the memory phase which occurs at the end immune response after the decline in lymphocyte number. All aspects of the mammalian immune system seem to be involved in the clearance of acute LCMV 7
infection. Both the innate (Welsh, Jr. and Zinkernagel, 1977; Zhou et al., 2005) and the adaptive immune systems are triggered by LCMV infection, and the adaptive immune response elicits both humoral (Moskophidis and Lehmann-Grube, 1984) and cell- mediated responses (Cole et al., 1972).
The early response to LCMV involves the production of type I interferon (IFN)
and the induction of NK cells. This phase of the immune response lasts for about four
days after viral infection. Type I IFN is an antiviral cytokine that is induced to high levels
following acute infection with LCMV and peaks two days after infection (Merigan et al.,
1977). LCMV infection of mice deficient in type I IFN or type I IFNR signaling generate
higher LCMV titers. These titers differed among strains of LCMV, suggesting that
different strains had different susceptibilities to the antiviral effects of type I IFN(Muller
et al., 1994; van den Broek et al., 1995). Furthermore, strains of LCMV that do not
normally disseminate into a chronic viral infection can become chronic in mice that lack
an IFN response. Type I IFNs also activate and induce the proliferation of NK cells
during this early phase of the immune response to LCMV(Biron et al., 1983; Biron et al.,
1984). However, these NK cells do not seem to play any role in the control of virus
infection, as mice depleted of NK cells have normal LCMV titers(Bukowski et al., 1983).
Furthermore, LCMV-infected cells are no more sensitive to NK cell-mediated lysis than
uninfected cells(Bukowski et al., 1985). In addition to its effect on NK cells, type I IFN is
also involved in the attrition of memory-phenotype T cells, which occurs around day 3 of
viral infection (McNally et al., 2001). In an acute infection, mice infected with LCMV
lose approximately 50% of their memory-phenotype (CD44 high) T cells. This attrition 8
affects CD8 T cells and CD4 T cells, but does not occur in type I IFNR KO mice. This
loss of memory-phenotype T cells is polyclonal, virus-specific T cells are not spared, and
it has been hypothesized to be important for the generation of space which may be
required for the impending immune response (Bahl et al., 2006). Other studies have also
indicated that type I IFNs have a direct effect in the clonal expansion of CD4 and CD8 T
cells during acute LCMV infection (Kolumam et al., 2005; Havenar-Daughton et al.,
2006). Mice deficient in type I IFN R had decreased survival of both CD4 and CD8 T
cells after LCMV infection. Overall, the signals provided by type I IFNs early after
LCMV infection help shape the immune response.
For the clearance of acute LCMV infection, cell-mediated immunity seems to be
more important than humoral immunity. Mice deficient in B cells clear acute LCMV infection, albeit more slowly than mice with an intact B cell compartment(Cerny et al.,
1988; Brundler et al., 1996). However, acutely infected mice deficient in CD8 T cells develop a long-term persistent infection (Lehmann-Grube et al., 1993). Although there isn’t an absolute requirement for B cells for the elimination of virus, LCMV stimulates B cell activation, and humoral immunity is protective against LCMV infection. Transfer of a neutralizing monoclonal Ab protects mice after i.c. infection with LCMV even if administered up to two days post-infection (Wright and Buchmeier, 1991). Antibody- mediated protection, however, may not be completely protective. Studies performed in suckling mice where antiviral antibodies were passively transferred to pups by LCMV-
immune mothers showed that only those pups with an intact CD8 T cell compartment
were protected from neonatal LCMV infection (Baldridge et al., 1997). In this same 9
study, similar results were obtained in adult mice that received neutralizing monoclonal antibodies for protection against Clone13. These adult mice were protected from persistent clone 13 infections only when CD8 T cells were present. These results suggest that humoral immunity must act in concert with cell-mediated immunity to confer protection from LCMV.
The difference between acute and persistent LCMV infection seems to stem from the ability of the host to generate a protective CTL response before the virus grows to high levels and induces clonal exhaustion. Antibodies that neutralize infecting particles limit early viral spread and allow the cell-mediated response to develop. The cell- mediated, or T cell response to LCMV develops during the induction phase of the infection. Acute LCMV infection induces a massive T cell response(Masopust et al.,
2007). There is an expansion in the number of CD4 and CD8 T cells seen 6 days post- infection and peaks at 8 to 9 days post-infection. High numbers of LCMV-specific CTL can be detected at the peak of the response and CD8 T cells respond more vigorously than CD4 T cells in the induction phase (Moskophidis and Lehmann-Grube, 1989;
McFarland et al., 1992; Varga and Welsh, 1998b). LCMV titers begin to fall at day 5 of the acute infection and the virus is basically cleared before the peak in T cell number and peak in CTL response. The peak in T cell number also corresponds to peaks in the production of IFN-γ, IL-2 production by T cells, and Ab production by B cells(Gessner et al., 1990; Kasaian and Biron, 1989; Moskophidis and Lehmann-Grube, 1984).
Following the induction phase, at 8 to 10 days post-infection, T cells become highly sensitive to apoptosis in LCMV-infected mice (Razvi and Welsh, 1993). This 10
apoptosis can be induced through activation-induced cell death (AICD), a phenomenon that occurs when IL-2 and antigen receptor stimulation results in the generation of highly activated T cells, and when subsequent stimulation through their TCR induces a
Fas/FasL- or TNF-TNFR-mediated death (Nagata, 1997; Zheng et al., 1995). Although
AICD is not the mechanism responsible for the silencing of the immune response, it seems to correspond with a state of immune deficiency, where the T cells fail to respond to other normally mitogenic stimuli. Following the apoptosis phase of the acute LCMV response, there is a precipitous decline in lymphocyte number. This phase is known as the immune silencing phase. During this phase there is a high level of apoptosis as detected by loss of DNA content in cells and TUNEL, an assay that measures DNA fragmentation
(Razvi et al., 1995; Lohman et al., 1996; Christensen et al., 1996). This apoptosis peaks at
11 days post-infection and correlates with the loss of T cells and B cells observed during this silencing phase at the end of the immune response. Despite this silencing, it is during this phase of the immune response that neutralizing antibodies to LCMV can first be detected (Battegay et al., 1993).
Memory, or the ability to respond more rapidly during a secondary response, is one of the hallmarks of the mammalian immune system. Memory to LCMV infection has been very well characterized. Acute infection with LCMV in normal mice results in the development of a long-lasting T cell immunity (Jamieson and Ahmed, 1989). According to limiting dilution assays, MHC tetramer staining, ELISPOT analysis, and TCR spectratyping, memory CTL frequencies remain high after LCMV infection and stably remain high for the lifetime of the animal (Selin et al., 1996; Altman et al., 1996; Murali- 11
Krishna et al., 1998; Lin and Welsh, 1998). A similar stability of CD4 T cells has also been reported (Varga and Welsh, 1998b). B cell memory has also been characterized following acute LCMV infection of mice. Many of these studies used LCMV-specific
ELISPOT assays where LCMV-specific Ab-producing cells could be enumerated by their ability to react with LCMV antigens. These analyses revealed that the bone marrow was the major site in which memory B cells resided and, through limiting dilution analysis, that the memory B cell precursor frequency is 1 in 20,000 spleen cells(Slifka et al., 1995;
Slifka and Ahmed, 1996a). Further studies of memory B cell formation following acute
LCMV infection in mice revealed a critical role for CD40 signaling (Whitmire et al.,
1996). LCMV-infected mice deficient in CD40L had skewed production of Ab isotypes compared to LCMV-infected wild type (WT) mice and were completely deficient in memory B cell generation.
This section was meant to provide a broad overview of the LCMV model of virus infection. In the following sections I will explore in further detail different aspects of the mammalian immune system and describe specific aspects of the immune response to
LCMV as they relate to my thesis work.
B. T cell subsets. T cells are lymphocytes associated with the adaptive immune system and generate cell-mediated immunity. They are categorized by the type of T cell receptor (TCR) they express and also by their coreceptors, either CD4 or CD8. TCRs are generated by germline gene rearrangements in three gene segments, the variable (V), diversity (D), and joining (J) segments. These VDJ rearrangements occur during T cell 12
development. The TCR exists as a heterodimer with a heavy and a light chain. Heavy
chains have all three V, D, and J segments whereas light chains have only the V and J
segments (Fields and Mariuzza, 1996). There are two types of light chain, alpha and
gamma, and two types of heavy chain, beta and delta. Gamma-delta T cells are more
fixed in their TCR gene rearrangement and have been considered to overlap the innate
and adaptive immune responses (O'Brien et al., 2007; Born et al., 2007). Their importance in immune responses has not yet fully been appreciated, but studies in
vaccinia virus-infected mice showed that gamma-delta T cells can have a significant
impact on early virus replication, with a corresponding increase in their ability to be cytotoxic and in their ability to secrete IFN-γ (Selin et al., 2001). Alpha-beta T cells comprise the major population of T cells in the immune system. They recognize antigenic peptides that have been processed from larger proteins and presented on major
histocompatibility complex (MHC) molecules (Zinkernagel and Doherty, 1974). T cells
with the CD4 co-receptor recognize MHC class II molecules that generally present peptides that have been processed from exogenous proteins; that is, proteins from outside the cell. CD8 co-receptor-expressing T cells recognize MHC class I molecules that generally present peptides from endogenous proteins(Germain, 1994). However, MHC class I proteins can present exogenous protein antigens via a process known as cross- presentation (Bevan, 1976; Rock and Shen, 2005). MHC class I molecules are presented by all somatic cells whereas MHC class II molecules are generally expressed by specialized cells of the immune system known as antigen presenting cells (APC)
(Germain, 1994). Once a T cell recognizes an MHC-peptide complex, the TCRs on the 13
surface of the T cell aggregate. At this point, other molecular interactions between co-
stimulatory molecules occur, and a complex signaling cascade leads to the activation of
the T cell(Bromley et al., 2001). Both TCR signaling and co-stimulation are required for
T cell activation(Mondino and Jenkins, 1994). T cell activation is characterized by a set
of clonal expansions and programmed differentiation where the T cells become effector
cells. Effector T cells are defined by the cytokines they express and the functions they
perform.
CD8 T cells are thought to be involved in what is known as a Th1-type immune
response. CD8 T cells that become effector cells following activation are generally
cytotoxic and can produce some Th1-type cytokines, including cytopathic antiviral
cytokines such as TNF and IFN-γ. Some studies have also suggested that CD8 T cells can
also express Th2-type cytokines such as IL-4(Mosmann et al., 1995). CD8 T cells are
also able to produce the cytokine IL-2, which promotes their own survival (Kasaian and
Biron, 1989; Su et al., 1998). Cytotoxicity by CD8 T cells is mediated by their ability to
secrete lytic granules that contain granzymes and perforin. Besides granule exocytosis
and cytokine secretion, CD8 T cells can also lyse target cells by providing ligands for
death receptors such as Fas.
CD4 T cells have been categorized into a few major subsets. CD4 T cells are
often referred to as helper T cells, because the cytokines produced by these T cells help
direct an immune response toward fighting intracellular or extracellular pathogens, or
they help prevent tissue damage by suppressing other effector cells in tissues locally.
CD4 T cells, upon initial encounter with their MHC II-bound antigen, will secrete IL-2, 14
upregulate IL-2R, and divide. These CD4 cells are termed Th0 cells and are cells that can
differentiate in to any of a number of effector cell types, depending on the cytokines in
the local environment(Sad and Mosmann, 1994).
CD4 T cell immune responses vary and have been categorized by the cytokines
produced during the response. These cytokines play an important role in directing the
immune response. Th1 and Th2 cytokine responses are probably the best studied
responses. For many years, all CD4 T cell responses were lumped into either Th1 or
Th2(Mosmann et al., 1986). However, as more cell types and cytokines were discovered, this Th1/Th2 paradigm has shifted to include other Th responses and, as time goes on, more responses may be described. Cytokines produced by Th1 CD4 T cells are IFN-γ,
TNF, and IL-12(Mosmann et al., 1986; Mosmann and Coffman, 1989). Cytokine receptor signaling, particularly IFN-γ receptor, within the CD4 T cell results in the upregulation of
T-bet, a transcription factor responsible for Th1 development. Cytokines produced by
Th2 CD4 T cells are IL-4, IL-5, and IL-13(Mosmann et al., 1986; Mosmann and
Coffman, 1989; Mosmann and Moore, 1991). IL-4 receptor signaling in CD4 T cells results in expression of GATA-3, a transcription factor that is responsible for Th2 cell development. Interestingly, T-bet and GATA-3 suppress each other’s expression to further promote either a Th1 or Th2 response, respectively(Grogan and Locksley, 2002).
But beside these effects, the cytokines produced by helper CD4 T cells have other functions which change the nature of the immune response. For instance, these cytokines have a major effect on B cell Ab class switching(Stavnezer, 1996). 15
Other CD4 T cells are known for their ability to suppress or regulate immune
responses. Regulatory T cells suppress immune responses locally(Bluestone and Abbas,
2003). They can produce IL-10 and TGF-beta, two cytokines that prevent activation of
other T cells. Th17 cells are another CD4 T cell subset that has been recently described.
These cells are characterized by their secretion of IL-17 and are hypothesized to be
important in the response to extracellular bacteria. This hypothesis is based on the fact
that secretion of IL-17, IL-22, and IL-23 by Th17 cells recruits neutrophils, promotes
epithelial cell growth and differentiation, and forms tighter junctions between epithelial
cells, all which combat bacterial growth. However, these T cells have gained further notoriety due to their association with autoimmune or inflammatory conditions, such as arthritis, inflammatory bowel disease, psoriasis, and multiple sclerosis (Reiner, 2007).
Classifying functional subsets of CD4 T cells is becoming more and more difficult as
CD4 T cell research continues. Cytokines produced by one subset are overlapping with the cytokines produced by other subsets. Thus, these categories (Th0, Th1, Th2, T reg, and Th17) are used as a guide and will more than likely be refined as more knowledge is gained and more subsets are defined. This thesis will describe another function of CD4 T cells, cytotoxicity, which had previously only been associated with CD8 T cells in vivo.
Cytotoxic CD4 T cells add another layer to the complex and varied nature of the CD4 T cell immune response.
C. Cytotoxic CD4 T cells. The ability of “killer” CD8+ T cells to lyse virus- infected cells has been well characterized in vitro and in vivo, but the “helper” CD4+ T 16
cell subset is usually associated with the production of cytokines which provide
secondary activation or survival signals to APCs, B cells, and CD8 T cells. Some twenty
years ago, however, CD4 T cells cultivated in vitro were found to have cytolytic potential
and to lyse targets in an MHC class II-dependent manner (Jacobson et al., 1984;
Lukacher et al., 1985; Maimone et al., 1986; Hou et al., 1993). Cytotoxic CD4 T cell
clones were generated, for example, against influenza virus in the mouse and against
measles virus, HIV, and Epstein Barr virus in humans. For a period of time it was
uncertain whether CD4 T cell cytotoxicity was an artifact of long term in vitro culture
(Bourgault et al., 1989), but eventually CD4 killer cell activity was found immediately ex vivo in cells that had been activated in vivo with keyhole limpet hemocyanin (KLH) (Erb et al., 1990) or during HIV infection (Appay et al., 2002). In contrast to the killing mediated by CD8 T cells, however, CD4 T cell-mediated killing had not been convincingly shown to occur in vivo.
In vitro cytotoxicity studies have suggested that CD4 killer T cells, when present, may employ similar mediators of cytotoxicity as CD8+ CTL, using TNF-α- (Tite, 1990),
Fas ligand- (FasL-) (Stalder et al., 1994), and perforin-dependent mechanisms (Williams and Engelhard, 1996). Initially it was thought that CD4 T cell killing was mediated exclusively by TNF-α or FasL and that perforin was only expressed in CD8 T cells and
NK cells. However, CD4 T cell clones were eventually shown to sometimes express perforin (Williams and Engelhard, 1997), and ex vivo studies from HIV-infected patients showed that CD4 T cell-mediated cytotoxicity could occur through a granule exocytosis mechanism involving perforin (Appay et al., 2002). Other mechanisms may also be 17
involved. Two studies reported the involvement of TRAIL in the killing of tumor cells by cytotoxic CD4 T cells (Thomas and Hersey, 1998; Kayagaki et al., 1999). It is also notable that CD4 killer T cell activity in vitro has been described as a potential function of CD4+CD25+ regulatory T cells (Janssens et al., 2003).
Studies by Lukacher, et al. showed that, after adoptively transferring an influenza- specific CD4 T cell clone into mice infected with two different strains of influenza virus, only the strain of virus recognized by the clone was cleared (Lukacher et al., 1985). This was taken as evidence that the antiviral effect of the clone may not involve release of soluble mediators but instead was due to exquisite recognition of cells expressing the cognate ligand, consistent with a direct cytotoxicity mechanism. What remains, however, is a clear demonstration of antigen-dependent, class II-dependent CD4 T cell-mediated killing in vivo.
D. The CD4 T cell response to LCMV As described in section B, LCMV infection of mice is a widely-used model system to study viral infections. The LCMV infection stimulates strong CD8 and CD4 T cell responses against a number of defined
MHC class I- (Whitton et al., 1988; van der Most et al., 1998) and MHC class II-
(Oxenius et al., 1995; Varga and Welsh, 1998a) presented epitopes. CD4 T cell numbers increase over the course of an acute LCMV infection. As assessed by intracellular cytokine stain, the highest number of epitope-specific cytokine producing CD4 T cells can be detected at Day 9 of acute IP infection of mice with LCMV Armstrong (Varga and
Welsh, 1998a). LCMV encodes two known MHC class II-restricted epitopes in I-Ab- 18
expressing (C57BL/6) mice (Oxenius et al., 1995). The more dominant GP61 response can account for around 8 % of the CD4 T cells during an acute infection while the subdominant NP309 response can account for around 2% of CD4 T cells during acute
LCMV infection (Varga and Welsh, 1998a). Acute infection with LCMV Armstrong stimulates a strong predominantly Th1 type immune response when administered i.v. or i.p., and CD4 T cells can secrete IFN-γ, IL-2, and TNF in response to LCMV peptide stimulation (Whitmire et al., 1998; Varga and Welsh, 2000).
A transgenic CD4 T cell, called SMARTA, that is specific for the LCMV GP61
peptide in the context of I-Ab has been generated (Oxenius et al., 1998a). These
SMARTA cells help B cells produce isotype-switched antibodies against LCMV GP.
They were able to enhance peptide-pulsed target lysis mediated by CD8 T cells in vitro and exhibited low levels of MHC class II-restricted cytotoxicity in vitro. SMARTA Tg mice, which are deficient in CD8 T cells, could not control acute LCMV infection.
As described above, LCMV-specific CD4 T cell numbers peak at day 9 post- infection. This peak is followed by contraction and the emergence of a long-lived CD4 memory cell pool(Varga and Welsh, 1998b). LCMV memory CD4 T cells rely on IL-15 and IL-7 for their generation and survival (Purton et al., 2007; Lenz et al., 2004). More recent studies using adoptive transfers of SMARTA transgenic CD4 T cells show that the virus-specific CD4 T cell proliferation is nonlinear and CD4 T cells do not start dividing until at least two days after viral infection. This is followed by a burst of proliferation and an approximately 150-fold expansion. Cytokine expression from these SMARTA CD4 T cells changes over the course of infection where early after infection they primarily 19
produce IFN-γ only and at later time points they begin to secrete both TNF and IFN-γ.
The level of IFN-γ expression by these CD4 T cells also increases over the course of the
infection, and T cells from later time points required 6-fold less peptide for the
stimulation of IFN-γ production, suggesting an increase in functional avidity. Using a
limiting dilution of these SMARTA T cells in adoptive transfers followed by intracellular cytokine stain allowed for the enumeration of the naive precursor frequency of a virus-
specific CD4 T cell, which was determined to be 1 in 105 cells (Whitmire et al., 2006).
CD4 T cells are not necessary for clearance of acute LCMV infection
(Moskophidis et al., 1987; Ahmed et al., 1988). However, the absence of CD4 T cells resulted in a slight 2-fold reduction in the number of CD8 effector T cells. In contrast,
CD4 T cells are necessary for viral clearance and for the generation of an effective CD8
T cell response when higher doses of virus or when more disseminating strains of LCMV are used. Battegay et al. showed that infection of CD4-deficient mice with low doses of
LCMV WE resulted in normal viral clearance and normal CD8 T cell induction.
However, infection of CD4 knock-out (KO) mice with higher doses of WE or infection
with the more quickly replicating Docile strain resulted in viral persistence and the
absence of a protective CTL response(Battegay et al., 1994). Further examination of this
effect, using CD4 Ab-depleted mice infected with another highly disseminating strain of
the virus, Clone 13, basically repeated this result(Matloubian et al., 1994).
A type of clonal exhaustion of CD4 T cells has also been reported following chronic LCMV infection. In mice infected with Docile strain of LCMV, CD4 T cells lost
the ability to stimulate B cell Ab production 6 weeks after infection. In these same 20
studies, CD4 transgenic SMARTA T cells were also analyzed for loss of function after adoptive transfer into Docile-infected mice. SMARTA cells were rendered unresponsive in a similar time period and lost their ability to upregulate activation markers and their ability to secrete IL-2 (Oxenius et al., 1998b). Along these lines, CD4 T cells from mice challenged with clone 13 had a reduced capacity to secrete IFN-γ in response to peptide stimulation as compared to Armstrong-infected mice(Varga and Welsh, 2000). Perforin
KO mice infected with the Armstrong strain develop a chronic infection resulting from the limited ability of perforin KO CD8 T cells to kill virus-infected cells. CD4 T cells in these mice also lost their ability to secrete IL-2, but not IFN-γ or TNF(Fuller and Zajac,
2003).
CD4 T cells, in addition to their role in sustaining CD8 T cell responses during chronic LCMV infection, also seem to play a role in the priming of the CD8 T cell response and the formation of good CD8 T cell memory to LCMV. A closer examination of the frequency of memory CD8 T cells in CD4-deficient mice following acute LCMV infection revealed that these frequencies were much lower than frequencies in WT mice
(von Herrath et al., 1996). This reduction in frequency was of functional consequence when mice were immunized with LCMV-protein-expressing vaccinia virus recombinants and subsequently rechallenged i.c. with LCMV. Mice deficient in CD4 T cells were not protected from i.c. rechallenge. These results suggested that CD4 T cells were required
for the generation of a protective CD8 memory response. A second study looked
specifically at the ability of these “helpless” CD8 T cells to divide and secrete cytokines
in response to secondary challenge. LCMV-specific CD8 T cells primed in the absence of 21
CD4 T cells had a reduced ability to expand in response to secondary challenge (Janssen et al., 2003). CD4 T cells were later shown to be required during CD8 T cell priming, at the beginning of the response, and were dispensable during the actual memory response.
This study used an adoptive transfer system where CD8 T cells were taken from CD4- deficient and sufficient mice that were immunized with LCMV or with LCMV-protein- expressing vaccinia virus recombinants. These CD8 T cells were transferred into CD4- deficient hosts and challenged with various LCMV-protein-expressing recombinant microorganisms. CD8 T cells from CD4-deficient mice mounted a defective response against the challenge with less protection and lower amount of cell expansion (Shedlock and Shen, 2003). Further examination of the phenotype of memory CD8 T cells that develop in CD4-deficient mice revealed that these CD8 T cells had lower expression of
CD62L, CD44, CD122 and CD127 than memory T cells generated in WT animals. These
T cells were also less able to produce IL-2 and TNF upon stimulation. This phenotype is more consistent with the phenotype of a naïve T cell as opposed to a memory-phenotype cell (Khanolkar et al., 2004; Fuller et al., 2005).
E. B cells and their activation. B cells are the lymphocytes that comprise the humoral immune system. Different B cell subsets have been associated with both adaptive and innate immune responses. Unlike the TCR, the B cell receptor (BCR) can be secreted, and the secreted BCR, known as antibody (Ab), is the product of a B cell response. B cells can be categorized in two distinct manners: first, by their location in lymphatic and peripheral tissues and, second, by their antigen receptors and co-receptors. 22
The major B cell compartments in the spleen and lymph nodes are comprised of
follicular B cells (Martin and Kearney, 2000). These are the conventional B cells that are
found in all lymphoid organs in an anatomical location known as the B cell or lymphoid
follicle. These cells align opposite T cells in the lymphoid follicle and their activation is
T dependent (TD), i.e. their activation is dependent on T cell help. Besides anatomical
location, they are differentiated from other B cells in lymphatic organs by their high expression of CD21, a complement receptor, and of CD23, a low affinity Fc epsilon receptor. Activation of these cells results in the formation of a germinal center (GC) response which can take several days to occur (MacLennan, 1994a). A second major
subset of B cells is marginal zone (MZ) B cells (Pillai et al., 2005). These surround the
lymphoid follicle in the spleen and filter antigens that enter lymphoid tissues via the
bloodstream, providing a first line of defense against pathogens. These cells are T
independent (TI), as they do not rely on T cell help for their activation and can be rapidly
activated during the early immune response. MZ B cells are also differentiated from other
B cell subsets by lower expression of CD23 and high expression of CD21. Both follicular
B cells and MZ B cells develop in the bone marrow and migrate as mature cells into the
spleen and lymph nodes. These cells may have a common precursor. Another major B
cell population is B1 B cells (Martin and Kearney, 2001). B1 B cells generate TI immune
responses and are located in the peritoneal and pleural cavities. Their developmental
origins remain controversial, but B1 B cells are formed during development from fetal
liver and may have an origin separate from other B cells. What is clear is that B1 BCRs
are somewhat restricted and recognize conserved epitopes shared between commensal 23
and parasitic organisms in the gut. The conserved nature of their BCR allows B1 B cells
to give rise to natural or innate Ab. This natural Ab could be considered the very first
form of humoral immunity to a pathogen because it exists before infections actually
occur and has been shown to reduce infection levels of viruses and bacteria (Ochsenbein
et al., 1999a).
Activation of B cells follows a similar mechanism as T cell activation. B cells
bind antigens by the BCR. BCR aggregation and crosslinking lead to intracellular
signaling events that cause a programmed differentiation and clonal expansion of the B
cell. B cell differentiation can be mediated by CD4 T cell help via costimulation and
cytokine secretion. The type of help received dictates the type of differentiation that occurs, meaning that Th1 T cell help will result in one type of B cell response and Th2 T cell helps will result in a different B cell response (Sher and Coffman, 1992). Following activation, B cells have two major fates. The first fate allows a B cell to secrete its BCR and become a plasma cell. Plasma cells do not divide and are thus at the end stage of their
differentiation. The second fate is to become a memory B cell. Memory B cells are
bipotent and upon restimulation can give rise to more memory cells or give rise to newly
formed plasma cells.
Part of the B cell differentiation process allows B cells to secrete their BCR.
Secreted BCR or Ab is a shortened form of the BCR with the transmembrane domain of
the receptor removed and a secretory signal sequence exposed. Antibody, except when it
is in the form of secreted dimeric IgA or pentameric IgM, is a homodimeric protein
connected by disulfide bonds. Each half of the homodimer has a larger heavy chain and a 24
smaller light chain connected to each other by dilsulfide bonds(Alzari et al., 1988).
Secreted Ab makes its way throughout the body via the circulatory system and lymphatics. Upon binding to a pathogen, Ab has several of its own effector functions.
First, it can bind to and directly inactivate pathogens by blocking entry or attachment receptors. Second, it can bind to pathogens and facilitate complement-mediated lysis.
Third, Ab can lead to sequestration of the pathogen through more efficient opsonization,
as scavenger cells such as macrophages have receptors that detect Ab. Fourth, other cells
such as mast cells also have Ab-binding receptors that bind the constant (Fc) region of
secreted Ab and these Fc receptors, if crosslinked by antigen, can initiate intracellular
signaling and cellular activation.
B cells have the capability to undergo two rounds of germline gene
rearrangement. The first round of rearrangement occurs during development and results
in the generation of the BCR. The second round occurs following B cell activation and
results in the class switching of the BCR. A third change in BCR genes can occur via
point mutations following activation and results in a higher affinity BCR (Li et al., 2004).
Like the TCR, the BCR subunits also derive from VDJ recombination events that occur
during development. The heavy and light chains of the Ab must undergo successful VDJ
recombination for B cell maturation to be completed. The larger heavy chains are
rearranged first and, if recombination is successful, the heavy chain proteins are paired
with temporary surrogate light chain proteins while light chain gene rearrangement
occurs. Two light chain genes are encoded in both mice and humans. Gene rearrangement
tends to occur in the kappa light chain first and, if this rearrangement is unsuccessful, the 25
lamba light chain gene is rearranged. Successful rearrangement of the light chain and
subsequent protein expression without self-reactivity signifies the end of naïve B cell
maturation. These naïve effector B cells are released from the bone marrow to encounter
antigen (Meffre et al., 2000).
In addition to heavy and light chains, descriptions of Ab structure can also be
made by referring to their antigen-binding or variable region and by their constant (Fc)
regions. The variable region is formed by germ line gene rearrangements in the V region
during VDJ recombination. B cells further differentiate through germline gene
rearrangements in their Fc region in a process known as class switch recombination, a
highly regulated process that involves the looping out of gene segments in order to
produce phenotypically different antibodies (Stavnezer, 1996). This process has been
extensively studied in mice. B cells develop in the bone marrow expressing only IgM. As
B cells mature, they express IgD as well, but eventually lose IgD expression following
activation (Monroe et al., 1983; Monroe and Cambier, 1982). Depending on the type of
help they receive, B cells then switch to expression of IgG1, IgG2a, IgG2b, IgG3, IgA, or
IgE isotypes in mice. IgM is the first Ab secreted by B cells and is thus the first Ab
detected during an immune response. Secreted IgM (sIgM) can exist as a monomer or as
a pentamer. Many viral infections that induce Th1 immune responses elicit IgG2a and
IgG3 production by B cells via IFN-γ R signaling. Th2 responses produce IgG1, IgA and
IgE responses, which depend on signaling from IL-4, TGF-beta, IL-5, and IL-13
(Stavnezer, 1996). 26
A third change in B cell germline gene sequence occurs during an active immune
response in distinct lymphoid compartments called germinal centers (GC). Here, B cells
undergo clonal expansion and somatic hypermutation (SHM), where the BCR V region
undergoes point mutations in hope of forming higher affinity interactions with antigen, a
process called affinity maturation (MacLennan, 1994b). Antigen in the GC is presented to
B cells by follicular dendritic cells (FDC), which can hold antigen in immune complexes
on their surface for long periods of time. Besides BCR signaling, the GC reaction is
dependent on T cells which align with B cells within the GC. B cells in the GC reaction
have phenotypically higher expression of MHC class II and can present antigens taken
from FDC to CD4 T cells that are induced to provide CD40L. CD40-CD40L interactions
are critical for GC reactions, and mice lacking CD40 also lack GC (Van den Eertwegh et
al., 1993). Other TNF family members such as TACI, BR3, and BCMA and their ligands
BLyS and APRIL also seem to play some role in the GC reaction (Miller et al., 2006).
For instance, mice treated with TACI-Fc blocked BLyS stimulation and eliminated the
GC reaction in the spleen (Yan et al., 2000). B cells in the GC can produce a more potent
BCR as a result of SHM, but they also can produce a non-functional or a self-reactive
BCR. Because of the nature of SHM and the selection of B cells based on their ability to
compete for limiting amounts of antigen, high amounts of apoptosis of lower affinity B
cells occur in GC reactions (MacLennan, 1994a). The exact molecular mechanism for
this apoptosis is not fully defined but positive signaling in higher affinity clones seems to
induce the expression of anti-apoptotic molecules. One of these molecules, BCL-XL, is up-regulated in B cells by CD40 stimulation and BCR crosslinking (Choi et al., 1996). 27
Overexpression of BCL-XL reduced apoptosis in GC B cells and lowered the affinity of the resulting B cells (Takahashi et al., 1999). Taken together these results suggest that low affinity clones with insufficient BCR and CD40 signaling cannot up-regulate BCL-
XL and die in the GC. The Fas death receptor is up-regulated on all GC B cells, but it
does not play a role in GC-associated apoptosis (Smith et al., 1995). In addition to clonal
expansion, SHM, and affinity maturation; GC reactions ultimately induce differentiation
of B cells into memory or plasma cells.
Serum antibodies are secreted by activated B cells that have differentiated into
plasma cells. Some plasma cells are short-lived and contribute to clearance of the primary
viral infection, whereas others are long-lived and can produce neutralizing Ab for several
months or years following the primary infection (Manz et al., 1997; Slifka et al., 1998).
One would think that, except in the case of persistent infection or secondary challenge,
serum Ab would disappear after a few months, but this is not the case, as titers can be
maintained throughout a lifetime without obvious additional antigen stimulation (Manz et
al., 1998). Several mechanisms have been proposed to explain this observation. One is
that some plasma cells are long living and sufficient to maintain serum Ab (Slifka et al.,
1998). Another model suggests that persisting or cross-reactive antigen induces memory
cells to divide and differentiate into plasma cells throughout an organism’s lifetime
(Ochsenbein et al., 2000; Traggiai et al., 2003). I will propose an additional model based
on the observations of hypergammaglobulinemia associated with virus infections and on
the work presented in this thesis that describes a virus-induced polyclonal B cell
activation. 28
F. The B cell response to LCMV. The humoral immune response to LCMV is
delayed in comparison to the cell-mediated response. Antibody responses are generated
against all viral proteins but not all of these Ab are neutralizing(Battegay et al., 1993).
Antibodies that react with LCMV can be detected by day 7 following infection
(Moskophidis and Lehmann-Grube, 1984; Battegay et al., 1993). However, neutralizing
Ab to acute LCMV infection cannot be detected until 30 days following initial exposure
to the virus, long after virus has been cleared from the host (Battegay et al., 1993; Recher
et al., 2004). The non-neutralizing Ab can bind to antigenic determinants on both LCMV
NP and GP whereas neutralizing Ab is specifically reactive with LCMV GP1 (Battegay
et al., 1993; Bruns et al., 1983; Wright et al., 1989). LCMV neutralizing Ab prevents
binding of GP1 with its receptor, alpha dystroglycan(Cao et al., 1998). LCMV stimulates an IgM response followed by a class-switched IgG response, where IgG1, IgG2a, IgG2b and IgG3 reactive with LCMV are generated(Slifka et al., 1995). Secretory IgD, which is considered an oddity for most B cell responses, can also be detected as a result of acute
LCMV infection(Moskophidis et al., 1997). IgG2a is the predominant Ab produced that binds LCMV (Whitmire et al., 1996). CD4 T cell help is required for optimal class- switched B cell responses to LCMV, and this help seems to come from CD4 T cell provision of CD40L. CD40LKO mice mount a reduced IgG response to LCMV with a complete absence of IgG1 production (Whitmire et al., 1996). Following reinfection with
LCMV, memory B cells respond rapidly, and LCMV-specific ASC numbers in the spleen 29
peak at day 5 post-infection. These numbers are then dramatically reduced at day 15 in
the spleen, corresponding to an increase in ASC number in the BM (Slifka et al., 1995).
Neutralizing antibodies have been shown to be important in protection against
reinfection with LCMV and for protection against lethal LCMV infection, but the
humoral response to acute LCMV is not required for viral clearance, as seen in B cell-
deficient mice (Thomsen and Marker, 1988; Wright and Buchmeier, 1991) (Cerny et al.,
1988; Cerny et al., 1986). Other studies have shown that mice deficient in B cells
succumb to i.c. inoculation of LCMV Armstrong with the same kinetics as WT mice,
suggesting that LCMV-induced immune pathology is not altered by B cells (Planz et al.,
1997). However, mice deficient in B cells can develop a persistent LCMV infection more
readily than WT mice. CTL memory in B cell KO mice is not well maintained and mice can develop persistent infections. Experiments performed in genetically B cell-deficient
mice infected with the LCMV Traub strain i.v. showed that B cell KO mice had a delayed
but detectable CTL response early after viral infection. In these mice, virus levels
dropped but were still detectable early after infection. Eventually, however, viral titers
rose to high levels in the blood and lung 30 days post-infection, resulting in persistent
infection in the B cell KO mice, but not WT mice. Corresponding to this recrudescence
of virus was the loss of the CTL response (Thomsen et al., 1996). In agreement with this
study, these same genetically B cell-deficient mice, when infected with high dose
LCMV-WE or the rapidly growing strain Docile, developed a persistent infection and lost
virus-specific CTL. However, if B cell-deficient mice were first primed with low dose
LCMV-WE they were able to efficiently clear a high dose rechallenge, suggesting that 30
memory CTL can actually be generated and maintained in the absence of B cells during a low dose infection, but these CTL are lost if higher doses of virus are present(Brundler et al., 1996). B cells may function as APC for CD4 T cells which are also required for the maintenance of an effective CTL response. However, the defective CTL response and persistent viral infection in B cell KO mice may be the result of a reported reduction in the overall CTL number in LCMV-infected B cell KO mice at the peak of the response and an improper rescue of memory CTL during the attrition phase of the immune response to acute LCMV infection in B cell KO mice (Asano and Ahmed, 1996). Thus, in the absence of B cells, the balance between the number of CTL and viral replication may be tipped in favor of virus replication and CTL become susceptible to clonal exhaustion.
Consistent with this hypothesis, other studies on the exhaustion of CD8 T cells following adoptive transfer into congenital LCMV carrier mice have shown that pre-activated B cells can prevent CD8 T cell exhaustion in these mice and that the effect is most likely due to the production of neutralizing Ab, which lowers viral burden (Planz et al., 1997;
Hunziker et al., 2002). It still remains to be clarified exactly what B cells provide in these situations that prevents clonal exhaustion.
LCMV-specific plasma B cells home to the bone marrow, where they are maintained (Slifka et al., 1995). Using limiting dilution analysis, the precursor frequency of memory B cells in the spleen was determined to be about 1 in 20,000 splenocytes or about 1 in 10,000 B cells (Slifka and Ahmed, 1996a). LCMV memory B cells also reside in the bone marrow. This was concluded based on a virus-specific enzyme-linked immunospot assay (ELISPOT) to determine precursor frequency, which relies on a 31
limiting dilution analysis of ASC and is predicated on the fact that LCMV memory B cells will become Ab-secreting plasma cells after in vitro stimulation. Thus, the Ab- secreting daughter cells of LCMV-specific memory B cells are actually counted in the assay. The phenotype of mouse memory B cells is not identifiable by flow cytommetry, and there is not much known about the LCMV memory B cell response. There are no means to detect LCMV-specific B cells, and there have been no reports of fluorescently labeled virions or virus proteins that could be used to “see” virus-specific B cells. Much of what is known about the B cell response to LCMV was discovered by examining
LCMV-specific Ab with enzyme-linked immunosorbant assays (ELISA), ELISPOT, or virus neutralization assays. However, some recent reports on the B cell response to
LCMV have used B cell hybridomas to generate LCMV-specific B cell transgenic mice.
Two transgenic mice with B cells that generate neutralizing IgM antibodies against LCMV have been created. These mice were created by targeting the IgM heavy chain and kappa light chain genes from the LCMV-specific KL25 B cell hybridoma into mice. The resulting mice are called H25, which expresses the heavy chain of the KL25
IgM only, and HL25, which expresses both the heavy and light chains of the KL25 IgM.
H25 mice mounted a neutralizing Ab response 8 days after acute LCMV infection, and this improved the host’s ability to clear the virus. HL25 mice spontaneously produced neutralizing IgM that protected mice immediately after LCMV infection but did not generate a CTL response, presumably because viral antigen was cleared before it could be processed by APC (Seiler et al., 1998). HL25 B cell transgenic mice were also used to assay B cell tolerance induction in congenitally-infected LCMV carrier mice by crossing 32
the HL25 transgenic mice with congenitally-infected mice. Virus isolated from the resulting HL25 transgenic LCMV carrier mice was a neutralization resistant variant of
LCMV, and no evidence of B cell tolerance was observed in these mice (Seiler et al.,
1999). This study highlights how difficult it is to generate B cell tolerance to virus infections, because the transgenic carrier mice did not delete what should be considered a
“self reactive” anti-LCMV transgenic B cell response. Another LCMV-specific B cell transgenic mouse was generated by targeting only the variable and diversity regions of the heavy chain genes from the KL25 B cell hybridoma into mice(Hangartner et al.,
2003). The resulting mice, called TgH(KL25), have B cells that can generate a more dynamic neutralizing Ab response against LCMV and class switch from IgM to IgG.
These mice rapidly generate neutralizing Ab responses after the third day of infection.
TgH(K25) mice had reduced viral loads and generated a CTL response during acute
LCMV infection. Additionally, these mice avoided clonal exhaustion when challenged with a high dose of virus due to Ab-mediated reduction in viral load. Generation of a TD
IgG response was observed in TgH(K25) mice but interestingly, TgH(25) transgenic mice also generated a TI IgM response to LCMV when CD4 T cells were absent(Hangartner et al., 2003). This was an unexpected result because previous investigations of anti-LCMV responses performed in CD4-deficient mice did not detect a TI neutralizing Ab response
(Ahmed et al., 1988).
G. Hypergammaglobulinemia and virus-induced polyclonal activation of B cells. The adaptive immune system is designed to mount a rapid antigen-specific 33
response to pathogens by way of clonal expansions of T and B cells bearing discrete
antigen-specific receptors. For several decades, however, it has been noted that a number
of human viral infections, including those caused by Epstein Barr virus (EBV),
cytomegalovirus (CMV), human immunodeficiency virus (HIV), and hepatitis C virus
(HCV), seem to elicit Ab responses to non-viral antigens, including self-antigen(Hu et al.,
1986; Burgio and Monafo, 1983; Martinez-Maza et al., 1987; Hutt-Fletcher et al., 1983;
Rosa et al., 2005). In some cases, there is a demonstrable hypergammaglobulinemia and a
presumed but mostly undefined polyclonal activation of B cells. During the course of my
thesis work, autoantibody production associated with this phenomenon was observed in mice infected with LCMV, vaccinia virus, and vesicular stomatitis virus (VSV), where
autoantibody production mirrored some of the autoantibody specificities detected in human infections (Hunziker et al., 2003; Ludewig et al., 2004). This presumed polyclonal
B cell activation associated with virus infection is of great medical interest because it
may be involved in the initiation of autoimmunity or contribute to the long-term maintenance of B cell memory.
Clonal B cell activation requires a signal though a specific antigen receptor (BCR)
and costimulation by a secondary receptor. These two signals result in the specific
activation of one B cell. Polyclonal B cell activation is not specific and can occur through
many proposed mechanisms, including: signals generated by innate pattern recognition
molecules, such as Toll-like receptors (TLR); mitogenic cytokine signals, such as those
provided by IL-4; non-specific T cell help, such as the provision of costimulation ligands;
cell transformation, which occurs following EBV infection; and superantigen signals, that 34
occur following binding of staphylococcal protein A to the BCR. All of these can lead to
the polyclonal activation of B cells. It is not clear whether the two signal model of
activation is required for polyclonal B cell activation, but constant with all of these
polyclonal activators is global B cell activation and the absence of specific BCR
triggering, meaning that both naïve- and memory-phenotype cells could potentially be
activated.
Polyclonal activation of B cells without BCR triggering has been reported in
vitro with human memory B cells stimulated by TLR agonists or with CD4 T cell
help(Bernasconi et al., 2002). In these studies, B cells were isolated from blood and
separated based on CD27 expression into naïve-phenotype (CD27-) and memory- phenotype (CD27+) B cells. The memory-phenotype B cells were also separated into IgG
and IgM receptor-expressing cells. CD27 expression on human B cells has been shown to
correlate with previous antigen encounter and subsequent somatic mutation (Klein et al.,
1998). Once separated, the B cells were stimulated with either TLR agonists or by alloreactive T cells that could provide CD40L and cytokines to the B cells subsets. These stimulations can be called polyclonal stimuli because they are not specific for any one B cell. Despite a presumed requirement for BCR triggering, the memory B cells responded to the TLR stimulation and to the surrogate T cell help without Ab receptor triggering.
The B cells divided, changed their surface phenotype, and secreted Ab in response to the polyclonal stimuli. Of note, IgG-expressing B cells were more responsive to these stimuli than IgM-expressing cells. These findings led the investigators to believe that there were conditions under which B cells could become activated in the absence of BCR 35
stimulation. They proposed that polyclonal activation of memory B cells may account for the maintenance of memory B cells and the maintenance of serum Ab levels throughout an organism’s lifetime (Bernasconi et al., 2002; Traggiai et al., 2003).
The biological significance of polyclonal B cell activation can be speculated to be beneficial or detrimental. The above study illustrates that the outcome of polyclonal B cell activation can be beneficial and potentially lead to sustained B cell memory.
However, because polyclonal B cell activation is non-specific, auto-reactive B cell clones may also be maintained by polyclonal activation, a potentially harmful effect. However,
B cell maintenance may not be the only biological significance for polyclonal B cell activation.
There are many ways to polyclonally activate B cells, and many infectious agents have been presumed to induce polyclonal B cell activation because these infections have been associated with the induction of hypergammaglobulinemia.
Hypergammaglobulinemia may be caused by polyclonal B cell activation but, for the most part, these B cells have not been identified. Some have proposed that this hypergammaglobulinemia is meant to increase the levels of natural Ab that can eliminate some of the early spread of infectious agents (Ochsenbein et al., 1999a). Natural antibodies are germ-line-encoded antigen recognition molecules that bind conserved patterns in many pathogens. Hypergammaglobulinemia induced by polyclonal B cell activation, because it is non-specific, could lead to increased natural Ab production. It has also been proposed that polyclonal activation, because it activates all B cells, may induce the production of some pathogen-specific antibodies, although at low frequency. 36
However, these low frequencies may be relevant early in infection when the infectious particles are low in number(Ochsenbein et al., 1999b). Elevations in natural Ab and elevations in pathogen-specific antibodies are further examples of how polyclonal B cell activation could be beneficial, but because many of the antibodies produced by polyclonal B cell activation would not be pathogen-specific, some infectious agents may have evolved to induce polyclonal B cell activation as a means to dilute the specific Ab response, thus allowing more time for pathogen growth in the host (Hunziker et al., 2003;
Minoprio et al., 1988).
Several microorganisms, including viruses, bacteria, and protozoans can induce hypergammaglobulinemia. The mechanisms used to induce hypergammaglobulinemia are shared between different classes of microorganism. For instance, different pattern recognition molecules can be triggered by viruses, bacteria, and protozoans. Due to the complex nature of polyclonal B cell activation and the many pathogens that induce hypergammaglobulinemia, for the remainder of this section, I will focus primarily on virus-induced hypergammaglobulinemia and describe virus infections where, for the most part, polyclonal B cell activation has been presumed, but not clearly documented.
Some viruses such as EBV directly infect B cells. In these cases, activation of the infected B cell may be due to mitogenic signals generated during the infection process or due to the transformation of B cells by the virus. EBV infection can transform B cells and can cause B cell lymphomas(Kuppers, 2003). EBV does not seem to selectively infect certain B cells, and this transformation can thus be considered a type of polyclonal activation. Transformation may not be the only way EBV can polyclonally activate 37
infected B cells. In vitro infection of B cells with the P3HR-1 strain of EBV, a strain that cannot transform B cells, showed that these infected B cells would readily down-regulate
IgD and elevate thymidine incorporation more than uninfected B cells if they were exposed to supernatants from in vitro cultures of activated T cells (Hu et al., 1986).
Division, as measured by thymidine incorporation, and down-regulation of IgD are consistent with events that would occur with other polyclonal activators. EBV also encodes a protein LMP1 which acts as a constitutively active CD40 signal in infected B cells and causes those B cells to proliferate (Kilger et al., 1998). Although it has not formally been tested, LMP1 may have caused the polyclonal B cell activation induced by the P3HR-1 EBV strain. Not only does direct EBV infection cause polyclonal B cell division but EBV infection of human lymphocytes leads to polyclonal Ab production
(Rosen et al., 1977). This polyclonal Ab production includes the production of
“heterophile antibody” which reacts with the red blood cells of other species(Paul and
Bunnell, 1974; Duverlie et al., 1989). The production and detection of this heterophile Ab was exploited by clinicians as a means to diagnose infection with EBV in a test known as a Paul-Bunnell antibody test (Burgio and Monafo, 1983). This test, or variations of it, is still used to diagnose EBV infection in humans. EBV is one of the only human viral infections where polyclonal B cell activation and hypergammaglobulinemia are both described. For other infections, hypergammaglobulinemia is observed, but polyclonal B cell activation per se has not been formally described.
Other viruses do not readily infect B cells, yet still evoke hypergammaglobulinemia in infected patients. Patients with CMV mononucleosis present 38
with unusually high numbers of Ab-secreting lymphocytes as well as increases in Ab irrelevant to viral antigens. Peripheral leukocytes from CMV seronegative donors produced Ab when exposed to both live and UV-irradiated CMV in the relatively short in vitro culture where the generation of a specific antiviral response is unlikely and, in this case, viral infection of B cells was not driving Ab production (Hutt-Fletcher et al., 1983).
The specific mechanism by which CMV activates human B cells remains unclear.
HCV is another virus that seems to induce polyclonal B cell activation in infected patients. For HCV, hypergammaglobulinemia is known as cryoglobulinemia.
Cryoglobulinemia is a high frequency of serum Ab that undergo a reversible precipitation when exposed to cold temperatures. The resulting cryoglobulinemia seems to stem from increased IgG and IgM levels that are generated by a naïve-phenotype (CD27-) population of B cells (Racanelli et al., 2006). HCV interaction with CD81 receptor on B cells might be the trigger that induces cryoglobulinemia. In vitro studies show that stimulation of CD81 on naïve-phenotype B cells leads to their activation (Rosa et al.,
2005). The HCV protein E2 binds CD81 and uses it as an entry receptor. Based on these observations, it was proposed that HCV E2 protein binding of CD81 could activate B cells and cause the cryglobulinemia. The nonspecific nature of the cryoglobilinema was speculated to be due to that fact that naïve-phenotype B cells express more CD81 than previously-activated B cells. CD81 stimulation of naïve B cells by HCV E2 would therefore effect non-HCV-specific B cells more than the activated HCV-specific B cells and generate the non-specific cryoglobulinemia. 39
Another example of human viral infection that is associated with
hypergammaglobulinemia is HIV. HIV may employ multiple mechanisms to induce
polyclonal B cell activation. Some of the earliest reports in AIDS research showed that
patients infected with HIV had a hypergammaglobulinemia and presumed polyclonal B
cell activation (Lane et al., 1983). Based on these clinical observations, researchers tested
the ability of HIV to stimulate B cells in vitro, similar to the studies performed with
CMV. Peripheral blood B cells exposed to the HIV virus in vitro became activated and
divided, as seen by Ab secretion and thymidine incorporation(Schnittman et al., 1986).
Others later found that B cells isolated from the peripheral blood of HIV positive
individuals were more likely to display activation markers and had elevated levels of spontaneous Ab-producing cells than uninfected individuals (Martinez-Maza et al.,
1987). These results showed that B cell exposure to HIV could result in polyclonal B cell activation in vitro but the mechanism of HIV-induced polyclonal B cell activation both in vitro and in vivo remained undefined.
HIV-associated hypergammaglobulinemia is seen throughout the course of an
HIV infection from initial virus exposure to the development of AIDS. Early in infection, hypergammaglobulinemia is hypothesized to be T cell-dependent because some studies found that HIV-infected cultures containing CD4 T cell clones could induce a high frequency of a polyclonal population of B cells to secrete Ab in vitro (Macchia et al., 1993; Chirmule et al., 1993). Several interactions were identified that govern this T
cell-dependent antibody secretion. Some claimed that TNF expression on infected CD4 T
cells was required. Antibody blockade of either TNF or TNF receptor by mAb treatment 40
eliminated Ab production by activated B cells (Macchia et al., 1993). In this study,
CD40L expression was not required on T cells. Other groups performed trans-well assays using both whole virus or the HIV protein gp160 and in these studies it was found that
Ab secretion in vitro was dependent on three things: direct contact between T and B cells, the expression of CD40L on T cells, and IL-6 receptor stimulation (Chirmule et al.,
1993). The discrepancies in these results remain unresolved and with the exception of
CD40L it is not clear that either study ruled out possible overlapping roles of TNF and
IL-6. These discrepancies may also be due to the fact that the experiments were done in vitro with T cell lines grown under different conditions. Nevertheless, CD4 T cells were found to be critical to induce this in vitro Ab secretion by HIV in both studies.
As mentioned above, hypergammaglobulinemia is seen both early and late during
HIV infection, when CD4 T cells are nearly absent. The hypergammaglobulinemia seen in AIDS patients who lack CD4 T cells may be driven by other mechanisms. Cytokine involvement, specifically from IL-15, has been associated with AIDS-related hypergammaglobulinemia. Clinically, AIDS patients experiencing hypergammaglobulinemia had elevated levels IL-15 (Kacani et al., 1997). It was hypothesized that the elevated levels of IL-15 may have some involvement in late-stage
HIV-induced hypergammaglobulinemia. Consistent with this hypothesis, exposure to heat-inactivated HIV along with IL-15 resulted in a dose-dependent increase of thymidine incorporation in human peripheral blood B cells in vitro (Kacani et al., 1999).
Based on these results, the hypergammaglobulinemia seen during the late stages of HIV infection may be due to IL-15 stimulation of B cells. More recently, gp120, a protein 41
from the envelope of HIV, was also shown to activate naïve B cells through a mannose binding receptor. Stimulation of this receptor induced DNA synthesis and Ab production by these B cells (He et al., 2006). This mannose binding receptor-mediated polyclonal activation may play a role in both the early HIV- and late AIDS-associated hypergammaglobulinemia.
In addition to clinical observations of hypergammaglobulinemia, several studies have noted this phenomenon in mouse models of viral infection. Similar to the observations made in humans, hypergammaglobulinemia is described but polyclonal B cell activation in vivo remains undefined except in cases where B cells are directly infected by the virus. Infection of mice with lactate dehydrogenase-elevating virus
(LDV) is one example of murine infection that induces hypergammaglobulinemia.
Expression of IgG2 isotypes were specifically elevated in LDV-infected mice experiencing hypergammaglobulinemia (Li et al., 1990). This Ab response could occur in the absence of a virus-specific response (Rowland et al., 1994). It was not strain dependent and even occurred in immunocompromised nude mice (Coutelier et al., 1990).
However, in contrast to the results in nude mice, LDV-induced hypergammaglobulinemia was found to be partially dependent on CD4 T cells as seen by in vivo depletion studies, where depletion of CD4 T cells in LDV-infected mice greatly reduced the levels of IgG production (Li et al., 1990). Expression of carbohydrates on the surface of the LDV envelope protein seemed to be the viral determinant that induced hypergammaglobulinemia (Plagemann et al., 2000). Viruses that lack two of these glycans had greatly reduced hypergammaglobulinemia. In some strains of mice, IL-6 was 42
found to be important for LDV-induced hypergammaglobulinemia. Mice lacking IL-6 had reduced hypergammaglobulinemia compared to WT mice following infection with
LDV, but IL-6 KO mice bred onto the common C57BL/6 strain did not have reduced levels of hypergammaglobulinemia (Markine-Goriaynoff et al., 2001). Thus, the dependence on IL-6 for LDV-induced hypergammaglobulinemia varied, based on the genetic background of the mouse.
LDV is just one of many viruses that cause hypergammaglobulinemia in mice.
Early studies included infection with LDV, LCMV, adenovirus, murine coronavirus
MHV-A59, and Sendai virus (Coutelier et al., 1988). Murine cytomegalovirus (MCMV) and murine gammaherpesvirus 68 (MHV68), two herpesviruses respectively used as mouse models of CMV and EBV in humans, were later also identified as viruses that could induce a hypergammaglobulinemia in mice (Price et al., 1993; Stevenson and
Doherty, 1998). With the exception of Sendai virus, which has not yet been tested, all of these infections in mice require CD4 T cells to induce hypergammaglobulinemia
(Coutelier et al., 1994; Stevenson and Doherty, 1999; Coutelier et al., 1990; Price et al.,
1993; Lardans et al., 1996). In each of these cases depletion of CD4 T cells resulted in a dramatic reduction in hypergammaglobulinemia. With the exception MHV68, where B cells can be directly activated by a viral protein or perhaps by viral infection, the specific mechanisms driving hypergammaglobulinemia in these other mouse models of infection remained unknown. Polyclonal B cell activation is presumed to be the mechanism, but, again, it has not been clearly demonstrated. 43
Several studies were undertaken using the LCMV model of mouse infection that
led to the presentation of an entirely different mechanism to explain the induction of
hypergammaglobulinemia. During the course of my thesis work, Hunziker et al.
performed seminal studies in which they investigated conditions under which LCMV
could induce hypergammaglobulinemia and the effects hypergammaglobulinemia had on
autoantibody induction (Hunziker et al., 2003). Also concurrent with my thesis, Recher et al. investgated the effects of hypergammaglobulinemia on the generation of virus-specific
neutralizing Ab (Recher et al., 2004). Initial experiments were performed by Hunziker et
al. using different strains of LCMV: Armstrong, Docile, Clone 13, and WE. While all
strains induced hypergammaglobulinemia at the same dose of infection in the same strain
of mice, only the less virulent Armstrong strain and the more virulent LCMV-WE strain
were cleared from infected mice and did not establish a persistent infection. Of the two
strains, LCMV-WE generated a longer lasting and greater level of
hypergammaglobulinemia and grew to higher titers in mice than Armstrong. This
correlation between viral load and hypergammaglobulinemia was noted, and was
confirmed by generating reassorted viruses, where the polymerase that controls viral
replication for LCMV was swapped between the two viruses. In this case, LCMV
Armstong with the WE viral replication gene induced the same level of
hypergammaglobulinemia and had the same viral load as the original LCMV-WE, and
LCMV-WE with the Armstrong viral replication gene resembled the
hypergammaglobulinemia and viral loads generated by the original Armstrong strain.
LCMV-WE was used in subsequent experiments because it induced higher levels 44
hypergammaglobulinemia than Armstrong, and Hunziker et al. proposed that use of a
persistent strain like Docile or Clone 13 may complicate experimental results due to the
potential effects of clonal exhaustion and chronic stimulation of LCMV-specific B cells
in persistently infected mice.
Hunziker et al. next infected several inbred mouse strains as well as several
strains of knock out mice with LCMV-WE to determine if there was anything specifically
required from virus-infected hosts that might induce hypergammaglobulinemia. Mice
lacking cytokine responses, such as type I IFN R, IL-6, and IL-12 as well as mice
defective in Th skewing cytokines, such as IL-4(Th2-inducing) and IFN-γ (Th1-inducing)
all experienced hypergammaglobulinemia following LCMV-WE infection. These results
indicated that these cytokines were not required for LCMV- induced
hypergammaglobulinemia. Furthermore, every strain of mouse infected experienced
hypergammaglobulinemia, suggesting that the phenomenon was not laboratory strain-
specific.
Previous studies on the hypergammaglobulinemia associated with adenovirus
infection found that up to 90% of the Ab produced did not bind viral particles (Coutelier
et al., 1988). To confirm that non-virus-specific antibodies could also be produced during the LCMV-induced hypergammaglobulinemia, Hunziker et al showed that quasimonoclonal mice, in which 80% of the B cells were all of the same known specificity and not LCMV-specific, had increased levels of the quasimonoclonal Ab species after infection with LCMV-WE. Additionally, WT mice infected with LCMV-
WE developed antibodies that reacted with other viruses such as VSV and influenza and 45
some mice developed self-reactive antibodies that bound thyroglobin, DNA, and insulin.
These additional results lend evidence to the potential biological significance of
polyclonal activation described earlier and show that non-virus-specific antibodies can be
produced as a result of LCMV-induced hypergammaglobulinemia.
CD4 T cells were found to be important for hypergammaglobulinemia induced by
other viruses. Hunziker et al found that CD4 T cells were also required for LCMV-
induced hypergammaglobulinemia because CD4 T cell KO mice did not experience
hypergammaglobulinemia following LCMV-WE infection. However, the role of CD4 T
cells was refined in these studies using mice deficient in CD40L, a key costimulatory
molecule for CD40 on B cells. CD40 is required for proper B cell activation, including
GC and memory formation(Foy et al., 1994). These CD40L KO mice also did not experience hypergammaglobulinemia following LCMV-WE infection. Hunziker et al.
next asked if any CD4 T cell could induce hypergammaglobulinemia or if virus-specific
CD4 T cells were required. To test this, they transferred virus-specific transgenic
SMARTA CD4 T cells into congenital LCMV carrier mice and into mice that
ubiquitously express LCMV GP and assayed for hypergammaglobulinemia. In both
cases, the SMARTA T cell transfer resulted in hypergammaglobulinemia. However,
when a non-LCMV-specific transgenic T cells was transferred, no Ab production was
seen. These results refined the CD4 T cell requirement that had been previously proposed
by others. For LCMV-induced hypergammaglobulinemia, CD40L and virus-specific CD4
T cells are required. 46
Thus far, Hunziker et al identified CD40L and virus-specific CD4 T cells as
requirements for LCMV-induced hypergammaglobulinemia and noted a correlation between high viral loads and hypergammglobulinemia. Non-virus-specific Ab production
had also been demonstrated. These observations, when taken together, led Hunziker et al. to propose that in situations where there is high antigen load, non-virus-specific B cells must be able to present viral antigens to virus-specific CD4 T cells which induces Ab production and hypergammaglobulinemia. To show that non-LCMV-specific B cells had the ability to present LCMV antigens, they isolated B cells from the LCMV-WE-infected quasimonoclonal mice and exposed them to a T cell hybridoma that was specific for
LCMV in vitro. Quasimonoclonal B cells from LCMV-WE-infected mice but not from uninfected mice stimulated the T cell hybridoma. These results led Hunziker et al. to conclude that non-LCMV-specific B cells could present viral antigens to CD4 T cells.
If the BCR is not used to internalize viral antigens, B cells may use complement
or Fc receptors to endocytose virus or viral antigens. Alternatively, viral antigen may be
internalized by general pinocytosis. To investigate these possibilities, mice missing C3 (a
ligand for CD23), the complement receptor CR2, CD19, Fc gamma receptor, and IgM
(mice that do not express IgM but do express IgD and can switch to IgG) were infected
with LCMV-WE by Hunziker et al. and assayed for hypergammaglobulinemia. In every
case, the LCMV-WE-infected knock out mice still experienced
hypergammaglobulinemia. These results suggested that none of the traditional antigen
uptake mechanisms used by B cells could account for how B cells presented antigen to T
cells. Hunziker et al. concluded from these studies that generalized pinocytosis was most 47
likely the entry route for viral antigen. The final conclusions made by Hunziker et al were
that LCMV-induced hypergammaglobulinemia was the result of high in vivo antigen
loads and these high viral loads allowed non-LCMV-specific B cells to present LCMV
antigens to LCMV-specific CD4 T cells that activated those B cells by a CD40L- dependent mechanism.
Another study of LCMV-induced hypergammaglobulinemia, also performed concurrent with my thesis work, investigated the effect of hypergammaglobulinemia on the generation of a virus-specific neutralizing Ab response (Recher et al., 2004). Previous observations had been made that LCMV-infected mice lacking CD8 T cells had reduced hypergammaglobulinemia and an earlier onset of virus-specific neutralizing Ab (Battegay et al., 1993). The CD4 T cell requirement for hypergammaglobulinemia had already been established by Hunziker et al. and CD8 T cells were not required in those studies
(Hunziker et al., 2003). It was reasoned that, under normal circumstances, CD8 T cells may help control viral titers and prevent CD4 T cell anergy or unresponsiveness. Thus,
CD8 T cells, by controlling the viral loads, allowed for the generation of a good CD4 T cell response and subsequent hypergammaglobulinemia. Supporting this, Recher et al. found virus-specific CD4 T cell responses, as measured by IFN-γ production, were diminished in the CD8- depleted animals (Recher et al., 2004). Thus, the LCMV-specific
CD4 T cell response correlated with hypergammaglobulinemia but was inversely correlated with neutralizing Ab titer. To determine a direct causal relationship between
CD4 T cells and hypergammaglobulinemia, an adoptive transfer study was performed where transgenic LCMV-specific SMARTA CD4 T cells were transferred into LCMV- 48
infected CD8 KO mice. Mice that did not receive the CD4 T cell transfer did not
experience hypergammaglobulinemia and rapidly generated LCMV-specific neutralizing
Ab. Mice that received SMARTA cells had a robust hypergammaglobulinemia and did
not generate neutralizing Ab. Parallel experiments were performed using a virus lacking
GP61, the epitope recognized by SMARTA T cells. These mice also rapidly generated a
neutralizing Ab response and had moderate hypergammaglobulinemia. These results
show that virus-specific T cell help causes hypergammaglobulinemia but delays the formation of neutralizing antibodies. Recher et al. reasoned that during the course of an
LCMV infection, they could enhance the onset of neutralizing Ab production by reducing
the virus-specific CD4 T cell response. They reduced the CD4 T cell response by
performing a partial Ab-mediated depletion of CD4 T cells in vivo before infection of mice with LCMV. By this method, a reduction in the number of virus-specific CD4 T cells resulted in a reduction of hypergammaglobulinemia and the earlier onset of neutralizing Ab. Recher et al. finally concluded that specific CD4 T cell help induces hypergammaglobulinemia and delays a neutralizing Ab response. These results, in combination with the results of Hunziker et al., suggest that this delay may be due to distraction of CD4 T cells by non-virus-specific B cells presenting viral antigen.
These recent studies on virus-induced hypergammaglobulinemia in LCMV- infected mice have concluded that the hypergammaglobulinemia was dependent on virus- specific CD4 T cells and CD40L interactions, that it occurred more extensively in the presence of high antigen loads, that it required antigen presentation by B cells, and that it interfered with the generation of specific anti-viral neutralizing antibodies, perhaps 49
suggesting that T cells were distracted by antigen-presenting B cells of irrelevant BCR
specificities (Recher et al., 2004; Hunziker et al., 2003). These studies were done concurrently with my own work and focused primarily on Ab production without examining the process of B cell activation and differentiation. Thus, the dynamics and magnitude of polyclonal B cell activation and proliferation under conditions of viral infection remained unclear.
H. Thesis objectives. I wanted to examine some of the interactions that occur between T cell and B cells during viral infections. To accomplish this, I performed adoptive-transfer studies, where B cells were coated with MHC class II-presented viral antigens and transferred into virus-infected hosts. These studies allowed me to define an
MHC class II-restricted killing in vivo and polyclonal B cell activation during virus infection.
Hypergammaglobulinemia following viral infections has been described in humans and mice. This hypergammaglobulinemia seems to associate with viral infections that grow to high titer in infected hosts. In several instances, CD4 T cells were required to induce hypergammaglobulinemia. Studies done in LCMV-infected mice concurrent with my thesis have shown that hypergammaglobulinemia, in addition to its association with high viral loads, is dependent on CD4 T cells, CD40-CD40L interaction, and presentation of viral epitopes on B cells. The hypergammaglobulinemia is inversely correlated with the virus-specific neutralizing Ab response and it was suggested that the hypergammaglobulinemia is caused by B cells that are not virus-specific. These studies 50
focused primarily on Ab production without examining the process of polyclonal B cell activation and differentiation.
Conditions transpiring in infections with high antigen loads may allow B cells to present viral antigens in a BCR-independent manner and become activated. This has been proposed to be the cause of hypergammaglobulinemia but the ability of these viral antigen-presenting B cells to become activated in vivo has not been formally tested. For my thesis, I questioned the fate of a polyclonal population of B cells presenting viral epitopes after adoptive transfer into virus-infected hosts. My thesis work shows that this polyclonal population of B cells has two fates. Some are killed in vivo by CD4 T cells
(CHAPTER III). This work is the first evidence of MHC class II-restricted killing in vivo during viral infection. However, B cells that survive the cytotoxicity undergo a dynamic polyclonal activation manifested by proliferation, changes in phenotype, and Ab production (CHAPTER IV). This work is a careful examination of the activation of the B cells that are responsible for virus-induced hypergammaglobulinemia.
51
CHAPTER II
MATERIALS AND METHODS
A. Mice. C57BL/6J, Tnfsf1atml-Mak (TNFR-1-/-p55-/-), B6.129S7-
Ifngr1
Tnfrsf6lpr (lpr), C57BL/6Ncr-Tnfrsf5tm1Kik (CD40-/-), and B6x129 TNF-α -/- and +/- mice were bought from the Jackson Laboratory (Bar Harbor, ME) and have been maintained in the animal colony at UMass Medical School. B6.Cg-IgHaThy-1aGpi-la/J (IgHa) mice,
which express an allotypic immunoglobulin marker, were purchased from the Jackson
Laboratory (Bar Harbor, ME). B6.SJL (Ly5.1) mice were purchased from Taconic Farms
(Germantown, NY). MyD88-/- mice(Adachi et al., 1998) were kindly given to me by Dr.
K. Alugupalli and then maintained in the animal colony at UMass Medical School. The
progeny of heavy chain VHSm7 knock-in and light chain Vk1AJk2 knock-in F1 mice
(T,G)A-L, generated by Dr. Joan Press (Brandeis University, Waltham, MA), were
kindly given to me by Dr. Robert Woodland (University of Massachusetts Medical
Center, Worcester, MA). IFN-α/β R knockout (R KO) mice (also abbreviated as IFN-α/β
R / , strain 129) were also provided by Dr. R. Woodland and then maintained in the animal colony at UMass Medical School.
52
B. Viruses, protein antigen, and mouse inoculation. Mice were inoculated
intraperitoneally (i.p.) with 4 x 105 PFU in 0.1 ml of LCMV, strain Armstrong, passaged
in baby hamster kidney (BHK21, American Type Culture Collection) cells in Dulbecco’s
medium (Invitrogen, Carlsbad, CA) containing 10% fetal calf serum (FCS). GP61-80 is
an LCMV-encoded I-Ab – restricted class II MHC peptide with the amino acid sequence
GLNGPDIYKGVYQFKSVEFD; NP309-328 is an LCMV-encoded I-Ab – restricted
class II MHC peptide with the sequence SGEGWPYIACRTSVVGRAWE; GP118-125 is
a subdominant LCMV-encoded Kb – restricted class I MHC peptide with the sequence
ISHNFCNL, and HA209-224(Cauley et al., 2002) is an influenza virus strain A/x31
encoded I-Ab – restricted class II MHC peptide derived from the hemagluttinin protein
with the sequence SLYVQASGRVTVSTRR. (T,G)A-L is a multivalent antigen with the
following structure: poly-L-(Tyr, Glu)-poly-DL-Ala--poly-L-Lys.
C. Preparation of cells. Single cell lymphocyte suspensions were prepared from spleens, peripheral blood, or bone marrow. In all cases erythrocytes were lysed using a
0.84% NH4Cl solution. Spleens were crushed using frosted microscope slides in
RPMI1640 (Invitrogen) and then filtered through nylon mesh. Peripheral blood was taken
from the tail vein into heparinized tubes. Bone marrow was obtained from femur and
tibia by removing both ends of the bone, followed by lavage with RPMI1640
(Invitrogen). For organ harvesting and subsequent adoptive transfer, media was not
supplemented. For organ harvesting at the termination of experiments, RPMI was
supplemented with 10% FCS. 53
D. Antibodies and Reagents. In order to identify the type of donor target cells
for some in vivo cytotoxicity assays, surface stains were performed as described below
using fluorescently labeled mAbs specific for Ly5.1 (A20), I-Ab (AF6-120.1), B220
(RA3-6B2), and CD19 (1D3), all purchased from BD PharMingen (San Diego, CA). For intracellular cytokine stains, the cells were stained for 30 min at 4°C with combinations
of fluorescently labeled mAbs specific for CD4 (L3T4), CD8α (53-6.7) or CD8β (Ly-3),
and IFN-γ (XMG1.2), all purchased from BD PharMingen, and for granzyme B (GB12),
purchased from Caltag Laboratories (Burlingame, CA).
For some in vivo cytotoxicity assays, CD4+ cells were depleted in vivo by i.p.
inoculation of 0.5 mg of mAb GK1.5 (Wilde et al., 1983) on both two and one days prior
to target cell transfer. Depletion of CD4 T cells was confirmed by cell surface stain for
L3T4, a different anti-CD4 Ab. CD8α cells were depleted using 2.5 mg mAb Lyt-2 (2.43;
ref. (Sarmiento et al., 1980) on both five and one days prior to target cell transfer.
Depletion of CD8 T cells was confirmed by cell surface stain for CD8β (Ly-3). mAb
GK1.5 (anti-CD4) was given to me by Dr. Kenneth Rock (UMass Medical School
Worcester, MA) and mAb Lyt-2 (2.43) was prepared by me from ascites.
For analysis of cells following adoptive transfer, single cell leukocyte suspensions were prepared. Following pre-incubation with 1 µl of Fc block (2.4G2) in 96-well plates
containing 100 µl of fluorescence activated cell sorting (FACS) buffer (Hank’s Balanced
Salt Solution [HBSS], 2% FCS, 0.1% NaN3), the cells were stained for 30 min. at 4°C
with combinations of fluorescently labeled mAbs specific for donor cell markers listed 54
below, B220 (RA3-6B2), CD19 (1D3), I-Ab (AF6-120.1), Fas/CD95 (Jo2), CD4 (RM4-
5), all purchased from BD PharMingen, and IgM (R6-60.2), IgD (11-26), and GL-7, all
purchased from eBioscience (San Diego, CA). Biotinlyated peanut agglutinin from
Vector Labs (Burlingam, CA) was used in combination with a streptavidin Pacific Blue
conjugate from Molecular Probes (Carlsbad, CA). Biotinylated monoclonal Ab 8A7 that
recognizes the variable region of the heavy and light Ab chains of (T,G)A-L transgenic
mice was generated by Dr. Joan Press and was kindly given to us by Dr. Robert
Woodland. This Ab was also used in conjunction with a streptavidin Pacific Blue
conjugate from Molecular Probes. Cells were washed and fixed in BD cytofix buffer for
5 min. at 4°C.
For some adoptive transfer experiments CD4+ T cells were depleted in vivo by
i.p. inoculation with 0.5 mg of mAb GK1.5(Wilde et al., 1983) one day prior to and three
days following adoptive transfer. Depletion of CD4 T cells was confirmed by cell surface
stain with a different anti-CD4 mAb, RM4-5. Blockade of costimulation was performed
by administering 0.5 mg of mAb to CD154, clone MR1(Noelle et al., 1992), i.p. one day
prior to adoptive transfer.
E. Flow cytometry. Freshly stained samples were analyzed using a BD
Biosciences LSRII and FACS Diva Software, FACSCalibur and Cellquest Pro Software
(San Diego, CA), or FlowJo (Treestar, Inc.) software.
55
F. Intracellular cytokine stain. LCMV peptide-specific, IFN-γ-secreting CD4+
and CD8+ T cells were detected using the Cytofix/Cytoperm Kit Plus (with GolgiPlug;
BD PharMingen), as described previously (Kim and Welsh, 2004). Briefly, cells were
incubated with 5 µM of synthetic peptide, 10 U/ml human rIL-2 (BD PharMingen), and
0.2 µl of GolgiPlug for 5 h at 37°C. To visualize overall functionality of CD4+ and CD8+
T cells, cells were incubated with 1 µg of purified anti-mouse CD3ε mAb (145-2c11; BD
PharMingen). Following pre-incubation with 1 µl of Fc block (2.4G2) in 96-well plates
containing 100 µl of FACS buffer (HBSS, 2% FCS, 0.1% NaN3), the cells were stained
for 30 min at 4°C with combinations of fluorescently labeled mAbs specific for CD4
(L3T4), CD8α (53-6.7) or CD8β (Ly-3), and IFN-γ (XMG1.2), all purchased from BD
PharMingen, and for granzyme B (GB12), purchased from Caltag Laboratories
(Burlingame, CA).
G. In vivo cytotoxicity assay. In vivo cytotoxicity was determined using
splenocytes coated with peptide epitopes and differentially labeled with the fluorescent dye 5(6) carboxyfluorescein diacetate succinimidyl ester (CFSE, Molecular Probes,
Eugene, OR), as previously described (Oehen et al., 1997). Splenocytes from naïve mice were separated and labeled in RPMI1640 (Invitrogen) with indicated peptides (0.5 µg/ml;
45 min at 37°C, 5% CO2). Populations were then labeled with different concentrations of CFSE (1µM and 0.4µM for two targets, or 1µM, 0.4µM, and 0.15µM for three targets) in HBSS (Invitrogen), combined, and injected i.v. into mice. Spleen lymphocytes were isolated 18 h later, and the ratio of CFSE low/CFSE high cells was determined by flow 56
cytometry. In order to identify the type of donor target cells for some assays, surface
stains were performed as described below using fluorescently labeled mAbs specific for
Ly5.1 (A20), I-Ab (AF6-120.1), Fas/CD95 ( Jo2), B220 (RA3-6B2), and CD19 (1D3), all
purchased from BD PharMingen (San Diego, CA). The percentage of specific killing
was calculated as follows: (1 - (ratio immune/ratio naive)) x 100. Ratio = number of
events LCMV peptide-coated target/number of events reference target.
H. Plaque assay. LCMV titers were determined by plaque assay. Organ
supernatants were obtained by grinding tissues in a mechanized mortar and pestle
apparatus. The organ slurry was then spun for 20 min at 1000 x G at 4°C to remove
cellular debris. Supernatants were collected and frozen at -80°C. Vero cell (ATCC)
monolayers, grown in 6-well plates (BD Biosciences), were infected with 10-fold
dilutions of freshly thawed organ supernatants for 90 min at 37° C in MEM (Invitrogen).
Monolayers were overlayed with a 1:1 mixture of 0.8% Seakem agarose (FMC
Bioproducts, Rockland, ME) and 2x EMEM (Biowhitaker, Walkersvile, MD)
supplemented with 6% FCS and antibiotics. The plates were then incubated and stained
on day four with a 1% neutral red stain (Sigma) included in the 1:1 mixture of 0.8%
Seakem agarose and 2x EMEM supplemented with 6% FCS and antibiotics.
I. Adoptive transfer of antigen-loaded cells. Donor splenocytes containing B
cells were isolated, charged with an immunogenic peptide, labeled with CFSE ( 2µM ),
and transferred i.v. into naïve or day 8 to 11 LCMV-infected mice, as previously 57
described(Kim et al., 2002). Briefly, splenocytes from naïve C57BL/6, naive αβTCR-/- ,
or naïve IgHa mice (all Ly5.2) were separated and labeled with 0.5 µg/ml peptide for 45
min at 37°C, 5% CO2. Populations were then labeled with 2 μM CFSE (Molecular
Probes, Eugene, OR) and 2 x 107 cells were injected i.v. into Ly5.1+ C57BL/6 mice. At no time were these donor B cells exposed to fetal calf serum (FCS), as FCS exposure caused a background proliferation likely due to the serum antigens. Six to seven days after transfer donor cells were identified and analyzed by flow cytometry, using fluorescently labeled mAbs specific for Ly5.1 (A20) or Ly5.2 (104) (BD PharMingen,
San Diego, CA).
J. Antibody and antigen specific ELISA. Enzyme linked immunosorbent assays
(ELISA) were done using a modification of the method described previously(Szomolanyi-Tsuda et al., 1998). Briefly, polyvinyl 96-well plates (Falcon,
Franklin Lakes, NJ) were coated with 100 µl of diluted anti-IgM, anti-IgG2aa, anti-IgG, or (T,G)A-L (1µg/ml in carbonate buffer), incubated overnight at 4° C, and then blocked
with 5% non-fat dry milk in whey dilution buffer (PBS containing 4% whey and 0.5%
Tween-20). Serum samples (diluted in whey dilution buffer) were added for 2 hours at
room temperature. Bound Ab was detected by using biotin-conjugated Abs to the mouse
IgMa allotype, IgG2aa allotype, or IgG (1µg/ml in whey dilution buffer) (BD
Pharmingen), and streptavidin-conjugated horseradish peroxidase (1µg/ml in whey dilution buffer) (Vector Laboratories, Burlingame, CA). Plates were developed using 58
3,3’,5,5’-tetramethylbenzidine tablets according to the manufacturers protocol (Sigma,
St. Louis, MO), and absorbence was read at 450 nm using a microplate reader.
K. Antibody and antigen specific ELISPOT. Enzyme linked immunospot assays (ELISPOT) were performed to determine ASC number. IgMa-, IgG2aa- and Anti-
(T,G)A-L IgG and IgM secreting ASC numbers were measured by ELISPOT using a
modified method for measuring influenza-specific ASC(Guay et al., 2004). Multiscreen
HA plates (Millipore, Bedford, MA) were coated with 100 µl anti-mouse IgM clone R6-
60.2, anti-Mouse IgG2aa clone 8.3 (both from BD Pharmingen) or (T,G)A-L (diluted to 1
µg/ml in PBS) overnight at 4°C, and blocked the following day with 200 µL RPMI 1640
(Invitrogen) containing 10% fetal calf serum (FCS) for 30 minutes at 37°C. Cells were
plated in duplicate at several dilutions in 200 µL RPMI 1640 containing 10% FCS and
incubated for 4 hours at 37°C. Bound Ab was detected using biotin-anti-mouse IgMa
(clone DS-1, BD Pharmingen) or IgG2a (clone 8.3) diluted to 1 mg/ml in PBS containing
1% FCS and 0.1% Tween-20, and by streptavidin-horse radish peroxidase (HRP, diluted
to 1µg/ml in PBS containing 1% FCS and 0.1% Tween-20; Vector Laboratories). Plates
were developed using 3-amino-9-ethylcarbazole (AEC) substrate (Sigma) according to
the manufacturer’s protocol. For some assays streptavidin-ammonium peroxidase
(Invitrogen) diluted to (1ug/ml as above) and was used in combination with BCIP/NBT substrates (Sigma) according to the manufacturer’s protocol. Anti-(T,G)A-L bound Ab
was detected using anti-mouse IgG or IgM directly conjugated to ammonium peroxidase 59
(Southern Biotech, Birmingham, AL) in combination with BCIP/NBT substrates. Spots were counted using a dissection microscope.
L. Statistical analysis Where appropriate, statistics were calculated by performing student’s t-tests with Microsoft Excel software. A difference was considered significant when p=<0.05. 60
CHAPTER III
MHC CLASS II-RESTRICTED CYTOTOXICITY IN VIVO
FOLLOWING LCMV INFECTION
Antiviral MHC class II-restricted cytotoxic CD4 T cells have previously been
described in vitro in cell cultures (Jacobson et al., 1984; Lukacher et al., 1985; Maimone
et al., 1986; Hou et al., 1993). However, the detection of this cytotoxic function was
deemed to be an artifact of in vitro culture (Bourgault et al., 1989). Cytolytic activity
could be detected in vitro if CD4 T cells were intoduced to a strong polyclonal stimulus in vivo (Erb et al., 1990), but only recently was antiviral cytolytic activity detected directly ex vivo in freshly isolated cells (Appay et al., 2002).
Despite a large amount of in vitro data, the existence of MHC class II-restricted killing in vivo was still in doubt at the initiation of my thesis work. Elimination of MHC class II epitope-expressing target cells in LCMV-infected mice would provide evidence that class II-restricted killing can occur during viral infection. The LCMV infection stimulates strong CD8 and CD4 T cell responses against a number of defined MHC class
I- (Whitton et al., 1988; van der Most et al., 1998) and MHC class II- (Oxenius et al.,
1995) presented epitopes. The two MHC class II I-A -restricted epitopes for LCMV are
NP309-328 and GP61-80 (Oxenius et al., 1995), and up to 10% of CD4 T cells can be specific to these epitopes during an acute infection in C57BL/6 mice(Varga and Welsh,
1998a). Due to such a high frequency of effector CD4 T cells, LCMV infection of mice 61
provides an attractive model system to demonstrate MHC class II-restricted killing in vivo. In this chapter, I provide evidence for MHC class II-restricted killing in vivo by performing in vivo cytotoxicity assays in LCMV-infected C57BL/6 mice (Oehen et al.,
1997). By using this approach, I am able to identify mechanisms used by cytotoxic CD4
T cells to eliminate target cells in vivo, and in these assays MHC class II-expressing B cells that present viral antigens become targets for MHC class II-restricted killing.
A. Elimination of LCMV MHC Class II peptide-coated target cells in infected animals. I first examined killing of GP61-coated targets in vivo by transferring
Ly5.1 splenocytes into day 14 LCMV-infected congenic C57BL/6 (Ly5.2) mice. After 20 hours in vivo, about 52% of the MHC class II high donor cells (lymphocyte gated) labeled with GP61 peptide were lost when compared to the reference influenza A HA209 peptide-coated donor population (Fig. III-1A and Table III-1A). In contrast to these results, GP61-coated MHC class II low donor cells in the same recipient mouse were not eliminated (Fig. III-1B and Table III-1A). In vivo cytotoxicity against class II high GP61- coated targets was also observed at day 9 postinfection (Table III-1B). As a second approach for doing in vivo cytotoxicity assays, I used donor cells from αβ T cell KO mice in my assay. Due to the absence of T cells, most of the donor cells from αβ T cell KO mice constitutively expressed class II MHC and could be evaluated directly without staining for class II MHC in the in vivo cytotoxicity assay. This technique demonstrates a similar loss of the GP61-coated population when compared to the reference population
(Fig. III-1C and Table III-1C, D). This killing was also seen when the CFSE label was 62
63
64
reversed such that the GP61-coated population was CFSE high and the reference
population was CFSE low (killing of GP61-coated targets CFSE high = 48 % vs. killing
of GP61 coated target cells CFSE low = 47%). A reduced level of cytotoxicity (16%) was
seen after incubation for 6 hours in vivo compared to 48% killing seen after 20 hours.
The time necessary to lyse targets may reflect the mechanism used to kill or it may
represent the amount of time it takes for a CD4 CTL to find its target or hop from one
target to another. Lower levels of in vivo cytotoxicity were directed against the LCMV-
subdominant MHC class II epitope, NP309, at day 14 post-infection (Table III-1C).
When αβ T cell KO donor cells were stained for expression of the B cell markers CD19
and B220, 80% of the donor lymphocytes and 60% of the donor leukocytes based on
forward and side light scattering profiles were double positive. By comparing the
cytotoxicity between these B cells and all other lymphocyte-gated donor cells, B cells had
a 53% loss of the GP61-coated population whereas none (- 2%) of the other donor cells
were killed (Table III-1D). Taken together, these data demonstrate the existence of viral antigen-specific MHC class II-restricted killing in vivo.
B. CD4 cell-dependent cytotoxicity. Since MHC class II-peptide complexes are recognized by CD4 T cells, I determined whether CD4 T cells were acting as the effectors by depleting hosts of CD4 cells using the monoclonal Ab GK1.5 prior to
injection of the donor populations. CD4 depletion was highly effective, as shown in Fig.
III-2. In naïve mice, equal numbers of the injected GP61- and reference peptide-coated
65
66
populations of αβ T cell KO splenocytes were observed by flow cytometry (Fig. III-2,
Day 0). At Day 14, a selective loss of the GP61-coated population was observed in
LCMV-infected mice with an intact CD4 cell population (Fig. III-2, Day 14 LCMV,
Table III-1E). In infected mice depleted of CD4 T cells two days prior to transfer, the
GP61-coated targets were not eliminated, and their profiles resembled those in naïve mice in that the GP61-coated population was about equal in size to the reference
population (Fig. III-2, Day 14 LCMV –CD4, Table III-1E). The amount of killing was
reduced about 5-fold, from 41% in controls to 8% in CD4-depleted animals (Table III-
1E). In a separate experiment, depletion of CD8α cells in vivo did not prevent the
elimination of GP61-coated targets (Table III-1F). These experiments indicate that the
killing of GP61-coated αβ T cell KO splenocytes is dependent on CD4 T cells and not
CD8 T cells.
C. The role of Fas-FasL interactions in MHC class II-restricted killing.
Several effector mechanisms have been demonstrated in the lysis of target cells in vitro
by CD4 T cells, including FasL, TNF-α, and perforin. In order to implicate a mechanism
for CD4-directed MHC class II-restricted killing in vivo during the LCMV infection, I used splenocytes from Fas mutant lpr mice mixed with Ly5.1 splenocytes as donors and transferred these into either naïve or LCMV-infected mice. I was thus able to distinguish killing of Fas mutant lpr targets from wild type targets in the same mouse based on the congenic marker. Lpr targets were killed less efficiently than wild type targets (26 versus
41%, respectively) (Table III-1G). A similar pattern was seen on day 9 of infection 67
(Table III-1H). These results suggested that engagement of Fas-FasL may be required for
some but not all of the class II-restricted killing in vivo. To examine the role of FasL, I
used αβ T cell KO splenocytes as donors and either C57BL/6 or FasL-mutant gld mice as hosts in my assay. αβ T cell KO splenocytes were killed less efficiently in LCMV- infected gld hosts compared to LCMV-infected C57BL/6 hosts (16 versus 31%, respectively) (Table III-1I). These results complement the results using lpr targets and further indicate a role for Fas-FasL in some, but not all, of the MHC class II-restricted killing. Analysis of the intracellular cytokine profiles of gld mice revealed no deficiency in the frequencies of CD4 T cells producing IFN-γ (Fig. III-3) in response to stimulation by GP61, when compared to wild type mice. Additional analysis showed that Fas expression was selectively elevated on the remaining GP61-coated target B cell population (Fig. III-4). Those GP61-coated B cells that survived MHC class II-restricted killing had a higher mean fluorescence intensity (MFI) of Fas expression than HA209- coated B cells or GP61-coated B cells from uninfected mice.
D. The role of TNF-TNFR interactions in MHC class II-restricted killing.
FasL and TNF-α can sometimes exert complimentary or redundant functions, and both have been implicated in CD4 T cell killing in vitro (Tite, 1990; Stalder et al., 1994). I therefore performed the class II killing assay using TNFR1 KO splenocytes as donors or
TNF-α KO mice as hosts. TNFR1 KO target splenocytes were killed at the same level of their wild type counterparts (Table III-J), suggesting that TNF was not required for this in vivo cytotoxicity. Killing of wild type targets was observed in LCMV-infected TNF-α 68
69
70
KO mice (15±6%, n=5), though it was not as robust as in littermate controls (29±8%,
n=5), but this difference could be attributed to significant differences in the number of
GP61-specific T cells in vivo (1.6 x 106, n=2, in wild type vs. 7.6 x 105, n=2, in TNF-α
KO).
E. The role of perforin in MHC class II-restricted killing. I also examined
MHC class II-restricted killing in perforin KO mice and found virtually no killing in six out of six perforin KO mice at Day 9, despite the presence of CD4 T cells that produced
IFN-γ in response to GP61 (3.5±1% of CD4 T cells, n=6). Only one of four surviving mice at day 14 had >7% killing. These experiments, however, were complicated by the fact that perforin is needed for viral clearance, and the resulting high antigen load in vivo may cause immune suppression or may have competed for the cytotoxic CD4 T cells and result in reduced clearance of the transferred cell population. To reduce viral load in
those assays, I transferred 5x105 GP33-specific P14 transgenic T cells into perforin
knock-out mice 1 day prior to infection and assayed for class II-restricted killing on day
14 post-infection. This treatment resulted in a 15-fold increase in leukocyte number over the normally lymphopenic perforin KO spleen, but failed to reveal in vivo class II- restricted cytotoxicity (-12±3%, n=3). These results suggest a role for perforin in in vivo cytolysis, but a complete clarification of the role of perforin might be better analyzed in another system where perforin is not essential for the clearance of an acute infection.
LCMV infection in perforin KO mice results in the establishment of a persistent state where CD4 T cell function is impaired and clonal exhaustion of CD4 T cells may occur 71
(Fuller and Zajac, 2003). It is unclear whether perforin KO CD4 T cells have been clonally exhausted at the time these in vivo cytotoxicity assays were performed.
However, preliminary results in assays of polyclonal B cell activation suggest that CD4 T cells from LCMV-infected perforin KO mice are still functional because they have the ability to induce the proliferation of GP61-coated B cells in vivo. Thus, CD4 T cells from perforin KO mice can interact with adoptively transferred B cells and may only be impaired in their ability to lyse these antigen-presenting B cells, leading to the conclusion that perforin is required for MHC class II-restricted killing.
F. The role of CD40-CD40L interactions in MHC class II-restricted killing.
CD40-CD40L interactions are vital for B cell activation. Mice that are incapable of signaling through CD40 may have an even greater level of MHC class II-restricted cytoxicity because the target B cell population will be missing a potent survival signal.
To examine the role of CD40-CD40L interactions on class II-restricted killing, I performed MHC class II-restricted in vivo cytotoxicity assays in mice treated with 1.5 mg of anti-CD40L Ab clone MR1, which has previously been shown to block CD40-CD40L interactions in vivo(Noelle et al., 1992). The GP61-coated target population was eliminated in anti-CD40L treated mice at a similar frequency as untreated control mice
(Table III-2A and C). I also performed MHC class II-restricted in vivo cytotoxicity assays using CD40KO target cells and again found no difference in specific lysis of
GP61-coated CD40KO targets compared with WT targets (Table III-2B and D).
Interestingly, despite equal levels of killing, CD40KO target cells did not upregulate Fas 72
73
expression (Fig. III-5), suggesting that engagement of CD40 on B cells is required for the
upregulation of Fas seen in WT cells that survive MHC class II-restricted cytotxicity
(Fig. III-4). However, because GP61-coated CD40KO and WT B cells are eliminated at
equal frequency, these results lead me to believe the CD40-CD40L interactions do not provide any additional survival signals that protect target B cells from CD4-mediated
cytotoxicity.
G. The role of interferon-gamma in MHC class II-restricted killing. IFN-γ is an antiviral cytokine that is produced by CD4 T cells during the acute LCMV response(Varga and Welsh, 1998a). LCMV Armstrong and WE replicate to higher levels in IFN-γ-depleted mice. (Moskophidis et al., 1994; Muller et al., 1994). IFN-γ is also important for the clearance of other viruses, such as MCMV although in this system, the
IFN-γ is produced by NK cells(Orange and Biron, 1996). I tested whether or not IFN-γ was involved in MHC class II-restricted killing by transferring IFN-γ R KO targets that were Ly5.2 positive and WT targets that were Ly5.1 positive into the same LCMV- infected hosts. I found that IFN-γ R KO targets were killed at a similar frequency as their
WT counterparts (55% vs. 49%, Table III-2E), lending strong evidence that IFN-γ is not the primary mechanism used by CD4 T cells to kill their targets nor does it provide a survival signal to targets.
74
75
H. Chapter summary. Previous in vitro studies have shown that CD4 T cells
can kill virus infected targets, but this MHC class II restricted cytotoxicity had never
been shown to occur in vivo. It is reasonable to suppose that CD4 mediated cytotoxicity
could be detected in vivo in infections that generate strong CD4 T cell responses. LCMV
generates such a response. Using in vivo cytotoxicity assays, I specifically asked what
would happen to MHC class II expressing cells that present viral antigens following adoptive transfer into LCMV-infected hosts.
Approximately one day after adoptive transfer, there was a significant loss of the cells expressing the I-Ab -presented MHC class II epitopes GP61 or NP309 in LCMV-
infected mice compared to cells coated with a peptide from an unrelated virus. Thus,
MHC class II-restricted killing can occur in vivo during viral infection. This killing is
specific, because only virus-specific targets were eliminated and they were only
eliminated in infected mice. Mice depleted of CD4 T cells prior to adoptive transfer did
not eliminate virus-specific targets, suggesting that CD4 T cells are required for this
killing. Depletion of CD8 T cells prior to adoptive transfer reveals that CD8 CTL are not
involved. GP61-coated target cells were killed at a greater frequency than NP309-coated
cells, and this may be a reflection of the number of cytotoxic CD4 T cells stimulated by
these two epitopes. The frequency of effector CD4 T cells detected by intracellular
cytokine staining in previous studies showed a similar dominance of the GP61 response
over the NP309 response(Varga and Welsh, 1998a). GP61 is a more stimulatory epitope
than NP309 for both the induction of CD4 CTL and cytokine secretion. 76
As a result of my studies I have also identified the mechanism of MHC class II- restricted killing. As with previously published in vitro data, CD4 T cell-dependent cytotoxicity cannot be solely explained by one mechanism, but I have identified Fas-FasL interactions as one of the mechanisms and my results also strongly implicate a role for perforin. In vivo cytotoxicity assay using gld (FasL-deficient) hosts and lpr (Fas- deficient) targets showed a reduced elimination of GP61-coated target cells in comparison to WT. Assays done in perfoin KO mice showed no elimination of GP61- coated target cells despite the ability of perforin KO CD4 T cells to functionally interact with antigen-presenting B cells.
In these assays, B cells expressing viral antigens were targeted for lysis. Killing was restricted to MHC class II high cells that express the prototypic B cell markers B220 and CD19. The killing of B cells in LCMV-infected mice opposes the proposition that antigen-presenting B cells can be induced during virus infection to undego polyclonal activation and cause virus-induced hypergammaglobulinemia. However, my results show cytotoxicity against B cell targets is incomplete and I will show in the next chapter that the surviving B cells that express viral antigens undergo polyclonal activation in LCMV- infected hosts. 77
CHAPTER IV
THE POLYCLONAL ACTIVATION OF B CELLS FOLLOWING LCMV
INFECTION
Hypergammaglobulinemia following viral infection has been observed for many years in humans and in mice (Rosen et al., 1977; Hutt-Fletcher et al., 1983; Racanelli et al., 2006; Lane et al., 1983; Li et al., 1990; Coutelier et al., 1988; Price et al., 1993;
Stevenson and Doherty, 1998). This hypergammaglobulinemia is presumed to be the
result of a polyclonal activation of B cells during viral infection. However, with the
exception of the viruses like EBV that directly infect and transform B cells, polyclonal B
cell activation as a result of virus infection has not formally been described. Recently, a correlation between the hypergammaglobulinemia associated with viral infection and high titer virus infection was made. Based on this correlation, a mechanism has been proposed to explain hypergammaglobulinemia where B cells, regardless of their antigen specificity, can take up viral antigen and present it to CD4 T cells which, in turn, induce
Ab production in a CD40-dependent manner (Hunziker et al., 2003). If hypergammaglobulinemia is the result of polyclonal B cell activation, then B cells presenting viral antigens should be activated upon transfer into a virus-infected host by virus-specific CD4 T cells within that host. The MHC class II-restricted cytotoxicity described in chapter III seems incompatible with this concept because viral-antigen- expressing B cells were killed by CD4 T cells. However, the results in chapter III show 78
that the CD4-mediated cytotoxicity against virus epitope-presenting B cell targets was
incomplete and, in this chapter, I show that the surviving B cells become polyclonally activated in a CD4 T cell- and CD40-dependent manner which was also proposed to be responsible for virus-induced hypergammaglobulinemia. Unlike previous studies of virus-induced hypergammaglobulinemia, I provide a detailed analysis of virus-induced polyclonal B cell activation, including: proliferation, changes in phenotype, and Ab production.
A. Proliferation of LCMV class II epitope-expressing B cells in vivo in
LCMV-infected animals. The cytotoxicity against B cells expressing MHC class II antigens described in chapter III was incomplete. I asked what would happen to the remaining population of antigen-labeled B cells in LCMV-infected mice if they were left in vivo for several days. To examine the effects of virus infection on the antigen- presenting naïve B cell population, CFSE-labeled splenocytes from αβ TCR KO (Ly5.2) mice were transferred into day 10 LCMV-infected or uninfected congenic B6.SJL
(Ly5.1) mice. Day 10 was chosen because it is about at the peak of the LCMV-specific
CD4 T cell response (Varga and Welsh, 1998a), because viral antigens are mostly cleared by this time, and because the incubation period of the donor B cell population in vivo would overlap with the timepoints described in chapter III . Because of the absence of T cells in αβ TCR KO mice, B cells were enriched in the donor population. Six days after transfer, the CFSE profiles of donor Ly5.2+ B cells, identified by costaining with the
prototypic B cell markers CD19 and B220, were compared between LCMV-infected 79
80 81
mice and uninfected mice. B cells coated with LCMV-specific peptides GP61 and NP309
were induced to divide, as indicated by loss of CFSE label (Fig. IV-1a), and proliferate,
meaning an increase in number (Table IV), in LCMV-infected mice but not in uninfected mice. Many of the cells that survived the predictable CD4 T cell-mediated cytotoxicity reported earlier in chapter III underwent division, demonstrating a high level of polyclonal activation apparently independent of BCR specificity. Calculations made by
FlowJo software, which analyzes the CFSE profile based on the size of each peak, show that 64±11% of the remaining GP61-coated B cells and 19±4% of the remaining NP309- coated B cells that survived the initial cytotoxicity went on to divide in vivo following transfer into LCMV-infected hosts.
Donor cell division and proliferation reflected the CD4 T cell dominance hierarchy of these two LCMV-encoded peptides (Varga and Welsh, 1998a), as B cells coated with GP61 peptide proliferated more than those coated with NP309 peptide. This suggests that the proportion of B cells that proliferate may be affected by the frequency of CD4 T cells available to recognize the peptide. Donor B cells coated with an LCMV
MHC class I-restricted subdominant peptide GP92 were not induced to divide (Fig. IV-
1). This subdominant class I peptide was used as a control, as it did not sensitize B cells to in vivo cytotoxicity by CD8 T cells, in contrast to cells coated with the LCMV immunodominant class I peptides (Barber et al., 2003). Uncoated donor cells or donor cells coated with HA209, an irrelevant MHC class II-restricted peptide from influenza A virus, did not divide (Fig. IV-1) or proliferate (Table IV). Similar experiments using cells from C57BL/6 (Ly5.2) naïve mice as donors and Ly5.1 mice as hosts yielded similar 82
results. Also, GP61-coated donor B cells did not divide following adoptive transfer into influenza A virus-infected mice. In most experiments, peptide-coated B cells were
adoptively transferred at day 9 or 10 of LCMV infection, but LCMV class II peptide-
coated B cells also proliferated if they were adoptively transferred into hosts at day 4, day
7, day 14 and day 20 of LCMV infection, and they still proliferated, but to a lower degree, in the memory state at day 118 post-infection. Taken together, these data demonstrate a peptide-specific MHC class II-dependent polyclonal activation of B cells during LCMV infection.
The proliferation of the LCMV MHC class II peptide-coated B cells was evaluated in two ways. First, I calculated the number of donor B cells and found that
there was an overall increase in the number of GP61 peptide-coated donor B cells in
LCMV-infected mice compared to all other treatments (Table IV-1). Nevertheless, I was
concerned that the “take” of transferred cells found following adoptive transfer may vary
among LCMV-infected animals and uninfected animals. I thus also calculated the ratio of
donor B cells to all other non-B donor cells to monitor the proportional increase of B
cells within the donor cell population. These calculations show that the ratio of αβ TCR
KO donor GP61-coated B cells to non B cells was about 10:1 in LCMV-infected mice,
and the ratio of NP309-coated B cells to non B cells was 5:1, whereas the ratio of donor
B cells to non B cells was about 2:1 in control groups. Similar results were obtained
when C57BL/6 or B cell congenic (IgHa) donors were used; the ratio of GP61-coated B
cells to non B cells was 3:1, in contrast to a 1:1 or 1:2 ratio in control hosts. C57BL/6
donor cell ratios were lower than those for the αβ TCR KO donors because of B cell 83
dilution by donor T cells. These calculations clearly show that the overall number of
LCMV MHC class II peptide-coated donor B cells increases in LCMV-infected mice.
Further, the B cell number seen six days after transfer is likely an underestimate of the
extent of proliferation, considering that about half of the B cells would be expected to die
due to CD4 cell-mediated cytotoxicity within the first 24 hrs of transfer, as described in
chapter III.
B. Characterization of the phenotypic changes in polyclonally activated B
cells. As these GP61-coated B cells underwent polyclonal activation in LCMV- infected mice, their surface phenotype changed in a manner characteristic of conventionally activated B cell populations (Monroe et al., 1983; Monroe and Cambier, 1983). These activated B cells up-regulated I-Ab (MHC class II) levels as they divided, and the entire
dividing GP61-coated population had higher levels of MHC class II than uncoated B
cells, with an average change in mean fluorescence intensity (MFI) of 48%, ranging from
21%-68% (Fig. IV-2A, representive of 3 experiments, n=4). By their fifth division GP61-
coated donor B cells had also down-regulated IgD, with an average decrease in MFI of
39%, ranging from 30%-43% (Fig. IV-2B, representative of 3 experiments, n=4), and up-
regulated IgM levels, with an average increase in MFI of 92%, ranging from 42%-126%
(Fig. IV-2C, representative of 4 experiments, n=7). Some of these B cells increased
surface expression of the GC markers. For these GC markers, I compared activated
GP61-coated B cells in LCMV-infected mice with uncoated B cells in LCMV-infected
mice and found that the dividing population of GP61-coated cells stained higher for both 84
85
GL7, with a 129% average increase in MFI, ranging from 31%-300% (Fig. IV-2D, representative of 3 experiments, n=3) and peanut agglutinin (PNA), with a 140% average increase in MFI, ranging from 121%-161% (Fig. IV-2E, representative of 2 experiments, n=3), consistent with entry into GCs.
As the LCMV-specific peptide-coated B cells divided, dynamic changes occurred in their surface BCR phenotype (Fig. IV-3). The least divided cells were IgD single positive. The next group, which had a moderate loss of CFSE, was IgD and IgM double positive. IgM single positive cells had an even greater CFSE loss. But the most divided cells, as determined by loss of CFSE, were negative for both IgD and IgM. This latter result could mean that those B cells have become plasma cells and down-regulated their
BCR or it could mean that those B cells have class-switched to a BCR that is not IgM or
IgD.
Because I had previously seen that B cells coated with LCMV epitopes up- regulated Fas expression (Fig. III-4), and because Fas was important for MHC class II- restricted killing, I hypothesized that these surviving B cells would be Fas low, which may be the reason they escaped the CD4-mediated cytotoxicity. Instead, I found that the polyclonally activated GP61-coated population in LCMV-infected hosts had elevated levels of Fas in comparison to GP61-coated cells in uninfected mice or uncoated or
HA209-coated B cells in LCMV-infected and uninfected hosts (Fig. IV-4). Furthermore,
Fas expression seemed to increase as the GP61-coated B cells divided. These results are consistent with the other phenotypic changes seen in the dividing GP61-coated population in Figure IV-2 because it has previously been reported that up-regulation of 86
87
88
MHC class II and increased binding of GL7 or PNA coincide with Fas expression on GC
B cells (Han et al., 1997).
I next questioned if there was preferential activation of a particular splenic B cell
subset or if both MZ and follicular B cells were equally induced to undergo polyclonal
activation. I therefore stained GP61-coated B cells from LCMV-infected mice for
expression of markers that are differentially expressed on these two populations: the
complement receptor, CD21, and the low affinity Fc epsilon receptor, CD23. I found that
about 75% of the transferred cell population was of the follicular phenotype
(CD21intermediate, CD23+) whereas only 3.5% had a MZ phenotype (CD21high, CD23-
). A third population of B cells was found to be negative for both markers. All
populations were capable of polyclonal activation in LCMV-infected mice (Fig. IV-5).
C. CD4 cell-dependent proliferation. To determine whether CD4 T cells
influenced the activation and proliferation of peptide-coated donor B cells, I depleted
host mice of CD4+ cells with mAb GK1.5 one day prior to and three days following the
adoptive transfer of peptide-coated αβ TCR KO splenocytes. The CD4 depletion was
highly effective (Fig. IV-6 small boxes). Mice with an intact CD4 cell population (Fig.
IV-6) had CFSE profiles similar to those previously shown in Fig. IV-1. In infected mice
GP61-coated B cells divided better than NP309-coated B cells (Fig. IV-6). Again, no
division was seen in uninfected mice. In infected mice that were depleted of CD4 cells, no division was detected, and the CFSE profiles resembled those from uninfected mice 89 90
91
(Fig. IV-6). These experiments indicate that the peptide-specific division of GP61- and
NP309- coated B cells was dependent on CD4 cells.
Although most of these adoptive transfers of antigen-loaded B cells were performed at day 9 or 10 post-infection, when LCMV is normally cleared, there remains the possibility that some mice will have low levels of viral replication at the time of adoptive transfer which may play a role in polyclonal B activation. Furthermore, the peak
in antiviral cytokine production also occurs around the time of adoptive transfer and it
could be argued that polyclonal activation occurs as a result of a “bystander” cytokine
effect and not via the direct recognition of viral antigens by CD4 T cells. The question
remained whether virus-specific CD4 T cells alone were sufficient to induce polyclonal B
cell activation. To address these possibilities, I questioned whether memory CD4 T cells
could induce polyclonal B cell activation by transferring GP61-coated B cells into
immune mice infected 118 days previously with LCMV. These LCMV-immune mice
should have undergone all phases of the acute immune response to LCMV and should be
left with a resting population LCMV-specific CD4 memory T cells. It is unlikely that
active viral replication is occurring at 118 days post acute LCMV Armstrong infection of
immunocompetent mice and it is unlikely that there would be any residual antiviral
cytokines at 118 days post-infection (Welsh, 2000). GP61-coated B cells underwent
several rounds of division in LCMV-immune mice after seven days in vivo, albeit, to a
lower degree than in acute animals (Fig. IV-7). These results suggest that neither active
viral replication nor virus-induced “bystander” cytokine effects are responsible for 92
93
polyclonal B cell activation and virus-specific memory CD4 T cells can induce polyclonal B cell activation.
D. The role of CD40-CD40L interactions in polyclonal B cell activation.
CD40 signaling is important for antigen-specific TD B cell activation, GC formation and memory formation (Borrow et al., 1996). CD40 stimulation has also been implicated in the polyclonal activation of human memory B cells in vitro (Bernasconi et al., 2002;
Traggiai et al., 2003) and for the hypergammaglobulinemia seen following high dose
LCMV infection (Hunziker et al., 2003). I tested the role of CD40 stimulation in my system by blocking CD40-CD40L interactions with mAb to CD40L. The CFSE profiles of donor cells from infected animals without anti-CD40L treatment demonstrated GP61- specific polyclonal activation (Fig. IV-8A), but anti-CD40L treatment almost completely inhibited the B cell division. Further experiments were done using GP61-coated spleen cells from CD40 KO mice as donors for transfer into LCMV-infected mice. These B cells did not undergo polyclonal B cell activation (Fig. IV-8B). Overall, these experiments identify CD40-CD40L interactions as a mechanism needed for the proliferation of GP61- coated B cells. These results contrast with the in vivo cytotoxicity assays, where lysis of
B cells was not blocked by this treatment (Table III-2, B and C).
E. The role of MyD88-dependent signaling in polyclonal B cell activation. Toll like receptor (TLR) signaling has been proposed to be a requirement for T cell dependent B cell responses, but the role of TLR signaling during antiviral immune responses may be 94
95
more complex than initially proposed (Pasare and Medzhitov, 2005; Guay et al., 2007).
Work from our lab has shown that MyD88-dependent signaling is not required for acute
T cell-dependent antiviral B cell responses but it is required for effectively generating
long-term Ab responses (Guay et al., 2007). TLR signaling has also been shown to
induce polyclonal B cell activation of human memory B cells in vitro (Bernasconi et al.,
2002). Based of these observations, I questioned whether TLR signaling would be
required for LCMV-induced polyclonal activation. I therefore transferred GP61-coated
donor cells from mice deficient in MyD88, a signaling adaptor for most TLR signaling,
into LCMV-infected hosts. B cells deficient in MyD88 underwent several rounds of division in LCMV infected mice, suggesting that MyD88-dependent TLR signaling within the B cell was not required for LCMV-induced polyclonal activation as manifested by proliferation (Fig. IV-9, upper panel). I next asked whether MyD88-dependent signaling was required in T cells for polyclonal B cell activation to occur. I transferred
GP61–coated WT B cells into LCMV-infected MyD88KO hosts. The caveat to this
experiment is that MyD88KO mice have a defective CD4 T cell response to LCMV
(Zhou et al., 2008). Despite this defect, GP61-coated B cells divided several times in
LCMV-infected MyD88KO hosts (Fig. IV-9, lower panel). These results suggest that
MyD88-dependent signaling is not required in B cells or T cells for virus-induced polyclonal B cell activation.
F. The role of type I interferon signaling in polyclonal B cell activation.
LCMV induces a potent type I IFN response 2 days post-infection, and LCMV replicates 96
97
to a higher extent in mice deficient in type I IFN signaling (Merigan et al., 1977; Muller
et al., 1994; van den Broek et al., 1995). Type I IFN produced as a result of virus
infection has been reported to enhance B cell responses. Type I IFN produced as a result of VSV infection was shown to mediate class switching to IgG and type I IFN produduced as a result of influenza virus infection up-regulated expression of the activation marker CD69 and costimulation ligand CD86 (Fink et al., 2006; Coro et al.,
2006). Influenza virus-induced type I IFN has also been shown to induce B cell activation in autoimmune disease-prone mice, resulting in a lupus-like disease. This disease was not seen after infection of type I IFN R KO animals bred onto the same genetic background
(Jorgensen et al., 2007; Woods et al., 2007). When given in conjunction with a protein antigen, recombinant type I IFN, enhances Ab production and GC formation(Le et al.,
2006) Based on these observations, I asked whether type I IFN signaling was important
to induce polyclonal B cell activation during LCMV infection. I therefore transferred
GP61-coated type I IFN receptor (type I IFN R) KO cells into LCMV-infected and
uninfected hosts. Type I IFN R KO cells had a similar division profile as their WT
counterparts (Fig. IV-10) indicating that type I IFN R signaling in B cells is not required
for polyclonal activation to occur.
G. The role of interferon-gamma signaling in polyclonal B cell activation.
IFN-γ is an antiviral cytokine that is produced by CD4 T cells during the acute LCMV
response (Varga and Welsh, 1998a). LCMV Armstrong and WE have higher viral
replication levels in IFN-γ-depleted mice (Moskophidis et al., 1994; Muller et al., 1994). 98 99
IFN-γ is the major switch factor for IgG2a production by B cells and induces higher
antigen presentation by B cells(Hawrylowicz and Unanue, 1988). The earliest reports
using recombinant IFN-γ showed that it increased the proliferating capacity of resting B
cells in vitro when given polyclonal stimuli like the TLR agonist lipopolysaccharide,
though IFN-γ could not induce proliferation on its own (Snapper et al., 1988). I therefore asked whether IFN-γR signaling was important for virus-induced polyclonal B cell proliferation by transferring GP61-coated IFN-γR KO donor cells into LCMV-infected hosts. The GP61-coated B cells underwent several rounds of division after six days in vivo (Fig. IV-11). Quantitatively, proliferation analysis of the CFSE profiles by FlowJo software did not reveal a major difference in the percentage of GP61-coated IFN-γR KO
B cells that underwent cell division compared to GP61-coated WT B cells (47% versus
50%, respectively). Subjectively, however, GP61-coated IFN-γR KO B cells appeared to have one or two fewer CFSE low peaks than WT B cells. More experiments using IFN-
γR KO need to be performed to elucidate this possible trend in cell division. Overall,
IFN-γR KO B cells can indeed divide in LCMV-infected hosts and the proportion of IFN-
γR KO B cells that divide is similar to WT GP61-coated B cells. Thus, IFN-γR signaling is not absolutely required in B cells for polyclonal activation.
H. Polyclonal activation of B cells from LCMV carrier mice. In the system of polyclonal activation of B cells described thus far, donor B cells were charged with peptide in vitro. I asked if B cells could be charged with viral antigens in vivo in the presence of high antigen loads and if this would lead to their polyclonal activation upon 100
101
exposure to helper CD4 T cells. To test for this, I used B cells from mice that were
congenitally infected with LCMV (LCMV-carrier mice). These mice maintain high
antigen loads in vivo, as they are immunologically tolerant to the LCMV infection and
generate neither a CD4 nor CD8 LCMV-specific T cell response (Buchmeier et al.,
1980). B cells from Ly5.2+ LCMV-carrier mice were therefore harvested, labeled with
CFSE, and directly transferred without any peptide treatment into LCMV-infected or
uninfected Ly5.1+ hosts. I detected 1 x 104 PFU in the serum (Fig. IV-12b) and 1 x 105
PFU of LCMV in the spleen (Fig. IV-12c) of donor mice that were used in two of these experiments. Donor B cells from LCMV-carrier mice divided in acute-LCMV-infected hosts (Fig IV-12) at levels higher than uncoated non-dividing B cells from uninfected mice, but at a lower degree than in vitro LCMV peptide-charged B cells (compare Fig.
IV-12 to Fig. IV-1). This division occurred in 4 out of 5 donor groups tested, and in these mice 7-15% of donor B cells divided, as calculated by FlowJo software’s proliferation algorithm, which analyzes CFSE profiles according to peak size. Studies in several mouse strains have predicted the number of B cells secreting Ab to LCMV in carrier mice to be less than 1 out of 1000 (Moskophidis and Lehmann-Grube, 1984;
Moskophidis et al., 1986), a frequency much lower than the 15% that became activated following adoptive transfer. Thus, this polyclonal B cell response must include predominantly cells whose BCR is not LCMV specific. It is possible that some of the dividing B cells were infected with LCMV, but the frequencies of those proliferating exceeded by about a log the reported frequencies (1.5%) of infected B cells in carrier mice (Homann et al., 2004), suggesting that at least some were cross-presenting antigens. 102
103
LCMV carrier B cells did not divide following transfer into previously uninfected
animals, despite the fact that the hosts may have become infected with LCMV for 6 days
due to carrier cell transfer. This lack of division between day 0 and day 6 is not surprising because the earliest detectable division for GP61-coated B cells was observed in the time period between day 4 and day 10 in LCMV-infected mice and it takes about 3 days for B cell division to get started (data not shown). Taken together, these results suggest that B cells, armed by high antigen levels in vivo, are able to undergo polyclonal activation
without apparent triggering of their BCR.
I. Antibody production by polyclonally activated B cells. The hypothesis
proposed based on work done by Hunziker et al states that hypergammaglobulinemia is
the result of an infection that grows to high titers in vivo (Hunziker et al., 2003). The high
antigenic burden allows B cells, regardless of BCR specificity, to take up viral antigens
and present them to virus-specific CD4 T cells, which, in turn, activate Ab production in
a CD40-dependent manner. Thus, the antigen presenting B cells that become polyclonally
activated following adoptive transfer into virus-infected hosts should be able to secrete
antibodies. I, therefore, tested whether the polyclonally activated GP61-presenting donor
B cells would secrete Ab. For these assays, I transferred GP61-coated and CFSE-labeled
congenic antibody-allotype-mismatched (IgHa) donor B cells into LCMV-infected or
uninfected Ly5.1 mice, which are IgHb. These donor B cells express a distinct epitope on
their immunoglobulin molecules that allows for their detection by ELISA separate from
host immunoglobulin. Seven days following transfer, the transferred IgHa donor B cells 104
displayed similar proliferation profiles as C57BL/6 and αβ TCR KO donor B cells (Table
IV-1). Moreover, the cells coated with GP61 but not HA209 peptide secreted Ab, as IgMa
was detected in the serum of LCMV-infected hosts that received GP61-coated
splenocytes (Fig. IV-13A). IgMa Ab was not detected from the B cells implanted into
uninfected mice, into hosts depleted of CD4 T cells, nor in hosts treated with anti-CD154
(Fig. IV-13B). IgMa ASC were also detected with ELISPOT analyses but only when
GP61-coated donor B cells were transferred into LCMV-infected hosts (Fig. IV-13C).
The polyclonal activation of GP61-coated B cells following LCMV infection was therefore not limited to proliferation but also included Ab secretion.
I next sought to determine if polyclonal activation of LCMV peptide-coated B cells would result in immunoglobulin class switching. Again, IgHa allotype-mismatched
spleen cells were coated with LCMV or influenza peptides or left uncoated and then
transferred into LCMV-infected and uninfected mice at day 10. Serum was harvested
seven days later and tested for the presence of IgG2aa Ab by ELISA. GP61- and NP309-
coated B cells produced IgG2aa Ab in LCMV-infected hosts but not in uninfected hosts
(Fig. IV-14, upper panel). Unlike the results obtained for IgM secretion, one of two mice
receiving uncoated IgHa B cells also had detectable levels of IgG2aa but these Ab levels
were lower than those produced by GP61-coated cells. It is unclear why uncoated B cells
that were not induced to divide or change phenotype would be induced to secrete Ab.
One could speculate that Ab secretion by uncoated B cells could occur due to in vivo
antigen uptake during the LCMV infection, similar to the proliferation of B cells taken
from LCMV carrier mice which were also antigen labeled in vivo. This model would 105 106
107
predict that polyclonal B cell activation can induce Ab production without proliferation.
It is also interesting to note that IgG2aa ASC were undetectable by ELISPOT analysis of
cells isolated from the spleens of LCMV-infected mice that had received GP61-coated B
cells. This result may indicate a homing of IgG secreting cells to another location such as
the bone marrow.
In light of these results I questioned whether these polyclonally activated IgG2a class-switched B cells home to the bone marrow, as has been previously described for virus-specific plasma cells (Slifka and Ahmed, 1996b). I therefore performed a similar adoptive transfer of antigen-coated B cells from the spleen into LCMV-infected or uninfected hosts and tested bone marrow for the presence of ASC. Bone marrow was harvested 10 days following adoptive transfer. ELISPOT analysis revealed that GP61- coated B cells isolated from bone marrow of LCMV-infected mice secreted both IgG2aa and IgMa (Fig. IV-14, lower panel). In this experiment uncoated B cells did not produce
Ab in LCMV-infected or uninfected mice. However, low levels of IgG2aa production could be detected when GP61-coated B cells were transferred into uninfected mice or when HA209-coated B cells were transferred into LCMV-infected mice. Neither of these treatments resulted in IgMa production and, in both cases, the IgG2aa production was less
than half of the IgG2aa secreted by GP61-coated B cells in LCMV-infected hosts. The
highest numbers of IgG-secreting cells were detected when GP61-coated B cells were
transferred into LCMV-infected mice. These results show that the polyclonally activated
population of B cells in virus-infected mice can mount a class-switched Ab response
which can be detected by both serum Ab and ASC assays. 108
J. Polyclonal activation of a transgenic B cell population. In the previous sections of this chapter I described the virus-induced activation of a polyclonal population of B cells. I have shown that when these B cells are loaded with virus-specific
MHC class II epitopes and transferred onto virus-infected hosts, they proliferate, change phenotype, and secrete Ab, just like B cells that are conventionally activated by specific signals through their BCR. It would be interesting to determine in future studies if some of the fine tuning that occurs when B cells are conventionally activated, such as SHM, also occurs during this polyclonal activation of B cells. Furthermore, the polyclonal activation of B cells has been hypothesized to result in the maintenance of memory B cells (Bernasconi et al., 2002). It would be interesting to determine in future studies if antigen-expressing memory B cells can become polyclonally activated following transfer into virus-infected hosts. Both questions have various technical challenges.
The detection of SHM in B cells presents its own challenges. SHM is generally detected by sequencing the variable regions of activated B cells. This requires the identification of the DNA sequence of the germ-line-encoded antigen binding region of the BCR (Jacob et al., 1991). Changes that occur in the variable region sequence during
SHM are monitored by DNA sequencing of B cells that are isolated after antigen has been introduced into the system. An alternative to this method of SHM detection is to have a B cell population whose antigen is known and whose germ-line-encoded variable region can be detected with an anti-idiotype Ab. With these, SHM could be measured by the ability of a B cell to bind its antigen but lose reactivity with the anti-idiotype Ab after 109
an immune response, suggesting that the variable region has changed from the germ-line- encoded sequence.
Other limitations make it difficult to detect memory B cell involvement in virus- induced polyclonal activation. Memory B cells in mice cannot unambiguously be detected by flow cytometry; there is no way to determine if memory B cells are included in the proliferating population. Methods to detect memory B cells have relied on their ability to respond rapidly to secondary antigen stimulation but a rapid response may or may not occur after polyclonal activation. The frequency of memory B cells generated after an LCMV-specific immune response has been determined to be 1 in 20,000 spleen cells, a frequency that would be difficult to detect in an adoptive transfer system (Slifka and Ahmed, 1996a).
A transgenic B cell population may provide the necessary requirements to answer whether SHM is occurring and whether memory B cells are involved in the polyclonal activation of B cells after virus infection. This transgenic system would provide a large population of B cells with known specificity and known BCR variable region. Thus,
SHM could be detected by changes in the variable region of the BCR and the polyclonal activation of memory B cells could be measured due to the high frequency of memory B cells that would be found in transgenic mice after immunization with cognate antigen. I therefore performed preliminary studies to test for LCMV-induced activation of a BCR transgenic B cell population. This BCR transgenic B cell recognizes a multivalent epitope
(T,G)A-L (Giorgetti and Press, 1994). (T,G)A-L is a synthetic peptide antigen which is a polymer of four amino acids: tyrosine, glutamic acid, alanine, and lysine. The (T,G)A-L 110
transgenic mouse is the result of an F1 cross between (T,G)A-L-specific heavy chain
VHSm7 and light chain Vκ1AJκ2 knock-in parents. Eighty percent of the B cells in F1
mice recognize (T,G)A-L, and they can be identified with a mAb, 8A7, that recognizes
the variable region of the (T,G)A-L-specific BCR. To test if these transgenic B cells
could undergo polyclonal activation in LCMV-infected mice, I coated splenocytes from
the transgenic F1 mice with GP61 and HA209 peptide and transferred these cells into
LCMV-infected or uninfected B6.129 hosts. Two weeks after adoptive transfer, serum, spleen, and bone marrow were harvested and analyzed for donor cell division and anti-
(T,G)A-L Ab production. As seen with every polyclonal donor B cell population tested, only GP61-coated (T,G)A-L B cells divided in LCMV-infected host as defined by
B220+, CD19+, and 8A7+ cells (Fig. IV-15). No other treatment resulted in donor cell division. Anti-(T,G)A-L antibodies were detected at highest levels in the sera of LCMV- infected mice that received GP61-coated B cells. Lower levels of anti-(T,G)A-L antibodies could be found in the sera of uninfected and LCMV-infected mice that
received HA209-coated cells or uninfected mice that received GP61-coated cells (Fig.
IV-16A). Anti-(T,G)A-L ASC numbers followed a similar pattern. High numbers of anti-
(T,G)A-L ASC were detected by ELISPOT in the spleen and bone marrow of mice
receiving GP61-coated B cells. Lower numbers of ASC were detected in uninfected and
LCMV-infected mice that received HA209-coated cells or uninfected mice that received
GP61-coated cells (Fig. IV-16B and C). In summary, an antigen-presenting transgenic B
cell population recognizing (T,G)A-L can be activated by LCMV in the same manner as
a polyclonal population of B cells. 111 112
113
The transgenic population of B cells will be useful to answer questions about
SHM and memory B cell involvement in virus-induced polyclonal activation. However, the results obtained with the transgenic B cell population also lend evidence to the fact that the virus-induced polyclonal activation of B cells is BCR independent. Using a polyclonal population of B cells, the high frequency of B cells that became polyclonally activated to proliferate and secrete Ab suggested that BCR specificity was not driving
LCMV-induced polyclonal activation. There still remained the possibility that a conserved germ-line-encoded BCR motif may have low affinity interactions with viral antigens, as has been described for VSV(Pinschewer et al., 2004). Although no such interaction has been described for LCMV, these low affinity interactions could potentially drive polyclonal B cell activation. However, when one considers that the
(T,G)A-L transgenic B cells were derived from a B cell that responded to a synthetic peptide antigen with repeat of four amino acids (poly-[tyrosine, glutamic acid] poly- alanine, poly-leucine), it makes it difficult to believe that the (T,G)A-L-specific BCR could be cross-reactive with LCMV viral antigens or that BCR specificity plays some role virus-induced polyclonal B cell activation. Note that (T,G)A-L transgenic B cells do not divide unless they are coated with viral antigens. If (T,G)A-L-specific BCR were reactive with LCMV, all (T,G)A-L transgenic B cells would divide during infection.
Therefore, the polyclonal activation of (T,G)A-L transgenic B cells indicates that virus- induced polyclonal B cell activation occurs regardless of BCR specificity.
114
K. Summary of Chapter IV. In this chapter I describe the fate of the antigen-
presenting B cells that survive the MHC class II-restricted killing that was described in
chapter III. These B cells presenting viral epitopes undergo a virus-induced polyclonal activation that is dependent on CD4 T cells and CD40-CD40L interactions. I provide a detailed analysis of the proliferation, phenotypic changes, and Ab production by these B cells. LCMV GP61- and NP309-coated B cells proliferate in LCMV-infected mice but not in uninfected mice. Uncoated B cells, B cells coated with LCMV MHC class I antigens, or B cells coated with an irrelevant influenza virus epitope do not divide in
LCMV-infected animals. As the GP61-coated B cells divided, their phenotype changed in a manner characteristic of conventionally activated B cells that are entering a GC reaction. These GP61-coated B cells were stimulated to secrete antibodies in LCMV- infected mice. The antibodies generated by polyclonally activated B cells were both IgM and class-switched IgG. Uncoated or HA209-coated B cells, which did not divide or change phenotype, did not produce the Ab to the same extent as polyclonally activated
GP61-coated B cells. CD4 T cell depletion or CD40-CD40L blockade resulted in the absence of both proliferation and Ab production by GP61-coated B cells. B cells from
LCMV carrier mice, where LCMV grows to high titers due to immunologic tolerance, undergo polyclonal activation in acute LCMV-infected hosts but not in uninfected hosts.
Thus, the polyclonal activation of B cells shown in chapter IV complements the
hypothesis proposed by Hunziker et al. and Recher et al. to explain the
hypergammaglobulinemia caused by LCMV infection. My adoptive transfer study
features a polyclonal population of B cells which present viral antigens regardless of 115
BCR specificity, it affirms the requirement for CD4 T cells and CD40L, it shows that the
viral antigen-presenting B cells secrete antibodies, and it shows that B cells, under
conditions of high viral load, take up and present viral antigens. However, unlike other
studies which focused mainly on the hypergammaglobulinemia itself, the work presented in this chapter examines the B cells that are responsible for hypergammaglobulinemia.
By focusing on the B cells that are responsible for hypergammaglobulinemia, specific questions could be asked about the mechanisms involved in virus-induced polyclonal activation. In these studies I found that MyD88-dependent signaling, Type I
IFN signaling and IFN-γ signaling are not required for polyclonal B cell activation. I also
show that polyclonal B cell activation can occur in a transgenic B cell population whose
BCR is not crossreactive with LCMV. The polyclonal activation of this transgenic
population shows that virus-induced polyclonal B cell activation is BCR-independent and
these transgenic B cells will undoubtedly be of great use in future studies of virus-
induced polyclonal B cell activation. 116
CHAPTER V
DISCUSSION
This thesis examines some of the interactions that occur between CD4 T cells and
B cells during a viral infection. I have described the existence of CD4-dependent killing of targets in vivo and the existence of polyclonal activation of B cells during LCMV infection. These two paradoxical fates occur as the result of BCR-independent viral antigen presentation on B cells. I used in vivo cytotoxicity assays that featured antigen- presenting B cells as targets to show that MHC class II-restricted killing can occur in vivo after viral infection. MHC class II-restricted killing is CD4-dependent and specific to B cells that present cognate viral antigens. CD4 T cells use Fas-FasL interactions and perforin as mechanisms for target cell lysis. I have also been able to characterize the dynamics and magnitude of a virus-induced polyclonal B cell activation through the use of a novel adoptive transfer system that also featured antigen-presenting B cells.
Polyclonal B cell activation is CD4- and CD40L-dependent and specific to B cells that present cognate viral antigens. Virus-induced polyclonal B cell activation during LCMV infection was hypothesized by others during my thesis studies to cause hypergammaglobulinemia (Hunziker et al., 2003). My studies, which focused on the polyclonal activation of B cells rather than the phenomenon of hypergammaglobulinemia, have definitively shown the existence and requirements for virus-induced polyclonal B cell activation and also the phenotypic, proliferative, and 117
functional changes that polyclonally activated B cells undergo. These dynamic
interactions that can occur between CD4 T cells and B cells during virus infections were
not fully appreciated previous to the work presented in this thesis.
A. The cytotoxic CD4 T cell response to LCMV. The results in chapter III show
that CD4-dependent MHC class II-restricted killing occurs in vivo. The ability to lyse
virus-infected cells is a function that has primarily been described for CD8 T cells. This
cytotoxic function had been described in CD4 T cells but only in vitro and usually only
following in vitro culture (Jacobson et al., 1984; Lukacher et al., 1985; Maimone et al.,
1986; Hou et al., 1993). One fate of the adoptively transferred antigen-presenting B cell
population used in this thesis was an MHC class II-restricted elimination by CD4 T cells
in vivo. By taking advantage of the LCMV system and the robust CD4 T cell response it induces, I was able to show by in vivo cytotoxicity assay the first evidence of MHC class
II-restricted killing in vivo during viral infection.
In vivo cytotoxicity assays using GP61- and NP309-coated target populations showed that only LCMV-specific cells are killed and only in LCMV-infected mice. The specific killing of LCMV class II peptide-coated target cells mirrored the dominance hierarchy of cytokine production induced by these two peptides, where GP61 induced more specific lysis than NP309. Killing was also restricted to MHC class II high cells and not MHC class II low cells. These results identify MHC class II high cells that present virus-specific epitopes as the targets for in vivo cytotoxicity. 118
The frequency of CD4 T cells that can secrete cytokines in response to LCMV
peptide stimulation may correlate to the frequency of CD4 T cells that can lyse antigen presenting targets. GP61-coated target cells were killed at a greater frequency than
NP309-coated cells (Table III-1). The frequency of effector CD4 T cells detected by intracellular cytokine staining in previous studies showed a similar dominance of the
GP61 response over the NP309 response (Varga and Welsh, 1998a). GP61 is a more
stimulatory epitope than NP309 for both the induction of CD4 CTL and cytokine
secretion. However, this correlation between the frequency of cytokine-producing cells
and cytotoxic CD4 T cells is not absolute. While GP61 remains dominant to NP309,
specific lysis of antigen-coated targets was seen both at the peak of the CD4 cytokine
response at day 9 of infection and several days later at day 14 post-infection, when the
frequency of cytokine-producing CD4 T cells is reduced. One would think that the most
killing would occur when the CD4 effector frequency was highest, but perhaps cytotoxic
CD4 T cell frequency peaks at a different time post-infection than the cytokine response.
These results may reflect a change in phenotype of CD4 T cells from cytokine producers
to CTL or they may indicate that cytotoxic CD4 T cells and cytokine-producing CD4 T
cells are two distinct subsets.
CD4 CTL have been found in many model systems. In most cases, cytotoxic CD4
T cells were only found in CD4 T cell lines or from isolated cells that had experienced
simulation through in vitro culture (Bourgault et al., 1989). Eventually cytoxicity was
seen directly ex vivo (Erb et al., 1990; Appay et al., 2002). Previous studies of MHC
class II-restricted killing often found that CD4-mediated killing would only occur in the 119
absence of CD8 T cells, as if the CD4 CTLS were compensating for the lack of CD8 CTL
(Hou et al., 1993; Williams and Engelhard, 1997). In the LCMV system, cytotoxic CD4
T cell activity could be detected in vitro using splenocytes isolated from i.p. infected
mice deficient in CD8 T cells because they lack MHC class I expression (Muller et al.,
1992). Intracranial (i.c.) LCMV infection in these mice led to the death of many infected
mice and a profound weight loss in the surviving animals (Quinn et al., 1993). CD4
depletion prior to i.c. infection prevented death and weight loss in these MHC class I
deficient animals. Death induced by i.c. infection with LCMV in WT animals is mediated
by CD8 CTL and the authors of this study suggested that in MHC class I KO mice, CD4
T cells may substitute for CD8 CTL. Adoptive transfer of increasing numbers of LCMV-
primed CD4 T cells derived from these CD8-deficient mice induced a higher incidence of
death in i.c. infected mice. This result suggested a causal relationship between cytotoxic
CD4 T cells and disease in these CD8-deficient mice (Quinn et al., 1993). Similar to my results, the CD4 CTL isolated from LCMV-infected MHC class I KO mice relied on Fas-
FasL interactions to kill targets in vitro (Zajac et al., 1996). Inhibitors of Fas prevented in
vitro lysis of target cells by CD4 T cells isolated from LCMV-infected MHC class I KO
mice. Furthermore, CD4-mediated death of i.c. infected mice did not occur if the MHC
class I KO mice were also deficient in Fas. It is noteworthy that these studies and my
study in vivo both found cytotoxic CD4 T cell activity that relied on Fas-FasL
interactions to eliminate targets. In my system, CD4 T cells were capable of killing in the
presence of a normal CD8 T cell compartment (Fig. III-1 and 2). Using in vivo Ab
depletion I found that CD4 cells were necessary and that CD8 T cells were not required 120
for MHC class II-restricted killing (Table III-1). This finding was critical because the
GP61 polypeptide has a subdominant MHC class I H2Kb-restricted epitope GP70 within
its sequence (Masopust et al., 2007).
In many of my assays, B cells were used as the target cell for MHC class II-
restricted killing. LCMV does not infect a high frequency of B cells, but one wonders if
MHC class II-restricted killing of B cells is an evolutionarily conserved mechanism to kill virus-infected B cells or other MHC class II presenting cells. Some viruses, most notably EBV, can infect B cells and CD4 CTL may be partially mediating clearance of
this viral infection. Lending evidence to this theory, EBV-specific CD4 CTL can be
generated from the blood of EBV seronegative donors when co-cultured with EBV-
infected B cells (Savoldo et al., 2002). The generation of CD4 CTL from seronegative
donors suggests that CD4 CTL may have mediated clearance of EBV in these patients.
The importance of this CD4–dependent cytotoxicity remains unclear, but the studies
showing requirements for CD4 T cells or for class II MHC in the control of viral
infections in vivo must now take this cytotoxic function into account.
B. Mechanisms by which CD4 T cells kill their targets. MHC class II-restricted
killing in vivo cannot be explained solely by one mechanism. This was also observed in previously published in vitro studies of MHC class II restricted killing (Tite, 1990;
Stalder et al., 1994; Williams and Engelhard, 1996). Here I identify Fas/FasL and
perforin as mechanisms responsible for killing. MHC class II-restricted killing in FasL
mutated gld host mice occured but the killing of targets was at lower frequency than in 121
wild type mice (Table III-1). Importantly, I found that the number of cytokine-producing
CD4 effector T cells was not reduced in LCMV-infected gld mice, a result which suggested that the reduction in killing in these mice was not due to differnces in virus- specific CD4 T cell number (Fig. III-3). The reciprocal assay, using Fas mutant lpr
targets in direct competition with WT targets, showed that Fas expression on the target
cells was necessary for optimal lysis (Table III-1). Thus, Fas-FasL interactions were
identified as a mechanism used by CD4 T cells for MHC class II-restricted cytotoxicity in vivo. This result was perhaps not surprising because Fas-FasL interactions were reported
to be the mechanism used for lysis by CD4 cell lines (Janssens et al., 2003).
Several studies using in vitro derived CD4 CTL have identified perforin as the
major mechanism for MHC class II-restricted cytotoxicity (Grossman et al., 2004a;
Williams and Engelhard, 1997; Williams and Engelhard, 1996; Yasukawa et al., 1993).
In agreement with these in vitro studies, I found no evidence of MHC class II-restricted
cytotoxicity in vivo in LCMV-infected perforin KO mice. The in vivo cytotoxicity assays
performed in perforin KO mice showed no specific lysis of the GP61-coated population,
even when P14 transgenic T cells were used to reduce viral loads. However, preliminary
data generated in LCMV-infected perforin KO mice that received P14 transgenic CD8 T
cells revealed that these mice could induce polyclonal B cell activation. I originally
hypothesized that, due to high Ag loads in perforin KO mice, CD4 T cells would be
engaged with LCMV antigens and would not get the opportunity to encounter the transferred target cell population. However, because there is sufficient T cell help to induce polyclonal B cell activation, it appears that CD4 cells in perforin KO mice can 122
interact with the transferred target cell population. Thus, CD4-mediated killing in vivo is
likely to include a role for perforin. The caveat remains that perforin KO CD4 T cells
may require more time to encounter and kill B cell targets and specific lysis may only be detectable after several days in vivo. However, this elimination would be masked by the onset of division by polyclonal activation. Others have shown that CD4 T cells lose their ability to secrete IL-2 in LCMV-infected perforin KO mice, but cytotoxicity was not tested (Fuller and Zajac, 2003). It is possible that CD4 T cells also lose cytotoxic ability
in perforin KO mice as a result of high antigen loads and clonal exhaustion. Despite these
caveats, the fact that perforin KO CD4 cells can mediate polyclonal B cell activation but
not B cell killing strongly suggests that these CD4 T cells are encountering B cell targets
and are functional in vivo. Since my initial report on MHC class II-restricted cytotoxicity
in vivo during viral infection, other groups have also published their own work on MHC
class II-restricted killing in vivo (Santosuosso et al., 2005; Brown et al., 2006). Studies
of antiviral CD4 CTL in influenza infection have also identified perforin as the
mechanism of killing (Brown et al., 2006). Based on these results and my own studies, I
conclude that perforin and Fas-FasL interactions are both used by CD4 T cells for MHC
class II-restricted killing in vivo during virus infection.
When the B cell targets that survived the CD4 dependent cytotoxicity were
stained for Fas expression, I found that the remaining GP61-coated target B cell
population expressed high levels of Fas. Elevated Fas was not seen on HA209-coated B
cells or GP61-coated B cells in uninfected mice (Fig. III-4). I had previously observed
that Fas-FasL interactions were partially responsible for MHC class II-restricted killing, 123
and, therefore, I did not expect to find surviving LCMV antigen-labeled target B cells with high Fas expression. Nevertheless, these Fas-expressing B cells were not eliminated.
These results may indicate that some antigen-presenting B cells had not yet encountered a
CD4 T cell with cytotoxic activity or they may indicate that some B cells can escape Fas- mediated killing. The expression of Fas without the induction of apoptosis has been seen in GC B cells which up-regulate Fas but do not require Fas expression to undergo apoptosis (Smith et al., 1995). In chapter IV I show that the B cells which survive the
MHC class II-restricted cytotoxicity become polyclonally activated and display a phenotype consistent with entry into GCs (Fig. IV-2). The antigen-presenting B cells that survive the CD4-dependent cytotoxicity may be protected due to their polyclonal activation and entry into GC reactions. This hypothesis is supported by evidence that human GC CD4 T cells have been shown to suppress Th1 cell responses in vitro. This in vitro suppression seemed to be mediated by Fas receptor ligation on the Th1 cells and also cytokine-mediated suppression via IL-10 and TGF-beta secretion by GC CD4 T cells
(Marinova et al., 2007). If these GC T cells are also able to suppress the cytotoxic CD4 T cell response in LCMV-infected mice, it would suggest that entry into GC as a result of polyclonal B cell activation prevents lysis of those B cells. Thus, the MHC class II- restricted killing of B cells may occur in a lymphoid compartment or location distinct from polyclonal B cell activation. In addition to protection conferred by GC Treg-type cells, other factors may protect B cells from MHC class II-restricted killing. For CD8- mediated killing, protection of APC target cells from granule exocytosis has been described where target dendritic cells that express serpin serine protease inhibitor 124
molecules can be protected from specific lysis by CD8 T cells (Medema et al., 2001; Bots et al., 2005). Serpin molecule expression in GC B cells has also been reported (Frazer et al., 2000; Paterson et al., 2007). It is unclear whether or not these serpins could prevent
CD4-dependent cytotoxicity but it remains a possibility and could potentially explain why some antigen-presenting B cell targets resist cytotoxicity.
Production of IFN-γ and TNF by CD4 T cells does not appear to mediate MHC class II-restricted killing, because targets deficient in TNFR1 and IFN-γ receptors are killed at similar frequencies as WT targets when they present viral epitopes in LCMV- infected animals (Table III-1 and 2). Based on these results it can also be concluded that neither of these receptors provides survival signals to B cells that prevent lysis by CD4 T cells. Similar results were obtained when CD40KO B cells were used as targets for MHC class II-restricted cytotoxicity (Table III-2). Although CD40 signaling was crucial for the polyclonal activation of B cells (Fig IV-8B), CD40 receptor signaling does not appear to provide a survival signal for B cell targets that prevents in vivo cytotoxicity.
CD40KO target B cells were incapable of upregulating Fas but were still eliminated in in vivo cytotoxicity assays (Fig. III-5 and Table III-2). The observation that
Fas was not up-regulated is perhaps not surprising, because CD40 signaling is required for the formation of GC, the site where B cells up-regulate Fas expression(MacLennan,
1994a). However, if GC B cells are somehow protected from MHC class II-restricted killing, as hypothesized earlier, then CD40 KO B cells, which cannot form GCs, should be more susceptible to lysis. My results show that CD40 KO B cells are not more susceptible than WT B cells to MHC class II-restricted killing (Table III-2). Furthermore, 125
CD40 KO B cells cannot undergo polyclonal B cell activation (Fig. IV-8B). Combined, these results suggest that polyclonal B cell activation does not induce the expression of anti-apoptotic molecules that can prevent CD4 killing. Other unknown factors must, therefore, be involved in the B cell fate decision.
As described above, CD4 CTL activity was detected in MHC class I-deficient animals infected with LCMV (Quinn et al., 1993; Zajac et al., 1996). Other studies using
LCMV GP61-specific SMARTA Tg CD4 T cells have shown that these transgenic CD4
T cells have cytotoxic ability ex vivo in chromium release assays (Oxenius et al., 1998a).
This cytotxicity only occurred when the targets expressed MHC class II with bound
GP61. Similar to my results, effector CD4 T cells isolated at day 8 and day 13 post- infection both displayed similar lytic activity ex vivo. These SMARTA transgenic CD4 T cells, therfore, could be used to alter the frequency of CD4 T cell precursors to determine if cytotoxic function correlates with the number of CD4 T cells. A correlation would indicate that T cells, and not the B cells themelves, control the fate of antigen-presenting
B cells during viral infection. Additionally, if two types of CD4 T cells (killers and helpers) can be defined phenotypically and separated, these SMARTA cells would be of great use in future studies to possibly generate responses that consist entirely of either
CD4 T cell-mediated killing or polyclonal B cell activation.
Based on previous in vitro studies of CD4 killing, it was speculated that cytotoxic
CD4 T cells may play a role in the regulation of immune responses (Hahn et al., 1995).
These studies proposed that CD4 CTL may be a type of Treg used to dampen the immune response. Similar to my results in LCMV infection, perforin and FasL have been shown 126
to be the effector mechanisms used by Treg-phenotype cells for specific lysis of B cells,
allogeneic targets, and DCs (Janssens et al., 2003; Grossman et al., 2004a; Grossman et
al., 2004b; Kawamura et al., 2006). Because CD4 killing can occur late in infection
during the silencing or contraction phase at the end of an immune response, it could be speculated that CD4 CTL may provide a mechanism that eliminates MHC II high cells
that are presenting viral antigens. Elimination of these cells would effectively stop CD4 T
cell stimulation and may lead to the silencing of a CD4 T cell response. In the GC
reaction, antigen contact provided to B cells by FDC is required for the BCR signaling
that maintains B cell turnover (MacLennan, 1994a). Perhaps CD4 CTL can eliminate
these FDC or B cells, thus terminating GC reactions during the silencing phase of the
immune response. However, results described earlier would suggest that GC CD4 T cells might prevent cytotoxicity in the GC (Marinova et al., 2007). CD4 CTL activity is also detected during the peak of the immune response that is associated with a transient period of immune deficiency in LCMV-infected hosts. This correlation makes it tempting to speculate that CD4 CTL may play some role in AICD of Fas-expressing CD8 T cells
(Nagata, 1997). In this case CD4 CTL would simply provide FasL to the AICD sensitive
CD8 T cells.
C. Analysis of the B cell population that undergoes polyclonal activation after LCMV infection. In chapter IV I demonstrate that, in the context of a viral infection in vivo, B cells presenting viral class II epitopes can be driven to proliferate and secrete Ab without any apparent viral engagement of the BCR. Because such a high 127
proportion of the B cells proliferate, I conclude that either the BCR is not stimulated at all
or else it is stimulated by very low affinity endogenous or self antigens. It appears that
sufficient help and stimulatory signals are provided by helper CD4 T cells and CD40-
CD40L interactions to drive the B cell activation and differentiation. This contrasts with
the classical models of T cell-dependent B cell activation, where the BCR is engaged by
viral antigens and where this interaction leads to endocytosis of the viral proteins, which
are processed and presented as peptides in the context of class II MHC molecules. In the
classical model, CD4 T cells recognize and provide help to these virus-specific B cells,
which get antigen-specific stimulation through their BCR, and the B cells then proliferate and differentiate into plasma cells and memory cells (Rajewsky, 1996).
Conditions transpiring in a virus-infected host may allow B cells that are not virus-
specific to present viral antigen and become activated without a strong BCR signal. This
mechanism has been suggested for a presumed virus-induced polyclonal B cell activation
as detected by hypergammaglobulinemia (Hunziker et al., 2003), but in chapter IV I
formally describe phenotypic changes in naïve B cells and demonstrate high levels of
activation and proliferation in vivo in the absence of apparent BCR stimulation.
The findings in chapter IV clarify a potential discrepancy between the MHC class
II-restricted killing described in Chapter III and work by Hunziker et al., who suggested
that B cells could be passively charged with viral antigens and be stimulated by cognate
CD4 T cells to produce Ab (Hunziker et al., 2003). A consequence of this phenomenon was a hypergammaglobulinemia induced by high dose LCMV infection, where antigen was plentiful (Hunziker et al., 2003). Seemingly at odds with that finding was my report 128
in chapter III showing that peptide-expressing B cells were killed by CD4 T cells during
LCMV infection. I report in chapter IV that, on further examination, there is a residual population of B cells that escape CD4 T cell-mediated cytotoxicity, and that these remaining B cells undergo a polyclonal activation, as shown by CFSE-loss proliferation assays, phenotypic changes consistent with entry into GCs, and Ab secretion. Although I have not clarified why some B cells die and others proliferate, my demonstrations of the proliferation and differentiation of B cells complement the conclusions of Hunziker et al, whose main focus was on the production of Ab. Both studies showed that B cells presenting class II antigens can be activated by CD4 T cells in a CD40-CD40L dependent manner without a defined involvement of the BCR. A broad based polyclonal activation of B cells was proposed by Hunziker et al, and my studies show that most of the B cells that escaped cytotoxicity can undergo proliferation, a result inconsistent with a role for the BCR, where some B cells would be expected to proliferate more so than others. My study also shows that without LCMV-specific class II peptide presentation, there is absolutely no B cell division, indicating that cytokines alone during an infection will not induce B cells to divide. Thus, my experimental strategies not only demonstrate the phenomenon of polyclonal B cell activation, but also track in vivo the differentiation and proliferation of the B cells that are being polyclonally activated. The potential significance of this polyclonal activation is documented in the introduction to this thesis and by studies of others (Bernasconi et al., 2002; Hunziker et al., 2003; Traggiai et al.,
2003; Ludewig et al., 2004; Recher et al., 2004). Some examples of the potential significance of virus-induced polyclonal B cell activation include the maintenance of 129
memory B cells, the induction of self-reactive B cells, and the distraction of CD4 T cells from virus-specific B cells resulting in the delay of neutralizing Ab formation.
D. The antibodies produced during virus-induced polyclonal B cell activation. Virus-induced polyclonal B cell activation resulted in the production of IgM and IgG2a. The production of IgM is perhaps not surprising because IgM responses are thought to be more innate and may result from cytokine stimulation that activates the more innate-like B1 and MZ B cells. However, this IgM response produced in my system of polyclonal B cell activation was T cell-dependent and did not occur in the absence of
CD4 cells or CD40 signaling (Fig. IV-9B and C). It should be noted that the IgM was produced only by GP61-coated B cells and only in LCMV-infected hosts. These data rule out the possibility that cytokine-activated innate-like B cells are responsible for the IgM production.
The hypergammalobulinemia detected following high dose infection in other studies primarily consisted of IgG (Hunziker et al., 2003). In my system, I was able to detect IgG production from a polyclonal population of IgHa allotype-mismatched B cells as well as from (T,G)A-L Tg B cells that were polyclonally activated during LCMV infection(Figs. IV-14 and IV-16). Based on these results, the peptide-coated B cell population that I transferred into hosts would be able to contribute to virus-induced hypergammaglobulinemia.
The fact that Tg B cells can divide and secrete Ab also provides evidence that
BCR specificity does not play a role in this virus-induced polyclonal B cell activation. 130
A caveat to using transgenic B cells specific to (T,G)A-L is that the level of Ab produced
by the transgenic B cells may be more affected by MHC class II-restricted killing than a
polyclonal population of B cells. Preliminary data using (T,G)A-L transgenic B cells as targets for in vivo cytotoxicity assays revealed that the transgenic population is more
susceptible to lysis by CD4 T cells than a polyclonal population of B cells. Thus, if cell number correlates with Ab production, the levels of Ab produced by the transgenic
population will never reach the levels produced by polyclonal IgHa B cells. Despite this caveat, this transgenic B cell system will provide a valuable source of donor B cells to explore SHM and memory B cell involvement in future studies of virus-induced polyclonal B cell activation
IgG2a production by the polyclonally activated B cell population was expected, because IFN-γ is the switch factor for IgG2a and IFN-γ is produced in abundance during
LCMV infection (Gessner et al., 1990). However, Figures IV-14 and IV-16 show that
IgG was also produced by B cells that did not present LCMV epitopes. These B cells did
not divide in LCMV-infected hosts but seemed to be able to produce antibodies, unlike
GP61-coated B cells which proliferated and produced antibodies. It could be speculated
that these B cells were activated by another mechanism such as TLR engagement, and the
IFN-γ produced during the LCMV infection led to the enhancement of IgG production
from these B cells, as has been previously reported in vitro (Snapper et al., 1988). This
was not the case with IgM production but IgM production is not enhanced by IFN-γ
signaling and TLR simulation (Snapper et al., 1988). Another alternative explanation is
that B cells can take up viral antigens in vivo during acute LCMV infection and this leads 131
to Ab production. Results from the polyclonal activation of B cells taken from
congenitally-infected LCMV carrier mice upon transfer into acute LCMV-infected hosts
provide evidence that in vivo antigen uptake can occur and lead to polyclonal activation.
It is important to note that in all cases where donor B cell IgG production was detected, the highest levels of IgG in the serum and the highest number of IgG secreting cells were found in LCMV-infected host mice that received GP61-coated B cells(Figs. IV-14 and
IV-16), and in all cases, these GP61-coated B cells had undergone more proliferation than any other donor B cell type (Table IV-1).
E. Implications of virus-induced polyclonal B cell activation. B cell activation is normally thought to require signals from the BCR, from CD40, and from cytokines secreted by activated CD4 T cells that recognize peptides displayed on B cell class II
MHC molecules. B cell activation can, however, occasionally be driven through less conventional BCR-independent pathways such as through toll-like receptor (TLR) ligation or direct MHC class II signaling (Chen et al., 1987; Pasare and Medzhitov, 2005;
Nashar and Drake, 2006). Although B cells which capture antigens with their BCR present peptides on class II molecules more efficiently, there is no barrier preventing B cells of other specificities to passively absorb and cross-present viral class II antigens, especially under cases of high viral antigen loads or persisting viral infections. Viral infection of B cells with irrelevant BCR specificities could also be a mechanism for B cell presentation of class II antigens. 132
Polyclonal activation of B cells without BCR triggering has been reported in
vitro with human B cells stimulated by TLR agonists or with CD4 T cell help(Bernasconi
et al., 2002), but my observation in vivo shows that TLR signaling through MyD88 was
not required within the B cell. It should be emphasized that polyclonal activation could
only be seen when B cells were presenting cognate viral antigen and that no other treatment ever resulted in any B cell division.
For significant levels of virus-induced polyclonal B cell activation in vivo, there should either be a high enough antigen load to allow for the presentation of viral peptides,
albeit probably inefficient, on the B cells or, alternatively, the infection of a high
frequency of B cells by the virus. LCMV infects B cells poorly, so presentation of
antigen from the high dose antigen environment may be a preferred mechanism in this
system (Buchmeier et al., 1980). Neutralizing Ab responses to LCMV are poor under
conditions of high dose inocula (Hunziker et al., 2003), and there may be a reduced
efficiency of CD4 T cells providing help to the virus-specific B cells due to the distraction of the T cells by other B cells cross-presenting antigen. It is noteworthy also
that an LCMV peptide (GP61) that induces a higher frequency CD4 T cell response
elicits more polyclonal activation of B cells than a peptide (NP309) that induces a lower
frequency CD4 T cell response (Fig. IV-1), implying that CD4 T cell help can be
limiting for both the BCR-non-specific as well as the BCR-specific B cell activation.
Hypergammaglobulinemia is associated with a number of human viral infections,
such as HIV, HCV, and EBV (Burgio and Monafo, 1983; Hutt-Fletcher et al., 1983; Hu
et al., 1986; Martinez-Maza et al., 1987; Rosa et al., 2005). All of these infections can 133
lead to high antigen loads. In the unique case of EBV infection, the direct virus-induced transformation of B cells (Kuppers, 2003) might be the cause for polyclonal B cell stimulation, but CD4 T cells may still assist in this activation if viral class II antigens get displayed on the cell surface. It is noteworthy that EBV-encoded LMP1 is a viral equivalent of activated CD40 (Kilger et al., 1998), shown to be important in my system of polyclonal B cell activation.
The observation of polyclonal B cell activation has implications for autoimmunity and for the long-term maintenance of B cell memory and serum Ab titers. Autoreactive
B cells might take up viral antigens and be activated in this manner. Indeed, the generation of auto-antibodies is a common feature in human and mouse viral diseases
(Hansen et al., 1998; Ludewig et al., 2004). Previous studies have also shown that a hemolytic anemia can be caused by LCMV-induced polyclonally activated B cells and that this process can be prevented by the depletion of CD4 T cells (Stellrecht and Vella,
1992). I suggest that B cells reactive with red blood cell antigens may have become polyclonally activated because of their presentation of LCMV antigens to LCMV-specific
CD4 T cells. In fact, auto-antibodies with specificities that cause human disease have been isolated from mice during LCMV-induced hypergammaglobulinemia (Hunziker et al., 2003).
What is not clear is why some antigen-pulsed B cells are killed by CD4 T cells and why others become polyclonally activated. These different fates may be determined by properties intrinsic to the T cell, as some T cells may help B cells while others kill B cells, or it may be a property of the B cell, with some B cells expressing higher levels of 134
anti-apoptotic molecules, or it may be a random stochastic process reflecting the number
of T cell hits on a B cell. Results using CD40KO B cells as targets suggest that the fate
decision is not due to entry in to GC. CD40KO B cells that cannot upregulate Fas, cannot
form GC, and cannot undergo polyclonal activation were killed at about the same
frequency as WT B cells. These results would suggest that survival signals in B cells
induced by GC formation or polyclonal activation do not confer protection from MHC
class II-restricted killing. These results may imply that different CD4 T cell subsets,
killers and helpers, dictate the fate of B cells. However, preliminary results using
transgenic B cells as targets for in vivo cytotoxicity assays suggest that there are B cell
intrinsic factors which make B cells more susceptible to specific lysis. Antigen-
presenting transgenic B cells, which are essentially a clonal population, were eliminated more than a polyclonal population of B cells espressing the same viral antigen in the same virus-infected mice. It may be that there are both B cell- and T cell-intrinsic factors which contribute to the fate of antigen presenting B cells in virus-infected mice. Of note, unlike in mice, activated human CD4 T cells express MHC class II (Lanzavecchia et al.,
1988) and one wonders whether CD4 T cells also could be subjected to polyclonal activation or MHC class II-restricted killing by a similar mechanism.
Serum antibodies are secreted by activated B cells that have differentiated into plasma cells. Some plasma cells are short lived and contribute to clearance of the primary viral infection, whereas others are long-lived and can produce neutralizing Ab for several months or years following the primary infection (Manz et al., 1997; Slifka et al., 1998).
One would think that, except in the case of persistent infection or secondary challenge, 135
serum Ab would disappear after a few months, but this is not the case, as titers can be
maintained throughout a lifetime without obvious additional antigen stimulation (Manz et
al., 1998). Several mechanisms have been proposed to explain this observation. One is
that some plasma cells are long living and sufficient to maintain serum Ab (Slifka et al.,
1998). Another model suggests that persisting or cross-reactive antigen induces memory
cells to divide and differentiate into plasma cells throughout an organism’s lifetime
(Ochsenbein et al., 2000; Traggiai et al., 2003). My results of virus-induced polyclonal B
cell activation provide an additional explanation. If memory B cells were polyclonally
activated following viral infection, they may be induced to divide and secrete Ab. Over
the course of an organism’s lifetime, the cumulative effect of constant exposure to new
pathogens and other antigens and the subsequent even low level polyclonal B cell activation might lead to turnover of memory B cells and serum Ab maintenance. It has
been shown that polyclonal activation of human memory B cells in vitro by TLR signaling or by mimicking CD4 T cell help with cytokines and CD40 engagement can occur in the absence of BCR signaling, and these signals were sufficient to sustain Ab titers in the culture fluid (Bernasconi et al., 2002; Traggiai et al., 2003). Interestingly, I
found that MyD88 was not required for polyclonal activation of B cells during LCMV
infection.
I have attempted to show that polyclonal B cell activation could lead to memory B
cell stimulation by transferring GP61-coated influenza virus- or polyomavirus-immune
splenocytes into LCMV-infected mice in some preliminary studies. With both donor
spleen cell populations, the same frequency of GP61-coated B cells was induced to 136
proliferate as seen in Figure IV-1, where GP61-coated naive donor spleen cells were transferred: when influenza virus- or polyomavirus-immune spenocytes were transferred,
GP61-coated B cells divided and uncoated or HA209-coated B cells did not. Due to the differences in proliferation amongst GP61-coated B cells and uncoated or HA209-coated
B cells in LCMV-infected mice, I expected to see differences in the level of Ab produced by GP61-coated donor populations as I had seen when I used naïve IgHa donor B cells
(Figs. IV-13 and IV-14). Despite the proliferation of GP61-coated cells, influenza virus- specific Ab responses could not be detected in any mice regardless of treatment. This result could be explained by a low frequency of influenza virus-specific memory B cells in the spleen of influenza immune mice although these donor mice had detectable levels of anti-influenza serum Ab prior to adoptive transfer. In contrast, polyomavirus VP1- specific Ab was detected in both LCMV-infected and uninfected host mice and no discernable pattern could be detected with regard to LCMV antigen presentation by donor
B cells. These results may be explained by the fact that polyomavirus can establish a persistent infection which could result in chronic B cell simulation (Moser and Lukacher,
2001). Similar studies have been done using mice infected with MHV68. Influenza virus-immune mice were infected with MHV68 and subsequently tested for hypergammaglobulinemia and anti-influenza Ab production (Stevenson and Doherty,
1999). These mice experienced hypergammaglobulinemia as a result of the MHV68 but the level of anti-influenza Ab did not increase (Stevenson and Doherty, 1999). These results and my own preliminary studies would suggest that memory B cells are not stimulated by virus-induced polyclonal B cell activation. However, as previously 137
discussed, the use of a transgenic population of memory B cells in my adoptive transfer system will clarify whether memory B cells can indeed be polyclonally activated by virus infections.
F. Conclusions. The data presented in this thesis help provide a better understanding of the interactions that occur between CD4 T cells and B cells during a viral infection. This thesis provides the first in vivo evidence for MHC class II-restricted killing, a function of CD4 T cells that was previously thought only to occur in vitro, and it describes a virus-induced polyclonal B cell activation, long thought to be the cause of hypergammaglobulinemia associated with human and murine virus infections. Several observations can be made from my studies where I used adoptive transfer systems to ask what would happen to viral-antigen-presenting B cells if they were transferred into virus-
infected hosts.
The first set of observations is that MHC class II-restricted killing can occur in
vivo and CD4 T cells are required for this killing. B cells that present viral antigens in
virus-infected hosts are specifically eliminated by CD4 T cells. CD4 T cells use Fas-FasL
interactions and perforin to kill target cells. The second set of observation is that those
viral-antigen-presenting B cells that survive MHC class II-restricted cytotoxicity undergo
a polyclonal activation manifested by proliferation and phenotypic change. Viral-antigen-
presenting B cells increase in number and display a GC phenotype. Complementing the
previous reports of virus-induced hypergammaglobulinemia, the virus-induced polyclonal
B cell activation is CD4 T cell-dependent and CD40-CD40L interactions are also 138
required. Virus-induced polyclonal B cell activation also results in Ab production, regardless of BCR specificity. B cells isolated from mice experiencing high viral loads also proliferate. Thus, virus-induced polyclonal B cell activation can explain the virus- induced hypergammaglobulinemia. Both are caused by B cells, which regardless of their specificity, take up viral antigen and present it to CD4 T cells leading to their subsequent activation. The studies performed in this thesis provide a careful analysis of this process.
Some of the mechanisms required for polyclonal B cell activation remain elusive, and an explanation as to why CD4 T cells kill some B cells and activate others is also still a mystery. However, the techniques and descriptions of MHC class II-restricted killing and polyclonal activation of B cells provided by this thesis should provide a base from which many future studies can be initiated. 139
CHAPTER VI
Reference List
Adachi,O., Kawai,T., Takeda,K., Matsumoto,M., Tsutsui,H., Sakagami,M., Nakanishi,K., and Akira,S. (1998). Targeted disruption of the MyD88 gene results in loss of IL-1- and IL-18-mediated function. Immunity. 9, 143-150.
Ahmed,R., Butler,L.D., and Bhatti,L. (1988). T4+ T helper cell function in vivo: differential requirement for induction of antiviral cytotoxic T-cell and antibody responses. J. Virol. 62, 2102-2106.
Ahmed,R., Salmi,A., Butler,L.D., Chiller,J.M., and Oldstone,M.B. (1984). Selection of genetic variants of lymphocytic choriomeningitis virus in spleens of persistently infected mice. Role in suppression of cytotoxic T lymphocyte response and viral persistence. J. Exp. Med. 160, 521-540.
Althage,A., Odermatt,B., Moskophidis,D., Kundig,T., Hoffman-Rohrer,U., Hengartner,H., and Zinkernagel,R.M. (1992). Immunosuppression by lymphocytic choriomeningitis virus infection: competent effector T and B cells but impaired antigen presentation. Eur. J. Immunol. 22, 1803-1812.
Altman,J.D., Moss,P.A., Goulder,P.J., Barouch,D.H., Heyzer-Williams,M.G., Bell,J.I., McMichael,A.J., and Davis,M.M. (1996). Phenotypic analysis of antigen-specific T lymphocytes. Science 274, 94-96.
Alzari,P.M., Lascombe,M.B., and Poljak,R.J. (1988). Three-dimensional structure of antibodies. Annu. Rev. Immunol. 6, 555-580.
Appay,V., Zaunders,J.J., Papagno,L., Sutton,J., Jaramillo,A., Waters,A., Easterbrook,P., Grey,P., Smith,D., McMichael,A.J., Cooper,D.A., Rowland-Jones,S.L., and Kelleher,A.D. (2002). Characterization of CD4(+) CTLs ex vivo. J. Immunol. 168, 5954- 5958.
Asano,M.S. and Ahmed,R. (1996). CD8 T cell memory in B cell-deficient mice. J. Exp. Med. 183, 2165-2174.
Bahl,K., Kim,S.K., Calcagno,C., Ghersi,D., Puzone,R., Celada,F., Selin,L.K., and Welsh,R.M. (2006). IFN-induced attrition of CD8 T cells in the presence or absence of cognate antigen during the early stages of viral infections. J. Immunol. 176, 4284-4295. 140
Baldridge,J.R., McGraw,T.S., Paoletti,A., and Buchmeier,M.J. (1997). Antibody prevents the establishment of persistent arenavirus infection in synergy with endogenous T cells. J. Virol. 71, 755-758.
Barber,D.L., Wherry,E.J., and Ahmed,R. (2003). Cutting edge: rapid in vivo killing by memory CD8 T cells. J Immunol 171, 27-31.
Barber,D.L., Wherry,E.J., Masopust,D., Zhu,B., Allison,J.P., Sharpe,A.H., Freeman,G.J., and Ahmed,R. (2006). Restoring function in exhausted CD8 T cells during chronic viral infection. Nature 439, 682-687.
Battegay,M., Moskophidis,D., Rahemtulla,A., Hengartner,H., Mak,T.W., and Zinkernagel,R.M. (1994). Enhanced establishment of a virus carrier state in adult CD4+ T-cell-deficient mice. J. Virol. 68, 4700-4704.
Battegay,M., Moskophidis,D., Waldner,H., Brundler,M.A., Fung-Leung,W.P., Mak,T.W., Hengartner,H., and Zinkernagel,R.M. (1993). Impairment and delay of neutralizing antiviral antibody responses by virus-specific cytotoxic T cells. J. Immunol. 151, 5408-5415.
Bernasconi,N.L., Traggiai,E., and Lanzavecchia,A. (2002). Maintenance of serological memory by polyclonal activation of human memory B cells. Science 298, 2199-2202.
Bevan,M.J. (1976). Cross-priming for a secondary cytotoxic response to minor H antigens with H-2 congenic cells which do not cross-react in the cytotoxic assay. J. Exp. Med. 143, 1283-1288.
Biron,C.A., Sonnenfeld,G., and Welsh,R.M. (1984). Interferon induces natural killer cell blastogenesis in vivo. J. Leukoc. Biol. 35, 31-37.
Biron,C.A., Turgiss,L.R., and Welsh,R.M. (1983). Increase in NK cell number and turnover rate during acute viral infection. J. Immunol. 131, 1539-1545.
Bluestone,J.A. and Abbas,A.K. (2003). Natural versus adaptive regulatory T cells. Nat. Rev. Immunol. 3, 253-257.
Born,W.K., Jin,N., Aydintug,M.K., Wands,J.M., French,J.D., Roark,C.L., and O'Brien,R.L. (2007). gammadelta T lymphocytes-selectable cells within the innate system? J. Clin. Immunol. 27, 133-144.
Borrow,P., Evans,C.F., and Oldstone,M.B. (1995). Virus-induced immunosuppression: immune system-mediated destruction of virus-infected dendritic cells results in generalized immune suppression. J. Virol. 69, 1059-1070. 141
Borrow,P., Tishon,A., Lee,S., Xu,J., Grewal,I.S., Oldstone,M.B., and Flavell,R.A. (1996). CD40L-deficient mice show deficits in antiviral immunity and have an impaired memory CD8+ CTL response. J. Exp. Med. 183, 2129-2142.
Borrow,P., Tishon,A., and Oldstone,M.B. (1991). Infection of lymphocytes by a virus that aborts cytotoxic T lymphocyte activity and establishes persistent infection. J. Exp. Med. 174, 203-212.
Bots,M., Kolfschoten,I.G., Bres,S.A., Rademaker,M.T., de Roo,G.M., Kruse,M., Franken,K.L., Hahne,M., Froelich,C.J., Melief,C.J., Offringa,R., and Medema,J.P. (2005). SPI-CI and SPI-6 cooperate in the protection from effector cell-mediated cytotoxicity. Blood 105, 1153-1161.
Bourgault,I., Gomez,A., Gomard,E., Picard,F., Levy,J.P., and Gomrad,E. (1989). A virus- specific CD4+ cell-mediated cytolytic activity revealed by CD8+ cell elimination regularly develops in uncloned human antiviral cell lines. J. Immunol. 142, 252-256.
Bromley,S.K., Burack,W.R., Johnson,K.G., Somersalo,K., Sims,T.N., Sumen,C., Davis,M.M., Shaw,A.S., Allen,P.M., and Dustin,M.L. (2001). The immunological synapse. Annu. Rev. Immunol. 19, 375-396.
Brooks,D.G., Trifilo,M.J., Edelmann,K.H., Teyton,L., McGavern,D.B., and Oldstone,M.B. (2006). Interleukin-10 determines viral clearance or persistence in vivo. Nat. Med. 12, 1301-1309.
Brown,D.M., Dilzer,A.M., Meents,D.L., and Swain,S.L. (2006). CD4 T cell-mediated protection from lethal influenza: perforin and antibody-mediated mechanisms give a one- two punch. J. Immunol. 177, 2888-2898.
Brundler,M.A., Aichele,P., Bachmann,M., Kitamura,D., Rajewsky,K., and Zinkernagel,R.M. (1996). Immunity to viruses in B cell-deficient mice: influence of antibodies on virus persistence and on T cell memory. Eur. J. Immunol. 26, 2257-2262.
Bruns,M., Cihak,J., Muller,G., and Lehmann-Grube,F. (1983). Lymphocytic choriomeningitis virus. VI. Isolation of a glycoprotein mediating neutralization. Virology 130, 247-251.
Buchmeier,M.J., de la Torre,J.C., and eters,C.J. (2001). Arenaviridae: The Viruses and Their Replication. In Fields' Virology, D.M.Knipe and P.Howley, eds. (Philadelphia: Lippincott Williams & Wilkins).
Buchmeier,M.J., Welsh,R.M., Dutko,F.J., and Oldstone,M.B. (1980). The virology and immunobiology of lymphocytic choriomeningitis virus infection. Adv. Immunol. 30, 275-331. 142
Bukowski,J.F., Warner,J.F., Dennert,G., and Welsh,R.M. (1985). Adoptive transfer studies demonstrating the antiviral effect of natural killer cells in vivo. J. Exp. Med. 161, 40-52.
Bukowski,J.F., Woda,B.A., Habu,S., Okumura,K., and Welsh,R.M. (1983). Natural killer cell depletion enhances virus synthesis and virus-induced hepatitis in vivo. J. Immunol. 131, 1531-1538.
Burgio,G.R. and Monafo,V. (1983). Infectious mononucleosis fifty years after the discovery of the Paul-Bunnell test. Infection 11, 1-5.
Burnet,F.M. (1991). The Nobel Lectures in Immunology. The Nobel Prize for Physiology or Medicine, 1960. Immunologic recognition of self. Scand. J. Immunol. 33, 3-13.
Cao,W., Henry,M.D., Borrow,P., Yamada,H., Elder,J.H., Ravkov,E.V., Nichol,S.T., Compans,R.W., Campbell,K.P., and Oldstone,M.B. (1998). Identification of alpha- dystroglycan as a receptor for lymphocytic choriomeningitis virus and Lassa fever virus. Science 282, 2079-2081.
Cauley,L.S., Cookenham,T., Miller,T.B., Adams,P.S., Vignali,K.M., Vignali,D.A., and Woodland,D.L. (2002). Cutting edge: virus-specific CD4+ memory T cells in nonlymphoid tissues express a highly activated phenotype. J. Immunol. 169, 6655-6658.
Cerny,A., Huegin,A.W., Sutter,S., Bazin,H., Hengartner,H.H., and Zinkernagel,R.M. (1986). Immunity to lymphocytic choriomeningitis virus in B cell-depleted mice: evidence for B cell and antibody-independent protection by memory T cells. Eur. J. Immunol. 16, 913-917.
Cerny,A., Sutter,S., Bazin,H., Hengartner,H., and Zinkernagel,R.M. (1988). Clearance of lymphocytic choriomeningitis virus in antibody- and B-cell-deprived mice. J. Virol. 62, 1803-1807.
Chen,Z.Z., McGuire,J.C., Leach,K.L., and Cambier,J.C. (1987). Transmembrane signaling through B cell MHC class II molecules: anti-Ia antibodies induce protein kinase C translocation to the nuclear fraction. J. Immunol. 138, 2345-2352.
Chirmule,N., Kalyanaraman,V.S., Lederman,S., Oyaizu,N., Yagura,H., Yellin,M.J., Chess,L., and Pahwa,S. (1993). HIV-gp 160-induced T cell-dependent B cell differentiation. Role of T cell-B cell activation molecule and IL-6. J. Immunol. 150, 2478-2486.
Choi,M.S., Holmann,M., Atkins,C.J., and Klaus,G.G. (1996). Expression of bcl-x during mouse B cell differentiation and following activation by various stimuli. Eur. J. Immunol. 26, 676-682. 143
Christensen,J.P., Ropke,C., and Thomsen,A.R. (1996). Virus-induced polyclonal T cell activation is followed by apoptosis: partitioning of CD8+ T cells based on alpha 4 integrin expression. Int. Immunol. 8, 707-715.
Cole,G.A., Nathanson,N., and Prendergast,R.A. (1972). Requirement for theta-bearing cells in lymphocytic choriomeningitis virus-induced central nervous system disease. Nature 238, 335-337.
Coro,E.S., Chang,W.L., and Baumgarth,N. (2006). Type I IFN receptor signals directly stimulate local B cells early following influenza virus infection. J. Immunol. 176, 4343- 4351.
Coutelier,J.P., Coulie,P.G., Wauters,P., Heremans,H., and van der Logt,J.T. (1990). In vivo polyclonal B-lymphocyte activation elicited by murine viruses. J. Virol. 64, 5383- 5388.
Coutelier,J.P., Johnston,S.J., El Idrissi,M.e., and Pfau,C.J. (1994). Involvement of CD4+ cells in lymphocytic choriomeningitis virus-induced autoimmune anaemia and hypergammaglobulinaemia. J. Autoimmun. 7, 589-599.
Coutelier,J.P., van der Logt,J.T., Heessen,F.W., Vink,A., and Van,S.J. (1988). Virally induced modulation of murine IgG antibody subclasses. J. Exp. Med. 168, 2373-2378.
Duverlie,G., Driencourt,M., Roussel,C., and Orfila,J. (1989). Heterophile IgM, IgA, and IgE antibodies in infectious mononucleosis. J. Med. Virol. 28, 38-41.
Ejrnaes,M., Filippi,C.M., Martinic,M.M., Ling,E.M., Togher,L.M., Crotty,S., and von Herrath,M.G. (2006). Resolution of a chronic viral infection after interleukin-10 receptor blockade. J. Exp. Med. 203, 2461-2472.
Erb,P., Grogg,D., Troxler,M., Kennedy,M., and Fluri,M. (1990). CD4+ T cell-mediated killing of MHC class II-positive antigen-presenting cells. I. Characterization of target cell recognition by in vivo or in vitro activated CD4+ killer T cells. J. Immunol. 144, 790- 795.
Fields,B.A. and Mariuzza,R.A. (1996). Structure and function of the T-cell receptor: insights from X-ray crystallography. Immunol. Today 17, 330-336.
Fink,K., Lang,K.S., Manjarrez-Orduno,N., Junt,T., Senn,B.M., Holdener,M., Akira,S., Zinkernagel,R.M., and Hengartner,H. (2006). Early type I interferon-mediated signals on B cells specifically enhance antiviral humoral responses. Eur. J. Immunol. 36, 2094- 2105. 144
Foy,T.M., Laman,J.D., Ledbetter,J.A., Aruffo,A., Claassen,E., and Noelle,R.J. (1994). gp39-CD40 interactions are essential for germinal center formation and the development of B cell memory. J. Exp. Med. 180, 157-163.
Frazer,J.K., Jackson,D.G., Gaillard,J.P., Lutter,M., Liu,Y.J., Banchereau,J., Capra,J.D., and Pascual,V. (2000). Identification of centerin: a novel human germinal center B cell- restricted serpin. Eur. J. Immunol. 30, 3039-3048.
Fuller,M.J., Hildeman,D.A., Sabbaj,S., Gaddis,D.E., Tebo,A.E., Shang,L., Goepfert,P.A., and Zajac,A.J. (2005). Cutting edge: emergence of CD127high functionally competent memory T cells is compromised by high viral loads and inadequate T cell help. J. Immunol. 174, 5926-5930.
Fuller,M.J. and Zajac,A.J. (2003). Ablation of CD8 and CD4 T cell responses by high viral loads. J. Immunol. 170, 477-486.
Germain,R.N. (1994). MHC-dependent antigen processing and peptide presentation: providing ligands for T lymphocyte activation. Cell 76, 287-299.
Gessner,A., Drjupin,R., Lohler,J., Lother,H., and Lehmann-Grube,F. (1990). IFN-gamma production in tissues of mice during acute infection with lymphocytic choriomeningitis virus. J. Immunol. 144, 3160-3165.
Giorgetti,C.A. and Press,J.L. (1994). A peptide sequence mimics the epitope on the multideterminant antigen (Tyr,Glu)-Ala-Lys that induces the dominant H10/V kappa 1+ primary antibody response. J. Immunol. 152, 136-145.
Grogan,J.L. and Locksley,R.M. (2002). T helper cell differentiation: on again, off again. Curr. Opin. Immunol. 14, 366-372.
Grossman,W.J., Verbsky,J.W., Barchet,W., Colonna,M., Atkinson,J.P., and Ley,T.J. (2004a). Human T regulatory cells can use the perforin pathway to cause autologous target cell death. Immunity. 21, 589-601.
Grossman,W.J., Verbsky,J.W., Tollefsen,B.L., Kemper,C., Atkinson,J.P., and Ley,T.J. (2004b). Differential expression of granzymes A and B in human cytotoxic lymphocyte subsets and T regulatory cells. Blood 104, 2840-2848.
Guay,H.M., Andreyeva,T.A., Garcea,R.L., Welsh,R.M., and Szomolanyi-Tsuda,E. (2007). MyD88 is required for the formation of long-term humoral immunity to virus infection. J. Immunol. 178, 5124-5131.
Guay,H.M., Panarey,L., Reed,A.J., and Caton,A.J. (2004). Specificity-based negative selection of autoreactive B cells during memory formation. J. Immunol. 173, 5485-5494. 145
Hahn,S., Gehri,R., and Erb,P. (1995). Mechanism and biological significance of CD4- mediated cytotoxicity. Immunol. Rev. 146, 57-79.
Han,S., Zheng,B., Takahashi,Y., and Kelsoe,G. (1997). Distinctive characteristics of germinal center B cells. Semin. Immunol. 9, 255-260.
Hangartner,L., Senn,B.M., Ledermann,B., Kalinke,U., Seiler,P., Bucher,E., Zellweger,R.M., Fink,K., Odermatt,B., Burki,K., Zinkernagel,R.M., and Hengartner,H. (2003). Antiviral immune responses in gene-targeted mice expressing the immunoglobulin heavy chain of virus-neutralizing antibodies. Proc. Natl. Acad. Sci. U. S. A 100, 12883-12888.
Hansen,K.E., Arnason,J., and Bridges,A.J. (1998). Autoantibodies and common viral illnesses. Semin. Arthritis Rheum. 27, 263-271.
Havenar-Daughton,C., Kolumam,G.A., and Murali-Krishna,K. (2006). Cutting Edge: The direct action of type I IFN on CD4 T cells is critical for sustaining clonal expansion in response to a viral but not a bacterial infection. J. Immunol. 176, 3315-3319.
Hawrylowicz,C.M. and Unanue,E.R. (1988). Regulation of antigen-presentation-I. IFN- gamma induces antigen-presenting properties on B cells. J. Immunol. 141, 4083-4088.
He,B., Qiao,X., Klasse,P.J., Chiu,A., Chadburn,A., Knowles,D.M., Moore,J.P., and Cerutti,A. (2006). HIV-1 envelope triggers polyclonal Ig class switch recombination through a CD40-independent mechanism involving BAFF and C-type lectin receptors. J. Immunol. 176, 3931-3941.
Homann,D., McGavern,D.B., and Oldstone,M.B. (2004). Visualizing the viral burden: phenotypic and functional alterations of T cells and APCs during persistent infection. J. Immunol. 172, 6239-6250.
Hotchin,J. (1962). The biology of lymphocytic choriomeningitis infection: virus-induced immune disease. Cold Spring Harb. Symp. Quant. Biol. 27, 479-499.
Hou,S., Fishman,M., Murti,K.G., and Doherty,P.C. (1993). Divergence between cytotoxic effector function and tumor necrosis factor alpha production for inflammatory CD4+ T cells from mice with Sendai virus pneumonia. J. Virol. 67, 6299-6302.
Hu,C.P., Aman,P., Masucci,M.G., Klein,E., and Klein,G. (1986). B cell activation by the nontransforming P3HR-1 substrain of the Epstein-Barr virus (EBV). Eur. J. Immunol. 16, 841-845.
Hunziker,L., Klenerman,P., Zinkernagel,R.M., and Ehl,S. (2002). Exhaustion of cytotoxic T cells during adoptive immunotherapy of virus carrier mice can be prevented by B cells or CD4+ T cells. Eur. J. Immunol. 32, 374-382. 146
Hunziker,L., Recher,M., Macpherson,A.J., Ciurea,A., Freigang,S., Hengartner,H., and Zinkernagel,R.M. (2003). Hypergammaglobulinemia and autoantibody induction mechanisms in viral infections. Nat. Immunol. 4, 343-349.
Hutt-Fletcher,L.M., Balachandran,N., and Elkins,M.H. (1983). B cell activation by cytomegalovirus. J. Exp. Med. 158, 2171-2176.
Jacob,J., Kelsoe,G., Rajewsky,K., and Weiss,U. (1991). Intraclonal generation of antibody mutants in germinal centres. Nature 354, 389-392.
Jacobson,S., Richert,J.R., Biddison,W.E., Satinsky,A., Hartzman,R.J., and McFarland,H.F. (1984). Measles virus-specific T4+ human cytotoxic T cell clones are restricted by class II HLA antigens. J. Immunol. 133, 754-757.
Jamieson,B.D. and Ahmed,R. (1989). T cell memory. Long-term persistence of virus- specific cytotoxic T cells. J. Exp. Med. 169, 1993-2005.
Janssen,E.M., Lemmens,E.E., Wolfe,T., Christen,U., von Herrath,M.G., and Schoenberger,S.P. (2003). CD4+ T cells are required for secondary expansion and memory in CD8+ T lymphocytes. Nature 421, 852-856.
Janssens,W., Carlier,V., Wu,B., VanderElst,L., Jacquemin,M.G., and Saint-Remy,J.M. (2003). CD4+CD25+ T cells lyse antigen-presenting B cells by Fas-Fas ligand interaction in an epitope-specific manner. J. Immunol. 171, 4604-4612.
Jorgensen,T.N., Roper,E., Thurman,J.M., Marrack,P., and Kotzin,B.L. (2007). Type I interferon signaling is involved in the spontaneous development of lupus-like disease in B6.Nba2 and (B6.Nba2 x NZW)F(1) mice. Genes Immun.
Kacani,L., Sprinzl,G.M., Erdei,A., and Dierich,M.P. (1999). Interleukin-15 enhances HIV-1-driven polyclonal B-cell response in vitro. Exp. Clin. Immunogenet. 16, 162-172.
Kacani,L., Stoiber,H., and Dierich,M.P. (1997). Role of IL-15 in HIV-1-associated hypergammaglobulinaemia. Clin. Exp. Immunol. 108, 14-18.
Kasaian,M.T. and Biron,C.A. (1989). The activation of IL-2 transcription in L3T4+ and Lyt-2+ lymphocytes during virus infection in vivo. J. Immunol. 142, 1287-1292.
Kawamura,K., Kadowaki,N., Kitawaki,T., and Uchiyama,T. (2006). Virus-stimulated plasmacytoid dendritic cells induce CD4+ cytotoxic regulatory T cells. Blood 107, 1031- 1038.
Kayagaki,N., Yamaguchi,N., Nakayama,M., Kawasaki,A., Akiba,H., Okumura,K., and Yagita,H. (1999). Involvement of TNF-related apoptosis-inducing ligand in human CD4+ T cell-mediated cytotoxicity. J. Immunol. 162, 2639-2647. 147
Khanolkar,A., Fuller,M.J., and Zajac,A.J. (2004). CD4 T cell-dependent CD8 T cell maturation. J. Immunol. 172, 2834-2844.
Kilger,E., Kieser,A., Baumann,M., and Hammerschmidt,W. (1998). Epstein-Barr virus- mediated B-cell proliferation is dependent upon latent membrane protein 1, which simulates an activated CD40 receptor. EMBO J. 17, 1700-1709.
Kim,S.K., Brehm,M.A., Welsh,R.M., and Selin,L.K. (2002). Dynamics of memory T cell proliferation under conditions of heterologous immunity and bystander stimulation. J. Immunol. 169, 90-98.
Kim,S.K. and Welsh,R.M. (2004). Comprehensive early and lasting loss of memory CD8 T cells and functional memory during acute and persistent viral infections. J. Immunol. 172, 3139-3150.
Klein,U., Rajewsky,K., and Kuppers,R. (1998). Human immunoglobulin (Ig)M+IgD+ peripheral blood B cells expressing the CD27 cell surface antigen carry somatically mutated variable region genes: CD27 as a general marker for somatically mutated (memory) B cells. J. Exp. Med. 188, 1679-1689.
Kolumam,G.A., Thomas,S., Thompson,L.J., Sprent,J., and Murali-Krishna,K. (2005). Type I interferons act directly on CD8 T cells to allow clonal expansion and memory formation in response to viral infection. J. Exp. Med. 202, 637-650.
Kotturi,M.F., Peters,B., Buendia-Laysa,F., Jr., Sidney,J., Oseroff,C., Botten,J., Grey,H., Buchmeier,M.J., and Sette,A. (2007). The CD8+ T-cell response to lymphocytic choriomeningitis virus involves the L antigen: uncovering new tricks for an old virus. J. Virol. 81, 4928-4940.
Kuppers,R. (2003). B cells under influence: transformation of B cells by Epstein-Barr virus. Nat. Rev. Immunol. 3, 801-812.
Lane,H.C., Masur,H., Edgar,L.C., Whalen,G., Rook,A.H., and Fauci,A.S. (1983). Abnormalities of B-cell activation and immunoregulation in patients with the acquired immunodeficiency syndrome. N. Engl. J. Med. 309, 453-458.
Lanzavecchia,A., Roosnek,E., Gregory,T., Berman,P., and Abrignani,S. (1988). T cells can present antigens such as HIV gp120 targeted to their own surface molecules. Nature 334, 530-532.
Lardans,V., Godfraind,C., van der Logt,J.T., Heessen,W.A., Gonzalez,M.D., and Coutelier,J.P. (1996). Polyclonal B lymphocyte activation induced by mouse hepatitis virus A59 infection. J. Gen. Virol. 77 ( Pt 5), 1005-1009. 148
Le,B.A., Thompson,C., Kamphuis,E., Durand,V., Rossmann,C., Kalinke,U., and Tough,D.F. (2006). Cutting edge: enhancement of antibody responses through direct stimulation of B and T cells by type I IFN. J. Immunol. 176, 2074-2078.
Lehmann-Grube,F., Lohler,J., Utermohlen,O., and Gegin,C. (1993). Antiviral immune responses of lymphocytic choriomeningitis virus-infected mice lacking CD8+ T lymphocytes because of disruption of the beta 2-microglobulin gene. J. Virol. 67, 332- 339.
Lenz,D.C., Kurz,S.K., Lemmens,E., Schoenberger,S.P., Sprent,J., Oldstone,M.B., and Homann,D. (2004). IL-7 regulates basal homeostatic proliferation of antiviral CD4+T cell memory. Proc. Natl. Acad. Sci. U. S. A 101, 9357-9362.
Li,X., Hu,B., Harty,J., Even,C., and Plagemann,P.G. (1990). Polyclonal B cell activation of IgG2a and IgG2b production by infection of mice with lactate dehydrogenase- elevating virus is partly dependent on CD4+ lymphocytes. Viral Immunol. 3, 273-288.
Li,Z., Woo,C.J., Iglesias-Ussel,M.D., Ronai,D., and Scharff,M.D. (2004). The generation of antibody diversity through somatic hypermutation and class switch recombination. Genes Dev. 18, 1-11.
Lin,M.Y. and Welsh,R.M. (1998). Stability and diversity of T cell receptor repertoire usage during lymphocytic choriomeningitis virus infection of mice. J. Exp. Med. 188, 1993-2005.
Lohman,B.L., Razvi,E.S., and Welsh,R.M. (1996). T-lymphocyte downregulation after acute viral infection is not dependent on CD95 (Fas) receptor-ligand interactions. J. Virol. 70, 8199-8203.
Ludewig,B., Krebs,P., Metters,H., Tatzel,J., Tureci,O., and Sahin,U. (2004). Molecular characterization of virus-induced autoantibody responses. J. Exp. Med. 200, 637-646.
Lukacher,A.E., Morrison,L.A., Braciale,V.L., Malissen,B., and Braciale,T.J. (1985). Expression of specific cytolytic activity by H-2I region-restricted, influenza virus- specific T lymphocyte clones. J. Exp. Med. 162, 171-187.
Macchia,D., Almerigogna,F., Parronchi,P., Ravina,A., Maggi,E., and Romagnani,S. (1993). Membrane tumour necrosis factor-alpha is involved in the polyclonal B-cell activation induced by HIV-infected human T cells. Nature 363, 464-466.
MacLennan,I.C. (1994a). Germinal centers. Annu. Rev. Immunol. 12, 117-139.
MacLennan,I.C. (1994b). Somatic mutation. From the dark zone to the light. Curr. Biol. 4, 70-72. 149
Maimone,M.M., Morrison,L.A., Braciale,V.L., and Braciale,T.J. (1986). Features of target cell lysis by class I and class II MHC-restricted cytolytic T lymphocytes. J. Immunol. 137, 3639-3643.
Manz,R.A., Lohning,M., Cassese,G., Thiel,A., and Radbruch,A. (1998). Survival of long- lived plasma cells is independent of antigen. Int. Immunol. 10, 1703-1711.
Manz,R.A., Thiel,A., and Radbruch,A. (1997). Lifetime of plasma cells in the bone marrow. Nature 388, 133-134.
Marinova,E., Han,S., and Zheng,B. (2007). Germinal center helper T cells are dual functional regulatory cells with suppressive activity to conventional CD4+ T cells. J. Immunol. 178, 5010-5017.
Markine-Goriaynoff,D., Nguyen,T.D., Bigaignon,G., Van,S.J., and Coutelier,J.P. (2001). Distinct requirements for IL-6 in polyclonal and specific Ig production induced by microorganisms. Int. Immunol. 13, 1185-1192.
Martin,F. and Kearney,J.F. (2000). B-cell subsets and the mature preimmune repertoire. Marginal zone and B1 B cells as part of a "natural immune memory". Immunol. Rev. 175, 70-79.
Martin,F. and Kearney,J.F. (2001). B1 cells: similarities and differences with other B cell subsets. Curr. Opin. Immunol. 13, 195-201.
Martinez-Maza,O., Crabb,E., Mitsuyasu,R.T., Fahey,J.L., and Giorgi,J.V. (1987). Infection with the human immunodeficiency virus (HIV) is associated with an in vivo increase in B lymphocyte activation and immaturity. J. Immunol. 138, 3720-3724.
Masopust,D., Murali-Krishna,K., and Ahmed,R. (2007). Quantitating the magnitude of the lymphocytic choriomeningitis virus-specific CD8 T-cell response: it is even bigger than we thought. J. Virol. 81, 2002-2011.
Matloubian,M., Concepcion,R.J., and Ahmed,R. (1994). CD4+ T cells are required to sustain CD8+ cytotoxic T-cell responses during chronic viral infection. J. Virol. 68, 8056-8063.
McFarland,H.I., Nahill,S.R., Maciaszek,J.W., and Welsh,R.M. (1992). CD11b (Mac-1): a marker for CD8+ cytotoxic T cell activation and memory in virus infection. J. Immunol. 149, 1326-1333.
McNally,J.M., Zarozinski,C.C., Lin,M.Y., Brehm,M.A., Chen,H.D., and Welsh,R.M. (2001). Attrition of bystander CD8 T cells during virus-induced T-cell and interferon responses. J. Virol. 75, 5965-5976. 150
Medema,J.P., Schuurhuis,D.H., Rea,D., van,T.J., de,J.J., Bres,S.A., Laban,S., Toes,R.E., Toebes,M., Schumacher,T.N., Bladergroen,B.A., Ossendorp,F., Kummer,J.A., Melief,C.J., and Offringa,R. (2001). Expression of the serpin serine protease inhibitor 6 protects dendritic cells from cytotoxic T lymphocyte-induced apoptosis: differential modulation by T helper type 1 and type 2 cells. J. Exp. Med. 194, 657-667.
Meffre,E., Casellas,R., and Nussenzweig,M.C. (2000). Antibody regulation of B cell development. Nat. Immunol. 1, 379-385.
Merigan,T.C., Oldstone,M.B., and Welsh,R.M. (1977). Interferon production during lymphocytic choriomeningitis virus infection of nude and normal mice. Nature 268, 67- 68.
Miller,J.P., Stadanlick,J.E., and Cancro,M.P. (2006). Space, selection, and surveillance: setting boundaries with BLyS. J. Immunol. 176, 6405-6410.
Minoprio,P., Burlen,O., Pereira,P., Guilbert,B., Andrade,L., Hontebeyrie-Joskowicz,M., and Coutinho,A. (1988). Most B cells in acute Trypanosoma cruzi infection lack parasite specificity. Scand. J. Immunol. 28, 553-561.
Mondino,A. and Jenkins,M.K. (1994). Surface proteins involved in T cell costimulation. J. Leukoc. Biol. 55, 805-815.
Monroe,J.G. and Cambier,J.C. (1982). Cell cycle dependence for expression of membrane associated IgD, IgM and Ia antigen on mitogen-stimulated murine B- lymphocytes. Ann. N. Y. Acad. Sci. 399, 238-254.
Monroe,J.G. and Cambier,J.C. (1983). Level of mIa expression on mitogen-stimulated murine B lymphocytes is dependent on position in cell cycle. J. Immunol. 130, 626-631.
Monroe,J.G., Havran,W.L., and Cambier,J.C. (1983). B lymphocyte activation: entry into cell cycle is accompanied by decreased expression of IgD but not IgM. Eur. J. Immunol. 13, 208-213.
Moser,J.M. and Lukacher,A.E. (2001). Immunity to polyoma virus infection and tumorigenesis. Viral Immunol. 14, 199-216.
Moskophidis,D., Battegay,M., Bruendler,M.A., Laine,E., Gresser,I., and Zinkernagel,R.M. (1994). Resistance of lymphocytic choriomeningitis virus to alpha/beta interferon and to gamma interferon. J. Virol. 68, 1951-1955.
Moskophidis,D., Cobbold,S.P., Waldmann,H., and Lehmann-Grube,F. (1987). Mechanism of recovery from acute virus infection: treatment of lymphocytic choriomeningitis virus-infected mice with monoclonal antibodies reveals that Lyt-2+ T 151
lymphocytes mediate clearance of virus and regulate the antiviral antibody response. J. Virol. 61, 1867-1874.
Moskophidis,D., Lechner,F., Pircher,H., and Zinkernagel,R.M. (1993). Virus persistence in acutely infected immunocompetent mice by exhaustion of antiviral cytotoxic effector T cells. Nature 362, 758-761.
Moskophidis,D. and Lehmann-Grube,F. (1989). Virus-induced delayed-type hypersensitivity reaction is sequentially mediated by CD8+ and CD4+ T lymphocytes. Proc. Natl. Acad. Sci. U. S. A 86, 3291-3295.
Moskophidis,D. and Lehmann-Grube,F. (1984). The immune response of the mouse to lymphocytic choriomeningitis virus. IV. Enumeration of antibody-producing cells in spleens during acute and persistent infection. J. Immunol. 133, 3366-3370.
Moskophidis,D., Lohler,J., and Lehmann-Grube,F. (1986). Antiviral antibody production in parenchymatous organs of lymphocytic choriomeningitis virus carrier mice. Enumeration of antibody-producing cells and immunohistochemical investigation of mononuclear cell infiltrates. Med. Microbiol. Immunol. (Berl) 175, 113-116.
Moskophidis,D., Moskophidis,M., and Lohler,J. (1997). Virus-specific IgD in acute viral infection of mice. J. Immunol. 158, 1254-1261.
Mosmann,T.R., Cherwinski,H., Bond,M.W., Giedlin,M.A., and Coffman,R.L. (1986). Two types of murine helper T cell clone. I. Definition according to profiles of lymphokine activities and secreted proteins. J. Immunol. 136, 2348-2357.
Mosmann,T.R. and Coffman,R.L. (1989). TH1 and TH2 cells: different patterns of lymphokine secretion lead to different functional properties. Annu. Rev. Immunol. 7, 145-173.
Mosmann,T.R. and Moore,K.W. (1991). The role of IL-10 in crossregulation of TH1 and TH2 responses. Immunol. Today 12, A49-A53.
Mosmann,T.R., Sad,S., Krishnan,L., Wegmann,T.G., Guilbert,L.J., and Belosevic,M. (1995). Differentiation of subsets of CD4+ and CD8+ T cells. Ciba Found. Symp. 195, 42-50.
Muller,D., Koller,B.H., Whitton,J.L., LaPan,K.E., Brigman,K.K., and Frelinger,J.A. (1992). LCMV-specific, class II-restricted cytotoxic T cells in beta 2-microglobulin- deficient mice. Science 255, 1576-1578.
Muller,U., Steinhoff,U., Reis,L.F., Hemmi,S., Pavlovic,J., Zinkernagel,R.M., and Aguet,M. (1994). Functional role of type I and type II interferons in antiviral defense. Science 264, 1918-1921. 152
Murali-Krishna,K., Altman,J.D., Suresh,M., Sourdive,D.J., Zajac,A.J., Miller,J.D., Slansky,J., and Ahmed,R. (1998). Counting antigen-specific CD8 T cells: a reevaluation of bystander activation during viral infection. Immunity. 8, 177-187.
Nagata,S. (1997). Apoptosis by death factor. Cell 88, 355-365.
Nashar,T.O. and Drake,J.R. (2006). Dynamics of MHC Class II-Activating Signals in Murine Resting B Cells. J. Immunol. 176, 827-838.
Noelle,R.J., Roy,M., Shepherd,D.M., Stamenkovic,I., Ledbetter,J.A., and Aruffo,A. (1992). A 39-kDa protein on activated helper T cells binds CD40 and transduces the signal for cognate activation of B cells. Proc. Natl. Acad. Sci. U. S. A 89, 6550-6554.
O'Brien,R.L., Roark,C.L., Jin,N., Aydintug,M.K., French,J.D., Chain,J.L., Wands,J.M., Johnston,M., and Born,W.K. (2007). gammadelta T-cell receptors: functional correlations. Immunol. Rev. 215, 77-88.
Ochsenbein,A.F., Fehr,T., Lutz,C., Suter,M., Brombacher,F., Hengartner,H., and Zinkernagel,R.M. (1999a). Control of early viral and bacterial distribution and disease by natural antibodies. Science 286, 2156-2159.
Ochsenbein,A.F., Pinschewer,D.D., Odermatt,B., Carroll,M.C., Hengartner,H., and Zinkernagel,R.M. (1999b). Protective T cell-independent antiviral antibody responses are dependent on complement. J. Exp. Med. 190, 1165-1174.
Ochsenbein,A.F., Pinschewer,D.D., Sierro,S., Horvath,E., Hengartner,H., and Zinkernagel,R.M. (2000). Protective long-term antibody memory by antigen-driven and T help-dependent differentiation of long-lived memory B cells to short-lived plasma cells independent of secondary lymphoid organs. Proc. Natl. Acad. Sci. U. S. A 97, 13263- 13268.
Oehen,S., Brduscha-Riem,K., Oxenius,A., and Odermatt,B. (1997). A simple method for evaluating the rejection of grafted spleen cells by flow cytometry and tracing adoptively transferred cells by light microscopy. J Immunol Methods 207, 33-42.
Oldstone,M.B. and Dixon,F.J. (1967). Lymphocytic choriomeningitis: production of antibody by "tolerant" infected mice. Science 158, 1193-1195.
Orange,J.S. and Biron,C.A. (1996). An absolute and restricted requirement for IL-12 in natural killer cell IFN-gamma production and antiviral defense. Studies of natural killer and T cell responses in contrasting viral infections. J. Immunol. 156, 1138-1142.
Oxenius,A., Bachmann,M.F., Ashton-Rickardt,P.G., Tonegawa,S., Zinkernagel,R.M., and Hengartner,H. (1995). Presentation of endogenous viral proteins in association with major histocompatibility complex class II: on the role of intracellular 153
compartmentalization, invariant chain and the TAP transporter system. Eur. J. Immunol. 25, 3402-3411.
Oxenius,A., Bachmann,M.F., Zinkernagel,R.M., and Hengartner,H. (1998a). Virus- specific MHC-class II-restricted TCR-transgenic mice: effects on humoral and cellular immune responses after viral infection. Eur. J. Immunol. 28, 390-400.
Oxenius,A., Zinkernagel,R.M., and Hengartner,H. (1998b). Comparison of activation versus induction of unresponsiveness of virus-specific CD4+ and CD8+ T cells upon acute versus persistent viral infection. Immunity. 9, 449-457.
Pasare,C. and Medzhitov,R. (2005). Control of B-cell responses by Toll-like receptors. Nature 438, 364-368.
Paterson,M.A., Horvath,A.J., Pike,R.N., and Coughlin,P.B. (2007). Molecular characterization of centerin, a germinal centre cell serpin. Biochem. J. 405, 489-494.
Paul,J.R. and Bunnell,W.W. (1974). The presence of heterophile antibodies in infectious mononucleosis. Am. J. Med. Sci. 267, 178-188.
Pfau,C.J., Valenti,J.K., Pevear,D.C., and Hunt,K.D. (1982). Lymphocytic choriomeningitis virus killer T cells are lethal only in weakly disseminated murine infections. J. Exp. Med. 156, 79-89.
Pillai,S., Cariappa,A., and Moran,S.T. (2005). Marginal zone B cells. Annu. Rev. Immunol. 23, 161-196.
Pinschewer,D.D., Perez,M., Jeetendra,E., Bachi,T., Horvath,E., Hengartner,H., Whitt,M.A., de la Torre,J.C., and Zinkernagel,R.M. (2004). Kinetics of protective antibodies are determined by the viral surface antigen. J. Clin. Invest 114, 988-993.
Plagemann,P.G., Jones,Q.A., and Cafruny,W.A. (2000). N-glycans on the short ectodomain of the primary envelope glycoprotein play a major role in the polyclonal activation of B cells by lactate dehydrogenase-elevating virus. J. Gen. Virol. 81, 2167- 2175.
Planz,O., Ehl,S., Furrer,E., Horvath,E., Brundler,M.A., Hengartner,H., and Zinkernagel,R.M. (1997). A critical role for neutralizing-antibody-producing B cells, CD4(+) T cells, and interferons in persistent and acute infections of mice with lymphocytic choriomeningitis virus: implications for adoptive immunotherapy of virus carriers. Proc. Natl. Acad. Sci. U. S. A 94, 6874-6879.
Price,P., Olver,S.D., Gibbons,A.E., and Shellam,G.R. (1993). B-cell activation following murine cytomegalovirus infection: implications for autoimmunity. Immunology 78, 14- 21. 154
Purton,J.F., Tan,J.T., Rubinstein,M.P., Kim,D.M., Sprent,J., and Surh,C.D. (2007). Antiviral CD4+ memory T cells are IL-15 dependent. J. Exp. Med. 204, 951-961.
Quinn,D.G., Zajac,A.J., Frelinger,J.A., and Muller,D. (1993). Transfer of lymphocytic choriomeningitis disease in beta 2-microglobulin-deficient mice by CD4+ T cells. Int. Immunol. 5, 1193-1198.
Racanelli,V., Frassanito,M.A., Leone,P., Galiano,M., De,R., V, Silvestris,F., and Dammacco,F. (2006). Antibody production and in vitro behavior of CD27-defined B-cell subsets: persistent hepatitis C virus infection changes the rules. J. Virol. 80, 3923-3934.
Rajewsky,K. (1996). Clonal selection and learning in the antibody system. Nature 381, 751-758.
Razvi,E.S., Jiang,Z., Woda,B.A., and Welsh,R.M. (1995). Lymphocyte apoptosis during the silencing of the immune response to acute viral infections in normal, lpr, and Bcl-2- transgenic mice. Am. J. Pathol. 147, 79-91.
Razvi,E.S. and Welsh,R.M. (1993). Programmed cell death of T lymphocytes during acute viral infection: a mechanism for virus-induced immune deficiency. J. Virol. 67, 5754-5765.
Recher,M., Lang,K.S., Hunziker,L., Freigang,S., Eschli,B., Harris,N.L., Navarini,A., Senn,B.M., Fink,K., Lotscher,M., Hangartner,L., Zellweger,R., Hersberger,M., Theocharides,A., Hengartner,H., and Zinkernagel,R.M. (2004). Deliberate removal of T cell help improves virus-neutralizing antibody production. Nat. Immunol. 5, 934-942.
Reiner,S.L. (2007). Development in motion: helper T cells at work. Cell 129, 33-36.
Rock,K.L. and Shen,L. (2005). Cross-presentation: underlying mechanisms and role in immune surveillance. Immunol. Rev. 207, 166-183.
Rosa,D., Saletti,G., De,G.E., Zorat,F., Comar,C., D'Oro,U., Nuti,S., Houghton,M., Barnaba,V., Pozzato,G., and Abrignani,S. (2005). Activation of naive B lymphocytes via CD81, a pathogenetic mechanism for hepatitis C virus-associated B lymphocyte disorders. Proc. Natl. Acad. Sci. U. S. A 102, 18544-18549.
Rosen,A., Gergely,P., Jondal,M., Klein,G., and Britton,S. (1977). Polyclonal Ig production after Epstein-Barr virus infection of human lymphocytes in vitro. Nature 267, 52-54.
Rowland,R.R., Even,C., Anderson,G.W., Chen,Z., Hu,B., and Plagemann,P.G. (1994). Neonatal infection of mice with lactate dehydrogenase-elevating virus results in suppression of humoral antiviral immune response but does not alter the course of 155
viraemia or the polyclonal activation of B cells and immune complex formation. J. Gen. Virol. 75 ( Pt 5), 1071-1081.
Sad,S. and Mosmann,T.R. (1994). Single IL-2-secreting precursor CD4 T cell can develop into either Th1 or Th2 cytokine secretion phenotype. J. Immunol. 153, 3514- 3522.
Santosuosso,M., Zhang,X., McCormick,S., Wang,J., Hitt,M., and Xing,Z. (2005). Mechanisms of mucosal and parenteral tuberculosis vaccinations: adenoviral-based mucosal immunization preferentially elicits sustained accumulation of immune protective CD4 and CD8 T cells within the airway lumen. J. Immunol. 174, 7986-7994.
Sarmiento,M., Glasebrook,A.L., and Fitch,F.W. (1980). IgG or IgM monoclonal antibodies reactive with different determinants on the molecular complex bearing Lyt 2 antigen block T cell-mediated cytolysis in the absence of complement. J. Immunol. 125, 2665-2672.
Savoldo,B., Cubbage,M.L., Durett,A.G., Goss,J., Huls,M.H., Liu,Z., Teresita,L., Gee,A.P., Ling,P.D., Brenner,M.K., Heslop,H.E., and Rooney,C.M. (2002). Generation of EBV-specific CD4+ cytotoxic T cells from virus naive individuals. J. Immunol. 168, 909-918.
Schnittman,S.M., Lane,H.C., Higgins,S.E., Folks,T., and Fauci,A.S. (1986). Direct polyclonal activation of human B lymphocytes by the acquired immune deficiency syndrome virus. Science 233, 1084-1086.
Seiler,P., Kalinke,U., Rulicke,T., Bucher,E.M., Bose,C., Zinkernagel,R.M., and Hengartner,H. (1998). Enhanced virus clearance by early inducible lymphocytic choriomeningitis virus-neutralizing antibodies in immunoglobulin-transgenic mice. J. Virol. 72, 2253-2258.
Seiler,P., Senn,B.M., Brundler,M.A., Zinkernagel,R.M., Hengartner,H., and Kalinke,U. (1999). In vivo selection of neutralization-resistant virus variants but no evidence of B cell tolerance in lymphocytic choriomeningitis virus carrier mice expressing a transgenic virus-neutralizing antibody. J. Immunol. 162, 4536-4541.
Selin,L.K., Santolucito,P.A., Pinto,A.K., Szomolanyi-Tsuda,E., and Welsh,R.M. (2001). Innate immunity to viruses: control of vaccinia virus infection by gamma delta T cells. J. Immunol. 166, 6784-6794.
Selin,L.K., Vergilis,K., Welsh,R.M., and Nahill,S.R. (1996). Reduction of otherwise remarkably stable virus-specific cytotoxic T lymphocyte memory by heterologous viral infections. J. Exp. Med. 183, 2489-2499. 156
Shedlock,D.J. and Shen,H. (2003). Requirement for CD4 T cell help in generating functional CD8 T cell memory. Science 300, 337-339.
Sher,A. and Coffman,R.L. (1992). Regulation of immunity to parasites by T cells and T cell-derived cytokines. Annu. Rev. Immunol. 10, 385-409.
Slifka,M.K. and Ahmed,R. (1996a). Limiting dilution analysis of virus-specific memory B cells by an ELISPOT assay. J. Immunol. Methods 199, 37-46.
Slifka,M.K. and Ahmed,R. (1996b). Long-term antibody production is sustained by antibody-secreting cells in the bone marrow following acute viral infection. Ann. N. Y. Acad. Sci. 797, 166-176.
Slifka,M.K., Antia,R., Whitmire,J.K., and Ahmed,R. (1998). Humoral immunity due to long-lived plasma cells. Immunity. 8, 363-372.
Slifka,M.K., Matloubian,M., and Ahmed,R. (1995). Bone marrow is a major site of long- term antibody production after acute viral infection. J. Virol. 69, 1895-1902.
Smith,K.G., Nossal,G.J., and Tarlinton,D.M. (1995). FAS is highly expressed in the germinal center but is not required for regulation of the B-cell response to antigen. Proc. Natl. Acad. Sci. U. S. A 92, 11628-11632.
Snapper,C.M., Peschel,C., and Paul,W.E. (1988). IFN-gamma stimulates IgG2a secretion by murine B cells stimulated with bacterial lipopolysaccharide. J. Immunol. 140, 2121- 2127.
Stalder,T., Hahn,S., and Erb,P. (1994). Fas antigen is the major target molecule for CD4+ T cell-mediated cytotoxicity. J. Immunol. 152, 1127-1133.
Stavnezer,J. (1996). Immunoglobulin class switching. Curr. Opin. Immunol. 8, 199-205.
Stellrecht,K.A. and Vella,A.T. (1992). Evidence for polyclonal B cell activation as the mechanism for LCMV-induced autoimmune hemolytic anemia. Immunol. Lett. 31, 273- 277.
Stevenson,P.G. and Doherty,P.C. (1998). Kinetic analysis of the specific host response to a murine gammaherpesvirus. J. Virol. 72, 943-949.
Stevenson,P.G. and Doherty,P.C. (1999). Non-antigen-specific B-cell activation following murine gammaherpesvirus infection is CD4 independent in vitro but CD4 dependent in vivo. J. Virol. 73, 1075-1079.
Su,H.C., Cousens,L.P., Fast,L.D., Slifka,M.K., Bungiro,R.D., Ahmed,R., and Biron,C.A. (1998). CD4+ and CD8+ T cell interactions in IFN-gamma and IL-4 responses to viral infections: requirements for IL-2. J. Immunol. 160, 5007-5017. 157
Szomolanyi-Tsuda,E., Le,Q.P., Garcea,R.L., and Welsh,R.M. (1998). T-Cell-independent immunoglobulin G responses in vivo are elicited by live-virus infection but not by immunization with viral proteins or virus-like particles. J. Virol. 72, 6665-6670.
Takahashi,Y., Cerasoli,D.M., Dal Porto,J.M., Shimoda,M., Freund,R., Fang,W., Telander,D.G., Malvey,E.N., Mueller,D.L., Behrens,T.W., and Kelsoe,G. (1999). Relaxed negative selection in germinal centers and impaired affinity maturation in bcl-xL transgenic mice. J. Exp. Med. 190, 399-410.
Thomas,W.D. and Hersey,P. (1998). TNF-related apoptosis-inducing ligand (TRAIL) induces apoptosis in Fas ligand-resistant melanoma cells and mediates CD4 T cell killing of target cells. J. Immunol. 161, 2195-2200.
Thomsen,A.R., Johansen,J., Marker,O., and Christensen,J.P. (1996). Exhaustion of CTL memory and recrudescence of viremia in lymphocytic choriomeningitis virus-infected MHC class II-deficient mice and B cell-deficient mice. J. Immunol. 157, 3074-3080.
Thomsen,A.R. and Marker,O. (1988). The complementary roles of cellular and humoral immunity in resistance to re-infection with LCM virus. Immunology 65, 9-15.
Tite,J.P. (1990). Evidence of a role for TNF-alpha in cytolysis by CD4+, class II MHC- restricted cytotoxic T cells. Immunology 71, 208-212.
Traggiai,E., Puzone,R., and Lanzavecchia,A. (2003). Antigen dependent and independent mechanisms that sustain serum antibody levels. Vaccine 21 Suppl 2, S35-S37. van den Broek,M.F., Muller,U., Huang,S., Aguet,M., and Zinkernagel,R.M. (1995). Antiviral defense in mice lacking both alpha/beta and gamma interferon receptors. J. Virol. 69, 4792-4796.
Van den Eertwegh,A.J., Noelle,R.J., Roy,M., Shepherd,D.M., Aruffo,A., Ledbetter,J.A., Boersma,W.J., and Claassen,E. (1993). In vivo CD40-gp39 interactions are essential for thymus-dependent humoral immunity. I. In vivo expression of CD40 ligand, cytokines, and antibody production delineates sites of cognate T-B cell interactions. J. Exp. Med. 178, 1555-1565. van der Most,R.G., Murali-Krishna,K., Whitton,J.L., Oseroff,C., Alexander,J., Southwood,S., Sidney,J., Chesnut,R.W., Sette,A., and Ahmed,R. (1998). Identification of Db- and Kb-restricted subdominant cytotoxic T-cell responses in lymphocytic choriomeningitis virus-infected mice. Virology 240, 158-167.
Varga,S.M. and Welsh,R.M. (1998b). Stability of virus-specific CD4+ T cell frequencies from acute infection into long term memory. J. Immunol. 161, 367-374. 158
Varga,S.M. and Welsh,R.M. (1998a). Detection of a high frequency of virus-specific CD4+ T cells during acute infection with lymphocytic choriomeningitis virus. J. Immunol. 161, 3215-3218.
Varga,S.M. and Welsh,R.M. (2000). High frequency of virus-specific interleukin-2- producing CD4(+) T cells and Th1 dominance during lymphocytic choriomeningitis virus infection. J. Virol. 74, 4429-4432.
von Herrath,M.G., Yokoyama,M., Dockter,J., Oldstone,M.B., and Whitton,J.L. (1996). CD4-deficient mice have reduced levels of memory cytotoxic T lymphocytes after immunization and show diminished resistance to subsequent virus challenge. J. Virol. 70, 1072-1079.
Welsh,R.M. (2000). Lymphocytic Choriomeningitis Virus as a Model for the Study of Cellular Immunology. In Effects of Microbes on the Immune System, M.W.Cunningham and R.S.Fujinami, eds. (Philadelphia: Lippincott Williams & Wilkins), pp. 289-312.
Welsh,R.M., Jr. and Zinkernagel,R.M. (1977). Heterospecific cytotoxic cell activity induced during the first three days of acute lymphocytic choriomeningitis virus infection in mice. Nature 268, 646-648.
Wherry,E.J., Ha,S.J., Kaech,S.M., Haining,W.N., Sarkar,S., Kalia,V., Subramaniam,S., Blattman,J.N., Barber,D.L., and Ahmed,R. (2007). Molecular signature of CD8+ T cell exhaustion during chronic viral infection. Immunity. 27, 670-684.
Whitmire,J.K., Asano,M.S., Murali-Krishna,K., Suresh,M., and Ahmed,R. (1998). Long- term CD4 Th1 and Th2 memory following acute lymphocytic choriomeningitis virus infection. J. Virol. 72, 8281-8288.
Whitmire,J.K., Benning,N., and Whitton,J.L. (2006). Precursor frequency, nonlinear proliferation, and functional maturation of virus-specific CD4+ T cells. J. Immunol. 176, 3028-3036.
Whitmire,J.K., Slifka,M.K., Grewal,I.S., Flavell,R.A., and Ahmed,R. (1996). CD40 ligand-deficient mice generate a normal primary cytotoxic T-lymphocyte response but a defective humoral response to a viral infection. J. Virol. 70, 8375-8381.
Whitton,J.L., Southern,P.J., and Oldstone,M.B. (1988). Analyses of the cytotoxic T lymphocyte responses to glycoprotein and nucleoprotein components of lymphocytic choriomeningitis virus. Virology 162, 321-327.
Wilde,D.B., Marrack,P., Kappler,J., Dialynas,D.P., and Fitch,F.W. (1983). Evidence implicating L3T4 in class II MHC antigen reactivity; monoclonal antibody GK1.5 (anti- L3T4a) blocks class II MHC antigen-specific proliferation, release of lymphokines, and binding by cloned murine helper T lymphocyte lines. J. Immunol. 131, 2178-2183. 159
Williams,N.S. and Engelhard,V.H. (1996). Identification of a population of CD4+ CTL that utilizes a perforin- rather than a Fas ligand-dependent cytotoxic mechanism. J Immunol 156, 153-159.
Williams,N.S. and Engelhard,V.H. (1997). Perforin-dependent cytotoxic activity and lymphokine secretion by CD4+ T cells are regulated by CD8+ T cells. J Immunol 159, 2091-2099.
Woods,A., Monneaux,F., Soulas-Sprauel,P., Muller,S., Martin,T., Korganow,A.S., and Pasquali,J.L. (2007). Influenza virus-induced type I interferon leads to polyclonal B-cell activation but does not break down B-cell tolerance. J. Virol. 81, 12525-12534.
Wright,K.E. and Buchmeier,M.J. (1991). Antiviral antibodies attenuate T-cell-mediated immunopathology following acute lymphocytic choriomeningitis virus infection. J. Virol. 65, 3001-3006.
Wright,K.E., Salvato,M.S., and Buchmeier,M.J. (1989). Neutralizing epitopes of lymphocytic choriomeningitis virus are conformational and require both glycosylation and disulfide bonds for expression. Virology 171, 417-426.
Wright,K.E., Spiro,R.C., Burns,J.W., and Buchmeier,M.J. (1990). Post-translational processing of the glycoproteins of lymphocytic choriomeningitis virus. Virology 177, 175-183.
Yan,M., Marsters,S.A., Grewal,I.S., Wang,H., Ashkenazi,A., and Dixit,V.M. (2000). Identification of a receptor for BLyS demonstrates a crucial role in humoral immunity. Nat. Immunol. 1, 37-41.
Yasukawa,M., Yakushijin,Y., Hasegawa,H., Miyake,M., Hitsumoto,Y., Kimura,S., Takeuchi,N., and Fujita,S. (1993). Expression of perforin and membrane-bound lymphotoxin (tumor necrosis factor-beta) in virus-specific CD4+ human cytotoxic T-cell clones. Blood 81, 1527-1534.
Zajac,A.J., Quinn,D.G., Cohen,P.L., and Frelinger,J.A. (1996). Fas-dependent CD4+ cytotoxic T-cell-mediated pathogenesis during virus infection. Proc. Natl. Acad. Sci. U. S. A 93, 14730-14735.
Zheng,L., Fisher,G., Miller,R.E., Peschon,J., Lynch,D.H., and Lenardo,M.J. (1995). Induction of apoptosis in mature T cells by tumour necrosis factor. Nature 377, 348-351.
Zhou, S., Kurt-Jones, E. A., Cerny, A., Chan, M., Bronson, R. T., and Finberg, R. W. MyD88 intrinsically regulated CD4 T cell responses. J.Immunol. 2008. Ref Type: In Press 160
Zhou,S., Kurt-Jones,E.A., Mandell,L., Cerny,A., Chan,M., Golenbock,D.T., and Finberg,R.W. (2005). MyD88 is critical for the development of innate and adaptive immunity during acute lymphocytic choriomeningitis virus infection. Eur. J. Immunol. 35, 822-830.
Zinkernagel,R.M. and Doherty,P.C. (1974). Restriction of in vitro T cell-mediated cytotoxicity in lymphocytic choriomeningitis within a syngeneic or semiallogeneic system. Nature 248, 701-702.