Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Modulation of human tumor antigen-specific T cell responses by programmed death-1 blockade
(USC Thesis Other)
Modulation of human tumor antigen-specific T cell responses by programmed death-1 blockade
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
MODULATION OF HUMAN TUMOR ANTIGEN-SPECIFIC T CELL
RESPONSES BY PROGRAMMED DEATH-1 BLOCKADE
by
Raymond M. Wong
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(MOLECULAR MICROBIOLOGY AND IMMUNOLOGY)
December 2006
Copyright 2006 Raymond M. Wong
Acknowledgements
Experiments in Chapter 3 were performed in collaboration with Ron Scotland and
Roy Lau. Antibodies to human PD-1 were kindly provided by Drs. Alan Korman
and Changyu Wang from Medarex, Inc. (Milpitas, CA). I thank the Beckman
Immune Monitoring Center at the University of Southern California for use of
facilities.
ii
Table of Contents
Acknowledgements ii
List of Tables iv
List of Figures v
Abstract ix
Chapter 1: Tumor immunology: the immune surveillance theory. 1
Chapter 2: PD-1 expression on peripheral blood melanoma
antigen-specific CTLs from melanoma patients 20
Chapter 3: PD-1 blockade augments the frequencies and absolute
numbers of functional melanoma antigen-specific CTLs 34
Chapter 4: PD-1 blockade augments CTL proliferation but does
not affect CTL apoptosis. 68
Chapter 5: PD-1 blockade augments T cell cytokine secretion and
does not alter surface marker expression on antigen-specific CTLs. 86
Chapter 6: CD4+ T cell help augments antigen-specific human
memory CTL expansion upon ex vivo peptide stimulation. 102
Chapter 7: Conclusion. 112
Materials and methods 126
Bibliography 137
iii
List of Tables
Table 1-1. Tumor antigens recognized by T cells. 6
Table 1-2. PD-1/PD-L1/PD-L2 expression on human cell types. 13
Table 2-1. Summary of PD-1 expression on uncultured
post-vaccination melanoma antigen-specific CTLs. 25
Table 2-2. Summary of PD-1 expression on uncultured
CD8+/tetramer- and CD4+ T cells. 26
Table 3-1. Effector cell count and percent CD4+ and CD8+
cells from PD-1-blocked and IgG4 control-treated cultures. 43
Table 4-1. Effector cell counts from PD-1-blocked and IgG4
control-treated cultures. 73
iv
List of Figures
Figure 1-1. Hypothetical model for PD-1-mediated interactions
between different cell types. 14
Figure 2-1. PD-1 expression on uncultured pre- and post-vaccination
melanoma antigen-specific CTLs. 24
Figure 2-2. Phenotype of DCs used as antigen-presenting cells
for in vitro CTL stimulation. 28
Figure 3-1. In vitro CTL stimulation model. 36
Figure 3-2. Timeline of in vitro CTL stimulation model. 37
Figure 3-3. PD-1 blockade increases the frequencies of
post-vaccination gp100- and MART-1-specific CTLs. 40
Figure 3-4. PD-1 blockade increases the frequencies of
pre-vaccination MART-1-specific CTLs. 41
Figure 3-5. Specificity of melanoma tetramer binding. 42
Figure 3-6. PD-1 blockade increases the absolute numbers of
CD3+, CD8+, CD4+, and melanoma tetramer+ CTLs. 44
Figure 3-7. Dose-dependent effect of the 5C4 anti-PD-1 antibody
on the frequency of melanoma antigen-specific CTLs. 45
Figure 3-8. Dose-dependent effect of the 5C4 anti-PD-1 antibody
on absolute numbers of melanoma antigen-specific CTLs. 46
Figure 3-9. A monovalent 5C4 anti-PD-1 F(ab) fragment
demonstrates similar effects compared to the corresponding
full-length antibody. 47
Figure 3-10. PD-1 blockade increases the frequencies of
IFN- γ-secreting melanoma antigen-specific CTLs from
post-vaccination specimens. 49
v
Figure 3-11. PD-1 blockade increases the frequencies of
IFN- γ-secreting melanoma antigen-specific CTLs from
a pre-vaccination specimen. 50
Figure 3-12. PD-1 blockade augments the generation of effector
cells that lyse peptide-pulsed T2 targets. 51
Figure 3-13. PD-1 blockade augments the generation of effector
cells that lyse melanoma antigen-expressing melanoma targets. 52
Figure 3-14. Peptide stimulation expands CTLs with diverse RE. 55
Figure 3-15. Identification of high RE tumor-reactive CTLs by
MHC:peptide tetramer and CD107a labeling. 56
Figure 3-16. PD-1 blockade increases the fraction of
post-vaccination melanoma antigen-specific CTLs that degranulate
against melanoma targets. 59
Figure 3-17. PD-1 blockade increases the fraction of
pre-vaccination melanoma antigen-specific CTLs that degranulate
against melanoma targets. 60
Figure 3-18. Antibody structure. 65
Figure 4-1. PD-1 blockade affects lymphocyte volume and
granularity during in vitro culture. 73
Figure 4-2 PD-1 blockade increases Ki67 expression in
melanoma antigen-specific CTLs. 74
Figure 4-3. PD-1 blockade increases Ki67 expression in
CD4+ T cells. 75
Figure 4-4. PD-1 blockade augments CFSE dilution by
peptide-stimulated melanoma antigen-specific CTLs 76
Figure 4-5. PD-1 blockade augments CFSE dilution in
CD4+ T cells. 77
vi
Figure 4-6. PD-1 blockade does not significantly affect
Annexin V/7-AAD labeling of peptide-stimulated melanoma
antigen-specific CTLs. 79
Figure 4-7. PD-1 blockade does not significantly affect
Annexin V/7-AAD labeling of CD4+ T cells. 80
Figure 4-8. PD-1 blockade does not significantly affect
active Caspase-3 expression in peptide-stimulated melanoma
antigen-specific CTLs. 81
Figure 4-9. PD-1 blockade does not significantly affect
active Caspase-3 expression in CD4+ T cells. 82
Figure 5-1. Cytokine repertoire and secretion kinetics of
PD-1-blocked and IgG4 control-treated CD3+ T cells. 90
Figure 5-2. PD-1 blockade does not alter melanoma
antigen-specific CD4+ Th1/Th2 balance. 92
Figure 5-3. IL-12 induces selective expansion of antigen-specific
CD4+ Th1 cells. 93
Figure 5-4. Mixed Th1/Th2 polarization of vaccine-induced
MART-1-specific CD4+ T cells in the peripheral blood of
vaccinated melanoma patients. 94
Figure 5-5. Mixed Th1/Th2 polarization of vaccine-induced
MART-1-specific CD4+ T cells after in vitro peptide stimulation. 95
Figure 5-6. Percent expression of surface markers on
melanoma antigen-specific CTLs. 97
Figure 5-7. Mean channel fluorescence of surface markers on
melanoma antigen-specific CTLs. 98
Figure 6-1. CD8+ T memory cells generated in CD4+/+ mice
are T helper-independent during ex vivo peptide stimulation. 104
vii
Figure 6-2. In vitro peptide stimulation model to assess the
requirement for CD4+ T help during human CD8+ T memory
responses during ex vivo peptide stimulation. 105
Figure 6-3. Influenza-specific human CD8+ T memory cells are
T helper-dependent. 107
Figure 6-4. Gp100-specific human CD8+ T memory cells are
T helper-dependent. 108
Figure 6-5. MART-1-specific human CD8+ T memory cells are
T helper-dependent. 109
viii
Abstract
Negative costimulatory signaling mediated via cell surface Programmed Death-1
(PD-1) expression modulates T and B cell activation and is critical for maintaining
peripheral tolerance. Abrogation of the PD-1 pathway may therefore be useful
strategy to enhance the induction of antigen-specific T cells by vaccination. In
these studies, we examined the effects of a fully human PD-1-abrogating antibody
on the in vitro expansion and functional profile of human CD8+ T cells (CTLs)
specific for the melanoma-associated antigens gp100 and MART-1. PD-1 blockade
during peptide stimulation augmented the absolute numbers of CD3+, CD4+,
CD8+, and MHC:peptide tetramer-binding CTLs. This correlated with increased
frequencies of IFN-gamma-secreting antigen-specific cells and augmented lysis of
peptide-pulsed targets as well as gp100+/MART+ melanoma cell lines. PD-1
blockade also increased the fraction of antigen-specific CTLs that recognized
melanoma targets by degranulation, suggesting increased recognition efficiency for
cognate peptide. The increased frequencies and absolute numbers of antigen-
specific CTLs by PD-1 blockade resulted primarily from augmented proliferation,
and not decreased apoptosis. Kinetic analysis of cytokine secretion demonstrated
that PD-1 blockade increased both Type-1 and Type-2 cytokine secretion, without
any apparent skewing of the cytokine repertoire. PD-1 blockade did not
significantly alter the phenotype of peptide-stimulated gp100- and MART-1-
specific CTLs, which were predominantly activated effector / effector memory cells
ix
characterized by a CD45RA(low), CD45RO(high), CCR7(low), CD62L(low), and
CD44(high), expression profile. Vaccine-induced gp100- and MART-1-specific
memory CTLs were also found to be dependent on CD4+ T help for optimal
expansion during in vitro peptide stimulation. These findings suggest that PD-1
blockade may augment CTL expansion directly and/or indirectly through the
augmented provision of CD4+ T help. These findings contribute to the basic
understanding of human CTL expansion and have implications for developing new
cancer immunotherapy strategies.
x
Chapter 1
Tumor immunology: the immune surveillance theory
Introduction:
The immune system serves to protect the body from pathogens and from tumor
development. This defense system depends upon the activity of white blood cells,
or leukocytes, which are derived from precursors in the bone marrow. These cells
give rise to polymorphonuclear leukocytes and macrophages of the innate immune
system and also to lymphocytes of the adaptive immune system. Together, these
cells form a protective barrier against disease. Immunity against microorganisms –
viruses, bacteria, pathogenic fungi, and parasites – represents the classical
immunological paradigm of “self vs. non-self.” Immunity against cancer is a
paradox in that tumor cells often do not fall into the conventional “non-self”
category in the same way that microorganisms do. Rather, tumor immunity appears
to follow a slightly different model of “self vs. altered self.” This concept of an
“altered self” has important implications for the immunotherapy of some human
cancers, as normal cellular proteins are often the basis of immunological
recognition of tumors.
1
What is the evidence that immunity controls cancer?
The immune surveillance theory, first proposed by Frank Burnet and Lewis Thomas
in the 1950s, hypothesizes that the immune system specifically recognizes and
destroys cancerous cells before they become harmful to surrounding tissue (Burnet,
1970). The introduction of this concept was followed by experiments aimed to
determine if hosts with impaired immune systems demonstrated increased
incidences of spontaneous and chemically-induced tumors. Early studies in
thymectomized and athymic nude mice did not provide consensus support for the
role of immunity in controlling tumor development (Grant, 1965; Stutman, 1975;
Nishizuka, 1965; Trainin, 1967; Sanford, 1973; Pantelouris, 1968). Since the
1990s, however, studies using alternative immune-compromised mouse models,
epidemiological studies of cancers among HIV and organ transplant patients, and
the discovery of human tumor antigens recognized by T cells provide evidence that
cellular and serologic immune responses against tumors do exist, as discussed
below.
Mice lacking either the Interferon (IFN)- γ receptor or Signal Transducer and
Activator of Transcription 1 (STAT1, the transcription factor that is crucial for IFN-
γ receptor signaling) were found to be 10-20 times more sensitive than wild-type
mice to the tumor-inducing capacity of 3-methylcholanthrene (MCA) (Kaplan,
1998). Perforin-deficient mice were also shown to develop significantly more
2
tumors than wild-type controls (van den Broek, 1996). Perforin is a key component
of the cytolytic granules of effector T cells and natural killer (NK) cells that
mediate killing of targets, including virus-infected cells and tumor cells (Peitsch,
1991; Lichtenheld, 1988). Additional evidence supporting the existence of tumor
immunosurveillance came from studies with mice deficient in recombination
activation gene (RAG)-1 or RAG-2. RAG-1 and RAG-2 are lymphocyte-specific
enzymes involved in end-joining of double-stranded DNA breaks, and are
responsible for generating antigen receptor diversity (Tonegawa, 1983; Davis,
1988). RAG-1- and RAG-2-deficient mice were shown to form more tumors than
wild-type mice when treated with MCA (Shankaran, 2001).
Immune suppression caused by Human Immunodeficiency Virus (HIV) infection
also provides evidence that the immune system protects against cancer. A study
examining the clinical behavior of malignant melanoma provided evidence that the
tumor demonstrates a more aggressive course in HIV patients compared to matched
HIV-negative patients (Rodrigues, 2002). Organ transplant patients – who undergo
treatment with immunosuppressive drugs such as cyclosporine and azathioprine –
also demonstrate increased frequencies of malignant melanoma and non-melanoma
skin cancers (Jensen, 1999; Naldi, 2000; Moloney, 2006). Taken together, the
observations from mouse models and human cancer patients provide a link between
immune function and cancer risk.
3
The identification of tumor infiltrating lymphocytes (TILs) from clinical tumor
specimens has provided additional evidence of immune responses against tumors.
Studies from the National Cancer Institute (Bethesda, MD, USA) in the 1980s
demonstrated that cellular infiltrates from human melanoma lesions contained TILs
that could be expanded ex vivo using Interleukin-2 (IL-2) and were then capable of
lysing autologous melanoma cells in culture (Topalian, 1989). The clinical use of
ex vivo-expanded TILs for melanoma immunotherapy followed. Adoptive transfer
of TIL populations after lymphodepleting chemotherapy has been shown to mediate
tumor regression in some patients with active metastatic melanoma (Aebersold,
1991; Huang, 2005; Zhou, 2005). Thus, amplification of endogenous anti-tumor T
cell responses via ex vivo manipulation demonstrates proof of concept that immune
responses against tumors naturally exist.
T cells recognize peptide epitopes derived from tumor antigens.
The molecular definition of tumor antigens recognized by T cells has provided new
possibilities for the development of effective immunotherapy for cancer. In 1991,
Thierry Boon and colleagues identified the first tumor-associated antigen
recognized by CD8+ cytolytic T cells (CTLs) (van der Bruggen, 1991). This
protein, termed “melanoma-antigen E-1” (MAGE-1), was found in a variety of
tumor types as well as normal testis and placental tissue, but no other normal tissue.
Other tumor antigens have since been identified using a variety of techniques
4
including (1) “reverse immunology,” whereby specific T cells are generated in vitro
against peptide sequences derived from serologically defined tumor antigens, (2)
acid stripping of peptides from tumor cells followed by mass spectroscopy, (3)
serological analysis of cDNA expression libraries (SEREX), (4) molecular cloning
(5) proteomics, and (6) DNA microarray analysis. To date, over 70 tumor antigens
have been identified for a multitude of cancers (Renkvist, 2001). Collectively,
these studies have demonstrated that true tumor regression antigens exist, and
provided additional rationale for their use in treating cancer.
Tumor antigens are broadly classified into two categories: (1) tumor-specific and
(2) tumor-associated. They can be thought of as comprising five specific groups
(Table 1-1). Tumor-specific antigens include the E6 and E7 oncoproteins of
Human Papilloma Virus (HPV), a transforming virus associated with human
cervical carcinomas (Bosch, 1995). Cells that are transformed by the same virus –
such as HPV – express antigens that are distinguished as “non-self” since they are
not present in normal tissue. Mutagenic events in tumors can give rise to novel
epitopes that are, in turn, also recognized as “non-self” within the context of MHC.
Tumors bearing such mutations are antigenically unique. Thus, immune
recognition is individually tumor-specific since the likelihood of the same mutation
occurring independently in tumors from different patients is low.
5
Table 1-1. Tumor antigens recognized by T cells (Modified from Renkvist,
2001).
Class Antigen Tumor
Viral HPV E6, E7 cervical
EBV LMP1, 2 lymphoma, nasopharyngeal, gastric
HTLV Env lymphoma, leukemia
Cancer-Testis MAGE melanoma, breast, colon, bladder,
head/neck
BAGE melanoma, breast, bladder
RAGE melanoma, breast, colon, bladder,
renal
GAGE melanoma, breast, lung, esophageal,
leukemia
LAGE lung
SAGE sarcomas
NY-ESO-1 melanoma, breast, lymphoma,
bladder, lung, prostate, ovarian,
SSX-2 thyroid
Differentiation Melan A/MART-1 melanoma
pMel17/gp100 melanoma
Tyrosinase melanoma
PSA prostate
CEA melanoma, breast, lung, pancreas,
gastric, rectum
TRP-1,2 melanoma
HER-2/Neu melanoma, breast, ovarian, gastric,
pancreatic
Mutations HPV E6, E7 cervical
MUM-1, 2, 3 melanoma
β-catenin melanoma
p16 melanoma
Carbohydrate MUC-1, 2 breast, colon, ovarian, lymphoma,
pancreas
6
Abnormal post-translational modifications of normal cellular proteins can also
result in antigenic differences between normal and tumor cells. Mucin 1 (MUC-1),
for example, is a transmembrane protein on ductal epithelial cells that is normally
heavily glycosylated (Lan, 1990). Loss of the glycosylation pattern exposes the
mucin peptide backbone to T cell recognition, which involves MHC-independent
interaction with the αβ TCR in a manner analogous to TCR interaction with
bacterial superantigens (Barnd, 1989). MUC-1 can also be a source of peptides
presented in association with MHC class I (Apostolopoulos, 1997).
Most tumor-associated antigens examined thus far result from de novo or over-
expression of normal cellular proteins. They are present on both transformed and
normal tissue, but their appearance on the latter are under conditions (such as lack
of MHC expression or low MHC:peptide concentration) that result in tolerance.
Cancer-testis antigens (such as the MAGE family, SSX-2 and NY-ESO-1) are
normally found only in specialized tissues such as the testes and placenta, but are
often aberrantly expressed in a variety of different tumor types (Chen, 1998).
Differentiation antigens are expressed only during a certain stage of tissue
development. Tumors arising from a particular tissue type will often express
differentiation antigens characteristic of that tissue. For example, Melan A/MART-
1 and pMel17/gp100 are melanosome-related differentiation antigens that are
commonly found in melanoma cells (Kawakami, 1996). These antigens may be
7
uniquely expressed by an individual tumor, although some appear to be shared
among tumors and tissues of varying origins.
Antigen-presenting cells and T cell activation.
Dendritic cells (DCs) are generally considered the most potent antigen-presenting
cells (APCs) in humans (Hart, 1997; Banchereau, 1998). DCs are derived from
myeloid progenitors within the bone marrow and initially have an immature
phenotype characterized by low surface expression of both major histocompatibility
complex (MHC) and B7 co-stimulatory molecules. Immature DCs are not potent
stimulators of naïve T cells, but are very active in capturing antigens by
phagocytosis and macropinocytosis (Sallusto, 1995). After persisting at sites of
infection for a variable length of time, immature DCs migrate via the lymphatics to
the secondary lymphoid tissues where they may attain a mature phenotype during
interaction with T cells (Mailliard, 2002; Schoenberger, 1998). Mature DCs no
longer take up antigen efficiently but express high levels of MHC class I and class
II proteins (Hart, 1997; Banchereau, 1998). They also express high levels of B7,
other co-stimulatory and adhesion molecules, and secrete chemokines that
specifically attract T cells (Hart, 1997; Banchereau, 1998). These properties help
explain their ability to prime naïve T cells and stimulate robust T cell clonal
expansion.
8
Specific rejection of tumor tissue requires that transformed cells be distinguished
immunologically from their normal counterparts. Peptide fragments presented on
the cell surface by MHC molecules is one basis for specific recognition. Tumor
cells present peptides derived from endogenous and foreign proteins, such as those
produced by viruses. However, tumor cells are often not intrinsically immunogenic
due to lack of B7 co-stimulatory molecules and adhesion molecules necessary to
interact with and activate naïve T cells (Chen, 1993). T cell receptor (TCR)
signaling in the absence of co-stimulation not only fails to activate naïve T cells, but
also leads to anergy (peripheral tolerance) (Appleman, 2003). Such self-tolerance
can be broken by an intermediary process of immune stimulation initiated by
professional APCs such as DCs. DCs presenting self peptides and expressing
sufficient levels of co-stimulatory molecules can activate naïve T cells (including
self-reactive T cells) at the tumor site or distantly in secondary lymphoid organs.
Cancer vaccine strategies must exploit this process directly or indirectly in order to
break tolerance and generate immune responses against tumor antigens.
Negative costimulation and the Programmed Death-1 pathway.
The CD28 family of costimulatory molecules is involved in fine-tuning immune
responses in both normal and pathological settings. These molecules provide
critical positive and negative signals that initiate, sustain, attenuate, and/or
terminate lymphocyte responses. Two major groups of costimulatory receptors that
9
modulate T cells have been described: (1) the Immunoglobulin (Ig) superfamily
that includes CD28, Programmed Death-1 (PD-1), Cytotoxic T Lymphocyte
antigen-4 (CTLA-4), B- and T-lymphocyte attenuator-4, B7-H3, and ICOS, and (2)
the Tumor Necrosis Factor (TNF) superfamily that includes OX40, CD27, 4-1BB,
CD30, and herpes-virus entry mediator (Peggs, 2005). Negative costimulatory
members of the Ig superfamily, CTLA-4 and PD-1, are known to restrict immune
responses against self antigens.
Programmed Death-1 (PD-1) is a recently defined Ig superfamily member related to
CD28 and cytotoxic T lymphocyte antigen-4 (CTLA-4). It is a 50-55 kDa
monomeric transmembrane protein that was originally identified in a mouse T cell
hybridoma undergoing activation-induced cell death (Ishida, 1992). PD-1
expression has been found on CD4+/CD8+ thymocytes, mature T and B cells
following activation, and occasionally macrophages (Keir, 2005). Experimental
evidence strongly suggests that the PD-1 pathway impacts negatively on T cell
activation (Agata, 1996; Nishimura, 2001). The PD-1 pathway is known to down-
modulate immune responses to self antigens, T cell proliferation, cytokine
secretion, and inhibit T cell-mediated lysis of tumor cells in mice (Nishimura, 1999;
Latchman, 2004; Nishimura, 2001; Salama, 2003; Curiel, 2003; Cai, 2004; Blank,
2006; Hirano, 2005; Iwai, 2005; Iwai, 2002). PD-1-mediated inhibition of T cells
appears to be dependent on T cell receptor (TCR) signaling. Ligation of the PD-1
10
molecule on T cells induces recruitment of Src-homology Phosphatase 1 and Src-
homology Phosphatase 2 to the immunoreceptor tyrosine-based switch motif of the
PD-1 cytoplasmic domain (Chemnitz, 2004). These phosphatases likely inhibit
proximal signaling kinases of the TCR pathway, thereby blocking TCR signal
transduction and leading to attenuated T cell activation (Chemnitz, 2004). Two
natural ligands for PD-1 have been described: Programmed Death Ligand-1 (PD-
L1; also known as B7-H1) and Programmed Death Ligand-2 (PD-L2; also known
as B7-DC), both being members of the Ig superfamily. PD-L1 is expressed on
resting and upregulated on activated B, T, myeloid, and DCs (Liang, 2003). PD-L2
is found primarily on macrophages and DCs, but low levels are occasionally seen
on activated T cells (Liang, 2003; Brown 2003). It has also been suggested that
undefined receptors, other than PD-1, may also interact with PD-L1 and/or PD-L2
(Greenwald, 2005; Chen, 2004).
Previously published and ongoing cancer vaccine trials have demonstrated that
CD4+ and CD8+ T cell responses to tumor antigens are often induced in patients
after vaccination (Mocellin, 2003; Rosenberg, 2004; Morse, 2005). As measured
by enzyme-linked immunospot (ELISPOT) assay and MHC:peptide tetramer
labeling from the peripheral blood (and occasionally sentinel immunized nodes and
tumors), the induction of immune responses to immunizing tumor antigens does not
clearly confer clinical benefit (Mocellin, 2003; Rosenberg, 2004; Morse, 2005). It
11
has been suggested that the majority of vaccine-induced T cells have low
recognition efficiency (RE; also known as “functional avidity”) for cognate peptide
presented endogenously by tumor cells, which may account for their lack of clinical
effect (Rubio, 2003; Stuge, 2004; Kohrt, 2005). The concentration of individual
peptide epitopes can vary on tumors depending on factors such as antigen
expression level, overall endogenous antigen processing capacity, and overall MHC
expression. High RE tumor antigen-specific CD8+ T cell (CTL) clones have been
suggested to require 10
-12
to 10
-14
molar cognate peptide concentration to induce
half-maximal target cell lysis in vitro, while low RE CTL clones require 10
-8
to 10
-9
molar concentration to achieve the same (Sadovnikova, 1996). Thus, cancer
vaccine strategies may be improved not only by increasing the total quantity of
induced T cells, but by also increasing their RE for cognate peptide presented by
tumor cells. It has been suggested that modulation of costimulatory signals – via
CTLA-4 blockade and enhanced CD80-mediated costimulation – in vaccinated
mice can selectively induce high RE CTLs that are specific for self and non-self
antigens (Oh, 2003; Hodge, 2005). We hypothesized that PD-1 blockade would
have a favorable effect on the expansion and RE of human CTLs specific for tumor-
associated antigens (i.e. self antigens). Herein, we examined the effects of PD-1
blockade on the expansion, phenotype, and RE of in vitro-stimulated CTLs from
vaccinated melanoma patients. The cytokine repertoire and secretion kinetics of
responding PD-1-blocked T cells were also assessed.
12
Table 1-2. PD-1/PD-L1/PD-L2 expression on human cell types (Greenwald,
2005).
13
Figure 1-1. Hypothetical model for PD-1-mediated interactions between
different cell types.
14
The role of CD4+ T help in CTL expansion and tumor immunity.
The production of antibodies is best adapted to generate immunity against
extracellular pathogens such as viruses and bacteria. Tumors, on the other hand,
typically express endogenous antigens that are mutated or true self proteins. As
such, T cell-mediated responses are best-suited for defense against tumors.
Traditionally, this function has been associated with MHC class I-restricted CTLs
with cytolytic activity. However, there is abundant evidence that CD4+ T helper
cells can play a role in protective anti-tumor responses. It has been suggested that
augmented cellular immunity against tumors may involve direct recognition of
MHC class II-restricted peptide epitopes at a tumor site, modulation of CTL
activation via induced DC maturation, and maintenance of CTL proliferation during
the effector phase of an immune response (Giuntoli, 2002; Knutson, 2005).
CD4+ T cells appear to orchestrate the maintenance of adaptive CTL responses.
Recent work has demonstrated that the establishment of long-term CTLs in mice is
dependent on cross-presentation of antigen to CD4+ T cells during the initial
priming phase of naïve CTLs (Shedlock, 2003; Janssen, 2005). Interestingly, once
memory CTLs in mice are established via CD4+ T help during primary activation,
CD4+ T help becomes dispensable during secondary encounter with antigen ex vivo
and also in vivo after adoptive transfer to host mice (Shedlock, 2003; Janssen,
2005). The applicability of this principle to human CTLs has yet to be established.
15
It has been previously reported that immune responses in cancer vaccine patients
skewed towards Th2-type responses are correlated with disease progression in
melanoma patients, while normal donors and patients who remained disease-free
exhibited highly-polarized Th1 or mixed Th1/Th2 responses (Tatsumi, 2002). Th1-
type CD4+ T cells secreting IFN- γ appear important to the optimal generation and
durability of tumor antigen-specific CTLs in vivo and may also recruit these
effector cells into the tumor microenvironment (Hung, 1998). Thus, there is a
strong rationale for the selective generation of tumor antigen-specific CD4+ Th1
cells by vaccination.
16
Overall goal:
To characterize the effects of PD-1 blockade on the expansion and functional
profile of melanoma antigen-specific CTLs. In this work we have developed an in
vitro stimulation model to assess the effects of antibody-mediated PD-1 blockade
on gp100- and MART-1-specific CTLs derived from the peripheral blood of
melanoma patients (pre- and post-vaccination). PD-1 blockade augmented the
expansion of functional melanoma antigen-specific CTLs and increased the
proportion of specific CTLs that recognized melanoma cells by degranulation.
Specific aims:
1. Analyze PD-1 expression on peripheral blood gp100- and MART-1-
specific CTLs from melanoma patients.
2. To determine the effect of antibody-mediated PD-1 blockade on the
expansion of gp100 and MART-1-specific CTLs.
• In Chapter 3, we established an in vitro model system to assess
the effects of PD-1 blockade on the expansion of melanoma
antigen-specific CTLs. As antigen-presenting cells, we used
monocyte-derived DCs pulsed with melanoma peptides. These
DCs were used to stimulate autologous peripheral blood CD3+ T
17
cells from vaccinated melanoma patients. After 10-11 days of
stimulation, effector cells were collected and analyzed by (1)
MHC:peptide tetramer labeling, (2) ELISPOT, (3) cytolysis assay,
and (4) degranulation assay.
3. To determine if PD-1 blockade augments the generation of antigen-
specific CTLs and/or decreases cell death.
• In Chapter 4, we analyzed proliferation and apoptosis of antigen-
specific CTLs treated with a PD-1-abrogating antibody.
Proliferation was measured by intracellular Ki67 staining and
CFSE dilution. Apoptosis was measured by Annexin V/7-AAD
labeling and intracellular active Caspase-3 expression.
4. To determine if PD-1 blockade alters the overall Type-1 vs. Type-2
cytokine balance of human CD3+ T cells and antigen-specific CD4+
Th1/Th2 ratio.
5. To examine the overall phenotype of PD-1-blocked gp100- and MART-1
specific CTLs.
18
6. To determine if the expansion of human memory CTLs specific for
influenza and melanoma antigens is dependent on CD4+ T help during ex
vivo stimulation with peptide.
• In Chapter 6, we developed an in vitro stimulation model similar
to that used to study the effect of PD-1 blockade on antigen-
specific CTL expansion. We assessed requirement of human
peripheral blood memory CTLs for CD4+ T help during ex vivo
antigen-driven expansion.
19
Chapter 2
PD-1 expression on peripheral blood melanoma antigen-specific CTLs
from melanoma patients
Introduction:
Programmed Death-1 (PD-1) is a recently defined immunoglobulin (Ig) superfamily
member related to CD28 and cytotoxic T lymphocyte antigen-4 (CTLA-4). It is a
50-55 kDa transmembrane protein that was originally identified in a mouse T cell
hybridoma undergoing activation-induced cell death (Ishida, 1992). PD-1
expression has been found on CD4+/CD8+ thymocytes, mature T and B cells
following activation, and occasionally macrophages (Keir, 2005; Agata, 1996;
Nishimura, 2001). Experimental evidence strongly suggests that the PD-1 pathway
impacts negatively on T cell activation (Greenwald, 2005; Chen, 2004). The PD-1
pathway is known to downmodulate immune responses to self antigens, T cell
proliferation, cytokine secretion, and inhibit T cell-mediated lysis of tumor cells in
mice (Nishimura, 1999; Latchman, 2004; Nishimura, 2001; Salama, 2003; Curiel,
2003; Cai, 2004; Blank, 2006; Hirano, 2005; Iwai, 2005; Iwai, 2002).
Expression of PD-1 during activation of naïve and memory CTLs in humans may
inhibit their activation and expansion, especially when antigen is presented by PD-
Ligand-expressing antigen presenting cells such as DCs. Activated T cells also
20
express PD-L1 and occasionally PD-L2 (Liang, 2003; Brown 2003). It has been
suggested that T cell:T cell interactions via PD-1 naturally occur, and results in
inhibition of T cell proliferation and cytokine secretion (Seo, 2006). Our first goal
was to analyze variable expression of PD-1 on melanoma antigen-specific CTLs
from the peripheral blood of melanoma patients, and PD-L1/PD-L2 expression on
monocyte-derived DCs that were used as antigen-presenting cells in our in vitro
stimulation model described in the next chapter.
21
Results:
PD-1 is expressed on Pre- and Post-vaccination melanoma antigen-specific
CTLs.
We analyzed PD-1 expression on post-vaccination CTLs from 9 melanoma patients.
Patient specimens were selected on the basis of positive melanoma MHC:peptide
tetramer staining, in order to obtain measurable populations of antigen-specific
CTLs for PD-1 expression analysis. MHC:peptide tetramers are recombinant MHC
molecules folded with a peptide of interest (i.e. from gp100 and MART-1), and are
used to detect antigen-specific T cells among heterogeneous cell populations
(Altman, 1996). Pre-vaccination tetramer labeling for Patients 1-9 were below
detectable levels, and therefore were not analyzed for PD-1 expression. Significant
levels of peripheral blood tumor antigen-specific CTLs are occasionally seen in
melanoma patients without vaccine manipulations (Khong, 2002). One such
specimen (Patient 10 listed in Table 2-1) with a significant frequency of MART-1-
specific CTLs (2.4% of CD8+ T cells) was identified in our clinical specimen
inventory and was also analyzed for PD-1 expression.
Melanoma antigen-specific (gp100 and MART-1) CTLs in freshly-thawed,
uncultured PBMC specimens were detected with HLA-A*0201-PE tetramers and
co-labeled with an anti-PD-1-FITC antibody. As shown in Figure 2-1, tetramer+
CTLs from fresh PBMCs expressed PD-1. PD-1 expression on the post-vaccination
22
tetramer+ CTLs (Patients 1-9) varied considerably between patients, ranging from
30.8 to 84.6% positive in the post-vaccination specimens we analyzed (Table 2-1).
MART-1-specific CTLs from Patient 10 – who demonstrated a significant
frequency of pre-existing MART-1-specific CTLs (i.e. before vaccine
manipulations) – were 100% positive for PD-1 expression. The average PD-1
expression on tetramer+ CTLs from the 10 analyzed patient specimens was 58.7%
+/- 23.7.
PD-1 expression was also analyzed on the CD8+/tetramer- and CD4+ populations
for each patient specimen (Table 2-3). The fraction of PD-1 expressing cells among
the CD8+/tetramer- populations ranged from 0% to 40.8%, with an average of
14.7% +/- 13.4. The fraction of PD-1-expressing cells among the CD4+
populations ranged from 0.8% to 5.8%, with an average of 4.3% +/- 1.6. The
observed variations in PD-1 expression across patient samples for melanoma
tetramer+ CTLs, melanoma tetramer- CTLs, and CD4+ cells may partly reflect
patient-specific differences in PD-1 gene polymorphisms (Kroner, 2005).
23
Figure 2-1. PD-1 expression on uncultured post-vaccination melanoma
antigen-specific CTLs. Tetramer+ CTLs that were acquired by gating on the
CD3+/CD8+ population. Data shown represent the lowest and highest percent PD-
1 expression among the patient specimens analyzed. Solid gray histograms: isotype
controls. Black histograms: PD-1 expression.
24
% Tetramer + Fraction PD-1+
Patient 1 (gp100-2M; post) 1.2 30.8%
Patient 2 (gp100-2M; post) 1.4 84.6%
Patient 3 (gp100-2M; post) 0.2 50.0%
Patient 4 (gp100-2M; post) 0.6 40.0%
Patient 5 (gp100-2M; post) 0.2 50.0%
Patient 6 (MART-27L; post) 0.6 83.3%
Patient 7 (gp100-2M; post) 0.3 66.6%
Patient 8 (gp100-2M; post) 0.2 50.0%
Patient 9 (MART-27L; post) 0.3 33.3%
Patient 10 (MART-27L; pre) 2.4 100%
Avg: 58.7% +/- 23.7
Table 2-1. Summary of PD-1 expression on uncultured pre- and post-
vaccination melanoma antigen-specific CTLs. PD-1 expression was measured on
peripheral blood tetramer+ CTLs that were acquired by gating on the CD8+
population. Post: post-vaccination. Pre: pre-vaccination.
25
Fraction PD-1+ Fraction PD-1+
(CD8+/Tetramer-) (CD4+)
Patient 1 (post) 8.4% 3.8%
Patient 2 (post) 1.9% 5.8%
Patient 3 (post) 9.7% 4.6%
Patient 4 (post) 6.1% 3.4%
Patient 5 (post) 9% 4.0%
Patient 6 (post) 40.8% 7.2%
Patient 7 (post) 17% 4.3%
Patient 8 (post) 0.0% 0.8%
Patient 9 (post) 21.7% 4.7%
Patient 10 (pre) 32.8% 4.3%
Avg: 14.7% +/- 13.4 4.3 % +/- 1.6
Table 2-2. Summary of PD-1 expression on uncultured CD8+/tetramer- and
CD4+ T cells. PD-1 expression was measured on peripheral blood CD8+/tetramer-
and CD4+ T cells. Post: post-vaccination. Pre: pre-vaccination.
26
PD-L1 is expressed on human monocyte-derived DCs.
Adherent human monocytes cultured in vitro with GM-CSF and IL-4 for 6-7 days
have an intermediate immature/mature phenotype and have a high capacity to
capture and process soluble antigen (Sallusto, 1995). DCs found in lymphoid
tissues vary in maturation state (Wilson, 2003; Shortman, 2002). As such, we used
monocyte-derived DCs with an intermediate immature/mature phenotype as
antigen-presenting cells for our in vitro stimulation model. These DCs expressed
CD80, CD86, HLA-A,B,C, HLA-DR, and PD-L1 (Figure 2-2). The expression of
PD-L1 on monocyte-derived DCs suggests that DCs intrinsically possess inhibitory
functions that balance the strength of MHC:peptide- and costimulatory signals
delivered to responding T cells.
27
Figure 2-2. Phenotype of DCs used as antigen-presenting cells for in vitro CTL
stimulation. DCs were differentiated using GM-CSF and IL-4 (1000 U/ml each),
and stained with fluorescently-conjugated antibodies to measure expression of
maturation markers. Grey: isotype controls. Black: PD-1.
28
Discussion:
Efficient secondary expansion of antigen-specific memory T cells is important for
long-term protection against tumors and viral infections. We show here that such T
cells in the peripheral blood of vaccinated melanoma patients express PD-1 to
varying degrees. The level of PD-1 expression on gp100- and MART-1-specific
CTLs ranged from 30.8% to 84.6%. This did not appear to correlate with the
frequency of tetramer+ cells, suggesting that increased PD-1 expression is unlikely
the result of increased antigen-specific CTL induction by vaccination.
Interestingly, pre-vaccination MART-1-specific CTLs from Patient 10 were 100%
positive for PD-1 expression, unlike the 9 post-vaccination specimens (Patients 1-9)
analyzed. Significant frequencies of tumor antigen-specific CTLs are occasionally
observed in melanoma patient peripheral blood prior to vaccine manipulations
(Khong, 2002). In our clinical specimen inventory, such samples are rare.
However, one pre-vaccination sample was located and analyzed for PD-1
expression (Patient 10).
It remains to be determined if endogenous (i.e. pre-vaccination) MART-1- and
other tumor antigen-specific CTLs from cancer patients are consistently at or near
100% positive for PD-1 expression. Such information may reveal a potential
immune-escape mechanism, whereby endogenous T cell responses against tumors
are induced to express high levels of PD-1. In a recent study in mice, it was
29
suggested that chronic stimulation of CTLs by Lymphocytic Choriomeningitis
Virus (LCMV) infection results in marked upregulation of PD-1 (~100% PD-1
positive) on LCMV- specific CTLs (Barber, 2006). Furthermore, PD-1 expression
is markedly upregulated on HIV-specific CTLs in HIV+ patients – with an average
~90% PD-1 positive – and is associated with disease progression (Day, 2006). It
remains to be determined if tumor burden also results in similar upregulation of PD-
1 on tumor-antigen specific T cells. A possible method to address that question
would be to measure PD-1 expression on TILs or splenocytes from mice with
experimentally-induced tumors.
On a similar note, repetitive immunization may also induce PD-1 upregulation on
antigen-specific T cells, analogous to CTLs that are chronically exposed to LCMV
or HIV (Barber, 2006; Day, 2006). Cancer vaccine regimens, such as those
referenced in this publication, often involve repetitive immunizations – from
weekly to monthly treatments (Lee, 2001; Lau, 2001; Gajewski, 2001; Cebon,
2003; Peterson, 2003; Wong, 2004). In Table 2-1, all post-vaccination melanoma
antigen-specific CTLs analyzed expressed PD-1 to varying degrees (~30-85%
positive). Analysis of PD-1 expression on antigen-specific CTLs arising from acute
infection – such as influenza whereby the infecting virus is cleared – from the
patient specimens analyzed may allow us to determine if repetitive vaccinations
with peptides is associated with PD-1 upregulation on cognate CTLs. It is expected
30
that influenza-specific CTLs from the same patients analyzed here would express
low levels of PD-1 (e.g. <10% positive), similar to LCMV-specific CTLs in mice
during acute infection stages (Barber, 2006).
Upregulation of PD-1 has been suggested to be indicative of T cell exhaustion,
characterized by impaired proliferative capacity and effector function (Barber,
2006; Day, 2006). Hence, repetitive exposure to antigen may generally lead to
exhaustion of T cells. Taken together, these observations may bring to question the
clinical immunization protocols currently being used cancer patients, whereby T
cells are repetitively exposed to immunizing antigens. Nonetheless, the data
presented in the next chapter demonstrate that antibody-mediated PD-1 abrogation
enhances the in vitro expansion and effector function of melanoma peptide-
stimulated CTLs.
For comparison, PD-1 expression was also measured on the CD8+/tetramer- and
CD4+ populations from each patient specimen. Percent PD-1 expression was
significantly lower on these populations, compared to tetramer+ CTLs. The
fraction of PD-1-expressing cells among the CD8+/tetramer- populations ranged
from 0% to 40.8%, with an average of 14.7% +/- 13.4. The fraction of PD-1-
expressing cells among the CD4+ populations ranged from 0.8% to 5.8%, with an
average of 4.3% +/- 1.6. These are compared to an average of 58.7% +/- 23.7 PD-1
31
expression on melanoma antigen-specific CTLs. These observations suggest that
antibody-mediated PD-1 abrogation during stimulation with melanoma peptides in
vitro and in vivo would preferentially affect melanoma antigen-specific CTLs, and
that toxicity of a PD-1-abrogating antibody may thus be limited due to
comparatively low expression of PD-1 on the general CD4+ and CD8+ T cell
population.
The observed variations in PD-1 expression across patient samples for melanoma
tetramer+ CTLs, melanoma tetramer- CTLs, and CD4+ cells may partly reflect
patient-specific differences in PD-1 gene polymorphisms (Kroner, 2005).
It was previously reported that the human PD-1 gene is polymorphic at an enhancer
within the fourth intron (Kroner, 2005). This single nucleotide polymorphism
disrupts binding of the Runx1 transcription factor, thereby altering regulation of
gene expression (Prokunina, 2002). Differences in PD-1 gene polymorphism may
partially explain the variation in percent PD-1 expression on various T cell
populations between patients.
The expression of PD-1 on melanoma antigen-specific T cells may limit their
expansion in vitro and also in vivo. Ligation of PD-1 receptors on T cells by PD-
L1, PD-L2, and possibly additional undefined ligands, can occur during DC:T cell
interactions as well as during T cell:T cell interactions (Brown, 2003; Seo, 2006).
32
The relative impact of the two types of interactions on antigen-specific T cell
expansion has yet to be defined. Nonetheless, we have demonstrated that PD-1 is
expressed on circulating melanoma antigen-specific CTLs from the peripheral
blood of melanoma patients before and after vaccination. In the next chapter, we
assess the effects of antibody-mediated PD-1 blockade on the in vitro expansion
and functional capacity of these CTLs.
33
Chapter 3
PD-1 blockade augments the frequencies and absolute numbers of functional
melanoma antigen-specific CTLs
Introduction:
In the previous chapter, we demonstrated that peripheral blood gp100- and MART-
1-specific CTLs from vaccinated melanoma patients expressed PD-1.
Ligation of PD-1 on T cells has been suggested to inhibit glucose metabolism and
kinase activity associated with T cell receptor signaling – both being critical
processes for T cell activation and subsequent clonal expansion (Parry, 2005).
Hence, PD-1 expression on melanoma antigen-specific CTLs may limit their
expansion during antigen-driven expansion. Ligation of PD-1 receptors on T cells
by PD-L1, PD-L2, and possibly additional undefined ligands, can occur during
DC:T cell interactions as well as during T cell:T cell interactions (Brown, 2003;
Seo, 2006).
It is generally understood that the strength of stimulatory and inhibitory signals
from DCs affects T cell clonal expansion and effector function. PD-ligand
expression on DCs represents an inhibitory signal that is transduced to T cells
during antigen-presentation, and thus modulates the net stimulatory capacity of
DCs. It is also now known that T cells have autoregulatory characteristics,
34
including “fratricidal” interactions mediated via death receptor interactions.
Pathways involved in T cell:T cell-mediated death include the Fas pathway and the
TNF-alpha-related apoptosis inducing ligand (TRAIL) pathway (Li, 2002; Janssen,
2005). The PD-1:PD-ligand pathway represents another T cell:T cell regulatory
mechanism. Thus, PD-ligands can regulate T cell activity at the level of DC:T cell
interactions and T cell:T cell interactions.
We hypothesized that in vitro expansion of melanoma antigen-specific CD8+ T
cells would be augmented by antibody-mediated PD-1 blockade. A PD-1-
abrogating antibody (known as antibody 5C4 from Medarex, Inc., Milpitas, CA)
was used to assess if PD-1 blockade during peptide stimulation would enhance the
expansion of functional antigen-specific CTLs.
35
Figure 3-1. In vitro CTL stimulation model. Peptide-pulsed monocyte-derived
DCs were used to stimulate autologous peripheral blood CD3+ T cells. CD3+ T
cells were affinity purified from peripheral blood mononuclear cells.
36
Figure 3-2. Timeline of in vitro CTL stimulation model. Peptide-pulsed DCs
were used to stimulated autologous CD3+ T cells. Accumulation of Type-1 and
Type-2 cytokines were measured from supernatant samples taken on days 2, 4, 6, 8,
and 11 of culture. On days 10-11, effector cells were harvested for functional
analysis.
37
Results:
PD-1 blockade augments the frequencies and absolute numbers of functional
melanoma antigen-specific CTLs.
Using MHC:peptide tetramers, we found that PD-1 blockade increased the
frequencies of post-vaccination gp100- and MART-1-specific CTLs (leftward
column in Figure 3-3), compared to IgG4 control-treated cells (center column in
Figure 3-3). The frequency of pre-vaccination MART-1-specific CTLs was also
increased by PD-1 blockade (Figure 3-4). Binding with a pan-MHC class negative
control tetramer was 0.0% for all samples (Figure 3-5). Increases in absolute
numbers of CD3+ cells ranged from approximately 2- to 4-fold, CD8+ cells
approximately 1.5- to 5-fold, CD4+ cells approximately 1.2- to 4-fold, and
tetramer+ CTLs approximately 5- to 35-fold (Figure 3-6). PD-1 blockade did not
significantly affect the percent CD4+ and CD8+ expression, compared to controls
(Table 3-1). The ratio of CD4+ to CD8+ expression did not significantly change
after 11 days of culture compared to the starting population on the initial day (day
0) of culture (Table 3-1). This suggests that PD-1 blockade augments the
proliferation of both the total CD4+ and CD8+ cell populations, as will be discussed
in Chapter 4.
The augmented expansion of gp100- and MART-1-specific CTLs was dependent on
specific peptide stimulation, as PD-1-blocked cells stimulated with a control HIV
38
peptide did not demonstrate augmented expansion of CTLs specific for gp100 and
MART-1 (rightward columns in Figure 3-3 and rightward columns on individual
plots in Figure 3-6). As determined by MHC:peptide tetramer labeling and cell
counting, the minimal effective dose of the 5C4 PD-1-abrogating antibody was
between 10
-1
to 10
-2
µg/ml final concentration, with saturation reached at 10
µg/ml
(Figures 3-7 and 3-8). A monovalent 5C4 PD-1-abrogating F(ab) fragment was
also tested in our model, and demonstrated similar activity to the corresponding full
length antibody (Figure 3-9).
39
Figure 3-3. PD-1 blockade increases the frequencies of post-vaccination gp100-
and MART-1-specific CTLs. HLA-A*0201 MHC:peptide tetramer labeling of
peptide-stimulated CTLs. CD3+ cells were stimulated with autologous peptide-
pulsed DCs for 11 days prior to labeling. Data shown are for 2 patient specimens (7
total), representing gp100- and MART-1-specific immune responses. Stimulation
conditions are shown above each column, with stimulating peptides shown in
parentheses. Numbers shown in each dot plot represent the percentage of tetramer+
CTLs among the total CD3+ population.
40
Figure 3-4. PD-1 blockade increases the frequencies of pre-vaccination
MART-1-specific CTLs. HLA-A*0201 MHC:peptide tetramer labeling of
peptide-stimulated CTLs from a pre-vaccine specimen (Patient 10 listed in Table 2-
1). CD3+ cells were stimulated with autologous peptide-pulsed DCs for 11 days
prior to labeling. Stimulation conditions are shown above each column, with
stimulating peptides shown in parentheses. Numbers shown in each dot plot
represent the percentage of tetramer+ CTLs among the total CD3+ population.
41
Figure 3-5. Specificity of melanoma tetramer binding. HLA-A*0201
MHC:peptide tetramer labeling of peptide-stimulated CTLs. CD3+ cells were
stimulated with autologous peptide-pulsed DCs for 11 days prior to labeling. Data
shown are representative of 2 patient specimens. Stimulation conditions are shown
above each column, with stimulating peptides shown in parentheses. Numbers
shown in each dot plot represent the percentage of tetramer+ CTLs among the total
CD3+ population. Top panel: melanoma tetramer. Bottom panel: Pan-MHC class
I negative control tetramer.
42
CD3+ count CD3+ count
Day 0 Day 11
PD-1 (gp100-2M) 24x10
6
72x10
6
IgG4 (gp100-2M) 24x10
6
39x10
6
PD-1 (HIV) 24x10
6
60x10
6
Percent CD4+ / CD8+ Percent CD4+ / CD8+
Day 0 Day 11
PD-1 (gp100-2M) 41.0% / 32.3% 41.2% / 36.0%
IgG4 (gp100-2M) 41.0% / 32.3% 42.0%
/ 35.0%
PD-1 (HIV) 41.0% / 32.3% 44.4%
/ 31.5%
Table 3-1. Effector cell count and percent CD4+ and CD8+ cells from PD-1-
blocked and IgG4 control-treated cultures. Total CD3+ effector cell numbers
were calculated by standard trypan blue exclusion. Shown is a representative
patient specimen showing total cell numbers and percent CD4+/CD8+ expression
on the initial day of culture (day 0) and 11 days thereafter. Percent CD4+ and
CD8+ cells were determined by flow cytometry as part of MHC:peptide tetramer
labeling assays. Percent CD4+ and CD8+ expression on day 11 were used to
determine the relative absolute numbers of CD4+, CD8+, and tetramer+ CTLs after
culture (Figure 3-6). Culture conditions are shown on the left, with stimulating
peptides shown in parentheses.
43
Figure 3-6. PD-1 blockade increases the absolute numbers of CD3+, CD8+,
CD4+, and melanoma tetramer+ CTLs. Total numbers of CD3+ effector cells
for each condition was acquired by standard trypan blue exclusion. Data shown are
for 7 patient specimens and were calculated by normalizing the percent CD8+,
CD4+, and tetramer+ staining to the total CD3+ effector cell count. Culture
conditions are shown on the abscissa for each plot, with stimulating peptides shown
in parentheses.
44
Figure 3-7. Dose-dependent effect of the 5C4 anti-PD-1 antibody on the
frequency of melanoma antigen-specific CTLs. The minimal effective dose of
the 5C4 PD-1-abrogating antibody in our in vitro peptide stimulation model was
between 10
-1
to 10
-2
µg/ml final concentration, with saturation reached at 10
µg/ml.
A total of three patient samples were completed with similar results.
45
Figure 3-8. Dose-dependent effect of the 5C4 anti-PD-1 antibody on absolute
numbers of melanoma antigen-specific CTLs. Absolute numbers of tetramer+
cells were calculated by normalizing the frequencies of tetramer+ cells by the total
CD3+ effector cell count. The minimal effective dose of the 5C4 PD-1-abrogating
antibody in our in vitro model was between 10
-1
to 10
-2
µg/ml final concentration,
with saturation reached at 10
µg/ml. A total of three patient samples were
completed with similar results.
46
Figure 3-9. A monovalent 5C4 anti-PD-1 F(ab) fragment demonstrates similar
effects compared to the corresponding full-length antibody. CD3+ T cells were
stimulated with autologous peptide-pulsed DCs for 11 days, and treated with
varying doses of a PD-1-abrogating F(ab) fragment.
47
Increases in the frequency of melanoma antigen-specific IFN- γ-secreting CTLs
were also seen in ELISPOT assays, confirming that PD-1 blockade augmented the
generation of cells that exert effector function (Figure 3-11). Augmented reactivity
to both the heteroclitic (gp100-2M and MART-27L) and native (“wild-type”)
gp100
209-217
and MART-1
26-35
(gp100-WT and MART-WT) peptides was observed.
Heteroclitic peptide analogs bind more stably to MHC than their native sequences
(Ruppert, 1993) and have been used as vaccines in cancer patients to induce T cells
capable of lysing tumor cells ex vivo (Valmori, 1998; Fong, 2001). Furthermore,
PD-1-blocked CTLs demonstrated increased specific lysis of peptide loaded T2
targets as well as gp100+/MART-1+ melanoma cells (Figures 3-12 and 3-13).
Taken together, these data show that PD-1 blockade during specific peptide
stimulation increased the total quantities of melanoma antigen-specific cytokine-
secreting cells and augmented the generation of tumor-cytolytic effectors.
48
Figure 3-10. PD-1 blockade increases the frequencies of IFN- γ-secreting
melanoma antigen-specific CTLs from post-vaccination specimens. CD3+ cells
were stimulated with autologous peptide-pulsed DCs for 11 days prior to assay.
Individual patient specimens are shown in rows (rightmost labels). Data shown are
for 2 patient specimens (7 total), representing gp100- and MART-1-specific
immune responses. Stimulation conditions are shown above each column, with
stimulating peptides shown in parentheses. Reactivity was assayed against
autologous peptide-pulsed CD3- PBMCs. Target peptides used were the
heteroclitic gp100 (gp100-2M) and MART-1 (MART-27L) peptides, their
corresponding native peptides (WT; “wild type”), and a control HIV RT peptide.
49
Figure 3-11. PD-1 blockade increases the frequencies of IFN- γ-secreting
melanoma antigen-specific CTLs from a pre-vaccination specimen. CD3+ cells
(Patient 10 listed in Table 2-1) were stimulated with autologous peptide-pulsed DCs
for 11 days prior to assay. Reactivity was assayed against autologous peptide-
pulsed CD3- PBMCs. Target peptides used were the MART-1 (MART-27L)
peptides, the corresponding native peptide (WT; “wild type”), and a control HIV
RT peptide.
50
Figure 3-12. PD-1 blockade augments the generation of effector cells that lyse
peptide-pulsed T2 targets. T2 is an HLA-A*0201+ TAP-deficient lymphoma cell
line, thus its surface MHC molecules are empty and can be efficiently loaded with
exogenous peptide (Wei, 1992). CD3+ cells were stimulated with autologous
peptide-pulsed DCs for 11 days prior to assay. Data shown are for 2 patient
specimens (7 total), representing gp100- and MART-1-specific immune responses.
Stimulation conditions are shown above each column, with stimulating peptides
shown in parentheses. Target peptides used were the heteroclitic gp100 (gp100-
2M) and MART-1 (MART-27L) peptides, their corresponding native peptides (WT;
“wild type”), and a control HIV RT peptide.
51
Figure 3-13. PD-1 blockade augments the generation of effector cells that lyse
melanoma antigen-expressing melanoma targets. Lysis of HLA-
A*0201+/gp100+/MART-1+ melanoma targets (526mel and 624mel) after two
consecutive cycles of stimulation with autologous peptide-pulsed DCs. Melanoma
line A-375 (HLA-A*0201+/gp100-/MART-1-) was used as a negative control. All
cell lines were negative for surface PD-L1 and PD-L2 expression (data not shown).
Stimulation conditions are shown above each column, with stimulating peptides
shown in parentheses. Data shown are for 2 patient specimens (3 total),
representing gp100- and MART-1-specific immune responses.
52
PD-1 blockade expands CTL clones that degranulate against melanoma
targets.
It is increasingly acknowledged that antigen-specific T cells can have diverse
requirements for cognate peptide (the epitope that is recognized by a specific T cell)
concentration on targets (Rubio 2003; Stuge, 2004; Korht, 2005; Zeh, 1999; Dutoit,
2001; Bullock, 2001; O’connor, 2002; Molldrem, 2003; Yang, 2002). T cell
recognition efficiency (RE, also known as “functional avidity”) refers to a specific
T cell’s sensitivity to different stimulatory peptide concentrations on antigen-
presenting cells or target cells (Rubio, 2003; O’connor, 2002). We use the term
recognition efficiency (as opposed to “functional avidity”) in our study to describe
the functional interaction between effector T cells and cognate peptide-bearing cells
(Rubio, 2003). In a tumor setting (illustrated in Figure 3-13), low RE CTLs can be
defined by their requirement for high peptide concentration on cells for activation,
and therefore are typically not activated by antigen-expressing tumor targets
(Rubio, 2003; Yang, 2002). The concentration of individual peptide epitopes can
vary on tumors depending on factors such as antigen expression level, overall
endogenous antigen processing capacity, and overall MHC expression. High RE
CTL clones have been suggested to require 10
-12
to 10
-14
molar cognate peptide
concentration to induce half-maximal target cell lysis in vitro, while low RE CTL
clones require 10
-8
to 10
-9
molar concentration to achieve the same (Sadovnikova,
1996). Low RE CTLs are expanded during in vitro peptide stimulation (Yang,
53
2002), and also appear to predominate in the peripheral blood of melanoma patients
vaccinated with peptides (Rubio 2003; Stuge, 2004; Korht, 2005). While the exact
mechanism is not clear, it has been suggested that high densities of relevant
peptide(s) on professional antigen-presenting cells, such as DCs, paradoxically
drive the preferential expansion of low RE CTLs that are not efficiently activated
by tumor cells in vitro (Stuge, 2004). If only high RE CTLs are capable of
recognizing tumor targets, then optimal in vivo anti-tumor activity induced by
vaccination likely requires enrichment for high RE T cells.
54
Figure 3-14. Peptide stimulation expands CTLs with diverse RE. In vitro
peptide stimulation and clinical peptide vaccination induces antigen-specific CTLs
with diverse RE for cognate peptide. The result is that only a fraction of antigen-
specific CTLs have sufficient RE to recognize tumor targets (Yang, 2002; Rubio,
2003; Stuge, 2004).
55
Figure 3-15. Identification of high RE tumor-reactive CTLs by MHC:peptide
tetramer and CD107a labeling. CTLs that have sufficient RE to become activated
by endogenous levels of antigen presented by tumor targets degranulate (right).
This method allows for determination of the fraction of antigen-specific with high
RE for cognate peptide.
56
We used MHC:peptide tetramers to directly enumerate functional, high RE CTLs
on the basis of CD107a (also called LAMP-1; lysosomal-associated membrane
protein-1) externalization during incubation with antigen-expressing melanoma
targets in vitro (illustrated in Figure 3-14). During the process of target killing,
specialized secretory lysosomes (also called lytic granules) fuse with the T cell
membrane and release cytotoxic mediators including perforin and granzymes
(Trapani, 2002). CD107a is a secretory lysosome membrane protein that is
transiently externalized on the cell membrane during this degranulation process.
CD107a mobilization against antigen-expressing melanoma targets by cloned
gp100-specific CTL lines is closely associated with high RE for cognate peptide
(Rubio, 2003; Stuge, 2004). Thus, the externalization of CD107a can be a surrogate
marker to identify individual antigen-specific CTLs with high RE (among a
population with diverse RE) for cognate peptide presented endogenously by tumor
cells.
The data in Figure 3-15 show that, after one cycle of peptide stimulation, PD-1
blockade enriched for gp100- and MART-1-specific CTLs (tetramer+) that
degranulated after exposure to HLA-matched, gp100+/MART-1+ melanoma cell
lines (526mel and 624mel). Trace CD107a staining of tetramer+ CTLs was seen
against the gp100-/MART-1- melanoma cell line A-375 (<0.3% staining), likely
due to necrosis of cells during culture and resulting in non-specific uptake of anti-
57
CD107a antibody. Not only was there an increase in the frequency of
tetramer+/CD107a+ cells in PD-1-blocked cultures, but also an increase in the
proportion of tetramer+ cells that externalized CD107a+. Hence, these data suggest
that PD-1 blockade altered the functional repertoire of melanoma peptide-
stimulated CTL populations, enriching for clones that are activated by endogenous
levels of antigen presented by melanoma cells (i.e. high RE CTLs).
58
Figure 3-16. PD-1 blockade increases the fraction of post-vaccination
melanoma antigen-specific CTLs that degranulate against melanoma targets.
Degranulation (CD107a mobilization) by melanoma antigen-specific CTLs
stimulated with autologous peptide-pulsed DCs for 11 days prior to assay. Data
shown are for 2 patient specimens (6 total), representing gp100- and MART-1-
specific immune responses. Melanoma targets are shown above each column. The
fraction of tetramer+ cells that externalized CD107a are shown in parentheses in the
upper right quadrant of each dot plot. Tetramer+/CD107a+ CTLs were acquired by
gating on the CD8+ population.
59
Figure 3-17. PD-1 blockade increases the fraction of pre-vaccination
melanoma antigen-specific CTLs that degranulate against melanoma targets.
Degranulation (CD107a mobilization) by MART-1-specific CTLs stimulated with
autologous peptide-pulsed DCs for 11 days prior to assay. Data shown represent
CTLs from a pre-vaccination specimen (Patient 10 listed on Table 2-1). Melanoma
targets are shown above each column. The fraction of tetramer+ cells that
externalized CD107a are shown in parentheses in the upper right quadrant of each
dot plot. Tetramer+/CD107a+ CTLs were acquired by gating on the CD8+
population.
60
Discussion:
For the first time, we have shown that direct antibody-mediated blockade of the PD-
1 receptor during in vitro antigen stimulation augments the generation of functional
gp100- and MART-1-specific human CTLs that secreted IFN- γ in ELISPOT assays
and lysed antigen-expressing targets. Increases in the absolute numbers of CD8+,
CD4+, and tetramer+ CTLs were also observed. The augmenting effect of the 5C4
PD-1-abrogating antibody was seen using T cells from post-vaccination specimens,
as well as a single pre-vaccination specimen. It was interesting to note that, for the
pre-vaccination specimen, increases in the frequencies of IFN- γ-secreting
melanoma antigen-specific CTLs was seen only against the heteroclitic peptide
analogue (MART-27L), and not the corresponding native sequence (MART-WT)
(Figure 3-11). For post-vaccination specimens, PD-1 blockade increased the
frequencies of IFN- γ-secreting melanoma antigen-specific CTLs that recognized
both the heteroclitic peptide analogues and their corresponding native sequences
(Figure 3-10). Also, for the pre-vaccination specimen, background degranulation of
MART-1-specific CTLs against the MART-1- control target A-375 was
significantly higher (Figure 3-17) than seen using post-vaccination specimens
(Figure 3-16). Further experiments using pre-vaccination specimens are currently
being repeated to confirm these results. Nonetheless, as measured by MHC:peptide
tetramer labeling, PD-1 blockade clearly increased the frequency of MART-1-
specific CTLs from a patient specimen before vaccine manipulations (Figure 3-4).
61
These results suggest that PD-1 blockade augments both pre-existing endogenous as
well as vaccine-induced melanoma-antigen specific CTLs during in vitro peptide
stimulation.
While increases in absolute numbers of tetramer- CTLs and CD4+ T cells were
observed in PD-1-blocked cultures, we do not know the antigen specificities of
these cells – possibly endogenous antigens presented by the DCs that were used as
antigen-presenting cells, or possibly culture media proteins captured and processed
by the DCs. This increase in absolute numbers of CD4+ T cells leaves open the
possibility that the increases in frequencies and absolute numbers of melanoma
antigen-specific CTLs observed by PD-1 blockade is due, in part, to the augmented
provision of CD4+ T help. CD4+ T helper cells – in particular the T-helper-1 (Th1)
type – modulate the effectiveness and long-term survival of memory CTLs. This
may occur by (1) helping to initiate antigen-specific CD8
+
T cells by activating DCs
via CD40-CD40L interaction, (2) secreting cytokines such as IL-2, and (3)
providing direct cell-to-cell costimulation via CD27, CD134, and MHC class II
(Giuntoli, 2002; Hung, 1998; Schoenberger, 1998). Future studies to test whether
the CD4+ T helper component of bulk CD3+ T cells is necessary for PD-1
blockade-mediated augmentation of melanoma antigen-specific CTLs would
involve depletion of CD4+ cells from our in vitro model.
62
As determined by MHC:peptide tetramer labeling and cell counting, the minimal
effective dose of the 5C4 PD-1-abrogating antibody was between 10
-1
to 10
-2
µg/ml
final concentration, with saturation reached at 10
µg/ml (Figures 3-7 and 3-8).
These data suggest that the optimal dose of 5C4 antibody needed to augment the
generation CTLs by peptide vaccination would likely be in the range of 1-10 mg/kg
of body weight. These are expected to be within clinically relevant doses, as a
CTLA-4-abrogating antibody (antibody MDX-010 by Medarex, Inc., Milpitas, CA)
has been used in patients at doses of up to 3 mg/kg of body weight (Phan, 2003;
Attia, 2005; Sanderson, 2005). MDX-010 doses of up to 10 mg/kg of body weight
are currently being used in ongoing clinical cancer vaccine trials (personal
correspondence with Dr. Jeffrey Weber, USC/Norris Comprehensive Cancer
Center, Los Angeles, CA, USA).
A monovalent 5C4 PD-1-abrogating F(ab) fragment was also tested in our model,
and demonstrated similar affects on the frequencies of melanoma antigen-specific
CTLs, compared to the corresponding full-length antibody (Figure 3-9). The
monovalent F(ab) fragment is the single antigen-binding region of the full-length
PD-1-abrogating antibody (illustrated in Figure 3-18), and thus its only reactivity is
to the PD-1 molecule. Antibody Fc regions represent a potential stimulus for DC
activation through stimulatory FC γ receptors. It has been shown that activation of
stimulatory Fc γ receptors expressed on DCs can result in DC maturation, thus
63
enhancing their T cell activating capacity (Regnault, 1999). The data in Figure 3-9
show that the effect of the full length 5C4 anti-PD-1 antibody resulted from
blockade of the PD-1 receptor, and not via possible activation of stimulatory Fc γ
receptors expressed on DCs used in our model. A final concentration of 100 µg/ml
of F(ab) fragment was required to achieve the same effects as 10 µg/ml of full-
length antibody (Figure 3-9). The higher efficiency of PD-1 blocking by the full-
length anti-PD-1 antibody compared to the corresponding monovalent F(ab)
fragment is likely due to cross-linking of PD-1 molecules via divalent interactions.
A PD-1-abrogating F(ab’)2 fragment would be expected to have the same blocking
efficiency as the corresponding full-length antibody.
64
Figure 3-18. Antibody structure. The F(ab) fragments are antigen-binding
regions, and are used to determine if a corresponding full-length antibody has
stimulatory effects through Fc receptors expressed on cells of interest. The Fc
region of antibodies can possibly mature DCs via stimulatory Fc γ receptors
(Regnault, 1999).
65
PD-1 blockade also increased the fraction of antigen-specific CTLs (tetramer+) that
recognized antigen-expressing melanoma targets, as measured by CD107a
externalization. These data suggest an increase in RE for cognate peptide. It has
been suggested that CD107a externalization is a marker for high RE human T cells
with functional capacity against tumor targets (Rubio 2003; Stuge, 2004). T cell
RE – a T cell’s sensitivity to peptide concentration – may be an important
determinant in anti-tumor immunity (Rubio 2003; Stuge, 2004; Korht, 2005; Zeh,
1999; Dutoit, 2001; Bullock, 2001; O’connor, 2002; Molldrem, 2003; Yang, 2002).
As such, vaccine approaches to cancer immunotherapy should focus on not only
improving the quantitative expansion of tumor antigen-specific T cells, but also
augmenting T cell RE. While the exact mechanism is not clear, high antigen
concentrations on antigen-presenting cells, such as in the case of in vitro peptide
stimulations or clinical peptide vaccines, may predominantly expand T cells with
low RE for cognate peptide (Stuge, 2004). These low RE T cells may be ineffective
at tumor destruction in vivo and therefore may not confer clinical benefit. The data
presented in our study show that antibody-mediated PD-1 blockade during in vitro
peptide stimulation enhances the generation of high RE gp100- and MART-1-
specific CTLs. These high RE CTLs were directly enumerated on the basis of
CD107a mobilization to the cell membrane after exposure to gp100+/MART-1+
melanoma targets (Figure 3-16 and 3-17). Further evidence supporting the
generation of high RE tumor antigen-specific effectors was shown by increased
66
lysis of gp100+/MART-1+ melanoma targets by PD-1-blocked cells (Figure 3-13).
These data show that PD-1 blockade influences not only the quantity of melanoma-
antigen specific CTLs, but also their quality – i.e. their efficiency of tumor
recognition. Hence, augmentation in the fraction of tumor-reactive clones among
an antigen-specific CTL population is expected to provide enhanced anti-tumor
effects in vivo.
67
Chapter 4
PD-1 blockade augments CTL proliferation but does not
affect CTL apoptosis
Introduction:
In Chapter 3, tetramer labeling and ELISPOT assays showed that PD-1 blockade
during in vitro peptide stimulation enhanced the generation of human gp100- and
MART-1-specific CTLs. To elucidate if this was the result of augmented
proliferation and/or increased survival (i.e. decreased cell death), we analyzed: (1)
the fraction of dividing melanoma antigen-specific CTLs and (2) the fraction of
melanoma antigen-specific CTLs undergoing apoptosis during in vitro peptide
stimulation. We hypothesized that PD-1 blockade increases the fraction of dividing
antigen-specific T cells. However, there is no clear existing data suggesting that the
PD-1 pathway directly modulates apoptosis (Greenwald, 2005). Accordingly, we
sought to find out whether PD-1 blockade would reduce apoptosis of melanoma
antigen-specific CTLs in our in vitro peptide stimulation model.
We directly enumerated proliferating antigen-specific CTLs based on expression of
the proliferation-associated nuclear protein Ki67. Ki67 – a protein with a still
undefined cellular function – is only expressed during late G1, S, G2 and M phases
of the cell cycle (Endl, 2000). Thus, it can be used as a surrogate marker for
68
proliferating lymphocytes. When combined with MHC:peptide tetramer labeling,
we were able to assess proliferating melanoma antigen-specific CTLs at the level of
single cells. Dilution of Carboxyfluorescein diacetate succinimidyl ester (CFSE) in
PD-1-blocked melanoma antigen-specific CTLs was also assessed to: (1) determine
the fraction of dividing cells and (2) track the number of cell divisions. CFSE is a
membrane-permeable fluorescein-based vital dye that forms stable covalent
linkages with amines at neutral pH, and can be used for in vitro and in vivo tracking
of lymphocytes (Parish, 1999). As CFSE-labeled cells divide, CFSE is diluted
among the daughter cells. Distinct rounds of cell divisions can then be visualized
by flow cytometry, with typically up to 7-8 rounds of observable cell division
(Parish, 1999). Thus, analysis of CFSE dilution at a single chosen time point
represents a summary of cell divisions that have occurred within a population of
interest.
Also using MHC:peptide tetramers, we were able to assess viability of melanoma
antigen-specific T cells by Annexin V/7-aminoactinomycin D (7-AAD) labeling
and intracellular active Caspase-3 expression. In normal blood cells,
phosphatidylserine molecules are found on the inner leaflet of the cell membrane
(Op den Kamp, 1979). Entry of cells into apoptosis leads to membrane lipid
asymmetry, whereby phosphatidylserine molecules are externalized to the outer cell
membrane leaflet (Fadok, 1992). Annexin V is an anti-coagulant that binds to
69
phosphatidylserine molecules, and fluorescently-conjugated Annexin V molecules
are commonly used to detect viable cells entering the early phases of apoptosis
(Andree, 1990). 7-AAD is a non-vital DNA dye used to detect cells with
compromised cell membranes, such as necrotic cells or cells in the late stages of
apoptosis (Schmid, 1992). Caspase-3 is considered an effector caspase that cleaves
a variety of cellular substrates including the cell survival factor BCL-2. Expression
of the active Caspase-3 enzyme is considered a marker for apoptosis activation
(Krajewska, 1997). Through the collective flow cytometry-based detections of the
aforementioned molecules, we were able to determine whether PD-1 blockade in
our in vitro peptide stimulation model augments the frequencies and absolute
numbers of melanoma antigen-specific CTLs via augmented proliferation and/or
decreased cell death.
70
Results:
PD-1 blockade augments proliferation of T cells.
We sought to determine if PD-1 blockade induced a larger proportion of tetramer+
CTLs to proliferate. PD-1-blocked lymphocytes exhibit marked increases in
volume and granularity on the day of peak proliferation (day 6 as measured by Ki67
expression), as measured by increased forward and side light scatter (Figure 4-1).
This is indicative of actively proliferating lymphocytes. The increase in volume and
granularity coincides with increased Ki67 expression by melanoma tetramer+ and
tetramer- CTLs on the same day (Figure 4-2). Both PD-1-blocked and IgG4
control-treated tetramer+ CTLs proliferated in response to melanoma peptide
stimulation, as measured by Ki67 expression during culture (Figure 4-2). Notably,
a larger percentage of PD-1-blocked tetramer+ CTLs expressed Ki67 (Figure 4-2).
This induction of proliferation was not sustained, however, as Ki67 expression in
both PD-1-blocked and IgG4 control-treated cells returned to background levels by
day 11 of stimulation. These data show that PD-1 blockade induced a higher
proportion of antigen-specific CTLs to exhibit a burst of proliferation in response to
peptide stimulation. Analysis of CFSE dilution in antigen-specific CTLs confirmed
these findings, as a higher fraction of PD-1-blocked tetramer+ CTLs demonstrated
increased dilution of CFSE (Figure 4-3). Of note, it appeared that both PD-1-
blocked and IgG4 control-treated melanoma antigen-specific CTLs underwent the
same numbers of cell divisions (6-7 divisions). Proliferation of tetramer- CTLs and
71
CD4+ T cells was also augmented by PD-1 blockade, as measured by both Ki67
expression and CFSE dilution (Figures 4-2, 4-3, and 4-5), demonstrating that PD-1
blockade affects both melanoma antigen-specific CTLs as well as the total CD8+
and CD4+ T cell populations.
72
Figure 4-1. PD-1 blockade affects lymphocyte volume and granularity during
in vitro culture. Stimulation conditions are shown in rows (rightmost labels). Data
shown represent the day of peak proliferation (day 6 as determined by Ki67
expression in Figure 4-2) and the day of harvest for antigen-specific functional
analysis (day 11). Forward scatter: cell volume. Side scatter: granularity.
Day 0 Day 6 Day 11
αPD-1 (MART-27L) 1.0 x 10
6
2.3x10
6
3.0x10
6
IgG4 (MART-27L) 1.0 x 10
6
1.3x10
6
1.0x10
6
Table 4-1. Effector cell counts from PD-1-blocked and IgG4 control-treated
cultures. Stimulation conditions are shown on the left. Total effector cell counts
were determined by standard trypan blue exclusion. Data listed are representative
of 4 patient specimens.
73
Figure 4-2. PD-1 blockade increases Ki67 expression in peptide-stimulated
melanoma antigen-specific CTLs. Peak Ki67 expression was seen on day 6 of
stimulation. Stimulation conditions are shown in rows (rightmost labels).
Rightmost dot plots show Ki67 expression in tetramer+ CTLs that were acquired by
gating on the CD8+ population. Corresponding isotype controls are shown in the
leftmost dot plots. The fraction of tetramer+ cells that expressed Ki67 are shown in
parentheses (upper right quadrants). Shown is a patient specimen (4 total) with a
MART-1-specific response.
74
Figure 4-3. PD-1 blockade increases Ki67 expression in CD4+ T cells. Peak
Ki67 expression was seen on day 6 of stimulation. Stimulation conditions are
shown in rows (rightmost labels). Rightmost dot plots show Ki67 expression in
CD4+ T cells. Corresponding isotype controls are shown in the leftmost dot plots.
Four patient specimens were analyzed with similar results.
75
Figure 4-4. PD-1 blockade augments CFSE dilution in peptide-stimulated
melanoma antigen-specific CTLs. CD3+ cells were stimulated with autologous
peptide-pulsed DCs for 11 days prior to assay. Treatment conditions are shown
above each dot plot. For each dot plot, the fractions of non-dividing melanoma
antigen-specific CTLs are shown in the rightmost boxes, dividing cells in the center
boxes, and rapidly-dividing cells in the leftmost boxes. Shown is a patient
specimen (4 total) with a MART-1-specific response.
76
Figure 4-5. PD-1 blockade augments CFSE dilution in CD4+ T cells. CD3+
cells were stimulated with autologous peptide-pulsed DCs for 11 days prior to
assay. Treatment conditions are shown above each dot plot. For each dot plot, the
fractions of non-dividing MHC:peptide tetramer+ cells are shown in the rightmost
boxes, dividing cells in the center boxes, and rapidly-dividing cells in the leftmost
boxes.
77
PD-1 blockade does not affect apoptosis of antigen-specific CTLs.
Despite the name, no clear evidence exists that the PD-1 pathway directly
modulates programmed cell death (Greenwald, 2005). In the previous section, we
demonstrated that a higher fraction of peptide-stimulated melanoma-antigen CTLs
underwent proliferation when treated with a PD-1-abrogating antibody. This, in
part, resulted in increased frequencies and absolute numbers of gp100- and MART-
1-specific CTLs after in vitro peptide stimulation. We also examined if PD-1
blockade also reduced the fraction of melanoma antigen-specific CTLs that
underwent apoptosis during proliferation, which could also account for the
augmenting effects of PD-1 blockade on frequencies and absolute numbers of
melanoma antigen-specific CTLs. Apoptosis was measured by Annexin V/7-AAD
labeling and intracellular active Caspase-3 expression. As shown in Figures 4-6
and 4-8, PD-1 blockade did not significantly affect the fraction of antigen-specific
CTLs that stained with Annexin V/7-AAD or expressed active Caspase-3. PD-1
blockade also did not significantly affect the fraction of CD4+ T cells that stained
with Annexin V/7-AAD or expressed active Caspase-3 (Figures 4-7 and 4-9).
78
Figure 4-6. PD-1 blockade does not significantly affect Annexin V/7-AAD
labeling of peptide-stimulated melanoma antigen-specific CTLs. CD3+ cells
were stimulated with autologous peptide-pulsed DCs. Effector cells were stained
with Annexin V and 7-AAD to detect morphological changes in the cell membrane
that are associated with apoptosis. Shown is an example of a patient with a MART-
1-specific response. A total of 4 patients were completed with similar results.
79
Figure 4-7. PD-1 blockade does not significantly affect Annexin V/7-AAD
labeling of CD4+ T cells. CD3+ cells were stimulated with autologous peptide-
pulsed DCs. Effector cells were stained with Annexin V and 7-AAD to detect
morphological changes in the cell membrane that are associated with apoptosis. A
total of 4 patients were completed with similar results.
80
Figure 4-8. PD-1 blockade does not significantly affect active Caspase-3
expression in peptide-stimulated melanoma antigen-specific CTLs. CD3+ cells
were stimulated with autologous peptide-pulsed DCs. Intracellular expression of
active Caspase-3 was measured on the day of peak proliferation (Day 6 as measured
by Ki67 expression) and 11 days after peptide stimulation. Shown is an example of
a patient with a MART-1-specific response. A total of 4 patients were completed
with similar results. Right panels: active caspase-3 expression. Left panels:
corresponding isotype controls.
81
Figure 4-9. PD-1 blockade does not significantly affect active Caspase-3
expression in CD4+ T cells. CD3+ cells were stimulated with autologous peptide-
pulsed DCs. Intracellular expression of active Caspase-3 was measured on the day
of peak proliferation (day 6 as measured by Ki67 expression) and 11 days after
peptide stimulation. A total of 4 patients were completed with similar results.
Right panels: active caspase-3 expression. Left panels: corresponding isotype
controls.
82
Discussion:
In our in vitro peptide stimulation model described in Chapter 3, antibody-mediated
PD-1 blockade increased the frequencies and absolute numbers of functional
gp100- and MART-1-specific CTLs. In Figure 4-1, PD-1-blocked lymphocytes
demonstrated a clear increase of large, granular cells on day 6 of culture, as
measured by increased forward and side light scatter. This is generally interpreted
as an increased fraction of proliferating lymphocytes, and correlates with increased
Ki67 expression, a nuclear proliferation-associated protein (Figures 4-2 and 4-3).
Measurement of intracellular expression of Ki67 showed that a higher fraction of
PD-1-blocked melanoma antigen-specific CTLs proliferated in response to peptide
stimulation. These observations that PD-1 blockade increased the fraction of
dividing antigen-specific CTLs is consistent with the previous finding that the PD-1
pathway inhibits cellular processes that are vital to T cell proliferation. It has been
demonstrated that PD-1 signaling inhibits glucose metabolism and PI3K/Akt
signaling in anti-CD3/CD28-stimulated CD4+ T cells (Parry, 2005). Akt is a
serine/threonine kinase that tranduces anti-apoptotic and/or proliferative signals in
T lymphocytes (Ahmed, 1997; Brennan 1997). Increases in the fraction of Ki67-
expressing tetramer- CTLs and CD4+ T cells from PD-1-blocked cultures are
consistent with our observations of increased absolute numbers of total CD8+ and
CD4+ cells, shown in Chapter 3.
83
CFSE dilution of PD-1-blocked, melanoma antigen-specific CTLs confirmed that a
higher fraction of CTLs underwent proliferation. CFSE dilution in PD-1-blocked
tetramer- CTLs and CD4+ T cells was also observed (Figures 4-4 and 4-5).
However, the total numbers of cell divisions did not appear to be altered for any cell
population, compared to IgG4 control-treated cells. These observations suggest that
although PD-1 induces a larger fraction of T cell clones to proliferate, it does not
alter the natural cycle of T cell proliferation. This is critical, as sustained
proliferation of T cells measured by Ki67 expression is associated with
lymphoproliferative disorders including large granular lymphocyte leukemia, T-cell
prolymphocytic leukemia, and non-Hodgkin’s lymphoma (de melo, 1992).
We found no significant effect of PD-1 blockade on apoptosis of antigen-specific
CTLs in our model. Annexin V/7-AAD staining and intracellular caspase-3
expression profiles were similar in both PD-1-blocked and IgG4 control-treated
cultures. Total effector cell counts for IgG4 control-treated cultures decreased
between days 6 and 11 of culture, suggesting a net loss of cells during that time
period (Table 4-2). This is likely explained by the cessation of proliferation, as
confirmed by lack of Ki67 expression by day 11 (Figure 5-2), while cell death was
still occurring, as measured Annexin V/7-AAD labeling (Figures 4-6 and 4-7).
Nonetheless, these data suggest that the predominant effect of PD-1 blockade in our
model is augmentation of T cell proliferation, not decreased cell death.
84
These findings are consistent with the hypothesis that an unknown receptor, distinct
from PD-1, may interact with PD-L1 and/or PD-L2 to transduce an apoptotic signal
to T cells (Dong, 2002). Taken together, this suggests that simultaneous antibody-
mediated blockade of PD-1 and PD-L1/PD-L2 may have additive and/or synergistic
effects on the frequencies and absolute numbers of antigen-stimulated T cells. If
PD-L1 and/or PD-L2 transduce apoptotic signals via a receptor distinct from PD-1,
then simultaneous blockade of PD-1, PD-L1, and PD-L2 would enhance antigen-
specific T cell generation by augmenting proliferation while concomitantly
inhibiting apoptosis via separate pathways. In other words, antibody-mediated
blockade of PD-1, PD-L1, and PD-L2 may not redundant. Genome-wide searches
using computational methods to predict protein-protein interactions is likely a
rational starting point for identifying additional PD-ligand-binding receptors
(Salwinski, 2003).
85
Chapter 5
PD-1 blockade augments T cell cytokine secretion and does not alter surface
marker expression on melanoma antigen-specific CTLs
Introduction:
Cytokine polarization appears to have significant implications for a variety of
diseases from autoimmune pathologies, allergies, and cancer. The nature of cellular
immune responses is determined in a large part by exposure to cytokines – in
particular those secreted by CD4+ T cells. Type-2 cytokines such as IL-4 and IL-5
secreted by CD4+ Th2 cells are required for optimal differentiation and
proliferation of B cells (Mosmann, 1989). Other Type-2 cytokines, in particular IL-
10 and TGF- β, have been demonstrated to inhibit antigen-specific CTL responses
against tumors (Taga, 1992; Li, 2006). Type-1 cytokines on the other hand, such as
IL-2 and TNF- α, are important for the proliferation of CTLs (Hung, 1998;
Kasahara, 2003). At the clinical level, it has been previously reported that immune
responses in cancer vaccine patients skewed towards CD4+ Th2-type responses are
correlated with disease progression in melanoma patients, while normal donors and
patients who remained disease-free exhibited highly-polarized CD4+ Th1 or mixed
Th1/Th2 responses (Tatsumi, 2002). Th1-type CD4+ T cells secreting IFN- γ
appear important to the optimal generation and durability of tumor antigen-specific
CTLs in vivo and may also recruit these effector cells into the tumor
86
microenvironment (Hung, 1998). Hence, cytokine milieu has a significant
influence on the expansion antigen-specific CTLs, and that vaccine approaches to
cancer immunotherapy should, in part, aim to optimize the in vivo cytokine
environment.
With the use of MHC:peptide tetramers, phenotypic characterization of antigen-
specific T cells can be assessed at the single-cell level. In Chapter 3, we
demonstrated that PD-1 blockade increased the absolute number and frequency of
functional gp100- and MART-1-specific CTLs. This was demonstrated by different
assays, including direct enumeration of antigen-specific CTLs by MHC:peptide
tetramers and IFN- γ ELISPOT. We also showed directly that the fraction of
antigen-specific T cells that degranulated against relevant melanoma targets was
increased by treatment with a PD-1-abrogating antibody. It is currently unknown if
PD-1 blockade affects the expression of surface markers on antigen-stimulated
CTLs, including PD-1. If PD-1 expression is still maintained on antigen-specific
CTLs, this would suggest that continuous treatment with the 5C4 PD-1-abrogating
antibody would be necessary to sustain their proliferation over multiple rounds of
antigen-driven expansion. A panel of surface markers was analyzed to assess if
PD-1 blockade altered the overall phenotype of antigen-specific CTLs in our in
vitro peptide stimulation model.
87
Results:
Cytokine repertoire and secretion kinetics of PD-1-blocked human CD3+ T
Cells.
Blockade of the PD-1 ligand, PD-L1, has been shown to augment IFN- γ and IL-2
secretion by murine CD8+ T cells stimulated in vitro with peptide-pulsed DCs
(Curiel, 2003). Furthermore, PD-L1 blockade has been shown to augment IFN- γ,
IL-13, and IL-10 secretion by bulk human peripheral blood mononuclear cells
(PBMCs) stimulated with protein (Cai, 2004). To our knowledge, extensive
analysis of the cytokine repertoire and secretion kinetics of PD-1-blocked human T
cells during in vitro peptide stimulation has not been previously reported. We
analyzed T cell culture supernatants for Type-1 and Type-2 cytokine content on
days 2, 4, 6, 8, and 11 of melanoma peptide stimulation, using a luminescence-
based cytokine array method. This method allows for high-throughput analysis of
multiple cytokines in solution from a single experimental specimen (dupont, 2005).
In PD-1-blocked cultures, we observed increased accumulation of IL-5, IL-13, GM-
CSF, and IFN- γ on day 11 of stimulation (Figure 5-1). Increases in IL-2, IL-10, and
TNF- α were seen on earlier days of stimulation, with each diminishing by day 11.
No significant amount of IL-4 or IL-12 was detected at any time point. These data
show that PD-1 blockade during in vitro peptide stimulation increases total cytokine
secretion by human CD3+ T cells without selectively enhancing either Type 1 (IL-
2, IL-12, GM-CSF, IFN- γ, TNF- α) or Type 2 (IL-4, IL-5, IL-10, IL-13) cytokines.
88
The increase in Tc1/Th1 cytokine production, in particular the lymphoproliferative
cytokine IL-2, may be a contributing factor towards the enhanced generation of
melanoma antigen-specific CTLs by PD-1 blockade in our mode, (Hung, 1998)
89
Figure 5-1. Cytokine repertoire and secretion kinetics of PD-1-blocked and
IgG4 control-treated CD3+ T cells. Both Type-1 (IL-2, IL-12, GM-CSF, IFN- γ,
TNF- α) and Type-2 (IL-4, IL-5, IL-10, IL-13) cytokine secretion is increased,
without skewing of the cytokine repertoire. Three patient specimens were assayed
with similar results.
90
We also found that PD-1 blockade augmented the generation of MART-1-specific
CD4+ Th1 and Th2 cells during in vitro peptide stimulation, but did not alter the
Th1/Th2 ratio (Figure 5-2). We found that IL-12 was able to selectively enhance
the generation of MART-1-specific Th1 cells during in vitro peptide stimulation
(while suppressing Th2 cell generation), demonstrating that post-vaccine peripheral
blood CD4+ T cells can still be polarized by exogenous cytokines (Figure 5-3).
This is a significant finding, based on our previous analysis of vaccine-induced
MART-1-specific CD4+ T cells in the peripheral blood of serologic HLA-DR4+
(HLA-DR7 and DR13 are members of the HLA-DR4 serologic family) melanoma
patients (Wong, 2004). A mixture of peripheral blood Th1/Th2 cells was induced
in patients vaccinated against the HLA-DR4-binding MART-1
51-73
T helper epitope
without cytokine adjuvants (Figures 5-4 and 5-5). This finding provides a rationale
for the addition of IL-12 as a cytokine adjuvant with MHC class II-restricted
peptide vaccines.
91
Figure 5-2. PD-1 blockade does not alter melanoma antigen-specific CD4+
Th1/Th2 balance. CD3+ cells were stimulated with autologous peptide-pulsed
DCs for 11 days prior to assay for IFN- γ and IL-5 secretion by ELISPOT.
Stimulation conditions are shown on the abscissa. Reactivity was assayed against
autologous peptide-pulsed CD3- PBMCs. Target peptides used were the HLA-
DR β1*0401-restricted MART-1
51-73
peptide, and HIV RT
171-190
control peptide.
Shown is a representative example of two patient specimens.
92
Figure 5-3. IL-12 induces selective expansion of melanoma antigen-specific
CD4+ Th1 cells. Fresh PBMCs from patients vaccinated with MART-1
51-73
were
incubated with the MART-1
51-73
peptide (10 µg/ml final) and IL-12 (10 ng/ml
final). Th1/Th2 reactivity was assayed against autologous peptide-pulsed PBMCs.
Target peptides used were the HLA-DR β1*0401-restricted MART-1
51-73
peptide,
and HIV RT
171-190
control peptide. Shown is a representative example of two
patient specimens.
93
Figure 5-4. Mixed Th1/Th2 polarization of vaccine-induced MART-1-specific
CD4+ T cells in the peripheral blood of vaccinated melanoma patients.
Measurement of CD4
+
T cell reactivity against MART-1
51-73
in uncultured PBMCs
from vaccinated melanoma patients. White bars: pre-vaccine. Black bars: post-
vaccine.
94
Figure 5-5. Mixed Th1/Th2 polarization of vaccine-induced MART-1-specific
CD4+ T cells after in vitro peptide stimulation. Measurement of CD4
+
T cell
reactivity against MART-1
51-73
after in vitro peptide stimulation. Left panel: IFN- γ
ELISPOT. Right panel: IL-5 ELISPOT. White bars: pre-vaccine. Black bars: post-
vaccine. Cross batch bars: 12+ months post-vaccine.
95
Surface marker expression on PD-1-blocked tetramer+ CTLs.
Surface marker analysis of PD-1-blocked tetramer+ cells was performed by flow
cytometry. Melanoma antigen-specific CTLs from vaccinated patients were
stimulated with autologous peptide-pulsed DCs for 11 days, as described in
Methods, and harvested for analysis. Phenotyping of both PD-1-blocked and IgG4
control-treated cells were carried out by combining MHC:peptide tetramer and
surface marker labeling. The summarized data in Figure 5-4 shows that peptide-
stimulated melanoma antigen-specific CTLs from both PD-1-blocked and IgG4
control-treated cultures exhibited a predominant activated effector / effector
memory phenotype, characterized by a CD45RA(low), CD45RO(high),
CCR7(low), CD62L(low), and CD44(high) expression profile. However, for the
markers we analyzed on day 11 of peptide stimulation, PD-1 blockade did not
appear to significantly alter their percent expression or mean channel fluorescence
(Figures 5-6 and 5-7). Of note, the expression of the adhesion molecules LFA-1,
LFA-2, CD43, and VLA-4 was not significantly altered in PD-1-blocked tetramer+
CTLs, suggesting that augmented recognition of melanoma targets by degranulation
(Figure 3-15) was likely not mediated by increased adhesion molecule expression.
96
Figure 5-6. Percent expression of surface markers on melanoma antigen-
specific CTLs. Data were acquired by gating on the CD8+/tetramer+ populations.
Black bars: PD-1-blocked cells. White bars: IgG4 control-treated cells. Data
shown are averaged from 4 patient samples.
97
Figure 5-7. Mean channel fluorescence of surface markers on antigen-specific
CTLs. Histograms were acquired by gating on the CD8+/tetramer+ populations.
PD-1-blocked cells are shown in solid grey. IgG4 control-treated cells are overlaid
in black. Data shown are representative of 4 patient samples.
98
Discussion:
The PD-1 pathway controls cytokine production in both CD8+ and CD4+ T cells
(Greenwald, 2005; Chen, 2004). To our knowledge, an extensive analysis of the
Type-1 vs. Type-2 cytokine repertoire and secretion kinetics of human T cells after
antibody-mediated PD-1 blockade has not been reported. In our in vitro peptide
stimulation model, we found that PD-1 blockade increased both Type-1 and Type-2
cytokine accumulation. However, neither the repertoire nor kinetics of cytokine
secretion was significantly altered. We observed no significant accumulation of IL-
12 or IL-4 in our model, which may explain the lack of effect of PD-1 blockade on
the overall Type-1 vs. Type-2 cytokine balance. IL-12 and IL-4 are Type-1- and
Type-2-polarizing cytokines, respectively (Paul, 1994). We also saw that PD-1
blockade enhanced the generation of MART-1-specific CD4+ Th1 and Th2 cells by
ELISPOT assay, but did not alter the Th1/Th2 ratio.
These data have translational significance towards anticipating the likely in vivo
effects of PD-1 blockade. The Type-1 vs. Type-2 cytokine balance has significance
in the expansion and function of antigen-specific CTLs. Type-2 cytokines, in
particular IL-10 and TGF- β, have been demonstrated to inhibit antigen-specific
CTL responses against tumors (Taga, 1992; Li, 2006). Furthermore, polarization
towards tumor antigen-specific CD4+ Th2 cells in the peripheral blood and tumors
of melanoma and renal cell carcinoma patients is associated with poor clinical
99
prognosis while polarization towards CD4+ Th1 cells is associated with positive
clinical prognosis (Tatsumi, 2002). Type-1 cytokines, in particular IL-2 and TNF-
α, support CTL proliferation and immune responses against viruses and tumors
(Hung, 1998; Kasahara, 2003). Our in vitro model suggests that the in vivo effect
of PD-1 blockade would be an augmentation in both Type-1 and Type-2 cytokine
secretion from the total T cell population. Because Type-1 polarization is
associated with a positive clinical prognosis in certain cancers, the use of a Type-1
polarizing cytokine such as IL-12 may provide an additive or synergistic effect on
antigen-specific CTL generation by PD-1 blockade. In Figure 5-3, IL-12 treatment
during in vitro peptide stimulation selective enhanced the generation of melanoma
antigen-specific CD4+ Th1 cells while suppressing Th2 cells.
The overall phenotypes of PD-1-blocked and IgG4 control-treated tetramer+ CTLs
were similar, both demonstrating an activated effector / effector memory phenotype
characterized by a CD45RA(low), CD45RO(high), CCR7(low), CD62L(low), and
CD44(high) expression profile on day 11 of peptide stimulation (Figure 5-6). They
were also similar in the mean channel fluorescence and percent expression of PD-1,
PD-L1, PD-L2, CTLA-4, CD44, CD69, CD25, CD125, CD28, CD137, LFA-1,
LFA-2, CD43, and VLA-4, (Figures 5-6 and 5-7). The increase in tumor
recognition efficiency by degranulation shown in Chapter 3 was therefore unlikely
to be associated with increased costimulatory (CD28, CD137) or adhesion molecule
100
expression (LFA-1, LFA-2, CD43, VLA-4). Another important finding is that PD-1
blockade during peptide stimulation did not inhibit the expression of PD-1 on
tetramer+ CTLs, suggesting that sustained expansion of PD-1-expressing CTLs in
vitro and in vivo likely requires continuous treatment with anti-PD-1 antibody. This
has implications for cancer vaccines, whereby repetitive immunization protocols are
often used to boost antigen-specific T cells. This also has implications for ex vivo
expansion of antigen-specific CTLs and TILs used for adoptive transfer therapies,
whereby T cells are expanded over weeks of culture before clinical use (Yee, 2002;
Dudley, 2005; Zhou, 2005). If the PD-1-abrogating antibody is withdrawn during
subsequent rounds of antigen-driven expansion, it is likely that PD-1-expressing
antigen-specific T cells would once again become susceptible to inhibition
mediated by the PD-1 pathway.
101
Chapter 6
CD4+ T cell help augments antigen-specific human memory CTL
generation upon ex vivo peptide stimulation
Introduction:
The production of antibodies is best adapted for generating immune responses
against extracellular pathogens such as viruses and bacteria. Tumors, on the other
hand, typically express endogenous antigens that are true self proteins. As such, T
cell-mediated responses are likely best-suited for defense against tumors.
Traditionally, anti-tumor function has been associated with MHC class I-restricted
CTLs with cytolytic activity. However, there is abundant evidence that CD4+ T
helper cells can play a relevant role in protective anti-tumor responses. It has been
suggested that augmented cellular immunity against tumors may involve direct
recognition of MHC class II-restricted peptide epitopes at a tumor site,
enhancement of CTL activation via CD40-mediated DC maturation, and
maintenance of CTL proliferation during effector phase immune responses
(Schoenberger, 1998, Giuntoli, 2002, Knutson, 2005).
CD4+ T cells appear to orchestrate the initiation and maintenance of adaptive CTL
responses. Recent work has demonstrated that the establishment of long-term
memory CTLs in mice is dependent on cross-presentation of antigen to CD4+ T
cells during the initial priming phase of naïve CD8+ T cells (Shedlock, 2003;
102
Janssen, 2005). Interestingly, once memory CTLs in mice are established via
CD4+ T help during priming, CD4+ T cells become dispensable during secondary
encounter with antigen either ex vivo or in vivo (Shedlock, 2003; Janssen, 2005).
The applicability of this phenomenon in humans has yet to be established. In this
study, we present evidence that human memory CTLs may not follow such a model,
and appear to have a more stringent requirement for CD4+ T help during secondary
antigen-driven expansion. In humans, both vaccine-induced influenza- and
melanoma antigen-specific CTLs arising from natural infection or vaccination with
attenuated virus appear to require CD4+ T help during ex vivo antigen stimulation
for optimal expansion.
In Chapter 3, we found that PD-1 blockade of human CD3+ T cells resulted in
augmented expansion of antigen-specific CD8+ CTLs. We also observed that
absolute numbers of CD4+ cells were also increased, suggesting the possibility that
PD-1 abrogation may indirectly augment antigen-specific CTL expansion via
augmented CD4+ T helper cell expansion. In this chapter, we assessed if vaccine-
induced gp100- and MART-1-specific human CTLs, and also influenza-specific
CTLs, require antigen-specific CD4+ T help to optimally proliferate in response to
antigen re-encounter ex vivo. This sheds light on the relative importance of CD4+ T
cells in human memory CTL maintenance, and the likely points to basic differences
between mouse and human immune system.
103
Figure 6-1. CD8+ T memory cells generated in CD4+/+ mice are T helper-
independent during ex vivo peptide stimulation. Antigen-specific CD8+ T
memory cells isolated from vaccinated wild type mice efficiently expand during ex
vivo peptide stimulation in the absence of CD4+ T cells. Conversely, CD8+ T
memory cells isolated from vaccinated CD4-/- mice are comparatively deficient at
proliferating during ex vivo peptide stimulation in the absence of CD4+ T cells
(Shedlock, 2003; Janssen, 2005).
104
Figure 6-2. In vitro peptide stimulation model to assess the requirement for
CD4+ T help during human CD8+ T memory responses during ex vivo peptide
stimulation. CD8+ T cells were stimulated with autologous peptide-pulsed DCs
alone, or with CD4+ T cells. CD4+ T cells were stimulated with the MART-1
51-73
T helper peptide epitope, using specimens from melanoma patients vaccinated with
the peptide (Wong, 2004).
105
Results:
CTLs specific for influenza and melanoma antigens require antigen-specific
CD4+ T help for optimal expansion during ex vivo peptide stimulation.
Using a stimulation model analogous to that used to study the effects of PD-1
blockade on the generation of antigen-specific CTLs, we determined that the ability
of both influenza- and melanoma antigen-specific CTLs to expand during ex vivo
stimulation was significantly impaired in the absence of activated CD4+ T help.
The frequencies of antigen-specific gp100-, MART-1, and influenza-specific CTLs
were assayed by IFN- γ ELISPOT assay. The frequencies of IFN- γ-secreting
antigen-specific CTLs stimulated by DCs cross-presenting activating helper
peptides to CD4+ T cells were significantly greater than that resulting from CTLs
stimulated alone (Figures 6-3, 6-4, and 6-5). Stimulation of CD4+ T cells with a
control helper peptide (derived from HIV RT) did not provide the same effect on
antigen-specific CTL generation, demonstrating that antigen-mediated activation of
CD4+ T helper cells is required to augment the generation of antigen-specific CTLs
in our model.
106
Figure 6-3. Influenza-specific human CD8+ T memory cells are T helper-
dependent. In the absence of CD4+ T help, influenza-specific CD8+ T memory
cells do not expand efficiently upon ex vivo peptide stimulation. Generation of
functional, IFN- γ-secreting cells was restored by the addition of CD4+ T cells that
were stimulated by a MART-1-derived class II helper peptide. Three patient
specimens were completed with similar results. Data shown represent # spots after
subtraction for background reactivity against an HIV peptide (< 10 spots).
107
Figure 6-4. Gp100-specific human CD8+ T memory cells are T helper-
dependent. In the absence of CD4+ T help, influenza-specific CD8+ T memory
cells do not expand efficiently upon ex vivo peptide stimulation. Generation of
functional, IFN- γ-secreting cells was restored by the addition of CD4+ T cells that
were stimulated by a MART-1-derived class II helper peptide. Two patient
specimens were completed with similar results. Data shown represent # spots after
subtraction for background reactivity against an HIV peptide (< 10 spots).
108
Figure 6-5. MART-1-specific human CD8+ T memory cells are T helper-
dependent. In the absence of CD4+ T help, influenza-specific CD8+ T memory
cells do not expand efficiently upon ex vivo peptide stimulation. Generation of
functional, IFN- γ-secreting cells was restored by the addition of CD4+ T cells that
were stimulated by a MART-1-derived class II helper peptide. Two patient
specimens were completed with similar results. Data shown represent # spots after
subtraction for background reactivity against an HIV peptide (< 10 spots).
109
Discussion:
The results in this chapter suggest that human memory CTLs induced by either
peptide vaccination (i.e. melanoma-specific) or arising from natural infection and/or
protein vaccination (i.e. influenza-specific) may be CD4+ T helper dependent
during the memory recall phase. Using samples from vaccinated melanoma
patients, we determined that during in vitro memory recall both melanoma- and
influenza-specific CTLs required activated antigen-specific CD4+ T helper cells for
the optimal expansion of functional IFN- γ-secreting effector cells. These findings
are consistent with a similar study of human CMV-specific peripheral blood CTLs
(Salkowitz, 2004).
The results in this chapter suggest that human memory CTLs specific for various
types of antigens may generally require CD4+ T help during secondary expansion.
This is contrary to that observed in mice, whereby helper-independent memory
CTLs are generated by vaccination of wild-type strains with virus-infected cells
(Shedlock, 2003; Janssen, 2004). In our in vitro PD-1 abrogation model described
in Chapter 3, we observed augmented expansion in absolute numbers of CD8+,
CD4+, and melanoma antigen-specific CTLs. We observed increased absolute
numbers of CD4+ cells in PD-1-blocked cultures, leaving the possibility that the
augmented frequencies and absolute numbers of gp100- and MART-1-specific
CTLs is, in part, due to the augmented provision of CD4+ T help. The observations
110
in this chapter suggest that human memory CTLs, at least in vitro, are indeed
dependent on the provision of CD4+ T help during memory recall responses and
can be augmented by co-culture with activated CD4+ T helper cells. Hence, the
augmented expansion of post-vaccination melanoma antigen-specific CTLs in our
model may be directly due to PD-1 abrogation in antigen-specific CTLs, and
indirectly due to PD-1 abrogation in CD4+ T helper cells. Studies to test the role of
CD4+ T cells in mediating the augmenting effects of PD-1 blockade would involve
depletion of CD4+ cells from our in vitro peptide stimulation model, described in
Chapter 3.
111
Chapter 7
Conclusion
The CD28 family of costimulatory molecules is involved in fine-tuning immune
responses in both normal and pathological settings. These molecules provide
critical positive and negative signals that initiate, sustain, attenuate, and/or
terminate lymphocyte responses. Two major groups of costimulatory receptors that
modulate T cells have been described: (1) the Ig superfamily that includes CD28,
PD-1, CTLA-4, B- and T-lymphocyte attenuator-4, B7-H3, and ICOS, and (2) the
TNF superfamily that includes OX40, CD27, 4-1BB, CD30, and herpes-virus entry
mediator (Peggs, 2005). Negative costimulatory members of the Ig superfamily,
CTLA-4 and PD-1, are known to restrict immune responses against self antigens.
As such, these molecules have emerged as potential therapeutic targets for treating
autoimmune disease and also for cancer immunotherapy. If expression of CTLA-4
and/or PD-1 can alter the expansion of tumor antigen-specific T cells and diminish
their functional recognition of tumor targets, then abrogation of those
immunoregulatory molecules might have clinical utility, especially in strengthening
recognition of known self-antigens used in many cancer vaccine strategies.
Recent studies in murine tumor model systems and in human cancer vaccine trials
have suggested that antibody-mediated blockade of the CTLA-4 pathway enhances
112
T cell activity and correlates with clinical benefit (Hodge, 2005; Phan, 2003;
Sanderson, 2005; Attia, 2005). Treatment with a human CTLA-4-abrogating
antibody alone or as an adjunct to a peptide vaccine caused significant evidence of
auto-immunity as well as anti-tumor responses in melanoma patients (Phan, 2003;
Sanderson, 2005; Attia, 2005). Interestingly, clinical response and time to relapse
appeared to correlate with autoimmunity associated with CTLA-4 abrogation
(Sanderson, 2005; Attia, 2005). The PD-1 pathway in T cells, which has a distinct
mechanism of action from that of CTLA-4 (Parry, 2005), is known to attenuate T
cell activation both in vivo and in vitro (Nishimura, 1999; Latchman, 2004;
Nishimura, 2001; Salama, 2003; Curiel, 2003; Cai, 2004; Blank, 2006; Hirano,
2005; Iwai, 2002; Iwai, 2005;). PD-1 knockout mice, like CTLA-4 knockouts,
exhibit an autoimmune phenotype and die early from extensive lymphocyte
infiltration of vital organs (Nishimura, 2001). That would suggest that blockade of
PD-1, like CTLA-4, may overcome tolerance mechanisms in vivo. Recent studies
also suggest that the PD-1 pathway may directly limit the expansion of antigen-
stimulated T cells. Blockade of PD-L1 was shown to increase the absolute numbers
of human tumor antigen-specific CTLs during in vitro antigen stimulation (Blank,
2006). It was also shown that PD-L1 blockade and, to a lesser extent, PD-1
blockade, restored in vivo proliferation of viral antigen-specific CTLs that were
exhausted by chronic infection in mice (Barber, 2006). To date, no study has
113
examined the effects of PD-1 blockade on the expansion and recognition efficiency
of human T cells specific for tumor antigens.
In this work, we used a human in vitro system to test whether PD-1 abrogation by a
fully human antibody would alter the expansion and function of human tumor
antigen-specific CD8+ T cells (CTLs). We found that PD-1 blockade during
peptide stimulation increased the absolute numbers of CD8+, CD4+, and melanoma
antigen-specific CTLs (Figure 3-6). These results correlated with increased
MHC:peptide tetramer labeling of melanoma antigen-specific CTLs and increased
frequencies of IFN- γ-secreting antigen-specific CTLs in ELISPOT assays (Figures
3-3, 3-4, and 3-10). This effect was seen using the full-length PD-1 antibody as
well as the corresponding monovalent F(ab) fragment (Figure 3-9).
PD-1 expression on uncultured T cell subsets differed between melanoma antigen-
specific (tetramer+) CTL populations, CD8+/tetramer-, and CD4+ populations. The
fraction of PD-1-expressing cells among the CD8+/tetramer- populations ranged
from 0% to 40.8%, with an average of 14.7% +/- 13.4 (Table 2-3). The fraction of
PD-1- expressing cells among the CD4+ populations ranged from 0.8% to 5.8%,
with an average of 4.3% +/- 1.6 (table 2-3). These are compared to an average of
58.7% +/- 23.7 PD-1 expression on melanoma antigen-specific CTLs (Table 2-1).
These observations suggest that antibody-mediated PD-1 abrogation during
114
stimulation with melanoma peptides in vitro and in vivo would preferentially affect
melanoma antigen-specific CTLs, and that toxicity of a PD-1-abrogating antibody
may thus be limited due to comparatively low expression of PD-1 on the general
CD4+ and CD8+ T cell population.
PD-1 blockade generated melanoma antigen-specific effectors that demonstrated
increased lysis of T2 targets pulsed with heteroclitic peptides used for vaccination
as well as gp100- and MART-1-expressing melanoma cell lines (Figures 3-11 and
3-12). PD-1-blocked tetramer+ cells demonstrated an activated effector / effector
memory phenotype characterized by a CD45RA(low), CD45RO(high), CCR7(low),
CD62L(low), and CD44(high) expression profile on day 11 of peptide stimulation,
but were not remarkably different from IgG4 control-treated cells (Figures 5-6 and
5-7). Both Type-1 and Type-2 cytokine secretion was increased in PD-1-blocked T
cells, suggesting that the PD-1 abrogating antibody used in our study does not alter
T cell polarization (Figures 5-1 and 5-2). This finding is of significance because in
some cancers, including melanoma, there is evidence that a Type-1 polarization of
tumor antigen-specific CD4+ T cells in the peripheral blood and in the tumor
microenvironment may be more effective in controlling tumors (Tatsumi, 2002).
Thus, there is a strong rationale for the adjunctive use of IL-12 in a clinical vaccine
setting – with or without antibody-mediated PD-1 blockade. In Chapter 5, we
115
found that IL-12 was capable of selectively enhancing MART-1-specific CD4+ Th1
cells while suppressing Th2 cells during in vitro peptide stimulation (Figure 5-3).
The lack of intrinsic Type-1-polarizing effects of antibody-mediated PD-1 blockade
suggests that the combinatorial use of IL-12 with a PD-1-abrogating antibody may
result in additive or synergistic anti-tumor immune responses. IL-12 has been used
as a cytokine adjuvant in peptide-based clinical cancer vaccine trials, and has been
demonstrated to increase immune and clinical responses (Lee, 2001; Peterson,
2003; Cebon, 2003; Gajewski, 2001). Furthermore, IL-12 has also been
demonstrated to inhibit activation-induced cell death of CD8+ and CD4+ T cells via
inhibition of caspase-dependent apoptotic pathways (Palmer, 2001; Lee, 2003). In
our in vitro peptide stimulation model, we found that PD-1 blockade augmented the
proliferation of melanoma antigen-specific CTLs without any significant affect on
apoptosis (Figures 4-1 to 4-4). Thus, it is anticipated that the combinatorial use of
IL-12 and antibody-mediated PD-1 blockade would have an additive/synergistic
effect on the frequencies and absolute numbers of melanoma-antigen-specific CTLs
in our in vitro model by augmenting proliferation of CTLs, augmenting CD4+ Th1
expansion, and inhibiting activation-induced apoptosis of T cells.
It has been suggested that CD107a externalization is a marker for high RE human T
cells with functional capacity against tumor targets (Rubio 2003; Stuge, 2004;
116
Khort, 2005). T cell RE – a T cell’s sensitivity to peptide concentration – may be
an important determinant in the functional recognition of tumors (Zeh, 1999;
Dutoit, 2001; Bullock, 2001; O’connor, 2002; Molldrem, 2003; Yang, 2002). As
such, vaccine approaches to cancer immunotherapy should focus on not only
increasing the frequencies of tumor antigen-specific T cells, but also augmenting T
cell RE. The data presented in our study show that antibody-mediated PD-1
blockade during in vitro peptide stimulation enhances the generation of high RE
gp100- and MART-1-specific CTLs. These high RE CTLs were directly
enumerated on the basis of CD107a mobilization to the cell membrane after
exposure to gp100+/MART-1+ melanoma targets (Figures 3-16 and 3-17). Further
evidence supporting the generation of high RE tumor antigen-specific effectors was
shown by increased lysis of gp100+/MART-1+ melanoma targets by PD-1-blocked
cells (Figure 3-13). These data show that PD-1 blockade during peptide stimulation
increases the quantity and quality of an anti-tumor T cell response, and has direct
implications for the design of novel cancer immunotherapy strategies.
The exact mechanism(s) of enhanced generation and enrichment of high RE CTLs
by PD-1 blockade in our model is currently unclear. We observed increased
absolute numbers of CD4+ cells in PD-1-blocked cultures, leaving the possibility
that the augmented frequencies and absolute numbers of gp100- and MART-1-
specific CTLs is, in part, due to the augmented provision of CD4+ T help. We
117
found that influenza and melanoma antigen-specific CTLs from humans were
dependent on activated CD4+ T cells for efficient expansion upon ex vivo peptide
stimulation (Figures 6-3, 6-4, and 6-5). This phenomenon suggests that human
memory CTLs, at least in vitro, may generally be “helper-dependent,” unlike CTLs
in experimental mouse models (Shedlock, 2003; Janssen, 2005). Hence, the
augmented expansion of post-vaccination melanoma antigen-specific CTLs in our
model may be directly due to PD-1 abrogation in antigen-specific CTLs, and
indirectly due to PD-1 abrogation in CD4+ T helper cells. The effect of CD4+ T
cell depletion in our in vitro model is currently being pursued.
It was intriguing that PD-1 blockade did not significantly affect apoptosis of
antigen-specific CTLs in our model. Rather, the most significant effect of PD-1
blockade was on T cell proliferation, measured by intracellular Ki67 expression and
CFSE dilution (Figures 4-2 and 4-4). This is consistent with an existing hypothesis
that an unknown PD-ligand-binding receptor, distinct from PD-1, may transduce an
apoptotic signal in T cells (Dong, 2002). To that end, identification of additional
PD-ligand-binding receptors that inhibit T cell activity may prove useful in
elucidating additional targets for cancer immunotherapy. If PD-L1 and/or PD-L2
induce apoptosis via another receptor distinct from PD-1, then simultaneous
blockade of PD-1 and PD-L1/PD-L2 may have additive/synergistic effects on T cell
generation by augmenting proliferation in addition to decreasing cell death. In
118
other words, blockade of PD-1, PD-L1, and or PD-L2 may not be redundant.
Genome-wide searches using computational methods to predict protein-protein
interactions may be a rational starting point for identifying additional PD-ligand-
binding receptors (Salwinski, 2003).
Information on the molecular and biochemical changes in T cells induced by the
PD-1 pathway remains limited. In vitro studies of non-specifically activated human
CD4+ T cells suggest that the PI3K/Akt pathway is inhibited by PD-1 ligation
(Parry, 2005). Akt is a serine/threonine kinase that tranduces anti-apoptotic and/or
proliferative signals in T lymphocytes (Ahmed, 1997; Brennan 1997). Flow-
cytometry based methods to detect changes in phospho-Akt levels can be used to
assess the effect of PD-1 blockade on antigen-stimulated CTLs. Akt activation is
mediated by phosphatidylinositol 3-kinase (PI3K) (Coffer, 1998), and together are
commonly referred to as the PI3K/Akt pathway. A number of selective small
molecule PI3K inhibitors have been described, including LY294002, quercetin,
myricetin, and staurosporine (Walker, 2000). It is expected that pre-treatment of T
cells with a PI3K inhibitor would abolish the augmenting effects of PD-1 blockade
in our in vitro stimulation model, via inhibition of Akt phosphorylation.
The study by Parry et al. also demonstrated that the CTLA-4 pathway inhibits Akt
phosphorylation via a distinct mechanism from the PD-1 pathway. Serine/threonine
119
phosphatase 2A (PP2A) associates with the CTLA-4 cytoplasmic domain and
results in downstream dephosphorylation of Akt in activated CD4+ T cells
(Chuang, 2000; Parry, 2005). CTLA-4 was expressed at low levels (<10%) on
melanoma antigen-specific CTLs after in vitro peptide stimulation, with or without
antibody-mediated PD-1 blockade (Figure 5-6). Taken together, this suggests that
CTLA-4 may predominately modulate CD4+ T cells. Indeed, the
lymphoproliferative disorder observed in CTLA-4 knockout mice is the result of
deregulated peripheral control of CD4+ T cell activation (Khattri, 1999). As such,
it is expected that combinatorial antibody-mediated PD-1 and CTLA-4 blockade
would have an additive/synergistic effect on antigen-specific CTL generation by
modulation of the CD4+ T helper component of the immune system, and not by
directly modulating costimulatory signaling in CD8+ T cells. Studies to examine
the combinatorial effects of antibody-mediated PD-1 and CTLA-4 blockade in vitro
are ongoing in our lab.
Simultaneous targeting of other costimulatory pathways in addition to the PD-1
pathway in T cells may be another plausible strategy to enhance the generation of
functional effector T cells. Potential targets include CD134 (OX40), CD137 (4-
1BB), and CD40. OX40 is a TNF superfamily receptor (TNFR) expressed on
CD4+ and CD8+ T cells (Murata, 2006). OX40 signaling has been shown to
augment CD4+ T cell survival through upregulation of the mitrochondrial
120
membrane-associated anti-apoptotic proteins Bcl-X
L
and Bcl-2 (Rogers, 2001). An
agonistic anti-OX40 antibody utilized as an adjunct to a whole cell vaccine in mice
enhances CD8+ T cell-mediated immunity directed against an immunizing tumor
antigen (HER-2/neu) (Murata, 2006). 4-1BB is another TNFR, and was initially
implicated in regulation of CD8+ T cell responses through, in part, control of Bcl-
X
L
expression (Hurtado, 1995; Tan, 1999). Recently, a recombinant 4-1BB-ligand-
expressing vaccinia virus was shown to enhance the induction of CD4+ and CD8+
T cells against carcinoembryonic antigen (CEA) in mice (Kudo-Saito, 2006). This
correlated with enhanced expression of Bcl-X
L
and BCL-2 in CEA-specific CD4+
and CD8+ T cells, accumulation of TILs in tumors, and tumor regression
(Kaufman, 2006). Furthermore, 4-1BB:4-1BBL interactions are involved in
sustaining CD8+ T cell proliferation and effector function mediated by direct
cell:cell interaction with CD4+ T cells (Giuntoli, 2002). CD40, another TNFR, is
expressed on CD8+ T cells upon activation (Bourgeois, 2002). Ligation of CD40
via CD40-ligand (CD40L) expressed on activated CD4+ T helper results in
augmented CD8+ T cell proliferation and cytokine secretion (Bourgeois, 2002).
These findings suggest that augmentation of CD8+ T cell responses via the
provision of CD4+ T help is mediated both indirectly (via maturation of DCs by
CD40L-expressing CD4+ T cells) and directly (via direct cell:cell interactions with
CD4+ T cells) (Schoenberger, 1998; Bourgeois, 2002). A CD40-agonizing
antibody has been shown to substitute for CD4+ T help during CD8+ T cell
121
induction in vivo (Bennett, 1998). Taken together, concomitant targeting of distinct
signaling pathways in T cells – i.e. blocking inhibitory receptors such as PD-1 in
addition to agonizing activating receptors such as OX40 – are intriguing, multi-
faceted approaches towards enhancing antigen-specific T cell induction.
Currently, it is unclear what effect PD-1 abrogation would have on the expansion
and function of T regulatory (Treg) cells. Tregs are specialized subsets of T cells
that suppress immune responses via: (1) direct cell:cell contact and (2) secretion of
immunosuppressive cytokines such as IL-10 and TGF- β (Sagakuchi, 2004;
Sagakuchi, 2005; Mills, 2004; Vigouroux, 2004). Naturally occurring Tregs,
characterized by a CD4+/CD25+/FoxP3+ expression profile, represent 5-10% of the
circulating CD4+ T cell population in humans and exert suppressive activity on
naïve and effector T cells via direct cell:cell contact (Sagakuchi, 2005). CD4 and
CD25 (IL-2 receptor alpha chain) expression is not directly involved in the
suppressive function of natural Tregs (Sagakuchi, 2005). However, FoxP3 is a
transcription factor that inhibits nuclear factor of activated T cells (NF-AT) and NF-
κB transcriptional activity in CD4+ T cells, genes that are essential for cytokine
gene expression and effector function (Betelli, 2005). However, it is currently
unclear how exactly FoxP3 directs the contact-dependent suppressive activity of
natural Tregs.
122
In our in vitro peptide stimulation model described in Chapter 3, the effect of
antibody-mediated PD-1 blockade was assessed in bulk CD3+ T cells that contain
subsets of melanoma antigen-specific CD8+ T cells. Despite the existence of
natural Tregs within the bulk human CD3+ T cell population, the net effect of PD-1
blockade was an augmentation of melanoma antigen-specific CD8+ T cell
expansion and functional capacity. It is unclear if the natural Treg subpopulation in
our model was also augmented by antibody-mediated PD-1 blockade. Expansion of
CD4+/CD25+/FoxP3+ T cells could be assessed by CFSE dilution analysis, to
determine if PD-1 blockade also affects natural Treg expansion in our model.
Another possibility is that PD-1 abrogation may inhibit natural Treg function. To
our knowledge, no information in the literature describes the specific molecules that
mediate the contact-dependent suppressive activity of natural Tregs. It is known
that PD-ligands expressed on activated T cells are involved in down-modulating T
cell proliferation and cytokine secretion via T cell:T cell interactions (Seo, 2006). It
remains to be determined if natural Treg cells are in any way involved in these
interactions. If expression of PD-ligands is necessary for the contact-dependent
suppressive function of natural Treg cells, then natural Treg cells isolated from PD-
L1 and/or PD-L2 knockout mice should have impaired suppressive activity in vitro
during T cell proliferation assays and also after adoptive transfer to hosts. Such a
study would implicate a role for PD-ligands in natural Treg suppressive function,
123
and would suggest that the augmenting effects of PD-1 abrogation is, in part,
resulted from inhibition of contact-dependent Treg interactions with effector T
cells.
The use of a PD-1-abrogating antibody ex vivo may also improve the growth and
functional profile of expanded human T cells used in current anti-cancer adoptive
transfer strategies (Dudley, 2005; Yee, 2002). A previously described clinical
protocol for ex vivo expansion of tumor antigen-specific CTLs involves three
consecutive cycles of stimulation using autologous peptide-pulsed DCs followed by
cloning of antigen-specific CTLs by limiting dilution (Yee, 2002). Limiting
dilution is carried out by expansion of single progenitor cells using anti-CD3
antibody, IL-2, and irradiated allogeneic peripheral blood mononuclear cells (Yee,
2002). The data presented in Chapter 3 suggest that antibody-mediated PD-1
blockade during simulation with autologous peptide-pulsed DCs augments the
generation of antigen-specific CTLs that recognize tumor cell lines by
degranulation. Thus, the use of a PD-1-abrogating antibody may facilitate the
isolation and cloning of functional antigen-specific CTL clones that are capable of
exerting effector function against tumors presenting cognate antigen. The clinical
effectiveness of adoptive transfer therapy using cloned antigen-specific CTLs as
well as ex vivo-expanded TILs has been mixed thus far (Aebersold, 1991; Huang,
2005; Zhou, 2005; Dudley, 2005; Yee, 2002). One possible obstacle towards the in
124
vivo effectiveness of adoptively transferred T cells may be the expression of PD-1.
Because PD-1 appears to be highly upregulated on antigen-specific CTLs that are
chronically exposed to antigen, it is likely that ex vivo-expanded T cells generated
via current clinical protocols also express PD-1 (Barber, 2006; Day, 2006). Hence,
the in vivo persistence of adoptively-transferred T cells may be limited, in part, by
the PD-1 pathway. In such a scenario, the use of a PD-1-abrogating antibody would
likely increase anti-tumor efficacy of adoptively-transferred T cells.
In conclusion, this work suggests that PD-1 blockade by a human antibody may be
a useful strategy to augment the generation of functional, antigen-specific CTLs
that efficiently recognize tumor targets. Based in part on the work herein, the
clinical utility of a PD-1-abrogating antibody in vivo should be tested alone and in a
phase I cancer vaccine trial.
125
Materials and methods
Generation of a fully human PD-1-abrogating monoclonal antibody.
Antibodies to human PD-1 and a matching IgG4 isotype control were kindly
provided by Medarex, Incorporated (Milpitas, CA). Anti-PD-1 antibodies were
generated in mice transgenic for human Ig loci by immunization with Chinese
hamster ovary (CHO) transfectants expressing human PD-1, followed by boosting
with PD-1/human IgG1 Fc fusion protein. Antibodies were screened for binding to
PD-1 transfectants and activated human T cells, and the ability to enhance T cell
proliferation and IFN- γ secretion in allogeneic DC:T cell mixed lymphocyte
reactions (MLR). The antibodies were also tested for lack of reactivity to the PD-1
homologues CD28, CTLA-4, and inducible T cell costimulator (ICOS). Anti-PD-1
antibody (clone 5C4), which demonstrated high affinity and specificity for PD-1,
was selected and expressed in CHO cells. This antibody inhibited binding of
soluble PD-L1/PD-L2-Ig to plate-bound PD-1 and to PD-1-expressing CHO cells,
confirmed by enzyme-linked immunosorbant assay and flow cytometry,
respectively (data not shown). A corresponding monovalent F(ab) fragment was
also tested, which displayed similar activity to the full-length 5C4 antibody in
allogeneic DC:T cell MLRs and antigen-specific stimulations described herein
(Figure 3-8). To reduce binding to Fc receptors, the 5C4 antibody was converted to
an IgG4 antibody with a hinge mutation (S228P) (Wang, C. et al.; manuscript
126
submitted). An IgG4 (S228P) isotype control antibody specific for Diphtheria toxin
– which demonstrated no effect on T cell proliferation – was generated and used in
this study.
Preparation of patient peripheral blood mononuclear specimens.
Apheresis specimens were collected from stage III/IV resected melanoma patients
(all HLA-A*0201+) who were vaccinated with the gp100
209-217(210M)
(gp100-2M)
and/or MART-1
26-35(27L)
(MART-27L) heteroclitic peptide analogs. Heteroclitic
peptide analogs bind more stably to MHC than their native sequences (Ruppert,
1993) and have been used as vaccines in cancer patients to induce T cells capable of
lysing tumor cells ex vivo (Valmori, 1998; Fong, 2001). All patients were
participants in clinical melanoma vaccine trials – including those published
previously (Lau, 2001; Lee, 2001) – conducted at the University of Southern
California/Norris Comprehensive Cancer Center (Los Angeles, CA). All patients
were required to comprehend and sign an informed consent form approved by the
Los Angeles County/University of Southern California Institutional Review Board
prior to treatment. Pre- and post-vaccination apheresis specimens were processed
to purify peripheral blood mononuclear cells (PBMCs) by sedimentation on
Lymphoprep (Greiner Bio-One, Longwood, FL) and extensive washing in Hank’s
Balanced Salt Solution (HBSS; Mediatech, Herndon, VA). Cells were frozen in
40% human AB serum (HS; Omega Scientific, Tarzana, CA), 50% RPMI 1640
127
medium (Mediatech) and 10% dimethyl sulfoxide (Sigma, St. Louis, MO). All
PBMCs were stored in a secure liquid nitrogen freezer at -168º C until use. T cells
from post-vaccination PBMC samples were used in this study.
Generation of dendritic cells and CTL stimulation.
DCs found in lymphoid tissue demonstrate varying maturational states (Shortman,
2002). As such, we used monocyte-derived DCs with an intermediate
immature/mature phenotype as antigen-presenting cells for our in vitro stimulation
model. Plastic adherent monocytes from PBMC specimens were cultured in X-
VIVO 15 (Cambrex, East Rutherford, NJ) supplemented with 1000 U/ml each of
rhGM-CSF (Berlex, Richmond, CA) and rhIL-4 (R&D, Minneapolis, MN). On day
six of culture, rhGM-CSF and rhIL-4 were replenished (1000 U/ml each). The
following day, DCs were pulsed (minimum 2 hr) with gp100-2M, MART-27L, or
control HLA-A2*0201-binding peptide HIV RT
476-484
(HIV) (10 µg/ml each). All
DCs were then harvested with warm Phosphate Buffered Saline (PBS; Mediatech),
washed and resuspended in AIM V medium (Invitrogen, Carlsbad, CA)
supplemented with 5% HS. As described previously, these DCs possess the ability
to capture soluble antigen and also express CD80, CD86, HLA-DR, and PD-L1
(Brown, 2003; Sallusto, 1995). We confirmed surface expression of these
molecules by flow cytometry (data not shown). All DCs were added
(2x10
4
/100µl/well) to 96-well round bottom cluster plates (Corning, Corning, NY).
128
Because PD-1-dependent interactions may occur naturally between different T cell
subsets (i.e. CD4+ and CD8+) (Seo, 2006), bulk CD3+ T cells were used as
responders for peptide stimulation. Autologous CD3+ cells were purified from
thawed PBMCs, using the MACS (Miltenyi Biotech, Auburn, CA) technique for
positive selection. Using the manufacturer’s protocol, a purity of ≥ 95% CD3+
cells was routinely achieved as confirmed by flow cytometry (data not shown). The
CD3- PBMC population was refrozen in 90% HS / 10% dimethyl sulfoxide for later
use as antigen-presenting cells in ELISPOT assays. The purified CD3+ cells were
washed and resuspended in AIM V 5% HS, anti-PD-1 antibody or matching IgG4
isotype control (10 µg/ml final each), and then added to peptide-pulsed DCs. All
cultures were incubated for 11 days at 37 °C / 5% CO
2
. For 11 day stimulations
(one cycle), exogenous cytokines were not added at any point.
Dendritic cell phenotyping.
DCs were cultured as described above, harvested, and washed with PBS + 1% FBS
(FACS buffer). Each sample was Fc-blocked with HS (10 µl) for 15 minutes at
room temperature. Thereafter, all samples were stained according to manufacturer
protocol with CD11c-PE-Cy5 and one of the following: CD80-PE, CD86-PE,
HLA-A,B,C-FITC, HLA-DR-PE (all from Pharmingen, San Diego, CA), PD-L1-
PE, or PD-L2-PE (both from Ebiosciences, San Diego, CA). Following incubation,
129
all samples were washed once with PBS and resuspended in FACS buffer prior to
analysis on an FC500 flow cytometer (Beckman-Coulter, San Diego, CA).
MHC:peptide tetramer labeling.
Aliquots of fresh PBMCs (3x10
6
cells/test) and peptide-stimulated (one cycle)
CD3+ effector cells (10
6
cells/test) were stained according to manufacturers’
protocols with gp100-2M, MART-27L HLA-A*0201 tetramer-PE, or pan MHC
class I negative control tetramer (all from Beckman-Coulter, San Diego, CA), CD3-
PE-Cy7, CD4-PE-Cy5 (both from Pharmingen), and CD8-ECD (Beckman-Coulter).
Following incubation, all samples were washed with PBS and fixed with PBS +
0.5% formaldehyde (minimum 1 hour at room temperature). Samples were
analyzed within 24 hours on an FC500 flow cytometer (Beckman-Coulter). PD-1
expression was measured in freshly-thawed post-vaccination PBMCs by using an
anti-PD-1-FITC antibody (Pharmingen).
For phenotyping of tetramer+ cells in freshly-thawed PBMCs and in vitro-
stimulated CD3+ effector cells, the following antibodies were used according to
manufacturers’ protocols: CD45RO-PE-Cy5, PD-1-FITC, CTLA-4-PE-Cy5,
CD44-FITC, CD69-PE-Cy5, CD28-FITC, CD137-PE-Cy5, CD25-PE-Cy5, PD-L2-
biotin, Streptavidin-FITC, PD-L1-PE-Cy7, LFA-1-PE-Cy5, LFA-2-FITC, CD43-
130
FITC, and VLA-4-PE-Cy5 (all from Pharmingen), CD122-FITC, and CD45RA-
FITC (both from Beckman-Coulter).
IFN- γ/IL-5 ELISPOT assays.
Cultured CD3+ effector cells were harvested following 11 day in vitro peptide
stimulation and added to ELISPOT IP plates (Millipore, Bedford, MA) in
concentrations of 10
5
and 3.3x10
4
/well (triplicates of 50 µl/well). Autologous CD3-
negative flow-through populations (2x10
5
/50 µl/well) pulsed with gp100 or MART-
1 melanoma peptides (heteroclitic or native) or control HIV peptide (10 µg/ml final
each) were used as antigen-presenting cells. The plates were incubated (20 h) at
37ºC / 5% CO
2
. Color development was performed with 3-amino-9-ethyl-carbazole
(AEC; Sigma). Spot numbers were determined with the aid of computer-assisted
video imaging analysis (Carl Zeiss ELISPOT Reader System; Oberkochen,
Germany), and normalized to 10
5
input cells if needed.
Chromium-release assay.
Peptide-stimulated CD3+ effector cells were plated in triplicate in 96-well U-
bottom microtiter plates (BD Biosciences; San Diego, CA). HLA-A*0201+ T2
target cells were pulsed overnight with gp100 or MART-1 melanoma peptides
(heteroclitic or native) or control HIV RT peptide. An aliquot (1x10
6
) of T2 cells
for each peptide condition was labeled with
51
Cr (Perkin Elmer, Wellesley, MA)
131
and washed with RPMI 2% HS-CM. All T2 cells were combined with K-562 cells
to inhibit natural killer cell-mediated lysis. Target cell mixtures were added (2500
T2 targets/25,000 K-562 cells/100 µl/well) to CD3+ effector cells.
To measure lysis of melanoma targets, PD-1-blocked and IgG4 control-treated
CD3+ effector cells were stimulated for two consecutive cycles (11 days then 7
days) using peptide-pulsed DCs. On day 3 of the second cycle of stimulation, IL-2
(100 U/ml) was added to all conditions. CD8+ cells were MACS-purified from the
total effector population prior to assay against melanoma targets. The HLA-
A*0201+ melanoma lines 526mel, 624mel, and A-375 were used as targets. Both
526mel and 624mel endogenously express the gp100 and MART-1 antigens
(Kawakami, 1994), and encode the gp100
209-217
and MART-1
26-35
immunodominant
epitopes restricted by HLA-A*0201
(Kawakami 1994 and 1995). Melanoma line
A-375 does not express gp100 or MART-1 (Kawakami, 1994), and was used as a
negative control. Each cell line was chromated using the same protocol as T2
targets, and combined with K-562 cells. Target cell mixtures were added (1000
melanoma targets/10,000 K-562 cells/100 µl/well) to CD3+ effector cells.
All plates were centrifuged (300g for 1 minute) and incubated for 5 h at 37ºC / 5%
CO
2
. Supernatants were collected (100 µl) and assayed for radioactivity using a γ-
counter (Packard Cobra-II; Perkin Elmer). Percent cytotoxicity was calculated as
132
follows: [(experimental release – spontaneous release) / (maximum release –
spontaneous release)] x 100.
CD107a mobilization assay.
Incubations were done in 96-well U-bottom microtiter plates. To each well, the
following were added in order: peptide-stimulated (one cycle) CD3+ effector cells
(1x10
6
/well), melanoma targets 526mel, 624mel, or A-375 (2x10
6
/well), and anti-
CD107a-APC antibody (10 µl/well; Southern Biotech, Birmingham, AL). Each
well was gently mixed with a pipette. The plates were centrifuged (300g for 1
minute) and incubated for 5 h at 37ºC / 5% CO
2
. Following incubation, all cells
were washed with PBS + 0.5 mM EDTA to disrupt cell-cell conjugates. Thereafter,
each sample was stained according to manufacturers’ protocols with gp100-2M or
MART-27L HLA-A*0201 tetramer-PE, CD3-PE-Cy7, CD8-ECD, and CD4-FITC.
All samples were then washed with PBS and fixed with PBS + 0.5% formaldehyde
(minimum 1 hour at room temperature). All samples were analyzed within 24
hours on an FC500 flow cytometer.
Bio-Plex cytokine array.
Culture supernatants from in vitro peptide stimulations were collected on days 2, 4,
6, 8, and 11 of stimulation and immediately frozen at -80ºC. Samples where
thawed to room temperature and assayed in duplicates for cytokine content using
133
the Th1/Th2 Bio-plex suspension array kit (Biorad, Hercules, CA). All samples
were analyzed on a Bio-Plex Array Reader (Biorad).
Detection of intracellular Ki67 in antigen-specific CTLs.
PD-1-blocked and IgG4 control-treated CD3+ cells were stimulated in vitro with
autologous melanoma peptide-pulsed DCs for 11 days as described above. Effector
cells were harvested at days 6 and 11 of culture and stained according to
manufacturers’ protocols with gp100-2M or MART-27L HLA-A*0201 tetramer-
PE, CD3-PE-Cy7, CD8-ECD, and CD4-PE-Cy5. All cells were then fixed and
permeabilized with Cytofix/Cytoperm solution (250 µl/sample; Pharmingen) for 20
minutes on ice. Fixed/permeabilized cells were then washed with Perm
Wash/Buffer (Pharmingen) and stained according to manufacturer protocol with
Ki67-FITC or matching IgG1 isotype control (both from Pharmingen). All cells
were then washed with Perm Wash/Buffer and resuspended in PBS + 0.5%
formaldehyde (minimum 1 hour at room temperature) prior to analysis by flow
cytometry.
CFSE dilution in antigen-specific CTLs was measured by labeling CD3+ cells with
0.5 µM CFSE (Invitrogen) for 10 minutes at 37ºC. Labeling was stopped with
100% FBS and subsequent washings with AIM V 5% HS. CFSE-labeled cells were
134
then stimulated with peptide-pulsed DCs as described above, and analyzed by flow
cytometry 11 days thereafter.
Analysis of apoptosis in antigen-specific CTLs.
CD3+ cells were stimulated with autologous peptide-pulsed DCs as described
above. Cells were harvested on days 6, and 11 for Annexin-V and 7-AAD staining.
Briefly, cells were stained with MHC:peptide tetramer-PE, CD4-PE-Cy7, and CD8-
ECD for 30 minutes at room temperature. All samples were then washed once with
FACS buffer, and placed on ice for 20 minutes. Thereafter, all cells were washed
once with PBS and once with Annexin-V wash buffer (Pharmingen). All samples
were then stained with Annexin-V-FITC (5 ul) and 7-AAD (20ul) (both from
Pharmingen) for 15 minutes at room temperature. All samples were resuspended in
Annexin-V wash buffer and analyzed immediately on an FC500 flow cytometer.
To measure intracellular expression of active Caspase-3, effector cells were
harvested at days 6 and 11 of culture and stained according to manufacturers’
protocols with gp100-2M or MART-27L HLA-A*0201 tetramer-PE, CD3-PE-Cy7,
CD8-ECD, and CD4-PE-Cy5. All cells were then fixed and permeabilized with
Cytofix/Cytoperm solution (250 µl/sample; Pharmingen) for 20 minutes on ice.
Fixed/permeabilized cells were then washed with Perm Wash/Buffer (Pharmingen)
and stained according to manufacturer protocol with Ki67-FITC or matching IgG1
135
isotype control (both from Pharmingen). All cells were then washed with Perm
Wash/Buffer and resuspended in PBS + 0.5% formaldehyde (minimum 1 hour at
room temperature) prior to analysis by flow cytometry.
CD8+/CD4+ T cell co-culture system.
DCs were cultured from monocyte precursors using GM-CSF and IL-4 (1000 IU
each) for 7 days. DCs were pulsed with the following peptides: (1) Influenza or
melanoma class I peptides alone, (2) class I + class II MART-1
51-73
helper peptide,
or (3) class I peptides + control HIV RT
171-190
class II peptide. Autologous CD8+
and CD4+ cells were MACS-purified and plated with DCs without exogenous
cytokines. After 10 days, CD8+ and CD4+ cells were MACS purified for
functional analysis by ELISPOT.
136
Bibliography
Aebersold, P., Hyatt, C., Johnson, S., Hines, K., Korcak, L., Sanders, M., Lotze, M,
Topalian, S., Yang, J., and Rosenberg, S.A. 1991. Lysis of autologous melanoma
cells by tumor-infiltrating lymphocytes: association with clinical response. J Natl
Cancer Inst. 83:932-7.
Agata, Y., Kawasaki, A., Nishimura, H., Ishida, Y., Tsubata, T., Yagita, H., and
Honjo, T. 1996. Expression of the PD-1 antigen on the surface of stimulated mouse
T and B lymphocytes. Int Immunol. 8:765-72.
Ahmed, N.N., Grimes, H.L., Bellacosa, A., Chan, T.O., and Tsichlis, P.N. 1997.
Transduction of interleukin-2 antiapoptotic and proliferative signals via Akt protein
kinase. Proc Natl Acad Sci USA. 94:3627-32.
Altman, J.D., Moss, P.A., Goulder, P.J., Barouch, D.H., McHeyzer-Williams, M.G.,
Bell, J.I., McMichael, A.J., and Davis, M.M. 1996. Phenotypic analysis of antigen-
specific T lymphocytes. Science. 274:94-6.
Andree, H.A., Reutelingsperger, C.P., Hauptmann, R., Hemker, H.C., Hermens,
W.T., and Willems, G.M. 1990. Binding of vascular anticoagulant alpha (VAC
alpha) to planar phospholipid bilayers. J Biol Chem. 265:4923-8.
Apostolopoulos, V., Karanikas, V., Haurum, J.S., and McKenzie, I.F. 1997.
Induction of HLA-A2-restricted CTLs to the mucin 1 human breast cancer antigen.
J Immunol. 159:5211-8.
Appleman, L.J., and Boussiotis, V.A. 2003. T cell anergy and costimulation.
Immunol Rev. 192:161-80.
Attia, P., Phan, G.Q., Maker, A.V., Robinson, M.R., Quezado, M.M., Yang, J.C.,
Sherry, R.M., Topalian, S.L., Kammula, U.S., Royal, R.E., et al. 2005.
Autoimmunity correlates with tumor regression in patients with metastatic
melanoma treated with anti-cytotoxic T-lymphocyte antigen-4. J Clin Oncol.
23:6043-53.
Banchereau, J., and Steinman, R.M. 1998. Dendritic cells and the control of
immunity. Nature. 392:245-52.
137
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. 439:682-7.
Barnd, D.L., Lan, M.S., Metzgar, R.S., and Finn, O.J. 1989. Specific major
histocompatibility complex-unrestricted recognition of tumor-associated mucins by
human cytotoxic T cells. Proc Natl Acad Sci USA. 86:7159-63.
Bennett, S.R., Carbone, F.R., Karamalis, F., Flavell, R.A., Miller, J.F., and Heath,
W.R. Help for cytotoxic-T-cell responses is mediated by CD40 signaling. Nature.
1998 Jun 4;393(6684):478-80.
Bettelli, E., Dastrange, M., and Oukka, M. 2005. Foxp3 interacts with nuclear factor
of activated T cells and NF-kappa B to repress cytokine gene expression and
effector functions of T helper cells. Proc Natl Acad Sci U S A. 102:5138-43.
Blank, C., Kuball, J., Voelkl, S., Wiendl, H., Becker, B., Walter, B., Majdic, O.,
Gajewski, T.F., Theobald, M., Andreesen, R., and Mackensen, A. 2006. Blockade
of PD-L1 (B7-H1) augments human tumor-specific T cell responses in vitro. Int J
Cancer. 119:317-27.
Bosch, F.X., Manos, M.M., Munoz, N., Sherman, M., Jansen, A.M., Peto, J.,
Schiffman, M.H., Moreno, V., Kurman, R., and Shah, K.V. 1995. Prevalence of
human papillomavirus in cervical cancer: a worldwide perspective. International
biological study on cervical cancer (IBSCC) Study Group. J Natl Cancer Inst.
87:796-802.
Brennan, P., Babbage, J.W., Burgering, B.M., Groner, B., Reif, K., and Cantrell,
D.A. 1997. Phosphatidylinositol 3-kinase couples the interleukin-2 receptor to the
cell cycle regulator E2F. Immunity. 7:679-89.
Brown, J.A., Dorfman, D.M., Ma, F.R., Sullivan, E.L., Munoz, O., Wood, C.R.,
Greenfield, E.A., and Freeman, G.J. 2003. Blockade of programmed death-1
ligands on dendritic cells enhances T cell activation and cytokine production. J
Immunol. 170:1257-66.
Bullock, T.N., Mullins, D.W., Colella, T.A., and Engelhard, V.H. 2001.
Manipulation of avidity to improve effectiveness of adoptively transferred CD8(+)
T cells for melanoma immunotherapy in human MHC class I-transgenic mice. J
Immunol. 167:5824-31.
138
Burstein, N.A. and Law, L.W. 1971. Neonatal thymectomy and non-viral mammary
tumours in mice. Nature. 231:450 −452.
Cai, G., Karni, A., Oliveira, E.M., Weiner, H.L., Hafler, D.A., and Freeman, G.J.
2004. PD-1 ligands, negative regulators for activation of naive, memory, and
recently activated human CD4+ T cells. Cell Immunol. 230:89-98.
Cebon, J., Jager, E., Shackleton, M.J., Gibbs, P., Davis, I.D., Hopkins, W., Gibbs,
S., Chen, Q., Karbach, J., Jackson, H., et al. 2003. Two phase I studies of low dose
recombinant human IL-12 with Melan-A and influenza peptides in subjects with
advanced malignant melanoma. Cancer Immun. 3:7.
Chan, T. O., S. E. Rittenhouse, and P. N. Tsichlis. 1999. AKT/PKB and other D3
phosphoinositide-regulated kinases: kinase activation by phosphoinositide-
dependent phosphorylation. Annu. Rev. Biochem. 68:965-1014.
Chemnitz, J.M., Parry, R.V., Nichols, K.E., June, C.H., and Riley, J.L. 2004. SHP-1
and SHP-2 associate with immunoreceptor tyrosine-based switch motif of
programmed death 1 upon primary human T cell stimulation, but only receptor
ligation prevents T cell activation. J Immunol. 173:945-54.
Chen, L., Linsley, P.S., and Hellstrom, K.E. 1993. Costimulation of T cells for
tumor immunity. Immunol Today. 14:483-6.
Chen, L. 2004. Co-inhibitory molecules of the B7-CD28 family in the control of T-
cell immunity. Nat Rev Immunol. 4:336-47.
Chen, Y.T., Gure, A.O., Tsang, S., Stockert, E., Jager, E., Knuth, A., and Old, L.J.
1998. Identification of multiple cancer/testis antigens by allogeneic antibody
screening of a melanoma cell line library. Proc Natl Acad Sci USA. 95:6919-23.
Chuang, E., Fisher, T.S., Morgan, R.W., Robbins, M.D., Duerr, J.M., Vander
Heiden, M.G., Gardner, J.P., Hambor, J.E., Neveu, M.J., and Thompson CB. 2000.
The CD28 and CTLA-4 receptors associate with the serine/threonine phosphatase
PP2A. Immunity. 13:313-22.
Coffer, P.J., Jin, J., and Woodgett, J.R. 1998. Protein kinase B (c-Akt): a
multifunctional mediator of phosphatidylinositol 3-kinase activation. Biochem J.
335:1-13.
139
Curiel, T.J., Wei, S., Dong, H., Alvarez, X., Cheng, P., Mottram, P., Krzysiek, R.,
Knutson, K.L., Daniel, B., Zimmermann, M.C., et al. 2003. Blockade of B7-H1
improves myeloid dendritic cell-mediated antitumor immunity. Nat Med. 9:562-7.
Davis, M.M., and Bjorkman, P.J. 1988. T-cell antigen receptor genes and T-cell
recognition. Nature. 334:395-402.
Day, C.L., Kaufmann, D.E., Kiepiela, P., Brown, J.A., Moodley, E.S., Reddy, S.,
Mackey, E.W., Miller, J.D., Leslie, A.J., DePierres, C., et al. 2006. PD-1 expression
on HIV-specific T cells is associated with T-cell exhaustion and disease
progression. Nature. 443:350-4.
de Melo, N., Matutes, E., Cordone, I., Morilla, R., and Catovksy, D. 1992.
Expression of Ki-67 nuclear antigen in B and T cell lymphoproliferative disorders.
J Clin Pathol. 45:660-3.
Dong, H., Strome, S.E., Salomao, D.R., Tamura, H., Hirano, F., Flies, D.B., Roche,
P.C., Lu, J., Zhu, G., Tamada, K., et al. 2002. Tumor-associated B7-H1 promotes
T-cell apoptosis: a potential mechanism of immune evasion. Nat Med. 8:793-800.
Dudley, M.E., Wunderlich, J.R., Yang, J.C., Sherry, R.M., Topalian, S.L., Restifo,
N.P., Royal, R.E., Kammula, U., White, D.E., Mavroukakis, S.A., et al. 2005.
Adoptive cell transfer therapy following non-myeloablative but lymphodepleting
chemotherapy for the treatment of patients with refractory metastatic melanoma. J
Clin Oncol. 23:2346-57.
dupont, N.C., Wang, K., Wadhwa, P.D., Culhane, J.F., and Nelson, E.L. 2005.
Validation and comparison of luminex multiplex cytokine analysis kits with
ELISA: determinations of a panel of nine cytokines in clinical sample culture
supernatants. J Reprod Immunol. 66(2):175-91.
Dutoit, V., Rubio-Godoy, V., Dietrich, P.Y., Quiqueres, A.L., Schnuriger, V.,
Rimoldi, D., Lienard, D., Speiser, D., Guillaume, P., Batard, P., et al. 2001.
Heterogeneous T-cell response to MAGE-A10(254-262): high avidity-specific
cytolytic T lymphocytes show superior antitumor activity. Cancer Res. 61:5850-6.
Endl, E., and Gerdes, J. 2000. The Ki-67 protein: fascinating forms and an unknown
function. Exp Cell Res. 257:231-7.
140
Fadok, V.A., Voelker, D.R., Campbell, P.A., Cohen, J.J., Bratton, D.L., and
Henson, P.M. 1992. Exposure of phosphatidylserine on the surface of apoptotic
lymphocytes triggers specific recognition and removal by macrophages. J
Immunol. 148:2207-16.
Flanagan, S.P. 1966. 'Nude', a new hairless gene with pleiotropic effects in the
mouse. Genet Res. 8:295 −309.
Fong, L., Hou, Y., Rivas, A., Benike, C., Yuen, A., Fisher, G.A., Davis, M.M., and
Engleman, E.G. 2001. Altered peptide ligand vaccination with Flt3 ligand expanded
dendritic cells for tumor immunotherapy. Proc Natl Acad Sci USA. 98:8809-14.
Fugmann ,S.D., Lee, A.I., Shockett, P.E., Villey, I.J., and Schatz, D.G. 2000. The
RAG proteins and V(D)J recombination: complexes, ends, and transposition. Annu
Rev Immunol. 18:495-527.
Gajewski, T.F., Fallarino, F., Ashikari, A., and Sherman, M. 2001. Immunization of
HLA-A2+ melanoma patients with MAGE-3 or MelanA peptide-pulsed autologous
peripheral blood mononuclear cells plus recombinant human interleukin 12. Clin.
Cancer Res. 7:895s-901s.
Giuntoli, II, R.L., Lu, J., Kobayashi, H., Kennedy, R., and Celis, E. 2002. Direct
costimulation of tumor-reactive CTL by helper T cells potentiate their proliferation,
survival, and effector function. Clin. Cancer. Res. 8:922-31
Grant, G.A. and Miller, J.F. 1965. Effect of neonatal thymectomy on the induction
of sarcomata in C57BL mice. Nature. 205:1124 −1125
Greenwald, R.J., Freeman, G.J., and Sharpe, A.H. 2005. The B7 family revisited.
Annu Rev Immunol. 23:515-48.
Hart, DN. 1997. Dendritic cells: unique leukocyte populations which control the
primary immune response. Blood. 90:3245-87.
Hirano, F., Kaneko, K., Tamura, H., Dong, H., Wang, S., Ichikawa, M., Rietz, C.,
Flies, D.B., Lau, J.S., et al. 2005. Blockade of B7-H1 and PD-1 by monoclonal
antibodies potentiates cancer therapeutic immunity. Cancer Res. 65:1089-96.
Hodge, J.W., Chakraborty, M., Kudo-Saito, C., Garnett, C.T., and Schlom, J. 2005.
Multiple costimulatory modalities enhance CTL avidity. J Immunol. 174:5994-
6004.
141
Huang, J., Khong, H.T., Dudley, M.E., El-Gamil, M., Li, Y.F., Rosenberg, S.A.,
and Robbins, P.F. 2005. Survival, persistence, and progressive differentiation of
adoptively transferred tumor-reactive T cells associated with tumor regression. J
Immunother. 28:258-67.
Hung, K., Hayashi, R., Lafond-Walker, A., Lowenstein, C., Pardoll, D., and
Levitsky, H. 1998. The central role of CD4
+
T cells in the antitumor immune
response. J Exp Med. 188:2357-68.
Hurtado, J.C., Kim, S.H., Pollok, K.E., Lee, Z.H., and Kwon, B.S. 1995. Potential
role of 4-1BB in T cell activation. Comparison with the costimulatory molecule
CD28. J Immunol. 155:3360-7.
Ishida, Y., Agata, Y., Shibahara, K., and Honjo, T. 1992. Induced expression of PD-
1, a novel member of the immunoglobulin gene superfamily, upon programmed cell
death. EMBO J. 11:3887-95.
Iwai, Y., Ishida, M., Tanaka, Y., Okazaki, T., Honjo, T., and Minato, N. 2002.
Involvement of PD-L1 on tumor cells in the escape from host immune system and
tumor immunotherapy by PD-L1 blockade. Proc Natl Acad Sci USA. 99:12293-7.
Iwai, Y., Terawaki, S., and Honjo, T. 2005. PD-1 blockade inhibits hematogenous
spread of poorly immunogenic tumor cells by enhanced recruitment of effector T
cells. Int Immunol. 17:133-44.
Janssen, E.M., Droin, N.M., Lemmens, E.E., Pinkoski, M.J., Bensinger, S.J., Ehst,
B.D., Griffith, T.S., Green, D.R., and Schoenberger SP. 2005. CD4+ T-cell help
controls CD8+ T-cell memory via TRAIL-mediated activation-induced cell death.
Nature. 434:88-93.
Jensen, P., Hansen, S., Moller, B., Leivestad, T., Pfeffer, P., Geiran, O., Fauchald,
P., and Simonsen, S. 1999. Skin cancer in kidney and heart transplant recipients and
different long-term immunosuppressive therapy regimens. J Am Acad Dermatol.
40:177-86.
Kaplan, D.H., Shankaran, V., Dighe, A.S., Stockert, E., Aguet, M., Old, L.J., and
Schreiber RD. 1998. Demonstration of an interferon gamma-dependent tumor
surveillance system in immunocompetent mice. Proc Natl Acad Sci USA. 95:7556-
61.
142
Kasahara, S., Ando, K., Saito, K., Sekikawa, K., Ito, H., Ishikawa, T., Ohnishi, H.,
Seishima, M., Kakumu, S., and Moriwaki, H. 2003. Lack of tumor necrosis factor
alpha induces impaired proliferation of hepatitis B virus-specific cytotoxic T
lymphocytes. J Virol. 77:2469-76.
Kawakami, Y., Eliyahu, S., Delgado, C.H., Robbins, P.F., Sakaguchi, K., Appella,
E., Yannelli, J.R., Adema, G.J. Miki, T., and Rosenberg, S.A. 1994. Identification
of a human melanoma antigen recognized by tumor-infiltrating lymphocytes
associated with in vivo tumor rejection. Proc. Natl. Acad. Sci. USA. 91: 6458-62.
Kawakami, Y., Eliyahu, S., Sakaguchi, K., Robbins, P.F., Rivoltini, L., Yannelli,
J.R., Appella, E., and Rosenberg, S.A. 1994. Identification of the immunodominant
peptides of the MART-1 human melanoma antigen recognized by the majority of
HLA-A2-restricted tumor infiltrating lymphocytes. J Exp Med. 180:347-52.
Kawakami, Y., Eliyahu, S., Jennings, C., Sakaguchi, K., Kang, X., Southwood, S.,
Robbins, P.F., Sette, A., Appella, E., and Rosenberg, SA. 1995. Recognition of
multiple epitopes in the human melanoma antigen gp100 by tumor-infiltrating T
lymphocytes associated with in vivo tumor regression. J Immunol. 154:3961-8.
Kawakami, Y., Robbins, P.F., and Rosenberg, S.A. 1996. Human melanoma
antigens recognized by T lymphocytes. Keio J Med. 45:100-8.
Keir, M.E., Latchman, Y.E., Freeman, G.J., and Sharpe, A.H. 2005. Programmed
death-1 (PD-1):PD-ligand 1 interactions inhibit TCR-mediated positive selection of
thymocytes. J Immunol. 175:7372-9.
Khattri, R., Auger, J.A., Griffin, M.D., Sharpe, A.H., and Bluestone, J.A. 1999.
Lymphoproliferative disorder in CTLA-4 knockout mice is characterized by CD28-
regulated activation of Th2 responses. J Immunol. 162:5784-91.
Khong, H.T., and Rosenberg, S.A. 2002. Pre-existing immunity to tyrosinase-
related protein (TRP)-2, a new TRP-2 isoform, and the NY-ESO-1 melanoma
antigen in a patient with a dramatic response to immunotherapy. J Immunol.
168:951-6.
Knutson, K.L., and Disis, M.L. 2005. Tumor antigen-specific T helper cells in
cancer immunity and immunotherapy. Cancer Immunol Immunother. 54:721-8.
143
Kohrt, H.E., Shu, C.T., Stuge, T.B., Holmes, S.P., Weber, J., and Lee, P.P. 2005.
Rapid assessment of recognition efficiency and functional capacity of antigen-
specific T-cell responses. J Immunother. 28:297-305.
Krajewska, M., Wang, H.G., Krajewski, S., Zapata, J.M., Shabaik, A., Gascoyne,
R., and Reed, J.C. 1997. Immunohistochemical analysis of in vivo patterns of
expression of CPP32 (Caspase-3), a cell death protease. Cancer Res. 57:1605-13.
Kroner, A., Mehling, M., Hemmer, B., Rieckmann, P., Toyka, K.V., Maurer, M.,
and Wiendl, H. 2005. A PD-1 polymorphism is associated with disease progression
in multiple sclerosis. Ann Neurol. 58:50-7.
Kudo-Saito, C., Hodge, J.W., Kwak, H., Kim-Schulze, S., Schlom, J., and
Kaufman, H.L. 2006. 4-1BB ligand enhances tumor-specific immunity of poxvirus
vaccines. Vaccine. 24:4975-86.
Lan, M.S., Batra, S.K., Qi, W.N., Metzgar, R.S., and Hollingsworth, M.A. 1990.
Cloning and sequencing of a human pancreatic tumor mucin cDNA. J Biol Chem.
265:15294-9.
Latchman, Y.E., Liang, S.C., Wu, Y., Chernova, T., Sobel, R.A., Klemm, M.,
Kuchroo, V.K., Freeman, G.J., and Sharpe, A.H. 2004. PD-L1-deficient mice show
that PD-L1 on T cells, antigen-presenting cells, and host tissues negatively regulates
T cells. Proc Natl Acad Sci U S A. 101:10691-6.
Lau, R., Wang, F., Jeffery, G., Marty, V., Kuniyoshi, J., Bade, E., Ryback, M.E.,
and Weber, J. 2001. Phase I trial of intravenous peptide-pulsed dendritic cells in
patients with metastatic melanoma. J Immunother. 24:66-78.
Lee, P., Wang, F., Kuniyoshi, J., Rubio, V., Stuges, T., Groshen, S., Gee, C., Lau,
R., Jeffery, G., Margolin, K., et al. 2001. Effects of interleukin-12 on the immune
response to a multipeptide vaccine for resected metastatic melanoma. J Clin Oncol.
19:3836-47.
Lee, S.W., Park, Y., Yoo, J.K., Choi, S.Y., and Sung YC. 2003. Inhibition of TCR-
induced CD8 T cell death by IL-12: regulation of Fas ligand and cellular FLIP
expression and caspase activation by IL-12. J. Immunol. 170:2456-60.
Li, J.H., Rosen, D., Sondel, P., and Berke, G. 2002. Immune privilege and FasL:
two ways to inactivate effector cytotoxic T lymphocytes by FasL-expressing cells.
Immunology. 105:267-77.
144
Li, M.O., Wan, Y.Y., Sanjabi, S., Robertson, A.K., and Flavell, R.A. 2006.
Transforming growth factor-beta regulation of immune responses. Annu Rev
Immunol. 24:99-146.
Liang, S.C., Latchman, Y.E., Buhlmann, J.E., Tomczak, M.F., Horwitz, B.H.,
Freeman, G.J., and Sharpe, A.H. 2003. Regulation of PD-1, PD-L1, and PD-L2
expression during normal and autoimmune responses. Eur J Immunol. 33:2706-16.
Lichtenheld, M.G., Olsen, K.J., Lu, P., Lowrey, D.M., Hameed, A., Hengartner, H.,
and Podack, E.R. 1988. Structure and function of human perforin. Nature.
335:448-51.
Mailliard, R. B., S. Egawa, Q. Cai, A. Kalinska, S. N. Bykovskaya, M. T. Lotze, M.
L. Kapsenberg, W. J. Storkus, and P. Kalinski. 2002. Complementary dendritic cell-
activating function of CD8+ and CD4+ T cells: helper role of CD8+ T cells in the
development of T helper type 1 responses. J Exp Med. 195:473-83.
Mills, K.H., and McGuirk, P. 2004. Antigen-specific regulatory T cells--their
induction and role in infection. Semin Immunol. 16:107-17.
Mocellin, S., Rossi, C.R., Nitti, D., Lise, M., and Marincola, F.M. 2003. Dissecting
tumor responsiveness to immunotherapy: the experience of peptide-based
melanoma vaccines. Biochim Biophys Acta. 1653:61-71.
Molldrem, J.J., Lee, P.P., Kant, S., Wieder, E., Jiang, W., Lu, S., Wang, C., and
Davis, M.M. 2003. Chronic myelogenous leukemia shapes host immunity by
selective deletion of high-avidity leukemia-specific T cells. J Clin Invest. 111:639-
47.
Moloney, F.J., Comber, H., O'Lorcain, P., O'Kelly, P., Conlon, P.J., and Murphy,
G.M. 2006. A population-based study of skin cancer incidence and prevalence in
renal transplant recipients. Br J Dermatol. 154:498-504.
Morse, M.A., Chui, S., Hobeika, A., Lyerly, H.K., and Clay, T. 2005. Recent
developments in therapeutic cancer vaccines. Nat Clin Pract Oncol. 2:108-13.
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.
7:145-73.
145
Murata, S., Ladle, B.H., Kim, P.S., Lutz, E.R., Wolpoe, M.E., Ivie, S.E., Smith,
H.M., Armstrong, T.D., Emens, L.A., Jaffee, E.M., et al. 2006. OX40 costimulation
synergizes with GM-CSF whole-cell vaccination to overcome established CD8+ T
cell tolerance to an endogenous tumor antigen. J Immunol. 176:974-83.
Naldi, L., Fortina, A.B., Lovati, S., Barba, A., Gotti, E., Tessari, G., Schena, D.,
Diociaiuti, A., Nanni, G., La Parola, I.L., et al. 2000. Risk of nonmelanoma skin
cancer in Italian organ transplant recipients. A registry-based study.
Transplantation. 70:1479-84.
Nishimura, H., Nose, M., Hiai, H., Minato, N., and Honjo T. 1999. Development of
lupus-like autoimmune diseases by disruption of the PD-1 gene encoding an ITIM
motif-carrying immunoreceptor. Immunity. 11:141-51.
Nishimura, H., and Honjo, T. 2001. PD-1: an inhibitory immunoreceptor involved
in peripheral tolerance. Trends Immunol. 22:265-8.
Nishimura, H., Okazaki, T., Tanaka, Y., Nakatani, K., Hara, M., Matsumori, A.,
Sasayama, S., Mizoguchi, A., Hiai, H., Minato, N., et al. 2001. Autoimmune dilated
cardiomyopathy in PD-1 receptor-deficient mice. Science. 291:319-22.
Nishizuka, Y., Nakakuki, K. and Usui, M. 1965. Enhancing effect of thymectomy
on hepatotumorigenesis in Swiss mice following neonatal injection of 20-
methylcholanthrene. Nature. 205:1236 −1238.
O'Connor, D.H., Allen, T.M., Vogel, T.U., Jing, P., DeSouza, I.P., Dodds, E.,
Dunphy, E.J., Melsaether, C., Mothe, B., Yamamoto, H., et al. 2002. Acute phase
cytotoxic T lymphocyte escape is a hallmark of simian immunodeficiency virus
infection. Nat Med. 8:493-9.
Oh, S., Hodge, J.W., Ahlers, J.D., Burke, D.S., Schlom, J., and Berzofsky, J.A.
2003. Selective induction of high avidity CTL by altering the balance of signals
from APC. J Immunol. 170:2523-30.
Op den Kamp, J,A. 1979. Lipid asymmetry in membranes. Annu Rev Biochem.
48:47-71.
Palmer, E.M., Farrokh-Siar, L., Maguire van Seventer, J., and van Seventer, G.A.
2001. IL-12 decreases activation-induced cell death in human naive Th cells
costimulated by intercellular adhesion molecule-1. I. IL-12 alters caspase
processing and inhibits enzyme function. J Immunol. 167:749-58.
146
Pantelouris, E.M. 1968. Absence of thymus in a mouse mutant. Nature.
217:370 −371.
Parish, CR. 1999. Fluorescent dyes for lymphocyte migration and proliferation
studies. Immunol Cell Biol. 77:499-508.
Parry, R.V., Chemnitz, J.M., Frauwirth, K.A., Lanfranco, A.R., Braunstein, I.,
Kobayashi, S.V., Linsley, P.S., Thompson, C.B., and Riley, J.L. 2005. CTLA-4 and
PD-1 receptors inhibit T-cell activation by distinct mechanisms. Mol Cell Biol.
25:9543-53.
Paul, W.E., and Seder, R.A. 1994. Lymphocyte responses and cytokines. Cell.
76:241-51.
Peggs, K.S., and Allison, J.P. 2005. Co-stimulatory pathways in lymphocyte
regulation: the immunoglobulin superfamily. Br J Haematol. 130:809-24.
Peitsch, M.C., and Tschopp J. 1991. Assembly of macromolecular pores by immune
defense systems. Curr Opin Cell Biol. 3:710-6.
Peterson, A.C., Harlin, H., and Gajewski, T.F. 2003. Immunization with Melan-A
peptide-pulsed peripheral blood mononuclear cells plus recombinant human
interleukin-12 induces clinical activity and T-cell responses in advanced melanoma.
J. Clin. Oncol. 21:2342-8
Phan, G.Q., Yang, J.C., Sherry, R.M., Hwu, P., Topalian, S.L., Schwartzentruber,
D.J., Restifo, N.P., Haworth, L.R., Seipp, C.A., Freezer, L.J., et al. 2003. Cancer
regression and autoimmunity induced by cytotoxic T lymphocyte-associated
antigen 4 blockade in patients with metastatic melanoma. Proc Natl Acad Sci USA.
100: 8372-7.
Prokunina, L., Castillejo-Lopez, C., Oberg, F., Gunnarsson, I., Berg, L.,
Magnusson, V., Brookes, A.J., Tentler, D., Kristjansdottir, H., Grondal, G., et al.
2002. A regulatory polymorphism in PDCD1 is associated with susceptibility to
systemic lupus erythematosus in humans. Nat Genet. 32:666-9.
Regnault, A., Lankar, D., Lacabanne, V., Rodriguez, A., Thery, C., Rescigno, M.,
Saito, T., Verbeek, S., Bonnerot, C., Ricciardi-Castagnoli, P., et al. 1999. Fcgamma
receptor-mediated induction of dendritic cell maturation and major
histocompatibility complex class I-restricted antigen presentation after immune
complex internalization. J Exp Med. 189:371-80.
147
Renkvist, N., Castelli, C., Robbins, P.F., and Parmiani, G. 2001. A listing of human
tumor antigens recognized by T cells. Cancer Immunol Immunother. 50:3-15.
Rodrigues, L.K., Klencke, B.J., Vin-Christian, K., Berger, T.G., Crawford, R.I.,
Miller, J.R. 3rd, Ferreira, C.M., Nosrati, M., and Kashani-Sabet, M. 2002. Altered
clinical course of malignant melanoma in HIV-positive patients. Arch Dermatol.
138:765-70.
Rogers, P.R., Song, J., Gramaglia, I., Killeen, N., and Croft, M. 2001. OX40
promotes Bcl-xL and Bcl-2 expression and is essential for long-term survival of
CD4 T cells. Immunity. 15:445-55.
Rosenberg, S.A., Yang, J.C., and Restifo, N.P. 2004. Cancer immunotherapy:
moving beyond current vaccines. Nat Med. 10:909-15.
Rubio, V., Stuge, T.B., Singh, N., Betts, M.R., Weber, J.S., Roederer, M., and Lee,
P.P. 2003. Ex vivo identification, isolation and analysis of tumor-cytolytic T cells.
9:1377-82.
Ruppert, J., Sidney, J., Celis, E., Kubo, R.T., Grey, H.M., and Sette, A. 1993.
Prominent role of secondary anchor residues in peptide binding to HLA-A2.1
molecules. Cell. 74:929-37.
Sadovnikova, E., and Stauss, H.J. 1996. Peptide-specific cytotoxic T lymphocytes
restricted by nonself major histocompatibility complex class I molecules: reagents
for tumor immunotherapy. Proc Natl Acad Sci USA. 93:13114-8.
Sakaguchi, S. 2004. Naturally arising CD4+ regulatory t cells for immunologic self-
tolerance and negative control of immune responses. Annu Rev Immunol. 22:531-
62.
Sakaguchi, S. 2005. Naturally arising Foxp3-expressing CD25+CD4+ regulatory T
cells in immunological tolerance to self and non-self. Nat Immunol. 6:345-52.
Salama, A.D., Chitnis, T., Imitola, J., Ansari, M.J., Akiba, H., Tushima, F., Azuma,
M., Yagita, H., Sayegh, M.H., and Khoury, S.J. 2003. Critical role of the
programmed death-1 (PD-1) pathway in regulation of experimental autoimmune
encephalomyelitis. J Exp Med. 198:71-8.
Salkowitz, J.R., Sieg, S.F., Harding, C.V., and Lederman, M.M. 2004. In vitro
human memory CD8 T cell expansion in response to cytomegalovirus requires
CD4+ T cell help. J Infect Dis. 189:971-83.
148
Sallusto, F., Cella, M., Danieli, C., and Lanzavecchia, A. 1995. Dendritic cells use
macropinocytosis and the mannose receptor to concentrate macromolecules in the
major histocompatibility complex class II compartment: downregulation by
cytokines and bacterial products. J Exp Med. 182:389-400.
Salwinski, L., and Eisenberg, D. 2003. Computational methods of analysis of
protein-protein interactions. Curr Opin Struct Biol. 13:377-82.
Samelson, L.E., Donovan, J.A., Isakov, N., Ota, Y., and Wange, R.L. 1995. Signal
transduction mediated by the T-cell antigen receptor. Ann N Y Acad Sci. 766:157-
72.
Sanderson, K., Scotland, R., Lee, P., Liu, D., Groshen, S., Snively, J., Sian, S.,
Nichol, G., Davis, T., Keler, T., et al. 2005. Autoimmunity in a phase I trial of a
fully human anti-cytotoxic T-lymphocyte antigen-4 monoclonal antibody with
multiple melanoma peptides and Montanide ISA 51 for patients with resected stages
III and IV melanoma. J Clin Oncol. 23:741-50.
Sanford, B.H., Kohn, H.I., Daly, J.J. and Soo, S.F. 1973. Long-term spontaneous
tumor incidence in neonatally thymectomized mice. J Immunol. 110:1437 −1439.
Schmid, I., Krall, W.J., Uittenbogaart, C.H., Braun, J., and Giorgi, J.V. 1992. Dead
cell discrimination with 7-amino-actinomycin D in combination with dual color
immunofluorescence in single laser flow cytometry. Cytometry. 13:204-8.
Schoenberger, S.P., Toes, R.E., van der Voort, E.I., Offringa, and R., Melief, C.J.
1998. Help for cytotoxic T-cell responses is mediated by CD40-CD40L
interactions. Nature. 393:480-3.
Seo, S.K., Seo, H.M., Jeong, H.Y., Choi, I.W., Park, Y.M., Yagita, H., Chen, L.,
and Choi, I.H. 2006. Co-inhibitory role of T-cell-associated B7-H1 and B7-DC in
the T-cell immune response. Immunol Lett. 102:222-8.
Shankaran, V., Ikeda, H., Bruce, A.T., White, J.M., Swanson, P.E., Old, L.J., and
Schreiber RD. 2001. IFNgamma and lymphocytes prevent primary tumour
development and shape tumour immunogenicity. Nature. 410:1107-11.
Shedlock, D.J., and Shen, H. 2003. Requirement for CD4 T cell help in generating
functional CD8 T cell memory. Science. 300:337-9.
Shortman, K., and Liu, Y.J. 2002. Mouse and human dendritic cell subtypes. Nat
Rev Immunol. 2:151-61.
149
Stuge, T.B., Holmes, S.P., Saharan, S., Tuettenberg, A., Roederer, M., Weber, J.S.,
and Lee, P.P. 2004. Diversity and recognition efficiency of T cell responses to
cancer. PLoS Med. 1:e28.
Taga, K., and Tosato, G. 1992. IL-10 inhibits human T cell proliferation and IL-2
production. J Immunol. 148:1143-8.
Tan, J.T., Whitmire, J.K., Murali-Krishna, K., Ahmed, R., Altman, J.D., Mittler,
R.S., Sette, A., Pearson, T.C., and Larsen, C.P. 2000. 4-1BB costimulation is
required for protective anti-viral immunity after peptide vaccination. J Immunol.
164:2320-5.
Tatsumi, T., Kierstead, L.S., Ranieri, E., Gesualdo, L., Schena, F.P., Finke, J.H.,
Bukowski, R.M., Mueller-Berghaus, J., Kirkwood, J.M., Kwok, W.W., et al. 2002.
Disease-associated bias in T helper type 1 (Th1)/Th2 CD4(+) T cell responses
against MAGE-6 in HLA-DRB10401(+) patients with renal cell carcinoma or
melanoma. J Exp Med. 196:619-28.
Thornberry, N.A., and Lazebnik, Y. 1998. Caspases: enemies within. Science.
281:1312-6.
Topalian, S.L., Solomon, D., and Rosenberg, S.A. 1989. Tumor-specific cytolysis
by lymphocytes infiltrating human melanomas. J Immunol. 142:3714-25.
Tonegawa, S. 1983. Somatic generation of antibody diversity. Nature. 302:575-81.
Trainin, N., Linker-Israeli, M., Small, M. and Boiato-Chen, L. 1967. Enhancement
of lung adenoma formation by neonatal thymectomy in mice treated with 7,12-
dimethylbenz(a)anthracene or urethan. Int. J. Cancer 2:326 −336.
Trapani, J.A., and Smyth, M.J. 2002. Functional significance of the
perforin/granzyme cell death pathway. Nat Rev Immunol. 2:735-47.
Stutman, O. 1975. Immunodepression and malignancy. Adv Cancer Res. 22:261-
422.
Valmori, D., Fonteneau, J.F., Lizana, C.M., Gervois, N., Lienard, D., Rimoldi, D.,
Jongeneel, V., Jotereau, F., Cerottini, J.C., and Romero, P. 1998. Enhanced
generation of specific tumor-reactive CTL in vitro by selected Melan-A/MART-1
immunodominant peptide analogues. J Immunol. 160:1750-8.
150
van den Broek, M.E., Kagi, D., Ossendorp, F., Toes, R., Vamvakas, S., Lutz, W.K.,
Melief, C.J., Zinkernagel, R.M., and Hengartner, H. 1996. Decreased tumor
surveillance in perforin-deficient mice. J Exp Med. 184:1781-90.
van der Bruggen, P., Traversari, C., Chomez, P., Lurquin, C., De, Plaen E., Van den
Eynde, B., Knuth, A., and Boon, T. 1991. A gene encoding an antigen recognized
by cytolytic T lymphocytes on a human melanoma. Science. 254:1643-7.
Vigouroux, S., Yvon, E., Biagi, E., and Brenner, M.K. 2004. Antigen-induced
regulatory T cells. Blood. 104:26-33.
Walker, E.H., Pacold, M.E., Perisic, O., Stephens, L., Hawkins, P.T., Wymann,
M.P., and Williams, R.L. 2000. Structural determinants of phosphoinositide 3-
kinase inhibition by wortmannin, LY294002, quercetin, myricetin, and
staurosporine. Mol Cell. 6:909-19.
Wei, M.L, and Cresswell P. 1992. HLA-A2 molecules in an antigen-processing
mutant cell contain signal sequence-derived peptides. Nature. 356:443-6.
Wilson, N.S., El-Sukkari, D., Belz, G.T., Smith, C.M., Steptoe, R.J., Heath, W.R.,
Shortman, K., and Villadangos, J.A. 2003. Most lymphoid organ dendritic cell
types are phenotypically and functionally immature. Blood. 102:2187-94.
Wong, R., Lau, R., Chang, J., Kuus-Reichel, T., Brichard, V., Bruck, C., and Weber
J. 2004. Immune responses to a class II helper peptide epitope in patients with stage
III/IV resected melanoma. Clin Cancer Res. 10:5004-13.
Yang, S., Linette, G.P., Longerich, S., and Haluska, F.G. 2002. Antimelanoma
activity of CTL generated from peripheral blood mononuclear cells after
stimulation with autologous dendritic cells pulsed with melanoma gp100 peptide
G209-2M is correlated to TCR avidity. J Immunol. 169:531-9.
Yee, C., Thompson, J.A., Byrd, D., Riddell, S.R., Roche, P., Celis, E., and
Greenberg, P.D. 2002. Adoptive T cell therapy using antigen-specific CD8+ T cell
clones for the treatment of patients with metastatic melanoma: in vivo persistence,
migration, and antitumor effect of transferred T cells. Proc Natl Acad Sci USA.
99:16168-73.
Zeh, H.J. 3rd, Perry-Lalley, D., Dudley, M.E., Rosenberg, S.A., and Yang, J.C.
1999. High avidity CTLs for two self-antigens demonstrate superior in vitro and in
vivo antitumor efficacy. J Immunol. 162:989-94.
151
Zhou, J., Dudley, M.E., and Rosenberg, S.A., and Robbins, P.F. 2005. Persistence
of multiple tumor-specific T-cell clones is associated with complete tumor
regression in a melanoma patient receiving adoptive cell transfer therapy. J
Immunother. 28:53-62.
152
Abstract (if available)
Abstract
Negative costimulatory signaling mediated via cell surface Programmed Death-1 (PD-1) expression modulates T and B cell activation and is critical for maintaining peripheral tolerance. Abrogation of the PD-1 pathway may therefore be useful strategy to enhance the induction of antigen-specific T cells by vaccination. In these studies, we examined the effects of a fully human PD-1-abrogating antibody on the in vitro expansion and functional profile of human CD8+ T cells (CTLs) specific for the melanoma-associated antigens gp100 and MART-1. PD-1 blockade during peptide stimulation augmented the absolute numbers of CD3+, CD4+, CD8+, and MHC:peptide tetramer-binding CTLs. This correlated with increased frequencies of IFN-gamma-secreting antigen-specific cells and augmented lysis of peptide-pulsed targets as well as gp100+/MART+ melanoma cell lines. PD-1 blockade also increased the fraction of antigen-specific CTLs that recognized melanoma targets by degranulation, suggesting increased recognition efficiency for cognate peptide. The increased frequencies and absolute numbers of antigen-specific CTLs by PD-1 blockade resulted primarily from augmented proliferation, and not decreased apoptosis. Kinetic analysis of cytokine secretion demonstrated that PD-1 blockade increased both Type-1 and Type-2 cytokine secretion, without any apparent skewing of the cytokine repertoire. PD-1 blockade did not significantly alter the phenotype of peptide-stimulated gp100- and MART-1-specific CTLs, which were predominantly activated effector / effector memory cells characterized by a CD45RA(low), CD45RO(high), CCR7(low), CD62L(low), and CD44(high), expression profile. Vaccine-induced gp100- and MART-1-specific memory CTLs were also found to be dependent on CD4+ T help for optimal expansion during in vitro peptide stimulation. These findings suggest that PD-1 blockade may augment CTL expansion directly and/or indirectly through the augmented provision of CD4+ T help.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Self-secretion of checkpoint blockade enhances antitumor immunity by murine chimeric antigen receptor-engineered T cells
PDF
Novel design and combinatory therapy to enhance chimeric antigen receptor engineered T cells (CAR-T) efficacy against solid tumor
PDF
Tri-specific T cell engager immunotherapy targeting tumor initiating cells
PDF
PD-L1-GM-CSF fusion protein-loaded DC vaccination activates PDL1-specific humoral and cellular immune responses
PDF
Engineering chimeric antigen receptor (CAR) -modified T cells for enhanced cancer immunotherapy
PDF
Tri-specific T cell engager immunotherapy targeting tumor initiating cells
PDF
Regulation of T cell HLA-DR by CD3 ζ signaling
PDF
Novel approaches of mobilizing human iNKT cells for cancer immunotherapies
PDF
Characterization of invariant natural killer T cells in a novel humanized HBV-transgenic model
PDF
Lym-1 epitope targeted chimeric antigen receptor (CAR) T cells for the immunotherapy of cancer
PDF
Mechanistic model of chimeric antigen receptor T cell activation
Asset Metadata
Creator
Wong, Raymond M.
(author)
Core Title
Modulation of human tumor antigen-specific T cell responses by programmed death-1 blockade
School
Keck School of Medicine
Degree
Doctor of Philosophy
Degree Program
Molecular Microbiology
Degree Conferral Date
2006-12
Publication Date
11/15/2006
Defense Date
09/27/2006
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
costimulation,OAI-PMH Harvest,PD-1,T cells,tumor antigen
Language
English
Advisor
Weber, Jeffrey (
committee chair
), Epstein, Alan L. (
committee member
), Kast, W. Martin (
committee member
), Tahara, Stanley M. (
committee member
)
Creator Email
raymondw@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-m148
Unique identifier
UC1443696
Identifier
etd-Wong-20061115 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-28256 (legacy record id),usctheses-m148 (legacy record id)
Legacy Identifier
etd-Wong-20061115.pdf
Dmrecord
28256
Document Type
Dissertation
Rights
Wong, Raymond M.
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Repository Name
Libraries, University of Southern California
Repository Location
Los Angeles, California
Repository Email
cisadmin@lib.usc.edu
Tags
costimulation
PD-1
T cells
tumor antigen