University of Southern California Dissertations and Theses

Novel approaches of mobilizing human iNKT cells for cancer immunotherapies

(USC Thesis Other)

Novel approaches of mobilizing human iNKT cells for cancer immunotherapies

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content -1-
Novel Approaches of Mobilizing Human iNKT Cells for
Cancer Immunotherapies

By

Xin Li

A Thesis Presented to the Faculty of

THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA

In Partial Fulfillment of the Requirements for the Degree
of
MASTER OF SCIENCE
in
Molecular Microbiology and Immunology

August 2019

-2-
TABLE OF CONTENTS

ABBREVIATIONS ....................................................................................................................... 4
LIST OF TABLES AND FIGURES .............................................................................................. 6
SUMMARY .................................................................................................................................. 7
CHAPTER 1: INTRODUCTION .................................................................................................. 9
1.1 iNKT Cells ......................................................................................................................... 9
1.2 CD1d and α-GalCer ........................................................................................................ 11
1.3 iNKT Cells in Cancer Immunotherapy ............................................................................ 15
1.4 Setbacks of α-GalCer in Clinical Trials ........................................................................... 18
1.5 Agonistic MAbs in Cancer Immunotherapy .................................................................... 19
CHAPTER 2: MATERIALS AND METHODS ............................................................................ 20
2.1 Animals ........................................................................................................................... 20
2.2 Cell Lines ........................................................................................................................ 20
2.3 Antibodies ....................................................................................................................... 20
2.4 Glycolipids ....................................................................................................................... 21
2.5 Isolation of Progenitor Cells from Mouse Bone Marrow ................................................ 21
2.6 Preparation of BMDCs .................................................................................................... 22
2.7 Verification of BMDCs Maturation .................................................................................. 22
2.8 In vivo B16F10 Metastasis Assay .................................................................................. 23
2.9 Statistical Analysis .......................................................................................................... 23
CHAPTER 3: RESULTS ........................................................................................................... 25
3.1 In the current setting, NU-α-GalCer shows comparable anti-tumor efficacy compared to
α-GalCer ................................................................................................................................ 25
3.2 Anti-iTCR antibody NKTT320 has synergistic anti-tumor activities with mIL-12 ........... 32
3.3 Anti-CD1d Antibody 51.1.3 is a potent agonist of hCD1d .............................................. 37
-3-
CHAPTER 4: DISCUSSION ..................................................................................................... 43
ACKNOWLEDGMENTS ........................................................................................................... 46
REFERENCES ......................................................................................................................... 47

-4-
ABBREVIATIONS

AA
Amino Acid
ADCC
Antibody-Dependent Cellular Cytotoxicity
ADCP
Antibody, Dependent Cellular Phagocytosis
APC
Antigen-Presenting Cell
B6
C57bl/6
Bcl-2
B-Cell Lymphoma 2
BMDC
Bone Marrow-Derived Dc
CD
Cluster of Differentiation
CD1d
Cluster of Differentiation 1 (Class D)
CDC
Complement-Dependent Cytotoxicity
CTLA-4
Cytotoxic T-Lymphocyte-Associated Protein 4
DC
Dendritic Cells
DMEM
Dulbecco’s Modified Eagle’s Medium
FACS
Fluorescence-Activated Cell Sorting
FAK
Focal Adhesion Kinase
GM-CSF
Granulocyte-Macrophage Colony-Stimulating Factor
hCD1d
Human Cluster Of Differentiation 1 (Class D)
HSV-1
Herpes Simplex Virus Type 1
i.p.
Intraperitoneal
i.v.
Intravenous
IFN
Interferon
IgE
Immunoglobulin E
IgG
Immunoglobulin G
IL
Interleukin
IL-12R
Interleukin 12 Receptor
ILCs
Innate Lymphoid Cells
iNKT
Invariant Natural Killer T
iTCR
Invariant T-Cell Receptor
Jak-2
Janus Kinase 2
KI
Knock-In
KO
Knock-Out
LPS
Lipopolysaccharide
mAb
Monoclonal Antibody
MAIT
Mucosal-Associated Invariant T
MAPK
Mitogen-Activated Protein Kinase
MHC
Major Histocompatibility Complex
mIL-12
Mouse Interleukin 12
MR1
MHC Class-I Related Protein
-5-
NK
Natural Killer
NKT
Natural Killer T
Nu-α-GalCer
Naphthylurea–Alpha-Galactosylceramide
PBS
Phosphate Buffered Saline
PD-1
Programmed Cell Death-1
PI3K
Phosphoinositide 3-Kinase
RAG1
Recombination-Activating Gene 1
RAG2
Recombination-Activating Gene 2
RBC
Red Blood Cell
RPMI
Roswell Park Memorial Institute
STAT4
Signal Transducer And Activator Of Transcription 4
T cell
Thymus Cell
TCR
T-Cell Receptor
Tg
Transgenic
Th-1
Type 1 T Helper
Th-2
Type 2 T Helper
Thr
Threonine
TLR
Toll-Like Receptor
TNF
Tumor Necrosis Factor
TRAIL
TNF-Related Apoptosis-Inducing Ligand
α-GalCer
Alpha-Galactosylceramide
β2M
Beta 2-Microglobulin

-6-
LIST OF TABLES AND FIGURES

Table 1. Comparison among CD1d molecules, MHC class I molecules and MHC class II
molecules.
Figure 1. Structure of CD1d/iNKT cells systems and α-GalCer.
Figure 2. The anti-tumor activities of iNKT cells.
Figure 3. The general B16F10 melanoma challenge workflow in this study.
Figure 4. Preparation of BMDCs to deliver glycolipids.
Figure 5. NU-α-GalCer shows comparable anti-tumor efficacy to α-GalCer in the B16F10
melanoma challenge experiment.
Figure 6. The statistics of the mouse weight and lung weight of vehicle control group, α-GalCer
treatment group and NU-α-GalCer treatment group.
Figure 7. Anti-iTCR antibody NKTT320 showed synergistic anti-tumor activities with mIL-12 in
the B16F10 melanoma challenge experiment.
Figure 8. The statistics of the mouse weight and lung weight of vehicle control group, NKTT320
mAb treatment group, mIL-12 treatment group and combinatorial treatment group.
Figure 9. Anti-CD1d Antibody 51.1.3 was shown to be a potent agonist of hCD1d/iNKT cells
system in the B16F10 melanoma challenge experiment.
Figure 10. The statistics of the mouse weight and lung weight of isotype control group, 51.1.3
mAb treatment group, α-GalCer treatment group and combinatorial treatment group.
-7-
SUMMARY

NKT cells are named due to its sharing cellular makers and biological properties of both T
cells and NK cells, within which a group expressing only a limited choice of αβ TCR are called
iNKT cells. Different from conventional T cells, iNKT cells recognize lipid-based antigen presented
by CD1d molecules on professional APCs with one of the well-studied lipid-based antigens being
α-GalCer. As the bridge between innate immunity and adaptive immunity, iNKT cells, once
activated, can effectively kill tumor cells both indirectly and directly. iTCR recognition of the
endogenous lipid presented by the CD1d on tumor cells can lead to the secretion of several direct
cell killing proteins like granzyme B and perforin, which have been shown to be important for the
tumor suppression activities of regulatory T cell. Also, the Th-1 pro-inflammatory cytokines
produced by iNKT cells can mediate the activation of multiple other tumor-killing cells like NK cells
and CD8+ T cells. In this process, the suppressive tumor microenvironment is maintained by the
introduction of positive feedback loops for DCs and NK cells activation and anti-tumor cytokine
cascade. The potent anti-tumor activities of iNKT cells have made it a favored target for future
cancer therapies designs. This study has explored different novel approaches of mobilizing
human iNKT cells for cancer immunotherapies. An α-GalCer analog, NU-α-GalCer, though
suggesting an improvement in activating iNKT cells in other studies, had no significantly better
performance based on the B16F10 melanoma challenges in this study. Other than that, two
monoclonal antibodies, anti-iTCR antibody NKTT320, and anti-CD1d antibody 51.1.3 were
studied for their agonist function in the CD1d/iNKT cells system. As a result, NKTT320 mAb failed
to show anti-metastatic activities, yet the inclusion of this mAb has enabled a better performance
of mIL-12. The most encouraging discovery of this study would be that of potent iNKT cells agonist,
anti-CD1d antibody 51.1.3. In the melanoma challenge, without any adjuvant, a single
administration of 51.1.3 mAb was able to introduce a significant decrease of the lung nodule
numbers (p=0.0069). Together, these studies have supported the rationale of continuously
-8-
searching for better iNKT cells agonists and have provided useful information for future iNKT cells
study and potential agents choices for clinical trials.

-9-
CHAPTER 1
INTRODUCTION

1.1 iNKT Cells

With the ability to recognizing and eliminating the invading pathogens and tumors, the
immune system acts as the guardian of our health. It functions as a vast orderly network,
comprising a variety of components including molecules, cells, and tissues. A successful immune
response involves both nonspecific and specific components, represented by innate immunity
and adaptive immunity. Innate immunity is an ancient evolutionary system that protects organisms
from a wide variety of pathogens without specificity. As a crucial part of innate immunity, cells like
macrophages, DCs, and NK cells can recognize the pathogen-associated molecular patterns via
pattern recognition receptors and quickly contribute to the innate immunity.

Adaptive immunity is defined by its capability of specific responses towards pathogens,
and the long term memory developed after that. It consists of two functional systems, the antibody-
mediated response that is carried out by B cells/plasma cells and antibodies, and the cell-
mediated response, in which T cells plays a essential part. The T cells respond upon pathogen
infection and tumors through an interaction between the TCRs expressed on the cell surface and
the peptide antigen presented by MHC proteins on the APCs. After activation and proliferation, T
cells interact with other cells either for cytotoxicity or regulatory purposes.

While most of the immune cells can be easily categorized into these two immune systems,
the boundary can be ambiguous for specific cell types, for example, MAIT cells, ILCs, and iNKT
cells (1–6).

-10-
MAIT cells are innate-like T cells, locating in the mucosal barrier. Instead of a variant TCR
repertoire, MAIT cells express a restricted TCR, Vα7.2-Jα33 paired with Vβ2 or Vβ13 in human,
suggesting a conserved antigen recognition (1). MAIT cell TCRs interact with MR1 and recognize
the bacterially-produced vitamin B metabolites and become activated (2). The activated MAIT
cells contribute to the anti-bacterial immunity through direct cell lysis, adaptive immunity
stimulation and an effector memory phenotype (3). Due to these indefinite features, MAIT cells
are considered to be a bridge between the innate and adaptive immunity.

iNKT cells are another important cell population that bridges the innate and adaptive
immune systems. The term 'NK T cells' was first used in 1995 to describe a subset of T cells that
express both cell surface markers of T cells and NK cells (5). Being able to express T cells surface
makers, αβ TCR, and a variety of NK cell-associated markers, like NK1.1 (CD161) enable NKT
cells to function in both innate and adaptive immunity pathways (6).

NKT cells can be further categorized into two main categories, type I NKT cells and type
II NKT cells. Type I NKT cells, also called iNKT cells, are NKT cells that express an iTCR (7). The
reason the cells are called invariant is due to the limited TCR repertoire these cells exhibit.
Through genetic recombination with the help of RAG1 and RAG2, the TCRs on the conventional
T cells can gain diversity and uniqueness. However, this step is strongly limited in the
development of iNKT cells (8). In humans, the α chain of TCR is of the type Vα24-Jα18; in mice,
it is Vα14-Jα18. While the β chains of both species are slightly more diverse, the choices are still
confined in a highly restricted repertoire (7).

-11-
1.2 CD1d and α-GalCer

While a variant TCR repertoire is able to engage a variety of antigenic, presented peptides
by polymorphic MHC, iNKT cells can only recognize the lipid or glycolipids presented through the
non-polymorphic CD1d molecule.

CD1d is a member of the CD1 family of antigen-presenting molecules that is expressed
on the surface of professional APCs, such as DCs, macrophages, and B cells. CD1d is related to
the class I MHC protein. Both MHC class I and CD1d are expressed at lower levels in many
epithelial cells. Also, structurally, they are both heterodimers composed of a β2M and a heavy
chain, of which the extracellular part folds into α1, α2, and α3 domains. In the 3D structure, the
α1 and α2 domains together form a groove for the antigen binding. Restricted by the conformation
of the groove, MHC class I molecules are able to bind peptides with mostly 8-10 AAs. The groove
of CD1d is larger, deeper, and much more hydrophobic, including two pockets that can bind alkyl
chains, the F′ and A′ pockets, which limit the antigens bound to CD1d to be lipids, glycolipids or
lipopeptides (9).

-12-
PRESENTING
MOLECULE
CD1d MHC CLASS I MHC CLASS II
DISTRIBUTION APCs All Nucleated Cells APCs
COMPOSITION One α Chain and β2M One α Chain and β2M One α Chain and
One β Chain
DOMAINS Three α Domains &
One β Domain
Three α Domains &
One β Domain
Two α and β
Domains
ANTIGEN Glycolipid 8-10 AAs 13-18 AAs
ANTIGEN SOURCE Endogenous and
Exogenous
Endogenous Exogenous
ANTIGEN BINDING
DOMAINS
α1 & α2 α1 & α2 α1 & β1
RECEPTOR iTCR Diverse TCR Diverse TCR
ACTIVATING
CELLS
NKT Cell CD8+ T Cell CD4+ T Cell

Table 1. Comparison among CD1d molecules, MHC class I molecules and MHC class II
molecules.

-13-
More and more of the CD1d binding properties of lipid-based antigens have been
discovered, which include endogenous glycolipids and exogenous glycolipids. One of the best-
known antigens of iNKT cells is α-GalCer. α-GalCer is derived from the deep sea sponge Agelas
mauritanius (10). α-GalCer carries a galactose head group that is α-anomerically linked to a
phytoceramide. The phytoceramide contains two acyl chains, visualized in the crystal structures
of both mouse and human CD1d bound to α-GalCer, are shown to fit into the two pockets of CD1d.
With phytoceramide bound in the groove, the galactose is exposed for TCR recognition and
interaction. The invariant α chain of the TCR is responsible for the recognition of the galactose,
whereas the TCR-β chain interacts with CD1d residues. Hence, an α-GalCer/CD1d/iTCR
interaction is established (11).

-14-

Figure 1. Structure of CD1d/iNKT cells systems and α-GalCer. A, The crystal structure of α-
GalCer/CD1d/iTCR complex. PDB ID: 4LHU (12). B, The chemical structure of α-GalCer. C, The
chemical structure of NU-α-GalCer. Created with BioRender.com and ChemDraw from
PerkinElmer Informatics.

-15-
1.3 iNKT Cells in Cancer Immunotherapy

iNKT cells can be activated via antigen presented by CD1d or by cytokines and co-
stimulatory molecules. First, the most common model would be the direct activation of iNKT cells
by CD1d-antigen complexes. The APCs endocytose the exogenous lipids and then present them
to iNKT cells. For endogenous lipids, signaling by several cell surface receptors, e.g., TLRs, is
capable of inducing the loading of self-lipids onto the cell surface CD1d to activate iNKT cells. It
is presumed that the tumor cells that originate from professional APCs present altered self lipids
on its CD1d, which can directly trigger the immune response of iNKT cells towards the tumor cells.
For indirect activation, iNKT cells can be activated by the cytokines secreted from TLR-stimulated
DCs. These cytokines are of the type I IFN, IL-2, IL-18. Moreover, some co-stimulatory signals
can also activate iNKT cells. Sometimes two or more activation signals can act on iNKT cells at
simultaneously in a synergistic manner (8, 13).

Upon activation, iNKT cells can participate in the suppression of tumors through direct
killing and indirect killing. For tumors expressing CD1d molecules on the cell surface, activated
iNKT cells can carry out the anti-tumor function through perforin, granzyme B, Fas-FasL
interaction, and TRAIL (14). At the same time, iNKT cells are able to produce a variety of cytokines
and stimulate the adaptive immunity for tumor killing, including IL-2, IL-3, IL-4, IL-5, IL-6, IL-9, IL-
10, IL-13, IL-17, and IL-21, IFN-γ, TNF-α and GM-CSF (15). Through these cytokines, iNKT cells
can interact with and activate a variety of immune cells, like DCs and NK cells. When the iTCR
recognizes the lipid-based antigen CD1d complex, iNKT cells are activated. Activation-induced
CD40L then binds CD40 on the DCs surface and provides a co-stimulatory signal to activate DCs.
iNKT cells and DCs can also be indirectly activated through the cytokines secreted by each other,
respectively. IL-12, produced upon DCs activation, can stimulate iNKT cells. IFN-γ and IL-4,
produced upon iNKT cells activation, can then stimulate DCs. Both activated cells, through
-16-
various cytokines, are potent to activate NK cells and CD8+ T cells which can directly kill tumor
cells. The positive feedback between both iNKT cells and DCs also exist between other
populations of cells like NK cells and DC cells, which can significantly augment the activation
signals and eventually benefit the anti-tumor activities. The ability to bridge innate and adaptive
immunity has made iNKT cells an exceptional anti-tumor agent of choice (7, 16).

-17-

Figure 2. The anti-tumor activities of iNKT cells. iNKT cells can recognize the CD1d/self
glycolipid complex on some tumor cells directly and kill the tumor through perforin, granzyme B
and T-cell intracellular antigen-1. Indirectly, iNKT cells can be activated by the CD1d/self
glycolipid complex presented by APCs and function to stimulate tumor-killing cells, like CD8+ T
cells and NK cells to suppress tumors. (Created with BioRender.com).

-18-
1.4 Setbacks of α-GalCer in Clinical Trials

The animal administration of α-GalCer has shown significant suppression of tumor
metastases. So far, there have been over 30 clinical trials using α-GalCer in human anti-tumor
immunotherapies (17). However, minimal success was reported for these clinical trials (17, 18).
The potential mechanisms for the low anti-tumor efficiency of α-GalCer in human clinics have
always been contentious, yet mainly focus on four aspects. First, α-GalCer has shown a much
higher affinity for mouse CD1d/iTCR system than the same system in human (19). Second, the
composition and functional properties of iNKT cells in humans and mice are different. Take the
liver as an example, both human liver and mouse liver have the highest iNKT cell concentrations.
However, in mice, the iNKT cells comprise 20 percent to 30 percent of the whole mononuclear
cell population of the liver. Yet in humans, the iNKT cells are only one to two percent of the liver
mononuclear cells (20). Third, the activation of iNKT cells induced by α-GalCer leads to the
production of both Th-1 and Th-2 cytokines. The anti-tumor functions of the pro-inflammatory Th-
1 cytokines are at risks of being counteracted by the anti-inflammatory Th-2 cytokines (8). Finally,
α-GalCer induces anergy in iNKT cells. It was reported that a single injection of α-GalCer induces
anergy of iNKT cells in mice for at least one month (21). The anergy theory was used to explain
this phenomenon: strong TCR signals can be delivered without proper co-stimulation.

To address the affinity problem and the composition problem that is causing the
inconsistency between mice experiments and clinical trials, a human CD1d/iTCR system was
introduced into B6 background mice by our lab (22, 23). Two humanized mouse strains were
generated, hCD1d-KI mice and hCD1d-Vα24Tg mice. hCD1d-KI mice with the hCD1d gene
knocked into the mouse genome, stably express hCD1d molecule. The hCD1d-Vα24Tg mice are
a more humanized strain based on hCD1d-KI mice, with the human iTCR α-chain Vα24Tg
integrated into the genome. With the support of the presentation of humanized CD1d and iTCR,
-19-
the development of iNKT cells was altered, resulting in a human-like cell abundance, composition
and pattern of coreceptor expression, offering a more relevant humanized mouse iNKT model to
study the human CD1d/iNKT cell system.

1.5 Agonistic MAbs in Cancer Immunotherapy

Monoclonal antibodies have been effective treatments in cancer immunotherapy and
quickly have become one of the largest classes of new drugs that obtained approval for cancer
treatment in the last decade. The functions of these anti-tumor monoclonal antibodies can be
categorized into three different groups: blocking/antagonistic antibodies, depleting antibodies,
and agonistic antibodies (24). A blocking antibody is an antibody that does not have a valid
biological function while binding to its receptor, yet does not allow the binding of other antibodies.
The most notable examples are the immune checkpoint inhibitors, anti-CTLA-4 antibody (25),
and anti-PD-1 antibody (26), both of Nobel Prize stature. A depleting antibody can deplete a
population of cells through a diversity of functional activities, including ADCC, ADCP, CDC, direct
cell killing by induction of apoptosis (24). A perfect example would be Rituximab, the monoclonal
anti-CD20 antibody, which can deplete the malignant B cell population through different functions
(27). When an agonistic antibody binds the receptor, it mimics the binding of the natural ligands
and leads to downstream signaling and the normal biological function of the cells. APX005M, a
potent CD40 agonist, developed by Apexigen, has shown positive phase Ib clinical trial data.
Targeting CD40, a common co-stimulatory protein on APCs mimics the binding of CD40 ligand to
CD40 on the cell surface and activates the corresponding downstream signaling. Moreover,
stimulated APCs can further activate the adaptive immune system to kill cancer cells (28). A
significant part of this study involved investigating the potential agonistic function of monoclonal
antibodies in cancer immunotherapy.

-20-
CHAPTER 2
MATERIALS AND METHODS

2.1 Animals

B6 background mice were purchased from The Jackson Laboratory (Bar Harbor, ME) and
bred locally. The generation of hCD1d-KI mice in the B6 background has been reported previously
(22). hCD1d-Vα24Tg mice were generated by crossing Vα24 transgenic mice with a B6
background and hCD1d-KI mice (29). Their genotypes were confirmed as previously described.
All animal procedures were approved by the Institutional Animal Care and Use Committee at the
University of Southern California.

2.2 Cell Lines

The mouse melanoma cell line B16F10 was kindly provided by Xue Huang and Si-Yi Chen
(University of Southern California, Los Angeles, CA). B16F10 cells were maintained in DMEM
(Invitrogen) supplemented with 5% heat-inactivated FBS, 2 mM L-glutamine, penicillin (100 U/mL),
and streptomycin (100 μg/mL). The cells were cultured at 37ºC in an incubator containing 6%
CO 2. Frozen aliquots of cells were prepared upon receipt.

2.3 Antibodies

The mouse monoclonal antibodies against hCD1d, CD1d 51.1.3 were kindly provided by
Steven Porcelli of the Albert Einstein College of Medicine, Bronx, NY. The recombinant
humanized mAb NKTT320, which binds selectively and with high affinity to the human iTCR (30)
-21-
iTCR is from NKT Therapeutics, Inc. The CD11c antibody, eF450-CD11c (N418) was purchased
from Thermo Fisher Scientific for flow cytometry.

2.4 Glycolipids

α-GalCer was purchased from Funakoshi Co., Ltd. NU-α-GalCer is from Dr. Dirk Elewaut of Ghent
University, Belgium.

2.5 Isolation of Progenitor Cells from Mouse Bone Marrow

Mice were anesthetized and sacrificed by anesthesia and cervical dislocation. Under
aseptic conditions, both femur and tibia were isolated, and the muscles surrounding the bones
were carefully stripped via fine surgical scissors. Both ends of of the tibia and femur were snipped.
The bones were held upright and a syringe needle was inserted into the upper end of the bone
and the bone marrow was flushed out into a 100 mm x 15 mm Petri dish (VWR) with complete
RPMI -1640 media (Corning Cellgro) containing 10% FBS, 100 U/ml penicillin, 100 mg/ml
streptomycin, 50 μM 2-ME, 10 mmol HEPES, MEM (1:50), and MEM nonessential AAs (1:100).
The single cell suspension was obtained by passing the cell suspension through a 70 μm cell
strainer (VWR). The single-cell suspension was centrifuged at 1000 rpm for 5 min, and the
supernatant was discarded. The cell pellet was resuspended with Tris-NH 4Cl RBC lysis buffer to
lyse the RBCs. One second centrifugation was performed, and the supernatant was discarded.
The pelleted cells were washed with 1 x PBS and collected.

-22-
2.6 Preparation of BMDCs

Cells suspended in complete RPMI-1640 media were distributed into 100 mm x 15 mm Petri
dishes (VWR) at a density of 2×10
6
cells/Petri dish. Subsequently, mouse GM-CSF and IL-4 were
added into the medium to a final concentration of 20 ng/mL and 10 ng/mL, respectively. The cells
were cultured at 37ºC in an incubator containing 5% CO 2. On day 3, the same amount of medium,
mouse GM-CSF, and IL-4 were supplemented with GM-CSF and IL-6. On days 6 and 8, half of
the culture supernatant was collected and centrifuged (note: many of the DCs developing in the
culture at this point will have detached or become loosely adherent). The cell pellet was
resuspended in 10 ml complete RPMI medium containing 20 ng/ml of GM-CSF and added back
to the original plate. On day 9, the cells were incubated with LPS for 24 h, and BMDCs were
obtained.

2.7 Verification of BMDCs Maturation

On day 1, one-fifth of the freshly isolated progenitor cells from mouse bone marrow were
collected, centrifuged and washed with PBS, and divided into several fractions of 5×10
5
cells/100
μL. Cells were incubated with anti-CD16/32 to block FcγRII/III, followed by staining with
fluorochrome-conjugated mAb for mouse antigens, eF450-CD11c (N418) and then examined by
FACS Canto II (BD Biosciences) and FlowJo software (Tree Star, Ashland, OR). On day 10, The
non-adherent or loosely adherent BMDCs are collected by gentle pipetting, centrifuged and
washed, stained and examined the same way to detect the CD11c expression on the cell surface.

-23-
2.8 In vivo B16F10 Metastasis Assay

To investigate the anti-tumor function of iNKT cells in humanized mice, B16F10 melanoma
challenge was adopted. The B16F10 melanoma model is a well-established model for human
melanoma and the anti-metastasis study of anti-tumor agents. Cultured B16F10 cells were
trypsinized, and a single-cell suspension was obtained. Age/gender-matched hCD1d-KI (6–8-
week-old) mice were administered with 10
5
B16F10 cells, 2 mg α-GalCer via tail vein. For BMDC-
mediated lipid delivery, on day 8 of the BMDC preparation, cells were pulsed with α-GalCer (100
ng/mL) or vehicle (0.025% of polysolvate 20) at 37ºC in a 5% (vol/vol) CO 2 incubator for 40 h, and
1 μg/mL of LPS was added for the final 16 h. After filtration through 70 μm cell strainer (VWR),
vehicle-BMDCs or α-GalCer–pulsed BMDCs were iv injected into mice along with B16F10 cells.
For groups with mIL-12 treatment, mIL-12 was administrated i.p. every day after day one.
Recipient mice were euthanized after two weeks. The mouse lungs were collected and perfused
with PBS, and the surface melanoma nodules were counted with a Omano OM4724 microscope
at 20 x magnification. The maximal amount of lung nodules of a mouse is set to 800 based on
previous experience. Also, the weight of the mice both before administration and before sacrificing
and the final lung weight were recorded for inter-group comparison.

2.9 Statistical Analysis

Fair intra-group comparisons are possible due to the age/gender-matched grouping
method. The nodule numbers were analyzed between any two groups for comparison of the anti-
tumor outcomes of different treatment. The data is assumed to be normally distributed. Standard
deviation is calculated and compared between groups. According to Levene’s test, the universal
significantly different standard deviation suggests unequal variances between groups. Hence,
Welch t-tests are applied. Based on the rough estimation of the lung photos and nodule numbers,
-24-
the one-direction hypothesis was made, and the one-tailed p-value was calculated. A p-value of
<0.05 was considered significant. If a p-value is less than 0.05, it is flagged with one star (*). If a
p-value is less than 0.01, it is flagged with two stars (**). If a p-value is less than 0.001, it is flagged
with three stars (***). If a p-value is less than 0.0001, it is flagged with four stars (****).

Statistics was also done using the data of mouse weights and final lung weights as
supporting evidence of the tumor development status. (No statistically significant difference was
shown and included in this paper). Ideally, along with the lung melanoma development, the mouse
will lose weight. The final lung weights and percentage of body weight of lungs will accordingly
increase.

-25-
CHAPTER 3
RESULTS

3.1 In the current setting, NU-α-GalCer shows comparable anti-tumor efficacy compared to
α-GalCer.

Figure 3. The general B16F10 melanoma challenge workflow in this study. Age/gender-
matched grouped mice were injected with B16F10 cells together with different treatments. Two
weeks later, the mice were sacrificed and the lungs were harvested for lung nodule counting.
Initial and final mouse weights and final lung weights were recorded for statistical purposes as
well. (Figures created with BioRender.com).

-26-
There are over 30 clinical trials using α-GalCer in human anti-tumor immunotherapies.
Despite its massive success in animal experiments, there are few encouraging reports about α-
GalCer in clinical trials. As described before, one of the problems might lie in the different
composition of iNKT cells in mice and humans, namely a different α-GalCer affinity of mouse and
human CD1d/iTCR systems. By introducing the hCD1d gene into the mouse genome, we have
created an experimental animal model sharing the same iNKT cells composition and α-GalCer
affinity as in humans. However, other problems α-GalCer might cause, such as antagonizing
activities of the secreted pro-inflammatory Th1-cytokines (e.g., IFN-γ and IL-12), and anti-
inflammatory Th-2 cytokines (e.g., IL-4), have led people on a search for improved α-GalCer
analogs, with a Th-1 biased cytokine release capability. NU-α-GalCer has been synthesized and
found to be potent for inducing Th-1 biased cytokine production, making it a potentially better
candidate for the anti-tumor response (31).

On the molecular level, NU-α-GalCer has a higher binding affinity to the CD1d/iTCR
system than does α-GalCer. The crystal structure of the α-GalCer/CD1d/iTCR interaction shows
strong hydrogen bonding between the human iTCR α chain AAs and the 2′′-OH, 3′′-OH and 4′′-
OH groups of the α-GalCer galactose head group. However, the 6′′-OH group is the only hydroxyl
group not forming a hydrogen bond with iTCR, suggesting a potential position of α-GalCer for
modification. In NU-α-GalCer, the 6′′-OH group is replaced with a hydrophobic aromatic group to
form a new hydrogen bond between the carbonyl oxygen of the urea linker connecting the
galactose and the naphthyl moieties with Thr159 of CD1d (31), which significantly increases the
buried surface areas between TCR-CD1d and glycolipid (a positive indicator for the interaction
strength) (32). Both cytokine production and molecular structure have suggested NU-α-GalCer is
a better agent for cancer immunotherapy than α-GalCer. Thus, in this study, a B16F10 melanoma
challenge was performed accordingly and considering the potency of NU-α-GalCer, a dosage of
0.1 μg lipid/mouse was applied, lower than the normal dosage of 2 μg lipid/mouse.
-27-

There have been arguments regarding the most efficient way to deliver the glycolipids to
the immune system. Compared to free soluble glycolipids injection, the DC-vehicled glycolipids
have shown better performance in terms of overcoming iNKT cell anergy and functional
persistence (16, 33). Anergy has always been another setback of the anti-tumor clinical
application of glycolipids, not only making the activation of iNKT cells inefficient but could also
potentially exacerbate the tumorigenesis. However, the lipid loading form in DCs- can effectively
prevent anergy. Moreover, in that study, the DC-vehicled α-GalCer can potently stimulate higher
numbers of cytokine-secreting splenocytes. In another experiment attempting to use α-GalCer for
immunization, the results showed that vaccination with α-GalCer-loaded tumor cells is sufficient
to protect mice from the same challenge for up to one year (34).

Thus, based on literature reports, in this study the glycolipids were delivered with BMDCs
to augment the anti-tumor function. Bone marrow progenitor cells were obtained from a hCD1d
KI mouse 11 days before the experiment for in vitro differentiation into BMDCs. Bone marrow
progenitor cells and BMDCs were stained for CD11c for flow cytometry and analyzed in the same
manner. The FACS results show a significant increase in the CD11c+ cell population, suggesting
a successful differentiation and an effective loading of the glycolipids. A total of 18 mice with
hCD1d-Vα24Tg genotype were prepared for the B16F10 metastasis assay. B16F10 cells were
mixed together with BMDCs-loaded glycolipids and administered to the mice via the tail vein.
Control group mice (n=8) were injected with the mixture of B16F10 cells and BMDCs. α-GalCer
treatment group mice (n=4) were injected with the mixture of B16F10 cells and α-GalCer-loaded
BMDCs. The NU-α-GalCer treatment group mice (n=6) were injected with the mixture of B16F10
cells and NU-α-GalCer-loaded BMDCs. Theoretically, any individual difference was eliminated
through age/gender-matched grouping.

-28-

Figure 4. Preparation of BMDCs to deliver glycolipids. A, The workflow of isolation of
progenitor cells from mouse bone marrow and induction of BMDCs. B, The comparison of
CD11c positive cell population between newly-isolated bone marrow progenitor cells and mature
BMDCs. After cytokine induction, DCs were successfully induced. (Created with BioRender.com).

-29-

Figure 5. NU-α-GalCer shows comparable anti-tumor efficacy to α-GalCer in the B16F10
melanoma challenge experiment. A, The statistical analysis of the lung nodule numbers of
different experimental groups. Error bars represent the standard deviation. B, the photos of the
lungs of different experiment groups.

-30-
As shown in the photos, mice from all of the groups developed B16F10 melanoma tumors
in the lungs; the qualitative differences among the groups is obvious. For the vehicle control group,
the mouse lungs are completely covered/filled with B16F10 melanoma tumors such that the lung
tissue and the original color of the lung are obscured. For the α-GalCer treatment group, there is
a significant decrease in the lung nodule numbers. The lung tissue and the normal color of the
lung are visible. The melanoma development of the NU-α-GalCer treatment group mouse lungs
was similar to the α-GalCer treatment group; however, slightly fewer nodules can be observed.
However, a decreased nodule size was observed in the NU-α-GalCer treatment group. A
statistical comparison of the nodule numbers showed a significant difference between the vehicle
control group and the other two groups, while there was no evidence supporting a statistically
significant improvement of the NU-α-GalCer treatment group. (vehicle control group vs. NU-α-
GalCer treatment group: p=0.0001, ***; vehicle control group vs. α-GalCer treatment group:
p<0.0001, ****; NU-α-GalCer treatment group vs. α-GalCer treatment group: p=0.2769, ns).
These results suggested that as potent as the anti-tumor activity of NU-α-GalCer, it is not
statistically significant with the well known iNKT cells agonist, α-GalCer.

-31-

Figure 6. The results of the mouse weights and lung weights of vehicle control group, α-
GalCer treatment group and NU-α-GalCer treatment group. A, The final mouse weights of
different experiment groups before lung dissections. B, The mouse weight change of different
experiment groups during the time of treatment. C, The final mouse lung weights of different
experiment groups. D, The percentage of total body weight of lungs of different experiment groups.
Error bars represent the standard deviation.

-32-
For weight differences, at the end of this experiment, α-GalCer and NU-α-GalCer
treatment groups mice were shown to be heavier than the vehicle control group mice. This is
reasonable considering the vehicle control group mice developed more severe melanoma
metastases. We observed the mouse weight changes, a more accurate measurement, and final
mouse weight trended the same with each treatment group. All the vehicle control group mice lost
weight while the other two groups were able to maintain their weight over two weeks (except one
outlier of the α-GalCer treatment group). The final lung weights and the percentage body weights
of lungs showed a perfect opposite trend. Due to the more extensive tumor development of the
melanoma, the vehicle control group mice had the heaviest lungs, both absolutely and relatively.
Thus, the weight data was perfectly supported and consistent with the melanoma metastasis
results derived from enumeration of lung nodules.

3.2 Anti-iTCR antibody NKTT320 has synergistic anti-tumor activities with mIL-12

NKTT320 is a humanized mAb targeting human iTCR. With another humanized iTCR mAb
NKTT120 is a potent depleting antibody for iNKT cells (35). NKTT320 is attracting attention for its
potential agonistic interaction with iNKT cells. The exposure of human iNKT cells to NKTT320
mAb was shown to lead to potential iNKT activation both in vitro and in vivo (30). Moreover, in the
in vitro experiment of incubating iNKT cells with NKTT320 mAb, the supernatant IFN-γ levels
show a significant dose-dependent increase (36). These results have suggested that NKTT320
mAb is an iNKT cell agonistic antibody. Considering the potent anti-tumor function of iNKT cells,
an attempt was made in this study to utilize NKTT320 immunomodulatory function for cancer
immunotherapy.

IL-12 is a pro-inflammatory cytokine and a potent anti-tumor candidate. IL-12 acts in a
heterodimeric form (IL-12 p70: IL-12 p35/p40). These can be induced through TLR pathways or
-33-
another cytokine network, such as CD40-CD40L. After release and binding to its receptor, IL-2
will start to carry out its function as an organizer of the Th-1 type anti-tumor immune response.
First, Jak-2 will be phosphorylated upon IL-12R-β receptor recognition of IL-12 p35, which then
binds to STAT4 to effect its phosphorylation. The phosphorylated STAT4 forms a homodimer and
enters the nucleus to promote the transcription of IFN-γ (37). In addition to being the transcription
factor for IFN-γ, IL-12 can also carry out its anti-tumor function through a variety of pathways, for
example, IL-12 can act as a cytotoxicity stimulator of activated NK cells, CD8+ T cells, and CD4+
T cells and it can also switch the antibody production of B cells from IgE towards IgG (38).
Moreover, one study showed that in vivo IL-12 augmented the response of iNKT cells to CD1d-
presented lipid-based antigen (39). Due to its potent immunomodulatory activity, IL-12 has been
used as a vaccine adjuvant for protection against bacteria, viruses and cancer. In an effort to
maximize tumor suppression and study the potential synergistic anti-tumor function of NKTT320
mAb and mouse IL-12 (mIL-12), a single treatment group and a combinatorial treatment group of
mIL-12 were tested. A total of 17 mice with hCD1d-Vα24Tg genotype were prepared for the
B16F10 metastasis assay. B16F10 cells were mixed together with mAb and administered to these
mice via the tail vein. The mIL-12 was administrated via i.p. Control group mice (n=4) were
injected with B16F10 alone. The NKTT320 mAb treatment group mice (n=3) were injected with
the mixture of B16F10 cells and NKTT320 mAb. The mIL-12 treatment group mice (n=5) were
injected with the mixture of B16F10 cells and mIL-12. NKTT320 and mIL-12 combinatorial
treatment group mice (5) were injected with the mixture of B16F10 cells, NKTT320 mAb, and mIL-
12.

-34-

Figure 7. Anti-iTCR antibody NKTT320 showed synergistic anti-tumor activities with mIL-
12 in the B16F10 melanoma challenge experiment. A, The statistical analysis of the lung
nodule numbers of different experimental groups. Error bars represent the standard deviation. B,
the photos of the lungs of different experiment groups.

-35-
As shown in the photos, mice from of the groups developed B16F10 melanoma in the
lungs. For the vehicle control group, the mouse lungs were extensively covered/filled with B16F10
melanoma obscuring both normal lung tissue and color. As for the NKTT320 mAb treatment group,
there is a noticeable decrease in the lung nodule numbers although one of the mice showed
melanoma-covered/filled lungs, just like the control mice of the vehicle alone group. For the mIL-
12 treatment group, there was a significant decrease in the lung nodule numbers. The lung tissue
and the original color of the lung were visible. The melanoma development of the NKTT320 and
mIL-12 combinatorial treatment group mouse lungs was similar to the mIL-12 treatment group.
However, slightly fewer nodules were observed. A statistical comparison of the nodule number
showed a significant difference between the vehicle control group and the two groups, including
mIL-12 as well as between NKTT320 treatment group and NKTT320 and mIL-12 combinatorial
treatment group. However, no evidence supported a statistically significant difference between
the vehicle control group and NKTT320 treatment group (vehicle control group vs. NKTT320 mAb
treatment group: p=0.2113, ns; vehicle control group vs. mIL-12 treatment group: p=0.0158, *;
vehicle control group vs. NKTT320 and mIL-12 combinatorial treatment group: p=0.0032, **;
NKTT320 treatment group vs. NKTT320 and mIL-12 combinatorial treatment group: p=0.0392, *).
These results confirmed the anti-tumor activity of mIL-12. Although there is no statistically
significant difference between the mIL-12 treatment group and the combinatorial treatment group,
a difference can still be observed between these two groups due to the presence of NKTT320
mAb. The addition of NKTT320 mAb helped improve the p-value from 0.0158 to 0.0032. Also,
based on what was shown in the photos, most of the NKTT320 treated samples showed a visually
significant difference compared to the vehicle control group samples. This suggested that the set
of maximal lung nodules numbers to 800 might compromise the power of statistics. Summing up
the above, there might be an unrevealed difference between the vehicle control group and
NKTT320 mAb treatment group. Moreover, the anti-iTCR antibody NKTT320 had synergistic anti-
tumor activities with mIL-12.
-36-

Figure 8. The results of the mouse weights and lung weights of vehicle control group,
NKTT320 mAb treatment group, mIL-12 treatment group and combinatorial treatment
group. A, The final mouse weights of different experiment groups before lung dissections. B, The
mouse weight change of different experiment groups during the time of treatment. C, The final
mouse lung weights of different experiment groups. D, The percentage of total body weight of
lungs of different experiment groups. Error bars represent the standard deviation.

-37-
For weight differences, at the end of this experiment, mIL-12, as well as NKTT320 and
mIL-12 combinatorial treatment group mice, were heavier than the NKTT320 mAb treatment
group mice, which is reasonable considering the NKTT320 mAb treatment group mice have
developed much more severe melanoma. The mouse weight changes continued a similar trend
through the final mouse weights. Two mIL-12 treatment mice had lost less weight than the
NKTT320 mAb treatment group mice. However, the vehicle control group mice weighed the most
among the four groups and had a lesser weight change than the NKTT320 mAb treatment group
mice, which was not expected. Theoretically, the vehicle control mice should have developed the
more severe melanoma, which supposed makes them weigh the least and lose the most weight.
This study attributed the abnormality in mouse weights in the vehicle control group to the imperfect
grouping. Despite the deviation shown in the mouse weight analysis, the final lung weight and the
percentage body weight of lungs consistently supported the melanoma metastasis results derived
from lung nodule number counting. Due to the higher development of the melanoma, the vehicle
control group mice had the heaviest lungs, both absolutely and relatively. The NKTT320 mAb
treatment group was the second highest, and the remaining two groups had both the lowest final
lung weights and percentage body weights of lungs. It is notable that the outlier in the previous
lung nodule number analysis was shown clearly in the lung weight study, suggesting a high level
of consistency and fidelity.

3.3 Anti-CD1d Antibody 51.1.3 is a potent agonist of hCD1d

Anti-CD1d mAb 51.1.3 has been frequently used in CD1d and iNKT cell studies. However,
with detection being major use of 51.1.3, there is little known about the antibody. Here, the
agonistic function of 51.1.3 was examined due to its resemblance to MHC class I molecules.

-38-
The anti-MHC class I antibodies can stimulate the survival and proliferation of targeted
cells once bound to the cell surface MHC class I molecules. In human, the natural anti-MHC class
I antibodies can be produced in several situations: transplantation, pregnancy, and blood
transfusion (40). The ligation of the antibodies of MHC class I molecules on endothelial cells can
further trigger downstream signaling of MHC class I in an antibody concentration-dependent way.
A low concentration of antibody will simulate a phosphorylation cascade through FAK, PI3K, and
Bcl-2, eventually leading towards cell survival. At a high antibody concentration this can promote
cell proliferation through the phosphorylation of MAPK (41).

There is no antibody agonist of CD1d molecules reported so far. To better understand the
function of 51.1.3 on a cellular level and potentially contribute to cancer immunotherapy, anti-
CD1d antibody 51.1.3 was administrated to mice together with B16F10 melanoma cells. We
wanted to maximize tumor suppression and study the potential synergistic anti-tumor function of
51.1.3 mAb and α-GalCer, thus a single treatment group and a combinatorial treatment group of
α-GalCer were used. For this study a total of 14 mice of hCD1d genotype were prepared for the
B16F10 metastasis assay. B16F10 cells were mixed together with the mAb or/and α-GalCer and
injected into the mice via the tail vein. Isotype group control mice (n=3) were injected with the
mixture solution of B16F10 cells and a mouse IgG2b isotype control. The mAb. 51.1.3 treatment
group mice (n=3) were injected with the mixture of B16F10 cells and 51.1.3 mAb. α-GalCer
treatment group mice (n=4) were injected with the mixture of B16F10 cells and α-GalCer. 51.1.3
and α-GalCer combinatorial treatment group mice (n=4) were injected with the mixture solution of
B16F10 cells, 51.1.3 mAb, and α-GalCer.

-39-

Figure 9. Anti-CD1d Antibody 51.1.3 was shown to be a potent agonist of hCD1d/iNKT cells
system in the B16F10 melanoma challenge experiment. A, The statistical analysis of the lung
nodule numbers of different experimental groups. Error bars represent the standard deviation. B,
the photos of the lungs of different experiment groups.

-40-
As shown in the photos, mice from all groups developed B16F10 melanoma tumors in the
lungs. For the isotype control group, the mouse lungs show noticeably the most B16F10
melanoma lung nodules. As for the 51.1.3 mAb treatment group, there was a noticeable decrease
in the lung nodule numbers compared to the isotype control group. For α-GalCer treatment group
as well as 51.1.3 mAb and α-GalCer combinatorial treatment group, minimal lung nodule numbers
were observed. Lung tissue and the normal tissue color of the lung are visible for most areas of
the lung; also decreased nodule size was observed in the combinatorial treatment group. A
statistical comparison of the nodule numbers shows a significant difference between the isotype
control group and every one of the three treatment groups. There is no evidence supporting a
statistically significant difference between 51.1.3 mAb treatment group and 51.1.3 mAb and α-
GalCer combinatorial treatment group. (isotype control group vs. 51.1.3 mAb treatment group:
p=0.0069, **; isotype control group vs. α-GalCer treatment group: p=0.0024, **; isotype control
group vs. as 51.1.3 mAb and α-GalCer combinatorial treatment group: p=0.0031, **; 51.1.3 mAb
treatment group vs. 51.1.3 mAb and α-GalCer combinatorial treatment group: p=0.0962, ns)
These results suggested the potent anti-tumor activities of the anti-CD1d antibody 51.1.3, were
statistically equivalent to α-GalCer, the well-known potent anti-tumor agent (p=0.2728, ns).

-41-

Figure 10. The results of the mouse weights and lung weights of isotype control group,
51.1.3 mAb treatment group, α-GalCer treatment group and combinatorial treatment group.
A, The final mouse weights of different experiment groups before lung dissections. B, The mouse
weight change of different experiment groups during the time of treatment. C, The final mouse
lung weights of different experiment groups. D, The percentage of total body weight of lungs of
different experiment groups. Error bars represent the standard deviation.

-42-
For weight differences, at the end of this experiment, the 51.1.3 mAb and α-GalCer
combinatorial treatment group mice were the heaviest followed by the α-GalCer treatment group
and the 51.1.3 mAb treatment group. The isotype control group had the lowest body weight, which
is reasonable considering the vehicle control group mice have developed much more severe
melanoma. The mouse weight changes followed a similar trend of final mouse weights, with
slightly more deviation. The final lung weights and the percentage of body weight of lungs showed
a perfect opposite trend. Due to the higher development of the melanoma, the vehicle control
group mice had the heaviest lungs, both absolutely and relatively. Thus, the weight data perfectly
supported and verified the melanoma metastasis results derived from lung nodule number
counting.

-43-
CHAPTER 4
DISCUSSION

Both in vitro and in vivo experiments indicate the important role that iNKT cells play in
tumor immunosurveillance. Being able not only to recognize and directly kill the CD1d-expressing
tumor cells but also respond to the CD1d-expressing APCs within the tumor microenvironment
and regulate the tumor growth via cytokine production, iNKT cells have shown great potential as
a cancer immunotherapy target for clinical research. There have been clinical trials administrating
soluble α-GalCer or α-GalCer pulsed DCs or activated iNKT cells for cancer treatment. In general,
these iNKT-based cancer immunotherapy trials have shown great potential to induce clinical
responses against many types of cancer (33). Although, the soluble α-GalCer study met with little
success and the administration even triggered an instant decrease of iNKT cells; this was then
attributed to apoptosis upon activation. Also, the remaining iNKT cells went into anergy and no
longer responded to the same administration. The solution to the problem of different outcomes
between in vivo mouse experiments and clinical trials is assumed to lie in the experimental models.

Humanized mouse models have been generated by our lab to address two potential
mechanisms causing the inconsistent results within different experimental models: the affinity
difference of α-GalCer and CD1d/iTCR system and the different iNKT cells composition between
humans and mice. The introduced gene, hCD1d and human iTCR α chain Vα24-Jα18 have not
only mimicked the affinity in humans but also lowered the mice iNKT cells abundance to observed
human levels by affecting the cell development, which makes the models perfect for iNKT cells in
vivo study (22, 29). On the other hand, the research of α-GalCer has also significantly been
hindered by other problems, including the iNKT cell anergy and Th-1/Th-2 cytokine induction it
causes. While the tumor progression after administration of α-GalCer loaded DCs suggests an
approach to avoid anergy effects, a new α-GalCer analog NU-α-GalCer is reported to cause a
-44-
Th-1 biased cytokine production (31). At the same time, the molecular structure study has proved
a higher affinity to the CD1d/iTCR system of NU-α-GalCer, making it a potential replacement for
α-GalCer in cancer therapies (32).

This study tested the in vivo anti-tumor activities of NU-α-GalCer using B16F10 melanoma
model in comparison to α-GalCer. The results showed an anti-metastatic ability of NU-α-GalCer
as potent as α-GalCer though statistics showed no improved benefit. However, a p-value of
0.2769 and a visible improvement of the lung status suggested a potential statistical difference if
more mice subjects were included. In the meanwhile, more titration is needed.

A novel attempt was made to stimulate iNKT cells in vivo using monoclonal antibodies.
Monoclonal antibodies have been widely adopted in cancer immunotherapies. For different
monoclonal antibodies, various functions have been exploited to either deplete or mobilize the
targeted cell population or block ligand/receptor interactions. Two monoclonal antibodies
targeting different components of the glycolipid/CD1d/iTCR system were introduced, anti-CD1d
antibody 51.1.3 mAb and anti-iTCR antibody NKTT320 mAb. A potential agonist function was
assumed for both antibodies. In the B16F10 melanoma challenges, adjuvants were used to
amplify the anti-tumor effect. mIL-12 was administrated i.p. together with NKTT320 mAb i.p. and
α-GalCer was administrated together with 51.1.3 mAb both i.v. The results revealed synergistic
anti-tumor activities of anti-iTCR antibody NKTT320 and mIL-12. Though the sole administration
of NKTT320 mAb did not show a statistically significant anti-tumor activity, again a p-value of
0.2113 and a visible improvement of the lung status suggested a potential benefit to further
investigate if more mice subjects are included. For the study of anti-CD1d antibody 51.1.3 mAb,
a direct anti-meta-statistical activity can be concluded based on the decrease of the data of lung
nodules numbers. Also, a synergy between 51.1.3 mAb and α-GalCer may be possible though
this was not suggested by statistics.
-45-

It is worth mentioning that in the B16F10 melanoma challenges, lung nodule numbers
were used as the readout of the tumor metastases. However, the metastatic capability of the
tumor can be compromised as the result of multiple factors, with iNKT cells being only one of
them. A more direct link is needed for attribution of weakened metastases to the function of iNKT
cells. Hence, in vitro and in vivo DCs/iNKT cells activation experiments in the future would be
necessary to confirm the role iNKT cells play in the B16F10 melanoma challenges.

In this study, three potential iNKT cells agonists were tested. So far, little is known about
the CD1d/iTCR interaction and how it triggers the anti-tumor activities of iNKT cells. A potent
agonist would significantly accelerate the future mechanism studies. This study, in summary,
helped identify potential cancer immunotherapy agents. More importantly, it has also rationalized
the exploitation of iNKT cells function in the fight against cancer. In fact, iNKT cells have been
shown to be crucial in terms of other aspects of human health and disease besides cancer.
Though more evidence is needed, it is widely accepted that iNKT cells play a regulatory function
in the suppression of autoimmunity (42). Also, being among the first cells to respond to an
infection, iNKT cells play a significant part in defending against infectious agents. Reports have
shown direct iNKT cell recognition of glycolipid antigens from pathogenic bacteria, for example,
Streptococcus pneumoniae (43). An anti-viral function is also reported by our lab based on the
ability of Herpes Simplex Virus type 1 (HSV-1) to downregulate CD1d expression to evade
immunity (44). Hence the study of iNKT cells agonists can be helpful to manipulate many functions
of iNKT cells besides the anti-tumor activities. To do that, a better understanding of
glycolipid/CD1d/iTCR system binding models and the regulatory mechanisms of iNKT cells and
cytokine production is needed.

-46-
ACKNOWLEDGMENTS

First and foremost, I would like to express my sincere gratitude to my advisor and the chair
of my committee Dr. Weiming Yuan for all his mentoring and support of my Master’s study and
research, for his patience, studiousness, enthusiasm, and inspiration. His guidance helped me in
all the time of research and writing of this thesis. I could not have imagined having a better mentor
for my Master’s study.

Moreover, I would like to thank the rest of my thesis committee: Dr. Chengyu Liang and
Dr. Stanley Tahara, for their encouragement, insightful comments, and questions.

I would also like to thank my co-workers in Dr. Yuan’s lab, Siyang Chen, Xiaotian Feng,
Yingting Zhang, Saki Watanabe, Ryan Springfield, Rongxi Zhao, Jae Ho, Chelsea Moon and Dr.
Rirong Yang who have helped create a harmonious and productive working environment and
supported me emotionally.

Last but not the least, I wish to thank my parents for their love and trust, without whom I
would never have had such a great opportunity studying and enjoying my life at the University of
Southern California.

-47-
REFERENCES

1. Chua, W., S. M. Truscott, C. S. Eickhoff, A. Blazevic, D. F. Hoft, and T. H. Hansen. 2012.
Polyclonal Mucosa-Associated Invariant T Cells Have Unique Innate Functions in Bacterial
Infection. Infect. Immun. 80: 3256–3267.
2. Nours, L., B. Meehan, L. Liu, M. Bhati, Z. Chen, L. Kostenko, R. Reantragoon, N. A.
Williamson, A. W. Purcell, N. L. Dudek, M. J. Mcconville, R. A. J. O. Hair, G. N. Khairallah, D. I.
Godfrey, and D. P. Fairlie. 2012. MR1 presents microbial vitamin B metabolites to MAIT cells.
Nature 491: 717–725.
3. Napier, R. J., E. J. Adams, M. C. Gold, and D. M. Lewinsohn. 2015. The Role of Mucosal
Associated Invariant T Cells in Antimicrobial Immunity. Front. Immunol. 6: 1–10.
4. Withers, D. R. 2016. Innate lymphoid cell regulation of adaptive immunity. Immunology 149:
123–130.
5. Makino, Y., R. Kanno, T. Ito, K. Higashino, and M. Taniguchi. 1995. Predominant expression
of invariant Vα14+ TCR α chain in NK1.1+ T cell populations. Int. Immunol. 7: 1157–1161.
6. Juno, J. A., Y. Keynan, and K. R. Fowke. 2012. Invariant NKT Cells : Regulation and Function
during Viral Infection. PLoS Pathog. 8.
7. Mcewen-smith, R. M., M. Salio, and V. Cerundolo. 2015. Europe PMC Funders Group The
regulatory role of invariant NKT cells in tumor immunity. Cancer Immunol. Res. 3: 425–435.
8. Kaer, L. Van, V. V Parekh, and L. Wu. 2011. Invariant natural killer T cells : bridging innate
and adaptive immunity. Cell Tissue Res. 343: 43–55.
9. Porcelli, S. A. 2005. Bird genes give new insights into the origins of lipid antigen presentation.
Proc Natl Acad Sci U S A 102: 8399–8400.
10. Takenori Natori, Masahiro Morita, Kohji Akimoto, Y. K. 1994. Agelasphins, novel antitumor
and immunostimulatory cerebrosides from the marine sponge Agelas mauritianus. Tetrahedron
-48-
50: 2771–2784.
11. Girardi, E., and D. M. Zajonc. 2012. Molecular basis of lipid antigen presentation by CD1d
and recognition by natural killer T cells. Immunol. Rev. 250: 167–179.
12. Uldrich, A. P., J. Le Nours, D. G. Pellicci, N. A. Gherardin, K. G. Mcpherson, R. T. Lim, O.
Patel, T. Beddoe, S. Gras, J. Rossjohn, and D. I. Godfrey. 2013. CD1d-lipid antigen recognition
by the gd TCR. Nat. Immunol. 14: 1137–1147.
13. Danië lle Horst, Ruben J. Geerdink, Anna M. Gram, A. J. S. and M. E. R. 2012. Hiding Lipid
Presentation: Viral Interference with CD1d-Restricted Invariant Natural Killer T (iNKT) Cell
Activation. Viruses 4: 2379–2399.
14. Bedard, M., M. Salio, and V. Cerundolo. 2017. Harnessing the Power of Invariant Natural
Killer T Cells in Cancer Immunotherapy. Front. Immunol. 8.
15. Seo, E. B. H., I. Kim, and I. J. C. Kang. 2019. Roles of NKT cells in cancer immunotherapy.
Arch. Pharm. Res. .
16. Nair, S., and M. V Dhodapkar. 2017. Natural Killer T Cells in Cancer immunotherapy. Front.
Immunol. 8: 1178–1195.
17. Waldowska, M., A. Bojarska-junak, and J. Roliński. 2017. A brief review of clinical trials
involving manipulation of invariant NKT cells as a promising approach in future cancer
therapies. Cent. Eur. J. Immunol. 42: 181–195.
18. Motohashi, S., and T. Nakayama. 2008. Clinical applications of natural killer T cell – based
immunotherapy for cancer. Cancer Sci. 99: 638–645.
19. Pellicci, D. G., O. Patel, L. Kjer-nielsen, S. S. Pang, L. C. Sullivan, K. Kyparissoudis, A. G.
Brooks, H. H. Reid, S. Gras, I. S. Lucet, R. Koh, M. J. Smyth, T. Mallevaey, J. L. Matsuda, L.
Gapin, J. Mccluskey, and D. I. Godfrey. 2009. Article Differential Recognition of CD1d- a -
Galactosyl Ceramide by the V b 8 . 2 and V b 7 Semi-invariant NKT T Cell Receptors. 47–59.
20. Lawrenczyk, A., S. Kim, X. Wen, R. Xiong, and W. Yuan. 2014. Exploring the Therapeutic
-49-
Potentials of iNKT Cells for Anti-HBV Treatment. Pathogens 3: 563–576.
21. Sullivan, B. A., and M. Kronenberg. 2005. Activation or anergy : NKT cells are stunned by α
-galactosylceramide. J. Clin. Invest. 115: 2328–2329.
22. Wen, X., P. Rao, L. J. Carreno, S. Kim, A. Lawrenczyk, S. A. Porcelli, P. Cresswell, and W.
Yuan. 2013. Human CD1d knock-in mouse model demonstrates potent antitumor potential of
human CD1d-restricted invariant natural killer T cells. Proc. Natl. Acad. Sci. 110: 2963–2968.
23. Wen, X., S. Kim, R. Xiong, M. Li, A. Lawrenczyk, X. Huang, S. Chen, P. Rao, G. S. Besra,
P. Dellabona, G. Casorati, A. Porcelli, O. Akbari, and M. A. Exley. 2015. A Subset of CD8 α β +
Invariant NKT Cells in a Humanized Mouse Model. J. Immunol. 195: 1459–1469.
24. Weiner, L. M., M. V Dhodapkar, and S. Ferrone. 2009. Monoclonal antibodies for cancer
immunotherapy. Lancet 373: 1033–1040.
25. Leach, D. R., M. F. Krummel, and J. P. Allison. 1996. Enhancement of Antitumor Immunity
by CTLA-4 Blockade. Science (80-. ). 271: 32–35.
26. Ishida, Y., Y. Agata, K. Shibahara, and T. Honjo. 1992. Induced expression of PD-1 , a
novel member of the immunoglobulin gene superfamily , upon programmed cell death. EMBO J.
1: 3887–3895.
27. Weiner, G. J. 2011. Rituximab : mechanism of action. Semin. Hematol. 47: 115–123.
28. Bjorck, P., E. Filbert, Y. Zhang, X. Yang, and O. Trifan. 2015. The CD40 agonistic
monoclonal antibody APX005M has potent immune stimulatory capabilities. J. Immunother.
Cancer 3: P198.
29. Wen, X., S. Kim, R. Xiong, M. Li, A. Lawrenczyk, X. Huang, S. Chen, P. Rao, G. S. Besra,
P. Dellabona, G. Casorati, A. Porcelli, O. Akbari, and M. A. Exley. 2015. A Subset of CD8 α β +
Invariant NKT Cells in a Humanized Mouse Model. J. Immunol. 195: 1459–1469.
30. Das, R., F. Scheuplein, N. P. Patel, P. Guan, R. G. Schaub, and K. E. Nichols. 2014.
NKTT320, a novel monoclonal antibody activates the immuno-stimulatory functions of human
-50-
invariant natural killer T cells. In AACR Annual Meeting.
31. Aspeslagh, S., Y. Li, E. D. Yu, N. Pauwels, M. Trappeniers, E. Girardi, T. Decruy, K. Van
Beneden, K. Venken, M. Drennan, L. Leybaert, J. Wang, R. W. Franck, S. Van Calenbergh, D.
M. Zajonc, and D. Elewaut. 2011. by an induced fit of CD1d prevent tumour metastasis. EMBO
J. 30: 2294–2305.
32. Aspeslagh, S., M. Nemčovič, N. Pauwels, K. Venken, J. Wang, S. Van Calenbergh, D. M.
Zajonc, and D. Elewaut. 2013. Enhanced TCR Footprint by a Novel Glycolipid Increases NKT-
Dependent Tumor Protection. J. Immunol. 191: 2916–2925.
33. Zhang, Y., R. Springfield, S. Chen, X. Li, X. Feng, R. Moshirian, R. Yang, and W. Yuan.
2019. α-GalCer and iNKT cell-based cancer Immunotherapy: Realizing the therapeutic
potentials. Front. Immunol. .
34. Kim, S., and S. Lalani. 2008. Glycolipid ligands of invariant natural killer T cells as vaccine
adjuvants. Expert Rev. Vaccines 7: 1519–1532.
35. Scheuplein, F., A. Thariath, S. Macdonald, A. Truneh, and R. Mashal. 2013. A Humanized
Monoclonal Antibody Specific for Invariant Natural Killer T ( iNKT ) Cells for In Vivo Depletion.
PLoS One 8: 1–11.
36. Das, R., F. Scheuplein, P. Guan, R. Schaub, and K. E. Nichols. 2015. Cancer
immunotherapeutic potential of NKTT320, a novel human invariant natural killer T-cell activating
monoclonal antibody. In AACR 106th Annual Meeting.
37. Lasek, W., R. Zagoz, and M. Jakobisiak. 2014. Interleukin 12 : still a promising candidate for
tumor immunotherapy ? Cancer Immunol. Immunother. 63: 419–435.
38. Robertson, M. J., and J. Ritz. 1996. Interleukin 12 : Basic Biology and Potential Applications
in Cancer Treatment. Oncologist 1: 88–97.
39. Brigl, M., L. Bry, S. C. Kent, J. E. Gumperz, and M. B. Brenner. 2003. Mechanism of CD1d-
restricted natural killer T cell activation during microbial infection. Nat. Immunol. 4: 1230–1237.
-51-
40. Atz, M. E., and E. F. Reed. 2014. Role of Anti-MHC Class I Antibody in Facilitating
Transplant Accommodation. Crit. Rev. Immunol. 6: 485–511.
41. Valenzuela, N. M., and E. F. Reed. 2015. Antibodies to HLA Molecules Mimic Agonistic
Stimulation to Trigger Vascular Cell Changes and Induce Allograft Injury. Curr. Transplant.
Reports 2: 222–232.
42. Tarazona, R., O. Delarosa, E. Peralbo, J. G. Casado, J. Peña, and R. Solana. 2003. Human
NKT cells in health and disease. Immunologia 22: 359–370.
43. Kinjo, Y., N. Kitano, and M. Kronenberg. 2013. The role of invariant natural killer T cells in
microbial immunity. J. Infect. Chemother. 19: 560–570.
44. Rao, P., X. Wen, J. H. Lo, S. Kim, X. Li, S. Chen, X. Feng, O. Akbari, and W. Yuan. 2018.
Herpes simplex virus-1 specifically targets human CD1d antigen presentation to enhance its
pathogenicity. J. Virol. 92: 1–12.

Abstract (if available)

Abstract NKT cells are named due to its sharing cellular makers and biological properties of both T cells and NK cells, within which a group expressing only a limited choice of αβ TCR are called iNKT cells. Different from conventional T cells, iNKT cells recognize lipid-based antigen presented by CD1d molecules on professional APCs with one of the well-studied lipid-based antigens being α-GalCer. As the bridge between innate immunity and adaptive immunity, iNKT cells, once activated, can effectively kill tumor cells both indirectly and directly. iTCR recognition of the endogenous lipid presented by the CD1d on tumor cells can lead to the secretion of several direct cell killing proteins like granzyme B and perforin, which have been shown to be important for the tumor suppression activities of regulatory T cell. Also, the Th-1 pro-inflammatory cytokines produced by iNKT cells can mediate the activation of multiple other tumor-killing cells like NK cells and CD8+ T cells. In this process, the suppressive tumor microenvironment is maintained by the introduction of positive feedback loops for DCs and NK cells activation and anti-tumor cytokine cascade. The potent anti-tumor activities of iNKT cells have made it a favored target for future cancer therapies designs. This study has explored different novel approaches of mobilizing human iNKT cells for cancer immunotherapies. An α-GalCer analog, NU-α-GalCer, though suggesting an improvement in activating iNKT cells in other studies, had no significantly better performance based on the B16F10 melanoma challenges in this study. Other than that, two monoclonal antibodies, anti-iTCR antibody NKTT320, and anti-CD1d antibody 51.1.3 were studied for their agonist function in the CD1d/iNKT cells system. As a result, NKTT320 mAb failed to show anti-metastatic activities, yet the inclusion of this mAb has enabled a better performance of mIL-12. The most encouraging discovery of this study would be that of potent iNKT cells agonist, anti-CD1d antibody 51.1.3. In the melanoma challenge, without any adjuvant, a single administration of 51.1.3 mAb was able to introduce a significant decrease of the lung nodule numbers (p=0.0069). Together, these studies have supported the rationale of continuously searching for better iNKT cells agonists and have provided useful information for future iNKT cells study and potential agents choices for clinical trials.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Comparative studies of coronavirus interaction with CD1d-restricted iNKT cells

PDF

Characterization of invariant natural killer T cells in a novel humanized HBV-transgenic model

PDF

Development of TCR-like antibody and novel chimeric antigen receptor for cancer immunotherapy

PDF

SARS-CoV-2 suppression of CD1d expression and NKT cell function

PDF

The role of envelope protein in SARS-CoV-2 evasion of CD1d antigen presentation pathway

PDF

The interaction of SARS-CoV-2 with CD1d/NKT antigen presentation pathway

Asset Metadata

Creator Li, Xin (author)

Core Title Novel approaches of mobilizing human iNKT cells for cancer immunotherapies

School Keck School of Medicine

Degree Master of Science

Degree Program Molecular Microbiology and Immunology

Publication Date 07/23/2019

Defense Date 05/29/2019

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag 51.1.3,B16F10 metastasis assay,BMDCs,cancer immunotherapies,CD1d,iNKT cells,ITCR,mIL-12,monoclonal antibody,NKTT320,NU-α-GalCer,OAI-PMH Harvest,α-GalCer

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Yuan, Weiming (committee chair), Liang, Chengyu (committee member), Tahara, Stanley (committee member)

Creator Email Xin.Li-2.GR@dartmouth.edu,xli102@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-185168

Unique identifier UC11660421

Identifier etd-LiXin-7571.pdf (filename),usctheses-c89-185168 (legacy record id)

Legacy Identifier etd-LiXin-7571.pdf

Dmrecord 185168

Document Type Thesis

Format application/pdf (imt)

Rights Li, Xin

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA