University of Southern California Dissertations and Theses

Enhancing the anti-cancer specificity of chimeric antigen receptor T cells through targeting HLA loss

(USC Thesis Other)

Enhancing the anti-cancer specificity of chimeric antigen receptor T cells through targeting HLA loss

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

ENHANCING THE ANTI-CANCER SPECIFICITY OF CHIMERIC ANTIGEN
RECEPTOR T CELLS THROUGH TARGETING HLA LOSS

by

Nan Jiang

A Thesis Presented to the
FACULTY OF THE USC School of Pharmacy
University of Southern California
In Partial Fulfillment of the
Requirements for the Degree
Master of Science
(Pharmaceutical Sciences)

August 2021

Copyright 2021 Nan Jiang

ii
Table of Contents
Acknowledgements ........................................................................................................................ iv
List of Figures ................................................................................................................................. v
List of Tables ................................................................................................................................ vii
List of Abbreviations ................................................................................................................... viii
Abstract ........................................................................................................................................... x
Chapter 1 Introduction .................................................................................................................... 1
1.1 Acute myeloid leukemia and its routine treatment .......................................................... 1
1.2 Non-Small Cell Lung Cancer .......................................................................................... 3
1.3 Chimeric Antigen Receptor (CAR) T cell therapy .......................................................... 4
1.3.1 CAR T cell therapy review ...................................................................................... 4
1.3.2 Chimeric Antigen Receptor Structure ..................................................................... 6
1.3.3 Development of chimeric antigen receptor T cell therapy ...................................... 7
1.3.4 CAR-T cells’ applications in AML ......................................................................... 9
1.3.5 CAR-T cells for treating NSCLC .......................................................................... 10
1.3.6 Limitation of CAR-T therapy ................................................................................ 11
1.3.7 Strategies for overcoming the side effects of CAR-T therapy .............................. 13
1.4 Inhibitory Chimeric Antigen Receptor (iCAR) T cell therapy ...................................... 14
1.4.1 inhibitory CAR ...................................................................................................... 14
1.4.2 Introduction of HLA class I loss in NSCLC .......................................................... 16
1.4.3 Introduction of HLA-DR (HLA class II) Loss in AML ........................................ 18
1.4.4 Limitations in previous work ................................................................................. 19
1.5 Purpose and significance ............................................................................................... 20
1.5.1 Generation of an inhibitory CAR targeting HLA-DR loss in AML ...................... 20
1.5.2 Generation of an inhibitory CAR targeting MHC class I loss in NSCLC ............. 20
Chapter 2 Enhancing the anti-CD33 CAR-T cell specificity against AML by targeting HLA class
II .................................................................................................................................................... 22
2.1 Functional testing of anti-CD33 CAR and anti-HLA-DR CAR in Jurkat cells ............ 22
2.2 Construction and functional verification of dual CARs by one-step transduction ........ 25
2.2.1 Plasmids construction ............................................................................................ 25

iii
2.2.2 Functional validation of dual CAR by one step transduction in Jurkat cells ......... 27
2.3 Re-construction of plasmids for one-step transduction by switching the order of anti-
CD33 activating CAR and anti-HLA-DR inhibitory CAR ................................................. 31
2.3.1 Structure of plasmids ............................................................................................. 31
2.3.2 Plasmids construction ............................................................................................ 31
2.3.3 Basic knowledge of PBMC ................................................................................... 31
2.3.4 Functional validation of dual CAR by one step transduction in PBMC cells ....... 32
Chapter 3 Enhancing the anti-EGFR CAR-T cell specificity by targeting HLA class I .............. 36
3.1 Construction of an anti-EGFR activating CAR and an anti-β2m inhibitory CAR ........ 36
3.2 Functional verification of the anti-EGFR activating CAR-Jurkat cells ........................ 39
3.3 Functional verification of anti-EGFR and anti-β2m dual CAR in Jurkat cells ............. 41
3.4. Functional verification of dual anti-CD19 activating CAR and anti-β2m inhibitory
CAR ..................................................................................................................................... 43
Chapter 4 Discussion .................................................................................................................... 45
4.1 Anti-CD33-28ζ, anti-HLA-DR-PD-1 dual CAR for AML ........................................... 45
4.2 Anti-EGFR-4-1BBζ, anti-β2m-PD-1 dual CAR for lung cancer .................................. 47
Materials and Methods .................................................................................................................. 49
References: .................................................................................................................................... 52

iv
Acknowledgements
First and foremost, I would like to express my sincere gratitude to my supervisor, Prof.
Jianming Xie, for his good instructions during my research careers and useful suggestions on my
thesis.
Second, I would like to express my heartfelt gratitude to my committee, Dr. Okamoto and
Dr. Zaro. Thank you for taking the time to review my thesis.
I really appreciate my instructors, Dr. Fan Fei, and Dr. Liang Rong. They have helped me
a lot in my life and studies. Without their help, I cannot finish my research successfully. Hope
they have a bright future.
Special thanks would be given to my parents and friends for their continuous support and
encouragement.

v
List of Figures
Figure 1. CAR structure .................................................................................................................. 5
Figure 2. Generations of CAR T cells ............................................................................................ 8
Figure 3. “On-target, off-tumor” effect......................................................................................... 12
Figure 4. Dual CAR activation: one antigen binding cannot activate CAR-T cells ..................... 14
Figure 5. Scheme diagram of PD-1 and CTLA-4 based inhibitory CAR targeting PSMA .......... 15
Figure 6. Scheme diagram of anti-CD19 CAR-T cells bearing a KIR/PD-1 inhibitory CAR...... 17
Figure 7. Scheme diagram of dual CAR recognizing HLA allele ................................................ 17
Figure 8. Cell staining and cell activation of single CAR- and dual CAR-Jurkat cells ................ 24
Figure 9. Construction of a single lentiviral vector encoding both an anti-CD33 CAR and an anti-
HLA-DR iCAR ............................................................................................................................. 26
Figure 10. Staining and activation assays of Jurkat cells expressing the anti-CD33 CAR with or
without the anti-HLA-DR iCAR ................................................................................................... 29
Figure 11. Construction of a new anti-CD33 CAR and testing of its expression and activation in
human PBMC................................................................................................................................ 34
Figure 12. Expression of anti-CD33-28ζ-P2A-anti-HLA-DR-PD-1 on Jurkat cells .................... 35
Figure 13. Construction of an anti-EGFR CAR and an anti- 2m iCAR ...................................... 38
Figure 14. Generation and activation of anti-EGFR CAR Jurkat cells......................................... 40
Figure 15. Generation and functional testing of Jurkat cells expressing an anti-EGFR CAR with
or without an anti-β2m iCAR ....................................................................................................... 42

vi
Figure 16. Generation and testing of Jurkat cells expressing an anti-CD19 CAR with or without
an anti-β2m iCAR ......................................................................................................................... 44

vii
List of Tables
Table 1. HLA-DR expression in clinic ......................................................................................... 19

viii
List of Abbreviations
AML Acute Myeloid Leukemia
HLA human leukocyte antigen
TAA tumor-associated antigens
FAB French-American-British classification
WHO World Health Organization
HSCT Hematopoietic Stem Cell Transplantation
BBB blood brain barrier
ATRA all-trans-retinoicacid
APL acute promyelocytic leukemia
NSCLC non-small-cell lung carcinoma
FDA Food and Drug Administration
ORR overall response rate
ADC adenocarcinoma
SqCC squamous-cell lung carcinoma
SCLC large-cell lung carcinoma
EGFR epidermal growth factor receptor
β2m beta 2 microglobulin
mTOR mammalian target of rapamycin
CAR chimeric antigen receptor
APC antigen presenting cells
IL-2 interleukin-2
TNF-α tumor necrosis factor-α

ix
IFN-γ interferon-γ
scFv single chain variable fragment
pMHC peptide-Major Histocompatibility Complex
CRS cytokine release syndrome
ICANS immune effector cell-associated neurotoxicity syndrome
iCAR inhibitory chimeric antigen receptor
PD-1 programmed cell death protein-1
CTLA-4 cytotoxic T-lymphocyte-associated antigen-4
PSMA prostate-specific membrane antigen
SIGLEC-3 sialic acid binding Ig-like lectin-3
ELISA Enzyme-Linked Immunosorbent Assay
PBMC peripheral blood mononuclear cell
PCR polymerase chain reaction

x
Abstract
Chimeric antigen receptor (CAR) T cells have shown selective and potential efficacy by
targeting tumor-associated antigens (TAA) on cancer cells. However, the practice of CAR-T cell
therapy is limited by the “on-target, off-tumor” effect resulting from shared tumor-antigen on
normal tissues. Since the major histocompatibility complex (MHC) is frequently downregulated
or lost in malignant cells, we propose to reduce the on-target off-tumor toxicity of CAR-T cells
by activating them against TAA while inhibiting them against MHC molecules in normal tissue
cells. Here, we report two inhibitory CARs (iCARs) targeting human leukocyte antigen DR
(HLA-DR), an MHC class II molecule, and β2 microglobulin (β2m), a component of MHC class
I molecules. We first verified that dual CAR-Jurkat cells, which express a CD28/CD3ζ-based
anti-CD33 CAR and a PD-1-based anti-HLA-DR iCAR, preferentially targeted CD33
+
HLA-
DR
neg
cells over CD33
+
HLA-DR
+
cells. We have also put in effort to construct new plasmids for
one-step lentiviral transduction of human primary T cells with the CAR and the iCAR. Second,
we constructed another iCAR targeting β2m. However, we found the dual CAR-Jurkat cells,
which expressed the 4-1BB/CD3ζ-based anti-EGFR CAR and the PD-1-based anti-β2m iCAR,
showed no inhibition effect on EGFR
+
β2m
+
cells compared to EGFR
+
β2m
neg
cells. There are
two possible reasons. One is that the 4-1BB co-stimulatory signal pathway cannot be inhibited
by PD-1. This can potentially be addressed by replacing 4-1BB with CD28 in the anti-EGFR
CAR. The other is that β2m in Jurkat cells is interfering with the function of the anti-β2m iCAR.
We propose to knock out β2m in Jurkat cells in future work

1
Chapter 1 Introduction
1.1 Acute myeloid leukemia and its routine treatment
Acute myeloid leukemia (AML), including many types of non-lymphatic cells, is a blood
cancer. The main feature is that abnormal cells grow fast in bone marrow and blood, which in
turn disturbs hematopoietic function
1
. In addition, cellular heterogeneity, another feature of
AML, results from malignant blood cells arising from different stages of hematopoietic
progenitor cells
2
.
The symptoms of AML, which include infection, bleeding, and intravascular coagulation,
are attributed to the replacement of normal blood cells with leukemic cells. For example,
infection may be caused by low white blood cell counts, whereas a lack of platelets would cause
bleeding
3
.
It is difficult to illustrate the cause and pathogenesis of AML; however, there is also some
progress in this field. Currently, it is believed that AML can be generated from interaction of the
environment with the genetic material from human cells
4
. First, ionizing radiation, for example,
α-ray and γ-ray, is thought to be an important factor that can cause AML. One important
example is that survivors of the nuclear attacks on Hiroshima and Nagasaki were more likely to
develop AML
5
. Second, chemical materials from daily supplies, such as hair dyes
6
, can have a
negative effect on the bone marrow and trigger AML.
Currently, there are two important criteria for the diagnosis of AML. One is the French-
American-British (FAB) classification. The other is the World Health Organization (WHO)
classification. The FAB classification was developed in the 1970s by a group of experts in
France, the U.S., and the U.K. They divided AML into eight different subtypes, M0 to M7, based
on the types of cells that the leukemia develops from and the degree of cell maturity. And it was

2
largely in view of how the leukemia cells looked like under the microscope after being stained
7
.
The FAB classification is useful, but it does not account for some factors, for example the related
gene mutation, known to affect the prognosis. In order to better understand the classification of
AML, WHO updated some factors into the classification system in 2018.
Treatment of AML is usually divided into two phases, where the first one is induction
chemotherapy and the second one is consolidation chemotherapy. Induction chemotherapy aims
to eliminate the leukemia until cancer cells can no longer be detected, known as complete
remission (CR). To cure the patients, consolidation chemotherapy is implemented to remove
hard-to-detect cancerous remains
8
.
Chemotherapy is a type of cancer treatment that abrogates fast-growing cancer cells. In
the past thirty-years of study of chemotherapy, single-agent administration or a combination of
several different medicines are used to treat cancers. The first case of study about treatment on
AML via chemotherapy was in the 1980s. Compared with patients receiving no chemical agents,
those administered with thioguanine and cytarabine represented better duration and their lifespan
were prolonged. Additionally, common chemotherapy drugs include hypomethylating agents,
like 5-azacitidine, and decitabine
8
. However, chemotherapy can also bring short or long-term
side effects on patients after long period treatment, such as neutropenia, leukopenia, and AML
relapse
9
.
Hematopoietic Stem Cell Transplantation (HSCT) can also be used to cure AML and
may often be adopted after chemotherapy. Stem cells are collected from a healthy donor and
injected into patients’ peripheral blood to provide normal blood cells for the patients. Since
HSCT shows more toxicity than immunosuppression therapy and chemotherapy, patients are
carefully selected based on benefit and risk assessments
10
.

3
Other drugs like arsenic trioxide and all-trans-retinoicacid (ATRA) are also a choice for
the treatment of AML. They can destroy AML cells and encourage them to develop into normal
white blood cells. An applicable subtype of AML for these drugs is acute promyelocytic
leukemia (APL)
11
.
However, clinical data showed that some patients with AML relapsed again and became
more severe cases of AML. Besides, chemotherapy, and some anti-tumor drugs not only
suppress tumor cells, but also kill normal cells, causing an adverse effect on the patients’ bodies.
Therefore, looking for a more targeted approach to treat AML is a priority.
1.2 Non-Small Cell Lung Cancer
Lung cancer is the leading cause of cancer-related mortality in United States
12
.
Among various environmental factors that may cause lung cancer, cigarette smoking is
the most predominant factor—85% to 90% of patients were thought to have been smokers. Other
factors, like radon, or toxicity from some kinds of metal, including chromium and asbestos, have
also been proven to have a relationship with the occurrence of lung cancer
12
.
Non-small cell lung cancer (NSCLC) accounts for 85% of all lung cancers, and often
presents symptoms only when reaching an advanced stage. For NSCLC patients, a cough is the
most common symptoms. In addition, hemoptysis and chest pain are also common symptoms. To
provide evidence for a follow-up treatment scheme, doctors use positron-emission-tomography
(PET)-CT so that they can identify the stages of NSCLC in the patient
12
.
The treatment of NSCLC is stage-specific. Surgical excision is often employed for those
patients who are in the early stage and do not have any treatment prohibition. However, for
patients with recurrence and metastasis who have already undergone surgery, adjuvant
chemotherapy would be considered as a treatment option. In addition, chemotherapy has been

4
the primary approach to treat NSCLC patients at the unresectable stage for the last decade, but
only 15% of those patients were still alive after five years. Therefore, immunotherapy has been
introduced in the treatment of NSCLC. Monoclonal antibodies, like durvalumab, a checkpoint
inhibitor that blocks PD-L1, are approved by Food and Drug Administration (FDA). Chimeric
antigen receptor (CAR) T cell therapy has also been extensively studied
12
.

1.3 Chimeric Antigen Receptor (CAR) T cell therapy
1.3.1 CAR T cell therapy review
Chimeric Antigen Receptor T cell immunotherapy, also called CAR-T cell therapy, is a
novel treatment that can target cancer cells precisely. The mechanism is that T cells are
genetically engineered to express a chimeric antigen receptor, CAR, that specifically recognizes
a tumor associated antigen (TAA) in tumor cells. After binding to TAA, CAR-T cells will be
activated and prompted to kill the target cells. There are two main mechanisms for CAR-T cell
cytotoxicity. First, CAR-T cells secrete cytotoxic granules containing perforin and granzymes
via exocytosis. Those granules will induce pore formation on target cells through membrane
fusion. And then facilitate the formation of pro-apoptotic granzymes, which result in the lysis of
the target cell
13
. Second, the overexpressed TNF family ligands on the surface of CAR-T cells
will bind to its related receptors on tumor cells, which would induce the apoptosis of tumor cells
13
. The production of certain cytokines can also enhance the anti-tumor capability of CAR-T
cells. Specific cytokines not only make tumor cells more sensitive to IFN-γ, but they can also
alter the immunosuppression effects within the tumor microenvironment, and thereby increase
the cytotoxicity of CAR-T cells
13,14
.

5

Figure 1. CAR structure: the chimeric antigen receptor is commonly composed of binding domain (scFv), transmembrane
domain (TM), and signal domain (co-stimulatory domain and CD3ζ)

6
1.3.2 Chimeric Antigen Receptor Structure
CAR is usually structured in four different sections, which are: the antigen binding
domain; the hinge and transmembrane domains; and, the intracellular signaling domain (Figure
1). Each section has its specific function related to CAR T cell activation and proliferation.
Researchers could optimize the function of CARs by changing the details of each section of them.
The antigen binding domain of the CAR can recognize and bind to an antigen specifically.
In tradition, the binding domain is composed of a variable heavy chain (VH) and a variable light
chain (VL) of monoclonal antibodies. By connecting a flexible linker, they form a single chain
variable fragment (scFv)
15
. Through the function of scFv, CAR-T cells can recognize specific
antigens on target cells independently of Peptide-Major Histocompatibility Complexes (pMHCs).
Hinge and transmembrane domains bear the burden of connecting the extracellular
domain and the intracellular signaling domain. The hinge domain provides effective flexibility
for CARs to overcome spatial barriers and provides sufficient length for CAR to bind with the
antigen on target cells
16
. It is worth to notice that the different structures or lengths of hinge
domains can also influence the binding between CARs and antigens and affect downstream
signaling. In addition, researchers found that the hinge and transmembrane domains can also
have effects on cytokine production by CAR T cells
17,18
.
CARs are anchored to the cell membrane via a transmembrane domain, typically derived
from a Type-1 transmembrane protein, such as CD3ζ, CD28 and CD8
19
. One thing we need to
know is that different transmembrane domains could affect the function and stability of CAR on
the cell membrane. For example, a CAR with a CD28 transmembrane domain will be more
stable than those have a CD3ζ transmembrane domain
17
.

7
The signaling domain usually has an activation section and one or more co-stimulation
sections. Almost all CARs activate T cells via the CD3ζ-derived immunoreceptor tyrosine
19
.
However, ideally, CD3ζ alone cannot induce the activation of T cells, and the persistence, as
well as activation, of CARs will be limited. So, co-stimulation sections are introduced in the
signaling domain. The most widely used co-stimulation sections now are CD28 and 4-1BB
17
.
Based on the knowledge and use of these four sections of CARs, combined with clinical
data, researchers have expanded the applicability of CAR-T cell immunotherapy towards cancers
by changing the structure of each section to improve safety and reliability of these constructs.

1.3.3 Development of chimeric antigen receptor T cell therapy
The first generation of CAR T cells used CD3 zeta chain (CD3ζ) as the intracellular
domain (signaling domain) to activate the effector cells. To enhance the activation and
persistence of CAR T cells, researchers added one co-stimulatory domain (CD28, 4-1BB, or
OX40) into the intracellular domain, which is considered as the second generation of CAR T
cells. Also, they have developed a third generation of CAR T cells, where two or more co-
stimulatory domain are added in the signaling domain on the basis of the first-generation CAR
20

(Figure 2).

8

Figure 2. Generations of CAR T cells: first-generation CAR T cells contain the CD3ζ signaling domain. The second-generation
CAR T cells have one co-stimulatory domain besides CD3ζ. And two or more co-stimulatory domains are added in the third
generation of CAR T cells.

The first FDA approved CAR-T cell therapy was Tisagenlecleucel (Kymriah), an anti-
CD19 CAR specific for patients who suffer from B-cell acute lymphoblastic leukemia (B-ALL).
According to the clinical data, 81% of 75 children and young adults represented the overall
response rate (ORR), and 60% of them have shown complete remission (CR), which provided an
efficient way to treat chemotherapy-relapse patients. In addition, accompanied by the successful
utilization of Kymriah in the clinic, three additional types of anti-CD19 CAR-T cell therapy
products, axicabtagene ciloleucel (Yescarta), brexucabtagene autoleucel (Tecartus) and
lisocabtagene maraleucel (Breyanzi), have been developed and approved by FDA. 110 patients
with Diffuse Large B-Cell Lymphoma (DLBCL) were enrolled in the clinical trials of Yescarta,
and 83% of them showed a therapeutic effect with 58% of all patients achieving a CR, which is

9
similar in efficacy to the third product, lisocabtagene maraleucel
21,22
. Tecartus, a CD19-targeted
CAR-T cell therapy product, can be used to treat patients with mantle cell lymphoma (MCL)
23
.
Another approved CAR-T therapy is Idecabtagene vicleucel (brand name Abecma),
which is a B-cell maturation agent (BCMA)-specific CAR-T cell therapy. And, it is an efficient
product to treat relapsed and refractory multiple myeloma (RRMM)
24
.

1.3.4 CAR-T cells ’ applications in AML
CD33, also known as SIGLEC-3 (sialic acid binding Ig-like lectin 3), is a type of
transmembrane receptor which is expressed on cells of myeloid lineage. Due to its capability of
binding to sialic acid, CD33 is classified as a member of SIGLEC family of lectins. Additionally,
CD33 is always thought to be myeloid-specific, but it is also found on some types of lymphoid
cells
25
.
Any molecules with sialic acid residues, for example, glycoproteins and glycolipids, can
activate CD33. Upon binding, the immunoreceptor tyrosine-based inhibition motif (ITIM) of
CD33 could be phosphorylated and this would result in a cascade that inhibits the phagocytosis
effect by cells
26
. In addition, CD33 showed a higher surface expression on AML than on other
myeloid cells
27
.
In accordance with high CD33 expression on AML cells and low expression on normal
cells, Kenderian, et al., have constructed CD33-specific CAR-T cells which demonstrated a
significant eradication effect on leukemic cells. Their work displayed obvious de-granulation
(~90%), potent cytotoxicity at low ratios of effect cells to target cells (E:T ratio), extensive
proliferation, and robust cytokine production after co-culture of anti-CD33 CAR-T cells and
target cells (MOLM14). However, sequelae, including human lineage cytopenias and reduction

10
of myeloid progenitors resulted from treatment with anti-CD33 CAR-T cells. To increase the
adaptability of anti-CD33 CAR-T cell in clinic, they use RNA modification to achieve transient
expression of the CAR which make patients avoid the long-term suppression of myeloid cells
28
.
But the hematopoietic toxicity that resulted from utilization of anti-CD33 CAR-T cells acting on
CD33-expressing macrophages and Kupffer cells cannot be ignored. Hence, we aim to develop
an inhibitory CAR-T cell therapy that can alleviate the “on-target, off-tumor” effect.

1.3.5 CAR-T cells for treating NSCLC
The epidermal growth factor receptor (EGFR; ErbB-1; HER1 in humans) is a
transmembrane protein, which is a receptor and member of the epidermal growth factor (EGF)
receptor family
29
. When binding with its ligands, a series of downstream signaling pathways of
EGFR will be activated to promote cell proliferation and growth, block cell apoptosis, and resist
chemotherapy in cancer patient treatments
30
. It is reported that EGFR signaling is involved in
epithelial tissue maintenance and growth, and is also expressed in lung cancer
31
. Researchers
have found overexpression of EGFR on the cell surface of various solid tumors, especially in
non-small-cell lung cancer
32,33
.
Zhou, et al., have established an EGFR-specific (anti-EGFR-CD28-CD3ζ) CAR-T cell to
eliminate NSCLC, which achieved antitumor efficacy both in vivo and in vitro. Their modified T
cells were evaluated to have no or very minimal acute systemic toxicity in animal models
33
.
Besides, Li, et al., also developed anti-EGFR CAR-T cells (with the signaling domain of 4-1BB-
CD3ζ) which showed anti-cancer efficacy in vitro. A 3-fold cytotoxicity of anti-EGFR CAR-T
cells acting on target cells (H23 and H460) compared with control group has been shown up,
which reflected in target cell elimination and cytokine secretion. For specific cytokine secretion,

11
compared with anti-CD19 CAR-T cells (the control group), anti-EGFR CAR-T cells produced
more IL-4 (2- to 3- fold), IL-10 (~10-fold), and TNF-α (>5-fold)
34
.

1.3.6 Limitation of CAR-T therapy
CAR-T cell therapy has made a significant contribution toward the treatment of blood
cancers and some solid tumors. However, it now also seems to have severe side effects on
patients, including cytokine release syndrome, neurotoxicity, and still “on-target, off-tumor”
effects
35,36
.
Cytokine release syndrome, also called CRS, is a systemic inflammatory response caused
by a variety of factors, where CAR-T cell administration is one of them. Importantly, CAR-T
cells play an anti-cancer role by recognizing and binding to the antigens on target cells, which
will strongly activate the T cells to eliminate the target cells. During this procedure, CAR-T cells
rapidly expand, and activate caspase 3 via granzyme B release to cleave gasdermin E (GSDME),
a pore-forming protein expressed on B leukemic cells, which results in extensive pyroptosis of
target cells. Consequently, pyrotosis-released factors will activate caspase 1, which leads to the
cleavage of GSDME for macrophages
37
. At the same time, numerous immune cells besides
macrophages, like monocytes, are activated and appears to be the main reason for causing CRS.
Thus, the CRS reaction cannot only be considered as a side effect of CAR-T cell therapy, but
also as a symbol of the therapeutic effect on a tumor. In fact, CRS induced by CAR-T cells is not
yet completely avoidable. People should pay more attention on patients and give medical
intervention in these cases.
Immune effector cell-associated neurotoxicity syndrome (ICANS) is seldom occurring in
clinics. Its pathophysiology has not been understood completely. Cytokine-mediated toxicity,

12
disruption of blood brain barrier (BBB), and increased permeability of vascular are considered as
the reasons of ICANS, which may cause cerebral edema and death
35,38,39
. If patients suffer from
neurotoxicity during CAR-T cell therapy, hormone treatment like corticosteroids is the first
choice to suppress the disease in clinic
35
.
Also, immune escape caused by heterogeneity of antigen expression on solid tumors, and
hard infiltration of T cells resulted from specific structure of tumor microenvironment (TME)
40

make CAR-T cell therapy difficult in treatment of solid tumors.
“On-target, off-tumor” effect is the most serious toxicity in anti-tumor CAR-T cell
therapy, which resulted from the binding of CAR-T cells and following attack on normal cells
sharing the same antigen with tumor cells. A study
41
was ever found that high dose of CAR-T
cells injected in the patients’ bodies could evoke this toxicity, whereas low dose could not.
Therefore, besides the shared antigen expressed on the normal tissue, whether administrated dose
of CAR-T cells has exceeded the threshold is also supposed to be considered
36
(Figure 3).

Figure 3. “On-target, off-tumor” effect: due to the shared antigens expressed on tumor cells and normal cells, CAR T cells can
also be activated upon antigens on normal cells and eliminate the normal cells.

13

1.3.7 Strategies for overcoming the side effects of CAR-T therapy
The causes of CRS have not been understood completely. So the researchers divided the
patients into different grades according to their symptoms. Since even serious CRS patients
could be cured after treatments, doctors should monitor the development of CRS and be ready to
provide full spectrum of modern critical care
42
.
The CRS Grading scheme is composed of four levels. Patients with low grade CRS
usually receive antihistamines and antipyretics. However, severe CRS displays a life-threatening
situation. Patients with elevated serum IL-6 levels may contribute to the development of CRS
43
.
Therefore, patients who have developed grade three or four CRS toxicity are supposed to be
treated with tocilizumab. After the first dose of tocilizumab for a few hours to two days, CRS
related signs, such as fever, or hypotension, should disappear. Or a second administration of
tocilizumab is available if no obvious results appear after the first dose
42
.
For neurotoxicity caused by CAR-T cell administration, monoclonal antibodies represent
no evident effect on treatment of neurotoxicity due to the hindrance of passage through the
blood-brain-barrier (BBB). Hence, corticosteroids have played a more important role on the
treatment of neurotoxicity due to its good penetration effect of passing through BBB. Necessarily,
corticosteroids are supposed to be avoided as the first line treatment of CRS after getting CAR-T
cell therapy
42
.
To solve the problems in CAR-T cell therapy in solid tumors, Caruso H et al.
44
have
established Nimo-CAR T cells with sensitivity about different density of antigens, which achieve
to distinguish normal cells and tumor cells. Grada Z et al.
45
used bispecific chimeric antigen
receptors to enhance the specificity of immune cells to tumor cells.

14
There are several promising ways to solve the “on-target, off-tumor” effect. To enhance
the specificity of CAR-T cell therapy, Grada, et al., have established bispecific dual CAR-T cells,
which were composed of anti-CD19 CAR and anti-EGFR CAR. The dual CAR-T cells can
specifically be activated against CD19
+
EGFR
+
cells and protect the cells without one of the
target molecules (Figure 4)
45
.

Figure 4. Dual CAR activation: one antigen binding cannot activate CAR-T cells. Only two antigens binding can result in
full activation of CAR T cells.

Another method is to choose a more specific antigen for tumor cells to avoid the damage
to normal cells (Freyer and Porter 2020). Additionally, Caruso, et al., have constructed a type of
CAR-T cell targeting EGFR, which can distinguish normal tissue and cancer cells based on the
density of EGFR expression
44
.
In this study, we looked for an antigen that is expressed on normal cells, but not on tumor
cells. Then, we can design an inhibitory CAR to work in conjunction with an active CAR to help
enhance the specificity of the CAR-T cells.

1.4 Inhibitory Chimeric Antigen Receptor (iCAR) T cell therapy
1.4.1 inhibitory CAR

15
It has been reported that immune checkpoints in T cells play a crucial role of maintaining
self-tolerance and modulating the duration of the cells to reduce the damage of normal tissue.
Programmed cell death protein 1 (PD-1) and Cytotoxic T-lymphocyte-associated antigen 4
(CTLA-4), two currently well-studied immune checkpoints expressed on the surface of immune
cells, are important in immune response regulation, especially in T cell response termination
46
.
Fedorov et al. have constructed PD-1 and CTLA-4-based inhibitory CAR (iCAR) targeting
Prostate-Specific Membrane Antigen (PSMA) in 2013 to enhance the specificity of CAR-T
therapy
47
(Figure 5).

Figure 5. Schematic diagram of PD-1 and CTLA-4 based inhibitory CAR targeting PSMA.
Likewise, inhibitory CAR also consists of an scFv, a transmembrane domain, and a
signaling domain. In contrast to the activating CAR, the inhibitory CAR provides an inhibitory
signal to affect cells when binding to its antigen. Hui et al. reported that CD28, the costimulatory
receptor of activating CAR, is a primary target for PD-1-mediated inhibition in T cells
48
. Also,
many studies support that T cell receptor (TCR) and its downstream signal molecules are the
primary target of PD-1, whereas other studies consider both TCR and CD28 equally as the target
for PD-1
49-51
. Therefore, the PD-1 intracellular domain of the iCAR is expected to inhibit CD28
and CD3ζ mediated signaling of the activating CAR, which then protects the target cells (usually

16
normal tissue). Meanwhile, tumor cells with loss or downregulation of the antigen targeted by
the inhibitory CAR are expected to be eliminated.

1.4.2 Introduction of HLA class I loss in NSCLC
Human leukocyte antigen (HLA) system is a group of proteins encoded by MHC genes in
humans. It can be classified as HLA class I and HLA class II. HLA class I is expressed in all
human nucleated cells, and it bears the function of immune recognition.
HLA class I loss is commonly seen in tumor which can escape from cytotoxic T cells
attack. However, this feature could be used in immunotherapy. Recently, Tao L et al. (Figure 6)
reported that anti-CD19 CAR-T cells bearing a KIR/PD-1 inhibitory CAR could eliminate
CD19
+
HLA-C1
-
malignant B cells whereas protecting the CD19
+
HLA-C1
+
B cells. Their
results indicated a promising strategy for preventing the “on-target, off-tumor” effect
52
. Another
team focused on the clonal loss of heterozygosity (LOH) in cancer DNA, and they designed a
dual CAR that can recognize a pair of HLA. If both allele genes are normal, the dual CAR-T
cells will be inactive; if one HLA allele is lost, dual CAR-T cells will be activated to eliminate
the target cells. This targeting strategy not only provided an approach to treat cancer cells,
including breast, colon, brain, as well as lung cancers, but also offered a tactic to reflect the HLA
allele loss, which has been considered as a negative clinical correlate in immunotherapies
18

(Figure 7).

17

Figure 6. Schematic diagram of anti-CD19 CAR-T cells bearing a KIR/PD-1 inhibitory CAR.

Figure 7. Schematic diagram of dual CAR recognizing HLA allele. (Hwang S et al., 2021)

18
It is reported that 25% to 94% of NSCLC patients downregulate HLA class I molecules
53
.
Concerning the use of HLA class I loss in KIR/PD-1 inhibitory CAR in malignant B cells, we
can also use this feature to construct an anti-HLA class I CAR-T cell in the treatment of NSCLC.

1.4.3 Introduction of HLA-DR (HLA class II) Loss in AML
HLA-DR belongs to HLA class II, which is expressed in a few specialized cell types,
including dendritic cells, B cells, and macrophages. Like HLA class 1 loss in solid tumors, HLA-
DR (HLA class II) loss is common in immune cell cancers. It is reported that, among 117 cases
that were identified as acute myeloid leukemia (AML), 28 of them (~24%) had nearly complete
loss of HLA-DR expression on the cancerous cells. Other studies have shown HLA-DR loss in
15% to 17% cases
54,55
(Table 1).

19

Table 1. HLA-DR expression in clinical cases. (Higashi M et al, 2015)
In order to expand the application of inhibitory CAR, we intend to utilize HLA-DR (HLA
class II) loss to establish inhibitory CAR-T cells for enhancement of specificity of
immunotherapy.

1.4.4 Limitations in previous work
Previous studies have shown efficient efficacy of designing and constructing an
inhibitory CAR to enhance the specificity of CAR-T cell therapy. However, they also exposed
limitations in this approach.
(1) Anti-PSMA iCAR constructed by Fedorov, et al., lacks feasible antigens for the
inhibitory CAR. For example, PSMA is a TAA, but not an antigen on the surface of normal cells.

20
(2) Anti-CD19, anti-HLA-A/C dual CAR T cells established by Tao, et al., is also limited
in use by its antigen selection. HLA-A or HLA-C are expressed universally on normal cells, even
in T cells, which may cause adherence between the inhibitory CAR and cells in normal tissue,
resulting in a potential reduction of CAR-T cell trafficking in vivo. This paper also did not show
the inhibitory effect of dual CAR targeting HLA-C positive cells in the xenograft mouse model
in vivo.
(3) The study of Hwang, et al., also has several limitations. First, the near-ubiquitous
expression of HLA may cause a risk of adherence between dual activating/inhibitory CAR
targeting HLA and normal cells in endothelium and other tissues. Thus, T cell trafficking is
probably reduced in vivo. Second, dual CAR targeting antigens expressed on all normal cells
could conceivably result in tonic signaling by both receptors.

1.5 Purpose and significance
1.5.1 Generation of an inhibitory CAR targeting HLA-DR loss in AML
In this study, we aim to establish dual CAR-Jurkat and CAR-T cells that specifically
recognize HLA class II (inhibitory CAR) and CD33 (activating CAR) to enhance the specificity
of immune cells in anti-AML activity. Secondly, both CARs will be introduced in Jurkat cells
simultaneously with the help of a self-cleavage 2A peptide by one-step transduction. After
checking CAR expression and CAR activation by way of cell staining and Enzyme-Linked
Immunosorbent Assay (ELISA), we propose to introduce the CAR into PBMC cells so that we
can prepare to test killing of leukemia cells by CAR-T cells in xenograft mouse models.

1.5.2 Generation of an inhibitory CAR targeting MHC class I loss in NSCLC

21
We also constructed dual CAR-Jurkat cells containing an activating CAR specific for
EGFR and an inhibitory CAR for MHC class I, to investigate its anti-tumor selectivity. We first
expressed the anti-EGFR activating CAR in Jurkat cells. After that, an ELISA assay was carried
out to ascertain the activation of the CAR by detecting IL-2 secretion from a co-culture of CAR-
Jurkat cells and target cells. Second, the second CAR, an anti-β2m inhibitory CAR, was
introduced into the previous activating CAR-Jurkat cells. Consequently, an ELISA assay was
implemented again to probe the specificity of dual CAR-Jurkat cells on NSCLC with MHC class
I loss.

22
Chapter 2 Enhancing the anti-CD33 CAR-T cell specificity against
AML by targeting HLA class II
2.1 Functional testing of anti-CD33 CAR and anti-HLA-DR CAR in
Jurkat cells
Previously, our lab created two plasmids, one encoding an anti-CD33 activating CAR,
and the other, an anti-HLA-DR inhibitory CAR. The activating CAR consists of a CD8 leader
sequence, an HA epitope-tag, an anti-CD33 scFv, an IgG4-Fc hinge domain (IgG4h), a CD28
transmembrane domain, and CD28-CD3ζ intracellular domains (anti-CD33-28ζ). The inhibitory
CAR consists of a CD8 leader sequence, a FLAG-tag, an anti-HLA-DR scFv, and the PD-1
hinge, transmembrane, and intracellular domains (anti-HLA-DR-PD-1). The HA- and FLAG
epitope-tags were used for verifying the expression of both CARs.
The activating CAR, anti-CD33-28ζ, and the inhibitory CAR, anti-HLA-DR-PD-1, were
transduced into Jurkat cells in a two-step process. By a cell staining assay, the single CAR-
positive Jurkat cells reached ~60% of all cells, and the double positive CAR-Jurkat cells
accounted for 75.2% of all cells (Figure 8A). The target cells used in this experiment were HL60,
an HLA-DR-negative and CD33-positive cell line, and KG1, an HLA-DR-positive and CD33-
positive cell line.
Then, we examined the activation and inhibition of both CAR-Jurkat cells against
different target cells based on IL-2 production, measured by ELISA. As shown in Figure 8B,
against HL60 cells, single and dual CAR-Jurkat cells displayed similar IL-2 secretion (single:
~18 pg/mL, dual: ~22 pg/mL). However, against KG1 cells, dual CAR-Jurkat cells had a 70%
reduction of IL-2 secretion compared with single CAR-Jurkat cells. These results indicated that

23
the inhibitory CAR could inhibit the activation of dual CAR-T cells when binding to its antigen,
HLA-DR.
These experiments proved that the dual CAR system worked well in Jurkat cells, a
human T cell line. Next, we wanted to verify the dual CAR system in primary T cells from
human peripheral blood mononuclear cells (PBMC). However, with regard to the limited
survival period and persistence of PBMC, it is not a good way to transduce PBMC twice to
express both CARs. Hence, we intended to link two CARs via a self-cleavage peptide, p2A, and
try to achieve dual CAR transduction in one step.

24

Figure 8. Cell staining and cell activation of single CAR- and dual CAR-Jurkat cells. (A) Flow cytometric analysis of
transduced and un-transduced Jurkat cells. The anti-CD33 CAR was stained with a PE-labeled anti-HA tag antibody, and the
HLA-DR CAR was stained with an allophycocyanin (APC)-labeled anti-FLAG tag antibody. (N = 3). (B) The production of IL-2
for dual CAR Jurkat cells and single CAR Jurkat cells tested by ELISA. (N = 3).

25
2.2 Construction and functional verification of dual CARs by one-
step transduction
We constructed a plasmid encoding both CARs linked by a self-cleavage peptide, p2A. A
map of the former inhibitory CAR and the latter activating CAR is shown in Figure 9A. The
structure of both CARs has been described previously, except that the leader sequence was
changed from the CD8 leader sequence to the GM-CSF leader sequence. We also inserted a
Furin cleavage site upstream of p2A allowing removal of 2A residues attached to the C-terminus
of the former CAR (Figure 9A). For a negative control, we also constructed an anti-HLA-DR
inhibitory CAR with only the PD-1 hinge and transmembrane domains, but no intracellular
domain, named anti-HLA-DR-PD-1TM.
2.2.1 Plasmids construction
We first amplified DNA fragments of both inhibitory and activating CARs including
furin-p2A peptides via polymerase chain reaction (PCR) technology and verified the construct by
agarose gel electrophoresis (Figure 9B). Then, the DNA fragment of the inhibitory CAR
followed by Furin-p2A was digested with (BamH I, Nhe I), and the DNA fragment of the
activating CAR was digested with (Nhe I, EcoR I). Both DNA fragments were digested and
ligated with a pCDH linear vector (BamH I, EcoR I) to construct pCDH-anti-HLA-DR-PD-1-
Furin-P2A-anti-CD33-CD28-CD3ζ (referred to as DR-CD33-IgG4h). For the negative control,
we also constructed pCDH-anti-HLA-DR-PD-1-Furin-P2A-anti-CD33-CD28-CD3ζ (referred to
as TM-CD33-IgG4h). All the positive colonies were verified by colony PCR and sequencing
analysis (GENEWIZ) (Figure 9C, D).

26

Figure 9. Construction of a single lentiviral vector encoding both an anti-CD33 CAR and an anti-HLA-DR iCAR. (A)
Schematic diagram of anti-HLA-DR-PD-1-Furin-P2A-anti-CD33-CD28-CD3ζ. (B) Results of DNA electrophoresis. Lane 1:
CD8 Leader-FLAG tag-anti-HLA-DR-PD-1-(hinge, TM and Intra)-Furin-P2A; Lane 2: CD8 Leader-FLAG tag-anti-HLA-DR-
PD-1-(hinge and TM)-Furin P2A; Lane 3: CD8 leader-HA tag-anti-CD33-IgG4h-CD28-CD3ζ. M represents marker (1 kb plus
DNA ladder). (C, D) The colony PCR displays results of full-length DNA fragments. M represents marker (1 kb plus DNA
ladder). C: the bands in lane 3, 6 and 7 with red rectangles show correct DNA fragments of pCDH-anti-HLA-DR-PD-1TM-
Furin-P2A-anti-CD33-CD28-CD3ζ. D: bands in lane 5, 6, and 8 with red rectangles indicate correct DNA fragments of pCDH-
anti-HLA-DR-PD-1-Furin-P2A-anti-CD33-CD28-CD3ζ.

27
2.2.2 Functional validation of dual CAR by one step transduction in Jurkat cells
The expression of dual CAR on Jurkat cells was verified by cell staining. Since the
FLAG-tag and HA-tag are both theoretically expressed in the CAR, we stained the transduced
cells with an APC-labeled anti-FLAG antibody, plus an anti-HA antibody, and a phycoerythrin
(PE)-conjugated anti-Fc antibody to stain the transduced cells and examine CAR expression by
flow cytometric analysis (Figure 10A). The dual CAR expression of DR-CD33-IgG4h reaches
40% of all cells, and the dual CAR expression of TM-CD33-IgG4h is about 25% of all cells,
which will be used in subsequent experiments.
Then, we used an IL-2 ELISA to verify the function of dual CAR (Figure 10C). However,
we found the activation of dual CAR-Jurkat cells was very low, which showed no significant
differences with un-transduced Jurkat cells (plain) in the HL60 target cell group. In the KG1
target cell group, the IL-2 level secreted by DR-CD33-IgG4h CAR-Jurkat cells were about 10%
lower than TM-CD33-IgG4h CAR-Jurkat cells, much lower than the 70% reduction we observed
earlier with the two-step transduction experiment. This phenomenon might be due to the low
percentage of CAR-Jurkat cells used in the experiments. Hence, we enriched CAR-Jurkat cells
by magnetic cell sorting.
After cell sorting, the dual CAR expression of DR-CD33-IgG4h reaches to 74% of all
cells and the dual CAR expression of TM-CD33-IgG4h is about 67% of all cells (Figure 10B).
However, neither of them showed any significant activation compared to plain Jurkat cells based
on the IL-2 ELISA (Figure 10D).
We postulated that the low level of IL-2 secretion could be due to the low activation of
anti-CD33-28ζ. If this activating CAR was not cleaved by p2A peptide efficiently, it would have

28
a lower expression level compared to the former inhibitory CAR. In this case, the overall
activation of dual CARs may be too low for analysis.

29

Figure 10. Staining and activation assays of Jurkat cells expressing the anti-CD33 CAR with or without the anti-HLA-DR
iCAR. (A) Flow cytometric analysis of transduced and un-transduced (plain control) Jurkat cells. The CD33 CAR was stained
with a PE-labeled anti-HA tag antibody, and the HLA-DR CAR was stained with an APC-labeled anti-FLAG tag antibody. (N =
3) (B) Flow cytometric analysis of sorted DR-CD33-IgG4h, TM-CD33-IgG4h dual CAR Jurkat cells. (C) IL-2 secretion of TM-
CD33-IgG4h and DR-CD33-IgG4h CAR unsorted Jurkat cells in response to KG1 and HL60 target cell lines at an E:T ratio of
2:1 in 37℃ incubators overnight. The supernatant of the cell culture was collected and the concentration of IL-2 was measured
by ELISA. (N = 3) (D) IL-2 production stimulated by sorted TM-CD33-IgG4h and sorted DR-CD33-IgG4h CAR Jurkat cells

30
when co-culturing with KG1 and HL60 target cell lines at an E:T ratio of 1:1 overnight at 37℃. The supernatant was collected to
examine the IL-2 concentration by ELISA. (N = 3).

31
2.3 Re-construction of plasmids for one-step transduction by
switching the order of anti-CD33 activating CAR and anti-
HLA-DR inhibitory CAR
2.3.1 Structure of plasmids
To enhance the expression of the activating CAR (anti-CD33 CAR), we designed a new
pair of plasmids by switching the order of inhibitory CAR and activating CAR, which meant the
new plasmid contained two CARs and the anti-CD33 CAR was in front of the inhibitory CAR
linked by p2A. We also changed the leader sequence of the activating CAR from the GM-CSF
leader to the CD8 leader to ensure that there are no repeated DNA sequences for better virus
packaging (Figure 11A).
2.3.2 Plasmids construction
We first amplified the DNA fragments of the anti-CD33 CAR and the p2A-anti-HLA-DR
CAR, then digested them with (BamH I, Nhe I), and (Nhe I, EcoR I), respectively. Both DNA
fragments were ligated with a pCDH linear vector (BamH I, EcoR I) to construct pCDH-anti-
CD33-CD28-CD3ζ-P2A-anti-HLA-DR-PD-1 (referred to as CD33-IgG4h-DR). For a negative
control, we also constructed pCDH-anti-CD33-CD28-CD3ζ-P2A-anti-HLA-DR-PD-1 (without
the intracellular domain), (referred to as TM-CD33-IgG4h) (Figure 11B). Then, we implemented
colony PCR technology to examine positive colonies of CD33-IgG4h-DR and CD33-IgG4h-TM
(Figure 11C and 11D). Sequencing was then performed for validation (GENEWIZ).
2.3.3 Basic knowledge of PBMC
A peripheral blood mononuclear cell (PBMC) is every type of cell with a round nucleus
in peripheral blood. Those cells are composed of T and B cells (~80%), natural killer cells

32
(~10%) and monocytes (~10%)
56
. Considering the potential practical use in clinics, we chose to
verify our new constructs’ expression in PBMCs.

2.3.4 Functional validation of dual CAR by one step transduction in PBMC cells
Primary T cells in human PBMCs were activated with anti-CD3z/anti-CD28 beads, and
then transduced with lentivirus containing the new construct. By cell staining, we attempted to
detect the expression of CD33-IgG4h-DR (PBMC DR) and CD33-IgG4h-TM (PBMC TM)
(Figure 11E). Surprisingly, we could only observe the expression of anti-CD33 activating CAR,
but not anti-HLA-DR inhibitory CAR.
When the CAR-PBMC proliferated, we further implemented an IL-2 ELISA to ascertain
the cell activation, as shown in Figure 11F. In this assay, the target cells we used were MOLM13
cells, a CD33-positive and HLA-DR-negative cell line, and MOLM13-DR, a CD33-positive and
HLA-DR-positive cell line. Both PBMC TM and PBMC DR produced more than 100 pg/mL IL-
2 in the co-culture medium after 24 hours incubation. However, we cannot see any inhibition
effect, which is consistent with the absence of inhibitory CAR expression on the cell surface.
To address the low expression level of the latter CAR on the plasmid of one step
transduction on PBMC, we repeated the lentiviral transduction in Jurkat cells to verify the dual
CAR expression again (Figure 12A). The results of cell staining showed that both the activating
CAR and the inhibitory CAR were expressed on Jurkat cells. We considered to further optimize
our constructs to ensure the dual CAR are properly expressed in PMBC. After reviewing other
literature
18,52
on one-step transduction of dual CARs, we decided to reduce the length of the
entire structure to achieve better lentiviral packaging efficiency and dual CAR expression. We
previously used an IgG4-Fc hinge domain for anti-CD33 CAR which is 690 bp. Next, we

33
planned to use a shorter CD8α hinge (141 bp) to link the anti-CD33 scFv with the CD28
transmembrane domain in the activating CAR. This would also decrease the spacer distance
between CAR-T cells and target cells, resulting potentially in a higher potency of activation
57
.
This work is currently underway while the thesis is being prepared.

34

Figure 11. Construction of a new anti-CD33 CAR and testing of its expression and activation in human PBMC. (A)
Schematic diagram of anti-CD33-IgG4 hinge-Fc-CD28-CD3ζ P2A anti-HLA-DR-PD-1 CAR (referred to as CD33-IgG4h-DR)
and anti-CD33-IgG4 hinge-Fc-CD28-CD3ζ P2A anti-HLA-DR-PD-1 (without signal domain) CAR (referred to as CD33-IgG4h-
TM): Anti-HA and anti- FLAG tag antibodies were used for cell staining in the verification of CAR-expression. (B-D) B: DNA
gel electrophoresis results: M represents DNA marker (1kb plus DNA ladder). Lane 1 and lane2: HA tag-anti CD33-IgG4h-28ζ.
Lane 3 and 4, lane 5 and 6: P2A-FLAG tag-anti HLA-DR-PD-1 (hinge, TM, Intra), and P2A-FLAG tag-anti HLA-DR-PD-1
(hinge, TM), respectively. C: lanes 1 to 6 show DNA fragments of HA tag-anti CD33-IgG4h-28ζ-P2A-FLAG tag-anti HLA-DR-
PD-1 (hinge, TM, Intra). D: lanes 1 to 2 display HA tag-anti CD33-IgG4h-28z-P2A-FLAG tag-anti HLA-DR-PD-1 (hinge, TM).
(E) Flow cytometric analysis of transduced and un-transduced Jurkat cells. The expression of CD33 CAR and the HLA-DR
iCAR were verified by cell staining with a PE-labeled anti-HA tag antibody and an APC-labeled anti-FLAG tag antibody,

35
respectively. (F) Comparison of IL-2 secretion by Jurkat and PBMC of TM and DR in response to different target cell lines.
Effector cells and target cells were incubated with each of the six target cells (HL60, KG1, MOLM13, and MOLM13-DR) under
the E:T ratio of 1:1 overnight at 37℃. The culture supernatant was collected and then examined for its level of IL-2 secretion via
ELISA. (N = 3).

Figure 12. Expression of anti-CD33-28ζ-P2A-anti-HLA-DR-PD-1 on Jurkat cells. (A) Flow cytometric analysis of
transduced and un-transduced Jurkat cells. The CD33 CAR and the HLA-DR iCAR were verified by cell staining with a PE-
labeled anti-HA tag antibody and an APC-labeled anti-FLAG tag antibody, respectively.

36
Chapter 3 Enhancing the anti-EGFR CAR-T cell specificity by
targeting HLA class I
3.1 Construction of an anti-EGFR activating CAR and an anti-β2m
inhibitory CAR
Based on the previous anti-EGFR CAR-T therapy
33,34
, we utilized 4-1BB instead of
CD28 as the costimulatory domain to promote the survival and persistence of CAR-T cells. The
activating CAR consists of a CD8 leader sequence, a HA-tag, an anti-EGFR scFv, CD8α hinge,
and transmembrane domains, and 4-1BB-CD3ζ intracellular domains. Since different HLA
alleles of MHC class I share the same subunit β2m, we chose it as the target of the inhibitory
CAR. The inhibitory CAR is composed of a CD8 leader sequence, a FLAG-tag, an anti-β2m
scFv, and PD-1 hinge, transmembrane, and intracellular domains. The HA-tag and FLAG-tag are
used for verification of CAR expression (Figure 13A). For a negative control, we also
constructed an anti-β2m inhibitory CAR with only PD-1 hinge and transmembrane domains, but
no intracellular domain, named as anti-β2m-PD-1TM.
We first amplified DNA fragments of anti-EGFR scFv, anti-β2m scFv, and different
hinge-TM-signaling domains separately via PCR, and verified the constructs by agarose gel
electrophoresis (Figure 13B, D). Then, we used overlap PCR to construct full-length anti-EGFR
CAR and anti-β2m CAR (as well as anti-β2m-PD-1TM control) (Figure 13C, E). The DNA
fragments of the inhibitory CAR and activating CAR were digested with (BamH I, EcoR I), and
both DNA fragments were ligated with a pCDH linear vector (BamH I, EcoR I) to construct
pCDH-anti-β2m-PD-1 (referred to as anti-β2m CAR), and pCDH-anti-EGFR-4-1BB-CD3ζ

37
(referred to as anti-EGFR). The negative control was pCDH-anti-β2m-PD-1TM. All of the
positive colonies were verified by sequencing analysis (GENEWIZ) (Figure 13F, G, H).
We first transduced the activating CAR, anti-EGFR-4-1BB-CD3ζ, into Jurkat cells to
verify its activation. Next, the inhibitory CAR was transduced into anti-EGFR CAR-Jurkat cells
to verify its inhibitory effect.

38

Figure 13. Construction of an anti-EGFR CAR and an anti- 2m iCAR. (A) Schematic design of the anti-EGFR single CAR
and anti-β2m single CAR. (B-H) Results of DNA gel electrophoresis: M represents DNA marker (1kb plus DNA ladder). B: lane
1 and 2, lane 3 and 4, as well as lane 6 and 7, indicate, pairwise, anti-EGFR, 4-1BB-CD3ζ, and anti-β2m, respectively. D: lane 1
and lane 2, lane 4 and lane 5, display PD-1 (hinge-TM-Intra), and PD-1 (hinge-TM), respectively. C: lane 1 and lane 2 show anti-
EGFR-4-1BB-CD3ζ. E: lane 1 and lane 2, lane 3 and 4 indicate anti-β2m-PD-1, and anti-β2m-PD-1 (without the signaling
domain), respectively. F: red boxes show DNA fragments of anti-EGFR-4-1BB-CD3ζ. G: red boxes show pCDH-anti-β2m-PD-1
(without the signaling domain). H: red circle shows anti-β2m-PD-1.

39
3.2 Functional verification of the anti-EGFR activating CAR-Jurkat
cells
The expression of anti-EGFR CAR on Jurkat cells was detected by cell staining with anti-
HA antibody and PE-anti-Fc antibody, followed by flow cytometric analysis (Figure 14A). The
expression of anti-EGFR CAR reached ~80% of all cells, which were used in the following
ELISA assay.
Then, we used an IL-2 ELISA to verify the function of the anti-EGFR CAR targeting
EGFR positive cells H2347 (Figure 14B). The results indicated that the activation of anti-EGFR
CAR was significant compared with un-transduced Jurkat cells, representing the negative control
group. The anti-EGFR CAR-Jurkat cells secreted more than 400 pg/mL IL-2 after co-culture
with the target cells, H2347. The successful construction and expression of anti-EGFR CAR
propelled the introduction of the inhibitory CAR, anti-β2m CAR in following experiments.

40

Figure 14. Generation and activation of anti-EGFR CAR Jurkat cells. (A) Flow cytometric analysis of transduced and not
transduced (plain control) Jurkat cells. Both cells were stained with PE-labeled anti-HA tag antibody. (B) IL-2 production of anti-
EGFR CAR Jurkat cells in response to target cells. The effector (CAR-Jurkat) cells were co-cultured with target cells (H2347)
overnight at an E:T ratio of 1:1 in 37℃ incubators.

41
3.3 Functional verification of anti-EGFR and anti- β2m dual CAR in
Jurkat cells
After transduction of the anti-β2m inhibitory CAR into anti-EGFR single CAR-Jurkat
cells, the expression of the dual CAR in Jurkat cells was verified by cell staining. Since the
FLAG-tag and HA-tag were added in the CARs structures, respectively, we used APC-labeled
anti-FLAG antibody, anti-HA antibody and PE-anti-Fc antibody to stain the cells and testified
the expression by flow cytometric analysis (Figure 15A). The double positive dual CAR Jurkat
cells (anti-EGFR-4-1BBζ; anti-β2m-PD-1) accounted for ~99% of all stained cells compared
with un-transduced Jurkat cells, and the expression of the negative control (anti-EGFR-4-1BBζ;
anti-β2m-PD-1TM) cells also reached higher than ~70% of all cells. The high expression of
those CARs indicated that they did not need cell sorting to raise the ratio of CAR-positive cells.
Then, we used an IL-2 ELISA to verify the function of dual CAR (Figure 15B). The
target cells included H2347, H1975, and PC9. H1975 and PC9 cell were all EGFR-positive,
β2m-negative cells, and H2347 was EGFR+ β2m+ double positive cells. Unexpectedly, we
found the inhibition of dual CAR-Jurkat cells was not apparent, where co-culture incubation of
effector cells with β2m positive or negative target cells led to similar levels of IL-2 secretion.
The IL-2 secretion was ~ 400 pg/mL from the coculture of effector cells (anti-EGFR single CAR,
anti-EGFR; anti-β2m-PD-1TM, and antiEGFR-4-1BBζ; anti-β2m-PD-1 dual CAR) and target
cells (H2347, PC9, and H1975)

42

Figure 15. Generation and functional testing of Jurkat cells expressing an anti-EGFR CAR with or without an anti-β2m
iCAR. (A) Flow cytometric analysis of transduced and plain control of Jurkat cells. The anti-EGFR CAR and the anti-β2m (or
control without the signaling domain). iCAR were stained with PE-labeled anti-HA tag antibody, and APC-labeled anti-FLAG
tag antibody, respectively. (N = 3) (B) Comparison of IL-2 production by dual anti-EGFR, anti-β2m-PD-1 CAR and dual anti-
EGFR, anti-β2m-PD-1TM CAR Jurkat cells in response to different target cells. Effector cells were co-cultured with each of the
three target cells (H1975, PC9, and H2347) at an E:T ratio of 1:1 under 37℃ overnight.

43
3.4. Functional verification of dual anti-CD19 activating CAR and
anti-β2m inhibitory CAR
We postulated that the loss of inhibitory effect of dual anti-EGFR, anti-β2m-PD-1 CAR
targeting EGFR
+
β2m
+
cells could be explained by two reasons. The first one is that the β2m
may not be a good target of the inhibitory CAR. The second one is that the main targets of PD-1
are T cell receptor (TCR), including CD3ζ, and CD28, but not the 4-1BB co-stimulatory pathway
51
. Hence, we determined to transduce the anti-β2m-PD-1 inhibitory CAR into anti-CD19-CD28-
CD3ζ single CAR Jurkat cells to test our hypothesis (Figure 16A).
After cell staining, the anti-CD19 single CAR expression was about 95% of all cells and
the dual anti-CD19, anti-β2m CAR expression was about 70% of all cells, and those cells were
applied for functional research (Figure 16B). The target cells are K562-CD19 and Raji cell lines,
which were commonly used cells expressing CD19. We first implemented a cell staining
experiment to verify the β2m antigen expression on K562-CD19 cells and Raji cells. Figure 16C
showed that K562 is β2m-negative, whereas Raji is β2m-positive. In contrast to HLA-DR, β2m
is also expressed in Jurkat cells, so we used an anti-β2m PE-conjugate antibody to block the β2m
on the surface of Jurkat cells to prevent the anti-β2m CAR from inhibiting the effector cells
themselves.
Then, we used IL-2 ELISA to verify the function of dual CAR (Figure 16D). The results
showed that both Raji cells group and K562-CD19 cells group showed inhibition effect. Raji
cells has shown better activation of effector cells compared with K562-CD19, which may be
explained by higher CD19 expression on Raji cells. For the unexpected inhibition effect in
K562-CD19 group, this may be due to the incomplete blockage of β2m antigen on Jurkat cells,
which could induce self-inhibition.

44

Figure 16. Generation and testing of Jurkat cells expressing an anti-CD19 CAR with or without an anti-β2m iCAR. (A)
Schematic design of the anti-CD19 CAR, anti-β2m, and anti-β2mTM iCAR. (B) Flow cytometric analysis of anti-CD19 CAR
and anti-β2m CAR, where they were stained by PE-labeled anti-HA tag antibody, and APC-labeled anti-FLAG tag antibody,
respectively. (N = 3) (C) Flow cytometric analysis of target cells (K562-CD19 and Raji). Both kinds of target cells were stained
with anti-β2m antibody. (D) Comparison of IL-2 production by single anti-CD19 CAR and dual anti-CD19, anti-β2m CAR
Jurkat cells. The effector cells were incubated with two target cells (Raji and K562-CD19) at the E:T ratio of 1:1 and 1:5
overnight at 37℃. (N = 3). The number of target cells in the 1:5 ratio group is 5 times as many of the target cells in 1:1 ratio
group.

45
Chapter 4 Discussion
4.1 Anti-CD33-28ζ, anti-HLA-DR-PD-1 dual CAR for AML
The crux of the dual activating and inhibitory CAR design relies on the antigen choice.
An antigen up-regulated in tumor cells compared to normal tissue would be ideal for the design
of activating CAR, while an antigen down-regulated or lost in tumor cells compared to normal
cells would be an excellent choice for the inhibitory CAR construction. In this study, we first
demonstrated that the anti-HLA-DR inhibitory CAR inhibits anti-CD33 CAR activation
signaling in Jurkat cells. This approach enabled dual anti-CD33, anti-HLA-DR CAR-Jurkat cells
to preferentially target CD33
+
HLA-DR
neg
cells, which could reduce the “on-target, off-tumor”
effect.
Researchers have studied several constructions of iCARs. Fedorov, et al., constructed an
anti-CD19-28ζ, anti-PSMA-PD-1 dual CAR T cells, and the inhibitory anti-PSMA CAR
provided an efficient reduction of cytokine levels and cytotoxicity against CD19
+
PSMA
+
cells
compared to CD19
+
PSMA
neg
cells
48
. However, the tumor-associated antigen (TAA), PSMA, is
not expressed on normal tissues, which limited its application in vivo. Tao, et al., have
established an anti-CD19, anti-KIR-PD-1 dual CAR T cell showing an ideal inhibitory effect
against CD19
+
HLA-A/C
+
cells as well
52
. However, one limitation in this study is that HLA-A
and HLA-C are expressed ubiquitously on all types of normal cells, even in T cells, which may
result in a reduction of efficiency in CAR-T cell treatment. Additionally, Hwang, et al.,
18
have
established a dual CARs system that can specifically react against Allele A
+
Allele B
neg
cells
compared with Allele A
+
Allele B
+

18
. They have provided an efficient way to treat cancers.
However, the clinical use of their CAR T cells can potentially cause tonic signaling of both
chimeric antigen receptor due to the expression of targeting antigens on all normal tissues.

46
Compared with previous studies, our research has several advantages. The first is that
HLA-DR was found to be a novel antigen for inhibitory CAR. In contrast with PSMA, HLA-DR
is not a tumor-associated antigen, and thus it is also suitable for inhibitory CAR design. Unlike
MHC class I molecules, HLA-DR can only be expressed in a few specialized cell types like
dendritic cells, B cells, and macrophages, and activated T cells, which can potentially minimize
adherence between dual CAR-T cells and normal tissue cells. Most importantly, HLA-DR loss in
acute myeloid leukemia (AML) was reported from 15% to 24%
54,55
making it a good target for
the inhibitory CAR design. Given the limited proliferation capacity of myeloid cells, enhancing
the specificity of anti-CD33 activating CAR therapy can avoid excessive killing of normal
myeloid cells.
In this study, we first verified the function of anti-CD33-28ζ CAR and anti-HLA-DR-PD-
1 CAR, respectively, in Jurkat cells by two-step transduction. We showed that dual CAR-Jurkat
cells had an efficient inhibitory effect targeting CD33
+
HLA-DR
+
cells compared to CD33
+
HLA-
DR
neg
cells, where a 3-fold reduction of IL-2 secretion of inhibitory CAR stimulation, compared
to the activating CAR, was observed (Figure 8B). We then attempted to use the one-step
transduction method to transduce the dual CAR into human PBMC. To this end, we inserted a
self-cleavage peptide, p2A, between the inhibitory CAR (situated at the front) and the activating
CAR (situated at the rear) to express both CARs simultaneously. However, dual CAR-Jurkat
cells generated by one-step transduction were not efficacious against target cells, likely due to
the low expression level of the anti-CD33 activating CAR. To address this issue, we switched the
order of the inhibitory CAR and the activating CAR, and we transduced them into human
PBMCs directly. The results showed that the expression and activation of anti-CD33-28ζ CAR
were successful, where the CAR-PBMC cells secreted IL-2 to more than 300 pg/mL (Figure 11E,

47
F). However, the inhibitory CAR was not expressed on the surface of PBMC, which can
probably be explained by that the IgG4-Fc hinge in anti-CD33-28ζ CAR structure is longer than
CD8 hinge that may affect the efficiency of the secondary CAR.
Next, we wanted to focus on optimizing CAR structure to increase their activation and
the expression of the secondary CAR. Muller, et al., indicated that using CD8α hinge and
transmembrane domain in both the activating and inhibitory CARs could significantly increase
the inhibitory effect of inhibitory CAR bearing the PD-1 signaling domain
58
. We plan to test
whether replacing the IgG4-Fc hinge in the anti-CD33-28ζ CAR to the CD8α hinge can help the
expression of the secondary CAR. Also, we can add another co-stimulatory domain, 4-1BB, to
raise the survival and persistence of CAR-T cells
59
. Third, we want to expand the application of
the inhibitory CAR system into other types of effector cells, such as NK cells, and explore other
applicable targets for treating solid tumor cells that have lost expression of MHC class I
molecules.

4.2 Anti-EGFR-4-1BBζ, anti-β2m-PD-1 dual CAR for lung cancer
We also attempted to develop an anti-β2m inhibitory CAR to enhance the specificity of
anti-EGFR CAR-T cells towards non-small cell lung cancer （NSCLC ）cells with HLA class I
loss. EGFR is abnormally overexpressed on lung cancer cells, which means that it is a superior
target for CAR design. Previous studies reported that T cells expressing the 4-1BB/CD3ζ-based
CAR have shown prolonged persistence and survival compared with those expressing the
CD28/CD3ζ
59,60
. Therefore, we developed a 4-1BB/CD3ζ-based anti-EGFR CAR. Compared
with anti-CD33 CAR, anti-EGFR CAR displayed an ideal activation. The IL-2 secretion of anti-
EGFR CAR-Jurkat cells reached from 200 to 1000 pg/mL, whereas the anti-CD33 CAR-Jurkat

48
cells only produced ~30 pg/mL IL-2. However, we did not see any inhibitory effect from the
anti-β2m inhibitory CAR. There are two possible reasons. One is that the PD-1-mediated signal
pathway can only inhibit CD28 and CD3ζ, but not 4-1BB
59,61
. To solve this problem, we are
currently developing a new anti-EGFR CAR in which the 4-1BB co-stimulatory domain is
replaced by the CD28 co-stimulatory domain. The other possibility is that β2m on Jurkat cells
can interfere with the expression or function of the anti-β2m inhibitory CAR. This can
potentially be addressed by deleting β2m in Jurkat cells by the CRISPR/Cas9 gene knockout
technique.

49
Materials and Methods
Cell culture. Jurkat plain cells, CAR Jurkat cells, K562-CD19 cells, Raji cells, H2349 cells, PC9
cells, H1975 cells, PBMC, CAR PBMC, KG1 cells, Nalm6 cells, MOLM13 cells, MOLM13-DR
cells, and HL60 cells were cultured in RPMI-1640 medium with 10% Fetal Bovine Serum (FBS)
and Penicillin-Streptomycin-Glutamine (PSG, 100 unit/mL Penicillin or Streptomycin,
292ng/mL L-Glutamine). All cells were also incubated in incubator at 37℃ and 5% CO2 levels.
293T cells were seeded in DMEM high-glucose medium with 10% FBS solution and PSG
solution, and were also incubated in incubator with 37℃ and 5% CO2 concentration.
Generation of lentiviral vectors and calcium-phosphorus-based Jurkat cell transduction
and transfection. We transduced CAR plasmids into Jurkat cells via lentiviral vectors. Firstly,
we cultured 293T cells in a 100 mm dish until the number of cells reached 20 million in 30mL
DMEM medium. Next, we added 20,000 ng pREV, pVSVG, pPRE, and 40,000 ng of the
plasmid of interest to the 293T cells, and put the dish into an incubator at 37℃, and 5% CO2
levels. The medium was changed after 8 hours, and the cells were re-cultured it in the incubator.
After two days of transduction, the virus was collected with a 10K filter. The cells were
centrifuged at 1200g for 15min to concentrate the virus to a 500 μl final volume. Then, the virus
solution was added the virus solution into 0.5 mL with 1 million Jurkat cells in a well of 24-well
plate. The cells were centrifuge at 2000g for more than two hours. Cell were incubated in a cell
incubator.
Generation of lentiviral vectors and lipofectamine-based Jurkat cell transduction and
transfection. 293T cells were cultured DMEM/high glucose medium (HyClone™) containing 10%
FBS, and PSG (100 unit/mL Penicillin or Streptomycin, 292 ng/mL L-Glutamine) at 37℃

50
incubators. Forty-eight hours before cell transfection, 293T cells were seeded into a T-75 culture
flask until the number of cells reached ~3.5 × 10
6
cells/ml. Then, the medium was changed to
8mL of Reduced Serum Medium (Opti-MEM®1, 1X) for 30 minutes. In the transfection process,
we added 15000 ng of the plasmid of interest, with 7500 ng pPRE, pREV, and pVSVG plasmids
into Reduced Serum Medium containing 36 μl Lipofectamine® 2000 reagent (1 mg/mL). After
vortexing completely, the mixture was added to 293T cells for 4 to 6 hours, and the medium was
changed to 10 ml DMEM-based medium. After incubating at 37℃ for seventy-two hours, the
supernatant was collected, and the virus was precipitated with one-third volume of Lenti-X™
Concentrator (TaKaRa Bio USA, Inc) via 2 hours centrifuging at 2000g. The supernatant was
discarding after centrifuging, and the virus was resuspended the virus with RPMI-1640 medium,
and added into 0.5 ml of Jurkat 1.0 × 10
6
cells/ml, and were incubated in a treated 24-well plate.
Transformation A small tube of competent cells from -80℃ were incubated on ice for 15 min
to thaw it. After thawing, we put the ligation product of the plasmid of interest into competent
cells and incubated it on ice again for 30 min. Then, cells were heat shocked at 47℃ for 30 s to 1
min, and put back on ice for 2 min. 800μl of SOC medium was added and 1 μl ampicillin (100
μg/ml). Next, we put the competent cell tube into an incubator and shook it for 1 hour at 220 rpm
and 37℃. After that, 200μl of the solution was withdrawn and applied evenly on the surface of
the culture medium with ampicillin. The culture medium was incubated in 37℃ incubators
overnight. Finally, the bacterial clusters were picked and placed on the surface of the culture
medium, and the approximate sequence length was determined using colony PCR and DNA
sequencing technology.

51
ELISA assay We co-cultured the effector cells and target cells in U-shaped wells in a 96-well
plate with certain E:T ratio overnight or for 12 hours. Then, the plate was centrifuged at 1300
rpm for 3 min. We withdrew 100 μl surfactant of co-culture medium and tested their IL-2 levels
using the IL-2 ELISA test kit.
Cytotoxic effect of CAR-T cells on target cells. We sat up two group of cells. One group
(experimental group) was co-culturing of effector cells and target cells overnight. Another
(control group) was culturing the effector cells and target cells in separate wells. These cells
were also incubated in an incubator overnight, as the experimental group was. After co-culture,
we mixed the effector cells and target cells and placed them into FACS tubes immediately,
which was the 0 hours sample. Likewise, we took the cells from the experimental group into
FACS tubes, which was the 24 hours sample. Finally, we used flow cytometry analysis to verify
the cells status.
Cell Staining experiment for effect cells and target cells. To verify the expression of CARs
and antigens on effector cells and target cells, we used cell staining and flow cytometry analysis
to examine them. First, we acquired 1 to 2 million cells into each well of treated U-shape 96-well
plates and washed the cells with FACS buffer (2% FBS, 0.05% Na3N, 2mM EDTA) under the
condition of 1300 rpm, 3 mins. Then, 1 μl of the antibody solution or the anti-HA tag
(eBioscience™), APC anti-DYKDDDDK tag (BioLegend®) into 50 μl of FACS buffer. Aspirate
All of the solution from each well was aspirated and mixed. After that, the plate was incubated
on ice for 30 min. Next, we washed the cells with FACS buffer twice and resuspended the cells
in 400 μl FACS buffer into a FACS tube. For CAR T cells, they also needed to be stained by a
secondary antibody, PE (Fab’2-Donkey anti-Rabbit IgG H+L, eBioscience™). The signal was
examined by flow cytometry.

52
References:
1. Stone RM, O'Donnell MR, Sekeres MA. Acute myeloid leukemia. Hematology Am Soc
Hematol Educ Program. 2004:98-117.
2. Jongen-Lavrencic M, Sun SM, Dijkstra MK, Valk PJ, Lowenberg B. MicroRNA
expression profiling in relation to the genetic heterogeneity of acute myeloid leukemia.
Blood. 2008;111(10):5078-5085.
3. Pelcovits A, Niroula R. Acute Myeloid Leukemia: A Review. R I Med J (2013).
2020;103(3):38-40.
4. De Kouchkovsky I, Abdul-Hay M. 'Acute myeloid leukemia: a comprehensive review
and 2016 update'. Blood Cancer J. 2016;6(7):e441.
5. Bizzozero OJ, Jr., Johnson KG, Ciocco A. Radiation-related leukemia in Hiroshima and
Nagasaki, 1946-1964. I. Distribution, incidence and appearance time. N Engl J Med.
1966;274(20):1095-1101.
6. Rauscher GH, Shore D, Sandler DP. Hair dye use and risk of adult acute leukemia. Am J
Epidemiol. 2004;160(1):19-25.
7. Bennett JM, Catovsky D, Daniel MT, et al. Proposed revised criteria for the classification
of acute myeloid leukemia. A report of the French-American-British Cooperative Group.
Ann Intern Med. 1985;103(4):620-625.
8. Molica M, Breccia M, Foa R, Jabbour E, Kadia TM. Maintenance therapy in AML: The
past, the present and the future. Am J Hematol. 2019;94(11):1254-1265.
9. Wei AH, Dohner H, Pocock C, et al. Oral Azacitidine Maintenance Therapy for Acute
Myeloid Leukemia in First Remission. N Engl J Med. 2020;383(26):2526-2537.
10. Takami A. Hematopoietic stem cell transplantation for acute myeloid leukemia. Int J
Hematol. 2018;107(5):513-518.
11. Stahl M, Tallman MS. Differentiation syndrome in acute promyelocytic leukaemia. Br J
Haematol. 2019;187(2):157-162.
12. Duma N, Santana-Davila R, Molina JR. Non-Small Cell Lung Cancer: Epidemiology,
Screening, Diagnosis, and Treatment. Mayo Clin Proc. 2019;94(8):1623-1640.
13. Benmebarek MR, Karches CH, Cadilha BL, Lesch S, Endres S, Kobold S. Killing
Mechanisms of Chimeric Antigen Receptor (CAR) T Cells. Int J Mol Sci. 2019;20(6).
14. Labanieh L, Majzner RG, Mackall CL. Programming CAR-T cells to kill cancer. Nat
Biomed Eng. 2018;2(6):377-391.

53
15. Jayaraman J, Mellody MP, Hou AJ, et al. CAR-T design: Elements and their synergistic
function. EBioMedicine. 2020;58:102931.
16. Xie YJ, Dougan M, Jailkhani N, et al. Nanobody-based CAR T cells that target the tumor
microenvironment inhibit the growth of solid tumors in immunocompetent mice. Proc
Natl Acad Sci U S A. 2019;116(16):7624-7631.
17. Rafiq S, Hackett CS, Brentjens RJ. Engineering strategies to overcome the current
roadblocks in CAR T cell therapy. Nat Rev Clin Oncol. 2020;17(3):147-167.
18. Hwang MS, Mog BJ, Douglass J, et al. Targeting loss of heterozygosity for cancer-
specific immunotherapy. Proc Natl Acad Sci U S A. 2021;118(12).
19. Schepisi G, Cursano MC, Casadei C, et al. CAR-T cell therapy: a potential new strategy
against prostate cancer. J Immunother Cancer. 2019;7(1):258.
20. Yu S, Yi M, Qin S, Wu K. Next generation chimeric antigen receptor T cells: safety
strategies to overcome toxicity. Mol Cancer. 2019;18(1):125.
21. Sermer D, Brentjens R. CAR T-cell therapy: Full speed ahead. Hematol Oncol. 2019;37
Suppl 1:95-100.
22. Abramson JS. Anti-CD19 CAR T-Cell Therapy for B-Cell Non-Hodgkin Lymphoma.
Transfus Med Rev. 2020;34(1):29-33.
23. Han D, Xu Z, Zhuang Y, Ye Z, Qian Q. Current Progress in CAR-T Cell Therapy for
Hematological Malignancies. J Cancer. 2021;12(2):326-334.
24. Munshi NC, Anderson LD, Jr., Shah N, et al. Idecabtagene Vicleucel in Relapsed and
Refractory Multiple Myeloma. N Engl J Med. 2021;384(8):705-716.
25. Hernandez-Caselles T, Martinez-Esparza M, Perez-Oliva AB, et al. A study of CD33
(SIGLEC-3) antigen expression and function on activated human T and NK cells: two
isoforms of CD33 are generated by alternative splicing. J Leukoc Biol. 2006;79(1):46-58.
26. Zhao L. CD33 in Alzheimer's Disease - Biology, Pathogenesis, and Therapeutics: A
Mini-Review. Gerontology. 2019;65(4):323-331.
27. Willier S, Rothamel P, Hastreiter M, et al. CLEC12A and CD33 coexpression as a
preferential target for pediatric AML combinatorial immunotherapy. Blood.
2021;137(8):1037-1049.
28. Kenderian SS, Ruella M, Shestova O, et al. CD33-specific chimeric antigen receptor T
cells exhibit potent preclinical activity against human acute myeloid leukemia. Leukemia.
2015;29(8):1637-1647.
29. Herbst RS. Review of epidermal growth factor receptor biology. Int J Radiat Oncol Biol
Phys. 2004;59(2 Suppl):21-26.

54
30. Janne PA, Engelman JA, Johnson BE. Epidermal growth factor receptor mutations in
non-small-cell lung cancer: implications for treatment and tumor biology. J Clin Oncol.
2005;23(14):3227-3234.
31. Liu X, Wang P, Zhang C, Ma Z. Epidermal growth factor receptor (EGFR): A rising star
in the era of precision medicine of lung cancer. Oncotarget. 2017;8(30):50209-50220.
32. Lynch TJ, Bell DW, Sordella R, et al. Activating mutations in the epidermal growth
factor receptor underlying responsiveness of non-small-cell lung cancer to gefitinib. N
Engl J Med. 2004;350(21):2129-2139.
33. Zhou X, Li J, Wang Z, et al. Cellular immunotherapy for carcinoma using genetically
modified EGFR-specific T lymphocytes. Neoplasia. 2013;15(5):544-553.
34. Li H, Huang Y, Jiang DQ, et al. Antitumor activity of EGFR-specific CAR T cells
against non-small-cell lung cancer cells in vitro and in mice. Cell Death Dis.
2018;9(2):177.
35. Freyer CW, Porter DL. Cytokine release syndrome and neurotoxicity following CAR T-
cell therapy for hematologic malignancies. J Allergy Clin Immunol. 2020;146(5):940-948.
36. Bonifant CL, Jackson HJ, Brentjens RJ, Curran KJ. Toxicity and management in CAR T-
cell therapy. Mol Ther Oncolytics. 2016;3:16011.
37. Liu Y, Fang Y, Chen X, et al. Gasdermin E-mediated target cell pyroptosis by CAR T
cells triggers cytokine release syndrome. Sci Immunol. 2020;5(43).
38. Morris EC, Neelapu SS, Giavridis T, Sadelain M. Cytokine release syndrome and
associated neurotoxicity in cancer immunotherapy. Nat Rev Immunol. 2021.
39. Parker KR, Migliorini D, Perkey E, et al. Single-Cell Analyses Identify Brain Mural
Cells Expressing CD19 as Potential Off-Tumor Targets for CAR-T Immunotherapies.
Cell. 2020;183(1):126-142 e117.
40. Fuca G, Reppel L, Landoni E, Savoldo B, Dotti G. Enhancing Chimeric Antigen
Receptor T-Cell Efficacy in Solid Tumors. Clin Cancer Res. 2020;26(11):2444-2451.
41. Sun S, Hao H, Yang G, Zhang Y, Fu Y. Immunotherapy with CAR-Modified T Cells:
Toxicities and Overcoming Strategies. J Immunol Res. 2018;2018:2386187.
42. Shimabukuro-Vornhagen A, Godel P, Subklewe M, et al. Cytokine release syndrome. J
Immunother Cancer. 2018;6(1):56.
43. Pabst T, Joncourt R, Shumilov E, et al. Analysis of IL-6 serum levels and CAR T cell-
specific digital PCR in the context of cytokine release syndrome. Exp Hematol.
2020;88:7-14 e13.

55
44. Caruso HG, Hurton LV, Najjar A, et al. Tuning Sensitivity of CAR to EGFR Density
Limits Recognition of Normal Tissue While Maintaining Potent Antitumor Activity.
Cancer Res. 2015;75(17):3505-3518.
45. Grada Z, Hegde M, Byrd T, et al. TanCAR: A Novel Bispecific Chimeric Antigen
Receptor for Cancer Immunotherapy. Mol Ther Nucleic Acids. 2013;2:e105.
46. Pardoll DM. The blockade of immune checkpoints in cancer immunotherapy. Nat Rev
Cancer. 2012;12(4):252-264.
47. Themeli M, Kloss CC, Ciriello G, et al. Generation of tumor-targeted human T
lymphocytes from induced pluripotent stem cells for cancer therapy. Nat Biotechnol.
2013;31(10):928-933.
48. Fedorov VD, Themeli M, Sadelain M. PD-1- and CTLA-4-based inhibitory chimeric
antigen receptors (iCARs) divert off-target immunotherapy responses. Sci Transl Med.
2013;5(215):215ra172.
49. Chemnitz JM, Parry RV, Nichols KE, June CH, Riley JL. 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.
2004;173(2):945-954.
50. Yokosuka T, Takamatsu M, Kobayashi-Imanishi W, Hashimoto-Tane A, Azuma M,
Saito T. Programmed cell death 1 forms negative costimulatory microclusters that
directly inhibit T cell receptor signaling by recruiting phosphatase SHP2. J Exp Med.
2012;209(6):1201-1217.
51. Patsoukis N, Wang Q, Strauss L, Boussiotis VA. Revisiting the PD-1 pathway. Sci Adv.
2020;6(38).
52. Tao L, Farooq MA, Gao Y, et al. CD19-CAR-T Cells Bearing a KIR/PD-1-Based
Inhibitory CAR Eradicate CD19(+)HLA-C1(-) Malignant B Cells While Sparing
CD19(+)HLA-C1(+) Healthy B Cells. Cancers (Basel). 2020;12(9).
53. Baba T, Hanagiri T, Takenoyama M, et al. Identification of a lung cancer antigen evading
CTL attack due to loss of human leukocyte antigen (HLA) class I expression. Cancer Sci.
2010;101(10):2115-2120.
54. Higashi M, Tokuhira M, Fujino S, et al. Loss of HLA-DR expression is related to tumor
microenvironment and predicts adverse outcome in diffuse large B-cell lymphoma. Leuk
Lymphoma. 2016;57(1):161-166.
55. Nijland M, Veenstra RN, Visser L, et al. HLA dependent immune escape mechanisms in
B-cell lymphomas: Implications for immune checkpoint inhibitor therapy?
Oncoimmunology. 2017;6(4):e1295202.

56
56. Acosta Davila JA, Hernandez De Los Rios A. An Overview of Peripheral Blood
Mononuclear Cells as a Model for Immunological Research of Toxoplasma gondii and
Other Apicomplexan Parasites. Front Cell Infect Microbiol. 2019;9:24.
57. Hudecek M, Lupo-Stanghellini MT, Kosasih PL, et al. Receptor affinity and extracellular
domain modifications affect tumor recognition by ROR1-specific chimeric antigen
receptor T cells. Clin Cancer Res. 2013;19(12):3153-3164.
58. Muller YD, Nguyen DP, Ferreira LMR, et al. The CD28-Transmembrane Domain
Mediates Chimeric Antigen Receptor Heterodimerization With CD28. Front Immunol.
2021;12:639818.
59. Philipson BI, O'Connor RS, May MJ, June CH, Albelda SM, Milone MC. 4-1BB
costimulation promotes CAR T cell survival through noncanonical NF-kappaB signaling.
Sci Signal. 2020;13(625).
60. Cheng Z, Wei R, Ma Q, et al. In Vivo Expansion and Antitumor Activity of Coinfused
CD28- and 4-1BB-Engineered CAR-T Cells in Patients with B Cell Leukemia. Mol Ther.
2018;26(4):976-985.
61. Hui E, Cheung J, Zhu J, et al. T cell costimulatory receptor CD28 is a primary target for
PD-1-mediated inhibition. Science. 2017;355(6332):1428-1433.

Abstract (if available)

Abstract Chimeric antigen receptor (CAR) T cells have shown selective and potential efficacy by targeting tumor-associated antigens (TAA) on cancer cells. However, the practice of CAR-T cell therapy is limited by the “on-target, off-tumor” effect resulting from shared tumor-antigen on normal tissues. Since the major histocompatibility complex (MHC) is frequently downregulated or lost in malignant cells, we propose to reduce the on-target off-tumor toxicity of CAR-T cells by activating them against TAA while inhibiting them against MHC molecules in normal tissue cells. Here, we report two inhibitory CARs (iCARs) targeting human leukocyte antigen DR (HLA-DR), an MHC class II molecule, and β2 microglobulin (β2m), a component of MHC class I molecules. We first verified that dual CAR-Jurkat cells, which express a CD28/CD3ζ-based anti-CD33 CAR and a PD-1-based anti-HLA-DR iCAR, preferentially targeted CD33⁺HLA-DRⁿᵉᵍ cells over CD33⁺HLA-DR⁺ cells. We have also put in effort to construct new plasmids for one-step lentiviral transduction of human primary T cells with the CAR and the iCAR. Second, we constructed another iCAR targeting β2m. However, we found the dual CAR-Jurkat cells, which expressed the 4-1BB/CD3ζ-based anti-EGFR CAR and the PD-1-based anti-β2m iCAR, showed no inhibition effect on EGFR⁺ β2m⁺ cells compared to EGFR⁺ β2mⁿᵉᵍ cells. There are two possible reasons. One is that the 4-1BB co-stimulatory signal pathway cannot be inhibited by PD-1. This can potentially be addressed by replacing 4-1BB with CD28 in the anti-EGFR CAR. The other is that β2m in Jurkat cells is interfering with the function of the anti-β2m iCAR. We propose to knock out β2m in Jurkat cells in future work.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Enhancing the specificity and cytotoxicity of chimeric antigen receptor Natural Killer cells

PDF

Engineering chimeric antigen receptor (CAR) -modified T cells for enhanced cancer immunotherapy

PDF

Chimeric Antigen Receptor targeting Prostate Specific Membrane Antigen (PSMA)

PDF

Novel design and combinatory therapy to enhance chimeric antigen receptor engineered T cells (CAR-T) efficacy against solid tumor

PDF

Self-secretion of checkpoint blockade enhances antitumor immunity by murine chimeric antigen receptor-engineered T cells

PDF

Y-shaped DNA based pMHC nonamers for detecting low-affinity T cells

PDF

Generation and characterization of fully human anti-CD19 chimeric antigen receptor T (CAR-T) cells for the treatment of hematologic malignancies

PDF

Immunotherapy of cancer

PDF

Engineering chimeric antigen receptor-directed immune cells for enhanced antitumor efficacy in solid tumors

PDF

Lym-1 epitope targeted chimeric antigen receptor (CAR) T cells for the immunotherapy of cancer

PDF

T cell regulation of HLA-DR

PDF

Improving antitumor efficacy of chimeric antigen receptor-engineered immune cell therapy with synthetic biology and combination therapy approaches

PDF

Towards DNA-directed assembly of pMHC multimers for detection of low-affinity T cells

PDF

Deregulation of CD36 expression in cancer presents a potential targeting therapeutic opportunity

PDF

Investigating the effects of targeting CD99 on T cells to enhance their antileukemia activity

PDF

Generation and characterization of humanized anti-CD19 chimeric antigen receptor T (CAR-T) cells for the treatment of hematologic malignancies

PDF

Therapeutic resistance in acute lymphoblastic leukemia: cIAP2 inhibition sensitizes B cell acute lymphoblastic leukemia to anti-CD19 chimeric antigen receptor T cell

PDF

Generation and characterization of anti-CD138 chimeric antigen receptor T (CAR-T) cells for the treatment of hematologic malignancies

PDF

Regulation of T cell HLA-DR by CD3 ζ signaling

PDF

Enhancing the anti-HIV potency of eCD4-Ig by unnatural amino acid mutagenesis

Asset Metadata

Creator Jiang, Nan (author)

Core Title Enhancing the anti-cancer specificity of chimeric antigen receptor T cells through targeting HLA loss

School School of Pharmacy

Degree Master of Science

Degree Program Pharmaceutical Sciences

Degree Conferral Date 2021-08

Publication Date 07/28/2023

Defense Date 07/28/2021

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

Tag AML,CAR T cell therapy,CD33,EGFR,lung cancer,OAI-PMH Harvest,PD-1

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Xie, Jianming (committee chair), Okamoto, Curtis (committee member), Zaro, Jennica (committee member)

Creator Email jiangnan@usc.edu,nanjiang96@hotmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC15672114

Unique identifier UC15672114

Legacy Identifier etd-JiangNan-9954

Document Type Thesis

Format application/pdf (imt)

Rights Jiang, Nan

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 author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright. The original signature page accompanying the original submission of the work to the USC Libraries is retained by the USC Libraries and a copy of it may be obtained by authorized requesters contacting the repository e-mail address given.

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

Repository Email cisadmin@lib.usc.edu