Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Deregulation of CD36 expression in cancer presents a potential targeting therapeutic opportunity
(USC Thesis Other)
Deregulation of CD36 expression in cancer presents a potential targeting therapeutic opportunity
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
Deregulation of CD36 Expression in Cancer Presents a Potential Targeting Therapeutic
Opportunity
By
Yiting Meng
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 SCIENCE
August 2021
Copyright 2021 Yiting Meng
ii
Acknowledgements
Firstly, I would like to sincerely express my great appreciation to my advisor Dr. Houda
Alachkar of School of Pharmacy at the University of Southern California. Her guidance,
patience, encouragement, and enthusiasm always support me to finish my research.
Then I would like to thank other my thesis committee members, Dr. Curtis Okamoto, and Dr. Ian
Haworth for their insightful and encouraging comments to my thesis.
Next, I would like to thank my lab mates for their help on my work.
I would also like to thank my friend Antonina Nazarova for her encouragement.
Finally, I would like to thank my parents and my brother for their financial and emotional
support for my study.
iii
Table of Contents
Acknowledgements ......................................................................................................................... ii
List of Tables .................................................................................................................................. v
List of Figures ................................................................................................................................ vi
Abbreviations ................................................................................................................................ vii
Abstract .......................................................................................................................................... ix
Chapter 1: Introduction ................................................................................................................... 1
1.1 Tumorigenesis and Cancer Metabolism ................................................................................ 1
1.2 Expression and Function of CD36 in Healthy Tissues ......................................................... 3
1.3 The Roles of CD36 in Cancer ............................................................................................... 4
1.4 The Role of CD36 in Tumor Immune Microenvironment .................................................... 7
1.5 The Association Between CD36 Expression and Clinical Outcome in Cancer .................... 9
Chapter 2: Patterns of CD36 Expression in Healthy and Malignant Tissues Dictate Potential
Opportunities for Cancer Targeted Therapy ................................................................................. 11
2.1 Introduction ......................................................................................................................... 11
2.2 Experimental Approach and Methods ................................................................................. 13
2.2.1 CD36 Gene Expression Analysis ................................................................................. 13
2.2.2 DNA Methylation Analysis .......................................................................................... 14
2.2.3 Survival Analysis.......................................................................................................... 14
2.2.4 Tumor - Immune Interaction Analysis ......................................................................... 14
2.2.5 Genetic Mutations and Chromosome Abnormalities Analysis .................................... 15
2.2.6 Statistical Analysis ....................................................................................................... 15
2.3 Results ................................................................................................................................. 16
2.3.1 CD36 Expression Patterns in Normal Tissues.............................................................. 16
2.3.2 CD36 Expression Patterns in Cancer ........................................................................... 19
2.3.3 CD36 Promoter DNA Methylation Status in Different Cancers .................................. 21
2.3.4 CD36 Expression is Correlated with Clinical Outcomes in Specific Cancer Types .... 23
2.3.5 CD36 Expression Pattern in Normal Hematopoiesis ................................................... 25
iv
2.3.6 CD36 mRNA Abnormal Expression is Correlated with Immune Cells Infiltration, and
May Influence Patient’s Clinical Outcome ........................................................................... 28
2.3.7 CD36 Expression is Correlated with the Immune Suppression in Tumor
Microenvironment ................................................................................................................. 31
2.3.8 CD36 Gene is Frequently Mutated in Various Malignancies ...................................... 38
2.3.9 Clinical Attributes Associated with CD36 Alterations ................................................ 45
2.4 Discussion ........................................................................................................................... 47
Chapter 3. Functional Characterization of CD36 in Mouse Primary Cells .................................. 51
3.1 Introduction ......................................................................................................................... 51
3.2 Methods and Materials ........................................................................................................ 53
3.2.1 Short Hairpin RNA-mediated Knockdown of Cd36 Gene ........................................... 53
3.2.2 Lentiviral Vectors-mediated Cd36 Overexpression in Recombinant Plasmids ........... 55
3.2.3 Lentivirus Production ................................................................................................... 57
3.2.4 Loss and Gain Function of Cd36 in Mice Primary Cells in Vitro ................................ 58
3.2.4 qPCR Assessment of Cd36 Knockdown and Overexpression ..................................... 59
3.2.5 Western Blot Assessment of Cd36 Knockdown and Overexpression.......................... 60
3.2.6 Proliferation Assay ....................................................................................................... 61
3.2.7 Statistical Analysis ....................................................................................................... 61
3.3 Results ................................................................................................................................. 61
3.3.1 Assessment of Cd36 Knockdown in Mouse Primary Hematopoietic Cells ................. 61
3.3.2 Assessment of Cd36 Overexpression in Healthy Mouse Hematopoietic Cells ............ 63
3.3.3 Cd36 Knockdown Has Limited Effects on the Proliferation of Normal Mouse
Hematopoietic Cells .............................................................................................................. 64
3.4 Discussion ........................................................................................................................... 66
Chapter 4: Concluding Remarks ................................................................................................... 69
Bibliography ................................................................................................................................. 73
v
List of Tables
Table 2.1 Gene Expression for CD36 in Normal Tissues ............................................................. 17
Table 2.2 Gene Expression for CD36 in Normal Hematopoiesis ................................................. 27
Table 2.3 Correlation Analysis between CD36 Expression and Tumor Cells Immune Infiltration
Levels in TCGA Cancer................................................................................................................ 30
Table 2.4 Correlation Analysis between CD36 Gene and Related Gene Markers in T Cell
Exhaustion for TCGA Cancer ....................................................................................................... 34
Table 2.5 Correlation Analysis between CD36 Gene and Related Gene Markers in Treg cells for
TCGA Cancer ............................................................................................................................... 35
Table 2.6 Correlation Analysis between CD36 Gene and Related Gene Markers Associated with
TAM Polarization for TCGA Cancer (A) ..................................................................................... 36
Table 2.7 Correlation Analysis between CD36 Gene and Related Gene Markers Associated with
TAM Polarization for TCGA Cancer (B) ..................................................................................... 37
Table 2.8 CD36 Mutation Status in Different Cancers ................................................................. 40
Table 2.9 CD36 Alteration Associated Chromosomal Status ....................................................... 46
vi
List of Figures
Figure 2.1 CD36 pattern of expression in normal tissues ............................................................. 17
Figure 2.2 Analysis of CD36 differential expression in cancers .................................................. 20
Figure 2.3 Analysis of CD36 differential expression in cancers (supplementary) ....................... 21
Figure 2.4 Promoter methylation status of CD36 where CD36 has aberrant expression in tumor
tissues compared to healthy tissues............................................................................................... 22
Figure 2.5 Survival analysis of cancer patients with low or high CD36 mRNA expression ........ 24
Figure 2.6 Survival analysis of cancer patients with low or high CD36 mRNA expression
(supplementary) ............................................................................................................................ 25
Figure 2.7 CD36 pattern of expression in normal hematopoiesis ................................................. 26
Figure 2.8 Correlation between CD36 expression and immune infiltration level in different
cancers........................................................................................................................................... 29
Figure 2.9 Correlation between CD36 expression and immune markers in different cancers ..... 33
Figure 2.10 Patterns of CD36 mutations in cancer ....................................................................... 39
Figure 2.11 CD36 mutation associated tumor immune infiltration status .................................... 44
Figure 2.12 CD36 alterations associated chromosomal status...................................................... 46
Figure 3.1 The diagram of shCd36 design .................................................................................... 54
Figure 3.2 The diagram of colony PCR primers design in detecting shCd36 .............................. 55
Figure 3.3 The diagram of Cd36 molecular cloning ..................................................................... 56
Figure 3.4 The diagram of colony PCR Primer design in detecting Cd36 cDNA ........................ 57
Figure 3.5 The effect of Cd36 knockdown in mouse primary cells.............................................. 62
Figure 3.6 The effect of Cd36 overexpression in healthy mouse primary cells ........................... 64
Figure 3.7 Proliferation status of healthy mouse primary cells with reduced Cd36 expression ... 66
vii
Abbreviations
TCGA The Cancer Genome Atlas
ACC Adrenocortical carcinoma
BLCA Bladder Urothelial Carcinoma
LGG Brain Lower Grade Glioma
BRCA Breast invasive carcinoma
CESC Cervical squamous cell carcinoma and endocervical adenocarcinoma
CHOL Cholangiocarcinoma
LCML Chronic Myelogenous Leukemia
COAD Colon adenocarcinoma
ESCA Esophageal carcinoma
GBM Glioblastoma multiforme
HNSC Head and Neck squamous cell carcinoma
KICH Kidney Chromophobe
KIRC Kidney renal clear cell carcinoma
KIRP Kidney renal papillary cell carcinoma
LIHC Liver hepatocellular carcinoma
LUAD Lung adenocarcinoma
LUSC Lung squamous cell carcinoma
DLBC Lymphoid Neoplasm Diffuse Large B-cell Lymphoma
MESO Mesothelioma
MISC Miscellaneous
OV Ovarian serous cystadenocarcinoma
viii
PAAD Pancreatic adenocarcinoma
PCPG Pheochromocytoma and Paraganglioma
PRAD Prostate adenocarcinoma
READ Rectum adenocarcinoma
SARC Sarcoma
SKCM Skin Cutaneous Melanoma
STAD Stomach adenocarcinoma
TGCT Testicular Germ Cell Tumors
THYM Thymoma
THCA Thyroid carcinoma
UCS Uterine Carcinosarcoma
UCEC Uterine Corpus Endometrial Carcinoma
UVM Uveal Melanoma
ix
Abstract
CD36 is a cell surface receptor that is widely expressed in all kinds of cell types in the human
body, such as adipocytes, microphages, platelets, myocytes, and hepatocytes. The primary
function of CD36 is transporting fatty acids in lipid metabolism. So, as a fatty acid translocase,
CD36 is associated with several metabolic disorders and involved in the pathogenic mechanisms
of many diseases. The increased expression of CD36 is examined in breast cancer, oral cancer,
melanoma, prostate cancer, pancreatic cancer, glioblastoma, ovarian cancer, esophageal cancer,
and hematopoietic malignancies. In cancer cells, the upregulated CD36 level provides fuel to
support the increased cell proliferation, as well as the adaptation to the tumor microenvironment
(TME). Also, the decreased CD36 expression is reported in some metastatic tumors such as
colon cancer, and some cases of breast cancer, where the loss of CD36 is primarily found in the
endothelial cells resulting in an inhibition on thrombospondin receptor regulated antitumor
effects. CD36 has recently gained interest as a therapeutic target in acute myeloid leukemia
(AML). It was found to interact with apolipoprotein C2 to promote leukemia growth. However,
whether targeting CD36 in normal hematopoietic cells poses any toxicity remains to be
established in order to bring this therapeutic strategy to maturity.
In this thesis, I used public datasets to investigate the genomics, transcriptomics, and methylation
patterns of CD36 in both normal and cancer tissues. I noticed that CD36 deregulation was not
always consistent with its promoter methylation status. I found that CD36 expression was
upregulated in kidney renal clear cell carcinoma (KIRC) and glioblastoma; and the CD36
promoter was hypermethylated in KIRC. I also observed remarkably downregulated CD36
expression levels in breast cancer and lung cancer, where the CD36 promoter was
x
hypermethylated in both two cancer types. Besides, the tumor tissues exhibited lower CD36
expression compared with normal tissues for colon cancer, prostate cancer, and head and neck
cancer; yet the CD36 promoter was hypomethylated in all three cancer types. I next examined
the correlation between CD36 expression and the patient’s clinical outcome and the infiltration
level of tumor immune cells. The results indicated a significantly positive correlation between
CD36 expression and infiltration levels of tumor immune cells such as CD4+ T cells, CD8+ T
cells, neutrophils, natural killer cells, and dendritic cells during cancer development. Moreover, I
speculated that CD36 expression probably antagonized the T cell mediated cytotoxic effects on
malignant cells, since CD36 expression also showed a significantly positive correlation with
gene markers presented during T cell exhaustion, and the differentiation of Treg cells, and M2-
like tumor-associated macrophages. In addition, I investigated the role of Cd36 in normal
hematopoietic cells using gain- and loss-of-function approaches in murine normal hematopoietic
primary cells. The loss of function in Cd36 showed limited effects on the proliferation of mouse
healthy hematopoietic primary cells. Altogether, my work demonstrates that CD36 is highly
deregulated in cancer and within the cancer immune microenvironment. Yet targeting CD36 in
normal murine hematopoietic cells had limited effect on cell viability. Whether CD36 may
potentially present a viable therapeutic approach in specific malignancies, remain to be
investigated in preclinical models.
1
Chapter 1: Introduction
1.1 Tumorigenesis and Cancer Metabolism
In general, cancer refers to a group of abnormal cells that have uncontrolled growth and tend to
continually spread into surrounding tissues in the human body. During tumor progression and
metastasis, cancer cells need to be highly competitive in acquiring the nutrients and energy to
support their enhanced proliferation and invasion compared to normal cells. However, the tumor
niche is usually nutrient-poor and oxygen-deficient. In order to adapt to the harsh
microenvironment, cancer cells have to modify their metabolic networks at each stage of tumor
progression.
1
Therefore, metabolic reprogramming has been widely characterized as a hallmark
of cancer development.
1
In an earlier era, the Warburg effect described modified cellular metabolism in cancer cells,
which suggested an increased rate of glycolysis and lactate production regardless of the presence
of oxygen.
2
This increased lactate secretion in turn stimulates the acidification of cancer cells’
surrounding environment, which subsequently benefits tumor-stroma interactions and enhances
tumor invasion.
3
It was found that tumor-derived lactate contributed to the tumor associated
macrophage polarization partitioning to the M2-like phenotype that was favorable to tumor
progression and aggression.
4
Lately, the remarkably acidic regions are revealed to overlap with
tumor highly proliferative and invasive regions accompanied with increased risk of tumor
angiogenesis.
5
Therefore, compared with non-cancer tissues, extracellular acidity is also
recognized as a pathological feature of cancerous tissues. In lung cancer, for example, the acidic
2
extracellular pH maintains Lewis lung carcinoma (LLCm1) cells to passage for at least 28
generations, resulting in a more invasive and metastatic phenotype.
6
Tumor cells adapt to the surrounding microenvironment and include the large demand for
nutrients and oxygen. Compared with healthy tissues, tumor tissues appear to be more dense, and
have more extensive fibrosis, as well as more sophisticated blood and lymphatic vessel
constructions, which characteristics are conducive to tumor cells acquiring more nutrients during
cell survival and proliferation.
1, 7
Other evident nutrient composition alterations include the
relatively lower intracellular glucose, elevated glutaminolysis, and increased extracellular protein
secretion, which all support the viability of tumor cells.
1
Another main feature of an alteration in
metabolism in the tumor microenvironment is hypoxia. In this case, the prolyl hydroxylase
domain (PHD)-hypoxia-inducible factor (HIF) axis is significantly reduced in the tumor
microenvironment, which in turn promotes epithelial-mesenchymal transition, production of
nuclear matrix proteins, angiogenesis, and glycolysis during cancer invasion and metastasis.
1
Furthermore, the metabolic restrictions in the tumor microenvironment such as decreased
glucose level, acidic pHs, hypoxia, and other suppressive metabolites will cause barriers to
cancer immunotherapy.
8
Since both the nutrition-deprivation and reduction in metabolism can
inhibit the response of effector T cells in the tumor’s niche, and successfully prevent immune
cells attacking of tumor cells in an immunotherapeutic anticancer strategy.
8
For instance, when
the increased nutrient consumption, shifted towards the energetic demands from tumor cells, the
CD8+ T cells will not be able to obtain enough energy to support their own growth and
infiltration into tumor niches. Correspondingly, the reduced intratumoral CD8+ T cells will limit
3
functional T cells’-mediated anti-tumor responses.
9
Besides, the metabolic reprogramming in the
tumor microenvironment is also associated with increased T-cell exhaustion, that impairs the
anti-tumor immunity during tumor development.
10
For this reason, targeting cancer metabolism
should be a selective immunotherapeutic strategy in the future.
1.2 Expression and Function of CD36 in Healthy Tissues
CD36 was initially identified as a platelet glycoprotein IV (GPIV) meditating thrombospondin-1
(TSP-1) binding in platelets, so CD36 is also a TSP-1 receptor.
11
Later on, CD36 was discovered
as a fatty acid translocase (FAT) and a scavenger receptor participating lipid metabolism.
12
The
human CD36 gene is about 46kb long including 15 exons and is located at chromosome
7q11.2.
13
The CD36 receptor is an 88kDa transmembrane glycoprotein consisting of a large
extracellular domain with three cysteine-linked disulfide bridges and a short intracytoplasmic tail
at each C- and N- termini.
14
The extracellular domain of CD36 is heavily glycosylated in order to
interact with other transmembrane proteins such as integrins and tetraspanins,
15
and also contains
a hydrophobic region that binds to many hydrophobic molecules such as fatty acids,
phospholipids, and cholesterol.
13, 15
Moreover, CD36 associates with several ligands of other
receptors, such as those containing a TSP-1 repeat domain, through the electrostatic interactions
of extracellular amino acids.
16
The intracellular domain of CD36 plays a crucial role in signal
transduction by associating with the Src family of nonrecptor tyrosine kinase proteins, such as
FYN and LYN.
17
As a surface protein, CD36 is widely presented on different mammalian cell types and has a
unique function based on its location. The most abundant CD36 protein is detected in adipocytes,
4
where CD36 is primarily distributed in lipid rafts and cooperates with caveolin-1 to mediate the
uptake of fatty acids.
18
CD36-regulates fatty acid (FA) uptake and oxidation, thus providing
energy for many physiological activities in the human body. Therefore, consistent to high energy
demands, the high expression of the CD36 protein is often observed in cardiac and skeletal
muscle tissue. In myocytes, both insulin and contraction help to translocate intracellular CD36 to
the plasma membrane, thus contributing to cellular long-chain fatty acid uptake.
19
CD36 is also
found in myeloid cells such as platelets, macrophages, and erythrocytes. Here, CD36 mainly
mediates the cell-to-cell and cell-to-matrix interaction in platelets, as well as initiating the
signaling pathway of the monocytes’ oxidative burst.
20
CD36 has been described to be present on
microvascular endothelial cells as a fatty acid transporter that participates in the modulation of
angiogenic activities through the phospho-AMP-activated protein kinase (AMPK) pathway.
21
In
gut tissue, the higher expression of CD36 protein was examined in duodenal and jejunal
enterocytes, as well as in the proximal intestinal mucosa. The relatively high expression of CD36
in enteroendocrine cells of the intestine is explained by its function in long-chain fatty acid
absorption.
22
In normal tissues, CD36 expression is relatively low in hepatocytes; however, a
high-fat diet promotes CD36 protein expression through the activation of the upstream regulator,
PPAR𝛾 . Finally, the increased FA uptake and accumulated triglyceride will cause the
development of hepatic steatosis.
23
In the ovary, the CD36 protein is involved in ovarian
angiogenesis and folliculogenesis, as a TSP-1 receptor.
24
1.3 The Roles of CD36 in Cancer
CD36 has attracted attention in cancer research in recent years due to its role in fatty acid
transport as a scavenger receptor. The overexpression of CD36 is sufficient to satisfy the high
5
energy requirements and vigorous metabolic activities needed for the expansion of cancer cells.
25
For example, amplified CD36 expression was demonstrated in the self-renewing tumorigenic
cancer stem cell (CSC) population of glioblastomas, where CD36 binds to the oxidized
phospholipids presented in glioblastoma and promotes the exposure of glioblastoma cells to
oxidized low-density lipoprotein, supporting the proliferation of CSCs.
26
In ovarian cancer, the
upregulation of CD36 facilitates exogenous FAs uptake, thus playing an important role in the
metabolic adaptation of ovarian cancer cells in the adipocyte-rich microenvironment for ovarian
cancer in the peritoneal cavity.
27
Similarly, CD36-regulated FA uptake is an essential energy
source to maintain esophageal squamous cell carcinoma proliferation and invasion.
28
CD36 expression in cancer compared with healthy tissues has been reported for several
malignancies; however, these studies showed conflicting results, with CD36 either being
downregulated or upregulated in different cancer cell types.
29
Defective CD36 expression was
reported in invasive breast cancer based on Clezardin’s study in 1993, which suggested that
CD36 deficiency contributed to tumor progression and invasion.
30
Similarly, another study
implied that CD36 expression was reduced 30-100 fold in more aggressive MDA-MB-231 breast
cancer cell lines compared to less aggressive cells. It is also noticed that CD36 expression is
decreased in hormone-dependent T47-D and MCF-7 breast cancer cell lines in response to
estradiol.
31
Meanwhile, CD36 expression was observed to be highly expressed in breast cancer
tissues according to a study in 2016, in which the upregulated CD36 expression was correlated
with breast cancer cells expansion by actively promoting uptake of monosaturated fatty acids
during cancer cell migration.
32
Furthermore, the decreased expression of CD36 in endothelial
cells and stromal cells of the tumor microenvironment was reported compared to the tumour cells
6
themselves. CD36 acts as a thrombospondin-1 (TSP-1) receptor in the microvascular
endothelium, in which the conserved CLESH domain of CD36 interacts with the TSP type 1
repeat (TSR-1) domain leading to anti-angiogenic activity.
33, 34
The loss of CD36 in endothelial
cells lead to reduced interaction between CD36, as a TSP-1 receptor, and proteins with the TSR
domain. As a result, tumor angiogenesis was promoted due to the elimination of the antitumor
effect of TSR.
29
Likewise, the reduction of CD36 in stroma tissue was important in the maintenance of the
premetastatic niche and facilitated metastatic progression.
29, 35
In glioblastomas, it was suggested
that three type-1 repeats (3TSR) bind to CD36 and causes a cytotoxic effect on tumor cells
through the activation of tumor necrosis factor-related apoptosis-inducing ligand.
36
However, It
is also found that increased thrombospondin-2 (TSP-2) glycoprotein binds to the CD36 receptor,
leading to the activation of matrix metalloproteinase (MMP-2), followed by the migration of
prostate cancer cells.
37
The loss of CD36 expression was also reported in colorectal cancer
(CRC), and the progressively decrease in CD36 expression was also associated with poor
survival of CRC patitents.
38
It is suggested that CD36 acted as a tumor suppressor, inhibiting
aerobic glycolysis. Since CD36 interacts with Glypcian 4 (GPC4), and the ubiquitination of
GPC4 will interrupt the β-catenin/c-myc signaling axis, followed by the suppression of
downstream glycolytic target genes such as GLUT1, HK2, PKM2 and LDHA.
38
On the other hand, the overactivated CD36 receptor upregulates palmitate acid production
through an AKT/GSK-3β/β-catenin signaling pathway, therefore promoting the metastasis of
gastric cancer.
39
Also, CD36 overexpression has been implicated to correlate with tumor
7
metastasis in multiple cancer cell lines, such as human oral squamous cell carcinoma cell line
and human amelanotic melanoma cell line.
40, 41
Additionally, a recent study that leveraged the
cancer genome atlas data has shown an association between CD36 expression with epithelial-
mesenchymal transition (EMT).
42
In hepatocellular carcinoma, the increased CD36 regulated FA
uptake is involved in the activation of Wnt and TGF-β signaling pathways, thus conducive to
upregulate downstream EMT.
43
EMT is implicated in the favor of carcinogenesis and tumor
metastasis through increasing the resistance of tumor cells against toxic agent-mediated
apoptosis.
44
Therefore, CD36 has been repeatedly proposed as a biomarker in cancer initiation
and progression due to its tumorigenic properties, but the exact role that CD36 plays in the
mechanism of tumorigenesis still requires more investigation.
1.4 The Role of CD36 in Tumor Immune Microenvironment
CD36 modulates tumor immunity through the regulation of the inflammatory response, antigen
presentation, phagocytosis, and immune adaptation.
25
Inflammation is characterized as one risk
factor triggering cancer development and promoting all stages of tumorigenesis.
45
The ligands of
the scavenger receptor CD36 assembles with toll-like receptors 4 and 6, as heterodimers, to
stimulate the release of inflammatory cytokines in an innate immune response.
46
Also, CD36, as
a pattern-recognition receptor, mediates the conversion of soluble endogenous ligands such as
oxidized low-density lipoprotein into cholesterol crystals or amyloid fibrils, which results in
lysosomal disruption and activates NOD-like receptor protein-3-based (NLRP3) inflammation.
47
One recent study suggested that CD36-deficient mice exhibited relatively low risk to get
concanavalin A induced hepatitis compared to the CD36-positive mice; consistently, with
8
decreased immune infiltrating cells and reduced inflammatory mediators in CD36-deficient
mice.
48
CD36-mediated antigen recognition can be beneficial to immune tolerance. A study indicated
that the CD36 scavenger receptor mediates the transfer of medullary thymic epithelial cells
(mTECs) derived cell surface antigen to CD8𝛼 +
dendritic cells (DCs), in order to promote the
thymic T cells’ allo-tolerance during bone marrow transplantation in mice.
49
Additionally, a
large number of triglycerides were examined in tumor-infiltrated dendric cells (DCs) that are less
functional compared to tumor-free dendritic cells.
50
Although the specific mechanism of CD36
related to dysfunctional DCs is still unclear, it is possible that CD36 plays an important role in
lipid deposition, resulting in dysfunctional antigen presentation in the activation of DCs.
25
Besides, CD36 also participates in phagocytic activity within the tumor immune environment.
For example, the conjugation of CD36 and av𝛽 5 integrin can bind to epitopes presented on the
apoptotic cells, leading to activation of cytotoxic CD8+ T cells.
51
Also, CD36 can directly
cooperate with av𝛽 3 integrin of macrophages to consume apoptotic cells.
25
Therefore, CD36
expression may contribute to the eradication of some malignant or apoptotic cells through
phagocytosis.
In tumor tissues, CD36-driven lipid metabolic reprogramming can modulate the functions of
tumor-associated immune cells, thus leading to promotion of tumor growth and progression. For
example, CD36 mediates the uptake of triacylglycerol substrates and their subsequent lipolysis
by lysosomal acid lipase therefore promoting oxidative phosphorylation, and also increasing
spare respiratory capacity (SRC), together leading to the activation of the differentiation of M2-
9
like tumor-associated macrophages.
52
However, the increased M2 macrophage ratio is
responsible for metastatic cancer since M2 macrophages can support tumor growth by secreting
adrenomedullin (ADM) and vascular endothelial growth factors (VEGFs), as well as expressing
immunosuppressive molecules such as IL-10, programmed death-ligand 1 (PD-L1), and
transforming growth factor-β (TGFβ).
53, 54
In this case, targeting CD36 could be a
immunotherapeutic strategy for cancer treatment.
1.5 The Association Between CD36 Expression and Clinical Outcome in Cancer
Obese cancer patients tend to exhibit worse clinical outcomes compared with nonobese patients.
It is found that obese adipocytes upregulate free fatty acid translocase CD36 expression, giving
rise to cancer cells’ development of stemness features. In this case, CD36 upregulation appears
to be a potential association with poor prognosis and resistance to radiotherapy and
chemotherapy.
55
For example, CD36
+
leukemic stem cells residing in gonadal adipose tissue
builds a protective mechanism utilizing the continuous fuels from fatty acids to escape the
cytotoxic effects of chemotherapy.
56
Moreover, upregulated CD36 expression in cytarabine
(AraC)-resistant AML cells, accompanied with abundant oxidative activities in mitochondria,
consistent with their high mitochondrial fatty acid β-oxidation (FAO) and oxidative
phosphorylation (OXPHOS) status. The interruption of CD36-FAO-OXPHOS axis suggests a
remarkable enhancement of AraC-induced antileukemic effects.
57
The chemoresistance caused
by upregulated CD36 is also observed in the clinical specimens resected from pancreatic ductal
adenocarcinoma (PDAC) patients treated with gemcitabine adjuvant chemotherapy. Targeting
CD36 with small interfering RNA inhibited the synthesis of anti-apoptosis proteins and reduced
their chemoresistance in gemcitabine-resistant PDAC cell lines.
58
Besides, upregulated cDNA
10
expression levels of the FA transporter CD36 was reported in breast cancer cells that had
acquired resistance to lapatinib-targeted therapy. For HER2-positive breast cancer patients
treated by lapatinib, the endogenous FA lipogenic pathway was blocked by an HER2 inhibitor;
however, lapatinib-resistant cancer cells alternatively increase exogenous FA uptake through the
CD36 lipid transporter, thus assisting cancer cells to overcome the deprivation of de novo
lipogenesis and promoting cancer cells’ survival.
59
Together these results provide important
insights to anti-CD36 treatment in the alleviation of cancer drug resistance.
11
Chapter 2: Patterns of CD36 Expression in Healthy and Malignant Tissues Dictate
Potential Opportunities for Cancer Targeted Therapy
2.1 Introduction
Cellular metabolism reprogramming plays an important role in carcinogenesis and development,
since the alternate metabolic pathways will directly or indirectly induce oncogenic mutations
such as the abnormal expression of various genes and proteins, or uncontrolled cytokine
secretion and dysregulated signaling pathways.
60
The common cancer-associated features of
metabolic network alteration include reduced levels of oxygen, glucose and amino acids,
increased extracellular acidification, high demand of nutrients and nitrogen acquisition, and
unregulated metabolite-driven gene expression in the tumor microenvironment.
61
It is worth
noting that lipid metabolism reprogramming is among the most evident metabolic alterations in
cancer, since the altered lipid metabolism usually contributes to the cancer cells’ acquiring the
necessary nutrients from the harsh tumor microenvironment, and to utilize these nutrients to
support their survival and invasion.
61
In PDAC, oncogenic KRAS suppresses the expression of
hormone-sensitive lipase (HSL), which in turn prevents the lipolysis of excess lipids stored in
organelles as lipid droplets. Subsequently, the accumulated lipid droplets are utilized as fuel
sources by tumor cells, leading to increased FA oxidation and oxidative metabolism that drives
tumor cells’ migration.
62
In addition, the lipid metabolic alterations are also associated with
interfering with the immunosuppression of tumor cells. Because the modulation of lipid
metabolism will cause increased nutrients shifting toward tumor cells instead of immune cells in
TME, this switch often impedes the anti-tumor effects of functional immune cells.
63
One study
indicated that tumor cells actively responded to a high fat diet (HFD) by upregulating the
12
metabolic pathway that mobilizes free fatty acids, which results in more fuel sources partitioning
to cancer cells; however, the deficient FAs uptake by CD8
+
T cells would not be enough to meet
the energy consumption for their infiltration in a tumoral niche. As a result, HFD-induced
obesity reshapes lipid metabolism in the TME, then promotes tumor growth by inhibiting T cell
function and helping tumor cells to escape anti-tumor immunity.
9
Current studies have shown that CD36 as a fatty acid transporter protein plays a key metabolic
role in FA uptake and utilization. Except FA, the CD36 binding domain is able to recognize
multiple lipid ligands such as phospholipids, diacylglycerol, and cholesterol, and also bind to
native and oxidized lipoproteins.
15
In this case, CD36-dependent lipid metabolism implicates a
potential role in the pathogenesis of metabolic disorders. Indeed, it is identified that CD36
upregulation is correlated with the development of HFD-induced obesity and type 2 diabetes in a
mouse model. One study report that greatly increased CD36 mRNA expression is detected in
BDNF
M/M
mice that carry methionine instead of valine at codon 66 of human brain-derived
neurotropic factor (BDNF) gene. Compared with wild type BDNF
V/V
mice, BDNF
M/M
mice
exhibit significantly increased body weight; elevated CD36 expression facilitates FA uptake and
lipid accumulation, consequently leading to increased obesity and decreased glucose tolerance.
64
Additionally, CD36-mediated lipid accumulation may be associated with adaptive metabolic
plasticity and the immunosuppressive effects of tumor cells. For example, upregulated CD36 in
tumor infiltrated Treg cells supports mitochondrial fitness and biogenesis and elevates the NAD-
to-NADH ratio through the peroxisome proliferator-activated receptor-β (PPAR-β) signaling
pathway, which in turn, increases pyruvate production and then improves the survival of
intratumoral Treg cells in an acidic environment. Eventually, the enhanced proliferation of
13
intratumoral Treg cells will allow tumor cells to counteract immunosuppressive features in
TME.
65
Considering the increased evidence supporting the role of CD36 in cancer, and the wide interest
in targeting this receptor in a broad variety of malignancies, in this chapter, I conduct a
comprehensive analysis that explores public databases to identify patterns of aberrant expression
of CD36 in different types of cancer samples and normal tissues. My results establish that CD36
is either upregulated or downregulated in various cancer types, but the immune vulnerable
property of CD36 can be exploited in developing therapeutic approaches that target CD36.
2.2 Experimental Approach and Methods
2.2.1 CD36 Gene Expression Analysis
The expressions of CD36 mRNA in different tissues were obtained from the GETx Portal
database (https://www.gtexportal.org/home/). The expression values are shown in transcripts per
million (TPM). The analysis of CD36 differential expression in both healthy and primary tumor
tissues in 35 different cancer types retrieved from the cancer genome atlas (TCGA) was
performed by the UCSC Xena database (https://xenabrowser.net/),
66
and only the cancer types
with statistically significant differences in CD36 expression were presented. The CD36 mRNA
expression in normal hematopoiesis data was obtained from haemosphere
(https://www.haemosphere.org/).
67
The expression data was shown in transcripts per million
(TPM). GraphPad Prism 7.0 (GraphPad Software, Inc.) was used to generate the figures and
perform the statistical analyses.
14
2.2.2 DNA Methylation Analysis
The UALCAN database was used to analyze the differential promoter methylation status of
CD36 in both healthy and tumor tissues in multiple cancer types by comparing their medium
beta value.
68
Different beta value cut-off was considered to indicate unmethylation (beta value:
0), hypo-methylation (beta value: 0.25 - 0.3), hyper-methylation (beta value: 0.5 - 0.7), and fully
methylation (beta value: 1).
69, 70
2.2.3 Survival Analysis
Patient's mRNA expression level and clinical data in 32 different cancer types were downloaded
from the TCGA, Pan Cancer Atlas studies in cBioPortal (https://www.cbioportal.org/).
71, 72
I
created the overall survival curves for cancer patients by comparing groups based on their CD36
mRNA expression levels corresponding to normalized z-scores, in which samples with high
expression had a z-score above than zero and samples with low expression had negative z-score
values. Then, I generated the Kaplan-Meier survival curves using GraphPad Prism 8.0
(GraphPad Software, Inc.) by comparing the median survival in months between cancer patients
with up- or downregulated CD36.
2.2.4 Tumor - Immune Interaction Analysis
I used TIMER 2.0 database (http://timer.cistrome.org/), which implements a deconvolution
analytical method for visualization of the correlation between gene expression and the immune
cells infiltration level in diverse cancer types.
73-75
The statistically significant spearman's rho
values showing CD36 expression and its correlation with six types of immune infiltrated cells
(CD8+ T cells, CD4+ T cells, B cells, neutrophils, macrophages, and dendritic cells) across
15
various cancer types after adjusting for tumor purity, the correlation between immune infiltration
level and gene expression were presented in a heatmap. I also examined the correlation between
CD36 expression with the common gene markers presented in immune suppression. The effect
of CD36 gene mutations on immune cell infiltration across different cancer types and individual
immune cell types was also analyzed in TIMER 2.0. TIMER 2.0 used Kruskal-Wallis to
calculate the P value when comparing the log-fold changes of immune infiltration level between
tumor samples with altered and unaltered CD36 gene.
2.2.5 Genetic Mutations and Chromosome Abnormalities Analysis
I analyzed the data of 185 cancer studies obtained from TCGA PanCancer Atlas and other non-
redundant studies including all 48045 samples from cBioPortal to present the summary graphic
and table for all nonsynonymous mutations identified in CD36 gene. I obtained and compared
the value of CD36 mRNA expression across different genetic mutations and wild types for each
sample after analyzing these cancer studies. Moreover, by analyzing the 10967 samples in 32
studies from TCGA PanCancer Atlas, I scrutinized the chromosomal status between CD36 gene
altered groups and unaltered groups, and the Chi-squared test was used to compare their
differences.
2.2.6 Statistical Analysis
CD36 differential expression in both normal and tumor tissues was compared with the t test to
calculate the significance of difference between the two groups. PERL script with
Comprehensive Perl Archive Network (CPAN) module “Statistics::TTest”,
68
was used. Two-
tailed student t-test was used to identify the significance in the methylation status of CD36
16
analysis from the UALCAN database.
70
As for comparing the CD36 mRNA expression
according to the genetic mutation, I used GraphPad Prism 8.0 to perform Kruskal-Wallis test to
calculate the P value between groups and within groups. GraphPad Prism 7.0 also used a Log-
rank (Mantel-Cox) test to calculate the P value in our overall survival analysis. The statistical
significance threshold was set at p < 0.05 for all analysis.
2.3 Results
2.3.1 CD36 Expression Patterns in Normal Tissues
CD36 is widely distributed in all kinds of tissues of the body, where the CD36 gene expression
levels by measuring transcripts per million (TPM) were obtained from GTEx Portal. As shown in
Figure 2.1, the highest expressed CD36 gene was observed in adipose tissue (adipose -
subcutaneous: n = 663, median TPM = 535; adipose - visceral: n = 541, median TPM = 549) and
breast (breast - mammary tissue: n = 459, median TPM = 362.5). CD36 was also enriched in
heart muscle (heart - appendage: n = 429, median TPM = 80.18; heart - left ventricle: n = 432,
median TPM = 141.4), spleen (n = 241, median TPM = 106.6), and thyroid gland (n = 653,
median TPM = 70.97). The relatively lower expression of CD36 gene was found in skeletal
muscle (n = 803, median TPM = 43.80), lung (n = 578, median TPM = 43.41), and coronary
artery (n = 240, median TPM = 43.39). A little bit of CD36 gene expression could be detected in
other tissues, such as liver (n = 226, median TPM = 3.420), uterus (n = 142, median = 3.004),
prostate (n = 245, median TPM = 2.513), ovary (n = 180, median TPM = 2.259), pituitary (n =
283, median TPM = 1.191), and brain tissue (hypothalamus: n = 202, median = 0.9804). The
expression values of CD36 gene across multiple tissues were listed in Table 2.1.
17
Figure 2.1 CD36 pattern of expression in normal tissues
Gene expression for CD36 in various kinds of healthy tissues in the human body is determined by GTEx Portal.
The expression value is shown by TMP (Transcripts Per Million). Box plots are shown as median, 25th, and 75th
percentiles, and points are displayed as the outliers that are above or below 1.5 times the interquartile range.
Table 2.1 Gene Expression for CD36 in Normal Tissues
Tissue Type
Sample
Number
Transcripts Per Million
(TPM)
Adipose
Subcutaneous 663 535
Visceral 541 549
Breast Mammary 459 362.5
Heart
Left Ventricle 432 141.4
Appendage 429 80.18
Spleen 241 106.6
Thyroid Gland 653 70.97
Skeletal Muscle 803 43.80
Lung 578 43.41
Artery Coronary 240 43.39
Nerve Tibial 619 27.10
Esophagus
Musculars 515 26.26
Gastroesophageal Junction 375 25.78
Mucosa 555 2.852
Colon
Sigmoid 373 19.18
Transverse 406 15.36
18
Skin
Sun Exposed (Lower Leg) 701 17.48
Not Sun Exposed (Suprapubic) 604 10.83
Whole Blood 755 16.57
Small
Intestine
Terminal ILeum 187 15.99
Stomach 359 15.12
Bladder 21 13.10
Artery
Tibial 663 12.08
Aorta 432 7.709
Minor Salivary Gland 162 8.115
Vagina 156 7.657
Adrenal Gland 258 6.443
Pancreas 328 5.386
Kidney
Medulla 4 5.061
Cortex 85 0.8106
Fallopian Tube 9 4.784
Liver 226 3.420
Uterus 142 3.004
Testis 361 2.522
Prostate 245 2.513
Cervix
Ectocervix 9 2.346
Endocervix 10 2.232
Ovary 180 2.259
Pituitary 283 1.191
Brain
Hypothalamus 202 0.9804
Hippocampus 197 0.6567
Amygdala 152 0.5032
Spinal Cord (Cervical c-1) 159 0.4931
Caudate (Basal Ganglia) 246 0.3642
Frontal Cortex 209 0.2937
Anterior Cingulate Cortex (BA24) 176 0.2911
Cortex 255 0.2674
Substantia Nigra 139 0.2327
Cerebellum 241 0.2303
Nucleus Accumbens (Basal
Ganglia)
246 0.2228
Putamen (Basal Ganglia) 205 0.2042
Cerebellar Hemisphere 215 0.1952
Cells
Cultured Fibroblasts 504 0.5840
EBV-Transformed Lymphocytes 174 0.1610
19
2.3.2 CD36 Expression Patterns in Cancer
I further analyzed the difference of CD36 expression in different human tumors together with
their adjacent normal tissues using the UCSC Xena database. I observed that CD36 expression
was highly amplified in kidney renal clear cell carcinoma (4.8-fold, P < 0.0001; Figure 2.2A).
I also noticed the significantly upregulated CD36 expression from tumor samples in glioblastoma
multiforme (4.8-fold, P = 0.0044; Figure 2.2B). The downregulated CD36 expression was
observed in most cancer types. For example, I found that CD36 was greatly downregulated in
breast invasive carcinoma (27.1-fold, P < 0.0001; Figure 2.2C). The significantly reduced CD36
expression was also observed in the following tumors such as colon adenocarcinoma (8.0-fold, P
< 0.0001; Figure 2.2D), cholangiocarcinoma (15.0-fold, P < 0.0001; Figure 2.2E), lung
adenocarcinoma (12.6-fold, P < 0.0001; Figure 2.2F), lung squamous cell carcinoma (13.1-fold,
P < 0.0001; Figure 2.2G). Besides, reduced CD36 expression was also detected in samples taken
from tumor tissues in head and neck squamous cell carcinoma (2.2-fold, P = 0.0018; Figure
2.3A), kidney renal papillary cell carcinoma (2.2-fold, P < 0.0001; Figure 2.3B), prostate
adenocarcinoma (2.0-fold, P < 0.0001; Figure 2.3C), stomach adenocarcinoma (4.3-fold, P <
0.0001; Figure 2.2D), thyroid carcinoma (mean: 2.7-fold, P < 0.0001; Figure 2.2E) and uterine
corpus endometrial carcinoma (5.0-fold, P < 0.0001; Figure 2.3F).
20
Figure 2.2 Analysis of CD36 differential expression in cancers
CD36 expression data was obtained from UCSC Xena – GDC TCGA database. The expression level was shown
by FPKM-UQ (fragments per kilobase of transcript per million mapped reads upper quartile). The mean fold
change was compared between normal tissue and tumor tissue in (A) kidney renal clear cell carcinoma, (B)
glioblastoma, (C) breast invasive carcinoma, (D) colon adenocarcinoma, (E) cholangiocarcinoma, (F) lung
adenocarcinoma, and (G) lung squamous cell carcinoma. The differences between groups were analyzed by
unpaired t-test. (****, P < 0.0001; ***, P < 0.001; **, P < 0.01; *, P < 0.05)
21
Figure 2.3 Analysis of CD36 differential expression in cancers (supplementary)
CD36 expression data was obtained from UCSC Xena – GDC TCGA database. The expression level was shown
by FPKM-UQ (fragments per kilobase of transcript per million mapped reads upper quartile). The mean fold
change was compared between normal tissue and tumor tissue in (A) head and neck cancer, (B) kidney renal
papillary cell carcinoma, (C) prostate cancer, (D) stomach cancer, (E) thyroid cancer, and (F) endometrial cancer.
The differences between groups were analyzed by unpaired t-test. (****, P < 0.0001; ***, P < 0.001; **, P <
0.01; *, P < 0.05)
2.3.3 CD36 Promoter DNA Methylation Status in Different Cancers
To explore the epigenetic role of CD36 in the occurrence of cancer, I measured the methylation
status of CD36 in relation to tumor progression. I found that the CD36 promoter
hypermethylated status was corresponding to the lower CD36 gene expression in primary tumor
by comparison with healthy tissues for breast invasive carcinoma (median beta value 0.828 vs.
0.781, p < 0.0001; Figure 2.4A), lung adenocarcinoma (median beta value 0.744 vs. 0.641, p <
0.0001; Figure 2.4B), and lung squamous cell carcinoma (median beta value 0.796 vs. 0.666, p <
0.0001; Figure 2.4C). However, I noticed a significantly inconsistent association between the
CD36 gene expression and CD36 promoter methylation status in tumor tissues for several other
cancers. For example, CD36 promoter was found to be hypermethylated, yet the gene was
overexpressed in both kidney renal clear cell carcinoma (median beta value 0.8 vs. 0.872, p <
22
0.0001; Figure 2.4D). Also, I found that CD36 promoter was hypomethylated, but the gene
expression was suppressed in all colon adenocarcinoma (median beta value 0.766 vs. 0.837, p <
0.0001; Figure 2.4F), head and neck squamous cell carcinoma (median beta value 0.806 vs.
0.807, p = 0.0017; Figure 2.4G), and prostate adenocarcinoma (median beta value 0.878 vs.
0.887, p < 0.0001; Figure 2.4H).
Figure 2.4 Promoter methylation status of CD36 where CD36 has aberrant expression in tumor tissues
compared to healthy tissues
The beta value is used to evaluate the methylation level, in which 0 stands for unmethylation, beta value between
0.25 to 0.3 is considered as hypomethylation, and beta value between 0.5 to 0.7 is considered as
hypermethylation. The methylation analysis was performed in (A) breast invasive carcinoma, (B) lung
adenocarcinoma, (C) lung squamous cell carcinoma, (D) kidney renal clear cell carcinoma, (E) liver
hepatocellular carcinoma, (F) colon adenocarcinoma, (G) head and neck squamous carcinoma, and (H) prostate
adenocarcinoma. The statistical significance of median difference was analyzed by Student’s t-test. (****, P <
0.0001; ***, P < 0.001; **, P < 0.01; *, P < 0.05)
23
2.3.4 CD36 Expression is Correlated with Clinical Outcomes in Specific Cancer Types
To further investigate and characterize the prognostic value of CD36 in cancer patients, I
compared survival curves between patients with up- and downregulated CD36 in each different
cancer type from the TCGA Pan Cancer Atlas dataset. I found that the CD36 mRNA differential
expression was significantly associated with overall survival in several types of cancers. For
example, I observed that patients with lower CD36 expression exhibited a significantly higher
possibility of survival in both stomach carcinoma (median survival, OS: 42.54 vs. 18.35 months;
P = 0.0003; Figure 2.5A) and uveal melanoma (median survival, OS: 51.98 vs. 36.59 months; P
= 0.0039; Figure 2.5B). I also found that patients with lower expression levels of CD36 mRNA
showed slightly better survivability in several other types of cancer, such as bladder urothelial
carcinoma, (median survival, OS: 46.78 vs. 26.14 months; P = 0.0286; Figure 2.6A), colon
adenocarcinoma (median survival, OS: 92.74 vs. 67.30 months; P = 0.0055; Figure 2.6B), and
kidney renal papillary cell carcinoma (median survival, OS: undefined months; P = 0.0177;
Figure 2.6C). Adversely, the analysis results suggested that amplified CD36 mRNA expression
contributed to higher survival chances in pancreatic adenocarcinoma (median survival, OS:
19.50 vs. 43.79 months; P = 0.0230; Figure 2.5C). Also, patients with head and neck squamous
cell carcinoma with higher CD36 expression level tend to live longer. I analyzed the increased
and decreased CD36 mRNA expression in both HPV+ and HPV- patients, and I observed that
HPV+ patients exhibited longer overall survival in comparison with HPV- patients in general.
When I involved the factor of CD36 gene expression, I found that HPV- cancer patients with
intensified CD36 expression showed higher survival probability and the situation was similar in
HPV+ patients (median survival, OS: 35.54 vs. 67.86 months in HPV-, 68.48 vs. undefined
months in HPV+; P = 0.0122; Figure 2.5D). For kidney renal clear cell carcinoma, cancer
24
patients with upregulated CD36 mRNA level also showed longer overall survival compared with
patients with reduced CD36 mRNA expression (median survival, OS: 90.87 vs. undefined
months; P = 0.0291; Figure 2.6D).
Figure 2.5 Survival analysis of cancer patients with low or high CD36 mRNA expression
The overall survival curve of cancer patients with high (Z >0) and low (Z < 0) CD36 mRNA expression in (A)
stomach adenocarcinoma, (B) uveal melanoma, (C) pancreatic adenocarcinoma, and (D) head and neck squamous
cell carcinoma. The statistical significances were analyzed by log-rank (Mantel-Cox) test.
25
Figure 2.6 Survival analysis of cancer patients with low or high CD36 mRNA expression (supplementary)
The overall survival curve of cancer patients with high (Z >0) and low (Z < 0) CD36 mRNA expression in (A)
bladder urothelial carcinoma, (B) colon adenocarcinoma, (C) kidney renal papillary cell carcinoma, and (D)
kidney renal clear cell carcinoma. The statistical significance was analyzed by a log-rank (Mantel-Cox) test.
2.3.5 CD36 Expression Pattern in Normal Hematopoiesis
Next, I examined CD36 mRNA expression patterns in normal hematopoiesis using haemosphere
database. The expression level of CD36 mRNA was detected in B cell lineage, dendritic cell
lineage, eosinophil lineage, macrophage lineage, neutrophil lineage, NK cell lineage, and T cell
lineage (Figure 2.7). The relatively high expression of CD36 was found in dendritic cell lineage
including myeloid dendritic cell, CD123+ myeloid dendritic cell, and plasmacytoid dendritic
cell. CD36 expression also was enriched in macrophage lineage especially for monocytes, as
well as in NK cell lineage. A few expression levels of CD36 were examined in memory B cell,
naïve B cell, eosinophil, neutrophil, CD4+ T cell and CD8+ T cell.
26
Figure 2.7 CD36 pattern of expression in normal hematopoiesis
Gene expression for CD36 in hematopoiesis from healthy human tissue. The expression value was obtained from
the Haemopedia-Human-RNA-seq database and presented by the haemosphere database. Box plots are shown as
median, 25
th
and 75
th
percentiles, min and max. Points are displayed as the expression level of each single sample.
One-way ANOVA (Kruskai-Wallis test) was performed to compare the median difference of CD36 expression
among each cell type.
Memory B Cell
Naive B Cell
Myeloid Dendritic Cell
CD123+ Myeloid Dendritic Cell
Plasmacytoid Dendritic Cell
Eosinophil
Monocyte
Non-classical monocyte
Neutrophil
Natural Killer Cell
CD4+ T Cell
CD8+ T Cell
0
2
4
6
8
10
log2 (tpm+1)
Memory B Cell
Naive B Cell
Myeloid Dendritic Cell
CD123+ Myeloid Dendritic Cell
Plasmacytoid Dendritic Cell
Eosinophil
Monocyte
Non-classical monocyte
Neutrophil
Natural Killer Cell
CD4+ T Cell
CD8+ T Cell
P = 0.0007
27
Table 2.2 Gene Expression for CD36 in Normal Hematopoiesis
Cell Lineage Cell Type Expression Value
B cell lineage
Memory B cell
0.18177008
0.279170266
0.133334756
Naïve B Cell
0.134556497
0.169576854
0.177985027
0.045445825
0.171803692
Dendritic Cell Lineage
Myeloid Dendritic Cell
0.277466716
0.086241028
3.616737959
CD123+ Myeloid Dendritic Cell
2.999762857
4.449902967
Plasmacytoid Dendritic Cell
7.622123982
7.37265991
Eosinophil Lineage Eosinophil
0.090588626
0.824236436
Macrophage Lineage
Monocyte
5.919857058
6.55801769
6.469877036
5.464116895
6.508019369
Non-classical monocyte
3.733659815
3.112636502
Neutrophil Lineage Neutrophil
0.624386612
0.180974179
0.242162973
NK Cell Lineage Natural Killer Cell
1.933072672
2.984094021
3.001601713
1.858575266
4.581171506
T Cell Linage
CD4+ T Cell
0.259719056
0.127490028
0.175999732
0.090679677
0.083415993
CD8+ T Cell
0.164879146
0.120801741
0.176053247
0.041367607
0.229278112
28
2.3.6 CD36 mRNA Abnormal Expression is Correlated with Immune Cells Infiltration, and
May Influence Patient’s Clinical Outcome
I noticed an inconsistency between the CD36 mRNA expression in tumor samples and the
survival of these cancer patients in some cancer types. For example, I had observed the
deregulated CD36 expression in cancer patients in STAD, COAD, and KIRP; however, the
cancer patients exhibited longer survival with a lower expression of CD36 compared with those
who had a higher CD36 mRNA expression. In KIRC patients, where CD36 is generally
upregulated compared with normal tissue, high CD36 expression was still associated with longer
overall survival. I wondered whether CD36 expression in the tumor immune environment may
explain the conflicting correlations between CD36 expression and overall survival. I analyzed
the correlation of CD36 expression level with immune infiltration level in 40 types of cancer
with tumor purity adjustment. I identified a significantly positive correlation between CD36
expression and tumor immune infiltration levels. In the following cancers: COAD, KIRC, KIRP,
and STAD, Figure 2.8 showed the positive correlation between CD36 expression with either or
all CD8+ T cells, CD4+ T cells, B cells, and neutrophils infiltrations; however, CD36 is usually
minimally expressed in all four types of immune cells during healthy hematopoiesis.
Additionally, I also examined the significant association between CD36 expression with either or
all the four types of immune cell infiltrates in other cancers, particularly in BRCA-LumA, LGG,
LUAD, LUSC, PAAD, PRAD, and READ. While in KIRC, SKCM-metastasis, THCA, and
UVM, I also detected a negative correlation between CD36 expression with the immune
infiltration level of dendritic cells which have high expression of CD36 in normal tissues.
Differently with DC cells, I found that CD36 expression only has significantly positive
correlation with immune infiltration level in macrophages in different cancers, although CD36
29
expression was concentrated in macrophages of healthy tissue. I also summarized their
correlation coefficient in Table 2.3 for each type of cancer.
Figure 2.8 Correlation between CD36 expression and immune infiltration level in different cancers
The heatmap indicates the Spearman’s correlation coefficient (𝜌 ) of CD36 expression with immune infiltration
level in diverse cancer types with tumor purity adjustment. The red color refers to positive correlation, the blue
color refers to negative correlation, and the blank box means no significant correlation was examined. Also, the
more darkness of the color represents the higher possibility of correlation.
30
Table 2.3 Correlation Analysis between CD36 Expression and Tumor Cells Immune
Infiltration Levels in TCGA Cancer
CD8+ T Cells CD4+ T Cell B Cell Neutrophil Macrophage Dendritic Cell
Cancer Type rho P rho P rho P rho P rho P rho P
ACC (n=79) 0.445 0.000 -0.168 0.156 0.348 0.003 0.062 0.604 0.290 0.013 0.081 0.494
BLCA (n=408) 0.105 0.043 -0.108 0.038 -0.046 0.375 0.135 0.010 0.296 0.000 0.022 0.669
BRCA (n=1100) 0.140 0.000 0.078 0.013 -0.132 0.000 0.016 0.605 0.153 0.000 0.135 0.000
BRCA-Basal
(n=191)
0.085 0.265 0.128 0.091 -0.162 0.033 0.046 0.543 0.246 0.001 0.068 0.372
BRCA-Her2
(n=82)
0.009 0.941 0.038 0.752 -0.071 0.555 0.019 0.875 0.118 0.323 -0.078 0.515
BRCA-LumA
(n=568)
0.177 0.000 0.128 0.004 -0.214 0.000 0.216 0.000 0.103 0.019 0.343 0.000
BRCA-LumB
(n=219)
-0.048 0.506 0.104 0.150 -0.045 0.538 -0.115 0.111 0.081 0.266 0.068 0.346
CESC (n=306) 0.028 0.642 -0.164 0.006 -0.068 0.260 0.048 0.428 0.134 0.025 0.090 0.135
CHOL (n=36) 0.217 0.211 -0.172 0.324 0.319 0.062 0.128 0.463 -0.161 0.357 0.283 0.099
COAD (n=458) 0.224 0.000 0.346 0.000 -0.115 0.057 0.487 0.000 0.587 0.000 0.585 0.000
DLBC (n=48) 0.316 0.044 -0.017 0.915 -0.342 0.029 0.366 0.018 -0.040 0.802 0.182 0.255
ESCA (n=185) 0.118 0.113 -0.097 0.194 -0.076 0.313 0.154 0.039 0.285 0.000 0.212 0.004
GBM (n=153) 0.329 0.000 -0.313 0.000 0.058 0.497 0.154 0.073 0.165 0.054 0.133 0.120
HNSC (n=522) -0.055 0.224 -0.098 0.030 0.042 0.354 -0.017 0.704 0.308 0.000 0.076 0.092
HNSC-HPV-
(n=422)
-0.103 0.039 -0.024 0.629 0.024 0.634 0.047 0.350 0.262 0.000 0.030 0.547
HNSC-HPV+
(n=98)
-0.074 0.490 -0.217 0.042 0.076 0.481 -0.275 0.009 0.350 0.001 0.071 0.510
KICH (n=66) 0.198 0.114 -0.033 0.796 -0.086 0.497 0.083 0.511 0.196 0.118 -0.029 0.816
KIRC (n=533) 0.219 0.000 0.140 0.003 -0.086 0.064 0.215 0.000 0.144 0.002 -0.105 0.024
KIRP (n=290) 0.449 0.000 -0.054 0.391 0.275 0.000 0.357 0.000 0.288 0.000 0.393 0.000
LGG (n=516) -0.083 0.070 0.101 0.027 -0.003 0.954 0.209 0.000 0.144 0.002 0.114 0.013
LIHC (n=371) 0.205 0.000 -0.200 0.000 -0.126 0.019 0.059 0.272 -0.040 0.460 0.019 0.732
LUAD (n=515) 0.116 0.010 -0.030 0.503 -0.076 0.093 0.244 0.000 0.390 0.000 0.197 0.000
LUSC (n=501) 0.107 0.019 -0.052 0.253 -0.038 0.407 0.188 0.000 0.327 0.000 0.258 0.000
MESO (n=87) 0.143 0.190 -0.018 0.871 -0.045 0.684 0.034 0.756 0.334 0.002 0.135 0.220
OV (n=303) 0.122 0.054 -0.001 0.994 -0.267 0.000 0.248 0.000 0.271 0.000 -0.111 0.080
PAAD (n=179) 0.623 0.000 0.180 0.019 0.048 0.532 0.479 0.000 0.356 0.000 0.534 0.000
PCPG (n=181) 0.285 0.000 0.041 0.600 0.103 0.187 0.351 0.000 0.094 0.228 0.318 0.000
PRAD (n=498) 0.297 0.000 0.110 0.025 0.058 0.240 0.293 0.000 0.214 0.000 0.310 0.000
READ (n=166) 0.350 0.001 0.320 0.002 0.063 0.558 0.456 0.000 0.491 0.000 0.605 0.000
SARC (n=260) -0.005 0.938 0.296 0.000 -0.128 0.046 0.308 0.000 0.351 0.000 0.014 0.825
SKCM (n=471) -0.177 0.000 0.158 0.001 -0.187 0.000 0.277 0.000 0.275 0.000 -0.085 0.069
SKCM-Metastasis
(n=368)
-0.230 0.000 0.180 0.001 -0.155 0.003 0.225 0.000 0.296 0.000 -0.111 0.037
SKCM-Primary
(n=103)
-0.137 0.169 0.010 0.923 -0.362 0.000 0.305 0.002 0.186 0.061 -0.172 0.083
STAD (n=415) 0.429 0.000 0.300 0.000 0.052 0.310 0.405 0.000 0.582 0.000 0.404 0.000
TGCT (n=150) -0.217 0.008 0.146 0.077 -0.420 0.000 0.117 0.158 -0.010 0.902 0.006 0.943
THCA (n=509) 0.038 0.398 0.020 0.657 0.070 0.124 -0.246 0.000 0.399 0.000 -0.468 0.000
THYM (n=120) -0.230 0.013 -0.053 0.571 -0.146 0.119 0.138 0.141 0.182 0.052 -0.078 0.410
UCEC (n=545) 0.073 0.499 -0.030 0.780 0.076 0.482 0.147 0.171 0.247 0.020 0.224 0.036
UCS (n=57) 0.284 0.039 -0.209 0.133 -0.130 0.354 -0.319 0.020 0.119 0.394 0.155 0.269
UVM (n=80) 0.344 0.002 -0.150 0.194 -0.203 0.077 0.144 0.211 0.075 0.519 -0.372 0.001
31
2.3.7 CD36 Expression is Correlated with the Immune Suppression in Tumor
Microenvironment
To explore the association between CD36 expression and the characteristics of tumor immune
cells infiltration, I analyzed the correlation between CD36, and gene markers presented on
different subsets of immune cells using TIMER 2.0. I first investigated the correlation between
CD36 expression and the common gene markers including CD160, CD244, CD28, GZMB,
HAVCR2, KLRG1, LAG3, and PDCD1 located on anergic, exhausted, or senescent T cells. I
found that CD36 had a strong positive correlation with these gene markers in several cancers
such as BLCA, BRCA and its subtypes, COAD, KIRP, LGG, OV, PAAD, READ, STAD,
UCEC, and UVM (Figure 2.9A). These results indicated that CD36 is closely related to T cell
exhaustion in tumor suppression. I also observed the significantly negative correlation between
CD36 expression with the immune profiles’ gene markers in HNSC-HPV+ and TGCT; and for
KIRC and THCA, both positive and negative correlation between CD36 and immune gene
markers were observed, but the level of positive correlation was much higher in these two
cancers (Figure 2.9A). I also examined the correlation between CD36, and the gene markers
presented on Treg cells. In addition to the above cancer types shown to exhibit positive
correlation between CD36 and exhausted T cell gene markers, I also detected a significantly
positive correlation between CD36 and Treg cell gene markers in CESC, HNSC, LIHC, LUAD,
LUSC, PRAD, and SKCM and its subtypes (Figure 2.9B). These findings suggested the crucial
role of the CD36 gene in the immune suppression of tumor microenvironments regulated by Treg
cells. At last, I analyzed the relationship between CD36 expression and tumor associated
macrophages polarization by examining the correlation between CD36 and the gene markers
from monocytes, tumor associated macrophages (TAM), M1 macrophages, and M2
32
macrophages. The analysis results demonstrated a positive correlation between CD36 gene with
the gene markers of monocytes and TAM from the same cancer types which gene markers in
exhausted T cells and Treg cells had been verified positively correlated with CD36 (Figure 2.9C).
Moreover, I noticed the correlation between CD36 gene with the gene markers from M2
macrophages was much stronger than those presented on M1 macrophages (Figure 2.9C), which
discovery provided the clue that CD36 expression is involved in immune escape in the tumor
microenvironment for several types of cancers. I also listed the correlation coefficient between
CD36 and all these gene markers in Table 2.4 – 2.7 for each cancer.
33
Figure 2.9 Correlation between CD36 expression and immune markers in different cancers
The heatmap indicates the Spearman’s correlation coefficient (𝜌 ) of CD36 expression with the common immune
markers in T cell exhaustion (A), T reg cells (B), and macrophages polarization (C). The orange color refers to
positive correlation, the blue color refers to negative correlation, and the blank box means no significant
correlation was detected. Also, the more darkness of the color represents the higher possibility of correlation.
34
Table 2.4 Correlation Analysis between CD36 Gene and Related Gene Markers in T Cell
Exhaustion for TCGA Cancer
CD160 CD244 CD28 GZMB HAVCR2 KLRG1 LAG3 PDCD1
Cancer
Type
rho P rho P rho P rho P rho P rho P rho P rho P
ACC (n=79) 0.096 0.420 0.134 0.259 0.400 0.000 0.165 0.164 0.182 0.124 0.281 0.016 -0.100 0.402 0.051 0.671
BLCA (n=408) 0.045 0.385 0.066 0.203 0.158 0.002 0.007 0.896 0.147 0.005 0.130 0.013 0.036 0.488 0.041 0.436
BRCA
(n=1100)
0.166 0.000 0.062 0.049 0.194 0.000 -0.010 0.755 0.096 0.002 0.313 0.000 -0.130 0.000 -0.015 0.631
BRCA-Basal
(n=191)
0.140 0.065 -0.012 0.879 0.245 0.001 -0.110 0.148 0.153 0.043 0.297 0.000 -0.152 0.045 -0.210 0.006
BRCA-Her2
(n=82)
0.020 0.865 -0.014 0.907 0.148 0.215 -0.100 0.404 0.081 0.501 0.034 0.774 0.000 0.997 -0.088 0.461
BRCA-LumA
(n=568)
0.177 0.000 0.199 0.000 0.250 0.000 0.225 0.000 0.204 0.000 0.362 0.000 0.020 0.655 0.145 0.001
BRCA-LumB
(n=219)
0.061 0.402 0.032 0.657 0.163 0.024 -0.003 0.968 -0.068 0.348 0.267 0.000 -0.147 0.042 0.032 0.656
CESC (n=306) -0.084 0.165 0.068 0.257 0.190 0.002 -0.122 0.043 0.136 0.024 0.061 0.312 -0.155 0.010 -0.137 0.022
CHOL (n=36) 0.057 0.745 0.205 0.238 0.224 0.195 0.092 0.599 -0.014 0.938 0.455 0.006 0.041 0.817 0.111 0.525
COAD (n=458) 0.198 0.000 0.418 0.000 0.593 0.000 0.125 0.012 0.644 0.000 0.546 0.000 0.315 0.000 0.268 0.000
DLBC (n=48)
0.258 0.104 0.413 0.007 0.250 0.116 0.352 0.024 0.424 0.006 0.199 0.212 0.195 0.223 0.322 0.040
ESCA (n=185) -0.028 0.713 0.211 0.004 0.103 0.169 0.004 0.959 0.179 0.016 0.145 0.052 0.041 0.585 -0.025 0.735
GBM (n=153) -0.113 0.188 0.230 0.007 0.328 0.000 0.185 0.030 0.092 0.285 -0.001 0.988 0.141 0.101 0.082 0.340
HNSC (n=522) -0.040 0.377 0.024 0.596 0.090 0.047 -0.206 0.000 0.020 0.662 0.033 0.460 -0.195 0.000 -0.150 0.001
HNSC-HPV-
(n=422)
0.058 0.250 0.139 0.005 0.175 0.000 -0.136 0.007 0.073 0.144 0.151 0.002 -0.111 0.026 -0.045 0.367
HNSC-HPV+
(n=98)
-0.326 0.002 -0.365 0.000 -0.032 0.763 -0.423 0.000 -0.118 0.270 -0.245 0.021 -0.488 0.000 -0.444 0.000
KICH (n=66) 0.128 0.309 0.023 0.857 0.194 0.121 0.006 0.963 0.016 0.897 0.024 0.850 -0.096 0.447 -0.091 0.473
KIRC (n=533) 0.278 0.000 0.088 0.058 0.013 0.789 0.200 0.000 0.060 0.200 0.256 0.000 -0.153 0.001 -0.172 0.000
KIRP (n=290) -0.053 0.399 0.288 0.000 0.379 0.000 0.362 0.000 0.172 0.006 0.121 0.052 0.260 0.000 0.080 0.199
LGG (n=516) -0.143 0.002 0.150 0.001 0.364 0.000 0.224 0.000 0.119 0.009 0.059 0.195 0.225 0.000 0.229 0.000
LIHC (n=371) 0.049 0.363 0.156 0.004 0.235 0.000 0.127 0.019 0.054 0.318 0.084 0.118 0.034 0.527 -0.049 0.360
LUAD (n=515) 0.006 0.891 0.229 0.000 0.297 0.000 -0.072 0.109 0.336 0.000 0.282 0.000 -0.070 0.121 -0.064 0.153
LUSC (n=501) 0.002 0.962 0.198 0.000 0.239 0.000 -0.086 0.061 0.289 0.000 0.138 0.002 -0.132 0.004 -0.055 0.229
MESO (n=87) -0.004 0.974 -0.010 0.930 0.335 0.002 -0.089 0.416 -0.024 0.827 0.198 0.070 -0.284 0.008 -0.108 0.326
OV (n=303) 0.094 0.140 0.154 0.015 0.322 0.000 0.081 0.204 0.204 0.001 0.195 0.002 0.020 0.753 0.021 0.744
PAAD (n=179) 0.325 0.000 0.575 0.000 0.741 0.000 0.442 0.000 0.541 0.000 0.705 0.000 0.359 0.000 0.569 0.000
PCPG (n=181) 0.033 0.673 0.185 0.017 0.378 0.000 0.347 0.000 0.178 0.021 0.159 0.040 0.079 0.311 0.066 0.393
PRAD (n=498) 0.101 0.039 0.171 0.000 0.349 0.000 0.041 0.405 0.254 0.000 0.071 0.147 0.003 0.955 -0.019 0.700
READ (n=166) 0.193 0.023 0.308 0.000 0.419 0.000 0.084 0.325 0.502 0.000 0.396 0.000 0.343 0.000 0.265 0.002
SARC (n=260) 0.265 0.000 0.268 0.000 0.472 0.000 0.085 0.185 0.172 0.007 0.305 0.000 -0.190 0.003 -0.049 0.442
SKCM (n=471) 0.163 0.000 0.011 0.820 0.344 0.000 -0.053 0.258 0.181 0.000 0.206 0.000 -0.067 0.150 -0.066 0.157
SKCM-
Metastasis
(n=368)
0.163 0.002 -0.027 0.615 0.339 0.000 -0.051 0.341 0.159 0.003 0.215 0.000 -0.086 0.106 -0.088 0.097
SKCM-Primary
(n=103)
0.097 0.332 0.008 0.938 0.248 0.012 -0.188 0.059 0.042 0.674 0.060 0.547 -0.211 0.033 -0.206 0.038
STAD (n=415) 0.332 0.000 0.360 0.000 0.619 0.000 0.024 0.637 0.426 0.000 0.510 0.000 0.122 0.017 0.173 0.001
TGCT (n=150) -0.130 0.118 0.065 0.437 -0.040 0.633 -0.011 0.896 0.211 0.010 -0.042 0.612 -0.125 0.131 -0.186 0.024
THCA (n=509) 0.184 0.000 -0.157 0.000 -0.199 0.000 -0.072 0.114 -0.222 0.000 0.346 0.000 -0.278 0.000 -0.109 0.016
THYM (n=120) 0.173 0.065 0.004 0.969 -0.029 0.761 -0.246 0.008 0.236 0.011 0.193 0.039 -0.053 0.574 -0.063 0.504
UCEC (n=545) 0.074 0.204 0.091 0.121 0.269 0.000 0.172 0.003 0.225 0.000 0.153 0.009 0.095 0.105 0.216 0.000
UCS (n=57) 0.023 0.871 0.270 0.050 0.423 0.002 -0.039 0.780 0.430 0.001 0.149 0.287 0.001 0.993 0.139 0.321
UVM (n=80) 0.323 0.004 0.110 0.343 0.506 0.000 0.513 0.000 0.410 0.000 0.537 0.000 0.327 0.004 0.324 0.004
35
Table 2.5 Correlation Analysis between CD36 Gene and Related Gene Markers in Treg
cells for TCGA Cancer
CCR8 FOXP3 CD28 TGFB1
Cancer Type rho P rho P rho P rho P
ACC (n=79) 0.106 0.372 -0.157 0.185 0.385 0.001 0.159 0.179
BLCA (n=408) 0.111 0.033 0.157 0.003 0.258 0.000 0.083 0.111
BRCA (n=1100) 0.032 0.314 -0.017 0.594 0.213 0.000 0.156 0.000
BRCA-Basal (n=191) 0.218 0.004 0.101 0.184 0.170 0.025 0.263 0.000
BRCA-Her2 (n=82) 0.027 0.820 0.013 0.915 0.006 0.959 -0.057 0.637
BRCA-LumA (n=568) 0.123 0.005 0.075 0.088 0.173 0.000 0.142 0.001
BRCA-LumB (n=219) -0.041 0.575 -0.013 0.861 0.074 0.307 -0.068 0.350
CESC (n=306) 0.237 0.000 0.173 0.004 0.196 0.001 0.109 0.069
CHOL (n=36) 0.150 0.390 0.147 0.399 0.113 0.517 0.146 0.402
COAD (n=458) 0.440 0.000 0.390 0.000 0.309 0.000 0.461 0.000
DLBC (n=48) 0.069 0.667 -0.123 0.445 0.355 0.023 0.391 0.012
ESCA (n=185) 0.058 0.442 0.077 0.305 0.107 0.151 0.171 0.022
GBM (n=153) 0.095 0.269 0.006 0.947 0.000 0.997 0.108 0.211
HNSC (n=522) 0.185 0.000 0.067 0.139 0.176 0.000 0.130 0.004
HNSC-HPV- (n=422) 0.207 0.000 0.153 0.002 0.170 0.001 0.072 0.151
HNSC-HPV+ (n=98) 0.242 0.022 -0.078 0.465 0.271 0.010 0.148 0.167
KICH (n=66) -0.093 0.460 -0.116 0.356 0.137 0.275 0.201 0.109
KIRC (n=533) -0.048 0.300 -0.225 0.000 0.492 0.000 0.073 0.117
KIRP (n=290) 0.272 0.000 0.227 0.000 -0.066 0.288 0.309 0.000
LGG (n=516) 0.148 0.001 0.130 0.004 0.096 0.036 0.103 0.024
LIHC (n=371) 0.152 0.005 0.192 0.000 0.359 0.000 -0.130 0.015
LUAD (n=515) 0.157 0.000 0.028 0.535 0.277 0.000 0.183 0.000
LUSC (n=501) 0.136 0.003 0.036 0.432 -0.028 0.545 0.163 0.000
MESO (n=87) 0.099 0.368 -0.012 0.913 0.241 0.026 0.429 0.000
OV (n=303) 0.044 0.490 0.148 0.020 0.200 0.002 0.259 0.000
PAAD (n=179) 0.538 0.000 0.561 0.000 0.499 0.000 0.146 0.057
PCPG (n=181) 0.050 0.518 0.040 0.612 0.092 0.237 0.154 0.047
PRAD (n=498) 0.155 0.001 0.164 0.001 0.260 0.000 0.139 0.004
READ (n=166) 0.370 0.000 0.219 0.010 0.208 0.014 0.214 0.011
SARC (n=260) 0.146 0.022 0.069 0.285 0.254 0.000 -0.029 0.647
SKCM (n=471) 0.190 0.000 -0.003 0.946 0.229 0.000 0.171 0.000
SKCM-Metastasis (n=368) 0.143 0.007 -0.033 0.534 0.205 0.000 0.185 0.000
SKCM-Primary (n=103) 0.213 0.032 -0.028 0.782 0.214 0.030 0.028 0.781
STAD (n=415) 0.368 0.000 0.256 0.000 0.490 0.000 0.363 0.000
TGCT (n=150) -0.015 0.855 -0.037 0.659 0.031 0.714 0.344 0.000
THCA (n=509) -0.233 0.000 -0.387 0.000 0.220 0.000 -0.092 0.043
THYM (n=120) 0.135 0.149 0.166 0.077 0.434 0.000 0.241 0.009
UCEC (n=545) 0.187 0.001 0.143 0.015 0.160 0.006 0.186 0.001
UCS (n=57) -0.012 0.930 0.088 0.532 0.290 0.035 0.134 0.338
UVM (n=80) 0.198 0.085 0.193 0.092 0.269 0.018 0.095 0.409
36
Table 2.6 Correlation Analysis between CD36 Gene and Related Gene Markers
Associated with TAM Polarization for TCGA Cancer (A)
Cancer Type
Monocyte TAM
CD86 CSF1R CCL2 CD68 IL10
rho P rho P rho P rho P rho P
ACC (n=79) 0.094 0.430 0.124 0.295 0.064 0.588 0.386 0.001 0.165 0.163
BLCA (n=408) 0.136 0.009 0.207 0.000 0.219 0.000 0.186 0.000 0.181 0.000
BRCA (n=1100) 0.036 0.261 0.120 0.000 0.101 0.001 0.112 0.000 0.139 0.000
BRCA-Basal
(n=191)
0.154 0.042 0.266 0.000 0.113 0.138 0.203 0.007 0.269 0.000
BRCA-Her2 (n=82) 0.011 0.925 0.041 0.731 0.031 0.798 0.018 0.879 0.128 0.284
BRCA-LumA
(n=568)
0.145 0.001 0.166 0.000 0.257 0.000 0.212 0.000 0.216 0.000
BRCA-LumB
(n=219)
-0.110 0.127 -0.020 0.782 -0.069 0.340 0.015 0.838 0.019 0.797
CESC (n=306) 0.172 0.004 0.246 0.000 0.087 0.146 0.179 0.003 0.152 0.011
CHOL (n=36) 0.011 0.949 -0.112 0.522 -0.148 0.395 0.158 0.366 0.219 0.207
COAD (n=458) 0.663 0.000 0.605 0.000 0.618 0.000 0.453 0.000 0.608 0.000
DLBC (n=48) 0.129 0.422 0.391 0.011 0.322 0.040 0.354 0.023 0.420 0.006
ESCA (n=185) 0.240 0.001 0.192 0.010 0.227 0.002 0.153 0.041 0.210 0.005
GBM (n=153) 0.061 0.479 0.055 0.523 0.196 0.022 0.194 0.023 0.289 0.001
HNSC (n=522) 0.046 0.306 0.134 0.003 0.126 0.005 0.138 0.002 0.151 0.001
HNSC-HPV-
(n=422)
0.058 0.250 0.166 0.001 0.115 0.021 0.125 0.012 0.174 0.000
HNSC-HPV+ (n=98) 0.081 0.450 0.078 0.470 0.214 0.044 0.248 0.019 0.078 0.469
KICH (n=66) -0.064 0.613 -0.025 0.846 -0.134 0.288 0.080 0.527 0.080 0.529
KIRC (n=533) -0.001 0.980 0.039 0.403 0.347 0.000 -0.064 0.167 0.060 0.200
KIRP (n=290) 0.298 0.000 0.241 0.000 0.249 0.000 0.208 0.001 0.350 0.000
LGG (n=516) 0.154 0.001 0.085 0.062 0.198 0.000 0.228 0.000 0.189 0.000
LIHC (n=371) 0.071 0.186 0.099 0.067 -0.112 0.037 0.113 0.036 0.169 0.002
LUAD (n=515) 0.340 0.000 0.329 0.000 0.221 0.000 0.444 0.000 0.393 0.000
LUSC (n=501) 0.321 0.000 0.314 0.000 0.197 0.000 0.363 0.000 0.306 0.000
MESO (n=87) -0.078 0.478 0.005 0.967 -0.087 0.426 0.065 0.554 0.042 0.703
OV (n=303) 0.149 0.019 0.140 0.027 0.079 0.215 0.229 0.000 0.217 0.001
PAAD (n=179) 0.605 0.000 0.688 0.000 0.460 0.000 0.291 0.000 0.470 0.000
PCPG (n=181) 0.193 0.012 0.154 0.048 0.242 0.002 0.283 0.000 0.222 0.004
PRAD (n=498) 0.249 0.000 0.223 0.000 0.075 0.127 0.307 0.000 0.260 0.000
READ (n=166) 0.603 0.000 0.463 0.000 0.525 0.000 0.348 0.000 0.320 0.000
SARC (n=260) 0.222 0.000 0.368 0.000 0.221 0.000 0.194 0.002 0.260 0.000
SKCM (n=471) 0.244 0.000 0.308 0.000 0.259 0.000 0.063 0.178 0.224 0.000
SKCM-Metastasis
(n=368)
0.229 0.000 0.306 0.000 0.290 0.000 0.051 0.338 0.176 0.001
SKCM-Primary
(n=103)
0.117 0.242 0.220 0.026 0.146 0.142 -0.050 0.616 0.269 0.006
STAD (n=415) 0.486 0.000 0.614 0.000 0.471 0.000 0.279 0.000 0.473 0.000
TGCT (n=150) 0.152 0.066 0.413 0.000 0.484 0.000 0.224 0.006 0.099 0.232
THCA (n=509) -0.258 0.000 -0.156 0.001 -0.153 0.001 -0.187 0.000 -0.030 0.511
THYM (n=120) 0.316 0.001 0.454 0.000 0.349 0.000 0.224 0.016 0.083 0.379
UCEC (n=545) 0.271 0.000 0.168 0.004 0.171 0.003 0.159 0.007 0.284 0.000
UCS (n=57) 0.374 0.006 0.382 0.005 0.415 0.002 0.377 0.005 0.166 0.233
UVM (n=80) 0.466 0.000 0.272 0.017 0.190 0.098 0.189 0.099 0.430 0.000
37
Table 2.7 Correlation Analysis between CD36 Gene and Related Gene Markers
Associated with TAM Polarization for TCGA Cancer (B)
Cancer Type
M1 Macrophage M2 Macrophage
IRF5 NOS2 PTGS2 CD163 MS4A4A VSIG4
rho P rho P rho P rho P rho P rho P
ACC (n=79) -0.010 0.936 0.220 0.061 0.178 0.131 0.352 0.002 0.241 0.040 0.297 0.011
BLCA (n=408) -0.079 0.133 0.101 0.054 0.071 0.177 0.230 0.000 0.223 0.000 0.215 0.000
BRCA (n=1100) 0.028 0.373 0.172 0.000 0.213 0.000 0.149 0.000 0.252 0.000 0.161 0.000
BRCA-Basal
(n=191)
0.123 0.105 0.189 0.013 0.258 0.001 0.270 0.000 0.313 0.000 0.250 0.001
BRCA-Her2
(n=82)
0.001 0.995 0.236 0.046 0.070 0.557 0.087 0.468 0.050 0.677 0.018 0.881
BRCA-LumA
(n=568)
0.084 0.057 0.258 0.000 0.366 0.000 0.289 0.000 0.349 0.000 0.258 0.000
BRCA-LumB
(n=219)
-0.091 0.209 0.076 0.296 0.145 0.045 0.049 0.496 0.199 0.006 0.058 0.425
CESC (n=306) 0.239 0.000 0.093 0.122 -0.068 0.263 0.235 0.000 0.187 0.002 0.235 0.000
CHOL (n=36) 0.013 0.942 0.206 0.236 0.514 0.002 0.217 0.211 0.191 0.273 0.129 0.461
COAD (n=458) 0.287 0.000 -0.118 0.018 0.319 0.000 0.682 0.000 0.721 0.000 0.669 0.000
DLBC (n=48) 0.055 0.732 0.115 0.474 0.274 0.083 0.423 0.006 0.410 0.008 0.329 0.036
ESCA (n=185) 0.076 0.312 -0.143 0.056 0.135 0.071 0.175 0.019 0.266 0.000 0.227 0.002
GBM (n=153) 0.041 0.634 0.040 0.642 0.205 0.017 0.234 0.006 0.344 0.000 0.219 0.010
HNSC (n=522) 0.028 0.536 -0.044 0.334 0.067 0.139 0.173 0.000 0.128 0.004 0.154 0.001
HNSC-HPV-
(n=422)
0.081 0.104 0.038 0.451 0.015 0.767 0.164 0.001 0.129 0.010 0.153 0.002
HNSC-HPV+
(n=98)
0.027 0.803 -0.150 0.160 0.273 0.010 0.254 0.016 0.186 0.081 0.174 0.103
KICH (n=66) -0.040 0.751 0.147 0.242 0.458 0.000 0.109 0.390 0.141 0.263 0.079 0.533
KIRC (n=533) -0.165 0.000 0.431 0.000 0.160 0.001 0.133 0.004 0.152 0.001 0.007 0.889
KIRP (n=290) -0.121 0.052 0.081 0.197 0.197 0.001 0.445 0.000 0.343 0.000 0.360 0.000
LGG (n=516) 0.087 0.057 -0.032 0.484 0.152 0.001 0.317 0.000 0.340 0.000 0.163 0.000
LIHC (n=371) -0.009 0.873 -0.001 0.990 -0.001 0.987 0.304 0.000 0.194 0.000 0.114 0.035
LUAD (n=515) 0.088 0.050 0.274 0.000 0.005 0.905 0.463 0.000 0.506 0.000 0.460 0.000
LUSC (n=501) 0.028 0.535 0.010 0.820 0.239 0.000 0.395 0.000 0.397 0.000 0.402 0.000
MESO (n=87) -0.164 0.134 0.408 0.000 0.315 0.003 0.229 0.035 0.314 0.003 0.143 0.192
OV (n=303) 0.045 0.480 0.136 0.032 0.180 0.004 0.290 0.000 0.307 0.000 0.326 0.000
PAAD (n=179) 0.212 0.005 0.170 0.026 -0.091 0.236 0.602 0.000 0.625 0.000 0.548 0.000
PCPG (n=181) -0.014 0.861 0.094 0.229 0.216 0.005 0.307 0.000 0.297 0.000 0.249 0.001
PRAD (n=498) 0.206 0.000 0.075 0.129 0.056 0.252 0.299 0.000 0.281 0.000 0.305 0.000
READ (n=166) 0.048 0.573 -0.001 0.994 0.319 0.000 0.580 0.000 0.597 0.000 0.529 0.000
SARC (n=260) 0.254 0.000 0.288 0.000 0.107 0.096 0.327 0.000 0.449 0.000 0.325 0.000
SKCM (n=471) 0.079 0.093 0.148 0.002 0.221 0.000 0.332 0.000 0.308 0.000 0.212 0.000
SKCM-Metastasis
(n=368)
0.070 0.186 0.172 0.001 0.224 0.000 0.338 0.000 0.309 0.000 0.199 0.000
SKCM-Primary
(n=103)
-0.050 0.619 0.048 0.632 0.190 0.056 0.237 0.016 0.232 0.019 0.101 0.314
STAD (n=415) 0.167 0.001 -0.009 0.862 0.182 0.000 0.564 0.000 0.609 0.000 0.528 0.000
TGCT (n=150) 0.324 0.000 0.324 0.000 0.377 0.000 0.574 0.000 0.447 0.000 0.423 0.000
THCA (n=509) -0.358 0.000 0.196 0.000 -0.368 0.000 -0.060 0.189 -0.087 0.054 -0.157 0.001
THYM (n=120) 0.176 0.060 0.235 0.012 0.270 0.004 0.285 0.002 0.402 0.000 0.416 0.000
UCEC (n=545) -0.028 0.637 0.083 0.155 0.079 0.175 0.313 0.000 0.336 0.000 0.290 0.000
UCS (n=57) -0.195 0.161 0.208 0.136 -0.047 0.737 0.426 0.001 0.465 0.000 0.410 0.002
UVM (n=80) 0.398 0.000 0.239 0.037 0.245 0.032 0.543 0.000 0.534 0.000 0.305 0.007
38
2.3.8 CD36 Gene is Frequently Mutated in Various Malignancies
I also examined CD36 mutational status across several cancer types in the TCGA datasets.
Figure 2.10A shows the position and frequency of the mutations overlaid on CD36 protein
domain, in which the most frequent mutation type of CD36 was missense (153 cases), the second
one was truncating mutation (35 cases), I also found a few fusion mutations (4 cases) and
inframe mutations (2 cases). I next screened the CD36 mRNA expression relative to distinct
mutation type, I found that both missense and truncating mutations samples had slightly lower
CD36 expression compared with samples carrying the wild type CD36 (1.4-fold, P < 0.0001;
1.3-fold, P < 0.01), but the CD36 expression was greatly amplified from samples with fusion
mutations (9.0-fold, P < 0.01) (Figure 2.10B). The CD36 mutations observed in different cancer
types from TCGA, PanCancer studies were summarized in Table 2.7.
39
Figure 2.10 Patterns of CD36 mutations in cancer
(A) Diagram of CD36 point mutations in cancer. This diagram presented the sites and frequency of single
nucleotides alterations located on the CD36 chromosome including missense mutation (green spots), truncating
mutations (black spots), mutations (purple spots), and fusion mutations. (B) The raw data of gene expression for
CD36 mutations was downloaded from cBioPortal. The mRNA expression value was shown by RSEM (RNA-
Seq by Expectation-Maximization), and each point presented the expression level from a single sample. One-way
ANOVA (Kruskal-Wallis test) was performed to compare the group difference.
40
Table 2.8 CD36 Mutation Status in Different Cancers
Tissue Cancer Type Mutation Type Copy #
Protein
Change
Nucleotide
Change
Cervix,
Uterus and
Ovary
Cervical Squamous Cell Carcinoma Missense_Mutation diploid D308N c.922G>A
Serous Ovarian Cancer Missense_Mutation shallowdel I83F c.247A>T
Serous Ovarian Cancer Missense_Mutation diploid D270N c.808G>A
Serous Ovarian Cancer Missense_Mutation gain Y276C c.827A>G
Uterine Carcinosarcoma/Uterine
Malignant Mixed Mullerian Tumor
Missense_Mutation diploid R5W c.13C>T
Uterine Carcinosarcoma/Uterine
Malignant Mixed Mullerian Tumor
Fusion amp PVT1-CD36
Uterine Carcinosarcoma/Uterine
Malignant Mixed Mullerian Tumor
Missense_Mutation F379L c.1137C>A
Uterine Carcinosarcoma/Uterine
Malignant Mixed Mullerian Tumor
Missense_Mutation F186Y c.557T>A
Uterine Endometrioid Carcinoma Missense_Mutation diploid R96C c.286C>T
Uterine Endometrioid Carcinoma Missense_Mutation diploid P375H c.1124C>A
Uterine Endometrioid Carcinoma Missense_Mutation diploid S146F c.437C>T
Uterine Endometrioid Carcinoma Missense_Mutation diploid K403N c.1209G>T
Uterine Endometrioid Carcinoma Missense_Mutation diploid T419I c.1256C>T
Uterine Endometrioid Carcinoma Missense_Mutation diploid L414I c.1240C>A
Uterine Endometrioid Carcinoma Missense_Mutation diploid L414I c.1240C>A
Uterine Endometrioid Carcinoma Missense_Mutation diploid K39T c.116A>C
Uterine Endometrioid Carcinoma Missense_Mutation diploid A458D c.1373C>A
Uterine Endometrioid Carcinoma Missense_Mutation diploid D228N c.682G>A
Uterine Endometrioid Carcinoma Missense_Mutation diploid D228N c.682G>A
Uterine Endometrioid Carcinoma Missense_Mutation diploid S167F c.500C>T
Uterine Endometrioid Carcinoma Missense_Mutation diploid S468L c.1403C>T
Uterine Endometrioid Carcinoma Missense_Mutation diploid S468L c.1403C>T
Uterine Endometrioid Carcinoma Missense_Mutation diploid D282N c.844G>A
Uterine Endometrioid Carcinoma Missense_Mutation diploid S167Y c.500C>A
Uterine Endometrioid Carcinoma Frame_Shift_Del diploid N53Ifs*24 c.158del
Uterine Endometrioid Carcinoma Missense_Mutation diploid L138I c.412C>A
Uterine Endometrioid Carcinoma Missense_Mutation diploid T48I c.143C>T
Uterine Endometrioid Carcinoma Nonsense_Mutation diploid Q155* c.463C>T
Uterine Endometrioid Carcinoma Missense_Mutation diploid N428Y c.1282A>T
Uterine Endometrioid Carcinoma Missense_Mutation diploid Q433L c.1298A>T
Uterine Endometrioid Carcinoma Missense_Mutation diploid G442D c.1325G>A
Uterine Endometrioid Carcinoma Splice_Site diploid X144_splice c.430-2A>G
Uterine Endometrioid Carcinoma Missense_Mutation diploid F293C c.878T>G
Uterine Endometrioid Carcinoma Frame_Shift_Ins diploid I399Nfs*19 c.1195dup
Uterine Endometrioid Carcinoma Missense_Mutation diploid A145P c.433G>C
Uterine Endometrioid Carcinoma Missense_Mutation diploid L26R c.77T>G
Uterine Endometrioid Carcinoma Splice_Site diploid X204_splice c.610-1G>T
Uterine Endometrioid Carcinoma Missense_Mutation diploid S81G c.241A>G
Uterine Endometrioid Carcinoma Missense_Mutation diploid L405M c.1213C>A
Uterine Endometrioid Carcinoma Nonsense_Mutation diploid S274* c.821C>A
Uterine Endometrioid Carcinoma Missense_Mutation diploid F97S c.290T>C
Uterine Endometrioid Carcinoma Missense_Mutation gain M156T c.467T>C
Skin
Cutaneous Melanoma Missense_Mutation diploid S468L c.1403C>T
Cutaneous Melanoma Missense_Mutation diploid P255L c.764C>T
Cutaneous Melanoma Missense_Mutation diploid N82S c.245A>G
Cutaneous Melanoma Missense_Mutation diploid G378R c.1132G>A
Cutaneous Melanoma Missense_Mutation diploid V172G c.515T>G
Cutaneous Melanoma Missense_Mutation diploid G248C c.742G>T
Cutaneous Melanoma Nonsense_Mutation diploid S125* c.374C>A
Cutaneous Melanoma Splice_Site diploid X400_splice c.1200-1G>T
Cutaneous Melanoma Missense_Mutation gain D228N c.682G>A
Cutaneous Melanoma Missense_Mutation gain K213E c.637A>G
Cutaneous Melanoma Missense_Mutation gain D372N c.1114G>A
Cutaneous Melanoma Missense_Mutation gain G24D c.71G>A
Cutaneous Melanoma Missense_Mutation gain G378R c.1132G>A
Cutaneous Melanoma Missense_Mutation gain E364K c.1090G>A
Cutaneous Melanoma Missense_Mutation gain P395L c.1184C>T
Cutaneous Melanoma Nonsense_Mutation gain G129* c.385G>T
Cutaneous Melanoma Missense_Mutation D228N c.682G>A
Cutaneous Melanoma Splice_Region R234= c.702G>A
41
Cutaneous Melanoma Missense_Mutation G436E c.1307G>A
Cutaneous Melanoma Missense_Mutation I221R c.662T>G
Cutaneous Melanoma Missense_Mutation V172G c.515T>G
Cutaneous Squamous Cell Carcinoma Splice_Region R234= c.702G>A
Cutaneous Squamous Cell Carcinoma Missense_Mutation E397K c.1189G>A
Melanoma Missense_Mutation diploid P344S c.1030C>T
Melanoma Missense_Mutation diploid G181D c.542G>A
Melanoma Missense_Mutation diploid P296L c.887C>T
Melanoma Missense_Mutation diploid G287V c.860G>T
Melanoma Missense_Mutation diploid P193L c.578C>T
Melanoma Missense_Mutation diploid D352N c.1054G>A
Melanoma Missense_Mutation gain D270N c.808G>A
Melanoma Missense_Mutation gain P191S c.571C>T
Melanoma Missense_Mutation P344S c.1030C>T
Melanoma of Unknown Primary Missense_Mutation G378R c.1132G>A
Skin Cancer, Non-Melanoma Missense_Mutation D270N c.808G>A
Skin Cancer, Non-Melanoma Missense_Mutation P255S c.763C>T
Skin Cancer, Non-Melanoma Missense_Mutation M446I c.1338G>A
Skin Cancer, Non-Melanoma Missense_Mutation E305K c.913G>A
Skin Cancer, Non-Melanoma Splice_Site X419_splice c.1255-1G>A
Skin Cancer, Non-Melanoma Nonsense_Mutation Q400* c.1198C>T
Skin Cancer, Non-Melanoma Missense_Mutation G217R c.649G>A
Skin Cancer, Non-Melanoma Missense_Mutation K385I c.1154A>T
Merkel Cell Carcinoma Missense_Mutation V457A c.1370T>C
Lung
Lung Adenocarcinoma Splice_Site diploid X273_splice c.818+1G>T
Lung Adenocarcinoma Missense_Mutation diploid G199C c.595G>T
Lung Adenocarcinoma Missense_Mutation diploid D270Y c.808G>T
Lung Adenocarcinoma Missense_Mutation diploid T92M c.275C>T
Lung Adenocarcinoma Missense_Mutation diploid D358N c.1072G>A
Lung Adenocarcinoma Missense_Mutation diploid G420L c.1258_1259delinsTT
Lung Adenocarcinoma Missense_Mutation diploid V172F c.514G>T
Lung Adenocarcinoma Missense_Mutation gain A17S c.49G>T
Lung Adenocarcinoma Missense_Mutation gain G89S c.265G>A
Lung Adenocarcinoma Missense_Mutation gain N118H c.352A>C
Lung Squamous Cell Carcinoma Missense_Mutation diploid A11P c.31G>C
Lung Squamous Cell Carcinoma Missense_Mutation diploid R88K c.263G>A
Lung Squamous Cell Carcinoma Missense_Mutation diploid L328R c.983T>G
Lung Squamous Cell Carcinoma Frame_Shift_Del diploid P296Ifs*9 c.885_886del
Lung Squamous Cell Carcinoma Missense_Mutation gain D282A c.845A>C
Lung Squamous Cell Carcinoma Missense_Mutation gain F266L c.798C>A
Lung Squamous Cell Carcinoma Missense_Mutation gain N6S c.17A>G
Lung Squamous Cell Carcinoma Fusion gain SEMA3C-CD36
Lung Squamous Cell Carcinoma Missense_Mutation V451L c.1351G>C
Non-Small Cell Lung Cancer Splice_Site X30_splice c.-89-1G>A
Non-Small Cell Lung Cancer Splice_Site X30_splice c.-89-1G>A
Non-Small Cell Lung Cancer Splice_Site X30_splice c.-89-1G>A
Non-Small Cell Lung Cancer Splice_Site X30_splice c.-89-1G>A
Non-Small Cell Lung Cancer Missense_Mutation N134S c.401A>G
Non-Small Cell Lung Cancer Missense_Mutation N134S c.401A>G
Non-Small Cell Lung Cancer Missense_Mutation N134S c.401A>G
Colon and
Rectum
Colon Adenocarcinoma Missense_Mutation diploid S167Y c.500C>A
Colon Adenocarcinoma Missense_Mutation diploid I341V c.1021A>G
Colon Adenocarcinoma Missense_Mutation diploid Y192C c.575A>G
Colon Adenocarcinoma Missense_Mutation gain I373V c.1117A>G
Colorectal Adenocarcinoma Missense_Mutation R96C c.286C>T
Colorectal Adenocarcinoma Missense_Mutation R96C c.286C>T
Colorectal Adenocarcinoma Missense_Mutation F266L c.798C>A
Colorectal Adenocarcinoma Missense_Mutation D228N c.682G>A
Colorectal Adenocarcinoma Missense_Mutation A427T c.1279G>A
Colorectal Adenocarcinoma In_Frame_Del N206del c.615_617del
Colorectal Adenocarcinoma Frame_Shift_Del D219Ifs*3 c.654del
Rectal Adenocarcinoma Missense_Mutation gain I330V c.988A>G
Mucinous Adenocarcinoma of the Colon
and Rectum
Missense_Mutation diploid L295I c.883C>A
Breast
Breast Invasive Ductal Carcinoma Missense_Mutation diploid Q41L c.122A>T
Breast Invasive Ductal Carcinoma Missense_Mutation diploid R386Q c.1157G>A
Breast Invasive Ductal Carcinoma Missense_Mutation diploid S324T c.970T>A
Breast Invasive Ductal Carcinoma Missense_Mutation diploid D282N c.844G>A
42
Breast Invasive Ductal Carcinoma Missense_Mutation diploid S237C c.710C>G
Breast Invasive Ductal Carcinoma Missense_Mutation gain G336R c.1006G>A
Breast Invasive Ductal Carcinoma Nonsense_Mutation W239* c.717G>A
Invasive Breast Carcinoma Missense_Mutation gain G12R c.34G>A
Benign Phyllodes Tumor of the Breast Missense_Mutation A299V c.896C>T
Stomach
Stomach Adenocarcinoma Missense_Mutation diploid E240G c.719A>G
Stomach Adenocarcinoma Missense_Mutation diploid Q41H c.123A>C
Stomach Adenocarcinoma Missense_Mutation gain A99V c.296C>T
Stomach Adenocarcinoma Missense_Mutation A99V c.296C>T
Diffuse Type Stomach Adenocarcinoma Missense_Mutation diploid V198L c.592G>C
Intestinal Type Stomach
Adenocarcinoma
Missense_Mutation gain K394R c.1181A>G
Tubular Stomach Adenocarcinoma Missense_Mutation gain A99T c.295G>A
Brain
Glioblastoma Missense_Mutation diploid A11T c.31G>A
Glioblastoma Missense_Mutation diploid K223R c.668A>G
Glioblastoma Missense_Mutation diploid V434L c.1300G>T
Glioblastoma Multiforme Missense_Mutation diploid D244N c.730G>A
Glioblastoma Multiforme Missense_Mutation diploid E45K c.133G>A
Glioblastoma Multiforme Missense_Mutation diploid N321K c.963T>G
Glioblastoma Multiforme Missense_Mutation gain N159H c.475A>C
Prostate
Prostate Adenocarcinoma Missense_Mutation diploid S167F c.500C>T
Prostate Adenocarcinoma Missense_Mutation diploid M245I c.735G>A
Prostate Adenocarcinoma Splice_Region diploid X419_splice c.1255-3C>T
Prostate Adenocarcinoma Missense_Mutation gain Q150K c.448C>A
Prostate Adenocarcinoma Missense_Mutation Q150K c.448C>A
Prostate Neuroendocrine Carcinoma Missense_Mutation D219H c.655G>C
Neuron
Astrocytoma Missense_Mutation gain L392S c.1175T>C
Astrocytoma Missense_Mutation gain A11T c.31G>A
Astrocytoma Missense_Mutation gain R183K c.548G>A
Oligoastrocytoma Missense_Mutation diploid G89S c.265G>A
Medulloblastoma Nonsense_Mutation diploid G359* c.1075G>T
Neuroblastoma In_Frame_Ins diploid R368_P375dup c.1102_1125dup
Bladder
Bladder Urothelial Carcinoma Splice_Region shallowdel X336_splice c.1007-3C>T
Bladder Urothelial Carcinoma Missense_Mutation diploid Y340C c.1019A>G
Bladder Urothelial Carcinoma Missense_Mutation diploid S468L c.1403C>T
Bladder Urothelial Carcinoma Fusion gain SEMA3C-CD36
Bladder Urothelial Carcinoma Missense_Mutation V103A c.308T>C
Urachal Adenocarcinoma Missense_Mutation R407K c.1220G>A
Soft Tissue
Ewing Sarcoma Frame_Shift_Del I399Ffs*4 c.1195del
Myxofibrosarcoma Fusion amp CD36-ELMO1
Undifferentiated Pleomorphic
Sarcoma/Malignant Fibrous
Histiocytoma/High-Grade Spindle Cell
Sarcoma
Nonsense_Mutation gain E108* c.322G>T
Liver
Hepatocellular Carcinoma Missense_Mutation diploid L26V c.76C>G
Hepatocellular Carcinoma Missense_Mutation gain H242D c.724C>G
Esophagus
Esophageal Adenocarcinoma Frame_Shift_Del diploid I438* c.1312del
Esophageal Squamous Cell Carcinoma Missense_Mutation gain P395T c.1183C>A
Pancreas
Pancreatic Adenocarcinoma Missense_Mutation A17S c.49G>T
Ampullary Carcinoma Missense_Mutation I438M c.1314A>G
Kidney Papillary Renal Cell Carcinoma Nonsense_Mutation gain K385* c.1153A>T
Adenoid Adenoid Cystic Carcinoma Splice_Site X336_splice c.1006+2T>G
Head and
Neck
Head and Neck Squamous Cell
Carcinoma
Missense_Mutation diploid M455K c.1364T>A
43
Moreover, in the TIMER 2.0 database showing CD36 mutation frequency for each kind of
TCGA cancer, I observed that CD36 was most frequently mutated in UCEC from all different
TCGA cancer types. I next investigated the tumor immune infiltration level in UCEC by
comparing the wild type CD36 to mutated CD36, and I noticed that the tumor immune cells
infiltration level was slightly increased in mutated CD36. Higher immune infiltration level was
detected in dendritic cells from mutated CD36 compared with wild type CD36 (log2FC = 0.306-
fold, P = 0.013; Figure 2.11A), as well as in both CD8+ T cells (log2FC = 0.262-fold, P =
0.0034; Figure 2.11A) and B cells (log2FC = 0.145-fold, P = 0.024; Figure 2.11A). Moderately
elevated immune infiltration levels in CD4+ T cells for ESCA from samples with mutated CD36
gene was observed (log2FC = 0.499-fold, P = 0.049; Figure 2.11B). Although the mutation
frequency for LUAD was unavailable from TIMER2.0 dataset, it was found that the immune
infiltration level of dendritic cells was significantly reduced in LUAD by comparing the mutated
CD36 gene with wild type CD36 (log2FC = -1.102-fold, P = 0.017; Figure 2.11C).
44
Figure 2.11 CD36 mutation associated tumor immune infiltration status
Frequently mutated CD36 correlated tumor infiltration abundances was determined in (A) cancer uterine corpus
endometrial carcinoma (UCEC), (B) esophageal carcinoma (ESCA), and (C) lung adenocarcinoma (LUAD) by
TIMER 2.0. A Wilcoxon test was performed to compare the group difference between wild type and mutations.
45
2.3.9 Clinical Attributes Associated with CD36 Alterations
The CD36 gene is located on the long arm of chromosome 7 at band 11.2 (7q11.2) and is
encoded by 15 exons.
14
I examined the chromosomal status of 7q by comparing the CD36 altered
groups with unaltered groups. The results suggested that 39.02% of the samples (n = 48)
exhibited gain of chromosome and 7.32% of the samples (n = 9) exhibited loss of chromosome
in altered groups (n = 123); however, in unaltered groups (n = 8972), 26.97% of the samples (n =
2420) had 7q gain status and 5.52% of the samples (n = 495) had 7q lost status (P = 0.0048;
Figure 2.12A). The significantly increased gain of chromosomes in altered groups was also
observed in 6p and 9p. For 6p status, in the CD36 gene altered groups (n = 157), 17.2% of the
samples (n = 27) showed chromosomal gain status and 15.92% (n = 25) of the samples showed
chromosomal lost status; while in the unaltered groups, 12.73% of the samples (n = 1140) was
observed having a gain of chromosome and 10.08% of the samples (n = 903) was observed
having a loss of chromosome (P = 0.0062; Figure 2.12B). As for 9p status, 10.71% of the
samples (n = 18) had 9p gain status and 38.1% of the samples (n = 64) had lost status in altered
groups (n = 168), but I only observed 9p gain status in 7.08% of the samples (n = 636 samples)
and 9p lost status in 28.54% of the samples (n = 2563) in unaltered groups (n = 8979) (P =
0.0017; Figure 2.12C). Other loss of chromosomes in 6q, 16p, 17q, 21q, and 19p significantly
associated with CD36 gene alteration were summarized in Table 2.8 and Figure 2.12D-H.
46
Figure 2.12 CD36 alterations associated chromosomal status
The frequency of chromosomal status including gain, loss, and not called related to CD36 alterations was
exhibited in chromosome (A) 7q, (B) 6p, (C) 9q, (D) 6q, (E) 16p, (F) 17q, (G) 21q, and (H) 19p. CD36 alteration
consisted of CD36 mutations and copy number alterations.
Table 2.9 CD36 Alteration Associated Chromosomal Status
Chromosomal Status P-Value
Frequency in Altered Group Frequency in Unaltered Group
Gained Lost Not Called Gained Lost Not Called
7q Status 4.84E-3
39.02%
(48 samples)
7.32%
(9 samples)
53.66%
(66 samples)
26.97%
(2420 samples)
5.52%
(495 samples)
67.51%
(6057 samples)
5q Status 6.24E-3
5.99%
(10 samples)
32.34%
(54 samples)
61.68%
(103 samples)
7.92%
(702 samples)
22.08%
(1957 samples)
70.71%
(6202 samples)
6p Status 7.77E-3
17.2%
(27 samples)
15.92%
(25 samples)
66.88%
(105 samples)
12.73%
(1140 samples)
10.08%
(903 samples)
77.19%
(69.15 samples)
9p Status 1.71E-3
10.71%
(18 samples)
38.1%
(64 samples)
51.19%
(86 samples)
7.08%
(636 samples)
28.54%
(2563 samples)
64.37%
(5780 samples)
16p Status 8.52E-3
9.83%
(17 samples)
16.76%
(29 samples)
73.41%
(127 samples)
13.81%
(1303 samples)
10.09%
(952 samples)
76.1%
(7179 samples)
17q Status 9.89E-3
12%
(18 samples)
18.67%
(28 samples)
69.33%
(104 samples)
12.91%
(1101 samples)
10.85%
(925 samples)
76.25%
(6503 samples)
19p Status 6.94E-4
9.09%
(15 samples)
23.64%
(39 samples)
67.27%
(111 samples)
9.11%
(803 samples)
13.43%
(1184 samples)
77.45%
(6826 samples)
21 (21q) Status 1.50E-4
6.25%
(11 samples)
30.68%
(54 samples)
63.07%
(111 samples)
8.9%
(838 samples)
18.39%
(1731 samples)
72.71%
(6844 samples)
47
2.4 Discussion
Accumulated studies have provided that lipid metabolism reprogramming plays a crucial role in
the survival and invasion of cancer cells.
1
Thereby, the lipogenic characteristics of CD36 has
been increasingly involved in the lipid metabolic plasticity of tumor cells. Moreover, the
energetic sources supplied by fatty acid receptor CD36 contribute tumor cells against the
metabolic stress in TME, thus leading to build immune tolerance and escape immune scrutiny.
65
However, tumor surroundings are consisting of several components such as extracellular matrix,
fibroblasts, adipocytes, and immune cells. Where the CD36 is localized and how its expression is
altered in TME is still under investigation. In this study, I mainly focused on the correlation
between CD36 expression and immune cells infiltrated in the metastatic niche of tumor.
CD36 mRNA expression was detected in various healthy tissues especially in adipose tissue,
breast tissue, left ventricle of heart, thyroid gland, skeletal muscle, lung tissue, and coronary
artery. This expression pattern of tissue distribution is consistent with its function of facilitating
the translocation of fatty acid as a scavenger receptor.
76
The altered lipid metabolism is found in
the tumor progression for many cancer types such as kidney renal clear cell carcinoma, breast,
prostate, ovarian, stomach, and colorectal cancers.
77
Recent study has identified that high fat diet
induces the adaptive metabolism of cancer cells in tumor microenvironment, which leads to the
fuel sources shift to support the proliferation of tumor cells and survival but disturb the immune
niches of tumor microenvironment by impairing the anti-tumor functions of immune cells.
9
There is also one study indicating that upregulated CD36 alternates the lipid metabolism
mechanism for breast cancer cells, contributing to promote the lapatinib resistance of these
48
cancer cells.
59
These findings further emphasize the need to explore the role of the CD36 gene in
cancer initiation and progression.
CD36 mRNA expression level shows significant difference in various kinds of cancers. I
detected highly upregulated CD36 expression in KIRC and GBM, also repressed CD36 mRNA
level in BRCA, COAD, CHOL, LUAD, LUSC, STAD, THCA, HNSC, KIRP, PRAD, and
UCEC. I next scrutinized the methylation status of CD36 promoter DNA in these cancer types.
The results suggested that the decreased expression level of CD36 mRNA in BRCA, LUAD, and
LUSC was consistent with hypermethylated CD36 promoter. Also, the results from a recent
study specialized in CD36 methylation related to tumor progression reported that
hypermethylated CD36 associated with a low mRNA expression was detected in tumor samples
for lung cancer.
78
The treatment with decitabine, an inhibitor of DNA methylation and
chidamide, an HDAC inhibitor reduces the methylation of CD36 and activates the expression of
repressed CD36, resulting in a suppression of tumor growth. Yet, DNA methylation does not
fully explain CD36 expression patterns in cancer; as the case for cancer tissues from COAD,
HNSC, PRAD, KIRC, and LIHC, which suggested the deregulated CD36 expression in those
cancers, but could not be explained by CD36 promoter methylation.
When I further analyzed the differential expression of CD36 mRNA associated with the overall
survival for cancer patients, I found that CD36 expression had a different impact on the overall
survival rate for different types of cancers. The high expressed CD36 mRNA level predicted
worse overall survival in BLCA, COAD, KIRP, STAD, and UVM; while for HNSC, PAAD, and
49
KIRC, the lower expression of CD36 was associated with shorter overall survival. Besides, there
was a discrepancy for the overall survival in COAD, KIRC, KIRP, and STAD corresponding to
CD36 mRNA expression level in cancer patients. In order to better understand the prognostic
value of CD36 gene in cancer, I inspected the association between CD36 expression and the
immune cells infiltration levels across multiple types of cancer. I found that the CD36 gene
expression profile was significantly correlated to the density of tumor infiltrating immune cells
in several cancer types. During the tumor progression, the continuous antigens stimulation and
the reduced support from both helper T cells and cytokines usually cause the CD8+ T cells
eventually to differentiate into exhausted T cells, resulting in the destruction of anti-tumor
immunity.
10
In this study, it was also found that the CD36 gene expression was positively correlated with
gene markers of anergic T cells, exhausted T cells, and senescent T cells.
79
Additionally, a recent
study reveals that upregulated CD36 expression contributes to the survival of tumor cells by
supporting the metabolic adaptations of Treg cells in tumor microenvironment.
65
For this reason, I
examined the association between the gene markers of Treg cells and CD36 expression in cancer
types, which already suggested a strong correlation between CD36 expression and tumor
immune infiltrates. Then, the positive correlation between CD36 expression and gene markers
presented on Treg cells strengthened my speculation about the immunosuppressive role of CD36
expression. In addition, the much stronger correlation between CD36 and the gene markers of
M2 macrophages compared to M1 macrophages, suggested the potential role of CD36 related to
the polarization of tumor associated macrophages. Since the partitioning for TAM from M1
macrophages to M2 macrophages creates a friendly environment for cancer development.
80
50
Therefore, I surmised that the immune escape of tumor cells associated with CD36 expression
gives rise to the worst clinical survival in cancer patients from two aspects. On the one hand,
CD36 expression could be associated with the inhibition of anti-tumor immune cells such as
CD4+ T cells and CD8+ T cells; on the other hand, increased CD36 expression would facilitate
other tumor immune cells' alternative differentiation into supporting cancer cell growth.
However, the signaling pathway involved in the tumor immune reprogramming was not explored
in this study. The pathway analysis could provide more detailed information about the role of
CD36 involved in the tumor immune plasticity in further investigation. So far, the public data
analysis in this study only suggested the potential vulnerability of the CD36 molecule as a
biomarker in tumor immunosuppression.
51
Chapter 3. Functional Characterization of CD36 in Mouse Primary Cells
3.1 Introduction
Several studies have explored the functional and mechanistic roles of CD36 in specific cancer
types, increasingly indicating CD36 as a prognostic marker correlated with the metastatic
progression.
29
Increased palmitic acid or HFD-induced obesity specifically exacerbates the
metastatic process of tumor cells in a CD36-dependent manner; however, antagonist antibody
targeting CD36 protein remarkably inhibits metastasis in immunodeficient or immunocompetent
mice with xeno-engraftment with human oral cancer cells.
81
Recently, one study in our lab
suggested a novel association between fatty acid translocase CD36 upregulation with leukemia
development. It was found that upregulated APOC2 cooperates with CD36 contributing to
leukemia growth by promoting the LYN-ERK signaling mediated metabolic activities of tumor
cells. Consistently, the knockdown of CD36 reduced the tumor progression and promoted the
overall survival of AML mice.
82
Furthermore, in hematopoietic malignancies, the phenotypic diversifications of lipid metabolism
appear to vary greatly among cancer cells compared to normal cells. The alterations of metabolic
activities assist the development of chemotherapeutic resistance, leukemic stem cell stemness,
and cell proliferation, as well as the inhibition of cell apoptosis in cancerous cells.
83
For this
reason, CD36 overexpression mediated metabolic reprogramming has been increasingly
implicated as a critical sign of tumor cells exhibiting expansion and migration in hematopoietic
disorders. Indeed, increased CD36 was identified to support the chronic lymphocytic leukemia
(CLL) cells growth. The mechanism was explained that constitutively phosphorylated STAT3
52
protein bind to CD36 gene promoter and upregulate CD36 gene expression, eventually favoring
the fatty acid intake and metabolism in CLL cells.
84
Meanwhile, the scavenger function of CD36 receptor also shows an association with innate
immunity in healthy tissues, since one of characteristics of scavenger receptor is responsible for
the proper clearance of apoptotic cells generated during normal homeostasis.
51
Also, CD36
cooperating with 𝛼 𝑉 𝛽 5
Integrin will mediate dendritic cells phagocytosis induced malignant or
viral antigens presentation to CD8+ T cells in the context of MHC-I antigen. During the normal
hematopoiesis, the higher expression of CD36 is almost found in the mature subsets such as
megakaryocyte-erythroid progenitors, granulocyte-macrophage progenitors, and common
myeloid progenitors, as compared to the relatively lower expression in the primitive subsets
including hematopoietic stem cells, multipotent progenitors, and lymphoid-primed multipotent
progenitors.
85
It has been suggested that CD36 deficiency is related to the development of worse
malaria, because macrophage CD36 plays an important role in the clearance of parasitized
erythrocytes and prevents the severe parasitic infection.
86
Besides, CD36 mutation frequently
observed in African populations has advantage in recognizing sickle erythrocytes, implicating a
possible role in the pathogenesis of sickle cell disease.
51
Thus, it’s crucial to consider the impact
on normal hematopoietic cells in the therapeutic strategy of immune deficient diseases by
selectively targeting CD36.
In this chapter, the main goal is to characterize both deregulation and upregulation of mouse
Cd36 gene in primary bone marrow and spleen cells cultured in vitro. The efficiency of Cd36
53
knockdown and overexpression was evaluated by qPCR and Western Blot analysis. Also, the
proliferation status of mouse primary cells with the inhibition of Cd36 expression was assessed.
The expansion of mouse primary cells under the suppression of Cd36 indicated that specifically
targeting Cd36 would not be toxic to mouse normal cell growth. The results established the
possibility to build a leukemic mice model with alternated Cd36 expression, which model could
be used to investigate the functional and mechanistic role of CD36 in AML development in the
future.
3.2 Methods and Materials
3.2.1 Short Hairpin RNA-mediated Knockdown of Cd36 Gene
The lentiviral pLKO.1-TRC cloning vector was ordered from Addgene (plasmid #10878).
87
Two
pairs of functional shRNA constructs were designed and ordered from Integrated DNA
Technologies (IDT) (Figure 3.1 C-D). The two types of recombinant plasmids were presented in
Figure 3.1A-B.
54
Figure 3.1 The diagram of shCd36 design
The maps show the two types of plasmid constructions, which vector had Tet-on system for doxycycline-
inducible gene expression, puromycin resistance, and ampicillin resistance. (C-D) Format for shRNA oligo design
consisted of 5’- and 3’- overhang, targeting sequence, loop, reverse complement targeting sequence, and poly (A)
tail sequence, which sense and antisense shRNA oligos sequences targeting mouse Cd36 gene were found from
RNAi Consortium library, and the nucleotide sequences of other elements were provided by Addgene
manufactory protocol of lentiviral pLKO.1-TRC cloning vector.
Next, the recombinant plasmids containing shCd36 fragments were transformed into One Shot™
TOP10 Chemically Competent E. coli cells. Colony PCR with expected product size was used to
screen bacteria for positive single colonies in the LB-agar plate (Figure 3.2). The confirmed
single colony was then incubated and expanded for extracting plasmid DNA using ZymoPURE
Plasmid Miniprep Kit. At last, the purified plasmid DNA was sent to Sanger sequencing in
GENEWIZ. The reverse primer used in the Sanger sequence had a sequence as 5’
GAACGGACGTGAAGAATGTGC 3’.
55
Figure 3.2 The diagram of colony PCR primers design in detecting shCd36
The forward primer has a sequence exactly same as a fragment of conserved Human H1 promoter of pLKO.1-
TRC cloning vector, and the reverse primer has a sequence that is complementary to a fragment of conserved
hPGK promoter of pLKO.1-TRC cloning vector. The primer pairs span the shCd36 oligonucleotides and generate
a colony PCR product with 475bp.
3.2.2 Lentiviral Vectors-mediated Cd36 Overexpression in Recombinant Plasmids
The pLVX-AcGFP1-N1 lentiviral vector containing GFP reporter was used for expression of
Cd36 in mice primary cells. Also, another lentiviral expression vector pCDH-EF1-FHC was
ordered from Addgene (plasmid #64874).
88
The Mouse Cd36 RNA was extracted from a healthy
CL57BL/6J mouse liver tissue using RNA isolation kit (RNeasy mini kit; QIAGEN; Cat. nos.
74104) and the cDNA was synthesized by reverse transcription reaction through Superscript IV
First-Strand cDNA synthesis system (Cat. No. 18091050). The nucleotide sequence of mouse
Cd36 gene was obtained from e!Ensembl (https://uswest.ensembl.org/index.html). Next, Cd36
cDNA fragment was inserted into the EcoR1/BamH1 digestion sites of the two kinds of lentiviral
vectors. The diagram of recombinant plasmids was shown in Figure 3.3.
56
Figure 3.3 The diagram of Cd36 molecular cloning
The maps show Cd36 DNA fragments inserted with (A) pLVX-AcGFP1-N1 lentiviral vector and (B) pCDH-EF1-
FHC lentiviral vector. Both vectors have puromycin resistance and ampicillin resistance, but only the PLVX
vector contains green fluorescent protein (GFP) reporter. (C) The diagram shows the sequence of common
forward primer and unique reverse primer used in the Cd36 cloning.
Next, the ligation products were transformed into E. coli Top10 competent cells. Presence of
Cd36 gene in positive transformants was confirmed by colony PCR with expected product size
(Figure 3.4). Then, purified plasmid recombinants were isolated from positive transformants
using plasmid miniprep kit and sequencing with one vector specific primer by GENEWIZ.
57
Figure 3.4 The diagram of colony PCR Primer design in detecting Cd36 cDNA
The forward primer has a sequence exactly same as a fragment of conserved (A) PLVX-N1-CMV promoter and
(B) EF-1𝛼 promoter, and the reverse primer has a sequence that is complementary to a fragment of Cd36 cDNA.
The colony PCR product has a size of (A) 228bp for using PLVX vector and (B) 434bp for using PCDH vector.
3.2.3 Lentivirus Production
HEK 293T cell lines were cultured using Dulbecco’s Modified Eagle’s (DMEM, Thermo Fisher
Scientific) Medium supplemented with 10% inactivated fetal bovine serum (FBS, Invitrogen)
and 1% antibiotics (AA, Invitrogen) (D10) at 37℃ in a humidified incubator with an atmosphere
of 5% CO2. HEK 293T cells with 60-80% confluence was transfected with the pLKO.1-shCd36-
1 plasmid (shCd36-1), pLKO.1-shCd36-2 plasmid (shCd36-2), and shscramble plasmid as the
control group; as well as transfected with PLVX-mCd36 plasmid, PCDH-mCd36 plasmid, and
PLVX or PCDH empty vector as the control group. The manufactory protocol of CalPhos
Mammalian Transfection kit (Takara, Cat. nos. 631312) was followed in cell transfection. The
culture medium containing secreted virus was harvested twice, once at 48 hours of transfection
58
and second time at 72 hours post transfection. Next, all collected medium was filtered by
0.45𝜇𝑀 sterile filter and concentrated in PEG at 4℃ about 3 days. Then, the viral particles were
concentrated by centrifugation with a speed of 1500rpm for 30min at 4℃ and resuspended with
2mL RPMI 1640 medium (Thermo Fisher Scientific) supplemented with 20% FBS and 1%AA
(R20). The concentrated virus was stored in -80℃ bridge for further usage.
3.2.4 Loss and Gain Function of Cd36 in Mice Primary Cells in Vitro
Some wild-type C57BL/6J inbred mice expressing the CD45.2 allele that were at least 6-8 weeks
old were anesthetized using 1mL isoflurane and dissected in a biosafety cabinet. Other wild-type
C57BL/6J mice were engrafted with 200,000 FLT3-ITD/MLL-PTD mouse leukemia cells to
develop a xenograft leukemic mice model. The spleen was extracted and smashed to get spleen
cells, and the 0.5mL permanently attached syringe was used to flush bone marrow with
phosphate-buffered saline (PBS). All primary cells resuspended in PBS were filtered by 70𝜇 M
cell strainer (FALCON) and centrifuged with 1500 rpm for 5 min at room temperature. The cell
pellets were resuspended with R20 medium ready for infection.
For Cd36 knockdown, shCd36-1, shCd36-2, and shscramble lentivirus were separately used to
infect cells as 1:1 ratio on the 1
st
day. After 24 hours of 1
st
infection, lentivirus was added again,
still following 1:1 ratio in the 2
nd
infection. After 48 hours of 2
nd
infection, puromycin with a
final concentration of 2𝜇𝑔 /𝑚𝐿 was used to select the successfully infected cells, and the second
puromycin selection with the same concentration was performed after another 48 hours of first
selection. At the end of 2
nd
puromycin selection, the cells were treated with doxycycline with a
concentration of 1𝜇𝑔 /𝑚𝐿 to activate the Tet-on induced suppression effect. After 48 hours of
59
doxycycline treatment, half of cells were collected, and the cell pellets were resuspended in
500𝜇𝐿 Triazole and stored in -80℃ fridge in order to extract RNA later. Furthermore, another
half-cell pellet was collected after one more day of 1
st
collection and snapped frozen by dry ice
to run western blot.
For Cd36 overexpression, the cells were treated by the same way as knockdown until the second
puromycin selection. After 48 hours of 2
nd
puromycin selection, the fluorescence microscopy
was used to document fluorescent signals from cells with successfully transfected PLVX-mCd36
plasmid. The images were analyzed using NIH Image J software. Half of cells were also
collected and frozen by 500𝜇𝐿 Trizol in -80℃ fridge for qPCR quantification, then the remaining
cells were collected the next day and snap frozen by dry ice for western blot quantification.
3.2.4 qPCR Assessment of Cd36 Knockdown and Overexpression
Extraction of all RNA samples were performed based on the Trizol and chloroform extraction
protocol. Equal amount of RNA was used for cDNA synthesis using Superscript IV reverse
transcriptase (Invitrogen) following the manufacturer's protocol (cat. nos. 18091050). The qPCR
primer pair of CD36 was designed according to the nucleotide sequences found in NCBI
database (NC_000071.6), which forward primer has a sequence of 5’-
GCGACATGATTAATGGCACAG - 3’, and reverse primer has a sequence of 5’-
GATCCGAACACAGCGTAGATAG -3’. The primer sequences of reference gene Gapdh were
obtained from one study studying the gene expression in mouse heart,
89
which forward primer
has a sequence of 5’- CTCCCACTCTTCCACCTTCG -3’, and the reverse primer has a sequence
of 5’- GCCTCTCTTGCTCAGTGTCC -3’. The real-time quantitative PCR (qPCR) assay with
60
10𝜇𝐿 reaction solutions were performed using the ABI QuantStudio 12K Flex Real-Time PCR
system (Applied Biosystem). To check reproducibility, each assay was performed with technical
triplicates for each of the samples in the detection of Cd36 and Gapdh gene. The relative
transcripts levels of Cd36 were normalized to Gapdh and calculated using the 2
−∆∆𝐶𝑇
method.
3.2.5 Western Blot Assessment of Cd36 Knockdown and Overexpression
The collected cell pellets were lysed in the mixture of lysis buffer (Thermo Scientific, Cat. nos.
OG284643) and protease inhibitor. BCA assay (Pierce) was used to quantify the protein
concentration. Next, equivalent amounts of protein per sample was run on SDS polyacrylamide
tris-glycine gels (Bio-Rad, Cat. nos. 456-1094). The proteins were then transferred onto PVDF
membrane through wet tank blotting system with 20 voltages at 4℃ overnight. After protein
transferase, the membrane was blocked in 5% BSA/TBST for 1 hour at 4℃. Soon after, the
PVDF membrane was incubated with 10mL CD36 polyclonal antibody (Thermo Fisher
Scientific, 18836-1-AP) that was 1:1000 diluted by 5% BSA at 4℃ overnight. The next day, the
membrane was incubated with 10mL Horseradish peroxidase (HRP)-conjugated anti-Rabbit
secondary antibody that was 1:5000 diluted by 5% BSA at 4℃ for one hour. After that, luminal
chemiluminescence was used with a Bio-Rad Chemi-Doc imaging system to image blots of the
membrane. Once the bands were visualized, the membranes were then incubated with 1:5000
diluted conjugated GAPDH monoclonal antibody (Thermo Fisher Scientific, 1E6D9). At last, the
blot was re-imaged by Bio-Rad Chemi-Doc.
61
3.2.6 Proliferation Assay
During the Cd36 knockdown assay, a small percentage of cells in each group were converted to a
new suspension plate on the same day of doxycycline treatment to set up the proliferation assay.
This day was counted as day 0, and then cells were continued to count for 5 other days with
Trypan blue assay.
3.2.7 Statistical Analysis
In the proliferation assay, four replicates for each of the samples were counted every day. The
average of cell counts for each sample was calculated, then the cell numbers in the following 4
days were normalized to the 1
st
day separately. The relative change of cells growth compared to
1
st
day in doxycycline treated groups was presented in a histogram plot by GraphPad Prism 7.0,
in which error bars exhibited the standard deviation between the four replicates. One-way
ANOVA was performed to calculate the significant difference by comparing the treated and
control group in proliferation assay. The significance was considered for a P value less than 0.5.
3.3 Results
3.3.1 Assessment of Cd36 Knockdown in Mouse Primary Hematopoietic Cells
The efficiency of the Cd36 knockdown was then tested in the primary bone marrow and spleen
cells from a healthy mouse and in mouse splenic leukemia cells (MLL-PTD/FLT3-ITD cells).
Figure 3.5A exhibited the decreased Cd36 mRNA expression in bone marrow cells compared to
the control for the two pairs of shRNA, but only the second pair of shRNA exhibited suppression
effect on Cd36 expression in spleen cells (Figure 3.5B). Next, the effect of Cd36 knockdown
using shRNA lentiviral plasmids on the spleen primary cells of leukemic mice was assessed by
62
qPCR or Western blot. The first pair of Cd36-shRNA-infected spleen cells exhibited significant
reduction in the mRNA expression relative to the scramble sRNA infection, but the knockdown
effect was not observed in the second pair shRNA by qPCR (Figure 3.5C). However, the
knockdown effect using both two shRNA was confirmed by Western blot analysis (Figure 3.5D).
Figure 3.5 The effect of Cd36 knockdown in mouse primary cells
qPCR quantification of Cd36 knockdown efficiency in (A) healthy mouse bone marrow cells, (B) healthy mouse
spleen cells, and (C) mouse leukemic spleen cells. (D) Western blot quantification of Cd36 knockdown efficiency
in mouse leukemic spleen cells.
63
3.3.2 Assessment of Cd36 Overexpression in Healthy Mouse Hematopoietic Cells
The overexpression of Cd36 was measured twice in both bone marrow and spleen cells from a
healthy mouse. The remarkably elevated Cd36 mRNA expression level was identified in both
bone marrow and spleen cells infected by the two types of lentiviruses through qPCR assessment
(Figure 3.6A-B). Also, the amplificated Cd36 expression in spleen cells was consistent with the
Western blot analysis; meanwhile, the Western blot results only presented the overexpressed
Cd36 protein transduced with mCd36-PCDH recombinant plasmid in bone marrow cells (Figure
3.6C). I also confirmed the ectopic expression by immunofluorescence microscopy was to detect
the green fluorescence reported by the GFP in the PLVX vector in both bone marrow and spleen
cells. Intensive fluorescence was observed in the overexpressed Cd36 protein, and the controls
transduced with empty PLVX vector plasmids (Figure 3.6D).
64
Figure 3.6 The effect of Cd36 overexpression in healthy mouse primary cells
Measurement of Cd36 overexpression in healthy mouse (A) bone marrow and (B) spleen cells by qPCR. (C) The
expression levels of Cd36 in bone marrow and spleen cells were detected by western blot. (D) The green
fluorescence reported by PLVX lentiviral vector was observed under immunofluorescence microscopy
3.3.3 Cd36 Knockdown Has Limited Effects on the Proliferation of Normal Mouse
Hematopoietic Cells
Proliferation assay was performed on normal mouse primary hematopoietic cells infected by
doxycycline induced shCd36 and shscramble plasmids. Spleen cells infected by shCd36-1
plasmids presented a steady growth which was also observed in the control group of spleen cells
infected by shscramble plasmids (Figure 3.8A). In addition, spleen cells infected by the second
pair of shCd36 lentivirus exhibited cell growth pattern in the first four days but showed a slight
but not significantly different decrease in cell number in the 5
th
day compared with the control
cells. While further experiments are needed, my data suggest that the knockdown of CD36 has
65
limited effect on the proliferation of normal mice spleen cells. Differently, the growth situations
of bone marrow cells were not stable as spleen cells. It was surprising that the control group in
bone marrow cells appeared to stop proliferating from the 4
th
day, while the two experimental
groups showed the slightly increased tendency (Figure 3.8B). One reason causing the non-ideal
proliferation of bone marrow cells could be the very limited starting numbers of bone marrow
cells compared to the spleen cells. Meanwhile, the greatly reduced cell numbers were not
observed in bone marrow cells at the end of proliferation assay, instead the bone marrow cell
numbers were still increased compared to the 1
st
day in the experimental group. These results
suggested that mouse healthy cells would be able to grow even in the suppression of Cd36 gene.
66
Figure 3.7 Proliferation status of healthy mouse primary cells with reduced Cd36 expression
A continuous 5-day proliferation assay was performed on (A) healthy mouse primary spleen cells and (B) bone
marrow cells. The cell numbers among doxycycline treated groups were normalized to the first day and then were
compared.
3.4 Discussion
The cancer metabolic reprogramming facilitates leukemic stem cells (LSCs) to maintain AML
progression and leukemic cell survival causing relapse after treatment. The lipogenic property of
CD36 initiates cellular uptake of long chain FAs, and then shuffles FAs into endogenous
mitochondrial oxidative activities. So, AML cells are continuously supplied with fuels from FA
oxidation without requirement of de novo FA synthesis, ultimately bypassing the cytotoxic
agents and cell exhaustion.
90
Previously, targeting APOC2-CD36 metabolic axis was shown to
67
exhibits the antileukemic effects via the downregulation of the LYN-ERK signaling pathway.
82
Importantly, CD36 is also extensively distributed in different types of healthy tissues including
but not limited to the immune and other hematopoietic cells. Therefore, assessing the impact of
targeting CD36 on the normal tissue counterparts is crucial for developing a proper therapeutic.
Based on the proliferation assay performed in Cd36 knockdown mouse primary cells, I observed
limited effect on cell viability or proliferation. Whether other cell phenotypes are affected by this
genetic manipulation of the cells, remains to be explored. Yet, this is encouraging data
suggesting that Cd36 gene could be selectively targeted in the leukemic mouse model without
impairing the normal cells growth. Moreover, the loss function of Cd36 was characterized in
both bone marrow and spleen cells, where T cells were enriched in mice spleen cells.
Meanwhile, the proliferation assay would not tell me more information about the effects of Cd36
suppression on T cells. Thus, more specific assays will be required to monitor T cell responses
under the inhibition of Cd36.
However, there are still a couple deficiencies regarding this study. Usually, several cytokines and
hormones are involved in the signaling transduction that regulates the fates of hematopoietic
stem cells such as quiescence, self-renewal, differentiation, apoptosis, and mobilization in their
niches.
91
Characterizing the loss of function in Cd36 was performed in mouse primary cells in
this study, so, the survival and proliferative ability of primary hematopoietic cells in the absence
of cytokines or growth factors will be reduced in vitro than in vivo. Additionally, the infected
leukemic cells with confirmed knockdown of Cd36 could be injected into mice in the further
experiment, since the engraftment experiments using a mouse model could provide more
convincing results such as the symptoms of disease progression and the visible fluorescent
68
markers on antibodies by flow cytometry, all contributing to suggest the loss function of Cd36.
Moreover, trypan blue assay is the only way used to determine the ability of cell proliferation in
this study. The results would be more persuasive if I perform other cell viability assay such as
measuring ATP production, live-cell protease activity, tetrazolium or resazurin reduction over
time, as well as the analysis of apoptosis by flow cytometry.
Another deficiency for this study was reflected by the insufficient detection of overexpression
efficiency by both qPCR and western blot assessment, which could be explained by the limited
fresh cell numbers obtained from a younger mouse during cell infection assay. Also, it was hard
to maintain the fragile primary cells in about 10 days of cell culture, as a matter of course, only a
few primary cells were left to be assessed at the end of the experiment. Consequently, the high
background signals would interfere with the analysis results by qPCR and western blot. In the
future, an adequately starting number of primary cells are necessary to get expected efficiency of
overexpression, and more precise gene expression quantification could be used, for example,
flow cytometry quantification of the GFP-positive fraction of successfully infected mouse cells.
Lastly, the lentiviral infection assay could be less effective in primary cells compared to cell
lines since primary cells would not proliferate as fast as cell lines.
69
Chapter 4: Concluding Remarks
Tumorigenesis is dependent upon the oncogenic alterations during cellular metabolic network
reprogramming. Tumor cells frequently over activate PI3K-AKT-mTOR network in order to
stimulate the signaling pathway that controls cell proliferation, resulting in the increased
biosynthesis of nucleotides, proteins, and lipids. The loss of function in tumor suppressor p53
and the gain of function in MYC further promotes the anabolic growth including elevated
glycolysis and glutaminolysis, promoted lactate production, as well as vibrant serine metabolism
and mitochondrial metabolism.
92
The fundamentals of the reprogrammed cancer metabolism are
aimed to acquire more nutrients in order to satisfy the bioenergetic, biosynthetic, and redox
demands for tumor cells.
92
Thereby, lipid metabolism alteration is most noteworthy among
metabolic plasticity of tumor cells. Fatty acids as the primary energy fuels are highly demanded
for tumor cells, which synthesis requires sources of acetyl-CoA and oxidation of cytosolic
NADPH. For this reason, the lipid metabolism of tumor cells also heavily relies on mitochondrial
tricarboxylic acid (TCA) cycle and electron transport chain (ETC) activity for macromolecule
synthesis and ATP production.
92
There was one study suggesting that enhanced CD36 expression in tumor microenvironment
support the intratumoral Treg cells survival under lactic acid-enriched surroundings via
modulating mitochondrial fitness and oxidative phosphorylation. Since the elevated ETC activity
in active mitochondrial metabolism promotes NAD to NADH ratio that in turn supports the
conversion of lactic acid into pyruvate, which is beneficial to Treg cells in response to TME
imposed metabolic stress. Consequently, metabolic adaptation helps Treg cells to counteract
immunosuppressive features of TME.
65
These findings arouse my interest about the role of CD36
70
in the tumor immune environment. Next, I found that CD36 appeared to have a significant
correlation with the tumor immune cells infiltration levels in multiple cancer types, and which
interaction implicated a suppressive effect in tumor immunosuppression. Since CD36 expression
was highly associated with the gene markers associated with T cell exhaustion, Treg cells, and
M2-like macrophages differentiation. Although, there is one study that suggests that CD36
upregulation is involved in leukemia spreading by providing metabolic energy to cancer cells,
thus supporting their differentiation and proliferation. This study explains that CD36 mediated
fatty acid oxidation help leukemic stem cells niched in gonadal adipose tissue escape the immune
scrutiny, inducing chemoresistance in obese leukemia patients.
56
Surprisingly, the information
about CD36 related tumor immune cells infiltration for AML was unavailable in pan cancer data
analysis. To that end, whether and how CD36 upregulation correlates with the antitumor immune
response in AML immune microenvironment are still waiting to be determined.
In the pan cancer data analysis, the upregulated CD36 was found in cancer patient samples for
KIRC and GBM. Indeed, elevated CD36 expression was indicated to facilitate the lipid uptake of
GBM tumorigenic cancer stem cells (CSCs) and maintain their increased proliferation. Also,
CD36 siRNA knockdown decreases the self-renewal of CSCs and attenuates the tumor
progression.
26
Several other studies also have reported the increased CD36 expression in breast
cancer, prostate cancer, ovarian cancer, oral cancer, liver cancer, and acute myeloid leukemia.
29
Therefore, CD36 is increasingly recognized as a potential biomarker therapeutic target for cancer
treatment.
71
As above mentioned, small interfering RNA is one of the common methods in knockdown of
CD36, whose knockdown efficiency is also observed in GBM tumorigenic cancer stem cells
26
and PDAC cell lines
58
. In my bench work analysis, short hairpin RNA (shRNA) is an approach
to suppress mice Cd36 expression. Besides, the anti-CD36 monoclonal antibody such as clone
CRF D-2712 is also used to inhibit targeting CD36 expression in mice. Additionally, there are
other clinical trials targeting CD36 for cancer treatment. Based on the property of CD36 binding
to TSP-1 and then initiating the apoptosis of tumor vascular endothelial cells in tumor tissues,
there are some kinds of modified TSR peptide mimetics targeting CD36 exhibiting proapoptotic
effects in tumor cells such as ABT-510, ABT-526, and ABT-898. Pfizer also developed TSP-1
peptidomimetics including CVX-022 and CVX-045 presenting attenuated tumor progression by
targeting CD36 in solid tumor tissues.
25
Besides, a cyclic peptide called TAX2 derived from
CD47 that usually antagonizes CD36 activities, appears an activation on TSP-1/CD36 interaction
in tumor anti-angiogenesis therapy.
25
However, the CD36 receptor widely distributed on all
kinds of tissues in human body, it’s still required to consider whether and how to precisely target
the CD36 protein in specific human tumor tissues.
Furthermore, it’s worth knowing that CD36 is not the only player in the complicated tumor
metabolic network. Except previously reported, APOC2 cooperates with CD36 activates LYN-
ERK signaling pathway, resulting in the enhanced proliferation of leukemic cells.
82
CD36 also
can recruit 𝛼𝜈𝛽 5 or 𝛼𝜈𝛽 3 integrins during innate immunity phagocytosis or endogenous lipid
synthesis. Similarly, the interaction between CD36 and toll-like receptors 2/6 or 4/6 plays a key
role for innate immune response to foreign pathogens invasions. Besides, by coupling to the γ
subunit of the immunoglobulin Fc receptor (FcRγ), CD36 can activate the downstream Src-
72
family kinases, which following drives the internalization of CD36 and its bound ligands,
93
as
well as leads to promoted anti-angiogenesis and fatty acid oxidation accompanied with tumor
metastasis.
25
In addition to FcRγ, thrombospondin-1 binding to CD36 on microvascular
endothelial cells is followed by sequential phosphorylation and activation of p59
fyn
, which
continues to activate mitogen-activated protein kinase (MAPK) and increased expression of
caspase-3-like protease in cell apoptotic signaling pathway.
94
In hepatocellular carcinoma, one
study reports that elevated free fatty acid uptake by FA translocase CD36 will activate Wnt/TGF-
β signaling pathway, thus facilitating tumor metastasis via increased induction of epithelial-
mesenchymal transition (EMT).
43
From another study, CD36 mediated increased lipid uptake
help Treg cells adapt to the stress of acidic TME via activated CD36–PPAR-β axis signaling.
65
Also, elevated FA uptake by CD36 receptor has been shown to increase the phosphorylation of
AKT, which then inhibits glycogen synthase kinase 3β (GSK-3)/β-catenin degradation, thus
promoting gastric cancer metastasis.
43
In this case, the intersection role of CD36 in cancer
metabolic network further improves the difficulty in targeting CD36, since it’s highly possible to
disrupt the intersecting signaling pathway, then giving rising to severe side effects. Therefore,
it’s necessary to do more investigations specialized on the precise targeting CD36 in the future
clinical work.
73
Bibliography
1. Ohshima K, Morii E. Metabolic Reprogramming of Cancer Cells during Tumor
Progression and Metastasis. Metabolites. Jan 2 2021;11(1)doi:10.3390/metabo11010028
2. Vander Heiden MG, Cantley LC, Thompson CB. Understanding the Warburg effect: the
metabolic requirements of cell proliferation. Science. May 22 2009;324(5930):1029-33.
doi:10.1126/science.1160809
3. Liberti MV, Locasale JW. The Warburg Effect: How Does it Benefit Cancer Cells?
Trends Biochem Sci. Mar 2016;41(3):211-218. doi:10.1016/j.tibs.2015.12.001
4. Colegio OR, Chu NQ, Szabo AL, et al. Functional polarization of tumour-associated
macrophages by tumour-derived lactic acid. Nature. Sep 25 2014;513(7519):559-63.
doi:10.1038/nature13490
5. Rohani N, Hao L, Alexis MS, et al. Acidification of Tumor at Stromal Boundaries Drives
Transcriptome Alterations Associated with Aggressive Phenotypes. Cancer Res. Apr 15
2019;79(8):1952-1966. doi:10.1158/0008-5472.CAN-18-1604
6. Sutoo S, Maeda T, Suzuki A, Kato Y. Adaptation to chronic acidic extracellular pH
elicits a sustained increase in lung cancer cell invasion and metastasis. Clin Exp Metastasis. Feb
2020;37(1):133-144. doi:10.1007/s10585-019-09990-1
7. Farnsworth RH, Lackmann M, Achen MG, Stacker SA. Vascular remodeling in cancer.
Oncogene. Jul 3 2014;33(27):3496-505. doi:10.1038/onc.2013.304
8. DePeaux K, Delgoffe GM. Metabolic barriers to cancer immunotherapy. Nat Rev
Immunol. Apr 29 2021;doi:10.1038/s41577-021-00541-y
9. Ringel AE, Drijvers JM, Baker GJ, et al. Obesity Shapes Metabolism in the Tumor
Microenvironment to Suppress Anti-Tumor Immunity. Cell. Dec 23 2020;183(7):1848-1866 e26.
doi:10.1016/j.cell.2020.11.009
10. Franco F, Jaccard A, Romero P, Yu YR, Ho PC. Metabolic and epigenetic regulation of
T-cell exhaustion. Nat Metab. Oct 2020;2(10):1001-1012. doi:10.1038/s42255-020-00280-9
11. Asch AS, Barnwell J, Silverstein RL, Nachman RL. Isolation of the thrombospondin
membrane receptor. J Clin Invest. Apr 1987;79(4):1054-61. doi:10.1172/JCI112918
74
12. Nicholson AC, Febbraio M, Han J, Silverstein RL, Hajjar DP. CD36 in atherosclerosis.
The role of a class B macrophage scavenger receptor. Ann N Y Acad Sci. May 2000;902:128-31;
discussion 131-3.
13. Armesilla AL, Vega MA. Structural organization of the gene for human CD36
glycoprotein. J Biol Chem. Jul 22 1994;269(29):18985-91.
14. Rac ME, Safranow K, Poncyljusz W. Molecular basis of human CD36 gene mutations.
Mol Med. May-Jun 2007;13(5-6):288-96. doi:10.2119/2006-00088.Raae
15. Pepino MY, Kuda O, Samovski D, Abumrad NA. Structure-function of CD36 and
importance of fatty acid signal transduction in fat metabolism. Annu Rev Nutr. 2014;34:281-303.
doi:10.1146/annurev-nutr-071812-161220
16. Doi T, Higashino K, Kurihara Y, et al. Charged collagen structure mediates the
recognition of negatively charged macromolecules by macrophage scavenger receptors. J Biol
Chem. Jan 25 1993;268(3):2126-33.
17. Thorne RF, Law EG, Elith CA, Ralston KJ, Bates RC, Burns GF. The association
between CD36 and Lyn protein tyrosine kinase is mediated by lipid. Biochem Biophys Res
Commun. Dec 8 2006;351(1):51-6. doi:10.1016/j.bbrc.2006.09.156
18. Pohl J, Ring A, Korkmaz U, Ehehalt R, Stremmel W. FAT/CD36-mediated long-chain
fatty acid uptake in adipocytes requires plasma membrane rafts. Mol Biol Cell. Jan
2005;16(1):24-31. doi:10.1091/mbc.e04-07-0616
19. Koonen DP, Glatz JF, Bonen A, Luiken JJ. Long-chain fatty acid uptake and FAT/CD36
translocation in heart and skeletal muscle. Biochim Biophys Acta. Oct 1 2005;1736(3):163-80.
doi:10.1016/j.bbalip.2005.08.018
20. Trezzini C, Jungi TW, Spycher MO, Maly FE, Rao P. Human monocytes CD36 and
CD16 are signaling molecules. Evidence from studies using antibody-induced
chemiluminescence as a tool to probe signal transduction. Immunology. Sep 1990;71(1):29-37.
21. Bou Khzam L, Son NH, Mullick AE, Abumrad NA, Goldberg IJ. Endothelial cell CD36
deficiency prevents normal angiogenesis and vascular repair. Am J Transl Res.
2020;12(12):7737-7761.
75
22. Chen M, Yang Y, Braunstein E, Georgeson KE, Harmon CM. Gut expression and
regulation of FAT/CD36: possible role in fatty acid transport in rat enterocytes. Am J Physiol
Endocrinol Metab. Nov 2001;281(5):E916-23. doi:10.1152/ajpendo.2001.281.5.E916
23. Zhou J, Febbraio M, Wada T, et al. Hepatic fatty acid transporter Cd36 is a common
target of LXR, PXR, and PPARgamma in promoting steatosis. Gastroenterology. Feb
2008;134(2):556-67. doi:10.1053/j.gastro.2007.11.037
24. Osz K, Ross M, Petrik J. The thrombospondin-1 receptor CD36 is an important mediator
of ovarian angiogenesis and folliculogenesis. Reprod Biol Endocrinol. Mar 14 2014;12:21.
doi:10.1186/1477-7827-12-21
25. Wang J, Li Y. CD36 tango in cancer: signaling pathways and functions. Theranostics.
2019;9(17):4893-4908. doi:10.7150/thno.36037
26. Hale JS, Otvos B, Sinyuk M, et al. Cancer stem cell-specific scavenger receptor CD36
drives glioblastoma progression. Stem Cells. Jul 2014;32(7):1746-58. doi:10.1002/stem.1716
27. Ladanyi A, Mukherjee A, Kenny HA, et al. Adipocyte-induced CD36 expression drives
ovarian cancer progression and metastasis. Oncogene. Apr 2018;37(17):2285-2301.
doi:10.1038/s41388-017-0093-z
28. Yoshida T, Yokobori T, Saito H, et al. CD36 Expression Is Associated with Cancer
Aggressiveness and Energy Source in Esophageal Squamous Cell Carcinoma. Ann Surg Oncol.
Feb 2021;28(2):1217-1227. doi:10.1245/s10434-020-08711-3
29. Enciu AM, Radu E, Popescu ID, Hinescu ME, Ceafalan LC. Targeting CD36 as
Biomarker for Metastasis Prognostic: How Far from Translation into Clinical Practice? Biomed
Res Int. 2018;2018:7801202. doi:10.1155/2018/7801202
30. Clezardin P, Frappart L, Clerget M, Pechoux C, Delmas PD. Expression of
thrombospondin (TSP1) and its receptors (CD36 and CD51) in normal, hyperplastic, and
neoplastic human breast. Cancer Res. Mar 15 1993;53(6):1421-30.
31. Uray IP, Liang Y, Hyder SM. Estradiol down-regulates CD36 expression in human breast
cancer cells. Cancer Lett. Apr 15 2004;207(1):101-7. doi:10.1016/j.canlet.2003.10.021
32. Zhao J, Zhi Z, Wang C, et al. Exogenous lipids promote the growth of breast cancer cells
via CD36. Oncol Rep. Oct 2017;38(4):2105-2115. doi:10.3892/or.2017.5864
76
33. Silverstein RL, Febbraio M. CD36, a scavenger receptor involved in immunity,
metabolism, angiogenesis, and behavior. Sci Signal. May 26 2009;2(72):re3.
doi:10.1126/scisignal.272re3
34. Silverstein RL, Febbraio M. CD36-TSP-HRGP interactions in the regulation of
angiogenesis. Curr Pharm Des. 2007;13(35):3559-67. doi:10.2174/138161207782794185
35. Sleeman JP. The metastatic niche and stromal progression. Cancer Metastasis Rev. Dec
2012;31(3-4):429-40. doi:10.1007/s10555-012-9373-9
36. Choi SH, Tamura K, Khajuria RK, et al. Antiangiogenic variant of TSP-1 targets tumor
cells in glioblastomas. Mol Ther. Feb 2015;23(2):235-43. doi:10.1038/mt.2014.214
37. Chen PC, Tang CH, Lin LW, et al. Thrombospondin-2 promotes prostate cancer bone
metastasis by the up-regulation of matrix metalloproteinase-2 through down-regulating miR-
376c expression. J Hematol Oncol. Jan 25 2017;10(1):33. doi:10.1186/s13045-017-0390-6
38. Fang Y, Shen ZY, Zhan YZ, et al. CD36 inhibits beta-catenin/c-myc-mediated glycolysis
through ubiquitination of GPC4 to repress colorectal tumorigenesis. Nat Commun. Sep 4
2019;10(1):3981. doi:10.1038/s41467-019-11662-3
39. Pan J, Fan Z, Wang Z, et al. CD36 mediates palmitate acid-induced metastasis of gastric
cancer via AKT/GSK-3beta/beta-catenin pathway. J Exp Clin Cancer Res. Feb 4 2019;38(1):52.
doi:10.1186/s13046-019-1049-7
40. Sakurai K, Tomihara K, Yamazaki M, et al. CD36 expression on oral squamous cell
carcinoma cells correlates with enhanced proliferation and migratory activity. Oral Dis. May
2020;26(4):745-755. doi:10.1111/odi.13210
41. Chen M, Pych E, Corpron C, Harmon CM. Regulation of CD36 expression in human
melanoma cells. Adv Exp Med Biol. 2002;507:337-42. doi:10.1007/978-1-4615-0193-0_52
42. Nath A, Chan C. Genetic alterations in fatty acid transport and metabolism genes are
associated with metastatic progression and poor prognosis of human cancers. Sci Rep. Jan 4
2016;6:18669. doi:10.1038/srep18669
43. Nath A, Li I, Roberts LR, Chan C. Elevated free fatty acid uptake via CD36 promotes
epithelial-mesenchymal transition in hepatocellular carcinoma. Sci Rep. Oct 1 2015;5:14752.
doi:10.1038/srep14752
77
44. Mittal V. Epithelial Mesenchymal Transition in Tumor Metastasis. Annu Rev Pathol. Jan
24 2018;13:395-412. doi:10.1146/annurev-pathol-020117-043854
45. Greten FR, Grivennikov SI. Inflammation and Cancer: Triggers, Mechanisms, and
Consequences. Immunity. Jul 16 2019;51(1):27-41. doi:10.1016/j.immuni.2019.06.025
46. Stewart CR, Stuart LM, Wilkinson K, et al. CD36 ligands promote sterile inflammation
through assembly of a Toll-like receptor 4 and 6 heterodimer. Nat Immunol. Feb 2010;11(2):155-
61. doi:10.1038/ni.1836
47. Sheedy FJ, Grebe A, Rayner KJ, et al. CD36 coordinates NLRP3 inflammasome
activation by facilitating intracellular nucleation of soluble ligands into particulate ligands in
sterile inflammation. Nat Immunol. Aug 2013;14(8):812-20. doi:10.1038/ni.2639
48. Xu C, Zhang C, Ji J, et al. CD36 deficiency attenuates immune-mediated hepatitis in
mice by modulating the proapoptotic effects of CXC chemokine ligand 10. Hepatology. May
2018;67(5):1943-1955. doi:10.1002/hep.29716
49. Perry JSA, Russler-Germain EV, Zhou YW, et al. Transfer of Cell-Surface Antigens by
Scavenger Receptor CD36 Promotes Thymic Regulatory T Cell Receptor Repertoire
Development and Allo-tolerance. Immunity. May 15 2018;48(5):923-936 e4.
doi:10.1016/j.immuni.2018.04.007
50. Ramakrishnan R, Tyurin VA, Veglia F, et al. Oxidized lipids block antigen cross-
presentation by dendritic cells in cancer. J Immunol. Mar 15 2014;192(6):2920-31.
doi:10.4049/jimmunol.1302801
51. Febbraio M, Hajjar DP, Silverstein RL. CD36: a class B scavenger receptor involved in
angiogenesis, atherosclerosis, inflammation, and lipid metabolism. J Clin Invest. Sep
2001;108(6):785-91. doi:10.1172/JCI14006
52. Huang SC, Everts B, Ivanova Y, et al. Cell-intrinsic lysosomal lipolysis is essential for
alternative activation of macrophages. Nat Immunol. Sep 2014;15(9):846-55.
doi:10.1038/ni.2956
53. van Dalen FJ, van Stevendaal M, Fennemann FL, Verdoes M, Ilina O. Molecular
Repolarisation of Tumour-Associated Macrophages. Molecules. Dec 20
2018;24(1)doi:10.3390/molecules24010009
78
54. Jayasingam SD, Citartan M, Thang TH, Mat Zin AA, Ang KC, Ch'ng ES. Evaluating the
Polarization of Tumor-Associated Macrophages Into M1 and M2 Phenotypes in Human Cancer
Tissue: Technicalities and Challenges in Routine Clinical Practice. Front Oncol. 2019;9:1512.
doi:10.3389/fonc.2019.01512
55. Cao Y. Adipocyte and lipid metabolism in cancer drug resistance. J Clin Invest. Jul 2
2019;129(8):3006-3017. doi:10.1172/JCI127201
56. Ye H, Adane B, Khan N, et al. Leukemic Stem Cells Evade Chemotherapy by Metabolic
Adaptation to an Adipose Tissue Niche. Cell Stem Cell. Jul 7 2016;19(1):23-37.
doi:10.1016/j.stem.2016.06.001
57. Farge T, Saland E, de Toni F, et al. Chemotherapy-Resistant Human Acute Myeloid
Leukemia Cells Are Not Enriched for Leukemic Stem Cells but Require Oxidative Metabolism.
Cancer Discov. Jul 2017;7(7):716-735. doi:10.1158/2159-8290.CD-16-0441
58. Kubo M, Gotoh K, Eguchi H, et al. Impact of CD36 on Chemoresistance in Pancreatic
Ductal Adenocarcinoma. Ann Surg Oncol. Feb 2020;27(2):610-619. doi:10.1245/s10434-019-
07927-2
59. Feng WW, Wilkins O, Bang S, et al. CD36-Mediated Metabolic Rewiring of Breast
Cancer Cells Promotes Resistance to HER2-Targeted Therapies. Cell Rep. Dec 10
2019;29(11):3405-3420 e5. doi:10.1016/j.celrep.2019.11.008
60. Long J, Zhang CJ, Zhu N, et al. Lipid metabolism and carcinogenesis, cancer
development. Am J Cancer Res. 2018;8(5):778-791.
61. Pavlova NN, Thompson CB. The Emerging Hallmarks of Cancer Metabolism. Cell
Metab. Jan 12 2016;23(1):27-47. doi:10.1016/j.cmet.2015.12.006
62. Rozeveld CN, Johnson KM, Zhang L, Razidlo GL. KRAS Controls Pancreatic Cancer
Cell Lipid Metabolism and Invasive Potential through the Lipase HSL. Cancer Res. Nov 15
2020;80(22):4932-4945. doi:10.1158/0008-5472.CAN-20-1255
63. Guerra L, Bonetti L, Brenner D. Metabolic Modulation of Immunity: A New Concept in
Cancer Immunotherapy. Cell Rep. Jul 7 2020;32(1):107848. doi:10.1016/j.celrep.2020.107848
64. Yang J, Park KW, Cho S. Inhibition of the CD36 receptor reduces visceral fat
accumulation and improves insulin resistance in obese mice carrying the BDNF-Val66Met
variant. J Biol Chem. Aug 24 2018;293(34):13338-13348. doi:10.1074/jbc.RA118.002405
79
65. Wang H, Franco F, Tsui YC, et al. CD36-mediated metabolic adaptation supports
regulatory T cell survival and function in tumors. Nat Immunol. Mar 2020;21(3):298-308.
doi:10.1038/s41590-019-0589-5
66. Goldman MJ, Craft B, Hastie M, et al. Visualizing and interpreting cancer genomics data
via the Xena platform. Nat Biotechnol. Jun 2020;38(6):675-678. doi:10.1038/s41587-020-0546-8
67. Choi J, Baldwin TM, Wong M, et al. Haemopedia RNA-seq: a database of gene
expression during haematopoiesis in mice and humans. Nucleic Acids Res. Jan 8
2019;47(D1):D780-D785. doi:10.1093/nar/gky1020
68. Chandrashekar DS, Bashel B, Balasubramanya SAH, et al. UALCAN: A Portal for
Facilitating Tumor Subgroup Gene Expression and Survival Analyses. Neoplasia. Aug
2017;19(8):649-658. doi:10.1016/j.neo.2017.05.002
69. Men C, Chai H, Song X, Li Y, Du H, Ren Q. Identification of DNA methylation
associated gene signatures in endometrial cancer via integrated analysis of DNA methylation and
gene expression systematically. J Gynecol Oncol. Nov 2017;28(6):e83.
doi:10.3802/jgo.2017.28.e83
70. Shinawi T, Hill VK, Krex D, et al. DNA methylation profiles of long- and short-term
glioblastoma survivors. Epigenetics. Feb 2013;8(2):149-56. doi:10.4161/epi.23398
71. Gao J, Aksoy BA, Dogrusoz U, et al. Integrative analysis of complex cancer genomics
and clinical profiles using the cBioPortal. Sci Signal. Apr 2 2013;6(269):pl1.
doi:10.1126/scisignal.2004088
72. Cerami E, Gao J, Dogrusoz U, et al. The cBio cancer genomics portal: an open platform
for exploring multidimensional cancer genomics data. Cancer Discov. May 2012;2(5):401-4.
doi:10.1158/2159-8290.CD-12-0095
73. Li T, Fu J, Zeng Z, et al. TIMER2.0 for analysis of tumor-infiltrating immune cells.
Nucleic Acids Res. Jul 2 2020;48(W1):W509-W514. doi:10.1093/nar/gkaa407
74. Li T, Fan J, Wang B, et al. TIMER: A Web Server for Comprehensive Analysis of
Tumor-Infiltrating Immune Cells. Cancer Res. Nov 1 2017;77(21):e108-e110. doi:10.1158/0008-
5472.CAN-17-0307
80
75. Li B, Severson E, Pignon JC, et al. Comprehensive analyses of tumor immunity:
implications for cancer immunotherapy. Genome Biol. Aug 22 2016;17(1):174.
doi:10.1186/s13059-016-1028-7
76. Glatz JFC, Luiken J. Dynamic role of the transmembrane glycoprotein CD36 (SR-B2) in
cellular fatty acid uptake and utilization. J Lipid Res. Jul 2018;59(7):1084-1093.
doi:10.1194/jlr.R082933
77. Peck B, Schulze A. Lipid Metabolism at the Nexus of Diet and Tumor
Microenvironment. Trends Cancer. Nov 2019;5(11):693-703. doi:10.1016/j.trecan.2019.09.007
78. Sun Q, Zhang W, Wang L, et al. Hypermethylated CD36 gene affected the progression of
lung cancer. Gene. Dec 15 2018;678:395-406. doi:10.1016/j.gene.2018.06.101
79. Crespo J, Sun H, Welling TH, Tian Z, Zou W. T cell anergy, exhaustion, senescence, and
stemness in the tumor microenvironment. Curr Opin Immunol. Apr 2013;25(2):214-21.
doi:10.1016/j.coi.2012.12.003
80. Cheng H, Wang Z, Fu L, Xu T. Macrophage Polarization in the Development and
Progression of Ovarian Cancers: An Overview. Front Oncol. 2019;9:421.
doi:10.3389/fonc.2019.00421
81. Pascual G, Avgustinova A, Mejetta S, et al. Targeting metastasis-initiating cells through
the fatty acid receptor CD36. Nature. Jan 5 2017;541(7635):41-45. doi:10.1038/nature20791
82. Zhang T, Yang J, Vaikari VP, et al. Apolipoprotein C2 - CD36 Promotes Leukemia
Growth and Presents a Targetable Axis in Acute Myeloid Leukemia. Blood Cancer Discov. Sep
2020;1(2):198-213. doi:10.1158/2643-3230.bcd-19-0077
83. Pernes G, Flynn MC, Lancaster GI, Murphy AJ. Fat for fuel: lipid metabolism in
haematopoiesis. Clin Transl Immunology. 2019;8(12):e1098. doi:10.1002/cti2.1098
84. Rozovski U, Harris DM, Li P, et al. STAT3-activated CD36 facilitates fatty acid uptake
in chronic lymphocytic leukemia cells. Oncotarget. Apr 20 2018;9(30):21268-21280.
doi:10.18632/oncotarget.25066
85. Landberg N, von Palffy S, Askmyr M, et al. CD36 defines primitive chronic myeloid
leukemia cells less responsive to imatinib but vulnerable to antibody-based therapeutic targeting.
Haematologica. Mar 2018;103(3):447-455. doi:10.3324/haematol.2017.169946
81
86. McGilvray ID, Serghides L, Kapus A, Rotstein OD, Kain KC. Nonopsonic
monocyte/macrophage phagocytosis of Plasmodium falciparum-parasitized erythrocytes: a role
for CD36 in malarial clearance. Blood. Nov 1 2000;96(9):3231-40.
87. Moffat J, Grueneberg DA, Yang X, et al. A lentiviral RNAi library for human and mouse
genes applied to an arrayed viral high-content screen. Cell. Mar 24 2006;124(6):1283-98.
doi:10.1016/j.cell.2006.01.040
88. Yousefzadeh MJ, Wyatt DW, Takata K, et al. Mechanism of suppression of chromosomal
instability by DNA polymerase POLQ. PLoS Genet. Oct 2014;10(10):e1004654.
doi:10.1371/journal.pgen.1004654
89. Ruiz-Villalba A, Mattiotti A, Gunst QD, Cano-Ballesteros S, van den Hoff MJ, Ruijter
JM. Reference genes for gene expression studies in the mouse heart. Sci Rep. Feb 2
2017;7(1):24. doi:10.1038/s41598-017-00043-9
90. Kreitz J, Schonfeld C, Seibert M, et al. Metabolic Plasticity of Acute Myeloid Leukemia.
Cells. Jul 31 2019;8(8)doi:10.3390/cells8080805
91. Zhang CC, Lodish HF. Cytokines regulating hematopoietic stem cell function. Curr Opin
Hematol. Jul 2008;15(4):307-11. doi:10.1097/MOH.0b013e3283007db5
92. DeBerardinis RJ, Chandel NS. Fundamentals of cancer metabolism. Sci Adv. May
2016;2(5):e1600200. doi:10.1126/sciadv.1600200
93. Heit B, Kim H, Cosio G, et al. Multimolecular signaling complexes enable Syk-mediated
signaling of CD36 internalization. Dev Cell. Feb 25 2013;24(4):372-83.
doi:10.1016/j.devcel.2013.01.007
94. Kaur B, Cork SM, Sandberg EM, et al. Vasculostatin inhibits intracranial glioma growth
and negatively regulates in vivo angiogenesis through a CD36-dependent mechanism. Cancer
Res. Feb 1 2009;69(3):1212-20. doi:10.1158/0008-5472.CAN-08-1166
Abstract (if available)
Abstract
CD36 is a cell surface receptor that is widely expressed in all kinds of cell types in the human body, such as adipocytes, microphages, platelets, myocytes, and hepatocytes. The primary function of CD36 is transporting fatty acids in lipid metabolism. So, as a fatty acid translocase, CD36 is associated with several metabolic disorders and involved in the pathogenic mechanisms of many diseases. The increased expression of CD36 is examined in breast cancer, oral cancer, melanoma, prostate cancer, pancreatic cancer, glioblastoma, ovarian cancer, esophageal cancer, and hematopoietic malignancies. In cancer cells, the upregulated CD36 level provides fuel to support the increased cell proliferation, as well as the adaptation to the tumor microenvironment (TME). Also, the decreased CD36 expression is reported in some metastatic tumors such as colon cancer, and some cases of breast cancer, where the loss of CD36 is primarily found in the endothelial cells resulting in an inhibition on thrombospondin receptor regulated antitumor effects. CD36 has recently gained interest as a therapeutic target in acute myeloid leukemia (AML). It was found to interact with apolipoprotein C2 to promote leukemia growth. However, whether targeting CD36 in normal hematopoietic cells poses any toxicity remains to be established in order to bring this therapeutic strategy to maturity. ? In this thesis, I used public datasets to investigate the genomics, transcriptomics, and methylation patterns of CD36 in both normal and cancer tissues. I noticed that CD36 deregulation was not always consistent with its promoter methylation status. I found that CD36 expression was upregulated in kidney renal clear cell carcinoma (KIRC) and glioblastoma; and the CD36 promoter was hypermethylated in KIRC. I also observed remarkably downregulated CD36 expression levels in breast cancer and lung cancer, where the CD36 promoter was hypermethylated in both two cancer types. Besides, the tumor tissues exhibited lower CD36 expression compared with normal tissues for colon cancer, prostate cancer, and head and neck cancer; yet the CD36 promoter was hypomethylated in all three cancer types. I next examined the correlation between CD36 expression and the patient’s clinical outcome and the infiltration level of tumor immune cells. The results indicated a significantly positive correlation between CD36 expression and infiltration levels of tumor immune cells such as CD4+ T cells, CD8+ T cells, neutrophils, natural killer cells, and dendritic cells during cancer development. Moreover, I speculated that CD36 expression probably antagonized the T cell mediated cytotoxic effects on malignant cells, since CD36 expression also showed a significantly positive correlation with gene markers presented during T cell exhaustion, and the differentiation of Treg cells, and M2-like tumor-associated macrophages. In addition, I investigated the role of Cd36 in normal hematopoietic cells using gain- and loss-of-function approaches in murine normal hematopoietic primary cells. The loss of function in Cd36 showed limited effects on the proliferation of mouse healthy hematopoietic primary cells. Altogether, my work demonstrates that CD36 is highly deregulated in cancer and within the cancer immune microenvironment. Yet targeting CD36 in normal murine hematopoietic cells had limited effect on cell viability. Whether CD36 may potentially present a viable therapeutic approach in specific malignancies, remain to be investigated in preclinical models.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Investigating CD99 as a therapeutic target in acute myeloid leukemia
PDF
APOC2 presents a viable therapeutic target in cancer
PDF
Enhancing the anti-cancer specificity of chimeric antigen receptor T cells through targeting HLA loss
PDF
Investigating the effects of targeting CD99 on T cells to enhance their antileukemia activity
PDF
The role of CD99 in T cells
PDF
Clinical, functional and therapeutic analysis of CD99 in acute myeloid leukemia
PDF
MAO a deficient mice exhibit an altered immune system in the brain and prostate
PDF
Genomic and transcriptomic alterations of apolipoproteins genes in cancers
PDF
Immunotherapy of cancer
PDF
The effect of tumor-mediated immune suppression on prostate cancer immunotherapy
PDF
Optimization of nanomedicine based drug delivery systems for the treatment of solid tumors
PDF
Chimeric Antigen Receptor targeting Prostate Specific Membrane Antigen (PSMA)
PDF
Engineering chimeric antigen receptor (CAR) -modified T cells for enhanced cancer immunotherapy
PDF
Pancreatic cancer: a review on biology, genetics and therapeutics
PDF
Study of a novel near-infrared conjugated MAOA inhibitor, NMI, against CNS cancer by NCI60 data analysis
PDF
Investigating the effects of T cell mediated anti-leukemia activity in FLT3-ITD positive acute myeloid leukemia
PDF
The expression of human carboxylesterases in normal tissues and cancer cell-lines
PDF
The immunomodulatory effects of midostaurin on T cells
PDF
Structure-based computational analysis and prediction of TCR CDR3 loops in the TCR-peptide-MHC complex using solvation parameters and peptide molecular dynamics.
PDF
Decreased levels of expression of transmembrane protein 56 (TMEM56) in breast cancer tissues
Asset Metadata
Creator
Meng, Yiting
(author)
Core Title
Deregulation of CD36 expression in cancer presents a potential targeting therapeutic opportunity
School
School of Pharmacy
Degree
Master of Science
Degree Program
Pharmaceutical Sciences
Degree Conferral Date
2021-08
Publication Date
07/28/2021
Defense Date
07/25/2021
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
CD36,CD36 knockdown,CD36 overexpression,immune infiltration,immune suppression,normal hematopoietic cells,OAI-PMH Harvest,TCGA cancer
Format
application/pdf
(imt)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Alachkar, Houda (
committee chair
), Haworth, Ian (
committee member
), Okamoto, Curtis (
committee member
)
Creator Email
mengyiting407@gmail.com,yitingme@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-oUC15658936
Unique identifier
UC15658936
Legacy Identifier
etd-MengYiting-9913
Document Type
Thesis
Format
application/pdf (imt)
Rights
Meng, Yiting
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
Tags
CD36
CD36 knockdown
CD36 overexpression
immune infiltration
immune suppression
normal hematopoietic cells
TCGA cancer