The role of b cells in the pathogenesis of atherosclerosis 1

Therefore, we aim to address the following questions in atherosclerotic apoE -/- mouse model: i in which secondary lymphoid organs the antibody response against oxLDL takes place; ii whi

THE ROLE OF B CELLS IN THE PATHOGENESIS OF

YONG LOO LIN SCHOOL OF MEDICINE

NATIONAL UNIVERSITY OF SINGAPORE

2013

DECLARATION

I hereby declare that the thesis is my original work and it has been written by

me in its entirety I have duly acknowledged all the sources of information

which have been used in the thesis

This thesis has also not been submitted for any degree in any university

previously

-

Khoo Han Boon Lawrence

July 2013

Next, I’ll like to thank my labmates who accompanied me along the journey of my graduate studies Thank you Fiona, for being a steadfast partner during mice harvest and your help in dissecting the aortas Thank you Jocelyn, for the helpful discussions To Angeline, Kim, Kar Wai, Serena, Sheau Ying and Michael, thank you for the help and support in lab matters

I’ll also like to thank many friends I made in the Immunology Programme as well as in Singapore Immunology Network (SIgN) for the help they rendered in however big or small and not to mention the light-hearted moments! I’ll like to specially thank Dr Paul Hutchinson, Guo Hui and Fei Chuin for assistance in cell sorting experiments Also, special thanks to Dr Florent Ginhoux and Peter See for providing the CD11c-GFP-DTR mice

Last but not least, I’ll like to dedicate this thesis to my family and express my deepest gratitude to my parents for their unwavering support in the pursuit of my graduate studies

Table of Contents

DECLARATION ……… ii

ACKNOWLEDGEMENTS ……… … iii

TABLE OF CONTENTS ……… iv

ABSTRACT ……… xv

LIST OF TABLES xvii

LIST OF FIGURES xvii

LIST OF ABBREVIATIONS xxi

Chapter 1: Introduction 1.1 Main Immunological Features in Atherosclerosis 2

1.2 Experimental Mice Models of Atherosclerosis 5

1.3 B cell subset populations 7

1.3.1 B1 cells 7

1.3.2 B2 cells 10

1.3.2.1 Follicular B cells 10

1.3.2.2 Marginal Zone B cells 11

1.4 Humoral Immunity in Atherosclerosis 12

1.4.1 Localization of B cells in Atherosclerosis 12

1.4.2 Association of oxLDL-specific antibodies with Atherosclerosis 14

1.4.3 Cloning of oxLDL-specific B cells 18

1.4.4 B1 cell origin of oxLDL-specific monoclonal antibodies 19

1.4.5 Function of oxLDL-specific Antibodies 21

1.4.6 The role of B cells in Atherosclerosis: Pathogenic or protective? 22

1.5 Humoral B cell responses 24

1.5.1 Critical factors in differential antibody producing pathways 24

1.5.1.1 Antigen-BCR Interactions 25

1.5.1.2 Epstein-Barr virus-induced molecule 2 (EBI2) 25

1.5.1.3 B cell responses master regulators 27

1.5.2 Extrafollicular responses 29

1.5.3 Germinal centres 34

1.5.4 Long-lived plasma cells 40

1.5.4.1 Migratory competency 44

1.5.4.2 Survival factors for plasma cells 47

1.5.4.3 Role of CD28 on longevity of plasma cells 50

1.5.4.4 Cellular sources of survival factors 51

1.6 Objectives 53

Chapter 2: Materials & Methods

2.1 Mice 55

2.1.1 Generation of mice chimera 55

2.1.2 Splenectomy 55

2.2 Immunization 56

2.3 Adoptive transfer 56

2.4 Ezetimibe treatment 56

2.5 BrdU and EdU administration 56

2.6 Plasma collection 57

2.7 Cell isolation from tissues 57

2.7.1 Isolation of cells from spleen and lymph nodes 57

2.7.2 Isolation of cells from bone marrow 58

2.7.3 Isolation of cells from peritoneal 58

2.8 Viability cell count 58

2.9 Flow cytometry 59

2.9.1 Cell surface staining 59

2.9.2 Intracellular staining 59

2.9.3 EdU staining for flow cytometry 60

2.9.4 5-bromodeoxyuridine (BrdU) staining 60

2.9.5 Flow cytometry acquisition and analysis 61

2.9.6 Cell sorting 61

2.10 Enzyme-Linked Immunosorbent Assay (ELISA) 62

2.10.1 Quantification of total IgM and IgG 62

2.10.2 Measurement of anti-LDL and anti-oxLDL autoantibodies 62

2.10.3 Quantification of PC-specific antibodies 63

2.10.4 Measurement of OD and analysis 63

2.11 Enzyme-Linked Immunosorbent Spot (ELISpot) 64

2.11.1 Image acquisition and analysis 64

2.12 Cell culture 65

2.13 Immunofluorescence 66

2.13.1 Preparation of tissues for immunofluorescence staining 66

2.13.2 Immunofluorescence staining 66

2.13.2.1 Immunofluorescence staining with primary and/or secondary antibodies 66

2.13.2.2 Immunofluorescence staining with biotin-conjugated antibodies 67

2.13.2.3 Immunofluorescence staining with anti-mouse CD138 antibodies 67

2.13.2.4 Revealing EdU + cells in tissue sections 68

2.13.3 Image acquisition and analysis 68

2.14 Quantification of germinal center and extrafollicular responses 68

2.15 Quantification of lesion size 69

2.16 Oil Red-O staining 69

2.16.1 Image acquisition 69

2.17 Real-time Polymerase Chain Reaction (real-time PCR) 70

2.17.1 RNA Extraction 70

2.17.2 Reverse transcription 70

2.17.3 Primer Design 71

2.17.4 Real-time PCR Protocol 71

2.18 Mass spectroscopy analysis of samples 72

2.18.1 Extraction of sterols from tissue samples 72

2.18.2 High Performance Liquid Chromatography – Mass Spectrometry (HPLC-MS) method 72

2.19 Statistical Analysis 73

Chapter 3: Results

3.1 Characterization of antibody responses in plasma of atherosclerotic

experimental mice model, apoE -/- 75

3.1.1 Elevated total IgM and oxLDL-specific IgM autoantibodies in circulation with disease progression 75

3.1.2 No increase in PC-specific IgM autoantibodies in circulation 76

3.2.2 Evaluation of antibody-producing plasma cells in secondary

lymphoid organs in apoE -/- mice 84

3.2.3 Evaluation of antibody producing pathways in secondary

lymphoid organs in apoE -/- mice 86

3.2.3.1 Germinal center reactions increase significantly in the lymph nodes 86

3.2.3.2 Robust extrafollicular responses were observed in the

spleen of apoE -/- mice 91

3.2.4 Evaluation of antibody responses in secondary lymphoid organs

in apoE -/- mice 96

3.2.5 Summary 100

3.3 Evaluation of antibody producing pathways in contributing total IgM and oxLDL-specific IgM autoantibodies in spleen of apoE -/- mice 101

3.3.1 Splenic GCs from apoE -/- mice were not defective in producing isotype-switched plasma cells 101

3.3.2 Plasmablasts in extrafollicular responses were IgM + 104

3.3.3 Summary 108

3.4 Evaluation of molecular cues to direct extrafollicular responses in the spleen of apoE -/- mice 109

3.4.1 Increased ch25h mRNA expression in the spleen of apoE -/-mice 109

3.4.2 Increased oxysterol in the spleen of apoE -/- mice 110

3.4.3 Summary 112

3.5 Antibody production of B1a cells in apoE -/- mice 113

3.5.1 B1a cells were not expanded in PEC of apoE -/- mice 113

3.5.2 Increased splenic B1a cells differentiation into IgM + plasmablasts 114

3.5.3 Summary 118

3.6 Evaluation of the impact of splenic CD138 + ASCs on atherosclerosis 119

3.6.1 Adoptive transfer of splenic CD138 + ASCs into apoE

-/-mice 119

3.6.2 Summary 123

3.7 Evaluation of ASCs in bone marrow of apoE -/- mice 124

3.7.1 Accumulation of IgM + long-lived plasma cells in the bone

marrow of apoE -/- mice 124

3.7.2 Summary 127

3.8 Evaluation of IgM + plasmablasts migration from the spleen to bone

marrow of apoE -/- mice 128

3.8.1 Bone marrow was generating humoral responses in the absence

of spleen in apoE -/- mice 128

3.8.2 Evaluation of bone marrow compartment after disruption of humoral responses in the spleen 131

3.9 Evaluation of humoral responses being associated with

hypercholesterolemic setting in the apoE -/- mice 142

3.9.1 Evaluation of total IgM and oxLDL-specific IgM autoantibodies

in apoE -/- mice after ezetimibe treatment 143

3.9.2 Evaluation of splenic extrafollicular responses in apoE -/- mice after ezetimibe treatment 145

3.9.3 Evaluation of IgM + antibody secreting cells in the spleen of

apoE -/- mice after ezetimibe treatment 147

3.9.4 Evaluation of GC in the lymph node compartment of apoE

-/-mice 150

3.9.5 Evaluation of IgM + antibody secreting cells in the bone marrow

of apoE -/- mice after ezetimibe treatment 155

3.9.6 Summary 157

Chapter 4: Discussion

4.1 Summary of findings 159

4.2 Functional importance of IgM antibodies 161

4.3 Splenic humoral responses 163

4.4 Molecular cues in extrafollicular responses 164

4.5 Presence of oxLDL in plasma 166

4.6 Existence of tolerance mechanism? 168

4.7 Role of B cell subpopulations 169

4.8 Absence of oxLDL-specific humoral response in lymph node 171

4.9 Immune response or homeostatic response against oxidative modified lipids? 173

4.10 IgM + plasma cells in bone marrow 175

4.11 In situ generation of oxLDL-specific IgM plasma cells in bone marrow? 176

4.12 Trafficking to bone marrow 177

4.13 Selection into bone marrow compartment 179

4.15 Implication of study 180

4.14 Limitations of study 182

4.15 Future Work 183

4.15.1 Measurement of oxLDL in plasma 183

4.15.2 Importance of 7α, 25-OHC in mediating extrafollicular responses 183

4.15.3 Participation of B1a cells in splenic extrafollicular responses 184

4.15.4 Survival factors present in bone marrow in atherosclerosis 186

4.16 Conclusion 188

References 189

Appendix 1 Buffers and Media 227

Appendix 2 List of antibodies used in flow cytometry 228

Appendix 3 List of antibodies used in immunofluorescence 229

Appendix 4 List of primers used 230

ABSTRACT

Atherosclerosis is a chronic inflammatory disease of the arteries The presence of modified self-lipid specific autoantibodies such as anti-oxidized low-density lipoprotein (oxLDL) has been reported in clinical and experimental atherosclerosis, implying the involvement of humoral B cell responses However, the understanding of the mechanism(s) by which B cells differentiate into plasma cells in response to atherosclerosis and how these autoantibodies are sustained have been poorly explored Therefore, we aim to

address the following questions in atherosclerotic apoE -/- mouse model: (i) in which secondary lymphoid organs the antibody response against oxLDL takes place; (ii) which antibody producing pathway(s) – extrafollicular responses or germinal centre (GC) reaction is involved in the development of this B cell response; (iii) whether the bone marrow compartment may sustain this antibody response

In the present study, our ELISA analysis indicated elevated total and

oxLDL-specific IgM autoantibodies in the circulation of atherosclerotic apoE /- mice These observations corresponded to an increase in the population of B220-CD138+ plasma cells detected in the spleen and lymph nodes, suggesting that differentiation into plasma cells occurs at these sites However, the dominant antibody-producing pathway that took place in these secondary lymphoid organs was different, namely: robust extrafollicular responses in the

-spleen and increased GC reactions in the lymph nodes In our ELISpot and in

vitro studies, we identified the spleen but not the lymph nodes as a major

source of elevated circulating IgM antibodies Critically, the oxLDL-specific

IgM autoantibody response in apoE -/- mice was restricted to the splenic

compartment In support of our interpretation that IgM+ plasma cells were extrafollicular response-derived, GC reactions that were not dysfunctional in isotype-class switching and proliferating IgM+ plasmablasts in extrafollicular

responses were detected in the spleen of apoE -/- mice The expansion of the

extrafollicular response in apoE -/- mice was associated with an increased

generation and bioavailability of 7α, 25-OHC, the ligand for EBI2 expressed

on activated B cells which supports B cell migration to the bridging channel of follicles to participate in the extrafollicular responses B1a cells, which are implicated in the secretion of oxLDL-specific antibodies, were differentiating into IgM+ plasmablasts in the spleen This increase of oxLDL-specific IgM autoantibodies and antibody secreting cells (ASCs) in plasma and spleen of

apoE -/- mice, respectively, was also accompanied with an increased population

of oxLDL-specific ASCs in the bone marrow, suggesting their migration into this primary lymphoid organ Indeed, we demonstrated the appearance of new migrant IgM+ plasma cells into the bone marrow of apoE -/- mice as well as the presence of long-lived IgM+ plasma cells using EdU/BrdU labeling These

findings were further confirmed using an ldlr -/- bone marrow chimera reconstituted with CD11c-GFP-DTR in which disrupted splenic humoral responses led to decreased frequency of total IgM ASCs and new migrant BrdU+IgM+ plasma cells Lastly, we demonstrated that the IgM humoral response was associated with hypercholesterolemic condition in these mice

Collectively, our findings suggests that splenic extrafollicular responses, which are associated with hypercholesterolemia, contribute to elevated total and oxLDL-specific IgM autoantibodies and are sustained by long-lived plasma cells in the bone marrow

List of Tables

Table 1 Distribution and function of B1 cells in steady state ………… 9

Table 2 Linearization protocol 70

Table 3 Elongation protocol 71

List of Figures Figure 1 The pathogenesis of atherosclerosis …… ………….………… 4

Figure 2 Enzymatic actions of CH25H, CYP7B1 and HSD3B7 on oxysterol production 26

Figure 3 Roles of Bcl-6 and Blimp-1 in B cell differentiation 28

Figure 4 The main splenic compartments and their relation to CD11c hi plasmablasts-associated DCs 29

Figure 5 The four phases of splenic extrafollicular responses occur in distinct microenvironment 32

Figure 6 The germinal centre microenvironment 37

Figure 7 Three concepts for the maintenance of serum concentrations of specific antibody 40

Figure 8 Ontogeny of plasma cells 43

Figure 9 Model of lymphocyte egress 45

Figure 10 Plasma cell survival niche 52

Figure 11 Significant increases of total and anti-oxLDL IgM

autoantibodies in plasma of apoE -/- mice in late time course analysis 78

Figure 12 Significant increase of anti-LDL IgM autoantibodies in plasma

of apoE -/- mice in late time course analysis 79

Figure 13 No difference in PC-specific IgM antibodies in circulation of 24

weeks old apoE -/- mice 80

Figure 14 Lymph node hypertrophy in apoE -/- mice 83

Figure 15 Increased relative percentage of plasma cells in the spleen and

LNs of apoE -/- mice 85

Figure 16 Increased GC B cells in the lymph nodes, but not in the spleen

of apoE -/- mice 88

Figure 17 Increased germinal centers in the lymph nodes, but not in the

spleen of apoE -/- mice 89-90

Figure 18 Robust extrafollicular responses in the spleen of apoE

-/-mice 93-95

Figure 19 The spleen, but not the lymph nodes, is a source of increased

total IgM and anti-oxLDL specific IgM autoantibodies in apoE

-/-mice 98-99

Figure 20 GC B cells are not defective in isotype class-switching 103

Figure 21 IgM + plasmablasts were generated in extrafollicular responses

in the spleen of apoE -/- mice 106-107

Figure 22 Elevated 7α, 25-OHC oxysterol in the spleen of apoE

-/-mice 111

Figure 23 No difference in B1a cell population in peritoneal cavity of

apoE -/- mice 116

Figure 24 Increased population of B1a cells differentiating into IgM +

plasmablasts in the spleen of apoE -/- mice 117

Figure 25 No differences in lesion size of apoE -/- mice after adoptive transfer of ASCs 121-122

Figure 26 Increased IgM + long-lived plasma cells in bone marrow of

apoE -/- mice 126

Figure 27 The bone marrow compartment of splenectomized apoE -/- mice

in generating humoral responses in the absence of spleen 130

Figure 28 Increased total IgM, but not anti-oxLDL IgM, was observed in

circulation of ldlr -/- mice 133

Figure 29 Increased extent of splenic extrafollicular responses in ldlr

-/-mice 134

Figure 30 Increased accumulation of newly formed plasma cells in the

bone marrow of apoE -/- mice 136

Figure 31 Decreased IgM + plasma cells in bone marrow after depletion of CD11c-expressing cells 139-140

Figure 32 Total IgM and oxLDL-specific IgM autoantibodies were associated with hypercholesterolemia 144

Figure 33 Splenic extrafollicular responses were associated with hypercholesterolemia 146

Figure 34 Elevated total IgM and anti-oxLDL IgM autoantibodies were mostly contributed by proliferating IgM + ASCs 149

Figure 35 Decreased lymph nodes cellularity after ezetimibe treatment in

apoE -/- mice 152

Figure 36 Decreased GC B cells in the lymph nodes of apoE -/- mice after ezetimibe treatment 153

Figure 37 Germinal centres were smaller in the lymph nodes of apoE

-/-mice after ezetimibe treatment 154

Figure 38 Decreased frequency of anti-oxLDL ASCs in the bone marrow

of ezetimibe treated apoE -/- mice 156

Figure 39 Schematic diagram of humoral responses in sustaining the

elevated IgM and oxLDL-specific IgM autoantibodies responses in apoE

-/-mice 160

Figure 40 Differential adoptive transfer strategies of PEC B1 cells of IgM a allotype to investigate contribution of B1a cells to differentiate into IgM + ASCs in apoE -/- mice 186

List of Abbreviations

ABTS - 2,2'-azino-bis(3-ethylbenzothiazoline-6-sulphonic acid)

AEC - 3-amino-9-ethylcarbazole

AID - activation-induced deaminase

AIF - apoptosis-inducing factor

AMI – acute myocardial infarction

ApoB100 – apolipoprotein B100

APRIL - A Proliferation-inducing ligand

ASCs - antibody-secreting cells

BAFF - B cell activating factor

Bax - Bcl-2-associated X protein

BSA - bovine serum albumin

CAD - coronary artery disease

CGG - chicken gamma globulin

CFA – complete Freund’s adjuvant

DCs – dendritic cells

DCIR2 - DC-inhibitory receptor 2

DEPC - Diethylpyrocarbonate

DT - diphtheria toxin

DTR - diphtheria toxin receptor

EBI2 - Epstein-Barr virus-induced molecule 2

EDTA - ethylenediaminetetraacetic acid

EdU - 5-ethynyl-2’-deoxyuridine

ELISA - Enzyme-Linked Immunosorbent Assay

ELISpot - Enzyme-Linked Immunosorbent Spot

FACS - fluorescence-activated cell sorting

FBS - fetal bovine serum

FDCs - follicular dendritic cells

GC – germinal center

GFP - green fluorescence protein

HBSS - Hanks’ Balanced Salt Solution

HEL - hen egg lysozyme

HPLC-MS – high performance liquid chromatography mass spectrometry HPRT1 - hypoxanthine phosphoribosyltransferase 1

HRP - horseradish peroxidase

ICAM-1 - intercellular adhesion molecule 1

IFA – incomplete Freund’s adjuvant

IMT – intima-media thickness

iNKT – invariant natural killer T cells

IFN-γ - interferon-gamma

i.p - intraperitoneal

i.v – intravenous

IVIg - Intravenous immunoglobulin

LCMV - Lymphocytic choriomeningitis virus

LDL – low-density lipoprotein

LDLR – low-density lipoprotein receptor

LFA-1 - lymphocyte function-associated antigen 1

LRP - LDL receptor-related protein

LPS – lipopolysaccharide

LTβ-R - lymphotoxin beta receptor

LYVE-1 - Lymphatics vessel endothelial hyaluronan receptor MALT - mucosal-associated lymphoid tissue

M-CSF – macrophage colony stimulating factor

oxLDL – oxidized low-density lipoprotein

PALS - periarteriolar lymphoid sheaths

PAX5 - paired box protein 5

PBS - phosphate buffered saline

PC – phosphorylcholine

PCR - Polymerase Chain Reaction

PE – phycoerthrin

pMHC - peptide-major histocompatibility complex

PPS-3 - pneumococcal polysaccharide type 3

prdm1 - positive regulatory domain containing 1

ROS – reactive oxygen species

RPMI - Roswell Park Memorial Institute

SA – streptavidin

SHM - somatic hypermutation

SMCs - smooth muscle cells

SRBCs – sheep red blood cells

SR-BI - scavenger receptor class B, type I

Sx – splenectomized

S1P - sphingosine-1-phosphate

S1P1 - sphingosine 1 phosphate receptor-1

TACI - Trans-membrane activator and calcium modulator cyclophilin ligand

TI-1 - T-independent type 1

TI-2 - T-independent type 2

TLR - toll-like receptor

TSA – tyramide signal amplification

UPR - unfolded protein response

VCAM-1 - vascular cell-adhesion molecule-1

VLA-4 - very late antigen -4

vLDL - very low-density lipoprotein

XBP-1 - X-box binding protein-1

4-HNE – 4-hydroxynonenal

7α, 25-OHC - 7α, 25-dihydroxycholesterol

Chapter 1 Introduction

1.1 Main Immunological Features in Atherosclerosis

Cardiovascular disease (CVD) is a major health problem as it becomes the leading cause of mortality worldwide (Lopez et al., 2006) According to a recent WHO report released in 2011, it lead the proportion of non-communicable deaths under the age of 70 in 2008 as much as 39% (Alwan, 2011) Risk factors for CVD includes hypertension, high cholesterol and tobacco use (Dahlof, 2010)

Among CVD, atherosclerosis is a chronic inflammatory disease where accumulation of lipids results in asymmetrical focal thickenings in the intima layer of the arterial wall known as atherosclerotic lesions or plaque Such an occurrence typically takes place at atherosclerotic-susceptible segments of the aorta eg artery branching points at aortic arch and abdominal segments, which experience mechanistic hemodynamic forces (Dai et al., 2004)

Consequently, it results in activation of endothelial cells to promote adhesion of leukocytes and platelets, through the expression of adhesion molecules, in particular vascular cell-adhesion molecule-1 (VCAM-1) (Cybulsky and Gimbrone, 1991; Massberg et al., 2002) At the same time, the dysfunctional activation of endothelial cells also leads to increased permeability and retention of lipid components such as apolipoprotein B100 (ApoB100) in low-density lipoprotein (LDL) where it associates with proteoglycans in the extracellular matrix (Tabas et al., 2007) As a result, the retained pool of LDL in the intima layer of the arterial wall undergo oxidative modification by reactive oxygen species (ROS) and myeloperoxidases to form oxidized LDL (oxLDL) (Hansson and Hermansson, 2011) This process

further contributes to the activation of endothelial cells to promote monocyte adhesion (Berliner et al., 1990; Hansson and Hermansson, 2011; Shih et al., 1999) Both retention of lipids and oxidative factors play a significant role in the formation of foam cells, a critical early event in atherogenesis, as macrophage colony-stimulating factor (M-CSF) dependent monocyte-derived macrophages uptake oxLDL via scavenger receptors such as scavenger receptor class B, type I (SR-BI) and CD36 in the developing plaque (Greaves and Gordon, 2009; Smith et al., 1995) Many other immune cell types are also recruited such as neutrophils, mast cells, basophils, eosinophils and T cells, in particular of Th1 subset, but they are of less abundance than that of macrophages (Weber et al., 2008)

Over time, the early plaque matures into a more complex structure as more inflammatory cell subsets and lipids accumulate within the intima This results in the formation of a lipid-rich necrotic core due to failure or inefficient clearance of dead cells and is surrounded by a fibrous cap of smooth muscle cells (SMCs) (Tabas, 2010) In experimental animals, the SMCs forming the fibrous cap are recruited from the media to the intima layer, which is unlike in humans where there exist resident intima SMCs Although relatively stable, the developing plaque causes clinical manifestation of blood flow restricted stenosis, which potentially leads to tissue ischaemia However, the stability of the plaque could be compromised as matrix degrading proteases such as matrix metalloproteinases (MMPs) and cysteine proteases, and cytokines such

as interferon-gamma (IFN-γ) are constantly secreted in the plaque that results

in the thinning of fibrous cap (Amento et al., 1991; Galis et al., 1994; Hansson

et al., 1989) Physical disruption of plaque ensues, releasing pro-thrombotic

material allowing coagulation cascades to occur leading to arterial occlusion

or thrombosis As such, the sudden onset of acute myocardial infarction (AMI)

as well as strokes could take place, paradoxically, in the absence of severe

arterial stenosis (Figure1) (Hansson and Libby, 2006; Libby et al., 2011)

Figure 1 The pathogenesis of atherosclerosis a The recruitment of

leukocytes and platelets to atherosclerotic susceptible arterial region after activation of endothelial cells under pro-inflammatory conditions in

hypercholesterolemia b Increased permeability allowed accumulation of

monocyte-derived macrophages taking up modified cholesterol to become foam cells Continuous influx of cells, matrix deposition and recruitment of

SMCs leads to fibroproliferative progression of the plaque c The failure or

resolution of inflammation leads to the formation of lipid-rich necrotic core that is encapsulated by fibrous cap Neovascularization could occur at the

adventitia layer and leakage could result plaque haemorrhage d Thinning of

fibrous cap due to the action of metalloproteinases and cytokines such as

IFN-γ could lead to plaque rupture causing thrombosis (Adapted from Weber et al

2008)

1.2 Experimental Murine Models of Atherosclerosis

Like many other diseases, the direct examination of human atherosclerotic lesion represents obvious challenges such as ethical issues and availability of biopsy samples Transgenic animal models are, thus, constructed and utilized to yield important insights into different aspects of atherosclerosis This approach often relies on genetic ablation of critical genes

to achieve hypercholesterolemia unlike humans where the disease is multifactorial Exceptions include patients diagnosed with homozygous familial hypercholesterolemia – an autosomal dominant disorder, which do not result in having functional low-density lipoprotein receptor (LDLR), and of even rarer occurrence, homozygous familial null mutation of apoE (Nemati and Astaneh, 2010; Schaefer et al., 1986)

Of all the experimental models of atherosclerosis, apoE -/- and ldlr

-/-transgenic mice are most commonly used for research due to advantages such

as the lower cost relative to larger animals, space required for colony maintenance and reasonable time frame for disease development Although, their disease development overlaps to a certain extent in humans, they do not typically, however, develop unstable plaque, which leads to thrombosis or AMI (Bentzon and Falk, 2010)

Normal wild-type mice are highly resistant to atherosclerosis Only when fed with very high cholesterol diet containing cholic acids, certain susceptible inbred strain such as C57/BL6 developed foam cells layers However, the fatty streak lesions do not progress to fibrous plaque similar to human atherosclerotic lesions In addition, the very high cholesterol diet,

including unnatural component cholic acid to achieve fatty streak lesions is not physiological (Breslow, 1996)

The strain apoE -/- was generated through homologous recombination in embryonic stem cells (Plump et al., 1992; Zhang et al., 1992) The loss of

apoE results in the delayed clearance of plasma LDL and very low-density

lipoprotein (vLDL) via LDLR and LDL receptor-related protein (LRP) Consequently, when fed with normal/ chow diet, these mice display high plasma cholesterol levels between 300 to 500 mg/dL and notably, develop extensive plaque lesions similar to that of humans (Nakashima et al., 1994)

To accelerate the disease progression, apoE -/- mice could be placed on high fat

or Western diet which would significantly increase plasma cholesterol to

>1000 mg/dL (Plump et al., 1992)

LDLR is required for the clearance of plasma LDL through endocytosis into cells taking place mainly in the liver (Nemati and Astaneh, 2010; Willnow et al., 1999) Unlike spontaneous development of

hypercholesterolemia in apoE -/- mice, ldlr -/- mice have to be placed on high fat diet to develop hypercholesterolemia This is because apoB48, an isoform of

apoB100 associated with LDL, exists in ldlr -/- that could be cleared via LRP (Veniant et al., 1998) Though lesion development is slower than apoE-/- mice

placed on high fat diet, ldlr -/- mice present a distinct advantage over apoE mice which is that the transfer of bone marrow cells from apoE +/+ mice in a bone marrow chimera experiment, does not lead to reduced plasma lipids level

-/-and ameliorate atherosclerosis (Linton et al., 1995)

1.3 B cell subset populations

Before we begin to understand the humoral immunity in atherosclerosis, it is important to introduce the existence of different B cell sub-populations, which are developmentally divergent into 2 main populations – B1 and B2 subsets However, whether these two populations are derived from a common or different progenitors remained controversial (Dorshkind and Montecino-Rodriguez, 2007)

1.3.1 B1 cells

B1 cells could be found in many different tissues but they are particularly enriched in peritoneal and pleural cavities, and intestinal lamina

propria (Table 1) (Baumgarth, 2011) These B cells are termed “B1 cells”

because they are the first B cells to arise during fetal development on day 8.5

of mouse embryonic developmental stage in the splanchnopleura region prior

to fetal liver colonization (Dorshkind and Montecino-Rodriguez, 2007; Godin

et al., 1993) These B1 cells could be identified through their surface phenotypic markers CD19+CD11b+sIgMhisIgDlo In addition, the surface marker CD5 further delineates the B1 population into B1a (CD5+) and B1b (CD5-) sub-populations B1 cell populations, unlike B2 cells, are not

maintained through de novo influx of B1 cells to replenish the population but

through their “self-renewal” capacity This ability to replenish the B1 population is restricted to CD11b-CD5+/- cell populations as demonstrated in

in vivo transplantation experiments (Ghosn et al., 2008) Unexpectedly, it was

reported that the spleen is required for the maintenance of B1a cell number in the peritoneal cavity (Wardemann et al., 2002) However, it remained unclear

if the spleen is required for the generation or survival of peritoneal B1a cells

In addition, the maintenance of B1 cells is also regulated by circulating IgM

antibodies as µs -/- (does not secrete IgM) mice have increased peritoneal B1 population (Boes et al., 1998)

The B1 population belongs to the innate arm of immunity (Dorshkind and Montecino-Rodriguez, 2007) This is due to the fact that they are the main source of polyreactive natural IgM antibodies in circulation of germ-free and antigen-free mice (Bos et al., 1989; Ehrenstein and Notley, 2010; Haury et al., 1997) However, whether peritoneal B1 cells spontaneously secrete IgM is

contentious as in vitro culture of these B1 cells do not yield high amount of

IgM antibodies assessed by ELISA and formed small pinhead spots assessed

by ELISpot (Choi et al., 2012; Holodick et al., 2010; Kawahara et al., 2003) Recent studies demonstrated that these peritoneal cavity B1 cells migrate to the spleen upon activation and subsequently lose CD11b expression to become natural IgM-producing cells (Kawahara et al., 2003; Yang et al., 2007) The migration is primarily achieved through toll-like receptor (TLR) signaling to down-regulate integrin expression such as α4 and β1 integrin, for the detachment of peritoneal B1 cells to migrate along chemokine gradient and gain entry into the spleen (Ha et al., 2006)

Emerging views propose that B1a and B1b cell populations are functionally distinct from each other (Dorshkind and Montecino-Rodriguez, 2007) This functional differences was demonstrated using overexpression of CD19 in hCD19Tg mice, which have increased B1a population but severely reduced B1b population, and using CD19-/- mice which have B1b population but severely reduced B1a population (Haas et al., 2005) In this study, CD19-/-

mice were more susceptible to infection with S pneumoniae than hCD19Tg

and WT mice because of the lack of natural IgM antibodies from B1a cells However, CD19-/- mice, but not hCD19Tg mice, were provided with long-

term protection from infection with S pneumonia after immunization with

pneumococcal polysaccharide type 3 (PPS-3) suggesting B1b population provides adaptive PPS-specific antibody responses

Table 1 Distribution and function of B1 cells in steady state Adapted from

Baumgarth 2011

1.3.2 B2 cells

B2 cell population comprises of follicular B cells and marginal zone B cells, and they arise from bone marrow precursors After successful negative selection process in the bone marrow, most immature naive B cells (B220+IgM+) would migrate to the spleen, passing through transitional T1 and T2 stages before completing their development into circulating follicular B cells (CD23+CD21midIgDhi) and non-circulating marginal zone B cells (CD23-CD21hiIgDloIgMhi) (Pillai and Cariappa, 2009)

in a type I IFN-mediated manner (Swanson et al., 2010) In addition, it was observed that circulating B cells occupy a distinct perisinusoidal niche in the bone marrow and participated in TI immune responses against blood borne pathogens (Cariappa et al., 2005) Thus, follicular B cells generally contribute

to TD responses but not absolutely

1.3.2.2 Marginal Zone B cells

Marginal zone B cells are localized between the white pulp and red pulp region in the spleen, next to the marginal sinus where arterial blood enters into the spleen (Kraal, 1992) With a lower threshold of activation as compared to follicular B cells, marginal zone B cells are well positioned for early screening or encounter with blood borne pathogens and migrate to the bridging channel of the follicle to undergo proliferative burst and concomitantly, differentiate into plasmablasts to provide the initial wave of antibodies against invading pathogens (Martin and Kearney, 2002; Oliver et al., 1997) In addition, marginal zone B cells express high level of CD1d molecules, which allows them to present lipid antigens to invariant natural killer T cells (iNKT) to augment antibody responses through provision of cognate help (Barral et al., 2008; Leadbetter et al., 2008) Besides their involvement in TI antibody response, they play an intermediary role of shuttling into follicles, transporting captured antigen to FDCs in a CXCR5-dependent manner (Cinamon et al., 2008)

1.4 Humoral Immunity in Atherosclerosis

With respect to the formation of the atherosclerotic plaque and its progression, no detectable differences in lesion area were observed between

1997) This observation suggests that the both T- and B lymphocytes do not play a major role in plaque formation However, this observation does not necessarily preclude the absence of involvement of T- and B lymphocytes in atherosclerosis Of interest, B cells and their involvement in atherosclerosis will be discussed

1.4.1 Localization of B cells in Atherosclerosis

In apoE -/- mice, it was demonstrated that the localization of CD22+ B cells is mainly at the base of the plaque of both early fatty streak lesions and mature atherosclerotic plaques (Zhou and Hansson, 1999) The basal localization of CD22+ B cells corresponded to VCAM-1 expression, which was reported to be that from SMCs (Li et al., 1993) In addition, B cells have

also been reported to localize in the adventitia of old apoE -/- mice in the form

of T/B clusters throughout the arterial tree and tertiary lymphoid structures containing proliferating Ki-67+ B cells and CD138+ plasma cells in the abdominal region of aorta (Moos et al., 2005) These tertiary lymphoid structures formation was found to be dependent on the lymphotoxin beta receptor (LTβ-R) activation of medial smooth muscle cells (Grabner et al., 2009) Thus, both the presence of B cells in the intima and tertiary lymphoid structures in the adventitia, are indicative of local adaptive immune responses associated with atherogenesis (Neyt et al., 2012)

Similarly, in human atherosclerotic aorta tissues, CD19+ B cells were mainly found in the adventitia (Koch et al., 1990; Ramshaw and Parums, 1990) Tertiary lymphoid structures were also observed in minority of samples examined but they were associated with more advanced stage of the disease (Watanabe et al., 2007) To support that the development of tertiary lymphoid structures is considered a rare event, in biopsies samples of atherosclerotic aortic aneurysms in another study, lymphoid follicle structures were also observed for the minority (Walton et al., 1997) However, CD138+ cells were observed in adventitia of all samples and almost all samples showed an unrestricted usage of variable heavy chain gene usage suggesting that B cell responses were not resulting from limited repertoire of antigens (Walton et al., 1997)

Though, as described above, B cells are localized to atherosclerotic

lesion sites – in both plaque and adventitia for apoE -/- mice and adventitia for human, humoral immunity has been reported in the spleen (Caligiuri et al.,

2002) In that study, compared to sham-operated, splenectomized apoE -/- mice exhibited aggravated lesion size area Adoptive transfer of lymphocytes into

splenectomized apoE -/- recipients demonstrated that the protective effect was

due to B cells but not the T cells from apoE -/- mice In addition, B cells from

aged but not young apoE -/- mice mediated this effect In another study, an increase of 30-40% in lesion area in both proximal and distal aorta was

observed in bone marrow chimera ldlr -/- mice lacking B cells, revealing the involvement, or rather, the protective role of B cells (Major et al., 2002)

Therefore, the humoral immunity provided by B cells is not restricted

to local inflammatory sites of atherosclerotic lesions – at basal layer of plaque

or adventitia as tertiary lymphoid structures, but it is also provided by spleen

in a systematic manner

1.4.2 Association of oxLDL-specific antibodies with Atherosclerosis

Lipids such as the complex structure LDL becomes oxidized due to the action of ROS and in turn becomes immunogenic In essence, oxLDL is the product of oxidation of polyunsaturated fatty acid component where reactive lipid peroxidation products such as malondialdehyde (MDA) and 4-hydroxynonenal (4-HNE) form adducts on lysine residues on apoB100 protein resulting the formation of neo-epitopes (Hansson and Hermansson, 2011; Itabe et al., 2011) As a result, even native LDL-specific antibodies could also

be detected as these antibodies bind to the overlapping region of native LDL and lipid adducts A common way to prepare oxLDL for experimental studies

is to incubate native LDL with copper sulfate where copper ion (Cu2+) would initiate lipid peroxidation process that results in MDA-lysine and 4-HNE-lysine adducts on apoB100 (Palinski et al., 1990) Another commonly used method of preparation is the use of malonaldehyde to oxidatively modify LDL (MDA-LDL) Therefore, it is important to point out oxLDL is not a defined molecular species but an “umbrella” term to include a spectrum of LDL that underwent various physicochemical changes (Hansson and Hermansson, 2011)

Increased titers of oxLDL-specific autoantibodies are positively

correlated with the progression of disease It was reported that apoE -/- mice had increased autoantibodies against MDA-LDL and these autoantibodies were shown to be specific to atherosclerotic lesion when competitively inhibited with excess MDA-LDL (Palinski et al., 1994) In addition, titers of

MDA-LDL specific autoantibodies were found to be elevated in ldlr -/- mice placed on high fat diet compared to chow diet (Palinski et al., 1995) The increased MDA-LDL specific autoantibodies significantly correlated positively with elevated plasma cholesterol and also with the extent of atherosclerotic lesions (Palinski et al., 1995)

Nonetheless, contradictory studies have also emerged and showed that oxLDL-specific autoantibodies can negatively correlate with atherosclerotic lesions, mainly demonstrated in immunization studies Active immunization in

apoE -/- mice with MDA-LDL demonstrated a significant reduction in atherosclerotic lesion that was accompanied by elevated MDA-LDL specific IgG autoantibodies (George et al., 1998; Zhou et al., 2001) Immunization with phosphorylcholine (PC), the main epitope in oxidized phospholipids on

LDL molecule, also yielded the same effect on atherosclerotic lesion in apoE

-/-mice with increased PC- and oxLDL specific IgG /IgM being observed (Caligiuri et al., 2007) However, it is interesting to note that adjuvant frequently used in active immunization by themselves have atheromodulating effects (Khallou-Laschet et al., 2006) For example, mineral oil in both complete and incomplete Freund’s adjuvant (CFA and IFA, respectively) could increase the titer of MDA-LDL specific IgM responses that is associated with significant reduction in lesion size The same effect was found using

alum, but not CpG, as adjuvant This atheromodulating effect is not restricted

to apoE -/- mice but also observed in ldlr -/- mice (Binder et al., 2003) Active immunization without use of adjuvant, for example, through intravenous (i.v.) injection of oxLDL-pulsed dendritic cells (DCs) attenuated atherosclerotic lesion size with increased Cu2+ oxLDL, but not MDA-LDL specific IgG

responses detected in ldlr -/- mice (Habets et al., 2009)

Passive immunization also attenuates atherosclerosis Intravenous

immunoglobulin (IVIg) treatment on apoE -/- mice reduced lesion area at aortic root and arch regions (Nicoletti et al., 1998) This effect was sustained only when IVIg treatment was given once every 2 months in a 4 months experiment

or even administered only in the middle of the study suggesting that the presence of IVIg was required to inhibit aggravating factors in the disease Similarly, another study found that late administration of IVIg also conferred

protection in apoE -/- mice and the effect was lost with administration of Fab fragments suggesting the requirement for an intact antibody with the Fc region (Okabe et al., 2005) The requirement for Fc portion in mediating protective effects was also observed in earlier studies (Yuan et al., 2003) Further studies demonstrated that the protective effect of IVIg was complement dependent as

IVIg-treated apoE -/- ldlr -/- mice showed significant decrease in lesion area

compared to C3 -/- apoE -/- ldlr -/- mice (Persson et al., 2005) The protective effect was less clear when passive immunization of PC-specific IgM antibodies was

administered because decreased vein graft atherosclerosis in apoE -/- mice but

not in established native atherosclerotic apoE -/- mice was observed Neto et al., 2006) The authors, however, speculated that 4 weeks treatment