Download AGE-RELATED MACULAR DEGENERATION

 AGE‐RELATED MACULAR DEGENERATION: HISTOPATHOLOGICAL AND SERUM AUTOANTIBODY STUDIES Svetlana Cherepanoff Submitted to the Faculty of Medicine in fulfillment of the requirements of for the degree Doctor of Philosophy (Medicine) Department of Clinical Ophthalmology & Eye Health University of Sydney AUGUST 2007 For Rinjani and Ambika A C K N O W L E D G E M E N T S First and foremost, I would like to thank Soren Soengkoeng, my husband, for his endless patience, unwavering support, and willingness to do far more than his fair share of child rearing and housekeeping during the course of my candidature. Without him this thesis would not have been possible. To my parents, Mei Lan, Irina and Leonid, who have always encouraged me to pursue my interests, and who arrive each week to look after the girls and perform endless chores without question – I thank each of you for your kindness, generosity and tireless support. I thank Mark Gillies, my supervisor, for his enthusiastic support of all ideas and potential projects, for introducing me to ophthalmology research and for graciously putting up with a “firebrand” student. Thanks must go especially to Shirley Sarks, who has been an extraordinary teacher and mentor, and who has utterly changed the way I understand this disease. I thank both Shirley and John Sarks for their generosity in sharing their tissue resources, their unique insights and their unrivalled expertise. Finally, a big thank you to all the people who enable the smooth running of both Save Sight Institute and the Prince of Wales Medical Research Institute – without you the experimental and written work could not have been produced. The National Health and Medical Research Council (Canberra, ACT Australia) provided financial support in the form of a PhD scholarship. i
D E C L A R A T I O N I hereby declare that this thesis is my own work, except where due acknowledgement is specified below. To the best of my knowledge, the thesis contains no material previously published or written by another person, nor material which to a substantial extent has been accepted for the award of any other degree or diploma of a university or other institute of higher learning. All experiments described in this thesis were performed at the Save Sight Institute laboratories (University of Sydney, Sydney Eye Hospital Campus, Sydney NSW Australia) and at the Prince of Wales Medical Research Institute laboratories (Randwick, NSW Australia), between 2001 and 2006. The histopathological observations in Chapter 3 were made by Dr Shirley Sarks (Prince of Wales Medical Research Institute, Randwick NSW Australia), and the electron micrographs in the same chapter were produced by A/Prof Murray Killingsworth (South Western Area Pathology Services, Liverpool NSW Australia). Dr Shirley Sarks produced the line drawings in Figure 2.2‐1 and Figure 3.1‐5. Ms Emma Kettle (Children’s Medical Research Institute, Westmead, NSW Australia) provided assistance with the immunohistochemical studies. Dr Karen Byth Wilson (NHMRC Clinical Trials Centre, University of Sydney, NSW Australia) performed the linear mixed effects modelling in Chapter 4. ii
Svetlana Cherepanoff August, 2007 A B S T R A C T B A C K G R O U N D : The accumulation of abnormal extracellular deposits beneath the retinal pigment epithelium characterises the pathology of early age‐related macular degeneration. However, the histopathological threshold at which age‐
related changes become early AMD is not defined, and the effect of each of the deposits (basal laminar deposit and membranous debris) on disease progression is poorly understood. Evidence suggests that macrophages play a key role in the development of AMD lesions, but the influence of basal laminar deposit (BLamD) and membranous debris on the recruitment and programming of local macrophages has not been explored. Although evidence also suggests that inflammation and innate immunity are involved in AMD, the significance of anti‐
retinal autoantibodies to disesase pathogenesis is not known. A I M S : (i) To determine the histopathological threshold that distinguishes normal ageing from early AMD; (ii) to determine the influence of BLamD and membranous debris on disease progression; (iii) to examine whether distinct early AMD phenotypes exist based on clinicopathological evidence; (iv) to determine the histopathological context in which Bruch’s membrane macrophages first found; (v) to examine the relationship between Bruch’s membrane macrophages and subclinical neovascularisation; (vi) to determine if the progressive accumulation of BLamD and membranous debris alters the immunophenotype of Bruch’s membrane macrophages and/or resident choroidal macrophages; (vii) to determine if the anti‐retinal autoantibody profile differs significantly between normal individuals and those with early AMD, neovascular AMD or geographic atrophy; (viii) to examine whether baseline anti‐retinal autoantibodies can predict progression to advanced AMD in individuals with early AMD; and (ix) to examine iii
whether baseline anti‐retinal autoantibodies can predict vision loss in individuals with neovascular AMD. M E T H O D S : Clinicopathological studies were performed to correlate progressive accumulation of BLamD and membranous debris to fundus characteristics and visual acuity, as well as to sub‐macular Bruch’s membrane macrophage count. Immunohistochemical studies were perfomed to determine whether the presence of BLamD and membranous debris altered the programming of Bruch’s membrane or resident choroidal macrophages. The presence of serum anti‐retinal autoantibodies was determined by western blotting, and the association with disease progression examined in early and neovascular AMD. R E S U L T S : The presence of both basal linear deposit (BLinD) and a continuous layer of BLamD represents threshold early AMD histopathologically, which was seen clinically as a normal fundus in the majority of cases. Membranous debris accumulation appeared to influence the pathway of progression from early AMD to advanced AMD. Bruch’s membrane macrophages were first noted when a continuous layer of BLamD and clinical evidence of early AMD were present, and increased with the amount of membranous debris in eyes with thin BLamD. Eyes with subclinical CNV had high macrophage counts and there was some evidence of altered resident choroidal macrophage programming in the presence of BLamD and membranous debris. Serum anti‐retinal autoantibodies were found in a higher proportion of early AMD participants compared with both controls and participants with neovascular AMD, and in a higher proportion of individuals with atrophic AMD compared to those with neovascular AMD. The presence of baseline anti‐retinal autoantibodies in participants with early AMD was not associated with progression to advanced AMD. Participants with neovascular AMD lost more vision over 24 months if they had IgG autoantibodies at baseline compared to autoantibody negative participants. iv
C O N C L U S I O N S : The finding that eyes with threshold early AMD appear clinically normal underscores the need to utilise more sophisticated tests to enable earlier disease detection. Clinicopathological evidence suggests two distinct early AMD phenotypes, which follow two pathways of AMD progression. Macrophage recruitment and programming may be altered by the presence of BLamD and membranous debris, highlighting the need to further characterise the biology of human resident choroidal macropahges. Anti‐retinal autoantibodies can be found in both control and AMD sera, and future approaches that allow the examination of subtle changes in complex repertoires will determine whether they are involved in AMD disease pathogenesis. v
P U B L I C A T I O N S Original Articles Cherepanoff S, Mitchell P, Wang JJ, and Gillies MC. Retinal autoantibody profile in early age‐related macular degeneration: Preliminary findings from the Blue Mountains Eye Study. Clin Exp Ophthalmol, 2006. 34(6): p. 590‐5. Cherepanoff, S and Gillies, MC, Circulating antiretinal autoantibodies and age‐
related macular degeneration: What is the link? Exp Rev Ophthalmol, 2007. 2(1): p. 27‐31. Sarks S, Cherepanoff S, Killingsworth M, and Sarks J. Relationship of basal laminar deposit and membranous debris to the clinical presentation of early age‐
related macular degeneration. Invest Ophthalmol Vis Sci, 2007. 48(3): p. 968‐77. Published Abstracts Cherepanoff S, Sarks J, Killingsworth M, and Sarks S. The Life Cycle of Soft Drusen. Invest Ophthalmol Vis Sci. 2007 48: E‐Abstract 3026. Cherepanoff S, Gillies M, Luo W, Chua W, Penfold PL. Anti‐Retinal Auto‐
Antibodies in Neovascular Age Related Macular Degeneration ‐ A Useful Prognostic Marker? Invest Ophthalmol Vis Sci. 2001; 42(4): S801. Abstract 4294 Gillies MC, Luo W, Chua W, Cherepanoff S, Billson FA, Mitchell P, Simpson J. The Efficacy of a Single Injection of Triamcinolone for Neovascular Age Related Macular Degeneration. One Year Results of a Randomised Clinical Trial: IVTAS. Invest Ophthalmol Vis Sci. 2001; 42(4): S522. Abstract 2808 Jordan B, Cvejic S, Trapaidze N, Cherepanoff S, Unterwald E, Devi LA. Agonist mediated internalization of the μ, κ and δ opioid receptor types. Society of Neuroscience Abstracts 1997; 2: 691.6. vi
T A B L E O F C O N T E N T S ACKNOWLEDGEMENTS............................................................................................................................ i DECLARATION ............................................................................................................................................ii ABSTRACT................................................................................................................................................... iii PUBLICATIONS...........................................................................................................................................vi TABLE OF CONTENTS .............................................................................................................................vii LIST OF FIGURES ......................................................................................................................................xii LIST OF TABLES....................................................................................................................................... xiii LIST OF TABLES....................................................................................................................................... xiii ABBREVIATIONS ..................................................................................................................................... xiv ABBREVIATIONS ..................................................................................................................................... xiv 1 INTRODUCTION ..................................................................................................................................... 1 2 BACKGROUND ........................................................................................................................................ 2 2.1 AGE‐RELATED MACULAR DEGENERATION: CLINICAL DEFINITIONS AND EPIDEMIOLOGY ........... 3 2.1.1 Clinical definition...................................................................................................................... 3 2.1.2 Disease classification ................................................................................................................. 5 2.1.3 Prevalence and incidence........................................................................................................... 6 2.1.4 Natural history .......................................................................................................................... 8 2.1.5 Risk factors .............................................................................................................................. 10 2.1.5.1 Smoking....................................................................................................................................... 10 2.1.5.2 Cardiovascular disease and cardiovascular risk factors ....................................................... 11 2.1.5.3 Inflammatory markers............................................................................................................... 12 2.1.5.4 Dietary factors............................................................................................................................. 13 2.1.5.5 Genetic factors ............................................................................................................................ 13 2.1.6 Estimated costs and disease burden......................................................................................... 16 2.1.7 Therapeutics ............................................................................................................................ 17 2.1.7.1 Preventative interventions ........................................................................................................ 17 2.1.7.2 Palliative interventions.............................................................................................................. 18 vii
2.2 THE “ANATOMY” OF AMD ......................................................................................................... 19 2.2.1 The macula in health and ageing ............................................................................................. 19 2.2.1.1 Anatomy & embryology of the human eye ............................................................................ 19 2.2.1.2 The retinal pigment epithelium................................................................................................ 21 2.2.1.3 Age‐related changes in the RPE ............................................................................................... 22 2.2.1.4 The choroid and Bruch’s membrane........................................................................................ 23 2.2.1.5 Age‐related changes in the choroid ......................................................................................... 25 2.2.1.6 Age‐related changes in Bruch’s membrane ............................................................................ 25 2.2.1.7 Immunological specialisations of the retina and choroid ..................................................... 28 2.2.2 The pathology of endstage AMD ............................................................................................. 29 2.2.2.1 Geographic atrophy of the RPE: the natural endstage of AMD........................................... 29 2.2.2.2 Choroidal neovascularisation and disciform scarring........................................................... 30 2.2.3 The basal deposits and early AMD.......................................................................................... 31 2.2.3.1 Basal laminar deposit is a marker of RPE degeneration ....................................................... 31 2.2.3.2 Basal linear deposit and soft drusen are the specific lesions of early AMD ...................... 32 2.2.3.3 Drusen classification .................................................................................................................. 35 2.2.3.4 Drusen biogensis ........................................................................................................................ 36 2.2.3.5 Histopathological classification schemes ................................................................................ 38 2.3 AMD DISEASE PATHOGENESIS: SUMMARY OF CURRENT CONCEPTS .......................................... 41 2.3.1 Oxidative stress ....................................................................................................................... 41 2.3.2 Ischaemia ................................................................................................................................. 43 2.3.3 Inflammation ........................................................................................................................... 43 2.4 MACROPHAGES AND IMMUNE COMPETENT CELLS IN AMD...................................................... 44 2.4.1 Morphological evidence ........................................................................................................... 44 2.4.1.1 Macrophages, Bruch’s membrane breaks and early AMD ................................................... 44 2.4.1.2 Macrophages in advanced AMD lesions................................................................................. 44 2.4.2 Macrophages and choroidal neovascularisation ...................................................................... 45 2.4.3 Choroidal macrophage recruitment and turnover and AMD.................................................. 47 2.4.3.1 Ccl‐2, Ccr‐2 and Cx3Cr1 knockout mice and AMD‐like lesions .......................................... 47 2.4.3.2 IL‐10 knockout mice and laser‐induced CNV ........................................................................ 49 2.4.3.3 Polymorphisms in leukocyte recruitment and turnover genes............................................ 49 2.4.3.4 Serum myeoloid cells and AMD .............................................................................................. 50 2.4.3.5 Macrophages in an immune‐privileged compartment.......................................................... 50 2.4.4 2.5 Dendritic cells and drusen biogenesis ..................................................................................... 52 INFLAMMATION AND AMD........................................................................................................ 54 2.5.1 Tissue evidence of inflammatory proteins ............................................................................... 54 2.5.1.1 Evidence of inflammation‐associated proteins in drusen..................................................... 54 viii
2.5.2 2.5.1.1.1 Immunoglobulins and complement in drusen and Bruch’s membrane ........................ 55 2.5.1.1.2 Amyloid, fibrinogen and other inflammatory proteins found in drusen...................... 57 Genetic associations................................................................................................................. 59 2.5.2.1 Single nucleotide polymorphisms in complement cascade and complement regulatory genes 59 2.5.3 2.6 Serum inflammatory markers in AMD................................................................................... 61 SERUM ANTI‐RETINAL AUTOANTIBODIES IN AMD..................................................................... 63 2.6.1 Autoantibodies and autoimmunity.......................................................................................... 63 2.6.2 Autoantibodies and ocular disease........................................................................................... 64 2.6.2.1 Anti‐recoverin autoantibodies, CAR and autoimmune retinopathy................................... 64 2.6.2.2 Retinal degenerations, autoimmune retinopathy and glaucoma......................................... 66 2.6.3 2.7 Autoantibodies in AMD.......................................................................................................... 66 THESIS AIMS ................................................................................................................................. 74 3 HISTOPATHOLOGICAL STUDIES ................................................................................................... 76 3.1 RELATIONSHIP OF BASAL LAMINAR DEPOSIT AND MEMBRANOUS DEBRIS TO THE CLINICAL PRESENTATION OF EARLY AGE‐RELATED MACULAR DEGENERATION ....................................................... 77 3.1.1 Introduction............................................................................................................................. 77 3.1.2 Methods ................................................................................................................................... 78 3.1.2.1 Patients and Eyes........................................................................................................................ 78 3.1.2.2 Histopathological Methods and Definitions........................................................................... 78 3.1.2.3 Clinical Parameters .................................................................................................................... 81 3.1.2.4 Statistical Methods ..................................................................................................................... 81 3.1.3 Results ..................................................................................................................................... 82 3.1.3.1 Basal laminar deposit and RPE changes.................................................................................. 82 3.1.3.2 BLamD – clinical correlations ................................................................................................... 91 3.1.3.3 Membranous debris‐ localization and correlation with BLamD.......................................... 91 3.1.3.4 Membranous debris – clinical correlations ............................................................................. 92 3.1.4 3.2 Discussion ............................................................................................................................... 98 CHOROIDAL AND BRUCH’S MEMBRANE MACROPHAGES IN EARLY AND ADVANCED AMD... 105 3.2.1 Introduction........................................................................................................................... 105 3.2.2 Methods ................................................................................................................................. 108 3.2.2.1 Eyes ............................................................................................................................................ 108 3.2.2.2 Clinical definitions ................................................................................................................... 108 3.2.2.3 Histopathological definitions and grading: BLamD............................................................ 109 3.2.2.4 Histopathological definitions and grading: membranous debris ...................................... 109 3.2.2.5 BrM macrophage counts ......................................................................................................... 109 3.2.2.6 Immunohistochemistry ........................................................................................................... 110 ix
3.2.2.7 3.2.3 Statistics ..................................................................................................................................... 111 Results ................................................................................................................................... 112 3.2.3.1 Groups I & II‐ normal and normal aged eyes....................................................................... 112 3.2.3.2 Groups III & IV– early AMD .................................................................................................. 112 3.2.3.3 Groups III & IV: BrM macrophages and subclinical CNV.................................................. 120 3.2.3.4 Group V – Geographic atrophy .............................................................................................. 125 3.2.3.5 Group VI – Disciform scarring ............................................................................................... 125 3.2.4 Discussion ............................................................................................................................. 128 4 SERUM AUTOANTIBODY STUDIES.............................................................................................. 131 4.1 COMPARISON OF THE ANTI‐RETINAL AUTOANTIBODY PROFILE IN EARLY, NEOVASCULAR & ATROPHIC AMD ......................................................................................................................................
132 4.1.1 Introduction........................................................................................................................... 132 4.1.2 Methods ................................................................................................................................. 133 4.1.2.1 Study serum .............................................................................................................................. 133 4.1.2.1.1 Blue Mountains Eye Study................................................................................................. 133 4.1.2.1.2 Intravitreal Triamcinolone Study...................................................................................... 134 4.1.2.1.3 Serum used in autoantibody assays ................................................................................. 135 4.1.2.2 Autoantibody assay ................................................................................................................. 135 4.1.2.2.1 Preparation of human retinal proteins for electrophoresis ........................................... 135 4.1.2.2.2 Electrophoresis and western blotting............................................................................... 136 4.1.2.2.3 Detection of serum anti‐retinal autoantibodies............................................................... 136 4.1.2.2.4 Antibodies used to detect serum autoantibodies............................................................ 137 4.1.2.3 Statistics ..................................................................................................................................... 138 4.1.3 Results ................................................................................................................................... 139 4.1.4 Discussion ............................................................................................................................. 149 4.2 ANTI‐RETINAL AUTOANTIBODIES AND DISEASE PROGRESSION IN EARLY AMD ..................... 152 4.2.1 Introduction........................................................................................................................... 152 4.2.2 Methods ................................................................................................................................. 152 4.2.2.1 Participants ............................................................................................................................... 152 4.2.2.2 Autoantibody assay ................................................................................................................. 152 4.2.2.3 Statistics ..................................................................................................................................... 153 4.2.3 Results ................................................................................................................................... 154 4.2.4 Discussion ............................................................................................................................. 156 x
4.3 ANTI‐RETINAL AUTOANTIBODIES AND VISION LOSS IN NEOVASCULAR AMD ........................ 157 4.3.1 Introduction........................................................................................................................... 157 4.3.2 Methods ................................................................................................................................. 158 4.3.2.1 Participants ............................................................................................................................... 158 4.3.2.2 Autoantibody assay ................................................................................................................. 158 4.3.2.3 Statistics ..................................................................................................................................... 158 4.3.3 Results ................................................................................................................................... 160 4.3.4 Discussion ............................................................................................................................. 164 5 SUMMARY & CONCLUSIONS......................................................................................................... 165 REFERENCES............................................................................................................................................. 168 APPENDICES............................................................................................................................................A‐H xi
L I S T O F F I G U R E S Figure 2.2‐1 Schematic diagram of BLamD and membranous debris ...................33 Figure 2.5‐1 The complement cascade ........................................................................56 Figure 3.1‐1 Basal Laminar Deposit ............................................................................84 Figure 3.1‐2 Continuous Late Basal Laminar Deposit..............................................86 Figure 3.1‐3 High magnification membranous debris .............................................88 Figure 3.1‐4 Basal mounds ..........................................................................................89 Figure 3.1‐5 Membranous debris: basal mounds and soft drusen .........................94 Figure 3.1‐6 Influence of basal deposits on the progression of AMD.................102 Figure 3.2‐1 Morphological and immunohistochemical features of normal (group I) and normal aged (group II) eyes ................................................................116 Figure 3.2‐2 Morphological and immunohistochemical features of eyes with early AMD (groups III and IV) .................................................................................118 Figure 3.2‐3 Morphological features of eyes with subclinical CNV.....................121 Figure 3.2‐4 CD68 and iNOS Immunohistochemistry in eyes with geographic atrophy (group V) or disciform scarring (group VI) ...................................126 Figure 4.1‐1 Western blot‐detected anti‐retinal autoantibodies in normal controls and participants with geographic atrophy ...................................................141 Figure 4.1‐2 Western blot‐detected anti‐retinal autoantibodies in participants with early AMD .........................................................................................................143 Figure 4.1‐3 Western blot‐detected anti‐retinal autoantibodies in participants with neovascular AMD.............................................................................................145 Figure 4.3‐1 Vision loss over 24 months in anti‐retinal autoantibody positive participants vs. anti‐retinal autoantibody negative participants (anti‐
IgGAM detected autoantibodies) ...................................................................162 Figure 4.3‐2 Vision loss over 24 months in anti‐retinal autoantibody positive participants vs. anti‐retinal autoantibody negative participants (anti‐IgG detected autoantibodies) .................................................................................163 xii
L I S T O F T A B L E S Table 2.2‐1 Histopathological grading scheme suggested by Sarks (1976)...........40 Table 2.2‐2 Histopathological grading scheme suggested by Curcio et al (1998) 40 Table 2.6‐1 Autoantibodies in AMD: summary of studies .......................................70 Table 3.1‐1 Clinical and histopathological characteristics of study eyes................83 Table 3.1‐2 Basal laminar deposit – clinical and histopathological correlations .96 Table 3.1‐3 Membranous debris – clinical and histopathological correlations.....97 Table 3.1‐4 Basal laminar deposit ‐ a summary......................................................103 Table 3.1‐5 Membranous debris – a summary ........................................................104 Table 3.2‐1 Clinical and histopathological features of study eyes........................114 Table 3.2‐2 BLamD, membranous debris and macrophages in early AMD .......115 Table 3.2‐3 BrM macrophages, membranous debris and subclinical CNV.........123 Table 3.2‐4 BrM macrophages and subclinical CNV in the fellow eye................124 Table 4.1‐1 Serum anti‐retinal autoantibodies in controls and in early, neovascular and atrophic AMD: summary of findings ....................................................140 Table 4.1‐2 Comparison of Anti‐IgGAM and anti‐IgG detected autoantibodies in control, early AMD, neovascular AMD and atrophic AMD participants 148 Table 4.2‐1 Baseline serum anti‐retinal autoantibodies and progression to advanced AMD at 5 and 10 years...................................................................155 Table 4.3‐1 Participant characteristics and baseline anti‐retinal autoantibody status...................................................................................................................161 xiii
A B B R E V I A T I O N S AMD age‐related macular degeneration APC antigen presenting cell ApoE apolipoprotein E BF factor B BLamD basal laminar deposit BLinD basal linear deposit CAR cancer‐associated retinopathy CC choriocapillaris CFH complement factor H CNV choroidal neovascularisation CRP C‐reactive protein FasL fas ligand GA geographic atrophy GFAP glial acidic fibrillary protein HLA human leukocyte antigen IL‐1 interleukin‐1 IL‐10 interleukin‐10 iNOS inducible nitric oxide synthase MAC membrane attack complex MCP‐1 macrophage chemoattractant protein‐1 MHC major histocompatibility complex MMP membrane metalloproteinases PDT photodynamic therapy PEDF pigment epithelium‐derived factor ROI reactive oxygen intermediate RPE retinal pigment epithelium TGFβ
transforming growth factor beta TIMP tissue inhibitors of metalloproteinases TLR‐4 toll like receptor‐4 VEGF vascular endothelial growth factor xiv
1
I N T R O D U C T I O N The macula is a part of the retina found at the posterior pole of the eye. It has the highest concentration of cone photoreceptors, and is responsible for high acuity colour vision. In age‐related macular degeneration (AMD), loss of the retinal pigment epithelium and the photoreceptors they support lead to loss of central vision, with devastating consequences for the affected individual. AMD is the leading cause of irreversible vision loss in industrialised populations, with the numbers of cases expected to grow due to population ageing. 1
2
B A C K G R O U N D 2
2.1 AGE-RELATED
MACULAR DEGENERATION: CLINICAL
DEFINITIONS AND EPIDEMIOLOGY
2.1.1 Clinical definition Drusen, atrophy of the retinal pigment epithelium (RPE) and disciform lesions were recognised clinical entities by the late 19th century. Donders described the clinical appearance of drusen in 1855 (reviewed in 1), while “symmetrical central chorio‐retinal disease occurring in senile persons” was described by Hutchinson and Tay in 1874 (reviewed in 2). In 1885, Haab described pigmentary and atrophic changes in the macular of patients over the age of 50, which was associated with central vision loss (reviewed in 3). However, Gass was first to suggest that drusen, senile macular degeneration and senile disciform degeneration were manifestations of a single disease entity 4. This entity has been variously named “senile macular degeneration”, “disciform degeneration” and “maculopathy”, but is currently known as “age‐related macular degeneration” (AMD). Clinically, AMD can be classified as early or advanced. The advanced disease is further categorised as neovascular (“wet” or “exudative”) or atrophic (“geographic atrophy”). The term “dry” AMD has been used by some investigators to describe both early AMD and geographic atrophy, i.e. the non‐
exudative forms of AMD. Geographic atrophy (GA) of the RPE is considered the natural endstage of the disease when choriodal neovascularisation (CNV) does not intervene 5. GA appears as discreet areas of pallor, due to loss of RPE and reduced retinal thickness. It commonly begins in the parafoveal region, and the foveal centre is often spared until late in the disease process 6‐8. Visual acuity may thus be preserved, although scotomas surrounding the point of fixation can be significant 9
. GA may follow pigmentary change and attenuation or may occur subsequent to drusen regression. GA may also follow involution of CNV, and CNV can develop 3
in an eye with GA 10. Sunness et al. found that the 4‐year cumulative incidence of CNV was 11% for eyes with GA, and 34% for eyes with GA if the fellow eye had CNV 11. Neovascular AMD results from the growth of new vessels from the choroid into the sub‐RPE or sub‐retinal space. These vessels are abnormally permeable, leading to the leakage of fluid or blood into the sub‐RPE (“occult” CNV) or sub‐retinal (“classic” CNV) space (reviewed in 12. Clinically, neovascular AMD can present as sub‐retinal fluid or haemorrhage, lipid exudates, pigment epithelial detachment or, in the later stages, as a fibrovascular “disciform” scar. Neovascular AMD is responsible for an estimated 75‐90% of the severe vision loss attributable to AMD 13
. Early AMD is defined as the presence of soft (or large) macular drusen with or without pigmentary changes. Soft drusen appear as ill‐defined yellow deposits while pigmentary abnormalities can include hypo‐ or hyperpigmentation. There is less agreement on whether pigmentary changes without drusen should be considered early AMD (see Section 2.1.4). Patients with early AMD often have good vision. There may be subtle deficits, such as reduced dark adaptation, which are revealed by psychophysical or electrophysiological testing. There is, thus, a wide range of clinical presentations of AMD, even within families 14
. Various definitions (of early AMD in particular) can lead to different prevalence estimates in population‐based studies. This “phenotypic heterogeneity” has lead to the development of a number of systems of disease classification. 4
2.1.2 Disease classification The Wisconsin Age‐Related Maculopathy Grading Scheme was developed to quantify retinal changes in the Beaver Dam Study, one of the largest and earliest population‐based studies of age‐related eye disease 15. This system uses grids with a radius of 3000μm, centred on the fovea, to define subfields in the macula of fundus photographs. Templates are then used to estimate drusen size and the area involved by drusen or RPE abnormalities. Drusen are graded on each of three criteria: size, type and area of macula involved. RPE changes are graded according to type of abnormality and area of macula involved. Early AMD was defined as: (i) the presence of hard or soft drusen with pigmentary changes; or (ii) the presence of soft, indistinct drusen only, in the absence of GA or exudative lesions. Despite achieving high agreement between trained graders, the Wisconsin system was not universally adopted. In the hope of creating common disease definitions that would allow comparisons between studies, the International Age‐Related Maculopathy Study Group brought together a number of leading investigators to develop the International Classification System 16. This system incorporated the Wisconsin Grading Scheme with other commonly used grading schemes. Early disease was termed “age‐
related maculopathy” (ARM), while advanced disease was termed “age‐related macular degeneration” (AMD). ARM was defined as: (i) the presence of drusen, “discrete whitish‐yellow spots” external to the neuroretina or RPE; or (ii) areas of hyperpigmentation associated with drusen; or (iii) areas of hypopigmentation associated with drusen. “Dry” AMD, or geographic atrophy, was defined as sharply demarcated areas of hypopigmentation, at least 175μm in diameter, in which the choroidal vessels are more visible than surrounding areas. “Wet” AMD, or “neovascular”, “disciform” or “exudative” AMD, was defined as: (i) RPE detachments, which may be associated with neurosensory retinal detachment; or (ii) subretinal or sub‐RPE neovascular membranes; or (iii) epiretinal, intraretinal, 5
subretinal or sub‐RPE scar/glial tissue/fibrin‐like deposits; or (iv) subretinal haemorrhages not related to other retinal vascular disease; or (v) hard (lipid) exudates related to other ARM findings, and unrelated to other vascular diseases. Of the longer‐running, large population‐based studies so far conducted, the Wisconsin Grading Scheme is used by the Beaver Dam Study (US) 17, the Blue Mountains Eye Study (Australia) 18 and the Melbourne Visual Impairment Project (Australia) 19. The Vitamin E, Cataract and Age‐related Maculopathy (VACT) Study (Australia) 20 and the Rotterdam Study (Netherlands) 21 utilise the International Classification System. The Age‐Related Eye Diseases Study (US) uses its own classification system 22, in which fundus features are classified into four categories based on drusen type and number and, unlike the other studies, incorporates a visual acuity score. A unified classification system has yet to be formulated. This is partly because new insights continue to be generated by ongoing epidemiological and clinicopathological studies. The findings from recent genetic studies are also likely to influence classification of disease subgroups. 2.1.3 Prevalence and incidence AMD is the commonest cause of blindness in industrialised populations 23. In 2005, Taylor and colleagues 24 reviewed the findings of the Melbourne Visual Impairment Project (VIP) and the Blue Mountains Eye Study (BMES), involving a total of 8909 participants, and estimated that 24 200 Australians are blind due to AMD, making it by far the commonest cause of blindness in people aged 40 and over. AMD was also found to be the most common cause of untreatable low vision, affecting an estimated 48 300 individuals. When findings from the Beaver Dam Study, the BMES and the Rotterdam study were pooled, AMD (early and 6
advanced) was present in 0.2% of 55‐64 year olds, rising to 13% of those aged 85 or older 25. Geographic atrophy increased from 0.04% to 4.2% for these age groups, while neovascular AMD increased from 0.17% to 5.8%. In general, population‐
based studies have shown that AMD is rare before age 55 and becomes more common after age 75 26. The Age‐Related Eye Diseases Study estimated that AMD affected over 8 million people in the US, with the advanced form affecting 1.75 million 27. This number is projected to increase 50% by 2020. Racial differences in both AMD prevalence and incidence exist. The frequency of neovascular AMD was higher in the Rotterdam population than the Beaver Dam or the BMES populations 25. However, in all three studies, the incidence and prevalence of neovascular CNV was higher in whites when compared to blacks 26. Although the frequency of early AMD in blacks appears to be similar to whites, the risk of progression to neovascular AMD is higher in whites. The early form is more commonly found than the late form, with both types increasing in frequency with age (17. The 5‐year overall incidence of advanced AMD (geographic atrophy and neovascular AMD) was 0.4% in the Beaver Dam Study, 0.7% in the BMES, and 1.1% in the Rotterdam study 28. In the longest running population‐based study (Beaver Dam), the 15‐year cumulative incidence for early AMD was 14.3% for early AMD and 3.1% for advanced AMD 29. AMD frequently affects both eyes. In the BMES cohort, 80% of participants with AMD (early and advanced) had bilateral disease 30. Once neovascular AMD is present in one eye, involvement of the second eye occurs in 82% of cases within four years in an Icelandic population 31. 7
2.1.4 Natural history Gass noted that drusen size, area and detectability altered over time, and that increased drusen area increased the risk of progression to advanced AMD 6. These observations were verified by later, large population‐based studies. Using the Wisconsin Grading System to grade retinal features, the Beaver Dam Study found that 14% of eyes with large (>125μm) drusen developed advanced AMD over 10 years 32. The Blue Mountains Eye Study, also using the Wisconsin Grading System, found that eyes with large drusen were six times as likely as eyes without large drusen to develop advanced AMD 33. The BMES also found that eyes with large, soft drusen were more likely to progress to neovascular AMD than atrophic AMD 33
. The Rotterdam Study, using the International Classification System 16, found that ten or more large drusen at baseline best predicted development of advanced AMD six and a half years later 34. Using these findings and their own, the Age‐
Related Eye Diseases Study (AREDS) investigators developed a severity scale for assessing AMD risk 35, which could produce a simplified score based on the presence of large drusen or pigmentary changes in each eye 36. The presence of large drusen is thus an established risk factor for progression to advanced AMD. However, it is not known whether there is an orderly progression from small, to intermediate and later, large drusen in the natural history of AMD. Klein et al. found that approximately 25% of large drusen found at baseline faded, or regressed over the same period, without progression to more advanced lesions in the Beaver Dam Study 32. The smaller, Chesapeake Bay Waterman study found that large drusen disappeared in 34% of eyes over a 5‐year period 37. While AREDS recognised intermediate drusen as a feature of early AMD 22, intermediate drusen were not significantly associated with progression from early to advanced AMD in the Cardiovascular Health and Age‐related Maculopathy (CHARM) Study 38. It should also be noted that while drusen size appears to be 8
the most consistently found predictor of progression, drusen appearance may also influence risk of progression. Tikellis et al. 38 found that eyes with soft indistinct drusen were more likely to progress to advanced AMD compared to eyes with soft, distinct drusen in the CHARM study. Small drusen (<65μm) are present in 90% of the population, are not considered AMD and do not increase the risk of developing AMD 32. However, the Chesapeake Bay study found that large numbers of small, hard drusen increased the risk of developing larger drusen over a 5‐year period 37. Over a 15‐year period, the Beaver Dam Study found that eyes with more than 8 small, hard drusen were more likely to develop large, indistinct drusen and/or pigmentary abnormalities compared to eyes with fewer than 8 small hard drusen (16.3% vs. 4.7%) 29. Thus it appears that large drusen do not irreversibly progress to advanced AMD, and that while a few small, hard drusen are not considered AMD, the presence of numerous small hard drusen increases the risk of developing larger drusen or pigmentary changes. The effect fading or regressing drusen has on the long‐term risk of AMD progression is unknown. It is also unclear whether numerous small hard drusen coalesce to form larger drusen, or whether the larger drusen arise separately in these eyes. The presence of pigmentary abnormalitites (hyperpigmentation or hypopigmentation) alone, without drusen, is not included as a definition of early AMD by all investigators, notably the International Classification System 16. However, in the three largest and longest running population‐based surveys eyes with pigmentary changes alone were included in the early AMD or age‐related maculopathy diagnostic group. Both the BMES 33 and AREDS 35 found that the risk of progression to advanced AMD is lower in eyes with hyperpigmentation alone group compared to eyes with large drusen. The BMES also found that eyes with hyperpigmentation alone did not show a strong predisposition for progession to 9
one or the other form of advanced AMD 33. However, in Beaver Dam Study, the 5‐ and 10‐ year incidence of advanced AMD is almost the same for large indistinct drusen and pigment changes 32, 39. The Rotterdam Study 34 also found that the most important predictors of progression to late AMD were either the presence of drusen involving more than 10% of macula area or the presence of retinal pigment changes 34. However, the BMES, Beaver Dam Study, Rotterdam Study and AREDS have all found that the risk of progression to advanced AMD is highest when abnormalities of the pigment epithelium are present in addition to large, soft drusen 29, 33‐35. Eyes with neovascular AMD are at highest risk of rapid vision loss, with 80‐90% of eyes losing at least 2 lines of vision within 2 years (reviewed in 12). Once neovascular AMD is present in one eye, the Macular Photocoagulation Study Group found that 42% of individuals with neovascular AMD will have second eye involvement within 5 years 40. Even after disciform scar formation, patients can continue to lose vision due to the development of expanding areas of GA around the scars 41. In eyes with GA, vision loss can also continue due to enlargement of the area of involvement. Sunness et al found that the median expansion rate for GA was 2.1mm2 per annum 42. 2.1.5 Risk factors 2.1.5.1
Smoking Smoking is the most consistently found modifiable risk factor and is associated with the prevalence of both early and advanced AMD (and to a lesser extent, incidence) 43‐49. Pooled data from the Beaver Dam Study, the BMES and the Rotterdam Study suggest that smokers were approximately 3‐4 times more likely 10
to have any form of AMD compared to non‐smokers 25. Current smokers were also 2‐3 times more likely to develop AMD over a 5‐6 year period compared to non‐
smokers 50. The association between smoking and early AMD appears to be less strong. The Beaver Dam Study found that smoking was associated with the development of soft drusen, but not pigment abnormalities, over a 5‐year period 51
. It has been postulated that smoking creates additional metabolic or biological stressors that contribute to the development of AMD. These include reductions in macular carotenoids, choroidal blood flow, serum antioxidant levels, and RPE anti‐oxidant defence systems 26. It appears however, that smoking cessation does not reverse the damage. In the Beaver Dam Study, participants who quit between baseline and the 5‐year follow up were no less likely to develop AMD than those who continued smoking 32. 2.1.5.2
Cardiovascular disease and cardiovascular risk factors The Rotterdam Study reported a 4.5‐fold increase in the risk of advanced AMD in participants with carotid bifurcation plaques (determined by ultrasound) 52. A 2.0‐
2.5 fold increased risk was also found for plaques in common carotid and lower extremities. However, when other studies have examined the association with stroke or myocardial infarction and AMD, the results have been less conclusive 53. The Beaver Dam Study found that higher pulse pressure and systolic blood pressure were associated with pigmentary abnormalities and neovascular AMD over a 10‐year period, although no association was found at the 5‐year follow up 54
. The Framingham, National Health and Nutrition Examination Survey and Rotterdam studies also found positive associations between hypertension and AMD (reviewed in 53). However, other studies have found no association between 11
blood pressure and AMD (reviewed in 26). Significantly, there is no current evidence suggesting that antihypertensive medications reduce the risk of AMD. The relationship between serum lipids and AMD risk is appears to be complex. While some studies have been unable to find an association (reviewed in 26), the Beaver Dam & Rotterdam studies both found that increased serum HDL‐
cholesterol or low LDL‐cholesterol was associated with prevalent and incident AMD. Additionally, the Beaver Dam Study also found an increased risk of early AMD with low total serum cholesterol and lower cholesterol/HDL‐cholesterol ratios 54. Surprisingly, these observations are the inverse of the relationship between lipids and cardiovascular risk. The use of lipid‐lowering agents does not appear to have a significant effect on AMD risk 26. Genetic studies have found associations between allelic variations in the apolipoprotein E (ApoE) gene and AMD. Apolipoproteins are involved in the transport and metabolism of lipids, and ApoE is particularly important in the central nervous system. Individuals with the epsilon 2 variant appear to have an increased risk of developing AMD while those with the epsilon 4 variant are protected 55‐60. While these observations have not been consistently found by all studies 61, 62, they are again the opposite of the relationship between ApoE genetic variants and cardiovascular disease. 2.1.5.3
Inflammatory markers The relationship between systemic markers of inflammation, such as C‐reactive protein (CRP), fibrinogen, cytokines and white cell count have been examined by a number of studies 63‐70. The results have been somewhat mixed, and are discussed in more detail in Section 2.5.3. 12
2.1.5.4
Dietary factors There does not appear to be a strong association between dietary fat and the risk of AMD (reviewed in 53), although the BMES found that high omega‐3 fatty acid intake may protect against early and advanced AMD 71. The BMES also found that the use of lipid lowering drugs (statins) was associated with a decreased 10‐year risk of soft drusen 72. Oral supplementation with antioxidants (vitamins A, C and E) and zinc reduced the risk of AMD progression and vision loss in the AREDS cohort 73. Diets high in antioxidant‐rich fruits and vegetables also appear to be associated with a lower risk of exudative AMD (reviewed in 53). The Nurse’s Health Study found that a higher intake of high glycaemic index carbohydrates was associated with increased risk of pigmentary changes over a 10‐year period 74. 2.1.5.5
Genetic factors AMD is not a single gene disease. Its late onset (not typically seen until the seventh decade) and phenotypic heterogeneity suggest that complex interactions between multiple genes and environmental exposures are likely to be involved. Twin studies have shown that the concordance rate for monozygotes is about twice that for heterozygotes, supporting a genetic component for the large drusen or multiple small drusen phenotypes in particular 75, 76. Familial aggregation studies suggest that a first degree relative of an individual with AMD is up to five times more likely to develop AMD than a family member of a control 77, 78, although there appears to be a large degree of variability in the genetic risk among families 79. Multiple sequence variations in many genes, either singly or more likely, in combination, are likely to exert subtle effects on the development of the disease. 13
Recently, candidate gene screening and genetic association (or linkage) studies have found several single‐nucleotide polymorphisms that are moderately or strongly associated with various AMD phenotypes. It is important to note that polymorphisms are present in the normal population, and only confer susceptibility. This is in contrast to gene mutations which, when inherited, always lead to disease. Technological advances have allowed the entire genomes of individuals to be rapidly scanned. Linkage studies compare the genomes of normal individuals with diseased individuals to identify regions which are sufficiently different and thus potentially implicated in disease pathogenesis. In AMD, the majority of studies support a linkage to the 1q31 region (reviewed in 80). A number of genes potentially important to AMD reside in this region. These include the genes for hemicentin, laminin C1 and C2, regulator of G‐protein signalling 16, oculomedin, glutaredoxin 2, proline arginine‐rich end leucine‐rich repeat protein and complement factor H (CFH). The next region of interest linkage studies identified was 10q26 (reviewed in 80). Within this region resides HTRA1, the gene for a serine protease involved in the degradation of extracellular matrix proteoglycans, and loc387715, whose function and gene product are unknown. Candidate gene screening studies select genes based on: (i) their functional importance in the pathobiology of a disease; (ii) their presence in susceptibility loci found in linkage studies; or (iii) their association with hereditary diseases that share similarities with the disease of interest. The sequences of the selected genes in normal individuals are compared with those with AMD to identify differences. Screening of genes involved in rare, hereditary degenerative maculopathies have yielded mixed results. The ABCA4 gene encodes a transporter protein involved in the visual cycle. While ABCA4 mutations give rise to Stargardt disease, the gene 14
does not appear to play a role in the majority of AMD cases (reviewed in 81). Deletions in the ELOVL4 gene, which is involved in fatty acid biosynthesis, result in Stargardt‐like macular dystrophy. A single nucleotide polymorphism in the ELOVL4 gene was associated with AMD in one population 80 but not another 82. A missense mutation in the fibulin‐5 gene is responsible for all cases of Malattia Leventinese, another rare hereditary maculopathy. The same mutation however, was present in only 1.7% of patients with AMD 83, and no associations with fibulin‐5 polymorphisms have been found. Candidate screening of genes based on their role in other diseases has identified associations between AMD and the ApoE gene, as previously discussed (Section 2.1.5.2). This approach has also identified: (i) two at‐risk polymorphisms in the fraktalkine (CX3CR1) gene 84; (ii) one at‐risk polymorphism in the toll‐like receptor 4 (TLR‐4) gene 85; (iii) and one at‐risk polymorphism and two protective polymorphisms in genes encoding human leukocyte antigens (HLA) 86. These genes encode proteins involved in inflammatory cell recruitment (CX3CR1) and phagocytosis (TLR‐4 and HLA). The possible functional implications of these allelic variants for AMD pathogenesis are discussed in Section 2.4.3.3. Screening of genes in the AMD susceptibility loci have found the strongest disease associations to date. Of these, a single nucleotide polymorphism at position 402 in the CFH gene has been robustly associated with both early and advanced AMD in a number of studies 82, 87‐94. CFH is a regulatory protein of the complement pathway, which prevents uncontrolled activation of complement cascades. These observations have lead to candidate gene screening of other members of the complement cascade. Both at‐risk and protective single nucleotide polymorphisms have been identified for complement factor B (BF), and complement component 2 (C2) 95. The pathobiological implications of complement pathway gene variants are discussed in Section 2.5. 15
Candidate studies of genes in the 10q26 region have found strong associations between a single nucleotide polymorphism in the promoter region of HTRA1 and AMD risk 96‐98. The potential functional implications of HTRA1 polymorphisms at the tissue level are discussed in Sections 2.2.3.1. Candidate gene studies of loc387715, the other gene located at 10q26, have also identified a single nucleotide polymorphism that is strongly associated with both early and advanced AMD 82, 99‐
103
. Although loc387715 mRNA is weakly expressed in retina, the gene product is yet to be identified. Because the loc387715 and HTRA1 genes are only 6bp apart, it is possible that the associations are due to one gene and not the other. While associations between a gene single nucleotide polymorphism and a disease do not necessarily infer causality, genetic studies have provided valuable insights, which will continue to inform research into pathogenesis. There is already some evidence that both the CFH and loc387715 polymorphisms are associated with AMD progression 94, 103, and it has been estimated that individuals homozygous for the at risk variants of CFH, loc 287715 and BF have a 250‐fold higher risk of developing AMD 104. It is thus likely that genetic testing in some form will be used in the near future, both to identify individuals most at risk of developing AMD and those most at risk of progression. In summary, the strongest risk factors thus identified for AMD are age, smoking and polymorphisms in the complement factor H gene. 2.1.6 Estimated costs and disease burden AMD has a significant impact on quality of life. Patients with mild AMD are estimated to lose an average 17% in quality of life, similar to the reduction experienced by patients with moderate cardiac angina or those with symptomatic 16
HIV. Those with moderate AMD have a 40% reduction in quality of life, akin to patients with severe angina or on renal dialysis. Severe AMD causes a 63% reduction in quality of life, similar to patients with end‐stage prostate cancer or those who have suffered a catastrophic stroke (reviewed in 105). The prevalence of AMD is expected to rise. In the US, for example, the population over 85 years is set to increase 107% by 2020 53. In Australia, AMD is estimated to cost $2.6 billion annually, projected to grow to $6.5 billion by 2025 (reviewed in 106
). The direct financial costs of blindness due to AMD was estimated at up to $22 507 per individual per year in 2000 107. Estimates of the economic burden of AMD using a value‐based medicine approach suggest it may have a $30 billion negative impact annually in the US 108 and $2.6 billion in Canada 109. The Australian study predicted a saving of $5.7 billion over 20 years if AMD progression could be reduced by as little as 10%. The search for therapeutic interventions is thus an urgent priority in industrialised countries. 2.1.7 Therapeutics 2.1.7.1
Preventative interventions The only modifiable risk factor for AMD is smoking. However, cessation of smoking has not been consistently shown to reverse risk 46, 47. The use of high dose antioxidant and zinc supplementation is not yet a part of standard clinical practice, since AREDS is the only prospective study that has found a reduction in the risk of AMD progression 110, 111. Ongoing population‐based studies and intervention trials will better define which nutrients confer the greatest protection and the best formulations or routes of administration for their use as supplements. 17
2.1.7.2
Palliative interventions For patients with neovascular AMD, a number of recent innovations have offered hope of slowing progression and stabilising, or even improving, vision. Photodynamic therapy (PDT) uses verteporfin, an intravenously administered photosensitising dye to target subfoveal CNV, and can be used concomitantly with intravitreal injection of triamcinolone, a corticosteroid. However, PDT treatment (with or without triamcinolone) only leads to modest reductions in vision loss and does not generally result in vision gain in the long term (reviewed in 112). The most promising therapeutic strategy to emerge in recent years is the use of engineered antibodies that target vascular endothelial growth factor (VEGF) and inhibit CNV. The dosing schedule and long term effects of a number of anti‐VEGF therapies are currently being evaluated, although several have been shown to slow disease progression and stabilise vision 112. To date, ranibizumab, a recombinant monoclonal antibody which recognises all isoforms of VEGF, is the only anti‐
VEGF therapy shown to reverse vision loss. The rate of vision loss was slowed in over 90% of participants over two years in a randomised trial, with up to a third of participants experiencing vision gain which was maintained over two years 113. There is currently no palliative therapy for GA and preventative interventions for all forms of AMD are limited. Advances in gene therapy and stem cell therapy have the potential to alleviate vision loss and the development of AMD lesions in the near future. However, AMD is a complex disease and a thorough appreciation of pathogenic mechanisms will necessarily underpin new therapies. The following sections overview normal macular anatomy and AMD histopathology, and summarise the current understanding of disease pathogenesis as well as its limitations. 18
2.2 THE “ANATOMY”
OF
AMD
2.2.1 The macula in health and ageing 2.2.1.1
Anatomy & embryology of the human eye The eyeball lies in the bony orbit, and consists of three layers of tissue: (i) the outermost fibrous tunic, composed of the sclera and cornea; (ii) the uveal tract, which forms the middle layer; and (iii) the retina, the innermost layer. The eye is divided into the anterior and posterior chambers by the iris and lens. The sclera and cornea together form an elastic envelope that gives the eye its shape. The uveal tract is composed of the iris, ciliary body and choroid. It carries the principle nerves and blood vessels of the eye, upon which the intraocular structures depend for their nourishment. The retina is an extension of the central nervous system, and extends from the optic nerve to the ora serrata. The retina and optic nerve are forebrain derivatives, with the optic nerve connecting the retina to the brain via the optic chiasm and optic tracts. In life, the retina is completely transparent, and in contact with Bruch’s membrane externally and the vitreous body internally. The macula, or macula lutea (“yellow spot”), is a shallow depression that lies in the central retina, 3.5mm lateral to the edge of the optic disc. High concentrations of the carotenoids lutein and zeaxanthin, known as macular pigment, give rise to the yellow colour of the macula. Although the macula area measures 5‐6 mm in diameter, it accounts for almost 10% of the visual field due to the concentration of photoreceptors. Microscopically, the retina has ten well‐defined layers. From the outermost to innermost, these are: (i) retinal pigment epithelium (RPE); (ii) photoreceptor outer segments; (iii) outer limiting membrane; (iv) outer nuclear layer, comprised of photoreceptor cell bodies; (v) outer plexiform layer; (vi) inner nuclear layer; (vii) inner plexiform layer; (viii) ganglion cell layer; (ix) nerve fibre layer; and (x) outer 19
limiting membrane. The rod and cone cells contain photopigments essential for the transduction of light into chemical and then electrical energy. Within the macula lies the fovea (1.5mm in diameter) and foveola or fovea centralis (0.35mm in diameter). The photoreceptor layer in the foveal region contains only cones ‐ approximately 100 000 cones within the total foveal area of 1.75mm2 114. The nerve fiber layer, ganglion cell layer, inner plexiform layer and inner nuclear layer are all absent in the fovea. Embryologically, the eyeball begins to form with the invagination of the optic vesicle into the optic cup at week four. Two neuroectodermal layers, in contact with each other at their apices, will have differentiated into the neurosensory retina internally and the RPE externally by week four. A thin capillary network appears and covers the entire optic cup after its formation. Primitive rod and cone inner, and later outer, segments appear in the seventh week, with the neuroblastic layers of the macula differentiating first. Pigmentation of the cells of the outer cup increases during the eighth week. The anterior margin of the cup grows forward during the tenth week to form the future ciliary body and iris. The inner plexiform layer forms in the 12th week, the end of embryonic life. During the fourth month, the outer plexiform layer is discernable and bipolar, amacrine, Muller and horizontal cells differentiate. The first retinal vessels also appear in the fourth month, and develop their adult pattern by the fifth month after birth. The RPE cells enlarge from the fourth to the eight month. The final arrangement of retinal cells and their processes is complete by the sixth month, with some further retinal differentiation after birth, particularly in the macular region. RPE cells, however, continue to increase in density between birth and two years of age 114. 20
2.2.1.2
The retinal pigment epithelium Approximately 4.2 ‐ 6.1 million RPE cells are present in the adult eye. Pigmentation varies depending on the region of retina, with the highest density of pigmentation at the macular region. RPE cells at the macula measure 14μm wide by 10‐14μm high, compared to flatter cells at the ora serrata that measure up to 60μm wide. RPE cells of the ora serrata are often multinucleate. It is generally accepted that in situ, RPE cells are post mitotic, and that adjacent cells slide in to replace cells that have died 114. The primary function of the RPE is to support the photoreceptors. They secrete and maintain both the inter‐photoreceptor matrix and Bruch’s membrane, regenerate bleached visual pigments, provide metabolic support by facilitating the movement of fluid and ions between the choroid and the photoreceptors, and phagocytose spent photoreceptor outer segments. The phagocytic load is great: in the rhesus monkey, for example, each RPE cell supports around 30 photoreceptors, which shed a combined 3000 membrane disks every day 115. In humans, each RPE cell supports approximately 45 photoreceptors, and a rod outer segment is fully phagocytosed and replenished every 10 days (reviewed in 116). To accomplish this, RPE cells have a highly developed lysosomal system. Outer segments are engulfed by phagosomes, which fuse with lysosomes to allow degradative enzymes to break down their contents. Undigested material becomes residual bodies, which become the substrate for lipofuscin in iron‐catalysed reactions in lysosomes. This autofluorescent lipid‐protein aggregate accumulates in the post‐mitotic RPE cells from the age of 16 months (reviewed in 2). In addition, the RPE releases cytoplasmic contents into Bruch’s membrane, to meet the demands of the high phagocytic load (reviewed in 117. The RPE secretes and maintains its own basement membrane and intercellular membrane, as well as the intercellular matrix of the photoreceptors and 21
choriocapillaris. These processes depend on a balance of RPE‐derived membrane metalloproteinases (MMPs) and tissue inhibitors of metalloproteinases (TIMPs) 118, and possibly other proteolytic enzymes 119. As well as providing the local cellular scaffolding, RPE cells regulate their own survival, together with that of the choriocapillaris, by the production of VEGF and pigment epithelial derived growth factor (PEDF), such that loss of RPE results in loss of not only the overlying photoreceptors, but the choriocapillaris as well 120. The RPE also regulates the choroidal vasculature, most likely by producing a balance of pro‐ angiogenic factors such as VEGF121, 122, and anti‐angiogenic factors, such as PEDF 123 and Fas ligand (FasL) 124. Additionally, RPE cells secrete a number of immunomodulatory factors that maintain outer retinal immune privilege and the immunological specialisations of the choroid (see Section 2.2.1.7). 2.2.1.3
Age‐related changes in the RPE A degree of polymorphism with regard to RPE cell shape, size, nuclei and degree of pigmentation occurs with advancing age. Cell nuclei become smaller and more basophilic 114. In the macular region, RPE cells become taller and narrower with normal ageing, and accumulate lipid and granular material, which is deposited in the inner collagenous zone of Bruch’s membrane. Lipofuscin, consisting of partially degraded cell membranes in lysosomes, accumulates and increases from around 1% of the cytoplasm during the first decade of life, to 19% during the ninth decade 125. Lipofuscin accumulates in cells that do not have a mechanism to deal with oxidised lipoproteins. These cells can be professional lipoprotein scavengers like macrophages, or non‐professional phagocytes such as central nervous system neurons and RPE. In RPE, it is most concentrated in the central retina, where photoreceptor cell density, and hence metabolic demand, is highest. Increasing lipofuscin content not only significantly reduces the volume of functional RPE cytoplasm, its generation also produces damaging reactive oxygen species 22
(reviewed in 116). Death of any critically damaged RPE cells results in increased demands on the remaining cells. Lipofuscin can slowly disappear once the overlying photoreceptors are lost. The residual bodies that give rise to lipofuscin retain degradative enzyme activity, implying that once the phagocytic demands on RPE are reduced, lipofuscin is eventually degraded, recycled or shed into Bruch’s membrane (reviewed in 117). Several histopathological studies have shown that an abnormal, thickened RPE basement membrane, together with loss of RPE basal infoldings, occurs in eyes after the seventh decade 10, 126, 127. Not only are these features indicative of cellular stress and loss of function, they were also found to be commonly associated with choroidal neovascularisation (see Section 2.2.2). Not surprisingly, RPE failure is considered a central event in the pathogenesis of AMD. 2.2.1.4
The choroid and Bruch’s membrane The retina has one of the highest metabolic demands of any tissue. While the inner retina derives its nutrition from the retinal vessels, the choroid functions principally to nourish the outer retina. It also acts as a conduit for vessels that supply the anterior eye. The choroidal stroma contains a significant number of pigmented cells – choroidal melanocytes – which are concentrated in the outermost layers 128. Microscopically, the choroid consists of: (i) Bruch’s membrane; (ii) choriocapillaris; (iii) stroma; and (iv) suprachoroidea. In ultrastructure, Bruch’s membrane is comprised of the basement membrane of the RPE, an inner collagenous layer, an elastic layer, an outer collagenous layer, and incorporates the basement membrane of choriocapillaris endothelial cells. It thus comprises the outer limit of the retina, and acts not only as a stabilising and anchoring structure, but also as a barrier and a filter 128. 23
Bruch’s membrane separates the photoreceptors and RPE from their primary source of nourishment, the choriocapillaris, and forms a permeable filter through which metabolic exchange can take place. Glycosaminoglycans, laminin, fibronectin and collagens I, III, IV, V, and VI are found in various combinations within the five layers of Bruch’s membrane 129. The composition and turnover of these extracellular matrix components can alter the structural integrity and permeability of Bruch’s membrane. Age‐related Bruch’s membrane thickening is attributed to compositional changes (Section 2.2.1.6), and abnormal extracellular Bruch’s membrane deposits are a feature of early AMD (see Section 2.2.3). Some of the material shed from RPE into Bruch’s membrane is probably cleared by the choriocapillaris by passive diffusion. Guymer et al suggest that choroidal endothelial cells or pericytes may also actively phagocytose RPE‐derived debris 117. Recent evidence points to a significant role for choroidal macrophages, a population of dedicated phagocytes, in keeping Bruch’s membrane debris‐free (discussed in Section 2.3). Embryologically, endothelial cells in the mesoderm adjacent to the RPE of the optic cup begin to differentiate during the fourth week. By the fifth week, the vascular net extends along the entire optic cup, coinciding with heavier pigmentation of the RPE. The RPE basement membrane and inner collagenous layer of Bruch’s membrane develop by the sixth week. Also at this time, rudimentary vortex veins appear and the choroidal capillary network becomes more developed. The elastic layer of Bruch’s membrane appears in the twelfth week. The elastic layer is one third (or less) as thick at the fovea than the periphery in the adult 130. Most of the choroidal vessel systems develop during the third and fourth months. Pigment in the choriodal melanocytes first appears in the fifth month. The choroidal melanocytes, unlike the RPE, derive from the neural crest. 24
The unique capillary system of the choroid resides in a single plane and is highly fenestrated on the Bruch’s membrane side. The blood flow rate through the choroid is relatively fast, resulting in lower oxygen exchange per millilitre of blood, and a relatively oxygen‐rich outer retina. In the macaque monkey, 90% of outer‐retinal oxygen is consumed by the photoreceptors 131. A rich nerve supply exists in the choroid, probably to control ocular blood flow, which also depends on intraocular as well as systemic blood pressure. Loss of the choroid, either experimental or disease‐related, results in atrophy of the overlying retina 128. A capillary‐free zone of 0.4mm exists within the macula, making it particularly vulnerable to age‐related reductions in choroidal blood flow 132. 2.2.1.5
Age‐related changes in the choroid There is a decrease in choroidal thickness from 200μm at birth to 80μm by age 90, due to a decrease in both choriocapillaris density and lumen diameter 5, 133 as well as increase in intercapillary pillar width 134. This is accompanied by an increase in Bruch’s membrane thickness 5, 133. The normal lobular arrangement, where a central arteriole feeds lobules of capillaries surrounded by venules 135, becomes tubular (reviewed in 117, thus reducing the surface area available for exchange. There is also a reduction of choriocapillary fenestrations as the overlying RPE becomes damaged or lost with age 5, 120. Correspondingly, there is a drop in macular choroidal blood flow (as measured by laser Doppler flowmetry), with delayed indocyanine green choriodal filling in normal individuals over 50 years 136
. 2.2.1.6
Age‐related changes in Bruch’s membrane Bruch’s membrane measures 2μm in thickness at birth and increases to 4 to 6μm by the tenth decade 133, 137, 138. Between 10 and 90 years, Bruch’s membrane 25
thickness doubles. These changes have been attributed to increased or altered protein content, increased lipid content and increased glycosaminoglycan size. On light and electron microscopy, there is accumulation of vesicular, granular, membranous and filamentous material, most prominently between the RPE basement membrane and the inner collagenous layer 129, 137, 139. There is increased collagen cross‐linking and possibly, glycosylation, which result in resistance to breakdown by RPE collagenases. Correspondingly, there is impaired collagen turnover and increased Bruch’s membrane collagen content (reviewed in 116). Two types of collagen accumulate: the normally present 64nm‐
periodicity collagen, and a 100‐140nm‐periodicity collagen described as “long‐
spacing collagen” 137, 140. Although long‐spacing collagen is also found in young, normal eyes in the outer collagenous zone, outer‐collagenous zone deposits steadily increase with age 140. Laminin, an important basement membrane structural component that facilitates RPE anchoring and choriocapillaris remodelling, is also increasingly found deposited in the inner collagenous zone, and between the outer collagenous zone and choriocapillaris basement membrane (reviewed by 117). Guo et al found reduced active forms of matrix metalloproteinases (MMPs) and increased inactive forms in the Bruch’s membrane of aged eyes, suggesting impaired extracellular matrix turnover 118. Bruch’s membrane hydroxyproline content decreases proportionate to tyrosine, methionine and phenylalanine in the macular region 141. Material resembling RPE phagosomal contents can be found in the inner collagenous layer, and later the elastic layer 142. There is also evidence of an age‐related increase in advanced glycation end products (products of non‐enzymatic glycosylation of long‐lived protein) in Bruch’s membrane 143. 26
Bruch’s membrane lipid content is negligible before the age of 30 by rises to 220mg/m2 by age 100, and consists primarily of phospholipids, triglycerides, fatty acids and free cholesterol, which suggest a cellular source 134, 144‐146. Peroxidised lipid content also increases with age 147, particularly at the macula, and appear to derive from the long‐chain polyunsaturated fatty acids found in outer segments. These age‐related changes in Bruch’s membrane composition and thickness alter its permeability. Fluid movement is significantly impeded by the fifth decade 148, particularly at the macula. Maximal resistance was found in the inner collagenous layer, and thought to be due to the entrapment of both lipid and visculo‐granular material 149. Diffusion of macromolecules 150 and amino acids 151 across Bruch’s membrane also follow a significant age‐dependent decline. The expression of vascular endothelial growth factor (VEGF) is increased in RPE, photoreceptor outer segments and the outer plexiform layer with increased Bruch’s membrane lipid content 152. VEGF is a potent promoter of angiogenesis, and its production is stimulated by hypoxia and ischaemia 153. Taken together, age‐related changes in the RPE, choriocapillaris and Bruch’s membrane result in reduced blood flow, impaired oxygen, fluid, amino acid and macromolecular exchange, dysregulated extracellular matrix turnover and alterations in RPE‐derived trophic factors. These are normal age‐related changes and do not by themselves constitute AMD. However, in susceptible individuals, it is within this compromised context that AMD develops. Inflammation is increasingly recognised as a significant factor in the pathogenesis of AMD (Sections 2.4 & 2.5). Before describing the pathological changes in early and advanced AMD (Sections 2.2.2 & 2.2.3), it is important to consider the specialised local immunological environment in which inflammation occurs. 27
2.2.1.7
Immunological specialisations of the retina and choroid Immune reactions within ocular tissues, while sharing many features with immune responses in other tissues, are atypical. This was first demonstrated when allografts transplanted into the anterior chamber of the eye survived better than those transplanted in skin 154. It is now understood that “ocular immune privilege” is the result of a complex network of interactions which allow the control of sight‐
threatening inflammation while affording immunity. This is made possible, in part, by the relative lack of lymphatics draining intraocular tissues,and the existence of the blood‐retina barrier (maintained by tight junctions between retinal vascular epithelium internally and the RPE externally). However, it is clear that immune privilege results from active as well as passive processes. Intraocular tissues express and secrete a number of molecules that suppress inflammation. Expression of major histocompatibility (MHC) molecules is limited to low levels of class I and virtually no class II in the anterior chamber, vitreous cavity and subretinal space 155, 156. Parenchymal ocular tissues express FasL, inducing apoptosis in invading activated T cells 157. Despite the presence of the blood‐retina barrier, ocular antigens can leave the eye compartment inside ocular‐
derived antigen presenting cells via the circulation 158. While the APCs travel to the spleen, they do not activate helper T‐cells. Rather, ocular‐derived APCs are programmed by a local microenvironment rich in immunosuppressive cytokines, such as TGFβ, vasoactive intestinal peptide and somatostatin (reviewed in 159). The end result is the production of CD8 suppressor cells and the suppression of both delayed type hypersensitivity and complement‐fixing antibody producton 160. While the location of the uveal tract is external to the blood‐retina barrier, several lines of evidence suggest that local immune reactions are also “atypical”. In the choroid, the local immune microenvironment appears to depend to a large degree 28
on RPE‐derived immunosuppressive factors. The RPE separates the retina from the highly fenestrated capillary network of the choriocapillaris. Due to the large amount of photoreceptor outer segments phagocytosed by RPE daily, and the constant release of cytoplasmic contents into Bruch’s membrane (Section 2.2.1.2), retinal antigens are readily available for antigen presentation in the choroid. It is thus important that inappropriate retinal antigen presentation resulting in sensitisation of the RPE or retina to autoimmune attack is prevented. RPE cells do not express MHC II molecules in the quiescent eye 161. While RPE can express pro‐inflammatory cytokines when stimulated, they constitutively express the anti‐inflammatory molecules transforming growth factor‐β1 (TGFβ1) 162, interleukine‐1 (IL‐1) receptor antagonist 163, somatostation, interleukin‐10 (IL‐10) and PEDF 164. These molecules are thought to modulate the immune microenvironment of Bruch’s membrane and the choriocapillaris, such that resident dendritic cells and APCs become tolerogenic. RPE compromise may thus have significant immunomodulatory consequences for the choroid and Bruch’s membrane (discussed further in Sections 2.4 and 2.5). 2.2.2 The pathology of endstage AMD 2.2.2.1
Geographic atrophy of the RPE: the natural endstage of AMD Gass used “geographic atrophy” to describe the slowly enlarging or coalescing areas of RPE atrophy, resulting in lesions with irregular boundaries 6. Within an area of atrophy, there is an absence of RPE cells, and loss of the overlying photoreceptors. The underlying choriocapillaris is atrophic or absent, highlighting its dependence on RPE‐derived maintenance factors 120. RPE cells at the edge of the atrophic area form a heaped‐up, hyperpigmented edge, which consists of large cells filled with lipofuscin and membrane‐bound bodies 165, with distortion of the 29
few remaining microvilli. Photoreceptors overlying the edge of atrophy are grossly abnormal, if present at all 8. 2.2.2.2
Choroidal neovascularisation and disciform scarring In the early stages of choroidal neovascularisation (CNV), capillary‐like vessels extend from the choroid into the inner Bruch’s membrane and the sub‐RPE space 7, 166
. Initially, the new vessels grow within the space normally occupied by basal linear deposit and drusen, i.e. along the inner layers of Bruchʹs membrane 167.
Endothelial sprouts continuous with pre‐existing choroidal vessels are surrounded by enlarged, activated pericytes, eventually forming lumina 167. The points of origin can vary from one to 12, with an average of 2.2 per eye 168. These vessels later become larger veins and arteries 126. Early CNV is rarely clinically discernable. Sarks found sub‐clinical CNV in a surprisingly high proportion (41.7%) of eyes with GA 10, suggesting that although a common occurrence, early CNV does not readily progress to neovascular AMD. Ultrastructural evidence supports two different phases of new vessel growth 167. A “low‐turnover” phase characterises the initial intra‐choroidal neovascularisation, while substantial sub‐RPE neovascularisation is accompanied by extensive endothelial cell proliferation and migration in a “high‐turnover” phase. In rare instances, CNV may encroach into the sub‐retinal space and join with retinal vessels 126. As discussed earlier (Section 2.1.1), once CNV is established in the sub‐RPE or sub‐
retinal space, vessels can leak or bleed, causing serous or haemorrhagic detachments of the RPE. Resolution of these events results in the proliferation of fibrous tissue and the formation of a sub‐RPE fibrovascular scar. Sarks noted that the choroid underlying disciform scars was thicker than that underlying areas of GA 10, although both are thin compared to normal 169. 30
The histopathological features of GA, CNV and disciform scars are thus well defined. However, there is less agreement regarding the histopathological definition of early AMD. This is partly because the earliest histopathological changes are not clinically evident. It was Gass who first proposed that neovascularisation originating from the choroid may occur secondary to accumulation of abnormal deposits beneath the RPE 6. 2.2.3 The basal deposits and early AMD 2.2.3.1
Basal laminar deposit is a marker of RPE degeneration In 1976, Sarks described a deposit found between the RPE plasma membrane and its basement membrane (Figure 2.2‐1), with a blue, striated appearance on picro‐
Mallory staining, which eventually came to be known as basal laminar deposit (BLamD) 10 (Figure 3.1‐2). The first systematic clinicopathological review of its kind, the study found that in 378 eyes, submacular BLamD correlated well with age, Bruch’s membrane hyalinisation and importantly, with the degree of RPE degeneration 10. When eyes in this series were organised by the appearance of BLamD, a histopathological grading system, comprised of five groups, emerged (detailed in Section 2.2.3.5). In 1986 Loffler et al confirmed that the BLamD consisted of long‐spacing collagen in a granular matrix, and was composed, in part, of normal RPE basement membrane molecules, such as collagen IV, laminin and heparin‐sulphate proteoglycan 170. On EM, the earliest evidence of BLamD had a remarkably similar appearance to normal RPE basement membrane 170. A late form of BLamD, found internal to the early form was noted subsequently, and found to correspond to severe RPE degeneration 8 (Figure 3.1‐2). Ultrastructurally, the late BLamD is 31
amorphous, with the appearance of having been laid down in waves (Figure 3.1‐
2). The similarity of BLamD to RPE basement membrane suggests it is perhaps an aberrant type of basement membrane material, secreted by distressed RPE. Hirata et al examined RPE secretion of radiolabelled proline in vitro and proposed that BLamD accumulation may be the result of aberrant degradation, rather than enhanced synthesis 171. There is some evidence of age‐related reduction in active MMP‐2 and increase in inactive degradative enzymes, suggesting reduced Bruch’s membrane extracellular matrix turnover 118. As discussed earlier, a single gene polymorphism in the promoter region of the HTRA1 gene has been associated with AMD risk (Section 2.1.5.5). The HTRA1 gene product is a member of the serine protease family, which includes proteins which facilitate the activity of matrix‐degrading enzymes (including MMPs) 172. The functional impact of the polymorphism appears to be HTRA1 over‐expression 97, although it is not yet clear how this impacts extracellular matrix turnover within Bruch’s membrane and BLamD deposition. Because material similar to BLamD can be found in the peripheral retina 173, and in the trabecular meshwork and cornea 140, it is not considered specific for AMD. However, it is the commonest histopathological finding in eyes with AMD 126, 169, 174
, and remains the most reliable marker of RPE degeneration 8, 175, 176. 2.2.3.2
Basal linear deposit and soft drusen are the specific lesions of early AMD Membranous material accumulates in the inner collagenous layer of Bruch’s membrane from the second decade onwards 177. Ultrastructurally, this material resembles coiled phospholipid membranes, and can be found in the RPE basement 32
membrane and between strands of early BLamD 3. Early investigators postulated that it derived from RPE, either as partially digested phagosomes 177 or selectively sequestered aliquots of cytoplasm shed into Bruch’s membrane 139, 177. Thin layer chromatography studies have shown that lipids of cellular origin, such as phospholipids, triglycerides, fatty acids and free cholesterol preferentially accumulated in macular Bruch’s membrane 144. Figure 2.2‐1 Schematic diagram of BLamD and membranous debris When membranous material collects in a continuous layer external to the RPE basement membrane, it is known as basal linear deposit (BLinD) (Figure 2.2‐1). This deposit is only visible on electron microscopy (EM). In contrast to BLamD, BLinD disappears when RPE is atrophied and overlying photoreceptors absent 178, suggesting that it is a product of the RPE‐photoreceptor complex. Curcio and Millican found that not only is BLinD specific for early AMD, but soft drusen, the hallmark lesions of early AMD, are composed of the same membranous material 33
176
or “membranous debris”179, 180. Thus, soft drusen can be understood as focal accumulations of BLinD external to the RPE basement membrane 175, 176, 179 (Figure 2.2‐1). Smaller focal collections of membranous debris can also accumulate internal to the RPE basement membrane, and are known as basal mounds (Figure 2.2‐1). The biochemical constituents of membranous debris are not fully characterised. Earlier histochemical and ultrastructural studies identified neutral lipids and cerebrosides/gangliosides 142 in drusen. Pauleikhoff and colleagues later showed that drusen also contained phospholipids 181. Ruberti et al used quick freeze/deep etch electron microscopy to find that lipids, including cholesterol, preferentially accumulated with age in a thin layer external to the RPE basement membrane. The authors proposed that lipid deposition in this layer may be the pre‐cursor event to BLinD formation (and later, soft membranous drusen) 146. Recently, Curcio and colleagues have shown that Bruch’s membrane and drusen contain solid lipid particles 182, esterified and unesterified cholesterol 145, 183 and lipoproteins 184, 185. Because they were also able to demonstrate the presence of microsomal triglyceride transport protein 186 and apolipoproteins C‐I, C‐II, E, and J 185 in RPE, these investigators have proposed that dysregulated lipid metabolism by RPE may be responsible for the accumulation of membranous debris in AMD. Although RPE apolipoprotein biochemistry is not well understood, genetic studies support a role for ApoE in AMD pathogenesis. In the central nervous system, ApoE modulates phospholipid and cholesterol transport as part of normal neuronal metabolism, as well as during neuronal remodelling. Certain ApoE gene polymorphisms, particularly the E4 allele, are linked to Alzheimer’s disease and atherosclerosis. Interestingly, the E4 allele of ApoE appears to be associated with a decreased risk of neovascular AMD 55, 58‐60, 187 (discussed in Sections 2.1.5.2 & 2.1.5.1). 34
Another mechanism by which membranous debris may form is by the non‐
apoptotic blebbing of damaged RPE cells, discussed further in Section 2.2.3.4. 2.2.3.3
Drusen classification Drusen are seen as discrete spots in the fundus clinically (see Section 2.1.2). Histopathologically, they consist of focal, discrete aggregates of abnormal extracellular material external to the RPE basement membrane. However, drusen can vary widely in their appearance in terms of size, shape, elevation, distribution and contents, and drusen nomenclature remains confusing. To a large degree, this is the result of the lack of availability of post mortem eyes which have been photographed during life, thus limiting clinicopathological correlations. Clinically, drusen have been described as: hard or soft; distinct or indistinct; large, intermediate or small; as well as by their fluorescein staining characteristics. Histopathologically, drusen have been described as: small hard (or hyalinised); soft extensive (or membranous); as well as by their ultrastructural and histological staining characteristics (with terms such as papillary drusen and basal laminar drusen). Hard drusen have a discrete, hyalinisted light‐microscopic appearance, while soft drusen are usually larger, have sloping margins, and membranous or granular contents. Large, population‐based studies have contributed a great deal to the understanding of the natural history of drusen. It is clear that the natural histories of small hard drusen and large soft drusen are distinct, and the associated AMD risk is also significantly different. Small, hard drusen are present in 94% of the population 29 and 80% of post mortem eyes 188. Over a 5 to 15 year period, eyes with fewer than 8 small hard drusen are no more at risk of developing AMD than normal eyes 29, 32‐34, 37, 39. Eyes with soft, indistinct drusen (>125μm), on the other 35
hand, are almost 18 times more likely to develop late AMD over 5 and 15 years 29, 37
. Unfortunately, studies investigating drusen contents often do not discriminate between small hard drusen and large soft drusen, while the results are generalised to explain AMD pathogenesis. This may be partly due to the vulnerability of soft drusen contents to loss during tissue processing, and partly because small hard drusen are much more commonly found in donor eyes. However, given the differences in clinical behaviour, there has been a surprising lack of large‐scale studies comparing the contents of hard and soft drusen. Hageman and Mullins found that the extracellular matrix protein vitronectin, and carbohydrate moieties stained by wheat germ agglutinin and limax flavus agglutinin were present in several drusen phenotypes 189, 190. This, and their reports of the presence of immunoglobulin G and complement components in both hard and soft drusen 191, 192
lead them to propose that phenotypic descriptions of drusen as “hard” or “soft”, “small” or “large” were meaningless. While this approach simplifies drusen nomenclature, it fails to address the clinical reality that a few small hard drusen are considered normal while large soft drusen are the hallmarks of early AMD, and carry a significant risk of progressing to advanced AMD. However, Hageman and Mullins point out that different drusen phenotypes may result not from differences in protein or carbohydrate composition, but from the amount and type of lipids present 189. 2.2.3.4
Drusen biogensis Evidence to date suggests the RPE plays a significant role in the biogenesis of both hard and soft drusen. Observations that RPE overlying drusen are cytologically abnormal in shape, size and pigment characteristics were made at least as early as 1905 193. Ultrastructural studies later showed that pigmentation, intercellular junctions, and organelle content and organelle distribution are altered in RPE 36
adjacent to drusen 142, 194, 195. As RPE becomes increasingly degenerate histopathologically, soft drusen regress and collapse, and hard drusen may disappear 196. Farkas and colleagues demonstrated the presence of degenerate organelles and possibly lysosomes in drusen 197, implicating the RPE as a source of drusen components. Although lysosomal activity was not demonstrated in a study by Burns and Feeney‐Burns 195, cytoplasmic debris was found in small drusen. In a later ultrastructural study, Ishibashi et al found that RPE cells in aged human eyes with drusen shed portions of their basal cytoplasm into Bruch’s membrane 198
. These “buds” of RPE cytoplasm contained many organelles, including secondary lysosymes, mitochondria, vesicles and vacuoles. In conditions of stress, mammalian cells may sequester non‐functional or toxic cytoplasmic components in discrete packets of their cytoplasm to be shed as membrane blebs 199, 200. There is in vitro evidence that oxidative damage causes increased RPE membrane blebbing 201
. Interestingly, RPE MMP‐2 activity is correspondingly reduced under the same conditions, suggesting a possible mechanism for increased membranous debris production and decreased degradation. A recent proteomic analysis of primarily hard drusen found increased carboxyethylpyrrole adducts (CEP protein adducts), products of docosahexaenoate lipid oxidation 202. Since docosahexanoate is the most abundant fatty acid in photoreceptor outer segments 203, the finding of CEP adducts in drusen suggests that drusen may contain indigestible, RPE‐derived photoreceptor elements. However, it is evident that apart from RPE derived components, hard drusen also contain molecules involved in extracellular matrix turnover, as well as acute phase reactants and other molecules of the innate immune system 202 (discussed further in Section 2.5.1.2). The possible involvement of choroidal dendritic cells in drusen biogenesis is discussed in Section 2.4.4. 37
2.2.3.5
Histopathological classification schemes As with clinical classifications, there is no universally accepted classification system for AMD histopathology. In particular, histopathological definitions of early AMD have varied. These include: (i) the type or number of drusen; (ii) the presence of basal deposits; (iii) RPE changes; and (iv) degree of photoreceptor degeneration 126, 133, 169, 175, 204. Histopathological grading is only meaningful when pathological features can be correlated to clinical (particularly fundus) features. Two large‐scale studies have attempted to do this. Sarks surveyed 378 eyes, all with clinical data, including fundus appearance, and in some cases, fluorescein angiography 10. Eyes with no BLamD were considered normal (group I). A normal fundus was seen in all eyes in this group. Eyes with patchy BLamD represented normal ageing (group II). The majority of eyes in this group had a normal fundus. Since approximately half of the eyes with a thin continuous layer of BLamD (group III), and all eyes with thick continuous BLamD (group IV) had clinical fundus changes (soft drusen and/or pigmentary disturbance), they were considered to have early AMD. Eyes with geographic atrophy (Group V) and disciform scarring (Group VI) were classified irrespective of BLamD appearance, although BLamD was incorporated into scar tissue in disciform lesions and persisted in areas of GA even in the absence of RPE. The Alabama Age‐Related Macular Degeneration Grading System for donor eyes was devised by Curcio and colleagues, and surveyed 34 eyes, 30 of which had accompanying clinical data 205. Clinicopathological correlations were made using 5 eyes. The grading in this study was based on the extent of the “basal deposits, a combination of BLamD and BLinD” (since BLinD can only be distinguished on EM). In both studies, normal and early AMD eyes were defined histopathologically. Tables 2.2‐1 and 2.2‐2 compare the two grading schemes. Although similar in many respects, Sarks’ grading scheme was used for the 38
histopathological studies in Chapter 3 due to its larger sample size and more extensive clinico‐pathological correlations. Despite the availability of clinical and histopathological classification systems, there is no universal definition of early AMD. Chapter 3.1 presents findings of a clinicopathological study that examined the influence of BLamD and membranous debris on the progression of AMD. The histopathological threshold at which ageing becomes early AMD is discussed. 39
Table 2‐1 Histopathological grading scheme suggested by Sarks (1976) 10 Group
I
II
III
IV
V
VI
BLamD
Absent
Patchy
Thin continuous
Thick continuous
Persists after loss of pigment
Incorporated into scar
RPE
Normal
3.6% with fine pigment disturbance
GA
Incorporated into scar
CNV
Nil
Nil
41.7%*
100%
46.7% fine 16.7% fine pigment pigment 11.7% coarse 78.6% coarse pigment
pigment
Nil
14.3%*
* subclinical CNV
Table 2‐2 Histopathological grading scheme suggested by Curcio et al (1998) 205 Grade
0
1
2
3
4
Basal deposits
None
Patchy
Thin continuous
Thick continuous
N/A
RPE
Uniform
Non‐uniform
Heaped or sloughed
Anterior migration
GA
CNV
None
Present
N/A
N/A
N/A
40
2.3 AMD
DISEASE PATHOGENESIS: SUMMARY OF CURRENT
CONCEPTS
Although the pathogenesis of AMD is incompletely understood, histopathological & epidemiological studies, together with newer molecular and genetic insights, show that AMD results from age‐related changes plus additional pathological events. Models of AMD pathogenesis fall into one of three main categories. The mechanisms described in each are not mutually exclusive, and all centre on interactions between photoreceptors, RPE, Bruch’s membrane, and choroid. 2.3.1 Oxidative stress Reactive oxygen intermediates (ROIs), are released under physiological conditions by mitochondrial oxidative phosphorylation and liver cytochrome p450‐related reactions. ROIs can be quenched or removed by antioxidant defence systems, which depend on the enzymes glutathione, superoxide dismutase and catalase, and on vitamins A, C and E, the carotenoids, bioflavinoids, selenium and zinc. Oxidative damage results if antioxidant defences are exceeded by the production of ROIs. Advancing age is accompanied by increased oxidative damage, due to an age‐
related reduction in antioxidant defences. For example, total plasma glutathione, oxidised glutathione and peroxidised lipid increase with age, while total serum vitamin C and E decrease (reviewed in 116). The retina is susceptible to oxidative damage due to the amount of light irradiation it receives, the relatively high concentration of local oxygen, the presence of peroxidation‐prone polyunsaturated fatty acids (in photoreceptor outer segments), the presence of photosensitising pigments and the high RPE phagocytic load. There is an age‐related decline in macular pigment (lutein and zeaxanthin) 206, RPE cell density, and RPE cell vitamin E 116. Conversely, there is a corresponding increase in RPE lipofuscin. In a 41
study of AMD donor eyes graded by post‐mortem fundus features, Decanini et al found that RPE expression of catalase, copper‐zinc superoxide dismutase and manganese superoxide dismutase increased with increasing AMD pathology 207, along with proteins associated with secondary defence against oxidative damage. These findings suggest oxidative defence is upregulated in response to increasing oxidative stress. However, even though protein levels are increased, there is evidence that at the activity of least one antioxidant enzyme, catalase, diminishes with age 208. Other ROI‐quenching enzymes, such as metallothionein 209 and heme oxygenase 210 appear to decrease in RPE with age. Data from epidemiologic studies tend to support a role for oxidative stress in AMD. Lower plasma glutathione reductase 211, and higher glutathione peroxidise 212
are associated with AMD. A number of population‐based studies have shown increased risk of AMD with low dietary intake of carotenoids (reviewed in 206), and AREDS (see Section 2.1.5) has reported a reduction in risk of progression to advanced AMD with dietary antioxidant supplementation 73. In addition, cigarette smoking, known to increase ROIs and reduce antioxidant defence systems, is a well‐established risk factor for AMD (see Section 2.1.3). Oxidative damage can result in destruction of cytoplasmic and nuclear structures, leading eventually to cell death. Apoptotic cells themselves release cytochrome c from their disintegrating mitochondria, leading to higher local levels of ROIs (reviewed in 213). Non‐lethal oxidative damage to RPE cells may result in cellular blebbing, which may contribute to the deposition of sub‐RPE membranous debris 214
. Oxidative damage may also result in the production of an abnormal extracellular matrix and thickened Bruch’s membrane, due to modifications in matrix proteins or their degradative enzymes. 42
2.3.2 Ischaemia The age‐related increase in lipid and protein content reduces Bruch’s membrane permeability to water soluble plasma molecules and amino acids, resulting in reduced metabolic exchange (Section 2.2.1). Increased Bruch’s membrane thickness results in a greater distance between the fenestrated choriocapillaris and the RPE, thus also impairing oxygen exchange. Age‐related reductions in choroidal blood flow, due to loss of choriocapillaris density and choroidal thinning, further exacerbate these changes. Even small reductions in the diffusional capacity of Bruch’s membrane or choriocapillaris blood flow can significantly affect photoreceptors and RPE, since the photoreceptors consume 90‐
100% of the oxygen delivered by the choriocapillaris 131. Reductions in gas exchange and in the bidirectional flow of molecules and fluid between the choroid and RPE, further promote the production of ROIs. 2.3.3 Inflammation Inflammation is a tightly controlled response to injury or infection, essential in restoring the balance between health and disease. It is becoming increasingly apparent that a breakdown in the regulation of the inflammatory response contributes to a number of chronic, degenerative diseases. The following sections review the growing evidence implicating inflammatory and immune‐mediated processes in AMD disease pathogenesis.
43
2.4 MACROPHAGES
AND IMMUNE COMPETENT CELLS IN
AMD
2.4.1 Morphological evidence 2.4.1.1
Macrophages, Bruch’s membrane breaks and early AMD Sarks noted that in eyes with early AMD, macrophages were found adjacent to breaks in Bruch’s membrane and in association with subclinical CNV, the earliest evidence of neovascularisation 10, 179. These observations were reinforced by similar follow‐up findings 127, 215. Stromal and choroidal leukocytes, including macrophages, were found in increased numbers by Penfold et al 216 once eyes developed a continuous layer of BLD. Mean macrophage counts were significantly higher in eyes with continuous BLD compared to normal aged eyes 217. Later reports found that macrophages appeared to be attracted to membranous debris deposition within Bruch’s membrane in early AMD 218. Macrophages were also found accompanying both “active” and “inactive” sublinical CNV 219. These morphological observations suggested a link between macrophage infiltration and the earliest stages of the neovascular process, possibly by causing breaks in Bruch’s membrane which allow the in‐growth of newly formed vessels. This hypothesis was strengthened by the known angiogenic influence of macrophages in other chronic inflammatory and fibroproliferative states, including rheumatoid arthritis and solid tumour growth 220, 221. 2.4.1.2
Macrophages in advanced AMD lesions An ultrastructural study of membranes from eyes with neovascular AMD found leukocytes and macrophages within neovascular structures, at breaks in Bruch’s membrane, and in contact with activated pericytes 222, further reinforcing the view that macrophages induced new vessel growth. Macrophage association with the neovascular process appeared to persist in burnt‐out disciform lesions 7. The most recent large histopathological survey of surgically excised subfoveal CNV lesions 44
confirmed that macrophages were the most frequently found cell type after RPE in both neovascular membranes and disciform scars 174. Interestingly, giant cells appear to the be commonest type of leukocyte found in eyes with GA 217, 223, whereas the breaks in Bruch’s membrane commonly found in eyes with disciform scars are absent. This observation further emphasises the association of Bruch’s membrane breaks with the in‐growth of neovascular tissue. Sarks proposed that macrophages and giant cells are attracted to Bruch’s membrane when the phagocytic capacity of RPE is exceeded 3. This idea has been supported by a number of recent observations on choroidal macrophage recruitment and turnover (see Section 2.4.3). 2.4.2 Macrophages and choroidal neovascularisation With compelling morphological data that macrophages were intimately involved with AMD lesions, and particularly the suggestion that they may facilitate neovascularisation, investigators sought evidence of macrophage expression of angiogenic mediators. Oh et al found that macrophages in surgically excised neovascular membranes, identified by CD68 labelling, expressed the pro‐
inflammatory cytokines interleukin‐1β (IL‐1β) and tumour necrosis factor alpha (TNFα), while RPE cells admixed in the same lesion expressed vascular endothelial growth factor (VEGF) 224. The authors suggested that the presence of activated, pro‐inflammatory macrophages may induce VEGF production by RPE, promoting choroidal neovascularisation. Grossniklaus et al also examined surgically excised CNV membranes, and observed that CD68‐positive macrophages expressed VEGF and tissue factor, while RPE cells expressed monocyte chemoattractant protein (MCP) 121. These authors concluded that RPE might be important for macrophage recruitment, and that recruited macrophages expressed two growth factors essential for neovascularisation. 45
Further evidence of the ability of macrophages to induce neovascularisation was provided by animal model studies. The animal model most commonly used to approximate the CNV found in AMD uses laser photocoagulation to cause breaks in Bruch’s membrane. Neovascular lesions typically form within one week of photocoagulation 225. Choroidal vascular endothelial cells migrated to into the subretinal space via the laser‐induced defects in Bruch’s membrane in monkey eyes three days after photocoagulation, and macrophage infiltration was observed three to seven days after photocoagulation 226. In the same study, VEGF expression was first found in macrophages and later in RPE and Muller cells. Similarly, in a rat model of laser‐induced CNV, macrophages in both the subretinal space and the choroid adjacent to the site of laser injury expressed VEGF three to seven days after photocoagulation 227. Tsutsumi and colleagues explored this relationship further when they compared macrophage infiltration and extent of CNV after laser photocoagulation in a Ccr‐2 knockout mouse model 228. Ccr‐2 is the receptor for macrophage chemoattractant protein‐1 (MCP‐1), is normally expressed by macrophages and is essential for macrophage trafficking. Both the number of infiltrating macrophages and the extent of CNV was significantly less in the knockout mice compared to wild type mice. These observations suggested that the neovascularisation process depended much more on recruited, and not resident, macrophages. Further strengthening this observation, Sakurai et al found that depletion of circulating monocytes using intravenous clodronate reduced both CNV lesion volume and leakage (as measured by fluorescein angiography) 229. To demonstrate that the recruited macrophages were derived from circulating monocytes, Caicedo et al used fluorescently‐labelled monocytes transplanted into the bone marrow of irradiated mice 230. Care was taken to reduce the laser intensity 46
and duration to limit its effect to Bruch’s membrane and minimise retinal injury. Blood‐derived monocytes in this study infiltrated the retina overlying CNV three days after laser photocoagulation. Activation of the overlying Muller cells (evidenced by their expression of c‐fos and phosphorylated extracellular signal‐
regulated kinase) occurred secondary to macrophage infiltration. Muller cell activation was abolished after depletion of circulating monocytes using clodronate. In summary, the laser‐induced model of CNV has produced a number of important observations. Firstly, it is likely that circulating monocytes, and not resident immune cells, are responsible for the inflammatory damage immediately after Bruch’s membrane disruption. Indeed, infiltrating macrophages may activate local cells 230. Secondly, recruited macrophages express VEGF and the extent of neovascularisation and leakage depends on the extent of macrophage trafficking and infiltration. However, laser photocoagulation tends to destroy the outer retina, along within Bruch’s membrane and choroid. In any case it is a significant, acute traumatic event very unlike the chronic, but sub‐lethal cellular changes that take place in AMD. Not surprisingly, the first cells to infiltrate laser‐induced CNV lesions are neutrophils, which are conspicuously absent in AMD lesions. As discussed below, the Ccr‐2 knockout mice used to illustrate the dependence of CNV on macrophage recruitment turn out to behave very differently when left to age without intervention. 2.4.3 Choroidal macrophage recruitment and turnover and AMD 2.4.3.1
Ccl‐2, Ccr‐2 and Cx3Cr1 knockout mice and AMD‐
like lesions In 2003, Ambati et al reported that knockout mice deficient in either MCP‐1 (Ccl‐2) or its receptor Ccr‐2, developed AMD‐like lesions when left to age beyond 9 47
months 231. These included RPE lipofuscin accumulation, RPE degeneration, photoreceptor fall out and the development of drusen and choroidal neovscularisation; features not seen in wild type mice even beyond 24 months of age. Evidence of IgG and C3c in choroidal vessel walls of the knockout mice suggested immune complex deposition. C5, serum amyloid P protein and advanced glycation endproducts were also immunolocalised on RPE or the choroids of knockout mice. In wild‐type mice, an age‐dependent increase in Ccl‐2 expression by RPE cells, together with an age‐dependent increase in choroidal macrophages, suggested that Ccl‐2 – Ccr‐2 interaction was critical to normal choroidal macrophage recruitment. This study also found that choroidal macrophages from wild type mice were able to degrade C5 and IgG deposited on the choroids or RPE of knockout mice. Together, data from this report suggests that resident choroidal macrophages play a critical role in the elimination opsonised debris in Bruch’s membrane. Macrophages expressing scavenger receptors for oxidised lipoproteins are seen found in human AMD eyes 232. Disruption of the normal recruitment or turnover of choroidal macrophages may lead to persistence of both the debris and the opsonising molecules C5 and IgG, resulting in lesions similar to those found in AMD. AMD‐like lesions were also found in mice deficient in Ccl‐2 and CXC3R1 (fractalkine, a chemokine receptor involved in leukocyte trafficking) 233, further reinforcing this view. These reports highlight the differences between age‐related pathological changes and those that result from laser disruption of Bruch’s membrane in mouse models. While CNV is reduced in Ccl‐2 (MCP‐1) knockout mice after laser photocoagulation, it develops spontaneously when the same mice are left to age beyond 9 months. 48
2.4.3.2
IL‐10 knockout mice and laser‐induced CNV An interesting, but counterintuitive finding in interleukin‐10 (IL‐10) knockout mice was recently reported by Apte et al 234. IL‐10 is a major immunosuppressive cytokine that programs macrophages along the non‐inflammatory, “repair” pathway (discussed in Section 2.4.3.4). These authors reported that the knockout mice developed less extensive laser‐induced CNV compared to wild type mice. Lack of IL‐10 would be expected to permit macrophage programming along the pro‐inflammatory pathway, resulting in more CNV. The authors explain their unexpected observations by proposing that it is the inhibition of normal macrophage recruitment (e.g. by local IL‐10) that promotes CNV. That is, a lack of macrophage recruitment results in increased debris deposition in Bruch’s membrane. This in turn results in the increased deposition of acute phase reactants and other opsonising components of the innate immune system, leading to further activation of damaging inflammatory cascades. 2.4.3.3
Polymorphisms in leukocyte recruitment and turnover genes A number of genetic studies have supported the observation that normal macrophage recruitment and function is important in the maintenance of a healthy and debris‐free Bruch’s membrane. Goverdhan et al examined polymorphisms in the human leukocyte antigen gene and AMD risk. The authors found one at‐risk allele and two protective alleles, as well as differential expression of HLA antigens in the choroid 86. HLAs are important molecules for macrophage recognition of antigen. The HLA genes are the most polymorphic in the human genome, and the mechanism by which the confer susceptibility to AMD is not yet clear. These findings, however, provide further evidence for the involvement of immune or inflammatory processes in AMD pathogenesis. 49
A single gene polymorphism in the CX3CR1 gene was also found to be associated with AMD 84. Donor eyes with the “at risk” allele had reduced fractalkine transcripts in the macula region. Since fraktalkine is an important chemokine in monocyte transendothelial migration 235, the authors concluded that aberrant macrophage recruitment may be implicated in AMD pathogenesis. Finally, a single gene polymorphism of the toll‐like receptor 4 (TLR‐4) gene was also associated with AMD 85. TLR‐4 plays an essential role in innate immunity, and mediates macrophage recognition and phagocytosis of microbial invaders 236. Interestingly, TLR‐4 appears to also involved in RPE photoreceptor phagocytosis and in cholesterol transport 237, 238. 2.4.3.4
Serum myeoloid cells and AMD There is some epidemiological evidence implicating circulating myeloid cells in AMD, suggesting a higher systemic inflammatory “set point” A higher white cell count was found in patients with neovascular AMD and disciform scars compared to controls in a small case‐control study in 1986 239. High white cell count was also associated with neovascular AMD 240, and with the 10‐year incidence of large (>125μm) drusen in the Beaver Dam Study 241. Finally, monocytes isolated from patients with neovascular AMD expressed more TNFα when stimulated with RPE blebs in vitro compared to controls 242. 2.4.3.5
Macrophages in an immune‐privileged compartment Cells of the macrophage/monocyte lineage can vary widely in function, morphology and immuno‐phenotype depending on the tissue microenvironment in which they are found. In broad terms, macrophages can be activated along a “classical” or “alternate” pathway (reviewed in 243). Classical activation follows stimulation with bacterial lipopolysaccharide and IFNγ or with TNFα. These macrophages express the enzyme inducible nitric oxide synthase (iNOS), which 50
allows them to convert arginine to nitric oxide, producing nitrite, peroxynitrites and superoxides. These powerful oxidants cause lipid peroxidation of affected cell membranes, leading to cell death. Alternate activation of macrophages switches on a different metabolic program, which is TGFβ1‐dependent. Macrophages thus programmed express the enzyme arginase, which converts arginine to ornithine, promoting cell division and repair. A rich network of macrophages and dendritic cells has been found in the choroid of mice 244. The dendritic cells appear to be in the “immature” phase of their life cycle, similar to the Langerhans cells of skin, suggesting that their function in the normal choroid is antigen capture and not antigen presentation. Although definitive data is not available for human eyes, it is likely that similar networks also exist. Resident choroidal dendritic cells and macrophages appear to be pre‐
programmed to resist classical activation, thus limiting inflammation and promoting repair. In culture, dendritic cells and macrophages isolated from rat and mouse choroid failed to present antigen 245. And, although the immunological specialisations of the choroid are less well characterised than the retina and other ocular compartments, there is evidence that local inflammation‐suppressing immune modulation exists, primarily via RPE cells (see Section 2.2.1.7). In culture, RPE produce of thrombospondin‐1 and TGFβ, and can directly suppress T‐cell activation 246. Mouse eye cup RPE cells produced PEDF and IL‐10, potent immunosuppressive cytokines. The supernatants from these RPE cells inhibited activated macrophages by reducing their production of IL‐12 (a pro‐
inflammatory cytokine) and increasing their production of IL‐10 (an immunosuppressive cytokine) 164. This effect was neutralised by anti‐PEDF antibodies. Ingestion of oxidised photoreceptor outer segments by the human RPE cell line ARPE‐19 in culture resulted in increased expression of MCP‐1, but not TNFα 122, suggesting that RPE can recruit macrophages without activating them. 51
In pathological circumstances, it appears that choroidal macrophages can be classically activated. Hattenbach et al found that macrophages in the eye of a 70 year old woman with neovascular AMD expressed iNOS 247. Endothelial cells within the neovascular membranes of the other 6 CNV specimens in this study also expressed iNOS. Although the macrophages in this study were identified by morphology alone, the results suggest that CNV‐associated macrophages in human eyes are phenotypically different to resident choroidal macrophages. Their ability to produce nitric oxide can not only damage local cells, but may also stimulate angiogenesis. Ando et al found that VEGF‐mediated retinal and choroidal neovascularisation in mice depended on local nitric oxide production 248. And finally, in the experimental autoimmune uveitis mouse model, choroidal macrophages (identified by ED1 labelling) expressing iNOS were seen 11 days after sensitisation with retinal S antigen, coinciding with the peak of inflammation 249
. Thus, iNOS‐expressing choroidal macrophages may be associated with both inflammation and neovascularisation. In the context of AMD, the influence of the progressive accumulation of extracellular deposits on local populations of macrophages, both recruited and resident, is not well defined. Section 3.2 examines the influence of BLamD and membranous debris on macrophages found within Bruch’s membrane and in the choroid. 2.4.4 Dendritic cells and drusen biogenesis Although dendritic cell networks are likely to be present in the normal human choroid, the functional role of choroidal dendritic cells is at present poorly characterised. Hageman and associates have proposed that sub‐RPE debris can attract the processes of cells on the choroid side of Bruch’s membrane and that these cells share immunophenotypic features with dendritic cells 250. Unfortunately 52
they reported CD68 and HLA‐DR immunoreactivity in drusen, an immunophenotype more typical of macrophages. These authors propose that dendritic cell processes are attracted to injured RPE or diffuse sub‐RPE debris, becoming a focus for deposition of inflammatory proteins which eventually results in the formation of drusen. While an interesting pathogenic mechanism, there has so far been insufficient evidence to support it. 53
2.5 INFLAMMATION
AND
AMD
2.5.1 Tissue evidence of inflammatory proteins 2.5.1.1
Evidence of inflammation‐associated proteins in drusen As discussed in Sections 2.2.3.2 and 2.2.3.4, drusen contain lipids and protein. The protein constituents of drusen so far identified can be roughly divided, with some overlap, into three groups: (i) cytoskeletal and basement membrane proteins, such as crystallins, collagens I‐V, laminin, fibronectin, vitronectin and proteoglycans 251, 252; (ii) proteins associated with lipid metabolism (apolipoproteins) 184, 253; and (iii) inflammation‐associated proteins, such as immunoglobulins, HLA‐DR, C‐reactive protein (CRP), vitronectin, complement cascade components, clusterin, crystallins, amyloid, serum amyloid P, albumin, factor X, thrombin and fibrinogen (see below). Cytoskeletal proteins and those associated with lipid turnover may result from extrusion by damaged RPE cells unable to meet the metabolic demands of constant photoreceptor outer segment turnover (see Section 2.2.3.4). Similarly, the production of excess basement membrane can be expected from epithelial cells under stress. However, it is the inflammation‐associated proteins that have the most obvious potential to influence the course of disease, due to their capacity to activate damaging inflammatory cascades. The following sections review the inflammation‐associated proteins found in drusen and Bruch’s membrane to date. It should be emphasised that interpretation and comparison of such studies is complicated by differences in drusen phenotype (see Section 2.2.3.3). Where defined, the majority of studies involve hard drusen, and often those found in the peripheral retina. Nonetheless, the findings have contributed to a growing understanding of the inflammatory pathways elicited by the presence of extracellular debris within the Bruch’s membrane‐choroid complex. 54
2.5.1.1.1 Immunoglobulins and complement in drusen and Bruch’s membrane Immunoreactivity to IgM was found to be strong in macular hard drusen by Newsome et al 251, while IgG immunoreactivity was prominent in Bruch’s membrane but less strong in drusen. Johnson et al 191 also demonstrated IgG immunoreactivity in small, hard drusen in normal aged eyes, as well as in the RPE that overly or flank these drusen. Interestingly, both the complement component C5 and the terminal complement complex (membrane attack complex) C5b9 were found in drusen and adjacent RPE, leading the authors to propose that antibody‐mediated complement attack of RPE may result in drusen formation. The complement system consists of at least 20 proteins, found in serum in inactive form (Figure 2‐2). The system is involved in both innate and adaptive immunity, and in the removal of both microbial invaders and extracellular debris. Formation of the final product, the membrane attack complex (C5‐9) is triggered by activation of the critical element, C3, which can occur in three main ways: (i) the classical pathway, by C1q binding to the Fc region of IgM or IgG; (ii) the alternate pathway, by microbial elements, aggregated immunoglobulins, polysaccharides and others; and (iii) the lectin pathway, by collectin binding to viral or bacterial carbohydrate‐containing proteins. The membrane attack complex (MAC) forms transmembrane channels when it binds to target cells, leading to cell lysis. Proteolytic fragments of the complement cascade, particularly those of C3 and C5, are potent inflammatory mediators, and can be activated by proteolytic enzymes within inflammatory exudates, independently of the complement cascade. C3a and C5a cause vasodilation and increased vascular permeability via mast cell degranulation. C3b and C3bi are opsonins, facilitating phagocytosis of their targets by macrophages and neurtrophils, which possess cell surface 55
C3b receptors. Additionally, C5a is a powerful promoter of leukocyte adhesion, chemotaxis and activation. To prevent self‐perpetuating inflammatory damage, the complement system is closely regulated by inhibitory proteins. These act primarily as inhibitors of C3 and C5 convertases, although regulatory proteins that prevent C1 binding to immune complexes or MAC formation also exist 254. IgG or IGM immune complex
Acute phase proteins
Classical Pathway
C1
Activated C1
C4 + C2
C4b2a
C4b2a3b
C3
C3b
C5
C5b
C5-9
Membrane Attack
Complex (MAC)
+C6 +C7 +C8 +C9
C3
C3b
C3bBb3b
C3bBb
Factor B
Accelerates decay of C3bBb
Microbial surfaces
Lipopolysaccharides
Alternative Pathway
Factor H
Figure 2.5‐1 The complement cascade Adapted from Cellular and Molecular Immunology 254 The presence of terminal complement components and IgG in drusen provided tissue evidence that the complement system may trigger inflammatory sequelae important in AMD pathogenesis via the classical 56
pathway (i.e. immune complexes). This hypothesis was further supported when two complement inhibitory proteins were found in aged normal eyes and eyes with GA: clusterin (in hard drusen) and vitronectin (in RPE surrounding hard drusen and sub‐RPE deposits resembling basal mounds) 192
. In a separate study, the same author group co‐localised IgG and C5b‐9 immunoreactivity within hard drusen (and on the internal aspect of choroidal capillary walls) 255.C5b‐9 immunoreactivity was also found on the basal RPE membrane, suggesting that RPE cells were targets of MAC‐
mediated attack. Interestingly, only C5 immunoreactivity was demonstrated in the one soft druse presented in this study. 2.5.1.1.2 Amyloid, fibrinogen and other inflammatory proteins found in drusen Immunofluorescence studies identified the coagulation proteins, thrombin, fibrinogen and factor X in drusen in 253. RNA transcripts for fibrinogen and factor X were amplified in the in RPE/choroid of AMD eyes in this study, suggesting local production. Factor X, thrombin and fibrinogen are proteases involved in the final common pathway of the clotting cascade. Since there is a well‐recognised association between vascular injury, inflammation and coagulation/fibrinolysis, these findings support a role for choroidal vascular dysfunction (in addition to inflammation) in AMD pathogenesis. C‐reactive protein, amyloid P component, α1‐antitrypsin and vitronectin are acute phase proteins, and have all, to various degrees, been localised in drusen by immunofluorescence techniques 190, 253. Acute phase proteins are inflammatory mediators released in response to both microbial invasion and tissue injury. One of their functions is to recognise molecules released by apoptotic or necrotic cells, such as nucleic acids, mitochondrial elements, membrane proteins, modified lipids, cholesterol, cytoskeletal proteins and proteins associated with cellular stress 256. Acute phase protein binding to 57
microbial or cellular elements activates the complement cascade by the classical pathway 257, 258. This process is called “opsonisation”, and allows the rapid removal of potentially immunogenic debris from the extracellular compartment by phagocytic cells. The presence of HLA‐DR, an MHC class II molecule required in the capture of opsonised antigen by macrophages, has also been demonstrated in drusen and sub‐RPE debris 253. The abnormal extracellular deposits in found in early AMD (BLamD and membranous debris), together with damaged RPE, are obvious candidates for binding by acute phase proteins, immunoglobulins and complement. That these deposits persist in AMD eyes despite the presence of opsonisation molecules, HLA‐DR and macrophages suggests dysfunction somewhere along the opsonisation‐phagocytosis axis. This may be in: (i) the control of the acute phase response or complement cascade; (ii) the strength of the acute phase response proportionate to the amount of debris present; (iii) failure of phagocytosis; or (iv) failure of local recruitment or turnover of phagocytes. As described in the previously (Section 2.4.2), human genetic evidence and evidence from transgenic mice implicate dysfunctional macrophage recruitment and turnover in AMD and the development of AMD‐like lesions. Human genetic evidence has also strongly implicated the complement cascade in AMD pathogenesis (Section 2.1.5.5). Interestingly, the genetic evidence implicates defective regulation of complement activation via the alternate pathway, and not the classical pathway. The next section discusses the functional consequences of polymorphisms in complement proteins. 58
2.5.2 Genetic associations 2.5.2.1
Single nucleotide polymorphisms in complement cascade and complement regulatory genes In 2005, four author groups independently published data reporting an association between AMD risk and a single nucleotide polymorphism in complement factor H (CFH) gene 87‐90. These observations are particularly strong, considering the participants were derived from four separate (American Caucasian) populations, and distinct methods of gene screening were used. The studies included participants with early, neovascular or atrophic AMD. CFH is the major soluble inhibitor of the alternate complement pathway, preventing C3 activation by proteolytic cleavage of C3b. Single nucleotide polymorphisms are variations in a single nucleotide within the genome, and are commonly found (approximately 1 in 1250 base pairs). The Tyr402His variant of CFH, in which histidine takes the place of tyrosine at amino acid 402, was found to increase an individual’s risk of AMD 2‐4 fold in heterozygotes and 5‐7 fold in homozygotes. The tyrosine‐histidine substitution is purported to affect the heparin and CRP binding site of CFH, diminishing its ability to keep the complement cascade in check. Association of the Tyr402His variant was found to be strongest for individuals with early and neovascular AMD. Studies in other ethnic groups have not corroborated closely the findings from the Caucasian populations. No association was found between the CFH Tyr402His polymorphism and Japanese patients with atrophic AMD 259, neovascular AMD or disciform scars 260. Additionally, the Tyr402His allele was associated with bilateral, but not unilateral early AMD in a US‐based Latino/Hispanic population 261. Even within Caucasian populations, disease 59
risk for the allelic variants has not yet been adequately tested in large prospective studies. More recently, analysis of polymorphisms in the Factor B (FB) and C2 genes found two “protective” and one “risk” haplotype 95. BF binds C3b, and its cleavage ultimately results in the formation of C3 convertase (C3Bb), while C2 is an activator of the classical pathway. The BF and C2 genes are situated only 500bp apart within the major histocomaptibility (MHC) class III locus, and the gene products have almost identical molecular structures, both regulating the production of C3 (the critical complement cascade component). Thus it is as yet unclear whether the association with AMD is due to the BF or C2 polymorphisms. The authors suggest however, that the BF polymorphism may be causal, since one of the protective alleles produces a gene product with reduced haemolytic activity, and BF immunoreactivity within retina was similar to that of CFH, C3 and the MAC 95. The genetic evidence appears to implicate the alternative complement pathway while earlier immunohistochemical and immunofluorescence studies implicated the classical pathway (which is activated by acute phase proteins such as β‐amyloid, IgG, CRP, serum amyloid P). As discussed earlier, CFH also binds CRP 262 and a recent study provides some clues on the way CFH‐CRP interactions may influence the development of AMD. When the distribution of CFH protein in the RPE, Bruch’s membrane and choroid was assessed in donor eyes with the “at risk” CFH Tyr402His variant, no differences were found compared to normal donor eyes 263. However, the Tyr402His homozygotes did have significantly more CRP protein deposition in the choroidal stroma than normals. Since the Tyr402His variant affects a CRP binding site, these findings suggest the associated increased AMD risk may be due to failure of CFH to bind and inactivate or remove CRP. 60
Excessive extracellular CRP deposition would act as a persistent inflammatory stimulus, and an activator of the classical complement pathway. Some have suggested that genetic evidence implicating the alternative pathway points to the possible involvement of pathogens, such as Chlamydia pneumoniae, in disease pathogenesis. There is some evidence that previous infection with c.pneumomiae is associated with AMD 264 and AMD progression 265, and c.pneumonia is more frequently found in neovascular lesions compared to normal eyes 266. Current evidence therefore, suggests both the classical and alternative pathways of complement activation may be implicated in AMD pathogenesis. 2.5.3 Serum inflammatory markers in AMD A number of population‐based studies and case‐control studies have examined the association between serum markers of inflammation and AMD. These studies were undertaken partly because of the strong evidence implicating serum markers of inflammation, such as CRP and homocysteine, in cardiovascular disease (reviewed in 267), and partly because of the evidence for acute phase protein deposition in Bruch’s membrane and drusen. While Smith et al found that serum fibrinogen levels greater than 4.5g/L were associated with both neovascular and atrophic late AMD in an Australian population‐based study 268, Dasch et al failed to find any association between early or late AMD and serum fibrinogen in a German population‐based study, once cardiovascular risk factors were adjusted for 65. A small study suggests reduced serum leptin, a protein involved in control of inflammation and lipid metabolism, was associated with early and advanced AMD 269. 61
Seddon and colleagues found a significant association between CRP and both “intermediate” and advanced AMD 67 in the AREDS study, and progression to advanced AMD in a cohort of participants with non‐exudative AMD 68. Apart from a small case‐control study in which participants with drusen, neovascular or atrophic AMD were found to have higher CRP and homocysteine levels than controls 270, others have not been able to replicate these findings. In the Cardiovascular Health Study, no association between CRP and either early 64, or advanced AMD 270 could be found. Dasch and colleagues also failed to find an association between CRP and either early or advanced AMD in the Muenster Ageing and Retina Study 65. A case‐control study of participants from the Beaver Dam population could not find an association between prevalent or incident early AMD and several inflammatory markers, including CRP, IL‐6, TNFα or amyloid A and 63. Wu et al found a small association between serum intercellular adhesion molecule‐1 levels and late AMD, but found no association between early or advanced AMD and CRP, and interleukin‐6, white cell count (WCC), fibrinogen, homocysteine, plasminogen activator inhibitor‐1 or von Willebrand factor in the Blue Mountains Eye Study population 70. It may be that factors associated with cardiovascular disease have an inverse, or at least more complex, association with AMD. Certainly polymorphisms in the ApoE (a lipoprotein), ABCA1 (a cholesterol transporter) and TLR4 (a receptor important in innate immunity and cholesterol efflux) genes which are associated with increased risk of atherosclerosis are protective for AMD 85. Autoantibodies, found in serum and directed against retinal and other proteins, have also been associated with AMD. However, their putative role in AMD pathogenesis remains controversial. 62
2.6 SERUM
ANTI-RETINAL AUTOANTIBODIES IN
AMD
2.6.1 Autoantibodies and autoimmunity Autoimmunity is classically described as the failure of self‐tolerance. The ability of the immune system to distinguish “self” from “non‐self” is essential to its normal function. Antibodies are produced by B cells as part of the acquired (or specific) immune response, after exposure to antigen. Antibody binding to the antigen (immune complex formation) activates the complement cascade, leading to complement‐mediated lysis of affected cells, and recruitment and activation of inflammatory cells (neutrophils and monocytes) 271. Autoimmune diseases arise when cells or antibodies of the immune system recognise and bind self‐antigens. Immunoglobulins that recognise self antigens are known as “autoantibodies” and are typically considered pathological. They can cause tissue damage in two recognised ways. Autoantibodies may form complexes with circulating antigens, which eventually deposit in vessel walls and/or tissues. An example of this occurs in systemic lupus erythematosus, whereby autoantibodies bound to DNA, nucleoproteins and other proteins deposit to cause nephritis, disseminated vasculitis and arthritis. Autoantibodies can also directly bind antigens in tissues to cause damage, such as when anti‐
type IV collagen antibodies cause pnuemonitis and glomerulonephritis in Goodpasture’s syndrome. Both mechanisms lead to the same downstream consequences, i.e. cell lysis, inflammatory cell recruitment and opsonisation and phagocytosis of affected cells. Direct binding to a tissue antigen can also alter the functional properties of a target protein. In myasthenia gravis, autoantibodies cause paralysis by binding acetylcholine receptors, preventing neurotransmission at muscle 63
endplates. In Grave’s disease, autoantibodies bind and activate receptors for thyroid stimulating hormone, leading to thyrotoxicosis. More recently however, autoantibodies have been found in association with a number of degenerative diseases, such as Alzheimer’s disease 272 and atherosclerosis 273. In the strictest definition of an autoimmune disease, the following criteria, postulated by Witebsky 274 should be satisfied: (i) presence of circulating autoantibodies; (ii) identification of the corresponding antigen(s); (iii) induction of autoreactivity against these antigens in an experimental model, leading to (iv) tissue pathology similar to the human disease; and (v) initiation of disease following adoptive transfer of the self‐
reactive antibodies or lymphocytes. Autoantibodies found in association with degenerative diseases do not of course, fulfill these criteria. Degenerative diseases cannot thus be considered primary autoimmune diseases. However, the findings do raise questions about definitions of autoimmunity, and about the involvement of the immune system in the setting of chronic degenerative disease. Autoantibodies have also been found in association with AMD. Before reviewing this evidence, autoantibody associations with several other ocular diseases will be discussed. 2.6.2 Autoantibodies and ocular disease 2.6.2.1
Anti‐recoverin autoantibodies, CAR and autoimmune retinopathy Probably the best known autoantibody in ocular disease is the anti‐recoverin autoantibody, which is well described in cancer‐associated retinopathy (CAR). This is a paraneoplastic syndrome in which autoantibodies against the 23kD protein recoverin cause unexplained vision loss in patients with 64
diagnosed or occult tumours 275, 276, primarily small cell lung cancers. Recoverin is a photoreceptor protein involved in the visual cycle. Tumour expression of recoverin was thought to elicit the production of anti‐recoverin antibodies which cross‐reacted with those found in photoreceptors, causing cell death 277‐279. Typically, CAR patients have profoundly abnormal electroretinograms. Anti‐recoverin autoantibodies were thus thought to be specific for CAR. However, later studies showed that anti‐recoverin autoantibodies could be found in other ocular diseases, and that autoantibodies against other proteins were found in CAR. Whitcup et al found anti‐recoverin autoantibodies in patients without cancers but with symptoms clinically similar to CAR 280. This entity became recognised as “recoverin‐associated retinopathy” or “autoimmune retinopathy” 281. Anti‐recoverin autoantibodies were also found in patients with retinitis pigmentosa, a hereditary retinal degenerative disorder 282. CAR patients were later found to have autoantibodies against enolase 283, 284 and heat shock protein 70 (hsp70) 285 (anti‐hsp 70 autoantibodies induced CAR‐like electroretinogram changes when were injected into the vitreous of rats 286). Up to 15 or more autoantigens have now been described in CAR 287. Melanoma‐associated retinopathy (MAR) is another paraneoplastic syndrome whereby tumour expression of aberrant proteins leads to the development of autoantibodies against retinal antigens 287. ERG changes differ to that of CAR, although multiple autoantigens have been reported, including a 35kD Muller cell antigen and a 22kD neuronal antigen 287, as well as putative antigens retinal transducin, rhodopsin, visual mitofilin and titin 288
. 65
2.6.2.2
Retinal degenerations, autoimmune retinopathy and glaucoma Immunohistochemical studies have shown increased serum reactivity against normal human retinae for a number of retinal diseases, including retinitis pigmentosa, diabetic retinopathy 289, 290 and cystoid macular oedema 291
. Autoantibodies against enolase were found to be associated with glaucoma 292‐294 and autoimmune retinopathy 295. More recent studies of serum from glaucoma patients have also revealed complex patterns of reactivity against retinal and optic nerve antigens 296, 297. These studies were the first to show that the same complex reactivities were present in up to 100% of normal controls. It is thus increasingly recognised that the presence of serum autoantibodies is not necessarily a sign of pathological autoimmunity. However, the significance of autoantibody associations in ocular diseases is still being clarified. 2.6.3 Autoantibodies in AMD Using direct immunofluorescence, Chant and colleagues noted in 1987 298, that IgM autoantibodies from two patients with “macular degeneration” bound to human retinal tissue. A larger study by Penfold et al 299 showed that autoantibody binding to human tissue was present in 20% of normal controls, 29% of patients with geographic atrophy, 52% of patients with disciform lesions, 62% of patients with drusen and 78% of patients with pigmentary disturbance. The autoantibody staining pattern on retinal tissue in this study resembled that of glial acidic fibrillary protein (GFAP), a cytoskeletal protein found in neurons, astrocytes and Muller cells. By contrast, a later study 300 found that 47% of patients with macular degeneration (including patients with neovascular AMD, atrophic AMD and 66
bilateral drusen) had autoantibodies directed mainly to photoreceptor outer segments. Western blot experiments by the same group showed that the autoantibodies reacted with a protein between 58‐62kD, and cross‐reacted with a bovine neurofilament protein. In 1993, Chen and co‐workers reported that serum anti‐retinal autoantibodies detected by western blotting were found in 66% of patients with “AMD”, compared with 18% of controls. The autoantibodies reacted with peptides and proteins occurring in five different molecular weight bands (Table 2.6‐1). Gu and colleagues used ELISA to test autoantibody binding to carboxyethylpyrrole protein (CEP) adducts 301, a product of lipid peroxidation which, although not specific to retina, was identified as a drusen component by earlier proteomic studies 202. They found that 53% of “AMD” patients (compared to 21% of controls) had high titres of anti‐CEP autoantibodies. Patel et al correlated immunohistochemical staining patterns of serum autoantibodies with AMD disease classification 302. Autoantibodies reacted with multiple retinal elements, including the inner and outer nuclear layers, the inner and outer plexiform layers and the ganglion cell layer (with some autoantibodies targeting membranes, others nucleoli and still others nuclear membranes or photoreceptor inner segments). Autoantibodies also reacted with RPE, Bruch’s membrane and choroid. The highest proportion of autoantibody reactivity was seen in patients with large drusen and pigmentary abnormalities. Joachim et al recently found complex anti‐retinal autoantibody binding patterns on western blots in both control and neovascular AMD sera 303. Using statistical software to discriminate western blot densitometry differences, AMD sera was found to have densitometry peaks at 46 and 52kD, and troughs at 18 and 36kD. Tandem liquid choromatography/mass 67
spectrometry was used by this author group to identify the 18kD antigen as α‐crystallin, the 46kD antigen as α‐enolase and the 52kD antigen as GFAP. Interpretation of these studies is complicated by several factors: (i) the different detection methods used (immunofluorescence, western blotting and ELISA); (ii) the different serum dilutions assayed (1:20 – 1:1000); and (iii) the various clinical AMD phenotypes (some poorly defined) involved in different studies (summarised in Table 2.6‐1). Most frustratingly, there appears to be not one or two, but many, retinal antigens implicated. Only a few of these are characterised or partially characterised 300, 301, 303. Many, if not all, of the antibodies/antigens so far identified have not been shown to be specific to AMD. Most perplexing of all, the very same autoantibodies are found in a large percentage of (and in some cases, all) normal control sera, making it difficult to argue that they are pathological. One explanation is that only high autoantibody titres are associated with disease. The evidence for anti‐CEP autoantibodies appears to support this view 301. However, Joachim et al found that certain autoantibodies were found in significantly lower amounts (based on western blot densitometry) in AMD sera compared to controls 303. This observation can only be explained by accepting that a least some of the autoantibodies detected are naturally occurring autoantibodies, or “natural autoantibodies”. Natural autoantibodies are present in all normal individuals 304, 305. They are encoded by germline B cell genes, and are produced in the absence of immunisation with antigen 306. Thus, they are considered part of the innate immune system and react with a large number of self‐epitopes 307. An individual’s natural autoantibody repertoire appears to be stable over time 308
. 68
Some natural autoantibodies recognise highly conserved pathogen‐
associated molecular patterns, such as phosphorylcholine, which allow them to neutralise invading pathogens rapidly. However they also react with cytoskeletal proteins, nuclear proteins and DNA 309, suggesting that they are involved in the removal of apoptotic cells, cellular debris and other altered self antigens. In fact, immunoadsorbent assays show that the majority of antibodies in human sera are self‐reactive 310. Mono‐reactive, highly specific antibodies generated by adaptive immunity appear to represent a very small proportion of total serum immunoglobulin. This is consistent with the emerging view that innate immunity evolved not only as first line defence against pathogens, but for the equally critical role of “housekeeping”, or keeping the extracellular compartment “debris‐free”.
69
Table 2‐3 Autoantibodies in AMD: summary of studies Year
Authors
AMD phenotype
1985
Chant et al
"macular degeneration" (n=2)
Direct immunofluorescence
using human retinae
not specified
1990
Penfold et al
Drusen (n=13)
Pigmentary disturbance (n=40)
Geographic atrophy (n=21)
Disciform scar (n=44)
Direct immunofluorescence
on human, rat and mouse
retinae
1991
1993
2003
2005
2006
Gurne et al
Chen et al
Gu et al
Patel et al
Joachim et al
Bilateral drusen (n=5)
Atrophic AMD (n=6)
Exudative AMD (n=19)
"AMD" (n=41)
"AMD" (n=19)
ARM (n=64)
CNV (n=51)
Method(s)
Serum dilution Autoantibody Ig
class
Autoantibody targets
% Postive
(disease sera)
% Positive (control
sera)
IgG, IgM
Rod outer segments [1]
50%
2%
1:10
not specified
Staining similar to GFAP labelling,
targeting predominantly astrocytes
29-78% [2]
20%
Indirect immunofluorescence
on human and monkey
retinae
1:2
1:4
IgG
Photoreceptor outer segments
(prominent); outer plexiform and
outer nuclear layers (weak)
47%
"weak or none"
Western blotting using
human, bovine and monkey
retinal homogenates
1:200
1:1000
IgG
Between 58-62kD. Cross reaction
with 68kD bovine neurofilament
protein
Western blotting
Not available [3]
Not available
28-32kD, 38-42kD, 48-52kD, 6265kD or 110-130kD
66%
18%
ELISA using synthesised
carboxyethylpyrrole (CEP)
protein adducts [4]
1:20
IgG
CEP adducts
53%
21% [5]
83-94% [6]
9%
100%
100%
CEP localised strongly to
photoreceptor outer segments and
RPE and weakly to inner plexiform
layer
Immunohistochemistry (to
localise CEP in mouse and
human retinae)
Indirect immunofluorescence
using murine retinal tissue
1:10
1:100
IgG
Choroid, Bruch's membrane, RPE,
photoreceptors, inner nuclear layer,
ganglion cell layer
Western blotting using
murine retinal homogenates
1:500
IgG
28-49kD
Western blotting using
bovine retinal homogenates
1:40
IgG
Complex staining, multiple
antigens in 10-180kD range
Wet AMD (n=39)
Liquid choromatography
tandem mass spectrometry
alpha-crystallin (18kD); alphaenolase (46kD); GFAP (52kD)
70
Table 2.6.1. Autoantibodies in AMD: summary of studies [1] It is not clear whether this was the staining pattern for the two macular degeneration patients or for the retinitis pigmentosa patients in this study [2] Highest for pigmentary disturbance, lowest for geographic atrophy [3] Chinese language journal [4] Product of lipid peroxidation [5] All controls had anti‐CEP autoantibodies. These percentages indicate high titre anti‐CEP autoantibodies [6] highest for bilateral drusen 71
In the context of AMD pathology, it is reasonable to postulate that the presence of abnormal extracellular debris (in the form of BLamD, BLinD and soft drusen), and cell signalling by damaged RPE, may lead to an increased production of natural autoantibodies, in tandem with other opsonising molecules of the innate immune system. The antigenic targets so far characterised for AMD seem to support this view. GFAP is a cytoskeletal protein normally expressed by astrocytes, and by Muller cells after retinal injury as part of the neuroprotective gliosis reaction Astrocyte and Muller cell GFAP expression has been shown to increase with age, and particularly in the presence of drusen 311. Alpha‐crytallin is a member of the small heat shock protein family, with molecular chaperone‐
like properties that prevent the precipitation of denatured proteins 312. Alpha‐
crystallin was one of the most frequently found proteins in drusen 202 and appears to accumulate in Bruch’s membrane and the choroid in AMD eyes 313
. Alpha‐enolase is an important glycolytic enzyme, as well as a heat shock protein, produced in response to hypoxic stress 314. Interestingly, higher titres of autoantibodies to α‐enolase are found in a number of autoimmune, degenerative and inflammatory conditions 315. And, as already discussed, CEP adducts are products of lipid peroxidation. The extracellular lipids found in drusen are vulnerable to oxidation, and CEP adducts have been found in drusen 202. Immunoglobulins themselves have also been found in drusen 191, 251, 253, Bruch’s membrane 251 and the abnormal RPE overlying drusen 191 in AMD eyes. If they are indeed natural autoantibodies, they may have a protective role in the removal of extracellular debris. Natural autoantibody‐mediated clearance of debris may be particularly relevant under pathological conditions that involve accumulation of stress‐induced structures, such as 72
atherosclerosis, where autoantibodies against heat shock proteins have been shown to prevent plaque formation (reviewed in 213). However, the setting of rapidly accumulating extracellular debris and an altered immune microenvironment may also favour the development of pathological antibodies. There is some evidence that pathological autoantibodies (such as rheumatoid factor) may derive from the natural autoantibody pool 316. The difficulty remains in determining which autoantibodies are pathological, and which are protective. One way is to examine whether there is “subclass restriction”, since restriction of autoantibodies to one IgG subclass is a feature of pathological autoimmunity 317
. Another way to determine whether autoantibodies are pathological is to determine their association with disease progression. Chapter 4 summarises preliminary studies that examine the autoantibody profile in normal controls and in individuals with early, neovascular and atrophic AMD. Autoantibody association with disease progression in early and neovascular AMD is also examined. 73
2.7 THESIS
AIMS
The past two decades have brought valuable insights to AMD disease pathogenesis. However, many questions remain. The definition of AMD phenotypes, particularly early in the disease process, requires clarification. The first part of Chapter 3 is a clinicopathological study of eyes with early AMD, which sought to answer the following questions: (i)
What is the histopathological threshold that distinguishes normal ageing from early AMD? (ii)
How do BLamD and membranous debris influence disease progression and what are their clinical correlates? (iii)
Is there clinicopathological evidence for distinct early AMD phenotypes? An emerging theme in the pathogenesis of AMD is the contribution of macrophages to the maintenance of choroidal/Bruch’s membrane health. The second part of Chapter 3 examines the influence of BLamD and membranous debris accumulation on both resident choroidal macrophages and Bruch’s membrane macrophages. The following questions were addressed: (i)
In what histopathological context are Bruch’s membrane macrophages first found? (ii)
What is the clinical fundus appearance when Bruch’s membrane macrophages are present? (iii)
What is relationship between Bruch’s membrane macrophages and subclinical neovascularisation? (iv)
Does the progressive accumulation of BLamD and membranous debris alter the immunophenotype of Bruch’s membrane macrophages and/or resident choroidal macrophages? 74
A further emerging theme is the recognition of the significant role inflammation plays in AMD pathogenesis. While there is growing evidence implicating altered innate immunity in AMD, the relationship between anti‐
retinal autoantibodies and AMD remains poorly defined. Chapter 4 summarises preliminary studies which addressed the following questions: (i)
Does the anti‐retinal autoantibody profile differ significantly between normal individuals and those with early AMD, neovascular AMD or geographic atrophy? (ii)
Can baseline anti‐retinal autoantibodies predict progression to advanced AMD in individuals with early AMD? (iii)
Can baseline anti‐retinal autoantibodies predict vision loss in individuals with neovascular AMD? 75
3
H I S T O P A T H O L O G I C A L S T U D I E S 76
3.1
RELATIONSHIP
OF BASAL LAMINAR DEPOSIT AND
MEMBRANOUS DEBRIS TO THE CLINICAL PRESENTATION OF
EARLY AGE-RELATED MACULAR DEGENERATION
3.1.1 Introduction The development of interventional and preventative strategies in AMD will depend on an understanding of the early stages of the disease. It is important, therefore, to describe the evolution of the two deposits pathologically significant to AMD, BLamD and membranous debris, in order to better understand the threshold at which aging becomes AMD and to correlate their further accumulation with changes in fundus appearance and visual acuity. The basal deposits are not directly visible on fundoscopy and hence their relationship to the clinical evolution of AMD is not well documented. The present study reviews a clinicopathologic series of 132 eyes in which diffuse basal deposits had been identified previously 10. Although membranous debris in the BLinD can be demonstrated only by electron microscopy (EM), the focal basal mounds and soft drusen can be identified by light microscopy, so that the respective influence each exerts on the progress of degeneration can also be assessed in histologic specimens. The type and thickness of BLamD are correlated with the sites of membranous debris accumulation and with the clinical findings. The threshold at which aging becomes early AMD is discussed. 77
3.1.2 Methods 3.1.2.1
Patients and Eyes In earlier studies a large clinicopathologic collection of aged eyes had been divided into histopathologic groups according to the appearance of the BLamD under the macula 10. Briefly, eyes with no BLamD (group I) and patchy BLamD (group II) were considered normal. The eyes chosen for the present study were those with a thin (group III) or thick (group IV) continuous layer of BLamD, considered to represent the pathologic changes preceding advanced AMD (groups V and VI). A total of 132 eyes belonging to 75 patients (69 males and 6 females) aged 67 to 97 years at death (mean 82 +/‐8 years) met the inclusion criteria (Table 3.1.1). All the patients had been examined clinically, and follow up ranged from 0 to 120 months (mean 24 +/‐32 months). Examination included best corrected visual acuity, direct fundoscopy and, unless unobtainable, fundus photography using the Zeiss 30° fundus camera. Fluorescein angiography was carried out where appropriate. Eyes in which the fundus could not be adequately visualized, or demonstrated other pathology, were excluded. The last examination ranged from two weeks to 75 months prior to death (mean 19 +/‐17 months). The study was conducted in keeping with the tenets of the Declaration of Helsinki and approved by the University of New South Wales Human Research Ethics Committee. Written consent was obtained from all patients. 3.1.2.2
Histopathological Methods and Definitions A review of archival material, including slides of 107 eyes prepared for light microscopy (LM) and electronmicrographs of 25 eyes, was conducted. Preparation 78
of eye tissue is described in detail elsewhere 10, 179. Briefly, for LM, serial sections 8μm thick were cut horizontally through the disc and macula and every 10th section stained and examined. Three sections closest to the centre of the fovea and 80μm apart were examined with a Leitz Dialux microscope. Objectives used were the APO 25x (NA=0.65) and PL APO 40x (NA‐0.75). The microscope was calibrated using a standard 0.01mm objective micrometer graticule (Olympus, Japan) and measurements were obtained with an eyepiece graticule, one division equaling 2.8μm with the 40x objective and 4.5μm with the 25x objective. Histologically BLamD was distinguished most readily with the picro‐Mallory method, an early type showing faint antero‐posterior striations and staining blue, while a later type was hyalinized and stained red. These staining characteristics are the result of the acid fuchsin (red staining) and aniline blue (blue staining) components of the picro‐Mallory stain. Acid fuchsin is a smaller molecule than aniline blue. Early type BLamD has a looser arrangement of molecules than late type, allowing the smaller acid fuchsin to wash away while trapping aniline blue. Late type BLamD is composed of more condensed material, imperveant to the large aniline blue molecule while trapping acid fuchsin. BLamD was then graded according to 2 parameters: thickness and staining characteristics. Maximum BLamD thickness was recorded as thin if it was less than half the height of the normal RPE (<7μm) and thick if greater (>7μm) 318. BLamD staining was graded in the following degree of progression: i) blue‐staining early type only (Figure 3.1‐
1A); ii) patchy late BLamD appearing either as small, rounded inclusions approximately 4μm in diameter lying within the early BLamD (Figure 3.1‐1C), or as larger nodular elevations occurring singly or in rows on the internal surface of the early BLamD, each nodule being overlain by a single RPE cell (Figure 3.1‐1D); (iii) continuous late BLamD comprising segments ≥ 250μm in length (Figure 3.1‐
2B,C). On electron microscopy (EM) the earliest form of BLamD was fibrillar and was continuous with the original basement membrane of the RPE, but the 79
predominant constituent of early BLamD was the banded form. This consists primarily of long‐spacing collagen (Figure 3.1‐1B) accounting for the striations seen on light microscopy. Late BLamD had a more condensed structure and appeared to be produced in waves as the overlying RPE retracted (Figure 3.1‐2 inset). Membranous debris was defined on EM as coiled membranes with a trilaminar appearance (Figure 3.1‐3) and was assessed by EM in 25 eyes (from 18 patients, mean age 79 +/‐7.14 years). The membranes often had a vesicular outline and may have contained lipids lost during processing 182, so that the term “membranous” is used herein as an ultrastructural description only. BLinD was defined as a layer of this membranous debris lying between the RPE basement membrane and the inner collagenous zone of Bruch’s membrane 126, 176, 179. The maximum number of layers of BLinD was recorded for each eye examined by EM. Basal mounds (Figure 3.1‐4) were defined on EM as pockets of membranous debris 176 lying internal to the RPE basement membrane and early BLamD. In histologic sections, they appeared a washed‐out pale blue with Picro‐Mallory staining, were at least half the height of an RPE cell, and were mostly 1‐2 RPE cells wide. For each eye the maximum number of mounds in any one of the 3 sections through the macula was recorded. Soft drusen were defined as focal accumulations of the BLinD with sloping margins measuring over half the height of the normal RPE, up to 350μm wide when confluent (Figure 3.1‐4), and up to 500μm wide in two cases with drusenoid detachment. Like basal mounds they stain blue with the picro‐Mallory method, but unlike basal mounds, soft drusen lay below the RPE basement membrane. Drusen contents were recorded as predominantly membranous or predominantly granular, those with granular contents being regarded as regressing. 80
Hard drusen (or “nodular drusen” 126) had a globular, hyalinised appearance and stained red with picro‐Mallory method. A few hard drusen are not considered part of AMD. One to two per section were seen in 85 study eyes. Histologic abnormalities of the RPE at the macula were graded as mild, moderate or severe depending on the degree of irregularity, hypertrophy and hyperpigmentation, or attenuation. 3.1.2.3
Clinical Parameters The best corrected visual acuity was converted to logMar for statistical purposes. The fundoscopic appearance of the macula was graded as: (i) normal, which included a few (< 5) small (<63μm) drusen; (ii) multiple small drusen (<63μm) involving an area >125μm; (iii) intermediate drusen (63 – 124μm) with or without pigment; (iv) large drusen ( ≥125μm) with or without pigment; or (v) pigment abnormalities alone. 3.1.2.4
Statistical Methods Statistics were performed using SPSS for Windows (v15.0; SPSS Inc, Chicago, IL, USA). Contingency tables of histopathological versus clinical parameters were evaluated by the χ2 test. Comparison of mean age and visual acuities was evaluated using the student’s t‐test. Correlations between BLamD type, BLamD thickness, and histopathologic RPE abnormalities were performed using Spearman’s correlation for non‐parametric data. A p ≤ 0.05 was considered statistically significant. 81
3.1.3 Results The clinical and histopathologic features of the study eyes are summarised in Table 1. 3.1.3.1
Basal laminar deposit and RPE changes The RPE was not normal even in eyes in which BLamD was exclusively of the early type, exhibiting a loss of uniformity and increase in pigmentation (Figure 3.1‐1A). However, it was the amount of late type BLamD that correlated positively with increasingly severe histological RPE abnormalities (r = 0.435, p< 0.001), first appearing as patchy late BLamD in the form of inclusions (Figure 3.1‐
1C) and then nodules (or “nodular excrescences” 319) (Figure 3.1‐1D). In the presence of severe RPE abnormality, late BLamD formed an unbroken layer exceeding 250μm in length (Figure 3.1‐2B&C) – termed “continuous late BLamD” here and also described as “diffuse thickening of the internal aspect of Bruch’s membrane” 204. The late form always lay on or near the internal aspect of the early type, closest to the base of the retracting RPE. BLamD in Group III eyes formed a thin layer, ranging in thickness from 1.0 – 7.0μm (3.2 +/‐ 1.6μm) and in 66 of 95 eyes (69%), was exclusively of the early type. Patchy late BLamD was present in the remaining 29 Group III eyes (31%). By contrast, continuous late BLamD was found in all 37 group IV eyes (100%), and ranged in thickness from 8 – 25μm (10.7 +/‐ 4.2μm). Overall, there was a strong positive correlation between BLamD thickness and the proportion of late type BLamD (r = 0.650, p < 0.001) (Table 3.1‐2). 82
Table 3‐1 Clinical and histopathological characteristics of study eyes Histopathological Group
III
IV
(thin continous BLamD) (thick continuous BLamD)
95/53
37/22
Mean +/- SD
81 +/-8
84 +/- 6
Range
67 - 97
74 - 92
49/4
20/2
21 +/- 18
12 +/- 8
0 - 36
1 - 28
23 +/- 32
28 +/- 31
0 - 120
0 - 99
Mean +/- SD
0.276 +/-0.228
0.611 +/- 0.401
Range
0.000 - 1.000
0.000 - 1.300
Normal
50
0
Pigment Only
16
20
Drusen* Only
18
0
Drusen* + Pigment
11
17
Number of eyes examined by
EM
21
4
BLamD thickness (microns)
3.2 +/-1.6
10.7 +/- 4.2
Early Only
66
0
Patchy late
29
0
Continuous Late
0
37
Number of eyes/patients
Age at death (years)
Sex
M/F
Interval between last visit
and death (months)
Mean +/- SD
Range
Mean +/- SD
Total clinical follow up
Range
Visual Acuity (logMar)
Fundus Appearance
BLamD type
*Clinical drusen defined as intermediate or large drusen
83
Figure 3.1‐1 Basal Laminar Deposit 84
Figure 3.1‐1 Basal Laminar Deposit (A) Early type BLamD forming a blue‐staining continuous layer beneath the RPE (asterisk). Note it is less than half the height of an RPE cell. Fundus at age 67 had shown a few small soft drusen. Patient died aged 71. Bar marker 25μm. (B) Electron micrograph illustrating changes between RPE and choriocapillaris (CC). Early BLamD (bracket) lies internal to RPE basement membrane (vertical arrows) and comprises mostly banded material resembling long‐spacing collagen. Other phenotypes comprise a darker and denser material with an enveloping rim of pale fibrillar material. Between the clumps of BLamD lie membrane fragments (horizontal arrow) that also form a layer external to the basement membrane, the basal linear deposit (asterisk), and can be traced even into Bruch’s membrane. Bar marker 1μm. (C) Early type BLamD containing small hyalinised clumps of late type (arrows). These stain red with picro‐Mallory and are found close to the RPE. Fundus at age 75 had shown early focal hyperpigmentation related to small soft drusen, patient died age 76. Bar marker 20μm. (D) Further build up of late BLamD forms nodular excrescences on internal surface of early type, each nodule overlain by single RPE cell. Fundus had shown pigment changes and small soft distinct drusen 4 years before death at age 85.Bar marker 50μm 85
Figure 3.1‐2 Continuous Late Basal Laminar Deposit 86
Figure 3.1‐2. Continuous Late Basal Laminar Deposit (A) Eye of 79‐year‐old man photographed 2 years before death, showing ring of pigment clumps around foveal perimeter (arrow). (B). Section through fovea of eye illustrated in (A), showing thick layer of late type BLamD. Pigment clumps in fundus correspond to large, hyperpigmented RPE cells (arrow). Bar marker 200μm Picro‐Mallory stain. (C). Higher magnification of (B). A thick confluent layer of late BLamD lies on internal surface of early type. Bar marker 30μm Picro‐Mallory stain INSET. Electron micrograph of late type, showing amorphous structure apparently laid down in waves as grossly abnormal RPE retracts from Bruch’s membrane. Magnification x 1200. 87
Figure 3.1‐3 High magnification membranous debris Electron micrograph of membranous debris (approx. X31 000) in the eye of an 81‐year‐old woman with normal fundus and 20/30 vision. Higher magnification (inset) shows two dark laminae enclosing a single electron lucent lamina (arrow), giving rise to a ʺtrilaminarʺ appearance (approx. X 100 000). 88
Figure 3.1‐4 Basal mounds 89
Figure 3.1‐4 Basal mounds (A, B, C) Section through macula of clinically normal eye of 79‐year‐old man. Thin BLamD (arrowheads) contains several basal mounds that can be recognized histologically as unstained spaces (arrows). Bar markers A 60μm, B 50μm, C 50μm Picro‐Mallory stain (D) Semithin section from eye of 80‐year‐old man with basal mounds (asterisk). Basal linear deposit can be recognized as a narrow interval beneath the RPE and BLamD (arrow). Bar marker 30μm. Methylene blue and basic fuchsin. E) Electron micrograph of basal mound internal to RPE basement membrane (arrows). Patient aged 83, fundus had shown small drusen. Asterisk = BLamD. Bar marker 5μm 90
3.1.3.2
BLamD – clinical correlations Clinically, 41 of 66 eyes (62%) with early BLamD had a normal fundus at the last clinical examination. Eyes with early BLamD alone were 10.4 times (95% CI 4.4 – 24.6, p < 0.001) more likely to have a normal fundus than eyes showing any late BLamD, clinical pigment changes being present in all those in which late BLamD was continuous. Only 10 eyes had clinical pigment changes without evidence of late BLamD, and 4 of these eyes were subsequently reclassified as adult vitelliform lesions. Increasing formation of late BLamD was associated with poorer vision, the mean logMar visual acuity for eyes with continuous late BLamD being 0.335 units worse than the other eyes (95% CI 0.194 – 0.477, p < 0.001). Continuous late BLamD also occurred in eyes that were on average 3 years older (95% CI 0.3 – 6 years, p = 0.013) than eyes with early BLamD (Table 3.1‐2). 3.1.3.3
Membranous debris‐ localization and correlation with BLamD Coiled membrane fragments were found at several levels reflecting their presumed transit, i.e., blebbing from the basolateral RPE surface, scattered within early BLamD, traversing the RPE basement membrane to form BLinD, and in the inner and outer collagenous layers of Bruch’s membrane (Figure 3.1‐1B). BLinD was present in all eyes examined by EM and ranged from 2 to 11 layers in maximum thickness. A greater quantity of debris formed basal mounds and soft drusen (Figure 3.1‐5). Since basal mounds were found only in the presence of BLinD, and since they can be recognized in histological sections, the mounds were considered a surrogate marker for the presence of BLinD at the light microscopic level. Membranous debris accumulation was limited to basal mounds in 44 eyes while in 50 eyes it extended to soft drusen formation. Although basal mounds were seen in the absence of soft drusen, soft drusen never occurred in the absence of basal mounds. 91
Basal mounds increased in number as early type BLamD thickened, but then plateaued with the formation of late BLamD. On EM there was also loss of some of their membranous contents. Additionally, soft drusen in eyes with continuous late BLamD were predominantly of the granular type (12/15, 80%). This gradual reduction in membranous debris mirrored the progressive degeneration of the RPE and fallout of photoreceptors. 3.1.3.4
Membranous debris – clinical correlations Although histopathologically, soft drusen as small as 20μm were found, they never occurred in the absence of larger drusen visible clinically. Thus, histopathological soft drusen correlated very well with clinically observed intermediate and large drusen (r = 0.953, p < 0.001). Only in 4 eyes (2 patients) were soft drusen (intermediate‐sized) found on histological examination when the fundus was clinically normal. The interval between last examination and death was 5 and 3 years in these cases. There were also 5 eyes from 3 patients with multiple small drusen clinically, that were found to be soft drusen on histopathology. In group III, the fundus appeared normal in 28 of 38 eyes (74%) in which membranous debris was limited to BLinD, confirming that BLinD in conjunction with early BLamD represents threshold AMD. Basal mounds did not appear until early BLamD had thickened sufficiently, indicating that membranous debris and early BLamD continue to develop together. However, in 18 of 24 (75%) of Group III eyes with basal mounds alone, the fundus remained normal. It was not until patchy late BLamD appeared that the majority of fundi exhibited abnormality. In Group IV, all eyes had continuous late BLamD and there was little further increase in membranous debris. In eyes in which the debris remained confined to 92
basal mounds, the mean segment length of late‐continuous BLamD was 413 μm longer (95% CI 87 – 740μm, p = 0.015) than in eyes with soft drusen. These eyes belonged to patients who were on average 6 years older (95% CI 2 – 9 years, p< 0.001) with poorer visual acuities (mean difference 0.402 logMar units; 95% CI 0.167 – 0.637, p< 0.001). Table 3.1‐3 summarizes the clinical findings of eyes in groups III and IV according to the amount of membranous debris present. 93
Figure 3.1‐5 Membranous debris: basal mounds and soft drusen 94
Figure 3.1‐5 Membranous drusen (A) Semithin section from 75‐year‐old man, with clinical large drusen. Mounds of membranous debris (basal mounds) (m) are seen above BLamD (asterisk). Basal linear deposit has built up into soft drusen (d). Bar marker 100μm. Methylene blue and basic fuchsin. (B) Electron micrograph of subclinical soft druse from eye illustrated in (A) above, demonstrating membranous contents. Asterisk indicates late BLamD overlying early BLamD. Bar marker 5μm. (C) Fundus of fellow eye of same patient at age 71, showing soft drusen measuring up to 2 vein widths (250μm). 95
Table 3‐2 Basal laminar deposit – clinical and histopathological correlations Early Only
BLamD
Patchy Late
Continuous Late
Normal
62.1%
31.0%
0.0%
Pigment changes only
15.2%
20.7%
54.1%
Drusen* only
16.7%
24.1%
0.0%
Drusen* with pigment
changes
6.1%
24.1%
45.9%
80 +/-8
82 +/-8
84 +/-6
Mean Visual Acuity
(logMar)
0.275 +/- 0.242
0.280 +/- 0.197
0.611 +/- 0.401
RPE Abnormalities
Mild
Moderate
Severe
7.0 +/- 2.3
6.0 +/-0.0
5.8 +/- 2.1
1.7 +/- 2.5
2.7 +/- 1.1
3.1 +/- 1.9
Fundus Appearance
Clinical
Mean Age (years)
BLinD** (no. layers)
Pathological
Membranous Debris Basal Mounds
Soft Drusen
No. of Eyes
Membranous contents
66
Contents becoming increasingly granular
29
*Clinical drusen defined as intermediate or large drusen
**Demonstrated on EM
37
96
Table 3‐3 Membranous debris – clinical and histopathological correlations Membranous Debris
Group III: Thin
Continuous
BLamD
Group IV: Thick
Continuous
BLamD
Number of
eyes
BLamD
thickness
(microns)
BLamD
Fundus Appearance*
Normal
Pigment Only
Drusen
Drusen + Pigment
Normal
Early Only 12
Pigment Only
Patchy Late 12
Continuous Late 0 Drusen
Drusen + Pigment
Normal*
Early Only 16
Pigment Only
Patchy Late 17
Continuous Late 0 Drusen
Drusen + Pigment
Normal
Early Only 0
Pigment Only
Patchy Late 0
Continuous Late 20 Drusen
Drusen + Pigment
Normal
Early Only 0
Pigment Only
Patchy Late 0
Continuous Late 17 Drusen
Drusen + Pigment
BLinD only
38
Early Only 38
2.5 +/- 1.2 Patchy Late 0
Continuous Late 0
Basal mounds only
24
2.9 +/- 1.4
Basal mounds + soft
drusen
33
4.5 +/- 1.6
Basal mounds only
20
11.3 +/- 5.3
Basal mounds + soft
drusen
17
9.9 +/- 2.1
28
10
0
0
18
6
0
0
4
0
18
11
0
20
0
0
0
0
0
17
Mean Age
(years)
Mean VA
(logMar)
Approx
Snellen
Equivalent
80 +/-8
0.245 +/- 0.206
6/10
20/30
83 +/-6
0.273 +/- 0.114
6/12
20/40
81 +/-8
0.311 +/- 0.298
6/12
20/40
86 +/-5
0.796 +/- 0.362
6/37.5
20/125
81 +/-5
0.394 +/- 0.337
6/15
20/50
*Clinical drusen defined as intermediate or large drusen
** From 2 patients. Eyes were normal at last clinical examination. Found to have soft drusen on histopathological examination 3 and 5 years later
97
3.1.4 Discussion Zarbin proposed that the biological changes associated with tissue aging, whilst present in AMD eyes, do not inevitably lead to AMD 116. The present study was undertaken to trace the evolution of BLamD and membranous debris in order to determine their relationship to aging and AMD. The study found that the first appearance of BLinD coincides with a continuous layer of early BLamD. Since membranous debris is not found in eyes without BLamD (group I) and occurs only as occasional, isolated fragments in eyes with patchy BLamD (group II) 196, the presence of both BLinD and continuous early BLamD represents threshold early AMD, and can be considered a continuum of normal aging. Also described as “incipient AMD” 168, 176, the findings confirm that the majority of these eyes have a normal fundus and good vision. The appearance of late BLamD signals severe RPE abnormality, and corresponds to clinical pigment changes. Once produced, BLamD is remarkably resilient, persisting even in areas of geographic atrophy and in disciform scars 8. BLamD is not lost during processing, remaining detectable using routine H&E and PAS staining and, although not specific for AMD, it is found in all eyes with the disease. These properties make it a useful and reliable histopathologic marker for AMD. In terms of the disease process, however, BLamD appears inert. Rather, it is the degree of membranous debris accumulation that appears to influence the course of disease. In the current study, the number of basal mounds increased as BLamD progressively thickened but, once RPE cells lost their ability to support the overlying photoreceptors, membrane production ceased. This was heralded by the appearance of continuous late BLamD and a loss of the membranous contents of the mounds. Soft drusen likewise became less membranous and more granular, interpreted as the onset of drusen regression. 98
However, in some eyes membranous debris production did not progress beyond the formation of basal mounds. Even in the ninth decade these eyes show no histopathological evidence of soft drusen, instead developing pigment changes associated with poor vision. Continuous late BLamD is always present in these eyes, and in longer segments than in eyes with soft drusen, suggesting longstanding RPE dysfunction. By contrast, soft drusen can occur in relatively young eyes with good vision, prior to the appearance of continuous late BLamD. Clinically, intermediate and large drusen are known to increase in size, number and confluence comparatively rapidly 178, implying a rapid outpouring of membranes. However, the resulting thickening of BLinD is not uniform, causing it to develop undulations, the larger of which become visible as soft drusen (Figure 3.1‐4). In these eyes, soft drusen do not appear to develop from an earlier stage when only basal mounds are present, i.e. soft drusen and basal mounds appear concurrently. Epidemiologic data support the concept of two pathways leading to the development of advanced AMD, depending on the amount of accumulated membranous debris. Eyes with a large amount of membranous debris, in the form of large drusen, are at highest risk of developing advanced AMD over a 5‐ or 10‐
year period 33, 320, particularly the blinding effects of CNV 37, 39, 321, 322. Eyes with pigment changes alone are at lower risk of developing advanced AMD over the same period 33, 320, consistent with an alternate and slower course of disease. This clinicopathologic study represents an overview of the basal deposits in the evolution of AMD, but makes no attempt to correlate with the clinical fundus grades or severity scales used in recent epidemiologic studies. It has two obvious limitations. Firstly, the sometimes long interval between the last examination and death, with drusen a particular concern since the clinical appearance can change relatively rapidly. With a few exceptions, however, there was good correlation 99
between histopathological soft drusen and clinical intermediate and large drusen, and the main study findings were unchanged by the exclusion of eyes with an interval longer than 36 months. Secondly, differences in the post‐fixation method for EM prevented comparison of membranous debris with other recent studies of this material 182. It should thus be emphasized that the terms “BLinD”, “basal mounds” and “soft drusen” are pathological descriptors of membranous debris location, and that the biochemical nature of this material has not been completely defined. BLamD and membranous debris may be products of two distinct cell‐survival strategies of RPE under stress. Early BLamD can be thought of as excess basement membrane secreted by the RPE, a common strategy employed by cells attempting to recover from injury, allowing them to remain attached to a tissue’s “scaffolding”. As BLamD thickens it progressively separates the basal RPE surface from its original basement membrane and choroidal blood supply, exacerbating the metabolic insufficiency caused by decreasing Bruch’s membrane permeability 148, 150, 151, 323. The production of late type BLamD signals a critical point in RPE damage at which the basement membrane material becomes more condensed and, once this forms continuous segments, the RPE cells become hyperpigmented, enlarge, lose their microvilli and round off. These phenotypic changes are initially accompanied by expression of the cytoskeletal protein vimentin 324. The RPE at this stage is severely compromised and can no longer support the photoreceptors. Eventually, affected RPE cells lose their anchoring to both the basement membrane and to adjacent cells and are shed into the sub‐
retinal space 3. The production of membranous debris, on the other hand, may reflect another survival mechanism adopted by RPE under stress. Ultrastructurally, the earliest membranes appear to bleb from the basolateral surface of the RPE. Non‐apoptotic 100
membrane blebbing has been observed in cultured RPE cells subjected to oxidative stress 201, and is considered a non‐specific cellular response to ischaemic, oxidative or other injury which allows the expulsion of damaged cellular constituents 199, 200. Other studies show that membranous debris contains solid lipid particles 182, and that dysregulated lipid trafficking by abnormal RPE may also contribute to its formation 185, 325. Since the appearance of BLinD coincides with a continuous layer of early BLamD, a threshold level of RPE injury (perhaps ischaemic) may trigger the production of membranous debris. In summary, late BLamD occurs in eyes with clinical pigment changes and indicates a severely compromised RPE. It is membranous debris, however, that appears to influence the course of the disease (Table 3.1‐4). Eyes in which membranous debris accumulation is limited to basal mounds present with pigment abnormalities alone. These eyes eventually develop “primary” or “drusen‐unrelated” 3, 8 geographic atrophy. If there is a more rapid accumulation of debris, early AMD presents with intermediate or large drusen, is at greater risk of CNV 37, 39, 321, 322 and earlier vision loss, and eventually progresses to “drusen‐
related” geographic atrophy 3. These two pathways share a common threshold stage when both BLinD and a continuous layer of early BLamD are present (Figure 3.1‐6). However, since the majority of these eyes will be normal in terms of fundus appearance and visual acuity, identifying “at‐risk” eyes will come to rely on more sophisticated clinical tests, possibly combined with genetic screening. 101
Figure 3.1‐6 Influence of basal deposits on the progression of AMD None or patchy early BLamD
Normal Aging
No BLinD
Cont thin early BLamD
Threshold early AMD
Fundus normal 74%
BLinD
Excess production MD
MD production remains limited
Older group, poorer vision
(High risk CNV)
Intermediate or large drusen
Continuous early BLamD
+/- Patchy late BLamD
Basal mounds
Continuous early BLamD
+/- Patchy late BLamD
Basal mounds
BLinD + Soft drusen
BLinD + Soft drusen
Continuous late BLamD
Basal mounds
Continuous late BLamD
Basal mounds
BLinD + Granular drusen
BLinD
Drusen-related
Geographic atrophy
Drusen-unrelated
Geographic atrophy
Intermediate or large drusen,
may show signs of regression
Pigment changes (100%)
Pigment changes (25%)
Pigment changes (100%)
= RPE basement membrane.
Continuous early BLamD together with BLinD represent the threshold for onset of early AMD. The pathway then followed depends on the degree of membranous debris production. A large build up manifests as intermediate or large drusen, whereas if limited to BLinD and basal mounds, the fundus initially remains normal. At the same time, BLamD reflects the degree of RPE damage, clinical pigment abnormalities becoming apparent when late BLamD appears. Both pathways then progress towards geographic atrophy. MD – membranous debris
102
Table 3‐4 Basal laminar deposit ‐ a summary BLamD Type
Early
Late
Site
Basement membrane‐like extracelluar material between RPE basal plasma membrane and its BM
Being a later development, lies above (internal to) the early form. Occurs in 3 stages in order of severity:
LM Appearance
EM Appearance
Clinical Correlation
Significance
Stains blue on P‐M. Shows feint vertical stirations and may contain small unstained spaces. May be absent, patchy or continous. When continuous, may be associated with early AMD. Reaches maximum height of 7mm. Exists in 3 forms: fibrillar, amorphous (light and dark) and polymerised (banded). The polymerised form is similar to long‐spacing collagen. The light amorphous form is continuous with the RPE BM.
All late BLamD stains red on P‐M. Late BLamD has a more compact appearance than early BLamD. It has a wave‐
like appearance, described previously as ʺflocculentʺ or ʺmultilaminarʺ.
(i) Inclusions of late BLamD: small, globular bodies in the order of 4mm lying within early BLamD.
(i) Inclusions are seen beneath a segment or portion of RPE plasma membrane which becomes indented. May appear within membranous debris.
Not seen. May be compatible with a normal fundus.
(ii) Nodular late BLamD: are nodular excrecenses up to 7mm (1 RPE cell) in size. The overlying RPE cell becomes hypertrophied and hyperpigmented initially, and eventually attenuates.
(ii) Appear multilaminar. Distinct from hard drusen.
May be compatible with a normal fundus or seen as focal hyperpigmentation.
Results from contiued production of late BLamD by a single RPE cell. (iii) Produced by several RPE cells. Not compatible with a normal fundus. Likely to have pigmentary changes without soft drusen.Usually seen as hyperpigmentation, followed by hypopigmentation and atrophy.
A sign of severe RPE dysfunction. Pre‐atrophy.
(iii) Continous (confluent) BLamD: also known as ʺdiffuse thickening of the internal aspect of Bruchʹs membraneʺ. Seen as a confluent layer internal to early BLamD, up to 7mm in thickness.
Not seen. May be compatible with a normal fundus.
The presence of fibronectin, laminin and glycoproteins imply it is abnormal basement membrane material. Considered inert, but reflects the escalating damage to RPE.
103
Table 3‐5 Membranous debris – a summary Relationship to RPE BM
Entity
Clinical Correlation
Basal Membrane Mounds
Rounded collections of membranous debris between the RPE basal plasma membrane and its BM. Small, unstained spaces with rounded margins between attenuated RPE cells internally, and early BLamD externally. RPE cells overlying mounds are grossly abnormal in shaped becoming attenuated over mounds but attached by their tight junctions. The smallest mound is at least half the height of an RPE cell. Mounds beneach adjacent cells may fuse, the largest up to 7 RPE cells wide and one RPE cell in height.
**
Patients with basal mounds less likely to have normal fundus. Increasing numbers of mounds correlates with poorer visual acuity. If over 30mm in diameter, may account for dot‐
like drusen.
Membrane Fragments
Membrano‐
vesicular material apparently transiting through RPE BM
Not seen
Vesicles are smaller and more compact than at other sites.
Not seen. Compatible with a normal fundus
Basal Linear Deposit
Forms a layer lying between the RPE BM and the inner collagenous zone of Bruchʹs membrane
Not seen in histological sections. In semithin sections, appears as an unstained narrow interval between the RPE and Bruchʹs membrane.
A layer of membranous debris, thickest when adjacent to soft drusen. Not seen. May be compatible with a normal fundus.
Soft Drusen
Focal accumulation of BLinD between the RPE BM and the inner collagenous zone of Bruchʹs membrane
Seen between the RPE and Bruchʹs membrane. May be all sizes , the smallest defined as at least half the height of an RPE cell. Regressing drusen may become granular.
Composed of membranous debris. Soft Drusen
Membrane Fragments
Apical
Not seen
EM Appearance
Not seen. Compatible with a normal fundus
Internal
External
LM Appearance
Bleb from basal surface of RPE cells. Found between strands of early BLamD
Membrane Fragments
Within
Site
Membrano‐
vesicular material between RPE basal plasma membrane and its BM
Apical Membrane Mounds
Withing the outer collagenous zone of Bruchʹs membrane
Not seen.
Not seen.
Found at apex of RPE, within the subretinal space
Only seen in well‐preserved eyes as empty spaces between shortened photoreceptor outer segments and apical surface of RPE
If over 30mm may account for dot‐
like drusen in patients with pigment clumps or adult vitelliform degeneration.
104
Mounds of debris overlying RPE cells which have lost their microvilli.
Significance
Forms a cleavage plane in which new vessels can spread.
Contribute to the increasing thickness of Bruchʹs membrane seen with age. May attract macrophages.
3.2 CHOROIDAL
AND
BRUCH’S
EARLY AND ADVANCED
MEMBRANE MACROPHAGES IN
AMD
3.2.1 Introduction Macrophages were first implicated in the pathogenesis of age‐related macular degeneration (AMD) when early morphological studies found them within drusen 179, 215, 218
and choroidal neovascular membranes 219. Significantly, macrophages were also found adjacent to breaks in Bruch’s membrane (BrM), through which subclinical new vessels grew from the choroid 10, 179, 222. Later studies supported the idea that macrophages facilitated choroidal neovascularisation (CNV) by interrupting BrM and releasing pro‐angiogenic factors such as vascular endothelium‐derived growth factor (VEGF) 121, 174, 226, 227, 229, 326, 327. Thus, macrophage infiltration of BrM or the subretinal space is considered pathological. However, another local population of macrophages can be found residing in the choroid of normal eyes. Extensive networks of resident choroidal macrophages have been demonstrated in normal mouse 244 and rat 328 eyes. Although their human equivalents are not well characterised, this population of macrophages appear to be programmed for phagocytosis without eliciting inflammation. Indeed, recent animal models of AMD suggests that the normal recruitment and turnover of resident choroidal macrophages may prevent the development of AMD‐like lesions 231, 233, presumably by facilitating the efficient removal of BrM debris. 105
It is not clear, however, how these two populations of macrophages – resident choroidal macrophages and macrophages that infiltrate BrM – are influenced by the progressive accumulation of the characteristic basal deposits found in the macula of human AMD eyes – basal laminar deposit (BLamD) and membranous debris. BLamD, an abnormal basement membrane‐like material, is found between the plasma membrane of the retinal pigment epithelium (RPE) and its basement membrane. Its accumulation correlates with progressive RPE abnormalities 5 and it is found in all eyes with early or advanced AMD 8, 126, 169, 174‐176. Membranous debris appears to be composed of membrane fragments 179, 180 and lipid 142, 181, and is specific for AMD 176. It is first found in eyes with early AMD as basal linear deposit (BLinD), a thin layer external to the RPE basement membrane 126, 176, 179. Membranous debris can also accumulate as small focal collections internal to the RPE basement membrane, forming basal mounds 329. Larger collections external to the RPE basement membrane become soft drusen 176. Macrophages are a highly heterogenous group of cells. Early inflammatory macrophages, for example, are phenotypically and functionally distinct from resident tissue macrophages. Pro‐inflammatory (“classically activated”) macrophages recruited to sites of injury or infection express inducible nitric oxide 106
synthase (iNOS), whereas resident macrophages do not 330. This enzyme allows macrophages to convert L‐arginine to nitric oxide, a damaging reactive nitrogen species with cytotoxic 331, 332 and pro‐angiogenic properties 220, 248. In the context of AMD, we hypothesise that: (i) resident choroidal macrophages do not express iNOS in normal eyes; (ii) the progressive accumulation of BLamD or membranous debris attracts macrophages to BrM and may favour the recruitment of classically activated, iNOS‐ expressing macrophages; (iii) the presence of BrM breaks and subclinical CNV is associated with a high BrM macrophage count; and (iv) iNOS‐
expressing macrophages are likely to be found where active neovascular processes are taking place. In this study, macrophages were identified using CD68 immunohistochemistry in a total of 16 human eyes, representing normal (2), normal aged (2), early AMD (7), geographic atrophy (2) and disciform scars (2). iNOS immunohistochemistry was used to determine whether macrophages were classically activated in these eyes. A submacular BrM macrophage count was undertaken in a further 125 eyes, representing normal (18), normal aged (20), early AMD (78) and geographic atrophy (9). CD68/iNOS immunohistochemistry and BrM macrophage counts were correlated with BLamD and membranous debris accumulation, and with subclinical CNV. Results were also correlated with clinical data, including age, visual acuity and fundus appearance. 107
3.2.2 Methods 3.2.2.1
Eyes Archival slides of human eyes were selected from a large clinicopathologic collection of over 600. Eyes were previously graded according to the appearance of BLamD 10 (see below). All of the eyes were examined (best corrected visual acuity and direct fundoscopy) and most photographed (Zeiss 30° fundus camera) during life. Eyes were paraffin embedded with 8μm serial sections cut horizontally through the disc and macula, and every 10th section stained using the picro‐
Mallory method 10. For histopathology and BrM macrophage counts, the best sub‐foveal sections of 125 eyes, representing histopathological groups I to V, were examined with a Leitz Dialux microscope, using the PL APO 40x (NA‐0.75) objective. For CD68 and iNOS immunohistochemistry, unstained sections of 16 eyes, belonging to histopathological groups I to VI, were used. Informed consent was obtained before the collection of all eyes. The study was carried out in keeping with the Declaration of Helsinki and was approved by the Human Research Ethics Committee of the University of NSW. 3.2.2.2
Clinical definitions Clinical data from the last follow up prior to death was used. Best corrected visual acuity was converted from Snellen to logMar for statistical purposes. The fundus was graded as: (i) normal, which included up to 5 small (<63mm) hard drusen; (ii) extensive small hard drusen; (iii) soft drusen (≥63μm) without pigment abnormalities; (iv) soft drusen (≥63μm) with pigment abnormalities; and (v) pigment abnormalities alone. 108
3.2.2.3
Histopathological definitions and grading: BLamD Eyes were graded based on the extent of submacular BLamD present, according to the system proposed by Sarks 10, modified by the inclusion of BLamD type. Early type BLamD stains blue with the picro‐Mallory method (Figure 3.2‐2a), while the late type stains red and is found internal to the early type 329 (Figure 3.2‐2b). Group I represented normal eyes and had no BLamD. Group II represented normal aged eyes in which there was patchy early BLamD. Groups III and IV represented early AMD. In group III, BLamD was present in a thin continuous layer, and consisted of early or patchy late type. In group IV, BLamD was present in a thick continuous layer, and consisted of continuous segments (>250μm) of the late type. Group V eyes had geographic atrophy and Group VI eyes had disciform scars. 3.2.2.4
Histopathological definitions and grading: membranous debris Due to the exclusive use of light microscopy in this study, the presence of BLinD, an ulstrastructural finding, was not determined. However, BLinD is found in all eyes in which there is a thin continuous layer of BLamD 329. Since soft drusen are large focal collections of membranous debris, and basal mounds are small focal collections, in eyes with early AMD (groups III and IV), the amount of submacular membranous debris was thus graded: soft drusen > basal mounds > no basal mounds (presumed BLinD). 3.2.2.5
BrM macrophage counts BrM macrophages were defined morphologically by their abundant pale pink‐
staining cytoplasm, large oval or kidney‐shaped nuclei and granular chromatin, and by their presence within the zone bounded by the inner margin of the inner collagenous zone and the outer margin of the choroidal capillaries/intercapillary pillars (Figure 3.2‐1b). Counts were not performed on group VI eyes, since the 109
large numbers of macrophages, fibroblasts and other cells within disciform scars is well established. An eyepiece graticule which equalled 280μm in total with the 40X objective was used to examine the foveal field and one graticule length on either side, giving a total of three fields (840μm). In eyes with hard drusen, soft drusen or RPE changes, additional fields (up to five, all within the inner macula) were examined. Cell counts were divided by the number of fields examined, multiplied by three, and expressed as mean count per three 40X fields. 3.2.2.6
Immunohistochemistry CD68 is a glycosylated protein (part of the scavenger receptor superfamily) found in all cells from the monocyte/macrophage lineage 333. CD68 immunohistochemistry is commonly used to identify macrophages/histiocytes in paraffin sections, and stains intracytoplasmic granules. Detailed methods for CD68 and iNOS immunohistochemistry, and the antibodies use, as described in the Appendices (A‐C). Briefly, sections were deparaffinised in xylene, rehydrated and treated in 1% hydrogen peroxide in alcohol for 30 min to block endogenous peroxidase. Non‐specific binding was blocked by incubation in 10% normal horse serum for 20 min. For iNOS detection, sections were incubated with monoclonal mouse anti‐human iNOS (Transduction Laboratories BD Biosciences, San Jose, CA, USA), diluted 1:400, for 1hr at 37oC. . For CD68 detection, antigen retrieval with pepsin digestion for 5 min at 37oC preceded the blocking step, which was followed by incubation with monoclonal mouse anti‐
human CD68 antibody (Zymed Invitrogen, Carlsbad, CA, USA), neat, for 1hr at 37oC. After washing with tris, sections were incubated with biotinylated anti‐
110
mouse IgG antibody (Vector Laboratories, Burlingame, CA, USA), diluted 1:200, for 30min at 37oC. Sections were then washed with tris, incubated with 1:500 ABC (Vector Laboratories) for 30 min at room temperature. For iNOS detection, this was followed by 3 changes of nickel ammonium phosphate tris solution and final incubation with DAB (Sigma‐Aldrich, St Louis, MO, USA) in nickel ammonium sulphate and peroxide solution for 10 min at room temperature, resulting in a dark grey colour product. For CD68 detection, incubation with ABC was followed by washing with tris and incubation with Vector SC Blue (Vector Laboratories) for 10 min at room temperature, resulting in a dark blue colour product. All sections were then washed with tris followed by distilled water, dehydrated and coverslipped in DPX (ProSciTech, Kirwan, QLD, Australia). 3.2.2.7
Statistics Statistical analysis was performed using SPSS for Windows (v15.0; SPSS Inc, Chicago, IL, USA). The student t test was used to compare mean BrM macrophage counts and visual acuity measurements between histopathological groups. Contingency tables of histopathological group versus the presence of subclinical CNV were evaluated by the χ2 and Fisher’s exact tests. A p value ≤ 0.05 was considered statistically significant. 111
3.2.3 Results 3.2.3.1
Groups I & II‐ normal and normal aged eyes The fundus was normal in all 18 eyes from group I (Table 3‐2.1). Histopathologically, the RPE appeared normal and BrM hyalinisation ranged from non‐existent (Figure 3.2‐a) to extending a short distance down the intercapillary pillars. Group II eyes were also fundoscopically normal, with the exception of two eyes with pigment abnormalities in which extensive hard drusen of intermediate size were present (Table 3.2‐1). BrM hyalinisation extended down the intercapillary pillars to reach the base of the choroidal capillaries, resulting in widened intercapillary spaces (Figure 3.2‐1b). BrM macrophages were absent in group I and II eyes (confirmed by CD68 immunohistochemistry), except in the two group II eyes with pigment changes, in which occasional macrophages were found beneath small hard drusen. Occassionally, cells with small nuclei and scanty cytoplasm were found in the place of choroidal capillaries (Figure 3.2‐1b). CD68 immunohistochemistry revealed numerous ramified cells in the choroid, within the choroidal melanocyte network, in all group I and II eyes (Figures 3.2‐1c & d). Expression of iNOS was not found in either BrM or choroidal macrophages, but was occasionally seen in the basal aspect of RPE cells (Figure 3.2‐1e). 3.2.3.2
Groups III & IV– early AMD In group III eyes, the fundus was normal in 15 (79%) eyes without basal mounds (presumed BLinD), and 6 (40%) eyes with basal mounds, while all group IV eyes had clinically evident pigment changes, with or without soft drusen. Resident choroidal macrophages ‐ CD68‐positive ramified cells among the choroidal melanocytes‐were found in all group III and IV eyes (Table 3.2‐2). BrM 112
macrophages were also seen in group III and IV eyes, and were found in the place of a choroidal capillary (Figure 3.2‐2a & c), or an intercapillary pillar (Figure 3.2‐
2b). Interestingly, BrM macrophages were only found in group III eyes with clinical pigment changes or soft drusen, apart from three eyes with normal fundus (which belonged to older patients). iNOS expression was again seen in the basal aspect of RPE cells. However, choroidal ramified cells also expressed iNOS (Figure 3.2‐2d) in group III and IV eyes, as did choroidal vascular endothelial cells (3.2‐2e), and perivascular ramified cells (3.2‐2f). 113
Table 3‐6 Clinical and histopathological features of study eyes Number of Eyes
Light
Immunohistomicroscopy
chemistry
Age at
Death (yrs)
logMar VA
Fundus appearance
Drusen*
Pigment
alone
alone
Group
Pathology
I
No BLamD
18
2
64.5 +/-9.6
0.080 +/0.105
18
(100%)
0
0
0
II
Patchy BLamD
20
3
74.4 +/-8.9
0.110 +/0.120
18
(90%)
0
2
(10%)
0
III
Thin continuous
BLamD
49
4
81.8 +/-7.4
0.277 +/0.213
21
(43%)
13
(27%)
10
(20%)
5
(10%)
IV
Thick continuous
BLamD
29
3
84.1 +/-6.3
0.593 +/1.164
0
0
17
(59%)
12
(41%)
V
Geographic
atrophy
9
2
80.9 +/-5.9
1.096 +/1.004
0
-
-
-
VI
Disciform scar
0
2
78.5 +/-2.1
2.000 +/0.000
0
-
-
-
Normal
Drusen* +
Pigment
BLamD basal laminar deposit; VA visual acuity
* intermediate or large drusen
114
Table 3‐7 BLamD, membranous debris and macrophages in early AMD Group
Pathology
Bruch's membrane macrophages
Choroidal macrophages
Cell count*
CD68 immunohistochemistry
iNOS immunohistochemistry
CD68 immunohistochemistry
iNOS immunohistochemistry
I
No BLamD
0
None
None
[occasional basal RPE expresson]
Ramified cells intermixed with
chroidal melanocytes
None
II
Patchy BLamD
0.03 +/-0.14
None
None
[basal RPE expresson]
Ramified cells intermixed with
chroidal melanocytes
None
III
Thin continuous
BLamD
3.14 +/-5.82
Round to epitheliod cells
within BrM
None
[basal RPE expresson]
Ramified cells intermixed with
chroidal melanocytes
Occassional ramified cells intermixed
with choroidal melanocytes
IV
Thick continuous
BLamD
7.14 +/-5.90
Round to epitheliod cells
within BrM
None
[basal RPE expresson]
Ramified cells intermixed with
chroidal melanocytes
[some large & plump]
Frequent ramified cells intermixed
with choroidal melanocytes, or in
perivascular region
[choroidal
endothelial cell expression]
V
Geographic
atrophy
2.72 +/-3.03
At edges of atrophy
None
Few cells beneath area of atrophy
None
VI
Disciform scar
-
Numerous, within fibrovascular
membrane/scar in both active CNV
and disciform scar
Little within CNV or disciform scar
Numerous, within fibrovascular
membrane/scar in both active CNV
and disciform scar
Perivascular ramified cells and
endothelial cell iNOS expression in
active CNV. Little in disciform scar.
BLamD basal laminar deposit; BrM Bruch's membrane
* Mean macrophage count per 3 (X40) fields
115
Figure 3.2‐1 Morphological and immunohistochemical features of normal (group I) and normal aged (group II) eyes ONL
a
RPE
CC
RPE
b
CC
*
c
d
RPE
CC
e
116
Figure 3.2‐1 Morphological and immunohistochemical features of normal (group I) and normal aged (group II) eyes (a) A group I eye from a 73 year old patient with a normal fundus. The RPE is normal, there is no BLamD and little BrM hyalinisation. (b) In group II eye from a 72 year old patient with a normal fundus, BrM hyalinisation extends down the intercapillary pillars (black arrows). Fibrous tissue (asterix) or small round cells (white arrow) appear in the place of lost choroidal capillaries. The black arrows also define the area within which BrM macrophage were counted. (c) Ramified cells with CD68 positive granules in the choroid of a group I eye from a 62 year old patient with a normal fundus, among choroidal melanocytes. (d) A group I eye from the same patient as (a), with numerous CD68‐positive choroidal cells intermixed with melanocytes (black arrows). An intravascular monocytoid CD68 positive cell is seen within a choroidal vessel (white arrow). (e) Occassionally, iNOS expression was seen in the basal aspect of RPE cells (arrows), such as in this group II eye from a 75 year old patient with a normal fundus. Neither BrM or choroidal cell iNOS expression was found in group I and II eyes. a‐b: picro‐Mallory stain; c‐e: CD68 and iNOS immunohistochemistry. The CD68 colour product is dark blue and the iNOS colour product is dark grey. Bar markers: approx. 100μm ONL – outer nuclear layer; RPE – retinal pigment epithelium; CC – choriocapillaris; iNOS – inducible nitric oxide synthase; BrM – Bruch’s membrane 117
Figure 3.2‐2 Morphological and immunohistochemical features of eyes with early AMD (groups III and IV) a
RPE
*
CC
b
RPE
CC
c
RPE
CC
d
RPE
CC
e
RPE
CC
f
118
Figure 3.2‐2 Morphological and immunohistochemical features of eyes with early AMD (groups III and IV) (a) BrM macrophages (black arrows) in the place of a lost choroidal capillary or eroding intercapillary pillars in a group III eye from an 88 year old patient with pigment changes and soft drusen. There is mild RPE disorganisation and thin continuous BLamD of the early (pale blue staining) type (square brackets). A soft druse (asterix) appears empty due to loss of lipid contents during processing. A pigment laden cell (white arrow) is also found replacing a choroidal capillary. (b) A group IV eye from an 89 year old patient with pigment changes only. The RPE appears more disorganised compared to the eye in (a) and there is thick continuous BLamD, with continuous segments of the late, red staining type (square brackets) type overlying the early type. A BrM macrophage is found apparently eroding an intercapillary pillar (arrow). (c) CD68 immunohistochemistry in a group IV eye from an 80 year old patient with soft drusen and pigment changes, confirming the presence of BrM macrophages (arrows). Note again the network of CD68 positive ramified cells in the choroid. (d) Ramified iNOS expressing cells in the choroid (white arrow), intermixed with melanocytes, in the same eye as in (c). Basal expression of iNOS by the RPE cells (black arrows) was also found in group III and IV eyes. (e) Choroidal vascular endothelial cell iNOS expression (arrow) in a group IV eye from a 74 year old patient with soft drusen and pigment changes. (f) iNOS expression by ramified cells found between choroidal melanocytes (white arrow) and near a choroidal vessel in the choroid (arrows) of the same eye as in (c). Bar marker: approx. 100μm 119
3.2.3.3
Groups III & IV: BrM macrophages and subclinical CNV Sixteen of the 125 eyes examined histologically had subclinical CNV. With the exception of two eyes from group V, the remainder belonged to groups III and IV. Overall, eyes with subclinical CNV had significantly more macrophages on average (p <0.001; mean difference 10.30 +/‐1.93; 95% CI = 6.19‐14.41) than eyes without (Table 3.2‐4). Macrophages in eyes with subclinical CNV were found eroding intercapillary pillars or BrM (Figure 3.2‐3a). Growth of fibrovascular tissue into the sub‐RPE space was seen in eyes with complete breaks in BrM (Figure 3.2‐3b). A lack of unstained tissue in these eyes prevented immunohistochemical characterisation of the macrophages. Subclinical CNV was associated with the amount of membranous debris present in eyes with thin continuous BLamD (group III), since it was only found in eyes with soft drusen (Table 3.2‐3). However, this relationship did not hold once thick continuous late BLamD was present (group IV) (Table 3.2‐3). Of the 23 pairs of eyes present in the study, when subclinical CNV was present in one eye, greater numbers of BrM macrophages were also found in the fellow eye (mean difference 9.18 +/‐2.52; 95% CI = 3.96‐14.41) compared to the fellow eyes of eyes without subclinical CNV (Table 3.2‐4). 120
Figure 3.2‐3 Morphological features of eyes with subclinical CNV a
RPE
CC
*
b
RPE
CC
*
121
Figure 3.2‐3 Morphological features of eyes with subclinical CNV (a) A group III eye from an 88 year old patient with soft drusen and pigment changes. Thin continuous BLamD of the early type (asterix) is overlain by mildly disorganised RPE. BrM macrophages are seen eroding BrM (arrow) beneath a soft druse (contents emptied by processing). Elsewhere in the same eye, fibrovascular tissue is found invading the sub‐
RPE space (not shown). (b) A group III eye from an 81 year old patient with soft drusen only. Fibrovascular tissue is seen invading the sub‐RPE space (asterix) via a focal break in BrM (arrow). Bar marker: approx. 100μm 122
Table 3‐8 BrM macrophages, membranous debris and subclinical CNV Group
BLamD
Extent
III
Thin
continuous
Type
Early +
Patchy late
Membranous
Debris
No. eyes
Age (yrs) logMar VA
BrM
macrophage
count**
p
Subclinical
CNV
No basal
mounds*
19
80 +/-8
0.185 +/0.179
0.8 +/-2.5
0
Basal mounds
only
15
84 +/-5
0.262 +/0.940
0.5 +/- 1.0
0
0.001
IV
Thick
continuous
Continuous
late
0.017
Soft Drusen
15
82 +/-8
0.374 +/0.272
8.7 +/-7.7
6
Basal mounds
only
17
87 +/-5
0.935 +/0.160
8.0 +/-6.6
5
80 +/-5
0.250 +/0.117
0.359
Soft Drusen
12
5.9 +/-4.8
BLamD basal laminar deposit ; VA visual acuity ; CNV choroidal neovascularisation
*presumed BLinD ** Mean macrophage count per 3 (X40) fields
***Fisher's exact p
123
p***
0.354
1
Table 3‐9 BrM macrophages and subclinical CNV in the fellow eye Number of eyes
BrM macrophage
count**
Subclinical CNV absent*
64
2.78 +/-4.07
Subclinical CNV present*
14
13.08 +/-6.97
Fellow eye subclinical CNV absent
20
4.24 +/-4.18
Fellow eye subclinical CNV present
3
13.42 +/-2.81
p
<0.001
0.002
CNV choroidal neovascularisation
* all eyes from groups III and IV
** Mean macrophage count per 3 (X40) fields
124
3.2.3.4
Group V – Geographic atrophy Compared to groups I – IV, the resident choroidal macrophage network appeared sparse immediately beneath areas of atrophy. BrM macrophages however, were found at the edges of areas of atrophy in group V eyes, aligned with the still‐viable RPE cells (Figure 3.2‐4a). iNOS expression was not found in BrM or the choroid of these eyes (Figure 3.2‐4b). 3.2.3.5
Group VI – Disciform scarring Immunohistochemistry performed on the 2 eyes with disciform scarring showed differences in CD68 and iNOS staining between active and inactive scars. In a clinically flat disciform lesion without haemorrhage or exudates, subretinal fibrous tissue surrounding a large feeder vessel was present. Numerous CD68 positive cells were present within the scar (Figure 3.2‐4c) but iNOS staining was minimal (Figure 3.2‐4d). By contrast, in an exudative disciform lesion, a fibrovascular membrane consisting of numerous small vessels and cells, some pigmented, was seen. CD68 positive cells were less numerous in this eye compared to the inactive scar (Figure 3.2‐4e), while plump ramified CD68 positive cells were noted in the choroid. However, large iNOS positive cells were found in the perivascular region of choroidal vessels (Figure 3.2‐4f). Choroidal endothelial cells in this region were also iNOS positive (Figure 3.2‐4f). 125
Figure 3.2‐4 CD68 and iNOS Immunohistochemistry in eyes with geographic atrophy (group V) or disciform scarring (group VI) RPE
CC
RPE
CC
126
Figure 3‐2.4 CD68 and iNOS Immunohistochemistry in eyes with geographic atrophy (group V) or disciform scarring (group VI) (a) CD68 positive cells were seen at the edges of areas of atrophy, in line with the remnant RPE (arrow) in a group V eye from an 84 year old man. Few ramified CD68 positive cells were found in the choroid immediately beneath the atrophic region. (b) The same eye in (a). iNOS expressing cells were not found within BrM or choroid. (c) A group VI eye from an 80 year old man with an “inactive” disciform scar. Numerous plump CD68 positive cells are present within the scar (arrows). (d) The same eye in (c). iNOS positive cells are absent. (e) A group VI eye from 76 year old man with exudative neovascular AMD, in which strongly CD68 positive cells are found within the scar (arrows) and in the choroid, in perivascular locations (arrowhead). (f) The same eye in (e). Choroidal perivascular cells (arrows) and endothelial cells express iNOS. Bar markers: approx. 100μm 127
3.2.4 Discussion The relationship between macrophages and the development of lesions specific to age‐related macular degeneration, particularly choroidal neovascularisation, is likely to be complex. Emerging evidence suggests that normal recruitment and turnover of the resident choroidal macrophage population may, in fact, protect against the development of AMD‐like lesions 231, 233. The current study suggests that resident choroidal macrophages are present in all eyes and, since they do not express iNOS, are phenotypically distinct from classically activated, inflammatory macrophages. The choroidal immune microenvironment appears to be tightly controlled. RPE cells express macrophage chemoattractant protein‐1 334, which facilitates macrophage recruitment. However, RPE cells also express immunomodulatory molecules such as complement factor H 335, transforming growth factor‐β1 162, interleukin‐1 receptor antagonist 163, pigment epithelium‐derived factor, somatostatin and interleukin‐10 164, preventing newly recruited macrophages from developing an inflammatory program. RPE cells can also suppress activated T cells 336 and initiate Fas‐Fas ligand mediated apoptosis of infiltrating immune cells 157. BrM macrophages were not found in significant numbers until eyes developed at least a thin, continuous layer of BLamD, together with a large amount of membranous debris (soft drusen). This suggests that while the RPE is relatively intact, local control of macrophage recruitment and programming is not significantly altered until a large amount of extracellular membranous/lipid material accumulates. However, once moderate to severe RPE abnormalities develop (thick continuous late BLamD), BrM macrophages increase irrespective of the amount of membranous debris present. Surprisingly, iNOS expression was not a feature of BrM macrophages in any eye. This suggests that although some 128
immunomodulatory capacity is lost once RPE are compromised, the mechanisms that prevent infiltration of BrM by classically activated macrophages appear robust. The same does not appear to hold for resident choroidal, or newly recruited choroidal macrophages, evidenced by their expression of iNOS once RPE abnormalities (continuous BLamD) and membranous debris are present. The presence of extracellular debris and dead or compromised cells (RPE) attracts components of the innate immune system, including acute phase proteins, natural autoantiabodies and complement components 337. Complement components and other acute phase proteins have been found in drusen 263 and BLamD 338. An excess build up of debris or inadequate regulation of innate immune activation may thus initiate damaging inflammatory cascades. The presence of iNOS expressing macrophages in the choroid of eyes with continuous BLamD and membranous debris suggests alteration of normal immune regulation within the choroid in early AMD. Macrophage nitric oxide production has cytotoxic amd proangiogenic effects in the choroid and retina 248 and is involved in pathological ocular neovascularisation 339
. Both iNOS and vascular endothelial growth factor expression have been noted in macrophages and endothelial cells in excised neovascular membranes 121, 247. The observation that a high BrM macrophage count is associated with subclinical CNV suggests that BrM macrophages may overcome local RPE immunomodulation and express iNOS in these eyes. Although this could not be confirmed in the current study, the presence of iNOS expressing cells in the choroid and scar of the active disciform lesion supports the idea that pro‐inflammatory macrophages are involved in active neovascular processes. Subclinical CNV is found in 41.7% of eyes with geographic atrophy 10, suggesting that neovascularisation is a self‐limiting process in many AMD eyes. However, 129
individuals may be predisposed to progression to neovascular membrane formation due to environmental and/or genetic risk factors. That the fellow eye of eyes with subclinical CNV also had high BrM macrophage counts lends support to this, and is also consistent with the clinical observation that these eyes are at higher risk of developing CNV 340. The significance of iNOS expression by RPE cells in normal aged eyes is unclear. In vitro, RPE cells express iNOS after exposure to ischaemic 341 or inflammatory 342 stimuli. As a part of normal ageing, ischaemia due to choroidal capillary fallout may stimulate RPE iNOS expression. Ideally, CD68 and iNOS double‐immunohistochemistry should be used for phenotypic studies. However, finding colour products easily co‐visualised in the presence of both RPE and melanocyte pigment proved challenging. Pigment also interferes with fluorophores, so that use of immunofluorescence double‐labelling can be equally problematic. It is possible some of the choroidal CD68 expressing cells in this study are activated dendritic cells and not macrophages, and that the perivascular iNOS expressing cells are pericytes. Further immunophentyping of the normal human resident choroidal macrophage population is needed to understand their role in homeostasis and AMD. In summary, while resident choroidal macrophages are present in all eyes, BrM macrophages are associated with accumulation of BLamD and membranous debris, corresponding to clinical evidence of early AMD. Choroidal macrophages in these eyes developed a pro‐inflammatory phenotype, suggesting an altered choroidal immune microenvironment. Macrophages are implicated in the neovascular process, in a mechanism possibly involving a switch in programming evidenced by iNOS expression. 130
4
S E R U M A U T O A N T I B O D Y S T U D I E S
131
4.1 COMPARISON
OF THE ANTI-RETINAL AUTOANTIBODY PROFILE
IN EARLY, NEOVASCULAR
&
ATROPHIC
AMD
4.1.1 Introduction A number of previous studies (reviewed in Section 2.6) have identified an association between serum anti‐retinal autoantibodies and early, neovascular and atrophic AMD. It appears more than one autoantibody may be involved, although their significance to AMD pathogenesis is still uncertain. This study used a single detection method, western blotting, to identify anti‐
retinal autoantibodies in the serum of normal individuals as well as those with early, neovascular or atrophic AMD. Since restriction of autoantibodies to one IgG subclass is a feature of pathological autoimmunity 317, the IgG subclass of autoantibodies was identified. 132
4.1.2 Methods 4.1.2.1
Study serum 4.1.2.1.1 Blue Mountains Eye Study Control, early AMD and GA serum came from the Blue Mountains Eye Study (BMES), a population‐based survey conducted in a geographically defined area west of Sydney, Australia 18. Non‐institutionalised, permanent residents of two postcode areas, aged 49 years or older, were invited to participate between 1992 and 1994. A total of 3654 (82.4% of the eligible) individuals were enrolled. Ocular examination and detailed grading of retinal photographs was conducted at baseline, and 5 343 and 10 years 344 later. The study was approved by the Western Sydney Area Health Service Human Research Ethics Committee and written consent was obtained from all participants. The Wisconsin Age‐related Maculopathy Grading System 15, with slight modifications, was used to define early AMD as either the presence of soft, indistinct or reticular drusen, or the presence of both RPE abnormalities and soft distinct drusen in the macula area (within a 3000 μm radius of the fovea). A total of 191 participants had signs of early AMD at baseline. The International Classification and Grading System 16 was used to define geographic atrophy as a discrete area of retinal de‐pigmentation characterised by sharp borders and the presence of visible choroidal vessels, occupying an area at least 175μm in diameter. A total of 39 participants had geographic atrophy at baseline. Normal controls were aged 60 years or over and had an absence of ocular disease. 133
4.1.2.1.2 Intravitreal Triamcinolone Study Serum for neovascular AMD came from the Intravitreal Triamcinolone Study (IVTAS), a randomised trial of a single dose of intravitreal triamcinolone acetonide (4mg) for neovascular AMD 345. Patients referred or presenting to the Retina Unit of the Sydney Eye Hospital (Sydney, Australia) were invited to participate if they met the following criteria: (1) newly diagnosed neovascular AMD with any classic component (≤3.5 Macular Photocoagulation Study disc areas, subfoveal or within 199μm from the foveal centre) 346; (2) aged 60 years or older; (3) had visual symptoms less than a year; (4) visual acuity 20/200 or better; (5) with clear media; (6) were offered and declined laser photocoagulation. The exclusion criteria were: (1) other serious eye disease, including diabetic retinopathy, hypertensive retinopathy, macular dystrophy, angioid streaks, high myopia (>8 dioptres), glaucoma (with glaucomatous field loss), epiretinal membranes, macular hole and nystagmus; (2) the use of systemic corticosteroids (prednisone ≥5mg/d) or any drug that affects the macula, including chloroquine, hydroxycholoroquine sulphate, thioridazine and cholorpromazine; (3) any condition, physical, mental or social, that would affect regular follow up; and (4) any condition that would prevent photographic or angiographic documentation. A total of 139 participants were enrolled in the study. Informed consent was obtained from each participant. The study was conducted in accordance with the principles of the Declaration of Helsinki and was approved by the Southern Eastern Sydney Area Health Service Research Ethics Committee. Baseline serum was collected before administration of triamcinolone or placebo. 134
4.1.2.1.3 Serum used in autoantibody assays Further inclusion criteria for all participants, from the BMES or IVTAS cohorts, included: (i) absence of autoimmune disease or history of autoimmune disease; (ii) serum that had undergone fewer than two freeze‐thaw cycles (to ensure integrity of serum immunoglobulins); and (iii) serum without evidence of haemolysis (to prevent interference with autoantibody binding). Available serum that fulfilled all the selection criteria totaled 47 participants with early AMD and 14 participants with GA from the BMES, and 74 participants with neovascular AMD from the IVTAS. A total of 16 normal controls from the BMES were also included. All serum was stored after collection at ‐80oC until use. 4.1.2.2
Autoantibody assay Detailed descriptions of protein isolation and quantification, electrophoresis, western blotting, detection antibodies, chemiluminescence reagents and associated buffers can be found in the Appendices (A&D‐G). 4.1.2.2.1 Preparation of human retinal proteins for electrophoresis Isolated human retinae within 12 hours post mortem, from 4 healthy donors (mean age 40.25 +/‐ 7.89 years) without eye disease, were homogenised in phosphate‐buffered saline (PBS) and lysis buffer containing salt buffer, sterile water, Mini‐Complete protease inhibitor cocktail (Boehringer Mannheim, Germany) and NP40. Homogenates were placed on an orbital shaker on ice at 50rpm for 20 minutes and centrifuged at 4000rpm for 5 minutes at 4oC. Supernatants were collected, pooled and stored at ‐20oC until use. Protein concentration was determined as 7mg/ml using Bio‐Rad’s protein assay (Bio‐Rad Laboratories, Hercules, California, USA), in microtitre plates with bovine serum albumin as standard. Retinal homogenates were diluted 1:2 in SDS‐glycine buffer 135
containing 0.02% β‐mercaptoethanol. Samples were heated for 2 minutes at 100 oC and clarified by centrifugation at 3000rpm for 3 minutes. 4.1.2.2.2 Electrophoresis and western blotting Approximately 3μg of protein was applied per 1 mm of SDS‐polyacrylamide gel. For the early experiments, a 12% resolving gel (4% stacking) was poured and set prior to electrophoresis. Because of the entire range of retinal proteins is present in whole retinal homogenates, these gels failed to separate the higher molecular weight proteins well. Later experiments used a 4‐20% linear gradient SDS‐
polyacrylamide gel (Ready Gel from Bio‐Rad Laboratories). All gels used had a single small well for loading the protein standard (Kaleidoscope precision plus protein standard, Bio‐Rad Laboratories) and a large “prep” well that ran almost the entire length of the gel which allowed retinal proteins from the same aliquot to be applied across the whole gel. Proteins were separated by electrophoresis at 200V, 100mA and at room temperature for 30‐40min (depending on ambient temperature). Gels were electro‐blotted onto Hybond P membrane (Amersham Pharmacia Biotech, Buckinghamshire, England) at 60V, 200mA for 4 hours on ice at 4oC. 4.1.2.2.3 Detection of serum anti‐retinal autoantibodies Transblots were incubated in 5% tween tris‐buffered saline (TTBS), ph 7.4, for 1 hour at room temperature to block non‐specific binding. Serum from each participant was reacted with the transblots overnight at room temperature. A multiscreen apparatus (Bio‐Rad Laboratories, see Appendix G) allowed the large area of blotted protein to be sequestered into individual chambers, such that a single blot could be used to test serum from up to 18 individuals. Each well had a total volume of 600μL, which also permitted significant savings in detection antibodies and reagents. From this step onwards, blots were kept in the multiscreen apparatus until the final wash step was complete. 136
An optimal serum dilution of 1:200 (in 0.05% TTBS) was found after trials to minimise background reactivity. After overnight incubation in serum, transblots were washed twice for 10 minutes with 0.05% TTBS, and once for 10 minutes with TBS. Transblots were incubated with the detection antibody (see below) for 1 hour at room temperature, followed by two 10‐minute washes in 0.05% TTBS, and a 10‐
minute wash in TBS. Blots were removed from the multiscreen apparatus at this point and incubated in enhanced chemiluminescence (ECL) western blotting detection reagent (Amersham Pharmacia Biotech) for 2 minutes in a dark room. Transblots were then placed in plastic sleeves and exposed to Hyperfilm ECL (Amersham Pharmacia Biotech) for 1‐3 minutes to detect autoantibody binding. Initially, the 3 minute exposure time was used. It was later found that a 1 minute exposure resulted in adequate films. Thus earlier blots appear much darker than later blots (Figures 4.1‐4.3). Equivocal ECL film results were adjudicated by two masked assessors. Dilution buffer without serum acted as a negative control in each experiment, and serum from a patient with known autoimmune retinopathy acted as a positive control. 4.1.2.2.4 Antibodies used to detect serum autoantibodies To detect autoantibody binding, transblots that had been incubated in participant serum overnight were incubated with a HRP‐conjugated secondary antibody. Initially, serum from each participant was tested with a goat anti‐human IgGAM (Zymed Invitrogen) detection antibody. This is a polyclonal antibody that recognises both heavy and light chains of human IgG, IgA and IgM. All tests were repeated at least once. Serum testing positive for anti‐retinal autoantibodies with the IgGAM detection antibody was then tested in 4 separate experiments, using fresh blots, with one of the following monoclonal antibodies: anti‐human IgG1, anti‐human IgG2, anti‐human IgG3 and anti‐human IgG4 (all HRP‐conjugated, mouse anti‐human, Zymed Invitrogen). Based on past reports, it was anticipated 137
that the majority of autoantibodies would belong to the IgG class. However, an unexpectedly small proportion of IgGAM‐detected autoantibodies were found to belong to one of the four IgG subclasses. It was subsequently decided that serum testing positive with the anti‐IgGAM detection antibody, but negative using anti‐
IgG1, IgG2, IgG3 and IgG4, would be re‐tested using HRP‐conjugated monoclonal mouse anti‐human IgM (Zymed Invitrogen). All detection antibodies were diluted 1:10 000 in TBS. 4.1.2.3
Statistics Contingency tables comparing the proportions of participants with serum autoantibodies in the following groups: (i) controls vs. early AMD; (ii) controls vs. GA; (iii) controls vs. neovascular AMD; (iv) early AMD vs. neovascular AMD; (v) early AMD vs. GA; and (vi) neovascular AMD vs. GA, were evaluated by the χ2 test, using SPSS (v15.0; SPSS Inc, Chicago, IL, USA). 138
4.1.3 Results Anti‐retinal autoantibodies were found in all groups, including 25% of controls, when the anti‐IgGAM detection antibody was used (Table 4.1‐1). Multiple bands of reactivity were frequently found in the same individual (Figures 4.1‐1 – 4.1‐3). The complexity of the reactivity patterns made comparisons between groups difficult. Autoantibodies reacted against antigens in the following molecular weight ranges: 70‐90kD, 40‐50kD and around 30kD. High molecular weight (>120kD) antigens were rare. Only a small proportion of serum tested positive for IgG autoantibodies with the monoclonal anti‐IgG1‐4 detection antibodies (Figures 4.1‐1 – 4.1‐3). IgG4 autoantibodies were rarest, found only in one participant with early AMD. Antigens were similar to those targeted by anti‐IgGAM detected autoantibodies. Of note, participants with early AMD and GA had the highest proportion of IgG autoantibodies and the largest numbers of participants with IgG autoantibodies belonging to more than one subclass (Table 4.1‐1). IgM autoantibodies were more frequently found in participants with neovascular AMD. However, it should be noted that IgM autoantibodies may be more prevalent than suggested in Table 4.1‐1, since only IgGAM positive and IgG1‐4 negative serum was tested. 139
Table 4‐1 Serum anti‐retinal autoantibodies in controls and in early, neovascular and atrophic AMD: summary of findings Mean Age (years
+/-SD)
Sex
(M/F)
Anti-IgGAM detected Anti-IgG detected
autoantibodies
autoantibodies (all
subclasses)
IgG autoantibodies by subclass
IgG1
IgG2
IgG3
IgG4
IgM
Non-IgG
Non-IgM
Control
(n=16)
66.1 +/- 5.8
6/10
4/16
25%
2
(13%)
0
0
2
0
0
2
Early AMD
(n=47)
72.9 +/- 6.8
17/30
27/47
(49%)
16
(34%)
5
8
11
1
1
10
(21%)
Neovascular AMD
(n=74)
76.9 +/- 7.2
31/43
26/74
(35%)
8
(11%)
2
7
0
0
7
11
(15%)
GA
(n=14)
85.3 +/- 6.1
4/10
7/14
(50%)
6
(43%)
4
4
5
0
1
0
(0%)
Number of participants in each group with IgG autoantibodies belonging to more than one subclass: Control= 0; Early AMD = 8; Neovascular AMD = 2; Geographic atrophy =
140
Figure 4.1‐1 Western blot‐detected anti‐retinal autoantibodies in normal controls and participants with geographic atrophy 1
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17 18 19 20
202
121
79
Anti-IgGAM
detection antibody
41
31.6
17.8
202
121
Anti-IgG1
detection antibody
79
41
31.6
17.8
202
121
79
Anti-IgG2
detection antibody
41
31.6
17.8
202
121
79
Anti-IgG3
detection antibody
41
31.6
17.8
202
121
79
Anti-IgG4
detection antibody
41
31.6
17.8
141
Figure 4.1‐1 Western blot‐detected anti‐retinal autoantibodies in normal controls and participants with geographic atrophy Rows 1‐8: blots of serum from normal controls. Rows 1‐4 represent serum with anti‐
IgGAM detected autoantibodies. Rows 5‐8 represent serum without anti‐IgGAM detected autoantibodies Rows 9‐20: blots of serum from GA participants. Rows 9‐16 represent serum with anti‐
IgGAM detected autoantibodies. Rows 17‐20 represent serum without anti‐IgGAM detected autoantibodies Each column represents western blot results from the same individual 142
Figure 4.1‐2 Western blot‐detected anti‐retinal autoantibodies in participants with early AMD 1
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
202
121
79
Anti-IgGAM
detection antibody
41
31.6
17.8
202
121
Anti-IgG1
detection antibody
79
41
31.6
17.8
202
121
79
Anti-IgG2
detection antibody
41
31.6
17.8
202
121
79
Anti-IgG3
detection antibody
41
31.6
17.8
202
121
79
Anti-IgG4
detection antibody
41
31.6
17.8
143
Figure 4.1‐2 Western blot‐detected anti‐retinal autoantibodies in participants with early AMD Rows 1‐27: early AMD participants with anti‐IgGAM detected autoantibodies Rows 28‐31: early AMD participants without anti‐IgGAM detected autoantibodies Each column represents western blot results from the same individual 144
Figure 4.1‐3 Western blot‐detected anti‐retinal autoantibodies in participants with neovascular AMD 205
123
1
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
79
40
Anti-IgGAM
detection antibody
31
16.9
202
121
79
Anti-IgG1
detection antibody
41
31.6
17.8
202
121
79
Anti-IgG2
detection antibody
41
31.6
17.8
202
121
Anti-IgG3
detection antibody
79
41
31.6
17.8
202
121
79
Anti-IgG4
detection antibody
41
31.6
17.8
145
Figure 4.1‐3 Western blot‐detected anti‐retinal autoantibodies in participants with neovascular AMD Rows 1‐26: neovascular AMD participants with anti‐IgGAM detected autoantibodies Row 27: positive control – patient with known autoimmune retinopathy Rows 28‐31: neovascular AMD participants without anti‐IgGAM detected autoantibodies Each column represents western blot results from the same individual 146
Participants with early AMD were four times more likely to have anti‐IGAM detected autoantibodies compared to controls (p= 0.0025; χ2=5.028; 95% CI = 1.1‐
14.4). Participants with neovascular or atrophic AMD were no more likely to have anti‐IgGAM or anti‐IgG(1‐4) detected autoantibodies than controls (Table 3.1‐2). Participants with GA were seven times more likely to have anti‐IgG(1‐4) detected autoantibodies compared to participants with neovascular AMD (Fisher’s exact p = 0.005; 95% CI = 1.9 – 26.3). Participants with early AMD were 1.6 times more likely to have anti‐IgGAM detected autoantibodies compared to participants with neovascular AMD (p = 0.016, χ2 = 5.813, 95% CI = 1.6 – 2.4), and three times more likely to have anti‐IgG(1‐
4) detected autoantibodies (p = 0.001, χ2 =, 95% CI = 1.6 – 7.1). 147
Table 4‐2 Comparison of Anti‐IgGAM and anti‐IgG detected autoantibodies in control, early AMD, neovascular AMD and atrophic AMD participants Control
IgGAM
autoantibodies
absent
IgGAM
autoantibodies
present
12
4
p
IgG
autoantibodies
absent
IgG
autoantibodies
present
14
2
29
18
14
2
66
8
14
2
7
6
66
8
7
6
30
17
7
6
30
17
66
8
0.025*
Early AMD
20
27
Control
12
4
0.115
0.436
Neovascular AMD
48
26
Control
12
4
1
0.092
0.143
GA
6
7
Neovascular AMD
48
26
0.005*
0.227
GA
6
7
Early AMD
20
27
0.535
0.817
GA
6
7
Early AMD
20
27
0.001*
0.016*
Neovascular AMD
48
26
148
p
4.1.4 Discussion The study confirms that complex repertoires of autoantibodies against retinal antigens can be found in all individuals, including those without evidence of eye disease. Despite this, individuals with early AMD had significantly more anti‐
retinal autoantibodies compared to normal controls and to participants with neovascular AMD. In early AMD, it is possible that anti‐retinal autoantibodies arise as a consequence of the accumulation of extracellular debris in the form of BLinD, BLamD and soft drusen. At least some of the constituents of these deposits have been shown to be retinal in origin (see Sections 2.2.3). Although retinal antigens regularly leave the ocular compartment via the blood inside ocular‐derived APCs 158, autoimmune responses are not mounted against them 160. This is because ocular‐derived APCs are programmed by a local microenvironment rich in immunosuppressive cytokines 159, to which RPE cells contribute significantly (see Section 2.2.1.7). RPE abnormalities are present in early AMD, and this may compromise their immunomodulatory capacity, favouring the formation of autoantibodies which, in this context could be considered pathological. Another explanation could be that the autoantibodies are part of the natural autoantibody repertoire, deployed to remove the build‐up of extracellular debris. In this context the autoantibodies may be protective, since the efficient removal of extracellular debris can prevent uncontrolled activation of inflammatory cascades (see Section 2.6). It is not clear why anti‐IgG detected autoantibodies were more frequently found in participants with GA compared to participants with neovascular AMD. The only previous study examining this group found the opposite – that patients with GA 149
were less likely to have immunofluorescence‐detected autoantibodies compared to those with early or neovascular AMD 299 (see Section 2.6.3). In GA, Muller cell end processes replace the absent RPE 311 and maintain the blood‐retina barrier. There is some remodelling at the edges of areas of atrophy 8 and it is possible that these liberate cellular debris. However, it is difficult to argue that the amount of remodelling in GA could exceed that in neovascular AMD, a group in which autoantibodies were found in surprisingly few individuals. One possible explanation for this observation is that perhaps the majority of autoantibodies in neovascular AMD belong to the IgM class. However, results from this study cannot determine if this is the case, since not all IgGAM‐detected autoantibodies were retested with the IgM detection antibody. The results do, however, highlight the need to assess not just the autoantibody target antigen, but the autoantibody immunoglobulin class and subclass in future studies. Changes in the whole autoantibody repertoire will also need to be assessed, since it may be complex, even in normal controls The numbers of individuals with IgG autoantibodies in this sample were small compared with previous studies (see Table 2.6‐1, Section 2.6). This may be due to the highly specific nature of the monoclonal antibodies used, and the relatively low concentration of serum tested. Because IgG autoantibodies were less common than expected, and because they targeted more than one antigen, it was difficult to assess subclass restriction. The significance of a larger number of participants with IgG autoantibodies belonging to more than one class in the GA and early AMD groups is not known and requires further investigation. This study is limited by the small control group, which is not age‐matched. The use of a large control group to establish normal autoantibody patterns of reactivity will be important in future studies. A number of limitations are inherent in the use 150
of Western blotting as a detection method. Firstly, proteins are separated under reducing conditions, such that their tertiary structure is lost, leading to loss of conformational epitopes. Secondly, when using retinal proteins from whole retinal homogenates, the molecular weight of autoantibody target proteins can only be approximated, since the number of proteins in each gel and transblot is large. Related to this limitation is the inability to measure autoantibody titres, since the identity of the target antigen is not known. Finally, it is likely that autoantibodies against oxidatively modified proteins and glycosylated proteins may be of more importance in AMD, and the concentrations of these proteins in normal retinae may be too minimal to allow autoantibody binding. The results of the following two sections should thus be interpreted with these limitations in mind. 151
4.2 ANTI-RETINAL
IN EARLY
AUTOANTIBODIES AND DISEASE PROGRESSION
AMD
4.2.1 Introduction Although no anti‐retinal autoantibodies were found to be specific for AMD in the previous study, they were more frequently found in individuals with early AMD than controls. This study examined whether the presence of baseline anti‐retinal autoantibodies in participants with early AMD was associated with progression to advanced AMD 5 and 10 years later. 4.2.2 Methods 4.2.2.1
Participants Participants with early AMD came from the BMES cohort as described in Section 4.1.2. Follow up examinations took place 5 and 10 years from date of enrolment, at which ocular examination and fundus photography was carried out. Photographs were graded using the Wisconsin Grading System by masked graders, and side‐
by‐side comparisons between baseline and 5‐ and 10‐ year photographs were made. 4.2.2.2
Autoantibody assay Autoantibodies were detected as described in Section 4.1.2. Detailed descriptions can also be found in the Appendices (A & D‐G). Since relatively few participants had IgG anti‐retinal autoantibodies, the relationship between IgGAM‐detected baseline autoantibodies and progression to advanced AMD was examined. 152
4.2.2.3
Statistics The χ2 test was used to evaluate the association between baseline autoantibodies and progression to advanced AMD at the 5‐ and 10‐year examinations, using SPSS (v15.0; SPSS Inc, Chicago, IL, USA). A p value of < 0.05 was considered significant. 153
4.2.3 Results A total of 11 participants (23.4%) with early AMD progressed to advanced AMD by the time the 10‐year follow up examinations were conducted, including 5 participants (10.6%) who developed advanced AMD 5 years after baseline (Table 4.2.1). Three of these 5 participants developed unilateral neovascular AMD, and two developed atrophic AMD in one eye and neovascular AMD in the fellow eye. No association as found between the presence of anti‐IgGAM detected autoantibodies at baseline and progression to advanced AMD at 5 years (χ2 = 0.015, Fisher’s exact p = 1.00). A further four participants (12.7%) progressed to advanced AMD at the 10‐year follow up examination. Two of these participants developed neovascular AMD and two developed atrophic AMD (Table 4.2.1). Additionally, 2 participants with neovascular AMD at the 5 year examination developed neovascular AMD in the fellow eye at the 10 year examination. There was no association between baseline autoantibodies and progression to advanced AMD over the 10 year‐period (χ2 = 0.156, Fisher’s exact p = 1.00). 154
Table 4‐3 Baseline serum anti‐retinal autoantibodies and progression to advanced AMD at 5 and 10 years. Advanced AMD at 5 years
Advanced AMD at 10 years
No
Yes
No
Yes
Autoantibodies
Absent
18
2
17
3
Autoantibodies
Present
24
3
24
3
RR
0.90
1.35
95% CI
0.16 - 4.90
0.30 - 6.00
chi square
0.150
0.156
Fisher's exact p
1.00
1.00
155
4.2.4 Discussion This study could not find an association between the presence of baseline anti‐
retinal autoantibodies and progression to advanced AMD over a 5 to 10 year period. There are several possible explanations for these findings. Firstly, although more participants with early AMD had anti‐retinal autoantibodies than controls, it is possible that these are a mix of both natural and pathological autoantibodies (as discussed in Section 4.1.4). Secondly, it is likely that any autoantibody‐AMD disease associations can only be found by examining changes in the whole repertoire, which may be subtle and would require larger sample sizes and more sophisticated tests. Finally, the autoantibody repertoire in each individual is likely to be dynamic, and may have changed many times in the 5 to 10 years between follow up visits. The study is limited by the small sample size, and larger studies with more frequent autoantibody assays between follow up visits are necessary to confirm the findings. It should also be noted that participants with pigmentary changes alone were not included in the original BMES definition of early AMD. Since this group may represent a separate early AMD phenotype (see Section 3.1), its inclusion in future comparative studies is important. 156
4.3 ANTI-RETINAL
NEOVASCULAR
AUTOANTIBODIES AND VISION LOSS IN
AMD
4.3.1 Introduction Patients with neovascular AMD are at high risk of rapid vision loss (Section 2.1.4). Identifying eyes most at‐risk of rapid deterioration is thus an important part of assessment and management. Although the previous study found that neovascular AMD participants were no more likely than controls to have anti‐
retinal autoantibodies, this study examined whether the presence of baseline anti‐
retinal autoantibodies was associated with disease progression over a 24‐month period. Ideally, this association should be studied in a cohort of individuals with neovascular AMD, followed up for 12 to 24 months, without any treatment or interventions. However, given the availability of palliative treatments for neovascular AMD, such a study would be unethical to undertake and difficult to recruit for. The study cohort was thus selected from the IVTAS. Vision loss was chosen as the outcome measure because it is a continuous variable which allowed the use of linear mixed effects modelling to compare autoantibody‐positive and autoantibody‐negative participants. This was the best available statistical method that could account for all of the following potential interactions, for each individual participant, at each follow‐up time point: (i) triamcinolone (or placebo) treatment and vision; (ii) anti‐retinal autoantibodies and treatment; and (iii) anti‐
retinal autoantibodies and vision. 157
4.3.2 Methods 4.3.2.1
Participants Participants with neovascular AMD came from the IVTAS, as described in Section 4.1.2. The study took place over 12 months and was later extended to 24 months. Participants were randomised to receive an injection of intravitreal triamcinolone or subconjunctival isotonic saline, given after baseline visual acuity (VA) measurements. Treatment assignment and masking protocols have been described in detail elsewhere 345. Best‐corrected visual acuity was measured by trained examiners using a back‐lit, self‐calibrating logMAR chart (Lighthouse International, LaDalle, IL, USA) at 2.40, 0.95 and 0.60 metres. VA was evaluated at baseline, 3, 6, 12, 18 and 24 months after entry into the study. Angiographic grading of CNV as classic or occult was based on Macular Photocoagulation Study guidelines 346. Serum was collected prior to the administration of triamcinolone or placebo. Of the 139 participants (151 eyes) enrolled in the trial, baseline serum from 74 participants (82 eyes) had undergone fewer than two freeze‐thaw cycles, was not haemolysed, and was thus included in this study. 4.3.2.2
Autoantibody assay Autoantibodies were detected as described in Section 4.1.2. Detailed descriptions can also be found in Appendices (A&D‐G). 4.3.2.3
Statistics S‐Plus (v6.2, Insightful Corporation, Seatle, WA, USA) was used to fit linear mixed effect models to VA. Eye identification number was incorporated as a random effect, treatment group and baseline autoantibody status as fixed effects, and month as a fixed covariate. In all cases, both two and three‐way interactions were 158
fitted and dropped from the final models if not significant. Two‐tailed tests with a 5% level of significance were used throughout. 159
4.3.3 Results Participant characteristics and baseline autoantibody status are summarised in Table 4.3‐1. Classic neovascular lesions were present in 59 eyes, and occult lesions in 23 eyes. There was no difference in gender, mean age (t72 = ‐1.21, p = 0.23) or rate of vision loss over 24 months (t286 = ‐1.05, p = 0.30) between the triamcinolone (n=32) and placebo (n=42) treated groups. There were no significant differences in age (p = 0.504) or sex (p = 0.266) between autoantibody‐positive and autoantibody‐negative participants. Autoantibodies appeared to be more frequently found in participants with classic neovascular lesions (12%) compared to those with occult lesions (4%), although this was not statistically significant (p = 0.431). Linear mixed effects modelling showed that anti‐IgGAM detected baseline anti‐
retinal autoantibodies were not associated with vision loss over a 24 month period (t288 = ‐0.12, p = 0.91) (Figure 4.3.1). However, participants with baseline anti‐IgG(1‐
4) detected autoantibodies lost, on average, 0.7 more letters of vision on a logMAR chart per month (or 8.4 more letters per year) than other participants (t289 = ‐2.17, p = 0.03), irrespective of treatment group (Figure 4.3‐2) 160
Table 4‐4 Participant characteristics and baseline anti‐retinal autoantibody status IgG1-4 detected autoantibodies
Number of
Patients/Eyes
M/F
Mean age
(years)
IgGAM-detected
autoantibodies
IgG1
IgG2
IgG3
IgG4
Total IgG
Placebo
42/42
18/24
77.4 +/- 6.6
16/42 (38%)
0
2
0
0
2 (5%)
Triamcinolone
40/32
13/19
75.3 +/- 7.8
10/32 (31%)
2
5
0
0
6 (9%)
161
Figure 4.3‐1 Vision loss over 24 months in anti‐retinal autoantibody positive participants vs. anti‐retinal autoantibody negative participants (anti‐IgGAM detected autoantibodies) 60
Anti-IgGAM detected autoantibodies absent
Anti-IgGAM detected autoantibodies present
Mean LogMar Acuity
50
40
30
20
p = 0.91
10
0
0
3
6
12
Month
162
18
24
Figure 4.3‐2 Vision loss over 24 months in anti‐retinal autoantibody positive participants vs. anti‐retinal autoantibody negative participants (anti‐IgG detected autoantibodies) 60
Anti-IgG detected autoantibodies absent
Anti-IgG detected autoantibodies present
Mean LogMar Acuity
50
40
30
20
p = 0.03
10
0
0
3
6
12
Month
163
18
24
4.3.4 Discussion That baseline IgG anti‐retinal autoantibodies were associated with more rapid vision loss over 24 months in participants with neovascular AMD is surprising. Few participants with neovascular AMD had IgG anti‐retinal compared to those with early AMD or GA (see Section 4.1). The finding may thus be an indication that in neovascular AMD, the anti‐retinal autoantibodies are pathological and not a part of the natural repertoire. It is also interesting that this association was not observed with anti‐IgGAM detected autoantibodies, which may have belonged to any of three immunoglobulin classes, and were thus less specific. The neovascular disease process involves the growth of abnormal and highly permeable new vessels into the sub‐RPE or sub‐retinal space. The high rate of extracellular matrix turnover and vascular remodelling in neovascular AMD lesions may liberate significant cellular debris. There is disruption of the external blood‐retina barrier, and the presence of local pro‐inflammatory cytokines 224. All of these factors may produce circumstances favourable to the initiation of autoimmune events. These explanations are at this stage, speculative, since the number of participants with IgG autoantibodies in this cohort was small and the findings will need to be confirmed in larger studies. However, the recruitment of participants with neovascular AMD into a study without interventions will continue to prove challenging. It should also be noted that the main outcome measure, visual acuity, may be influenced by other factors and may not be the most reliable indicator of disease progression. Outcome measures such as changes in lesion size and vessel leakage should be included in future studies where possible. 164
5
S U M M A R Y & C O N C L U S I O N S
165
The histopathological threshold at which normal ageing becomes early AMD is defined by the presence of both BLinD and a continuous layer of BLamD. Because these eyes appear clinically normal in the majority of cases, more sophisticated clinical tests will be required for early detection. Evidence for distinct early AMD phenotypes and two separate pathways along which early AMD can progress can be used to inform future clinical classifications systems, as well as genetic and epidemiological studies. BLamD and membranous debris accumulation may increase the recruitment of macrophages to sub‐macular Bruch’s membrane and alter the programming of resident choroidal macrophages. This observation requires confirmation, particularly in the context of subclinical CNV. Animal models that allow macrophage‐associated genes to be knocked out are likely to further elaborate normal resident choroidal macrophage function. However, interpretation of data from animal models will require better characterisation of human resident choroidal macrophages, about which little is known. The relationship between anti‐retinal autoantibodies and AMD is probably complex. It is unlikely that one or two autoantibodies will be specifically associated with AMD. Rather, future studies will need to assess the whole autoantibody repertoire to discern subtle changes in the reaction patterns. For these studies to be meaningful, a large control group is necessary, so that normal patterns of reactivity can be established. A number of approaches allow the examination of complex autoantibody repertoires each with strengths and limitations. Software designed to assess densitometry differences on western blots have been used 296. While this approach does not allow identification of the target antigens, it can be used to direct proteomic studies. A combination of 3‐D gel electrophoresis and liquid chromatography/mass spectrometry can then determine the amino acid sequence of the target proteins. This approach, 166 however, does not allow rapid screening of a large number of serum samples, a problem that can be overcome by the use of protein microarrays. Made‐to‐order arrays with thousands of purified proteins are becoming increasingly available and affordable. The difficulty with this approach is in determining the appropriate proteins for an autoantibody assay. It is likely that a combination of all three methods will determine the significance of anti‐retinal autoantibodies in AMD pathogenesis. 167 R E F E R E N C E S 1. Liem, A.T., Keunen, J.E., and van Norren, D., Foveal densitometry in adult‐onset diffuse drusen. American Journal of Ophthalmology, 1992. 114(2): p. 149‐57. 2. de Jong, P.T., Age‐related macular degeneration. New England Journal of Medicine, 2006. 355(14): p. 1474‐85. 3. Sarks, S.H. and Sarks, J.P., Age‐related maculopathy: Nonneovascular age‐related macular degeneration and the evolution of geographic atrophy, in Medical retina, S. Ryan, Editor. 2001, Mosby: St Louis. p. 1064‐1099. 4. Gass, J.D., Pathogenesis of disciform detachment of the neuroepithelium. American Journal of Ophthalmology, 1967. 63(3): p. Suppl:1‐139. 5. Sarks, S.H. Changes in the region of the choriocapillaris in aging and degeneration. in 23rd International Congress of Ophthalmology. 1978. Kyoto, Japan: Excerpta Medica. 6. Gass, J.D., Drusen and disciform macular detachment and degeneration. Archives of Ophthalmology, 1973. 90(3): p. 206‐17. 7. Green, W.R. and Key, S.N., 3rd, Senile macular degeneration: A histopathologic study. Transactions of the American Ophthalmological Society, 1977. 75: p. 180‐254. 8. Sarks, J.P., Sarks, S.H., and Killingsworth, M.C., Evolution of geographic atrophy of the retinal pigment epithelium. Eye, 1988. 2(Pt 5): p. 552‐77. 9. Sunness, J.S., The natural history of geographic atrophy, the advanced atrophic form of age‐related macular degeneration. Molecular Vision, 1999. 5: p. 25. 10. Sarks, S.H., Ageing and degeneration in the macular region: A clinico‐pathological study. British Journal of Ophthalmology, 1976. 60(5): p. 324‐41. 11. Sunness, J.S., Gonzalez‐Baron, J., Bressler, N.M., Hawkins, B., and Applegate, C.A., The development of choroidal neovascularization in eyes with the geographic atrophy form of age‐related macular degeneration. Ophthalmology, 1999. 106(5): p. 910‐9. 12. Pauleikhoff, D., Neovascular age‐related macular degeneration: Natural history and treatment outcomes. Retina, 2005. 25(8): p. 1065‐84. 13. Ferris, F.L., 3rd, Fine, S.L., and Hyman, L., Age‐related macular degeneration and blindness due to neovascular maculopathy. Archives of Ophthalmology, 1984. 102(11): p. 1640‐2. 14. De la Paz, M.A., Pericak‐Vance, M.A., Haines, J.L., and Seddon, J.M., Phenotypic heterogeneity in families with age‐related macular degeneration. American Journal of Ophthalmology, 1997. 124(3): p. 331‐43. 15. Klein, R., Davis, M.D., Magli, Y.L., Segal, P., Klein, B.E., and Hubbard, L., The wisconsin age‐related maculopathy grading system. Ophthalmology, 1991. 98(7): p. 1128‐34. 168 16. The International Age‐Related Maculopathy Study Group, An international classification and grading system for age‐related maculopathy and age‐related macular degeneration. The international arm epidemiological study group. Survey of Ophthalmology, 1995. 39(5): p. 367‐74. 17. Klein, R., Klein, B.E., and Linton, K.L., Prevalence of age‐related maculopathy. The beaver dam eye study. Ophthalmology, 1992. 99(6): p. 933‐43. 18. Mitchell, P., Smith, W., Attebo, K., and Wang, J.J., Prevalence of age‐related maculopathy in australia. The blue mountains eye study. Ophthalmology, 1995. 102(10): p. 1450‐60. 19. VanNewkirk, M.R., Nanjan, M.B., Wang, J.J., Mitchell, P., Taylor, H.R., and McCarty, C.A., The prevalence of age‐related maculopathy: The visual impairment project. Ophthalmology, 2000. 107(8): p. 1593‐600. 20. Robman, L.D., Tikellis, G., Garrett, S.K., Harper, C.A., McNeil, J.J., Taylor, H.R., and McCarty, C.A., Baseline ophthalmic findings in the vitamin e, cataract and age‐related maculopathy (vecat) study. Australian & New Zealand Journal of Ophthalmology, 1999. 27(6): p. 410‐6. 21. Hofman, A., Grobbee, D.E., Jong, P.T.V.M., and Ouweland, F.A., Determinants of disease and disability in the elderly: The rotterdam elderly study. European Journal of Epidemiology, 1991. 7(4): p. 403‐422. 22. Age‐Related Eye Disease Study Research Group, The age‐related eye disease study system for classifying age‐related macular degeneration from stereoscopic color fundus photographs: The age‐related eye disease study report number 6. American Journal of Ophthalmology, 2001. 132(5): p. 668‐681. 23. Evans, J.R., Risk factors for age‐related macular degeneration. Progress in Retinal & Eye Research., 2001. 20(2): p. 227‐53. 24. Taylor, H.R., Keeffe, J.E., Vu, H.T., Wang, J.J., Rochtchina, E., Pezzullo, M.L., and Mitchell, P., Vision loss in australia. Medical Journal of Australia, 2005. 182(11): p. 565‐8. 25. Smith, W., Assink, J., Klein, R., Mitchell, P., Klaver, C.C., Klein, B.E., Hofman, A., Jensen, S., Wang, J.J., and de Jong, P.T., Risk factors for age‐related macular degeneration: Pooled findings from three continents. Ophthalmology., 2001. 108(4): p. 697‐704. 26. Klein, R., Peto, T., Bird, A., and Vannewkirk, M.R., The epidemiology of age‐related macular degeneration. American Journal of Ophthalmology., 2004. 137(3): p. 486‐95. 27. Age‐Related Eye Disease Study Research Group, Potential public health impact of age‐
related eye disease study results: Areds report no. 11. Archives of Ophthalmology, 2003. 121(11): p. 1621‐4. 28. Tomany, S.C., Wang, J.J., Van Leeuwen, R., Klein, R., Mitchell, P., Vingerling, J.R., Klein, B.E.K., Smith, W., and De Jong, P.T.V.M., Risk factors for incident age‐related macular degeneration: Pooled findings from 3 continents. Ophthalmology, 2004. 111(7): p. 1280‐7. 29. Klein, R., Klein, B.E.K., Knudtson, M.D., Meuer, S.M., Swift, M., and Gangnon, R.E., Fifteen‐year cumulative incidence of age‐related macular degeneration: The beaver dam eye study. Ophthalmology, 2007. 114(2): p. 253‐62. 169 30. Wang, J.J., Mitchell, P., Smith, W., and Cumming, R.G., Bilateral involvement by age related maculopathy lesions in a population. British Journal of Ophthalmology, 1998. 82(7): p. 743‐7. 31. Gudnadottir, G.S., Magnusson, K.P., Stefansson, E., Jonasson, F., Helgadottir, G., and Sigurdsson, H., The time pattern of bilateral exudative age‐related macular degeneration. Acta Ophthalmologica Scandinavica, 2005. 83(3): p. 333‐6. 32. Klein, R., Klein, B.E.K., Tomany, S.C., Meuer, S.M., and Huang, G.‐H., Ten‐year incidence and progression of age‐related maculopathy: The beaver dam eye study. Ophthalmology, 2002. 109(10): p. 1767‐1779. 33. Wang, J.J., Foran, S., Smith, W., and Mitchell, P., Risk of age‐related macular degeneration in eyes with macular drusen or hyperpigmentation: The blue mountains eye study cohort. Arch Ophthalmol, 2003. 121(5): p. 658‐663. 34. van Leeuwen, R., Klaver, C.C., Vingerling, J.R., Hofman, A., and de Jong, P.T., The risk and natural course of age‐related maculopathy: Follow‐up at 6 1/2 years in the rotterdam study. Archives of Ophthalmology., 2003. 121(4): p. 519‐26. 35. Age‐Related Eye Disease Study Group, The age‐related eye disease study severity scale for age‐related macular degeneration: Areds report no. 17. Archives of Ophthalmology, 2005. 123(11): p. 1484‐98. 36. Age‐Related Eye Disease Study Research Group, A simplified severity scale for age‐related macular degeneration: Areds report no. 18. Archives of Ophthalmology, 2005. 123(11): p. 1570‐4. 37. Bressler, N.M., Munoz, B., Maguire, M.G., Vitale, S.E., Schein, O.D., Taylor, H.R., and West, S.K., Five‐year incidence and disappearance of drusen and retinal pigment epithelial abnormalities. Waterman study. Archives of Ophthalmology, 1995. 113(3): p. 301‐8. 38. Tikellis, G., Robman, L.D., Dimitrov, P., Nicolas, C., McCarty, C.A., and Guymer, R.H., Characteristics of progression of early age‐related macular degeneration: The cardiovascular health and age‐related maculopathy study. Eye, 2007. 21(2): p. 169‐76. 39. Klein, R., Klein, B.E., Jensen, S.C., and Meuer, S.M., The five‐year incidence and progression of age‐related maculopathy: The beaver dam eye study. Ophthalmology, 1997. 104(1): p. 7‐21. 40. Macular Photocoagulation Study Group, Risk factors for choroidal neovascularization in the second eye of patients with juxtafoveal or subfoveal choroidal neovascularization secondary to age‐related macular degeneration. Archives of Ophthalmology, 1997. 115(6): p. 741‐7. 41. Sarks, J., Tang, K., Killingsworth, M., Arnold, J., and Sarks, S., Development of atrophy of the retinal pigment epithelium around disciform scars. British Journal of Ophthalmology, 2006. 90(4): p. 442‐6. 42. Sunness, J.S., Margalit, E., Srikumaran, D., Applegate, C.A., Tian, Y., Perry, D., Hawkins, B.S., and Bressler, N.M., The long‐term natural history of geographic atrophy from age‐
related macular degeneration: Enlargement of atrophy and implications for interventional clinical trials. Ophthalmology, 2007. 114(2): p. 271‐7. 170 43. Klein, R., Klein, B.E., Linton, K.L., and DeMets, D.L., The beaver dam eye study: The relation of age‐related maculopathy to smoking. American Journal of Epidemiology, 1993. 137(2): p. 190‐200. 44. Seddon, J.M., Willett, W.C., Speizer, F.E., and Hankinson, S.E., A prospective study of cigarette smoking and age‐related macular degeneration in women. JAMA, 1996. 276(14): p. 1141‐6. 45. Thornton, J., Edwards, R., Mitchell, P., Harrison, R.A., Buchan, I., and Kelly, S.P., Smoking and age‐related macular degeneration: A review of association. Eye, 2005. 19(9): p. 935‐44. 46. Khan, J.C., Thurlby, D.A., Shahid, H., Clayton, D.G., Yates, J.R., Bradley, M., Moore, A.T., Bird, A.C., and Genetic Factors in, A.M.D.S., Smoking and age related macular degeneration: The number of pack years of cigarette smoking is a major determinant of risk for both geographic atrophy and choroidal neovascularisation. British Journal of Ophthalmology, 2006. 90(1): p. 75‐80. 47. Delcourt, C., Diaz, J.L., Ponton‐Sanchez, A., and Papoz, L., Smoking and age‐related macular degeneration. The pola study. Pathologies oculaires liees a lʹage. Archives of Ophthalmology, 1998. 116(8): p. 1031‐5. 48. Chan, D., Cigarette smoking and age‐related macular degeneration. Optometry & Vision Science, 1998. 75(7): p. 476‐84. 49. Vingerling, J.R., Hofman, A., Grobbee, D.E., and de Jong, P.T., Age‐related macular degeneration and smoking. The rotterdam study. Archives of Ophthalmology, 1996. 114(10): p. 1193‐6. 50. Tomany, S.C., Wang, J.J., van Leeuwen, R., Klein, R., Mitchell, P., Vingerling, J.R., Klein, B.E.K., Smith, W., and de Jong, P.T.V.M., Risk factors for incident age‐related macular degeneration: Pooled findings from 3 continents. Ophthalmology, 2004. 111(7): p. 1280‐1287. 51. Klein, R., Klein, B.E., and Moss, S.E., Relation of smoking to the incidence of age‐related maculopathy. The beaver dam eye study. American Journal of Epidemiology, 1998. 147(2): p. 103‐10. 52. Vingerling, J.R., Dielemans, I., Bots, M.L., Hofman, A., Grobbee, D.E., and de Jong, P.T., Age‐related macular degeneration is associated with atherosclerosis. The rotterdam study. American Journal of Epidemiology, 1995. 142(4): p. 404‐9. 53. Seddon, J.M. and Chen, C.A., The epidemiology of age‐related macular degeneration. International Ophthalmology Clinics., 2004. 44(4): p. 17‐39. 54. Klein, R., Klein, B.E.K., Tomany, S.C., and Cruickshanks, K.J., The association of cardiovascular disease with the long‐term incidence of age‐related maculopathy: The beaver dam eye study. Ophthalmology, 2003. 110(6): p. 1273‐1280. 55. Klaver, C.C., Kliffen, M., van Duijn, C.M., Hofman, A., Cruts, M., Grobbee, D.E., van Broeckhoven, C., and de Jong, P.T., Genetic association of apolipoprotein e with age‐
related macular degeneration. American Journal of Human Genetics., 1998. 63(1): p. 200‐6. 56. Schmidt, S., Saunders, A.M., De La Paz, M.A., Postel, E.A., Heinis, R.M., Agarwal, A., Scott, W.K., Gilbert, J.R., McDowell, J.G., Bazyk, A., Gass, J.D., Haines, J.L., and Pericak‐
Vance, M.A., Association of the apolipoprotein e gene with age‐related macular 171 degeneration: Possible effect modification by family history, age, and gender. Molecular Vision., 2000. 6: p. 287‐93. 57. Schmidt, S., Klaver, C., Saunders, A., Postel, E., De La Paz, M., Agarwal, A., Small, K., Udar, N., Ong, J., Chalukya, M., Nesburn, A., Kenney, C., Domurath, R., Hogan, M., Mah, T., Conley, Y., Ferrell, R., Weeks, D., de Jong, P.T., van Duijn, C., Haines, J., Pericak‐Vance, M., and Gorin, M., A pooled case‐control study of the apolipoprotein e (apoe) gene in age‐
related maculopathy. Ophthalmic Genetics., 2002. 23(4): p. 209‐23. 58. Baird, P.N., Richardson, A.J., Robman, L.D., Dimitrov, P.N., Tikellis, G., McCarty, C.A., and Guymer, R.H., Apolipoprotein (apoe) gene is associated with progression of age‐
related macular degeneration (amd). Human Mutation, 2006. 27(4): p. 337‐42. 59. Bojanowski, C.M., Shen, D., Chew, E.Y., Ning, B., Csaky, K.G., Green, W.R., Chan, C.C., and Tuo, J., An apolipoprotein e variant may protect against age‐related macular degeneration through cytokine regulation. Environmental & Molecular Mutagenesis, 2006. 47(8): p. 594‐602. 60. Tikellis, G., Sun, C., Gorin, M.B., Klein, R., Klein, B.E.K., Larsen, E.K.M., Siscovick, D.S., Hubbard, L.D., and Wong, T.Y., Apolipoprotein e gene and age‐related maculopathy in older individuals: The cardiovascular health study. Archives of Ophthalmology, 2007. 125(1): p. 68‐73. 61. Wong, T.Y., Shankar, A., Klein, R., Bray, M.S., Couper, D.J., Klein, B.E., Sharrett, A.R., and Folsom, A.R., Apolipoprotein e gene and early age‐related maculopathy: The atherosclerosis risk in communities study. Ophthalmology, 2006. 113(2): p. 255‐9. 62. Pang, C.P., Baum, L., Chan, W.M., Lau, T.C., Poon, P.M., and Lam, D.S., The apolipoprotein e epsilon4 allele is unlikely to be a major risk factor of age‐related macular degeneration in chinese. Ophthalmologica., 2000. 214(4): p. 289‐91. 63. Klein, R., Klein, B.E.K., Knudtson, M.D., Wong, T.Y., Shankar, A., and Tsai, M.Y., Systemic markers of inflammation, endothelial dysfunction, and age‐related maculopathy. American Journal of Ophthalmology, 2005. 140(1): p. 35.e1‐35.e. 64. Klein, R., Klein, B.E., Marino, E.K., Kuller, L.H., Furberg, C., Burke, G.L., and Hubbard, L.D., Early age‐related maculopathy in the cardiovascular health study. Ophthalmology., 2003. 110(1): p. 25‐33. 65. Dasch, B., Fuhs, A., Behrens, T., Meister, A., Wellmann, J., Fobker, M., Pauleikhoff, D., and Hense, H.W., Inflammatory markers in age‐related maculopathy: Cross‐sectional analysis from the muenster aging and retina study. Archives of Ophthalmology, 2005. 123(11): p. 1501‐
6. 66. Schaumberg, D.A., Christen, W.G., Buring, J.E., Glynn, R.J., Rifai, N., and Ridker, P.M., High‐sensitivity c‐reactive protein, other markers of inflammation, and the incidence of macular degeneration in women. Archives of Ophthalmology, 2007. 125(3): p. 300‐5. 67. Seddon, J.M., Gensler, G., Milton, R.C., Klein, M.L., and Rifai, N., Association between c‐
reactive protein and age‐related macular degeneration. JAMA, 2004. 291(6): p. 704‐10. 68. Seddon, J.M., George, S., Rosner, B., and Rifai, N., Progression of age‐related macular degeneration: Prospective assessment of c‐reactive protein, interleukin 6, and other cardiovascular biomarkers. Archives of Ophthalmology, 2005. 123(6): p. 774‐82. 172 69. Sivaprasad, S., Adewoyin, T., Bailey, T.A., Dandekar, S.S., Jenkins, S., Webster, A.R., and Chong, N.V., Estimation of systemic complement c3 activity in age‐related macular degeneration. Archives of Ophthalmology, 2007. 125(4): p. 515‐9. 70. Wu, K.H.C., Tan, A.G., Rochtchina, E., Favaloro, E.J., Williams, A., Mitchell, P., and Wang, J.J., Circulating inflammatory markers and hemostatic factors in age‐related maculopathy: A population‐based case‐control study. Investigative Ophthalmology & Visual Science, 2007. 48(5): p. 1983‐8. 71. Chua, B., Flood, V., Rochtchina, E., Wang, J.J., Smith, W., and Mitchell, P., Dietary fatty acids and the 5‐year incidence of age‐related maculopathy. Archives of Ophthalmology, 2006. 124(7): p. 981‐6. 72. Tan, J.S.L., Mitchell, P., Rochtchina, E., and Wang, J.J., Statins and the long‐term risk of incident age‐related macular degeneration: The blue mountains eye study. American Journal of Ophthalmology, 2007. 143(4): p. 685‐687. 73. Age‐Related Eye Disease Study Research, G., A randomized, placebo‐controlled, clinical trial of high‐dose supplementation with vitamins c and e, beta carotene, and zinc for age‐
related macular degeneration and vision loss: Areds report no. 8. Archives of Ophthalmology, 2001. 119(10): p. 1417‐36. 74. Chiu, C.J., Hubbard, L.D., Armstrong, J., Rogers, G., Jacques, P.F., Chylack, L.T., Jr., Hankinson, S.E., Willett, W.C., and Taylor, A., Dietary glycemic index and carbohydrate in relation to early age‐related macular degeneration. American Journal of Clinical Nutrition, 2006. 83(4): p. 880‐6. 75. Hammond, C.J., Webster, A.R., Snieder, H., Bird, A.C., Gilbert, C.E., and Spector, T.D., Genetic influence on early age‐related maculopathy : A twin study. Ophthalmology, 2002. 109(4): p. 730‐736. 76. Seddon, J.M., Cote, J., Page, W.F., Aggen, S.H., and Neale, M.C., The us twin study of age‐
related macular degeneration: Relative roles of genetic and environmental influences. Archives of Ophthalmology., 2005. 123(3): p. 321‐7. 77. Seddon, J.M., Ajani, U.A., and Mitchell, B.D., Familial aggregation of age‐related maculopathy. American Journal of Ophthalmology, 1997. 123(2): p. 199‐206. 78. Klaver, C.C., Wolfs, R.C., Assink, J.J., van Duijn, C.M., Hofman, A., and de Jong, P.T., Genetic risk of age‐related maculopathy. Population‐based familial aggregation study. Archives of Ophthalmology., 1998. 116(12): p. 1646‐51. 79. Assink, J.J., Klaver, C.C., Houwing‐Duistermaat, J.J., Wolfs, R.C., van Duijn, C.M., Hofman, A., and de Jong, P.T., Heterogeneity of the genetic risk in age‐related macular disease: A population‐based familial risk study. Ophthalmology, 2005. 112(3): p. 482‐7. 80. Conley, Y.P., Thalamuthu, A., Jakobsdottir, J., Weeks, D.E., Mah, T., Ferrell, R.E., and Gorin, M.B., Candidate gene analysis suggests a role for fatty acid biosynthesis and regulation of the complement system in the etiology of age‐related maculopathy. Hum. Mol. Genet., 2005. 14(14): p. 1991‐2002. 81. Klaver, C.C. and Allikmets, R., Genetics of macular dystrophies and implications for age‐
related macular degeneration. Developments in Ophthalmology., 2003. 37: p. 155‐69. 173 82. Conley, Y.P., Jakobsdottir, J., Mah, T., Weeks, D.E., Klein, R., Kuller, L., Ferrell, R.E., and Gorin, M.B., Cfh, elovl4, plekha1 and loc387715 genes and susceptibility to age‐related maculopathy: Areds and chs cohorts and meta‐analyses. Human Molecular Genetics, 2006. 15(21): p. 3206‐18. 83. Stone, E.M., Braun, T.A., Russell, S.R., Kuehn, M.H., Lotery, A.J., Moore, P.A., Eastman, C.G., Casavant, T.L., and Sheffield, V.C., Missense variations in the fibulin 5 gene and age‐
related macular degeneration. New England Journal of Medicine, 2004. 351(4): p. 346‐53. 84. Tuo, J., Smith, B.C., Bojanowski, C.M., Meleth, A.D., Gery, I., Csaky, K.G., Chew, E.Y., and Chan, C.C., The involvement of sequence variation and expression of cx3cr1 in the pathogenesis of age‐related macular degeneration. FASEB Journal., 2004. 18(11): p. 1297‐9. 85. Zareparsi, S., Buraczynska, M., Branham, K.E.H., Shah, S., Eng, D., Li, M., Pawar, H., Yashar, B.M., Moroi, S.E., Lichter, P.R., Petty, H.R., Richards, J.E., Abecasis, G.R., Elner, V.M., and Swaroop, A., Toll‐like receptor 4 variant d299g is associated with susceptibility to age‐related macular degeneration. Hum. Mol. Genet., 2005. 14(11): p. 1449‐1455. 86. Goverdhan, S.V., Howell, M.W., Mullins, R.F., Osmond, C., Hodgkins, P.R., Self, J., Avery, K., and Lotery, A.J., Association of hla class i and class ii polymorphisms with age‐related macular degeneration. Invest. Ophthalmol. Vis. Sci., 2005. 46(5): p. 1726‐1734. 87. Edwards, A.O., Ritter, R., 3rd, Abel, K.J., Manning, A., Panhuysen, C., and Farrer, L.A., Complement factor h polymorphism and age‐related macular degeneration. Science, 2005. 308(5720): p. 421‐4. 88. Haines, J.L., Hauser, M.A., Schmidt, S., Scott, W.K., Olson, L.M., Gallins, P., Spencer, K.L., Kwan, S.Y., Noureddine, M., Gilbert, J.R., Schnetz‐Boutaud, N., Agarwal, A., Postel, E.A., and Pericak‐Vance, M.A., Complement factor h variant increases the risk of age‐related macular degeneration. Science, 2005. 308(5720): p. 419‐21. 89. Hageman, G.S., Anderson, D.H., Johnson, L.V., Hancox, L.S., Taiber, A.J., Hardisty, L.I., Hageman, J.L., Stockman, H.A., Borchardt, J.D., Gehrs, K.M., Smith, R.J.H., Silvestri, G., Russell, S.R., Klaver, C.C.W., Barbazetto, I., Chang, S., Yannuzzi, L.A., Barile, G.R., Merriam, J.C., Smith, R.T., Olsh, A.K., Bergeron, J., Zernant, J., Merriam, J.E., Gold, B., Dean, M., and Allikmets, R., A common haplotype in the complement regulatory gene factor h (hf1/cfh) predisposes individuals to age‐related macular degeneration. PNAS, 2005. 102(20): p. 7227‐7232. 90. Klein, R.J., Zeiss, C., Chew, E.Y., Tsai, J.Y., Sackler, R.S., Haynes, C., Henning, A.K., Sangiovanni, J.P., Mane, S.M., Mayne, S.T., Bracken, M.B., Ferris, F.L., Ott, J., Barnstable, C., and Hoh, J., Complement factor h polymorphism in age‐related macular degeneration. Science., 2005. 308(5720): p. 385‐9. 91. Chen, L.J., Liu, D.T.L., Tam, P.O.S., Chan, W.M., Liu, K., Chong, K.K.L., Lam, D.S.C., and Pang, C.P., Association of complement factor h polymorphisms with exudative age‐related macular degeneration. Molecular Vision, 2006. 12: p. 1536‐42. 92. Sepp, T., Khan, J.C., Thurlby, D.A., Shahid, H., Clayton, D.G., Moore, A.T., Bird, A.C., and Yates, J.R., Complement factor h variant y402h is a major risk determinant for geographic atrophy and choroidal neovascularization in smokers and nonsmokers. Investigative Ophthalmology & Visual Science, 2006. 47(2): p. 536‐40. 174 93. DeAngelis, M.M., Ji, F., Kim, I.K., Adams, S., Capone, A., Jr., Ott, J., Miller, J.W., and Dryja, T.P., Cigarette smoking, cfh, apoe, elovl4, and risk of neovascular age‐related macular degeneration. Archives of Ophthalmology, 2007. 125(1): p. 49‐54. 94. Seddon, J.M., Francis, P.J., George, S., Schultz, D.W., Rosner, B., and Klein, M.L., Association of cfh y402h and loc387715 a69s with progression of age‐related macular degeneration. JAMA, 2007. 297(16): p. 1793‐1800. 95. Gold, B., Merriam, J.E., Zernant, J., Hancox, L.S., Taiber, A.J., Gehrs, K., Cramer, K., Neel, J., Bergeron, J., Barile, G.R., Smith, R.T., Group, A.M.D.G.C.S., Hageman, G.S., Dean, M., and Allikmets, R., Variation in factor b (bf) and complement component 2 (c2) genes is associated with age‐related macular degeneration. Nature Genetics, 2006. 38(4): p. 458‐62. 96. DeWan, A., Liu, M., Hartman, S., Zhang, S.S.‐M., Liu, D.T.L., Zhao, C., Tam, P.O.S., Chan, W.M., Lam, D.S.C., Snyder, M., Barnstable, C., Pang, C.P., and Hoh, J., Htra1 promoter polymorphism in wet age‐related macular degeneration. Science, 2006. 314(5801): p. 989‐
992. 97. Yang, Z., Camp, N.J., Sun, H., Tong, Z., Gibbs, D., Cameron, D.J., Chen, H., Zhao, Y., Pearson, E., Li, X., Chien, J., Dewan, A., Harmon, J., Bernstein, P.S., Shridhar, V., Zabriskie, N.A., Hoh, J., Howes, K., and Zhang, K., A variant of the htra1 gene increases susceptibility to age‐related macular degeneration. Science, 2006. 314(5801): p. 992‐3. 98. Yoshida, T., DeWan, A., Zhang, H., Sakamoto, R., Okamoto, H., Minami, M., Obazawa, M., Mizota, A., Tanaka, M., Saito, Y., Takagi, I., Hoh, J., and Iwata, T., Htra1 promoter polymorphism predisposes japanese to age‐related macular degeneration. Molecular Vision, 2007. 13: p. 545‐8. 99. Shuler, R.K., Jr., Hauser, M.A., Caldwell, J., Gallins, P., Schmidt, S., Scott, W.K., Agarwal, A., Haines, J.L., Pericak‐Vance, M.A., and Postel, E.A., Neovascular age‐related macular degeneration and its association with loc387715 and complement factor h polymorphism. Archives of Ophthalmology, 2007. 125(1): p. 63‐7. 100. Tanimoto, S., Tamura, H., Ue, T., Yamane, K., Maruyama, H., Kawakami, H., and Kiuchi, Y., A polymorphism of loc387715 gene is associated with age‐related macular degeneration in the japanese population. Neuroscience Letters, 2007. 414(1): p. 71‐4. 101. Pulido, J.S., McConnell, J.P., Lennon, R.J., Bryant, S.C., Peterson, L.M., Berger, P.B., Somers, V., and Highsmith, W.E., Relationship between age‐related macular degeneration‐
associated variants of complement factor h and loc387715 with coronary artery disease. Mayo Clinic Proceedings, 2007. 82(3): p. 301‐7. 102. Ross, R.J., Bojanowski, C.M., Wang, J.J., Chew, E.Y., Rochtchina, E., Ferris, F.L., III, Mitchell, P., Chan, C.‐C., and Tuo, J., The loc387715 polymorphism and age‐related macular degeneration: Replication in three case‐control samples. Invest. Ophthalmol. Vis. Sci., 2007. 48(3): p. 1128‐1132. 103. Schaumberg, D.A., Hankinson, S.E., Guo, Q., Rimm, E., and Hunter, D.J., A prospective study of 2 major age‐related macular degeneration susceptibility alleles and interactions with modifiable risk factors. Archives of Ophthalmology, 2007. 125(1): p. 55‐62. 104. Scholl, H.P.N., Fleckenstein, M., Issa, P.C., Keilhauer, C., Holz, F.G., and Weber, B.H.F., An update on the genetics of age‐related macular degeneration. Molecular Vision, 2007. 13: p. 196‐205. 175 105. Brown, M.M., Brown, G.C., Sharma, S., Stein, J.D., Roth, Z., Campanella, J., and Beauchamp, G.R., The burden of age‐related macular degeneration: A value‐based analysis.[erratum appears in curr opin ophthalmol. 2006 jun;17(3):265]. Current Opinion in Ophthalmology, 2006. 17(3): p. 257‐66. 106. Gehrs, K.M., Anderson, D.H., Johnson, L.V., and Hageman, G.S., Age‐related macular degeneration‐‐emerging pathogenetic and therapeutic concepts. Annals of Medicine, 2006. 38(7): p. 450‐71. 107. Wright, S.E., Keeffe, J.E., and Thies, L.S., Direct costs of blindness in australia. Clinical & Experimental Ophthalmology, 2000. 28(3): p. 140‐2. 108. Brown, G.C., Brown, M.M., Sharma, S., Stein, J.D., Roth, Z., Campanella, J., and Beauchamp, G.R., The burden of age‐related macular degeneration: A value‐based medicine analysis. Transactions of the American Ophthalmological Society, 2005. 103: p. 173‐84. 109. Brown, M.M., Brown, G.C., Stein, J.D., Roth, Z., Campanella, J., and Beauchamp, G.R., Age‐
related macular degeneration: Economic burden and value‐based medicine analysis. Canadian Journal of Ophthalmology, 2005. 40(3): p. 277‐87. 110. Evans, J.R., Antioxidant vitamin and mineral supplements for slowing the progression of age‐related macular degeneration. Cochrane Database of Systematic Reviews, 2006(2): p. CD000254. 111. Guymer, R.H. and Chong, E.W., Modifiable risk factors for age‐related macular degeneration. Medical Journal of Australia, 2006. 184(9): p. 455‐8. 112. Smith, B.T., Joseph, D.P., and Grand, M.G., Treatment of neovascular age‐related macular degeneration: Past, present and future directions. Current Opinion in Ophthalmology, 2007. 18(3): p. 240‐4. 113. Rosenfeld, P.J., Brown, D.M., Heier, J.S., Boyer, D.S., Kaiser, P.K., Chung, C.Y., Kim, R.Y., and Group, M.S., Ranibizumab for neovascular age‐related macular degeneration. New England Journal of Medicine, 2006. 355(14): p. 1419‐31. 114. Hogan, M.J., Alvarado, J.A., and Weddell, J.E., Chapter 9. Retina., in Histology of the human eye. An atlas and textbook. 1971, W.C. Saunders Company: Philadelphia. p. 393‐522. 115. Bok, D. and Young, R.W., Phagocytic properties of the pigment epithelium, in The retinal pigment epithelium, K.M. Zinn and M.F. Marmor, Editors. 1979, Harvard University Press: Cambridge, MA. p. 148‐74. 116. Zarbin, M.A., Current concepts in the pathogenesis of age‐related macular degeneration. Arch Ophthalmol, 2004. 122(4): p. 598‐614. 117. Guymer, R., Luthert, P., and Bird, A., Changes in bruchʹs membrane and related structures with age. Progress in Retinal & Eye Research, 1999. 18(1): p. 59‐90. 118. Guo, L., Hussain, A.A., Limb, G.A., and Marshall, J., Age‐dependent variation in metalloproteinase activity of isolated human bruchʹs membrane and choroid. Investigative Ophthalmology & Visual Science, 1999. 40(11): p. 2676‐82. 176 119. Elner, S.G., Elner, V.M., Kindzelskii, A.L., Horino, K., Davis, H.R., Todd, R.F., 3rd, Glagov, S., and Petty, H.R., Human rpe cell lysis of extracellular matrix: Functional urokinase plasminogen activator receptor (upar), collagenase and elastase. Experimental Eye Research., 2003. 76(5): p. 585‐95. 120. Korte, G.E., Reppucci, V., and Henkind, P., Rpe destruction causes choriocapillary atrophy. Investigative Ophthalmology & Visual Science, 1984. 25(10): p. 1135‐45. 121. Grossniklaus, H.E., Ling, J.X., Wallace, T.M., Dithmar, S., Lawson, D.H., Cohen, C., Elner, V.M., Elner, S.G., and Sternberg, P., Jr., Macrophage and retinal pigment epithelium expression of angiogenic cytokines in choroidal neovascularization. Molecular Vision., 2002. 8: p. 119‐26. 122. Higgins, G.T., Wang, J.H., Dockery, P., Cleary, P.E., and Redmond, H.P., Induction of angiogenic cytokine expression in cultured rpe by ingestion of oxidized photoreceptor outer segments. Investigative Ophthalmology & Visual Science, 2003. 44(4): p. 1775‐82. 123. Mori, K., Duh, E., Gehlbach, P., Ando, A., Takahashi, K., Pearlman, J., Yang, H.S., Zack, D.J., Ettyreddy, D., Brough, D.E., Wei, L.L., and Campochiaro, P.A., Pigment epithelium‐
derived factor inhibits retinal and choroidal neovascularization. Journal of Cellular Physiology, 2001. 188(2): p. 253‐63. 124. Kaplan, H.J., Leibole, M.A., Tezel, T., and Ferguson, T.A., Fas ligand (cd95 ligand) controls angiogenesis beneath the retina. Nature Medicine., 1999. 5(3): p. 292‐7. 125. Feeney‐Burns, L., Hilderbrand, E.S., and Eldridge, S., Aging human rpe: Morphometric analysis of macular, equatorial, and peripheral cells. Invest. Ophthalmol. Vis. Sci., 1984. 25: p. 195‐200. 126. Green, W.R. and Enger, C., Age‐related macular degeneration histopathologic studies. The 1992 lorenz e. Zimmerman lecture. Ophthalmology, 1993. 100(10): p. 1519‐35. 127. van der Schaft, T.L., Mooy, C.M., de Bruijn, W.C., and de Jong, P.T., Early stages of age‐
related macular degeneration: An immunofluorescence and electron microscopy study. British Journal of Ophthalmology, 1993. 77(10): p. 657‐61. 128. Hogan, M.J., Alvarado, J.A., and Weddell, J.E., Chapter 8. Choroid, in Histology of the human eye. An atlas and textbook. 1971, W.C. Saunders Company: Philadelphia. p. 320‐395. 129. Marshall, G.E., Konstas, A.G., Reid, G.G., Edwards, J.G., and Lee, W.R., Type iv collagen and laminin in bruchʹs membrane and basal linear deposit in the human macula. British Journal of Ophthalmology, 1992. 76(10): p. 607‐14. 130. Chong, N.H., Keonin, J., Luthert, P.J., Frennesson, C.I., Weingeist, D.M., Wolf, R.L., Mullins, R.F., and Hageman, G.S., Decreased thickness and integrity of the macular elastic layer of bruchʹs membrane correspond to the distribution of lesions associated with age‐
related macular degeneration. American Journal of Pathology, 2005. 166(1): p. 241‐51. 131. Ahmed, J., Braun, R.D., Dunn, R.J., and Linsenmeier, R.A., Oxygen distribution in the macaque retina. Invest Ophthalmol Vis Sci, 1993. 34: p. 516‐21. 132. Provis, J.M., Penfold, P.L., Cornish, E.E., Sandercoe, T.M., and Madigan, M.C., Anatomy and development of the macula: Specialisation and the vulnerability to macular degeneration. Clinical & Experimental Optometry, 2005. 88(5): p. 269‐81. 177 133. Ramrattan, R.S., van der Schaft, T.L., Mooy, C.M., de Bruijn, W.C., Mulder, P.G., and de Jong, P.T., Morphometric analysis of bruchʹs membrane, the choriocapillaris, and the choroid in aging. Investigative Ophthalmology & Visual Science., 1994. 35(6): p. 2857‐64. 134. Pauleikhoff, D., Harper, C.A., Marshall, J., and Bird, A.C., Aging changes in bruchʹs membrane. A histochemical and morphologic study. Ophthalmology, 1990. 97(2): p. 171‐8. 135. Yoneya, S., Tso, M.O., and Shimizu, K., Patterns of the choriocapillaris. A method to study the choroidal vasculature of the enucleated human eye. International Ophthalmology, 1983. 6(2): p. 95‐9. 136. Lutty, G., Grunwald, J., Majji, A.B., Uyama, M., and Yoneya, S., Changes in choriocapillaris and retinal pigment epithelium in age‐related macular degeneration. Molecular Vision, 1999. 5: p. 35. 137. Hogan, M.J. and Alvarado, J., Studies on the human macula. Iv. Aging changes in bruchʹs membrane. Archives of Ophthalmology, 1967. 77(3): p. 410‐20. 138. Sarks, S.H., New vessel formation beneath the retinal pigment epithelium in senile eyes. British Journal of Ophthalmology, 1973. 57(12): p. 951‐65. 139. Killingsworth, M.C., Age‐related components of bruchʹs membrane in the human eye. Graefes Archive for Clinical & Experimental Ophthalmology, 1987. 225(6): p. 406‐12. 140. van der Schaft, T.L., de Bruijn, W.C., Mooy, C.M., Ketelaars, D.A., and de Jong, P.T., Is basal laminar deposit unique for age‐related macular degeneration? Archives of Ophthalmology., 1991. 109(3): p. 420‐5. 141. Karwatowski, W.S., Jeffries, T.E., Duance, V.C., Albon, J., Bailey, A.J., and Easty, D.L., Preparation of bruchʹs membrane and analysis of the age‐related changes in the structural collagens. British Journal of Ophthalmology, 1995. 79(10): p. 944‐52. 142. Farkas, T.G., Sylvester, V., Archer, D., and Altona, M., The histochemistry of drusen. American Journal of Ophthalmology., 1971. 71(6): p. 1206‐15. 143. Handa, J.T., Verzijl, N., Matsunaga, H., Aotaki‐Keen, A., Lutty, G.A., te Koppele, J.M., Miyata, T., and Hjelmeland, L.M., Increase in the advanced glycation end product pentosidine in bruchʹs membrane with age. Investigative Ophthalmology & Visual Science, 1999. 40(3): p. 775‐9. 144. Holz, F.G., Sheraidah, G., Pauleikhoff, D., and Bird, A.C., Analysis of lipid deposits extracted from human macular and peripheral bruchʹs membrane. Archives of Ophthalmology, 1994. 112(3): p. 402‐6. 145. Curcio, C.A., Millican, C.L., Bailey, T., and Kruth, H.S., Accumulation of cholesterol with age in human bruchʹs membrane. Investigative Ophthalmology & Visual Science., 2001. 42(1): p. 265‐74. 146. Ruberti, J.W., Curcio, C.A., Millican, C.L., Menco, B.P., Huang, J.D., and Johnson, M., Quick‐freeze/deep‐etch visualization of age‐related lipid accumulation in bruchʹs membrane. Investigative Ophthalmology & Visual Science., 2003. 44(4): p. 1753‐9. 178 147. Spaide, R.F., Ho‐Spaide, W.C., Browne, R.W., and Armstrong, D., Characterization of peroxidized lipids in bruchʹs membrane. Retina, 1999. 19(2): p. 141‐7. 148. Starita, C., Hussain, A.A., Pagliarini, S., and Marshall, J., Hydrodynamics of ageing bruchʹs membrane: Implications for macular disease. Experimental Eye Research, 1996. 62(5): p. 565‐
72, 1996. 149. Starita, C., Hussain, A.A., Patmore, A., and Marshall, J., Localization of the site of major resistance to fluid transport in bruchʹs membrane. Investigative Ophthalmology & Visual Science, 1997. 38(3): p. 762‐7. 150. Moore, D.J. and Clover, G.M., The effect of age on the macromolecular permeability of human bruchʹs membrane. Investigative Ophthalmology & Visual Science, 2001. 42(12): p. 2970‐5. 151. Hussain, A.A., Rowe, L., and Marshall, J., Age‐related alterations in the diffusional transport of amino acids across the human bruchʹs‐choroid complex. Journal of the Optical Society of America, A, Optics, Image Science, & Vision, 2002. 19(1): p. 166‐72. 152. Rudolf, M., Winkler, B., Aherrahou, Z., Doehring, L.C., Kaczmarek, P., and Schmidt‐
Erfurth, U., Increased expression of vascular endothelial growth factor associated with accumulation of lipids in bruchʹs membrane of ldl receptor knockout mice. British Journal of Ophthalmology, 2005. 89(12): p. 1627‐30. 153. Witmer, A.N., Vrensen, G.F., Van Noorden, C.J., and Schlingemann, R.O., Vascular endothelial growth factors and angiogenesis in eye disease. Progress in Retinal & Eye Research., 2003. 22(1): p. 1‐29. 154. Medawar, P.B., Immunity to homologous grafted skin: Iii. The fate of skin homografts transplanted to the brain, to subcutaneous tissue, and to the anterior chamber of the eye. British Journal of Experimental Pathology, 1948. 29: p. 58‐69. 155. Roberge, F.G., Caspi, R.R., and Nussenblatt, R.B., Glial retinal muller cells produce il‐1 activity and have a dual effect on autoimmune t helper lymphocytes. Antigen presentation manifested after removal of suppressive activity. Journal of Immunology, 1988. 140(7): p. 2193‐6. 156. Helbig, H., Gurley, R.C., Palestine, A.G., Nussenblatt, R.B., and Caspi, R.R., Dual effect of ciliary body cells on t lymphocyte proliferation. European Journal of Immunology, 1990. 20(11): p. 2457‐63. 157. Griffith, T.S., Brunner, T., Fletcher, S.M., Green, D.R., and Ferguson, T.A., Fas ligand‐
induced apoptosis as a mechanism of immune privilege. Science., 1995. 270(5239): p. 1189‐
92. 158. Streilein, J.W., Immunological non‐responsiveness and acquisition of tolerance in relation to immune privilege in the eye. Eye, 1995. 9(Pt 2): p. 236‐40. 159. Streilein, J.W., Ksander, B.R., and Taylor, A.W., Immune deviation in relation to ocular immune privilege. Journal of Immunology, 1997. 158(8): p. 3557‐60. 160. Streilein, J.W., Ocular immune privilege: The eye takes a dim but practical view of immunity and inflammation. Journal of Leukocyte Biology, 2003. 74(2): p. 179‐85. 179 161. Baudouin, C., Fredj‐Reygrobellet, D., Jambou, D., Gastaud, P., and Lapalus, P., Hla dr and dq expression on human retinal pigment epithelial cells in vitro. Graefes Archive for Clinical & Experimental Ophthalmology., 1990. 228(1): p. 86‐9. 162. Hollborn, M., Enzmann, V., Barth, W., Wiedemann, P., and Kohen, L., Changes in the mrna expression of cytokines and chemokines by stimulated rpe cells in vitro. Current Eye Research, 2000. 20(6): p. 488‐95. 163. Holtkamp, G.M., Kijlstra, A., Peek, R., and de Vos, A.F., Retinal pigment epithelium‐
immune system interactions: Cytokine production and cytokine‐induced changes. Progress in Retinal & Eye Research., 2001. 20(1): p. 29‐48. 164. Zamiri, P., Masli, S., Streilein, J.W., and Taylor, A.W., Pigment epithelial growth factor suppresses inflammation by modulating macrophage activation. Investigative Ophthalmology & Visual Science, 2006. 47(9): p. 3912‐8. 165. Bressler, S.B., Bressler, N.M., Sarks, J.P., and Sarks, S.H., Age‐related macular degeneration: Nonneovascular early amd, intermediate amd, and geographic atrophy, in Retina, S. Ryan and A.P. Schachat, Editors. 2006, Elsevier Mosby: St Louis. p. 1041‐1074. 166. Sarks, S.H., Council lecture. Drusen and their relationship to senile macular degeneration. Australian Journal of Ophthalmology, 1980. 8(2): p. 117‐30. 167. Killingsworth, M.C., Angiogenesis in early choroidal neovascularization secondary to age‐
related macular degeneration. Graefes Archive for Clinical & Experimental Ophthalmology, 1995. 233(6): p. 313‐23. 168. Green, W.R., Histopathology of age‐related macular degeneration. Molecular Vision, 1999. 5: p. 27. 169. Spraul, C.W., Lang, G.E., Grossniklaus, H.E., and Lang, G.K., Histologic and morphometric analysis of the choroid, bruchʹs membrane, and retinal pigment epithelium in postmortem eyes with age‐related macular degeneration and histologic examination of surgically excised choroidal neovascular membranes. Survey of Ophthalmology., 1999. 44(Suppl 1): p. S10‐32. 170. Loffler, K.U. and Lee, W.R., Basal linear deposit in the human macula. Graefes Archive for Clinical & Experimental Ophthalmology, 1986. 224(6): p. 493‐501. 171. Hirata, A. and Feeney‐Burns, L., Autoradiographic studies of aged primate macular retinal pigment epithelium. Investigative Ophthalmology & Visual Science, 1992. 33(7): p. 2079‐90. 172. Grau, S., Richards, P.J., Kerr, B., Hughes, C., Caterson, B., Williams, A.S., Junker, U., Jones, S.A., Clausen, T., and Ehrmann, M., The role of human htra1 in arthritic disease. J. Biol. Chem., 2006. 281(10): p. 6124‐6129. 173. van der Schaft, T.L., de Bruijn, W.C., Mooy, C.M., and de Jong, P.T., Basal laminar deposit in the aging peripheral human retina. Graefes Archive for Clinical & Experimental Ophthalmology, 1993. 231(8): p. 470‐5. 174. Grossniklaus, H.E., Miskala, P.H., Green, W.R., Bressler, S.B., Hawkins, B.S., Toth, C., Wilson, D.J., and Bressler, N.M., Histopathologic and ultrastructural features of surgically excised subfoveal choroidal neovascular lesions: Submacular surgery trials report no. 7. Archives of Ophthalmology, 2005. 123(7): p. 914‐21. 180 175. Spraul, C.W., Lang, G.E., and Grossniklaus, H.E., Morphometric analysis of the choroid, bruchʹs membrane, and retinal pigment epithelium in eyes with age‐related macular degeneration. Investigative Ophthalmology & Visual Science., 1996. 37(13): p. 2724‐35. 176. Curcio, C.A. and Millican, C.L., Basal linear deposit and large drusen are specific for early age‐related maculopathy. Archives of Ophthalmology, 1999. 117(3): p. 329‐39. 177. Feeney‐Burns, L. and Ellersieck, M.R., Age‐related changes in the ultrastructure of bruchʹs membrane. American Journal of Ophthalmology, 1985. 100(5): p. 686‐97. 178. Sarks, J.P., Sarks, S.H., and Killingsworth, M.C., Evolution of soft drusen in age‐related macular degeneration. Eye, 1994. 8(Pt 3): p. 269‐83. 179. Sarks, S.H., Van Driel, D., Maxwell, L., and Killingsworth, M., Softening of drusen and subretinal neovascularization. Transactions of the Ophthalmological Societies of the United Kingdom, 1980. 100(3): p. 414‐22. 180. Sarks, S., Drusen and their relationship to senile macular degeneration. Australian Journal of Ophthalmology, 1980. 8: p. 117‐130. 181. Pauleikhoff, D., Zuels, S., Sheraidah, G.S., Marshall, J., Wessing, A., and Bird, A.C., Correlation between biochemical composition and fluorescein binding of deposits in bruchʹs membrane. Ophthalmology, 1992. 99(10): p. 1548‐53. 182. Curcio, C.A., Presley, J.B., Millican, C.L., and Medeiros, N.E., Basal deposits and drusen in eyes with age‐related maculopathy: Evidence for solid lipid particles. Experimental Eye Research, 2005. 80(6): p. 761‐775. 183. Curcio, C.A., Presley, J.B., Malek, G., Medeiros, N.E., Avery, D.V., and Kruth, H.S., Esterified and unesterified cholesterol in drusen and basal deposits of eyes with age‐
related maculopathy. Experimental Eye Research, 2005. 81(6): p. 731‐41. 184. Malek, G., Li, C.M., Guidry, C., Medeiros, N.E., and Curcio, C.A., Apolipoprotein b in cholesterol‐containing drusen and basal deposits of human eyes with age‐related maculopathy. American Journal of Pathology., 2003. 162(2): p. 413‐25. 185. Li, C.M., Clark, M.E., Chimento, M.F., and Curcio, C.A., Apolipoprotein localization in isolated drusen and retinal apolipoprotein gene expression. Investigative Ophthalmology & Visual Science, 2006. 47(7): p. 3119‐28. 186. Li, C.M., Presley, J.B., Zhang, X., Dashti, N., Chung, B.H., Medeiros, N.E., Guidry, C., and Curcio, C.A., Retina expresses microsomal triglyceride transfer protein: Implications for age‐related maculopathy. Journal of Lipid Research, 2005. 46(4): p. 628‐40. 187. Souied, E.H., Benlian, P., Amouyel, P., Feingold, J., Lagarde, J.P., Munnich, A., Kaplan, J., Coscas, G., and Soubrane, G., The epsilon4 allele of the apolipoprotein e gene as a potential protective factor for exudative age‐related macular degeneration. American Journal of Ophthalmology., 1998. 125(3): p. 353‐9. 188. Lewis, H., Straatsma, B.R., and Foos, R.Y., The prevalence of macular drusen in postmortem eyes. American Journal of Ophthalmology, 1986. 102(6): p. 801‐3. 189. Hageman, G.S. and Mullins, R.F., Molecular composition of drusen as related to substructural phenotype. Molecular Vision., 1999. 5: p. 28. 181 190. Hageman, G.S., Mullins, R.F., Russell, S.R., Johnson, L.V., and Anderson, D.H., Vitronectin is a constituent of ocular drusen and the vitronectin gene is expressed in human retinal pigmented epithelial cells. FASEB Journal., 1999. 13(3): p. 477‐84. 191. Johnson, L.V., Ozaki, S., Staples, M.K., Erickson, P.A., and Anderson, D.H., A potential role for immune complex pathogenesis in drusen formation. Experimental Eye Research., 2000. 70(4): p. 441‐9. 192. Johnson, L.V., Leitner, W.P., Staples, M.K., and Anderson, D.H., Complement activation and inflammatory processes in drusen formation and age related macular degeneration. Experimental Eye Research, 2001. 73(6): p. 887‐96. 193. Coats, G., The structure of the membrane of bruch and its relation to the formation of colloid excrescences. Royal London Ophthalm Hosp Rep, 1905. XVI: p. 164‐178. 194. Hogan, M.J., Role of the retinal pigment epithelium in macular disease. Transactions ‐ American Academy of Ophthalmology & Otolaryngology., 1972. 76(1): p. 64‐80. 195. Burns, R.P. and Feeney‐Burns, L., Clinico‐morphologic correlations of drusen of bruchʹs membrane. Transactions of the American Ophthalmological Society., 1980. 78: p. 206‐25. 196. Sarks, S.H., Arnold, J.J., Killingsworth, M.C., and Sarks, J.P., Early drusen formation in the normal and aging eye and their relation to age related maculopathy: A clinicopathological study. British Journal of Ophthalmology, 1999. 83(3): p. 358‐68. 197. Farkas, T.G., Sylvester, V., and Archer, D., The ultrastructure of drusen. American Journal of Ophthalmology., 1971. 71(6): p. 1196‐205. 198. Ishibashi, T., Patterson, R., Ohnishi, Y., Inomata, H., and Ryan, S.J., Formation of drusen in the human eye. American Journal of Ophthalmology, 1986. 101(3): p. 342‐53. 199. Malorni, W., Iosi, F., Mirabelli, F., and Bellomo, G., Cytoskeleton as a target in menadione‐
induced oxidative stress in cultured mammalian cells: Alterations underlying surface bleb formation. Chemico‐Biological Interactions, 1991. 80(2): p. 217‐36. 200. Malorni, W. and Donelli, G., Cell death. General features and morphological aspects. Annals of the New York Academy of Sciences, 1992. 663: p. 218‐33. 201. Marin‐Castano, M.E., Csaky, K.G., and Cousins, S.W., Nonlethal oxidant injury to human retinal pigment epithelium cells causes cell membrane blebbing but decreased mmp‐2 activity. Investigative Ophthalmology & Visual Science, 2005. 46(9): p. 3331‐40. 202. Crabb, J.W., Miyagi, M., Gu, X., Shadrach, K., West, K.A., Sakaguchi, H., Kamei, M., Hasan, A., Yan, L., Rayborn, M.E., Salomon, R.G., and Hollyfield, J.G., Drusen proteome analysis: An approach to the etiology of age‐related macular degeneration. Proceedings of the National Academy of Sciences of the United States of America., 2002. 99(23): p. 14682‐7. 203. Fliesler, A.J. and Anderson, R.E., Chemistry and metabolism of lipids in the vertebrate retina. Progress in Lipid Research, 1983. 22(2): p. 79‐131. 204. Bressler, N.M., Silva, J.C., Bressler, S.B., Fine, S.L., and Green, W.R., Clinicopathologic correlation of drusen and retinal pigment epithelial abnormalities in age‐related macular degeneration. Retina., 1994. 14(2): p. 130‐42. 182 205. Curcio, C.A., Medeiros, N.E., and Millican, C.L., The alabama age‐related macular degeneration grading system for donor eyes. Investigative Ophthalmology & Visual Science, 1998. 39(7): p. 1085‐96. 206. OʹConnell, E., Neelam, K., Nolan, J., Au Eong, K.G., and Beatty, S., Macular carotenoids and age‐related maculopathy. Ann Acad Med Singapore, 2006. 35(11): p. 821‐30. 207. Decanini, A., Nordgaard, C.L., Feng, X., Ferrington, D.A., and Olsen, T.W., Changes in select redox proteins of the retinal pigment epithelium in age‐related macular degeneration. American Journal of Ophthalmology, 2007. 143(4): p. 607‐15. 208. Liles, M.R., Newsome, D.A., and Oliver, P.D., Antioxidant enzymes in the aging human retinal pigment epithelium. Archives of Ophthalmology, 1991. 109(9): p. 1285‐8. 209. Tate, D.J., Jr., Newsome, D.A., and Oliver, P.D., Metallothionein shows an age‐related decrease in human macular retinal pigment epithelium. Investigative Ophthalmology & Visual Science, 1993. 34(7): p. 2348‐51. 210. Zarbin, M., Should corticosteroids be considered as part of the standard care with photodynamic therapy? Archives of Ophthalmology, 2006. 124(4): p. 563‐71. 211. Cohen, S.M., Olin, K.L., Feuer, W.J., Hjelmeland, L., Keen, C.L., and Morse, L.S., Low glutathione reductase and peroxidase activity in age‐related macular degeneration. British Journal of Ophthalmology, 1994. 78(10): p. 791‐4. 212. Delcourt, C., Cristol, J.P., Leger, C.L., Descomps, B., and Papoz, L., Associations of antioxidant enzymes with cataract and age‐related macular degeneration. The pola study. Pathologies oculaires liees a lʹage. Ophthalmology, 1999. 106(2): p. 215‐22. 213. Binder, C.J., Shaw, P.X., Chang, M.‐K., Boullier, A., Hartvigsen, K., Horkko, S., Miller, Y.I., Woelkers, D.A., Corr, M., and Witztum, J.L., Thematic review series: The immune system and atherogenesis. The role of natural antibodies in atherogenesis. J. Lipid Res., 2005. 46(7): p. 1353‐1363. 214. Handa, J.T., New molecular histopathologic insights into the pathogenesis of age‐related macular degeneration. International Ophthalmology Clinics, 2007. 47(1): p. 15‐50. 215. Sarks, S.H., Penfold, P.L., Killingsworth, M.C., Factors influencing choroidal neovascularization in senile macular degeneration, in Ophthalmology, K. Mizuno, Editor. 1985, Elsevier Science Publishers Biomedical Division. 216. Penfold, P., Killingsworth, M., and Sarks, S., An ultrastructural study of the role of leucocytes and fibroblasts in the breakdown of bruchʹs membrane. Australian Journal of Ophthalmology, 1984. 12(1): p. 23‐31. 217. Penfold, P.L., Killingsworth, M.C., and Sarks, S.H., Senile macular degeneration: The involvement of immunocompetent cells. Graefes Archive for Clinical & Experimental Ophthalmology, 1985. 223(2): p. 69‐76. 218. Killingsworth, M.C., Sarks, J.P., and Sarks, S.H., Macrophages related to bruchʹs membrane in age‐related macular degeneration. Eye, 1990. 4(Pt 4): p. 613‐21. 183 219. Sarks, J.P., Sarks, S.H., and Killingsworth, M.C., Morphology of early choroidal neovascularisation in age‐related macular degeneration: Correlation with activity. Eye., 1997. 11(Pt 4): p. 515‐22. 220. Xiong, M., Elson, G., Legarda, D., and Leibovich, S.J., Production of vascular endothelial growth factor by murine macrophages: Regulation by hypoxia, lactate, and the inducible nitric oxide synthase pathway. American Journal of Pathology., 1998. 153(2): p. 587‐98. 221. Crowther, M., Brown, N.J., Bishop, E.T., and Lewis, C.E., Microenvironmental influence on macrophage regulation of angiogenesis in wounds and malignant tumors. Journal of Leukocyte Biology., 2001. 70(4): p. 478‐90. 222. Penfold, P.L., Provis, J.M., and Billson, F.A., Age‐related macular degeneration: Ultrastructural studies of the relationship of leucocytes to angiogenesis. Graefes Archive for Clinical & Experimental Ophthalmology., 1987. 225(1): p. 70‐6. 223. Penfold, P.L., Killingsworth, M.C., and Sarks, S.H., Senile macular degeneration. The involvement of giant cells in atrophy of the retinal pigment epithelium. Investigative Ophthalmology & Visual Science, 1986. 27(3): p. 364‐71. 224. Oh, H., Takagi, H., Takagi, C., Suzuma, K., Otani, A., Ishida, K., Matsumura, M., Ogura, Y., and Honda, Y., The potential angiogenic role of macrophages in the formation of choroidal neovascular membranes. Investigative Ophthalmology & Visual Science., 1999. 40(9): p. 1891‐8. 225. Tobe, T., Ortega, S., Luna, J.D., Ozaki, H., Okamoto, N., Derevjanik, N.L., Vinores, S.A., Basilico, C., and Campochiaro, P.A., Targeted disruption of the fgf2 gene does not prevent choroidal neovascularization in a murine model. American Journal of Pathology, 1998. 153(5): p. 1641‐6. 226. Ishibashi, T., Hata, Y., Yoshikawa, H., Nakagawa, K., Sueishi, K., and Inomata, H., Expression of vascular endothelial growth factor in experimental choroidal neovascularization. Graefes Archive for Clinical & Experimental Ophthalmology, 1997. 235(3): p. 159‐67. 227. Yi, X., Ogata, N., Komada, M., Yamamoto, C., Takahashi, K., Omori, K., and Uyama, M., Vascular endothelial growth factor expression in choroidal neovascularization in rats. Graefes Archive for Clinical & Experimental Ophthalmology, 1997. 235(5): p. 313‐9. 228. Tsutsumi, C., Sonoda, K.H., Egashira, K., Qiao, H., Hisatomi, T., Nakao, S., Ishibashi, M., Charo, I.F., Sakamoto, T., Murata, T., and Ishibashi, T., The critical role of ocular‐
infiltrating macrophages in the development of choroidal neovascularization. Journal of Leukocyte Biology., 2003. 74(1): p. 25‐32. 229. Sakurai, E., Anand, A., Ambati, B.K., van Rooijen, N., and Ambati, J., Macrophage depletion inhibits experimental choroidal neovascularization. Investigative Ophthalmology & Visual Science, 2003. 44(8): p. 3578‐85. 230. Caicedo, A., Espinosa‐Heidmann, D.G., Pina, Y., Hernandez, E.P., and Cousins, S.W., Blood‐derived macrophages infiltrate the retina and activate muller glial cells under experimental choroidal neovascularization. Experimental Eye Research, 2005. 81(1): p. 38‐47. 231. Ambati, J., Anand, A., Fernandez, S., Sakurai, E., Lynn, B.C., Kuziel, W.A., Rollins, B.J., and Ambati, B.K., An animal model of age‐related macular degeneration in senescent ccl‐2‐ or ccr‐2‐deficient mice.[see comment]. Nature Medicine., 2003. 9(11): p. 1390‐7. 184 232. Kamei, M., Yoneda, K., Kume, N., Suzuki, M., Itabe, H., Matsuda, K.‐I., Shimaoka, T., Minami, M., Yonehara, S., Kita, T., and Kinoshita, S., Scavenger receptors for oxidized lipoprotein in age‐related macular degeneration. Investigative Ophthalmology & Visual Science, 2007. 48(4): p. 1801‐7. 233. Tuo, J., Bojanowski, C.M., Zhou, M., Shen, D., Ross, R.J., Rosenberg, K.I., Cameron, D.J., Yin, C., Kowalak, J.A., Zhuang, Z., Zhang, K., and Chan, C.‐C., Murine ccl2/cx3cr1 deficiency results in retinal lesions mimicking human age‐related macular degeneration. Investigative Ophthalmology & Visual Science, 2007. 48(8): p. 3827‐36. 234. Apte, R.S., Richter, J., Herndon, J., and Ferguson, T.A., Macrophages inhibit neovascularization in a murine model of age‐related macular degeneration. PLoS Medicine / Public Library of Science, 2006. 3(8): p. e310. 235. Ancuta, P., Rao, R., Moses, A., Mehle, A., Shaw, S.K., Luscinskas, F.W., and Gabuzda, D., Fractalkine preferentially mediates arrest and migration of cd16+ monocytes. Journal of Experimental Medicine, 2003. 197(12): p. 1701‐7. 236. Yu, F.‐S.X. and Hazlett, L.D., Toll‐like eeceptors and the eye. Invest. Ophthalmol. Vis. Sci., 2006. 47(4): p. 1255‐1263. 237. Elner, S.G., Petty, H.R., Elner, V.M., Yoshida, A., Bian, Z.M., Yang, D., and Kindezelskii, A.L., Tlr4 mediates human retinal pigment epithelial endotoxin binding and cytokine expression. Transactions of the American Ophthalmological Society, 2005. 103: p. 126‐35. 238. Kindzelskii, A.L., Elner, V.M., Elner, S.G., Yang, D., Hughes, B.A., and Petty, H.R., Toll‐like receptor 4 (tlr4) of retinal pigment epithelial cells participates in transmembrane signaling in response to photoreceptor outer segments. J. Gen. Physiol., 2004. 124(2): p. 139‐149. 239. Blumenkranz, M.S., Russell, S.R., Robey, M.G., Kott‐Blumenkranz, R., and Penneys, N., Risk factors in age‐related maculopathy complicated by choroidal neovascularization. Ophthalmology., 1986. 93(5): p. 552‐8. 240. Klein, R., Klein, B.E., and Franke, T., The relationship of cardiovascular disease and its risk factors to age‐related maculopathy. The beaver dam eye study. Ophthalmology, 1993. 100(3): p. 406‐14. 241. Klein, R., Klein, B.E., Tomany, S.C., and Cruickshanks, K.J., Association of emphysema, gout, and inflammatory markers with long‐term incidence of age‐related maculopathy. Archives of Ophthalmology., 2003. 121(5): p. 674‐8. 242. Cousins, S.W., Espinosa‐Heidmann, D.G., and Csaky, K.G., Monocyte activation in patients with age‐related macular degeneration: A biomarker of risk for choroidal neovascularization? Archives of Ophthalmology, 2004. 122(7): p. 1013‐8. 243. Mills, C.D., Kincaid, K., Alt, J.M., Heilman, M.J., and Hill, A.M., M‐1/m‐2 macrophages and the th1/th2 paradigm. Journal of Immunology, 2000. 164(12): p. 6166‐73. 244. McMenamin, P.G., Dendritic cells and macrophages in the uveal tract of the normal mouse eye. British Journal of Ophthalmology., 1999. 83(5): p. 598‐604. 245. Forrester, J.V., Lumsden, L., Duncan, L., and Dick, A.D., Choroidal dendritic cells require activation to present antigen and resident choroidal macrophages potentiate this response. British Journal of Ophthalmology, 2005. 89(3): p. 369‐77. 185 246. Zamiri, P., Masli, S., Kitaichi, N., Taylor, A.W., and Streilein, J.W., Thrombospondin plays a vital role in the immune privilege of the eye. Investigative Ophthalmology & Visual Science, 2005. 46(3): p. 908‐19. 247. Hattenbach, L.O., Falk, B., Nurnberger, F., Koch, F.H., and Ohrloff, C., Detection of inducible nitric oxide synthase and vascular endothelial growth factor in choroidal neovascular membranes. Ophthalmologica, 2002. 216(3): p. 209‐14. 248. Ando, A., Yang, A., Mori, K., Yamada, H., Yamada, E., Takahashi, K., Saikia, J., Kim, M., Melia, M., Fishman, M., Huang, P., and Campochiaro, P.A., Nitric oxide is proangiogenic in the retina and choroid. Journal of Cellular Physiology., 2002. 191(1): p. 116‐24. 249. Zhang, J., Wu, L.Y., Wu, G.S., and Rao, N.A., Differential expression of nitric oxide synthase in experimental uveoretinitis. Investigative Ophthalmology & Visual Science, 1999. 40(9): p. 1899‐905. 250. Hageman, G.S., Luthert, P.J., Victor Chong, N.H., Johnson, L.V., Anderson, D.H., and Mullins, R.F., An integrated hypothesis that considers drusen as biomarkers of immune‐
mediated processes at the rpe‐bruchʹs membrane interface in aging and age‐related macular degeneration. Progress in Retinal & Eye Research., 2001. 20(6): p. 705‐32. 251. Newsome, D.A., Hewitt, A.T., Huh, W., Robey, P.G., and Hassell, J.R., Detection of specific extracellular matrix molecules in drusen, bruchʹs membrane, and ciliary body. American Journal of Ophthalmology. 104(4), 1987: p. 373‐81. 252. Russell, S.R., Mullins, R.F., Schneider, B.L., and Hageman, G.S., Location, substructure, and composition of basal laminar drusen compared with drusen associated with aging and age‐related macular degeneration. American Journal of Ophthalmology, 2000. 129(2): p. 205‐
14. 253. Mullins, R.F., Russell, S.R., Anderson, D.H., and Hageman, G.S., Drusen associated with aging and age‐related macular degeneration contain proteins common to extracellular deposits associated with atherosclerosis, elastosis, amyloidosis, and dense deposit disease. FASEB Journal, 2000. 14(7): p. 835‐46. 254. Abbas, A.K., Lichtman, A.H., and Pober, J.S., Chapter 13. The complement system, in Cellular and molecular immunology. 1991, W.B. Saunders Company: Philadelphia. p. 259‐282. 255. Anderson, D.H., Mullins, R.F., Hageman, G.S., and Johnson, L.V., A role for local inflammation in the formation of drusen in the aging eye. American Journal of Ophthalmology., 2002. 134(3): p. 411‐31. 256. Abbas, A.K., Lichtman, A.H., and Pober, J.S., Chapter 11. Cytokines, in Cellular and molecular immunology. 1991, W.B.Saunders Company: Philadelphia. p. 225‐243. 257. Schultz, D.R. and Arnold, P.I., Properties of four acute phase proteins: C‐reactive protein, serum amyloid a protein, alpha 1‐acid glycoprotein, and fibrinogen. Seminars in Arthritis & Rheumatism., 1990. 20(3): p. 129‐47. 258. Bhakdi, S., Torzewski, M., Klouche, M., and Hemmes, M., Complement and atherogenesis: Binding of crp to degraded, nonoxidized ldl enhances complement activation. Arteriosclerosis, Thrombosis & Vascular Biology., 1999. 19(10): p. 2348‐54. 186 259. Fuse, N., Miyazawa, A., Mengkegale, M., Yoshida, M., Wakusawa, R., Abe, T., and Tamai, M., Polymorphisms in complement factor h and hemicentin‐1 genes in a japanese population with dry‐type age‐related macular degeneration. American Journal of Ophthalmology, 1074. 142(6): p. 1074‐6. 260. Uka, J., Tamura, H., Kobayashi, T., Yamane, K., Kawakami, H., Minamoto, A., and Mishima, H.K., No association of complement factor h gene polymorphism and age‐related macular degeneration in the japanese population. Retina, 2006. 26(9): p. 985‐7. 261. Tedeschi‐Blok, N., Buckley, J., Varma, R., Triche, T.J., and Hinton, D.R., Population‐based study of early age‐related macular degeneration: Role of the complement factor h y402h polymorphism in bilateral but not unilateral disease. Ophthalmology, 2007. 114(1): p. 99‐103. 262. Rodriguez de Cordoba, S., Esparza‐Gordillo, J., Goicoechea de Jorge, E., Lopez‐Trascasa, M., and Sanchez‐Corral, P., The human complement factor h: Functional roles, genetic variations and disease associations. Molecular Immunology, 2004. 41(4): p. 355‐67. 263. Johnson, P.T., Betts, K.E., Radeke, M.J., Hageman, G.S., Anderson, D.H., and Johnson, L.V., Individuals homozygous for the age‐related macular degeneration risk‐conferring variant of complement factor h have elevated levels of crp in the choroid. Proceedings of the National Academy of Sciences of the United States of America, 2006. 103(46): p. 17456‐61. 264. Kalayoglu, M.V., Galvan, C., Mahdi, O.S., Byrne, G.I., and Mansour, S., Serological association between chlamydia pneumoniae infection and age‐related macular degeneration. Archives of Ophthalmology., 2003. 121(4): p. 478‐82. 265. Robman, L., Mahdi, O., McCarty, C., Dimitrov, P., Tikellis, G., McNeil, J., Byrne, G., Taylor, H., and Guymer, R., Exposure to chlamydia pneumoniae infection and progression of age‐
related macular degeneration. American Journal of Epidemiology, 2005. 161(11): p. 1013‐9. 266. Kalayoglu, M.V., Bula, D., Arroyo, J., Gragoudas, E.S., DʹAmico, D., and Miller, J.W., Identification of chlamydia pneumoniae within human choroidal neovascular membranes secondary to age‐related macular degeneration. Graefes Archive for Clinical & Experimental Ophthalmology, 2005. 243(11): p. 1080‐90. 267. Blake, G.J. and Ridker, P.M., Inflammatory bio‐markers and cardiovascular risk prediction. Journal of Internal Medicine., 2002. 252(4): p. 283‐94. 268. Smith, W., Mitchell, P., Leeder, S.R., and Wang, J.J., Plasma fibrinogen levels, other cardiovascular risk factors, and age‐related maculopathy: The blue mountains eye study. Archives of Ophthalmology, 1998. 116(5): p. 583‐7. 269. Evereklioglu, C., Doganay, S., Er, H., Cekmen, M., Ozerol, E., and Otlu, B., Serum leptin concentrations are decreased and correlated with disease severity in age‐related macular degeneration: A preliminary study. Eye., 2003. 17(3): p. 350‐5. 270. Vine, A.K., Stader, J., Branham, K., Musch, D.C., and Swaroop, A., Biomarkers of cardiovascular disease as risk factors for age‐related macular degeneration. Ophthalmology, 2005. 112(12): p. 2076‐80. 271. Abbas, A.K., Lichtman, A.H., and Pober, J.S., Chapter 18. Diseases caused by humoral and cell‐mediated immune reactions, in Cellular and molecular immunology. 1991, W.B. Saunders Company: Philadelphia. p. 354‐376. 187 272. Bouras, C., Riederer, B.M., Kovari, E., Hof, P.R., and Giannakopoulos, P., Humoral immunity in brain aging and alzheimerʹs disease. Brain Research ‐ Brain Research Reviews, 2005. 48(3): p. 477‐87. 273. Hansson, G.K., Libby, P., Schonbeck, U., and Yan, Z.‐Q., Innate and adaptive immunity in the pathogenesis of atherosclerosis. Circ Res, 2002. 91(4): p. 281‐291. 274. Witebsky, E., Rose, N.R., Terplan, K., Paine, J.R., and Egan, R.W., Chronic thyroiditis and autoimmunization. Journal of the American Medical Association, 1957. 164(13): p. 1439‐47. 275. Thirkill, C.E., Roth, A.M., and Keltner, J.L., Cancer‐associated retinopathy. Archives of Ophthalmology, 1987. 105(3): p. 372‐5. 276. Thirkill, C.E., Tait, R.C., Tyler, N.K., Roth, A.M., and Keltner, J.L., The cancer‐associated retinopathy antigen is a recoverin‐like protein. Investigative Ophthalmology & Visual Science, 1992. 33(10): p. 2768‐72. 277. Adamus, G., Guy, J., Schmied, J.L., Arendt, A., and Hargrave, P.A., Role of anti‐recoverin autoantibodies in cancer‐associated retinopathy. Investigative Ophthalmology & Visual Science, 1993. 34(9): p. 2626‐33. 278. Adamus, G., Machnicki, M., and Seigel, G.M., Apoptotic retinal cell death induced by antirecoverin autoantibodies of cancer‐associated retinopathy. Investigative Ophthalmology & Visual Science, 1997. 38(2): p. 283‐91. 279. Adamus, G., Machnicki, M., Elerding, H., Sugden, B., Blocker, Y.S., and Fox, D.A., Antibodies to recoverin induce apoptosis of photoreceptor and bipolar cells in vivo. Journal of Autoimmunity, 1998. 11(5): p. 523‐33. 280. Whitcup, S.M., Vistica, B.P., Milam, A.H., Nussenblatt, R.B., and Gery, I., Recoverin‐
associated retinopathy: A clinically and immunologically distinctive disease. American Journal of Ophthalmology, 1998. 126(2): p. 230‐7. 281. Mizener, J.B., Kimura, A.E., Adamus, G., Thirkill, C.E., Goeken, J.A., and Kardon, R.H., Autoimmune retinopathy in the absence of cancer. American Journal of Ophthalmology, 1997. 123(5): p. 607‐18. 282. Heckenlively, J.R., Fawzi, A.A., Oversier, J., Jordan, B.L., and Aptsiauri, N., Autoimmune retinopathy: Patients with antirecoverin immunoreactivity and panretinal degeneration. Archives of Ophthalmology, 2000. 118(11): p. 1525‐33. 283. Adamus, G., Aptsiauri, N., Guy, J., Heckenlively, J., Flannery, J., and Hargrave, P.A., The occurrence of serum autoantibodies against enolase in cancer‐associated retinopathy. Clinical Immunology & Immunopathology, 1996. 78(2): p. 120‐9. 284. Adamus, G., Amundson, D., Seigel, G.M., and Machnicki, M., Anti‐enolase‐alpha autoantibodies in cancer‐associated retinopathy: Epitope mapping and cytotoxicity on retinal cells. Journal of Autoimmunity, 1998. 11(6): p. 671‐7. 285. Ohguro, H., Ogawa, K., and Nakagawa, T., Recoverin and hsc 70 are found as autoantigens in patients with cancer‐associated retinopathy. Investigative Ophthalmology & Visual Science, 1999. 40(1): p. 82‐9. 188 286. Ohguro, H., Ogawa, K., Maeda, T., Maeda, A., and Maruyama, I., Cancer‐associated retinopathy induced by both anti‐recoverin and anti‐hsc70 antibodies in vivo. Investigative Ophthalmology & Visual Science, 1999. 40(13): p. 3160‐7. 287. Alabduljalil, T. and Behbehani, R., Paraneoplastic syndromes in neuro‐ophthalmology. Current Opinion in Ophthalmology, 2007. 18(6): p. 463‐9. 288. Pfohler, C., Preuss, K.‐D., Tilgen, W., Stark, A., Regitz, E., Fadle, N., and Pfreundschuh, M., Mitofilin and titin as target antigens in melanoma‐associated retinopathy. International Journal of Cancer, 2007. 120(4): p. 788‐95. 289. Heckenlively, J.R., Aptsiauri, N., Nusinowitz, S., Peng, C., and Hargrave, P.A., Investigations of antiretinal antibodies in pigmentary retinopathy and other retinal degenerations. Transactions of the American Ophthalmological Society, 1996. 94: p. 179‐200; discussion 200‐6. 290. Attawia, M.A. and Nayak, R.C., Circulating antipericyte autoantibodies in diabetic retinopathy. Retina, 1999. 19(5): p. 390‐400. 291. Heckenlively, J.R., Jordan, B.L., and Aptsiauri, N., Association of antiretinal antibodies and cystoid macular edema in patients with retinitis pigmentosa. American Journal of Ophthalmology, 1999. 127(5): p. 565‐73. 292. Ikeda, Y., Maruyama, I., Nakazawa, M., and Ohguro, H., Clinical significance of serum antibody against neuron‐specific enolase in glaucoma patients. Japanese Journal of Ophthalmology., 2002. 46(1): p. 13‐7. 293. Maruyama, I., Ikeda, Y., Nakazawa, M., and Ohguro, H., Clinical roles of serum autoantibody against neuron‐specific enolase in glaucoma patients. Tohoku Journal of Experimental Medicine., 2002. 197(3): p. 125‐32. 294. Maruyama, I., Maeda, T., Okisaka, S., Mizukawa, A., Nakazawa, M., and Ohguro, H., Autoantibody against neuron‐specific enolase found in glaucoma patients causes retinal dysfunction in vivo. Japanese Journal of Ophthalmology., 2002. 46(1): p. 1‐12. 295. Weleber, R.G., Watzke, R.C., Shults, W.T., Trzupek, K.M., Heckenlively, J.R., Egan, R.A., and Adamus, G., Clinical and electrophysiologic characterization of paraneoplastic and autoimmune retinopathies associated with antienolase antibodies. American Journal of Ophthalmology, 2005. 139(5): p. 780‐794. 296. Grus, F.H., Joachim, S.C., Hoffmann, E.M., and Pfeiffer, N., Complex autoantibody repertoires in patients with glaucoma. Molecular Vision., 2004. 10: p. 132‐7. 297. Joachim, S.C., Grus, F.H., and Pfeiffer, N., Analysis of autoantibody repertoires in sera of patients with glaucoma. European Journal of Ophthalmology., 2003. 13(9‐10): p. 752‐8. 298. Chant, S.M., Heckenlively, J., and Meyers‐Elliott, R.H., Autoimmunity in hereditary retinal degeneration. I. Basic studies. British Journal of Ophthalmology, 1985. 69(1): p. 19‐24. 299. Penfold, P.L., Provis, J.M., Furby, J.H., Gatenby, P.A., and Billson, F.A., Autoantibodies to retinal astrocytes associated with age‐related macular degeneration. Graefes Archive for Clinical & Experimental Ophthalmology, 1990. 228(3): p. 270‐4. 189 300. Gurne, D.H., Tso, M.O., Edward, D.P., and Ripps, H., Antiretinal antibodies in serum of patients with age‐related macular degeneration. Ophthalmology, 1991. 98(5): p. 602‐7. 301. Gu, X., Meer, S.G., Miyagi, M., Rayborn, M.E., Hollyfield, J.G., Crabb, J.W., and Salomon, R.G., Carboxyethylpyrrole protein adducts and autoantibodies, biomarkers for age‐related macular degeneration. Journal of Biological Chemistry., 2003. 278(43): p. 42027‐35. 302. Patel, N., Ohbayashi, M., Nugent, A.K., Ramchand, K., Toda, M., Chau, K.Y., Bunce, C., Webster, A., Bird, A.C., Ono, S.J., and Chong, V., Circulating anti‐retinal antibodies as immune markers in age‐related macular degeneration. Immunology, 2005. 115: p. 422‐30. 303. Joachim, S., Bruns, K., Lackner, K., Pfeiffer, N., and Grus, F., Analysis of igg antibody patterns against retinal antigens and antibodies to alpha‐crystallin, gfap, and alpha‐enolase in sera of patients with ʺwetʺ age‐related macular degeneration. Graefeʹs Archive for Clinical and Experimental Ophthalmology, 2007. 245(5): p. 619‐626. 304. Avrameas, S., Natural autoantibodies: From ʹhorror autotoxicusʹ to ʹgnothi seautonʹ. Immunology Today., 1991. 12(5): p. 154‐9. 305. Avrameas, S. and Ternynck, T., Natural autoantibodies: The other side of the immune system. Research in Immunology., 1995. 146(4‐5): p. 235‐48. 306. Hayakawa, K., Asano, M., Shinton, S.A., Gui, M., Allman, D., Stewart, C.L., Silver, J., and Hardy, R.R., Positive selection of natural autoreactive b cells. Science, 1999. 285(5424): p. 113‐6. 307. Lacroix‐Desmazes, S., Kaveri, S.V., Mouthon, L., Ayouba, A., Malanchere, E., Coutinho, A., and Kazatchkine, M.D., Self‐reactive antibodies (natural autoantibodies) in healthy individuals. Journal of Immunological Methods., 1998. 216(1‐2): p. 117‐37. 308. Lacroix‐Desmazes, S., Mouthon, L., Kaveri, S.V., Kazatchkine, M.D., and Weksler, M.E., Stability of natural self‐reactive antibody repertoires during aging. Journal of Clinical Immunology., 1999. 19(1): p. 26‐34. 309. Coutinho, A., Kazatchkine, M.D., and Avrameas, S., Natural autoantibodies. Current Opinion in Immunology., 1995. 7(6): p. 812‐8. 310. Guilbert, B., Dighiero, G., and Avrameas, S., Naturally occurring antibodies against nine common antigens in human sera. I. Detection, isolation and characterization. Journal of Immunology., 1982. 128(6): p. 2779‐87. 311. Wu, K.H., Madigan, M.C., Billson, F.A., and Penfold, P.L., Differential expression of gfap in early v late amd: A quantitative analysis. British Journal of Ophthalmology., 2003. 87(9): p. 1159‐66. 312. Augusteyn, R.C., Alpha‐crystallin: A review of its structure and function. Clinical & Experimental Optometry, 2004. 87(6): p. 356‐66. 313. Nakata, K., Crabb, J.W., and Hollyfield, J.G., Crystallin distribution in bruchʹs membrane‐
choroid complex from amd and age‐matched donor eyes. Experimental Eye Research, 2005. 80(6): p. 821‐826. 190 314. Aaronson, R.M., Graven, K.K., Tucci, M., McDonald, R.J., and Farber, H.W., Non‐neuronal enolase is an endothelial hypoxic stress protein. Journal of Biological Chemistry, 1995. 270(46): p. 27752‐7. 315. Gitlits, V.M., Toh, B.H., and Sentry, J.W., Disease association, origin, and clinical relevance of autoantibodies to the glycolytic enzyme enolase. Journal of Investigative Medicine, 2001. 49(2): p. 138‐45. 316. Carayannopoulos, M.O., Potter, K.N., Li, Y., Natvig, J.B., and Capra, J.D., Evidence that human immunoglobulin m rheumatoid factors can be derived from the natural autoantibody pool and undergo an antigen driven immune response in which somatically mutated rheumatoid factors have lower affinities for immunoglobulin g fc than their germline counterparts. Scandinavian Journal of Immunology., 2000. 51(4): p. 327‐36. 317. Adamus, G., Antirecoverin antibodies and autoimmune retinopathy. Archives of Ophthalmology, 2000. 118(11): p. 1577‐8. 318. Dubovy, S.R., Hairston, R.J., Schatz, H., Schachat, A.P., Bressler, N.M., Finkelstein, D., and Green, W.R., Adult‐onset foveomacular pigment epithelial dystrophy: Clinicopathologic correlation of three cases. Retina, 2000. 20(6): p. 638‐49. 319. Green, W.R., McDonnell, P.J., and Yeo, J.H., Pathologic features of senile macular degeneration. Ophthalmology, 1985. 92(5): p. 615‐627. 320. Age‐Related Eye Disease Study Research Group, Risk factors for the incidence of advanced age‐related macular degeneration in the age‐related eye disease study (areds): Areds report no. 19. Ophthalmology, 2005. 112(4): p. 533‐539. 321. Bressler, S.B., Maguire, M.G., Bressler, N.M., and Fine, S.L., Relationship of drusen and abnormalities of the retinal pigment epithelium to the prognosis of neovascular macular degeneration. The macular photocoagulation study group. Archives of Ophthalmology., 1990. 108(10): p. 1442‐7. 322. Sarraf, D., Gin, T., Yu, F., Brannon, A., Owens, S.L., and Bird, A.C., Long‐term drusen study. Retina., 1999. 19(6): p. 513‐9. 323. Hillenkamp, J., Hussain, A.A., Jackson, T.L., Cunningham, J.R., and Marshall, J., The influence of path length and matrix components on ageing characteristics of transport between the choroid and the outer retina. Invest. Ophthalmol. Vis. Sci., 2004. 45(5): p. 1493‐
1498. 324. Guidry, C., Medeiros, N.E., and Curcio, C.A., Phenotypic variation of retinal pigment epithelium in age‐related macular degeneration. Investigative Ophthalmology & Visual Science., 2002. 43(1): p. 267‐73. 325. Li, C.‐M., Chung, B.H., Presley, J.B., Malek, G., Zhang, X., Dashti, N., Li, L., Chen, J., Bradley, K., Kruth, H.S., and Curcio, C.A., Lipoprotein‐like particles and cholesteryl esters in human bruchʹs membrane: Initial characterization. Invest. Ophthalmol. Vis. Sci., 2005. 46(7): p. 2576‐2586. 326. Oh, H., Takagi, H., Takagi, C., Suzuma, K., Otani, A., Ishida, K., Matsumura, M., Ogura, Y., and Honda, Y., The potential angiogenic role of macrophages in the formation of choroidal neovascular membranes. Invest. Ophthalmol. Vis. Sci., 1999. 40(9): p. 1891‐1898. 191 327. Shimada, H., Kawamura, A., Mori, R., and Yuzawa, M., Clinicopathological findings of retinal angiomatous proliferation. Graefes Archive for Clinical & Experimental Ophthalmology, 2007. 245(2): p. 295‐300. 328. Yang, P., de Vos, A.F., and Kijlstra, A., Macrophages and mhc class ii positive cells in the choroid during endotoxin induced uveitis. British Journal of Ophthalmology, 1997. 81(5): p. 396‐401. 329. Sarks, S., Cherepanoff, S., Killingsworth, M., and Sarks, J., Relationship of basal laminar deposit and membranous debris to the clinical presentation of early age‐related macular degeneration. Investigative Ophthalmology & Visual Science, 2007. 48(3): p. 968‐77. 330. Duffield, J.S., The inflammatory macrophage: A story of jekyll and hyde. Clinical Science, 2003. 104(1): p. 27‐38. 331. Aktan, F., Inos‐mediated nitric oxide production and its regulation. Life Sciences, 2004. 75(6): p. 639‐53. 332. Blaise, G.A., Gauvin, D., Gangal, M., and Authier, S., Nitric oxide, cell signaling and cell death. Toxicology, 2005. 208(2): p. 177‐92. 333. Holness, C.L. and Simmons, D.L., Molecular cloning of cd68, a human macrophage marker related to lysosomal glycoproteins. Blood, 1993. 81(6): p. 1607‐13. 334. Holtkamp, G.M., De Vos, A.F., Peek, R., and Kijlsta, A., Analysis of the secretion pattern of monocyte chemotactic protein‐1 (mcp‐1) and transforming growth factor‐beta 2 (tgf‐beta2) by human retinal pigment epithelial cells. Clinical & Experimental Immunology, 1999. 118(1): p. 35‐40. 335. Chen, M., Forrester, J.V., and Xu, H., Synthesis of complement factor h by retinal pigment epithelial cells is down‐regulated by oxidized photoreceptor outer segments. Experimental Eye Research, 2007. 84(4): p. 635‐45. 336. Kaestel, C.G., Lovato, P., Odum, N., Nissen, M.H., and Ropke, C., The immune privilege of the eye: Human retinal pigment epithelial cells selectively modulate t‐cell activation in vitro. Current Eye Research, 2005. 30(5): p. 375‐83. 337. Richards, A., Kavanagh, D., and Atkinson, J.P., Inherited complement regulatory protein deficiency predisposes to human disease in acute injury and chronic inflammatory states. Advances in Immunology, 2007. 96: p. 141‐77. 338. Lommatzsch, A., Hermans, P., Weber, B., and Pauleikhoff, D., Complement factor h variant y402h and basal laminar deposits in exudative age‐related macular degeneration. Graefes Archive for Clinical & Experimental Ophthalmology, 2007. 245(11): p. 1713‐6. 339. Sennlaub, F., Courtois, Y., and Goureau, O., Inducible nitric oxide synthase mediates the change from retinal to vitreal neovascularization in ischemic retinopathy.[comment]. Journal of Clinical Investigation., 2001. 107(6): p. 717‐25. 340. Macular Photocoagulation Study Group, Argon laser photocoagulation for neovascular maculopathy. Three‐year results from randomized clinical trials. Macular photocoagulation study group. Archives of Ophthalmology, 1986. 104(5): p. 694‐701. 192 341. Hangai, M., Yoshimura, N., Hiroi, K., Mandai, M., and Honda, Y., Inducible nitric oxide synthase in retinal ischemia‐reperfusion injury. Experimental Eye Research., 1996. 63(5): p. 501‐9. 342. Becquet, F., Courtois, Y., and Goureau, O., Nitric oxide in the eye: Multifaceted roles and diverse outcomes. Survey of Ophthalmology, 1997. 42(1): p. 71‐82. 343. Mitchell, P., Wang, J.J., Foran, S., and Smith, W., Five‐year incidence of age‐related maculopathy lesions: The blue mountains eye study. Ophthalmology, 2002. 109(6): p. 1092‐
1097. 344. Wang, J.J., Rochtchina, E., Lee, A.J., Chia, E.‐M., Smith, W., Cumming, R.G., and Mitchell, P., Ten‐year incidence and progression of age‐related maculopathy: The blue mountains eye study. Ophthalmology, 2007. 114(1): p. 92‐98. 345. Gillies, M.C., Simpson, J.M., Luo, W., Penfold, P., Hunyor, A.B., Chua, W., Mitchell, P., and Billson, F., A randomized clinical trial of a single dose of intravitreal triamcinolone acetonide for neovascular age‐related macular degeneration: One‐year results. Archives of Ophthalmology., 2003. 121(5): p. 667‐73. 346. Macular Photocoagulation Study Group, Subfoveal neovascular lesions in age‐related macular degeneration. Guidelines for evaluation and treatment in the macular photocoagulation study. Macular photocoagulation study group. Archives of Ophthalmology, 1991. 109(9): p. 1242‐57. 193 A P P E N D I C E S 194 APPENDIX A: ANTIBODIES Immunohistochemistry – primary antibodies Antibody
Clone
Raised in
Company
Catalog
number
Monoclonal anti-human CD68
KP1
Mouse
Zymed Invitrogen, Carlsbad, CA, USA
08-0125
Monoclonal anti-human iNOS
6
Mouse
BD Transduction Laboratories, San Jose, CA, USA
610328
Immunohistochemistry – secondary antibody Antibody
Clone
Raised in
Company
Catalog
number
Biotinylated monoclonal anti-mouse
IgG (H+L)
N0403
Horse
Vector Laboratories, Burlingame, CA, USA
BA-2000
Western blotting – detection antibodies Antibody
Peroxidase polyclonal anti-human
IgG+A+M (H+L)
Peroxidase monoclonal anti-human
IgG1
Peroxidase monoclonal anti-human
IgG2
Peroxidase monoclonal anti-human
IgG3
Peroxidase monoclonal anti-human
IgG4
Peroxidase monoclonal anti-human
IgM
Clone
Raised in
Company
Catalog
number
N/A
Goat
Zymed Invitrogen, Carlsbad, CA, USA
62-8320
HP6069
Mouse
Zymed Invitrogen, Carlsbad, CA, USA
05-3320
HP6014
Mouse
Zymed Invitrogen, Carlsbad, CA, USA
05-0520
HP6047
Mouse
Zymed Invitrogen, Carlsbad, CA, USA
05-3620
HP6025
Mouse
Zymed Invitrogen, Carlsbad, CA, USA
05-3820
HP6083
Mouse
Zymed Invitrogen, Carlsbad, CA, USA
05-4900
A APPENDIX B: CD68 IMMUNOHISTOCHEMISTRY 1. Xylene 2 x 3 min 2. 100% alcohol 2 x 3 min 3. 95% alcohol 3 min 4. 70% alcohol 3 min 5. Tapwater rinse 6. 1% peroxide solution 30 min ( 0.5ml peroxide in 49.5ml 50% alcohol) 7. Distilled water rinse 8. Tris rinse o
9. Pepsin digestion at 37 C 5 min 10. Tris rinse 11. 10% normal horse serum 20 min (1ml normal horse serum in 9 ml Tris) 12. Primary antibody : mouse anti‐human CD68 Neat 1 hr at 37oC (diluted in 1% normal horse serum) 13. Tris 3 x 5 min 14. Secondary antibody : Biotinylated anti‐mouse IgG 30 min at 37oC 1:200 15. Tris 3 x 5 min 16. ABC 1:500 30 min at RT 17. Tris 3 x 5 min 18. Vector SG Blue 10 min 19. Tris 3 x 5 min 20. Distilled water rinse 21. 70% alcohol 3 min 22. 95% alcohol 3 min 2 x 3 min 23. 100% alcohol 24. Xylene 2 x 3 min 25. Coverslip in DPX B APPENDIX C: iNOS IMMUNOHISTOCHEMISTRY 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. Xylene 100% alcohol 95% alcohol 70% alcohol Tapwater 1% peroxide solution ( 0.5ml peroxide in 49.5ml 50% alcohol) Distilled water Tris 10% normal horse serum (1ml normal horse serum in 9 ml Tris) Primary antibody : mouse anti‐human iNOS 1:400 (diluted in 1% normal horse serum) Tris Secondary antibody : Biotinylated anti‐mouse IgG 1:200 Tris ABC 1:500 Nickel ammonium sulphate Tris (100mg nickel ammonium sulphate in 250ml Tris) DAB made in nickel ammonium sulphate + 70μl H2O2 (1 DAB tablet dissolved in 15ml of nickel ammonium sulphate in Tris and filter) Tris Distilled water 70% alcohol 95% alcohol 100% alcohol Xylene Coverslip in DPX C 2 x 3 min 2 x 3 min 3 min 3 min rinse 30 min rinse rinse 20 min 1 hr at 37oC 3 x 5 min 30 min at 37oC 3 x 5 min 30 min at RT 3 x 5 min 10 min 3 x 5 min rinse 3 min 3 min 2 x 3 min 2 x 3 min APPENDIX D: PREPARATION OF RETINAL PROTEIN FOR ELECTROPHORESIS Dissection: 1 X human eye 1 X sterile dissection kit dissection tray PBS ~20mL 6 X 1mL eppendorfs for storage Perform in vacuum extraction hood Remove cornea and anterior chamber using curved scissors Remove vitreous Add PBS to the cup formed by remaining eye Tease out retina using fine forceps and cut at optic disc Divide retina into approx 6 pieces Lysis: Lysis buffer ~10mL Sterile 10mL syringe Sterile drawing up or 19G needle 2 X sterile 10mL centrifuge tubes per piece of retina 10 X 1mL sterile eppendorfs per piece of retina for microfuge 10 X 1mL sterile eppendorfs per piece of retina for storage human retina ice in styrofoam container 1000μL pipetteman & sterile tips In vacuum extraction hood: Draw up 5mL lysis buffer in syringe Add retina Homogenise by drawing up and down into 10mL centrifuge tube until homogenate macroscopically evenly mixed Place on orbital shaker on ice @ 50rpm for 20min Place in large centrifuge after balancing Spin @ 4000rpm for 5min @ 4oC Transfer supernatant into 10 eppendorf tubes Place in microfuge and spin @ 13000rpm for 5min @ RToC Transfer supernatant into 5‐10 eppendorf tubes Label and store at –20oC. D APPENDIX E: ELECTROPHORESIS & WESTERN BLOTTING Electrophoresis 1. Dilute retinal homogenates 1:2 in loading buffer (2X) 2. Heat samples for 2 minutes at 100 oC and clarify by centrifugation at 3000rpm for 3 minutes. 3. Assemble MiniProtean cell (BioRad Laboratories) for electrophoresis & fill with running buffer 4. Load samples in the prep well of a 4‐20% Tris‐Glycine polyacrylamide get in a Load Kaleidoscope precision plus protein standard (BioRad Laboratories) in the standards well 5. Run samples at 200V, 100mA and at room temperature for 30‐40min (depending on ambient temperature) Western blotting 1. Pre‐wet Hybond P membrane (Amersham Pharmacia Biotech) in methanol and allow to equilibrate for 10min in transfer buffer 2. Equilibrate gels in transfer buffer after electrophoresis 3. Assemble the Mini‐Protean cell is assembled for blotting & fill with transfer buffer and ice block 4. Blot at 60V, 200mA for 4hrs at 4 oC Autoantibody detection 1. Block transblots for 1hr in blocking buffer (5% tween tris‐buffered saline) at room temperature 2. Wash transblots in wash buffer 3x10min 3. Assemble multiscreen apparatus and insert transblot 4. Load each multiscreen well with serum diluted 1:200 in TBS and incubate overnight at room temperature 2x10min 5. Wash in wash buffer 6. Remove transblots from multiscreen apparatus and wash in TBS 1x10min 7. Mix together 10mls each of ECL solutions 1&2 (Amersham Pharmacia Biotech) 8. In dark room, incubate tranblot in ECL mixture for 2 min 9. Remove and blot with filter paper before inserting into plastic sleeves 10. Expose to ECL film for 1‐3min 11. Develop ECL film in photographic developer until a good signal can be seen and 12. Place film in photographic fixer for 5 sec 13. Rinse and dry film E A P P E N D I X F : B U F F E R S Lysis Buffer: 1ml 10X Salt buffer 9ml sterile H2O 1 X tablet miniComplete protease inhibitor cocktail 100μL NP40 Loading Buffer: 4 ml 10% (w/v) SDS 2 ml glycerol 1 ml 0.1% (w/v) bromophenol blue 2.5 ml 0.5M Tris‐HCl (pH 6.8) 0.2 ml β‐mercaptoethanol 10 ml deionised water Running buffer (10X): 29g Tris base 144g Glycine 10g SDS miliQ H2O to 1000ml Transfer buffer: 3g tris base 14.4g glycine 200mL methanol miliQ H2O to 1000ml Tris‐buffered saline (10X): 6.05g Tris base 43.9g NaCl milliQ H2O to 500ml Blocking buffer: 5ml Tween 20 1X TBS to 100ml Wash buffer: 400μL Tween 20 1X TBS to 800ml F A P P E N D I X G : M U L T I S C R E E N A P P A R A T U S An ECL film of a typical blot produced using the multiscreen apparatus. Protein standards at left. Photograph of apparatus courtesy of Bio‐Rad Laboratories. Hercules, California USA G