The clinical and angiographic features of AMD have been well described. In particular, Gass^20,21,22,23 helped to characterize the “dry” form of AMD including drusen, the extracellular deposits associated with the condition, and also described the contribution of the choroidal vasculature to the “wet” form (Figs. 23C.1, 23C.2 and 23C.3). The clinical features of this condition are not a subject of this Chapter and the reader is referred to Foundations, Volume 2, Chapter 12 for a more detailed discussion.

Drusen, the hallmark of AMD, are extracellular deposits of degraded normal cellular materials which accumulate between the basal lamina of the retinal pigment epithelium (RPE) and the inner collagenous layer of Bruch’s membrane^24,25(Fig. 23C.4). Drusen are predominantly composed of cholesterol, vitronectin, and apolipoproteins B and E (Table 23C.1).^24,26,27,28 In addition, amyloid,^25,29 a substance not normally found in healthy cells, as well as complement proteins (C5 and C5b-9), immunoglobulin light chains, and Factor X and their RNA transcripts have been identified in drusen and the RPE respectively.^25,30,31 Similar material has been identified the glomerular deposits of poststreptococcal and membranoproliferative glomerulonephritis type II (MPGN II) and in the neuritic plaques which characterize Alzheimer’s disease (AD)³² as well as in other diseases of aging²⁵ including atherosclerosis,^33,34 amyloidosis,³⁵ dense deposit disease,²⁵ and elastosis³⁶ (Table 23C.2). This suggests that these systemic disorders may have a similar pathophysiology. The association with MPGN II is particularly notable. In the majority of cases, MPGN II is characterized by an autoantibody against the C3bBb complex, which leads to uncontrolled activation of the alternative complement pathway. Renal lesions in MPGN II have been shown to be identical in composition and structure to drusen in AMD, supporting the hypothesis of the central importance of the complement system in AMD.³⁷ Recent studies have also implicated allelic variations in factor H in MPGN II, AMD, coronary artery disease, and atherosclerosis.^38,39,40,41 A common denominator in these seemingly diverse and unrelated ocular and systemic diseases may be the co-localization of factor H, C-reactive protein (CRP), and heparin on the arterial endothelium. This indicates that factor H may play an important role in the protection against complement mediated vascular damage with associated secondary exudation or neovascularization.^16,41

The juvenile onset macular dystrophies consist of a number of disorders which in most cases become manifest between the ages of 10 and 20 years. These conditions have been well characterized, in part because large families consisting of multiple generations have been available for study. Many specific genes^{42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63} or chromosomal regions^{64,65,66,67,68} have been identified and their discoveries are contributing to a better understanding of the basic underlying mechanisms of degenerative macular disease. In general these genes have not been implicated in a significant percentage of adult onset AMD cases. For more information regarding juvenile onset macular dystrophies the reader is referred to several excellent reviews on these important scientific developments.^69,70,71

Evidence supporting a genetic contribution is derived from studies of twins⁷² and first-degree relatives⁷³ of persons with AMD. These studies indicated that at least 25% of AMD may be genetically determined and that genetic factors play an important role in determining the onset and severity of the disease. Several families with AMD spanning multiple generations have been studied which have provided significant information regarding the genetics of AMD. Using the technique of linkage analysis and genome wide scanning two chromosomal regions 1q25-31 and 10q26 have been consistently associated with AMD.^{17,74,75,76,77,78,79,80,81,82}

Modern molecular biological techniques have made it possible to scan the entire genome for putative disease causing genes. Genome wide scans in AMD have implicated chromosomal regions1q25-31 and 10q26 as well as four genes associated with the inflammatory process: the chemokine receptor CX3CR1,^12,19 factor H,^13,15,16,17 factor B, and complement factors C2,¹⁴ C3,^83,84 and I⁸⁵ (Fig. 23C.1). SNPs identified within these genes are associated either with a predisposing (at risk) or protective effect for AMD. LOC387715, a gene of unknown function, has also been linked to AMD especially in association with a history of smoking.^18,86 Overall these studies have identified SNPs or haplotypes which are associated with at risk, protection, or modifying effects in upwards of 75% of AMD cases. What has emerged is the strong association of AMD with inflammation, the complement system, and smoking as summarized in Table 23C.3.

CX3CR1 is a macrophage chemokine receptor which may play a role in clearing age-related deposits. Two SNPs of CX3CR1, V249I, and T280M, are known to be associated with reduced chemoattraction of macrophages. The affected CX3CR1 receptors are fewer in number and have lower binding affinity with their substrate fractralkine. Tuo et al.¹⁹ showed that there was an association between these two SNPs and AMD. They proposed that a lower number of receptor binding sites and reduced binding affinity resulted in reduced macrophage recruitment and altered macrophage clearing of age-related sub-RPE deposits. In support of this theory, microdissection of retinal cells detected the M280 allele in 29.7% of eyes with AMD as well as reduced levels of CX3CR1 in the macular region of these eyes.¹² It is interesting to note 249I and 280M were protective against atherogenesis,^87,88 but I249 “not balanced” by M280 was associated with increased risk of CAD and re-stenosis following cardiac artery stenting.^89,90 Also, the haplotype 249I/280M increased progression to AIDS in HIV+ individuals,^91,92 but also increased favorable response to HAART⁹³ while 249I/280T was associated with long-term nonprogression.⁹⁴

The complement system is composed of a group of inactive plasma enzymes (zymogens) and cell membrane regulatory proteins that interact to maintain a balance between activation with phagocytosis of pathogens and inactivation with protection of self. In response to initiating stimuli, the complement cascade results in the production of peptide mediators of inflammation, phagocyte recruitment, opsonization, and cell lysis. The complement cascade is summarized in Figure 23C.4.

Complement factor H is a plasma protein and is one of the primary regulators of the alternative complement pathway.⁴⁰ Factor H binds with C3b on host cells which it is able to distinguish from C3b on pathogens through an interaction with host cell sialic acid. This prevents inadvertent host cell damage. Although it preferentially binds with membrane bound C3b, factor H is also able to modulate plasma C3b.

Various data support the hypothesis that particular DNA sequence variations of complement factor H may be associated with AMD. First, factor H is located in chromosome 1q31, a region known to link to a large family with AMD (ARMD1 locus).^{74,75,76,77,78,80} Second, complement has been identified in drusen, Bruch’s membrane, and the intercapillary pillars of the choriocapillaris.^25,95 In addition, drusen of persons with MPGN II, a renal condition known to be caused by a deficiency of factor H, have complement components similar to those associated with AMD.³² Third, risk factors associated with AMD, age, and smoking, affect levels of plasma factor H.⁹⁶

Four independent studies (Table 23C.3) tested this hypothesis and consistently identified a SNP in the factor H gene, Y402H, which significantly increases the likelihood of developing AMD. As a predisposing factor, this defect may explain up to 50% of the cases of AMD. Y402H is significantly and similarly associated with both drusen and advanced AMD, indicating that other genetic or environmental factors may be responsible for progression to advanced neovascular or atrophic AMD.⁹⁷ The results of these studies are summarized below.

Klein and associates¹⁶ performed a whole-genome case-control association study of over 100,000 SNPs in 96 white, nonhispanic patients and 50 similar controls and identified two SNPs which were significantly associated with AMD. Both are located on chromosome 1q31. Reconstruction of inferred haplotypes revealed that six SNPs within the factor H gene region associated in four distinct haplotypes or allelic variations. Homozygosity for one of these haplotypes increased the risk of disease by a factor of 7.4. Resequencing of the exons in this haplotype revealed 50 SNPs. One nonsynonymous SNP in exon 9, DNA sequence T1277C (substitution of cysteine for tyrosine at position 1277) caused an amino acid sequence change Y402H (substitution of histidine for tyrosine at position 402) in complement factor H which was strongly associated with AMD. The authors found this SNP was present in 97% of the chromosomes with the high-risk haplotype indicating this was a polymorphism underlying susceptibility to AMD.

Haines and coworkers¹⁶ studied 44 SNPs located on chromosome 1q32 in the ARMD1 locus. In two independent data sets (family-based and AMD cases and controls) the authors found multiple SNPs strongly associated with AMD. However, one haplotype composed of five of the SNPs was present in 46% of cases and 33% of controls. This haplotype was more frequent in diseased than nonaffected family members. DNA resequencing revealed only one sequencing variant, T1277C (DNA) or Y402H (factor H protein), which increased odds of AMD 2.45 to 5.57 times in heterozygotes with any form of AMD and homozygous with neovascular disease, respectively. Because residue 402 is located in the heparin and CRP binding site of factor H, the authors postulated the amino acid substitution could affect factor H function by altering its ability to bind CRP and heparin, and as a consequence, its affinity for C3b and its ability to prevent host damage.

Edwards et al.¹³ tested nonsynonymous SNPs within the ARMD1 locus for association with AMD in two independent case-control populations. Heterozygosity for histidine at 402 (Y402H) increased risk of AMD by a factor of 2.7. Genotype frequencies among cases with and without family history of AMD were similar, indicating that the Y402H polymorphism is a more general predisposing factor for AMD.

Hageman and associates^15,25 identified factor H and the membrane attack complex (MAC) in drusen, RPE basement membrane, and the choroidal capillary walls in the macular rather than the extramacular region.¹⁵ The authors also showed that Factor H was present in the intercapillary pillars and was synthesized locally by the RPE. Case-control allele association analysis of the factor H gene revealed several SNPs with highly significant associations with AMD. Haplotype analysis revealed one haplotype which increased risk, and two which decreased risk. Increased and decreased risk was more pronounced in homozygotes. These associations were observed to be stronger in individuals with drusen and neovascular AMD, rather than geographic atrophy. Half of individuals with AMD carried the same at-risk haplotype.

Others have confirmed the association of Y402H with AMD^17,98 and additional studies of various ethnic populations including Japanese,⁹⁹ French,¹⁰⁰ and British¹⁰¹ individuals have confirmed the association in groups with disparate heredity. Of interest is that the frequency of Y402H was much lower in Japanese, as is the incidence of advanced AMD. Additional risk haplotypes were present that appeared to be responsible for disease susceptibility in this population.⁹⁹ Allelic variations among other populations may contribute to the understanding of observed racial/ethnic variability in AMD prevalence not attributable to other known risk factors.¹⁰²