This study investigated the relationship between MPOD and
ApoE genotype in 302 healthy subjects between 21 and 66 years of age. The mean MPOD of all genotyped subjects at 0.5° retinal eccentricity was 0.38 ± 0.17 ODU, which is comparable to that in previous studies that used HFP to measure MPOD at this eccentricity.
6,41–45 Genotype data were available on 99.3% of our sample.
It is widely accepted that the effect exerted by the ε3 allele is neutral for a number of diseases and that the effects exerted by either the ε2 or the ε4 allele are either risky or protective, depending on the disease in question.
46–48 In the case of
ApoE and AMD, most studies, but particularly the more robust meta-analyses, have suggested that the ε4 allele acts in a dominant manner and that it has a protective effect regardless of the presence of one or two alleles.
48 Therefore, although it would have been preferable to compare samples of equal sizes, this was not possible because of the low in vivo haplotype frequencies of the ε2 and ε4 alleles in the population under investigation. As a result, we elected to divide the study sample into three genotype groups for purposes of analysis.
We found that there was a statistically significant association between MPOD and
ApoE genotype. Subjects who carried the ε4 allele had a higher MPOD than did noncarriers; the mean difference in the MPOD area between the ε4 allele group (group 3) and the other two groups was 0.18 ODU. This difference could not be attributed to differences in dietary intake or differences in serum concentrations of the macular carotenoids among the three groups. Importantly, the three analysis groups were also comparable to one another with respect to age, sex, BMI, and cigarette smoking, factors that may influence MPOD.
6 We did not observe a relative lack of MPOD in association with the ε2 allele, which is putatively, but unconvincingly, linked to an increased risk for AMD compared with the ε3 allele. However, in the three meta-analyses investigating the association between
ApoE and AMD, an increased odds ratio between 1.11 and 1.33 was observed in association with the ε2 allele; this finding was nonstatistically significant in one of these meta-analyses and barely significant in the other two.
48–50
MP carotenoids are derived entirely from diet, but macular levels of these carotenoids are subject to influence by genetic background, as shown in the classic twin study by Liew et al.
13 There is a large body of evidence to suggest that MP may protect against the development and/or progression of AMD.
51 This putative protective effect is based on the known properties of MP as a prereceptoral filter of actinic short-wavelength blue light and its antioxidant capacity, including the ability to quench singlet oxygen
52 and to inhibit the peroxidation of membranous phospholipids.
53 It has been estimated that MP absorbs approximately 40% of damaging short-wavelength irradiation before its incidence on the photoreceptors and the retinal pigment epithelium (RPE).
54 This is deemed to be particularly important because it has been shown by Ham et al.
55 that exposure to short-wavelength blue light can result in photochemical retinal injury in primates. In their study, they exposed rhesus monkey retinas to blue light for 1000 seconds, which resulted in atrophic AMD-like lesions and damage to the photoreceptor outer segments, cellular proliferation, and hypopigmentation of the RPE. They found that the threshold for such retinal damage was lowest for blue light than for other wavelengths of visible light.
55 In addition, it has been shown that the administration of antioxidants can prevent light-induced retinal damage in rat retinas.
56 In other words, there is a substantial body of evidence that cumulative exposure to short wavelength blue light is involved in the pathogenesis of AMD and that this mechanism of retinal injury can be prevented by the administration of antioxidants.
Beyond the contribution of MP to limiting blue light–induced photo-oxidative damage and the consequential generation of reactive oxygen intermediates (ROIs), its constituent carotenoids are also known to quench existing ROIs in the retina as reported by Khachik et al.,
57 who demonstrated the presence of direct oxidation products of L and Z in the retina. Indeed, oxidatively damaged photoreceptors cannot be completely digested by the apposing RPE, thus contributing to the accumulation of lipofuscin in this apposing cell layer.
58 Importantly, lipofuscin is a chromophore, rendering the blue light–filtering properties of MP all the more important.
59,60 Indeed, it has been hypothesized that the relative lack of lipofuscin in foveal RPE cells, when compared with surrounding RPE cells, is the result of protection conferred by MP against oxidation of the overlying photoreceptors and a consequential reduction in lipofuscin production at this location.
61,62
The absorption of the MP carotenoids, L and Z, from the gastrointestinal tract involves incorporation into micelles,
23 most likely followed by a facilitated, protein-mediated transport mechanism across human enterocytes.
63 Once absorbed into the bloodstream, L and Z are associated with plasma lipoproteins, being relatively equally distributed between LDL and HDL,
23–25 with a progressive decrease in the content of L and Z from light to dense LDL. However, some studies have reported that HDL is the preferential carrier of the MP carotenoids in plasma.
64–66 The uptake of these carotenoids into the retina is less well understood. Bernstein et al.
67 and Bhosale et al.
68 have identified specific xanthophyll-binding proteins, of which the Pi isoform of glutathione S-transferase has a high affinity for Z and
meso-Z, but not for L. In a recent publication, Bhosale et al.
68 have also identified a human retinal L-binding protein purified from peripheral human retina.
69 Connor et al.
65 have shown in a study of Wisconsin hypo-alpha mutant chicks, which have a genetic mutation resulting in very low circulating HDL levels, that retinal L accumulation was only 6% that of control chicks, suggesting an important role for HDL in the transport of L in serum, its capture by the retina, or both.
The
ApoE gene codes for apolipoproteins, which are the protein constituents of lipoproteins, particularly plasma chylomicrons, VLDL, and a subclass of HDL. Apolipoprotein E is essential for cholesterol transport and metabolism and for receptor-mediated uptake of specific lipoproteins.
70 It is an important regulator of cholesterol metabolism because of its affinity for apolipoprotein E-specific receptors in the liver and its affinity for LDL receptors in the liver and other peripheral tissues requiring cholesterol.
70 Apolipoprotein E is polymorphic and three common isoforms—E2, E3, and E4—which are coded for by three separate alleles, Apo ε2, Apo ε3, and Apo ε4. These alleles are differentiated on the basis of cysteine-arginine residue interchanges at positions 112 and 158 in the amino acid sequence.
19 As a result of this polymorphism, six common phenotypes exist—three homozygous phenotypes (ε3ε3, ε2ε2, ε4ε4) and three heterozygous phenotypes (ε2ε3, ε2ε4, ε3ε4).
ApoE polymorphisms result in differences in the metabolism of apolipoprotein E-containing lipoprotein particles.
71
In the retina, apolipoprotein E is synthesized in Müller cells and in the RPE, and apolipoprotein E has been identified in drusen.
72–74 Klaver et al.
15 suggested that a high degree of apolipoprotein E biosynthesis is required to support the high rate of photoreceptor renewal in the macular region. Indeed, clinical manifestations of retinal degeneration are exhibited in
ApoE-deficient mice that carry an
ApoE gene inactivated by gene targeting.
75 However, the association of the
ApoE gene with AMD in humans has been inconsistent, with some studies showing a protective effect for the ε4 allele and an increased risk in carriers of the ε2 allele,
14–17,21,49 whereas others have shown no such association with the disease.
76–82 In the three meta-analyses that have been reported on the association between
ApoE profile and AMD, a decreased odds ratio associated with the ε4 allele, ranging from 0.6 to 0.67, was observed, and this association was highly significant in all three meta-analyses.
48–50 Thus, the evidence to date favors a protective role for the Apo ε4 allele because several of the studies that did not show an overall association between the
ApoE gene and AMD have reported a nonsignificant trend toward a protective effect for this allele.
Selective binding of certain receptors to HDL particles enriched with apolipoprotein E has been demonstrated within the CNS.
83 These receptors include the LDL receptor and the LDL receptor-related protein. It has also been shown that there is a lack of binding of such receptors to HDL particles deficient in apolipoprotein E. This selectivity of the uptake mechanism may be dependent on the specific apolipoprotein E that forms part of the transporting lipoprotein molecule. This, in turn, could affect the ability of retinal receptors to bind with, and incorporate, such molecules, thus affecting the uptake of any other substances transported by these lipoproteins. Because Connor et al.
65 have demonstrated that retinal uptake of L is facilitated by an HDL-mediated transport system, it is reasonable to suggest that apolipoprotein E polymorphism may impact the method of capture of L by the retina. Certainly, our observation that the Apo ε4 allele is associated with a relative surplus of MP, but in a way that is unrelated to dietary intake or serum levels of its constituents, supports this view and represents a possible explanation for the putative protective role that the Apo ε4 allele plays in the pathogenesis of AMD.
In conclusion, we have shown that there is a statistically significant and positive association between MPOD and presence of at least one Apo ε4 allele in a cohort of 300 healthy subjects. This observation may relate to the role that apolipoprotein E4, coded for by the Apo ε4 allele, plays in the lipoprotein-mediated transport of L and Z in serum and the subsequent uptake of these carotenoids by the retina.
Supported by a European Union Strand 1 research grant and by a Bausch & Lomb Ireland research grant.
Disclosure:
E. Loane: None;
G.J. McKay: None;
J.M. Nolan, NutraSight Consultancy Limited (C);
S. Beatty, NutraSight Consultancy Limited (C)