Last year, four independent research groups reported a linkage between polymorphisms in the complement factor H (CFH) gene to age-related macular degeneration (AMD), the most common cause of blindness in people over 60. One particular SNP, which changed a tyrosine (Y402) to histidine, was strongly associated with disease in all the studies.

The results seemed clear enough, but now a more extensive examination of variability in the region has yielded a different take on the genetics of AMD. In two papers in the August 27 online edition of Nature Genetics, researchers report finding additional CFH sequence variants that are even more strongly associated with the risk of AMD than the Y402H allele. One of the papers, from a group of Boston researchers led by Johanna Seddon at the Massachusetts Eye and Ear Infirmary, also confirms variants in two other genes that act additively with CFH to determine the risk of AMD in individuals.

While not directly relevant to Alzheimer disease, the findings provide a cautionary tale for researchers searching for the genetic risk factors of complex diseases like AD. Determining genetic risk is an exceedingly complicated proposition, the study indicates, and finding all the risk alleles for common diseases will likely involve a thorough analysis of hundreds or thousands of cases.

In the study from the Seddon group, first author Julian Maller worked together with Mark Daly and colleagues at the Broad Institute of Harvard and MIT to survey 1,536 SNPs over three genes in more than 2,000 people (1,238 affected individuals and 934 controls). Unexpectedly, the strongest disease association they found within the CFH gene was not the previously observed Y402H SNP, but an intronic SNP that did not affect protein sequence.

To complicate matters even more, further SNP typing by Seddon and colleagues confirmed a second AMD locus on chromosome 10, and showed that two gene variants there were strongly associated with AMD. They also found two SNPs in a second complement-related locus, this one containing the gene for complement component 1 and complement factor B (C2/CFB). The five SNPs in three genes acted in additive fashion to increase the risk of AMD from less than 1 percent in people with the lowest-risk genotypes at each SNP, to more than 50 percent for those with five risky SNPs. Mixed genotypes of high- and low-risk SNPs displayed intermediate risk, forming a gradient of risk in the population. Finally, the authors estimated that the five common variants explain half the excess risk of AMD in siblings of affected individuals.

In the second paper, from Goncalo Abecasis and Anand Swaroop of the University of Michigan at Ann Arbor, a similarly intensive survey of SNPs in the CFH gene in 762 affected people and 268 controls turned up 20 polymorphisms that showed stronger association than Y402H. In this study, led by first author Mingyao Li, no single SNP accounted for disease susceptibility, but instead they found multiple SNPs making up four common haplotypes, two associated with increased risk, and two that were protective.

In both studies, the CFH variant showing the strongest association was an SNP in a non-coding region, raising questions of what change it would cause in the carrier. One possibility is that the variants affect complement factor H gene expression. Alleles that change a gene’s expression level without altering protein structure have been implicated in several other neurodegenerative diseases, revealing an increasing role for regulatory polymorphisms (see ARF related news story).

“Our results show that dissection of complex disease susceptibility loci will be a challenging process and that identification of strongly associated alleles, even when they are protein coding, should not preclude further detailed genetic analysis,” write Abecasis and Swaroop. They call for careful studies not just of SNP association, but of regional DNA sequence and gene expression patterns to get a full understanding of the contribution of the CFH gene.

Beyond the genetics lesson, could the role of complement in AMD also have something to teach us about AD? Resulting from the loss of retinal ganglion cells required for vision, AMD is a form of neurodegeneration. A role for the complement cascade was not suspected in the disease until the genetic association was made, but since then much work has cemented the idea that the macula is destroyed by complement-stimulated inflammatory and/or angiogenic processes. Complement activation and inflammation occur in AD, possibly because of Aβ production. (For more on the AD-complement angle, see comments below by Ben Barres and Tony Wyss-Coray on the potential roles of complement in AD neuropathology.) Recently, it was found that the characteristic protein deposits (Drusen) that mark AMD contain nonfibrillar amyloid (Luibl et al., 2006), making yet another intriguing link between AMD and AD pathology.—Pat McCaffrey

Comments

  1. Complement is activated in most forms of neurodegeneration and in AD in particular. Aβ aggregates (the particular conformation of which is ill-defined) can activate complement and may trigger the clearance of Aβ in solution via C3bi and complement receptors (Mac1/CR3) on microglia. Similarly, tangle ghosts may be a trigger for complement activation.

    It has also been suggested that injured cells or possibly Aβ aggregates bound to cell surfaces lead to activation of complement. This cell-bound activation of complement could lead to formation of the membrane attack complex (MAC) and further damage of cells, although studies in oligodendrocytes show that the MAC, which is a membrane pore, can also induce protective pathways in cells.

    Factor H is involved in the regulation of the alternative complement pathway and may influence any of the above activities. If Factor H does indeed have a role in the degeneration of cells in the eye, it may have a similar role in the hippocampus and cortex as well. At this point, however, there is little hard evidence from in-vivo studies to support a role for complement in neurodegeneration, and more studies need to be done.

  2. The Complement System and Age-related Macular Degeneration: What Does It Teach Us about Alzheimer Disease?
    Age-related macular degeneration (AMD) and Alzheimer disease (AD) are both neurodegenerative diseases of aging, with loss of photoreceptors and CNS neurons, respectively. A number of recent studies have shown that polymorphisms of several complement proteins in the alternative pathway of complement activation (CFB, C2, and CFH) enhance susceptibility to AMD. Somewhat similarly, in AD there is a profound increase in the levels of the initiating protein of the complement cascade called C1q, a prominent upregulation of which has also recently been reported to accompany glaucoma, which is a neurodegenerative retinal disease of aging. All of these changes, in AMD and AD, ultimately lead to activation of the pivotal complement protein called C3. Upon activation, C3 is fragmented into several pieces. One is called C3a, a small cytokine-like molecule that activates microglia and stimulates angiogenesis, whereas a larger fragment, called C3b, opsonizes the cell or debris where it was generated, leading to phagocytosis (or sometimes cell lysis).

    What does the association of complement pathway malfunction in AMD have to teach us about neurodegeneration in AD? First, complement activation has recently been linked to the strong stimulation of angiogenesis that contributes to the pathophysiology of AMD. When C3 activation is prevented in AMD, neovascularization is inhibited (Nozaki et al., 2006; Bora et al., 2005; Girardi et al., 2006). Is it possible that complement activation in AD could lead to abnormal angiogenesis that could contribute to AD pathophysiology? There is little data as of yet, but the possibility has been suggested previously (for instance, Zlokovic, 2005; Vagnucci and Li, 2003) and an increase in microvascularization in AD brain was reported previously by Perlmutter et al. (1990). Interestingly β amyloid peptide injected into a normal rat hippocampus has been found to stimulate angiogenesis (Zand et al., 2005) and fibrillar β amyloid is a well-known activator of C1q and thus the classical complement cascade.

    Second, complement activation leads to a dramatic activation of microglia, both in the retina and throughout the CNS. Microglia are a crucial source of many components of the complement cascade and other factors that may enhance inflammation. Amyloid-β also enhances microglial activation, and has been reported to be deposited at sites of complement activation in Drusen deposits in AMD (Johnson et al., 2002). In fact, amyloid-β triggers retinal pigment epithelial cells (a class of phagocytic cells within the retina implicated in retinal photoreceptor degeneration) to release VEGF and other factors that promote angiogenesis (Yoshida et al., 2005). Microglia have recently been shown to play a role in developmental blood vessel formation in the retina (Checchin et al., 2006), which is of interest given their close association with blood vessels and their activation in neurodegenerative diseases in general.

    Breakdown of the blood-brain barrier (and blood-retinal barrier) has been increasingly observed and invoked in AMD and AD. Newly formed vessels are leaky, as are microvessels involved in angiopathy observed in both AMD and AD. This may be crucially important in driving the pathology forward, as the blood is the major source of all complement proteins. Activated microglial cells and many neurons and glial cells also make certain complement components. If capillaries become leaky, key complement proteins such as C3 may leak in (and complement activation itself may in turn further enhance the leakiness of these vessels). C3 is the key complement protein that mediates complement activation in all pathways of complement activation. Its endogenous sources within the normal and injured brain are still quite mysterious. There is evidence that activated microglial cells and peripheral macrophages clearly make it; possibly other cell types can make it, as well. The brain cell type identities that produce each complement component are important questions for future work.

    The above observations suggest that age-related neurodegenerative diseases such as AMD and AD may share in common underlying mechanisms, particularly the activation of the complement pathways, leading to C3 activation, microglial activation, vascular changes and angiogenesis, amyloid production, and neuroinflammatory tissue damage. Sorting out the relative importance and sequence of these events is an important task for the future. But the linkage of AMD susceptibility to complement pathway polymorphisms clearly shows the important contribution of complement proteins in AMD to disease initiation and/or progression, and the high levels of C1q protein in AD brains suggest an important role for complement proteins in AD, as well.

    An important unanswered question is how complement pathway activation begins in AMD and AD. A primary way that the complement cascade becomes initiated is by the presence of abnormal cell surfaces which become opsonized by short pentraxins such as C-reactive protein or antibodies that in turn can both bind to C1q. In AMD, it is entirely unclear how the alternative pathway would become activated, as microbial surfaces are not implicated. The potential role of the classical complement cascade in AMD should now be examined. CFH not only controls activity of the alternative complement pathway but of the other complement pathways, as well, as its function is to inactivate the C3 complement protein that is central to all complement paths of activation. A still relatively underexplored question is the extent to which degenerating tissues become immunogenic. We have recently found that degenerating myelin in the peripheral nervous system is strongly immunogenic and that autoantibodies are generated to degenerating myelin that are actually necessary for its clearance (Vargas and Barres, SFN abstract, 2005 and submitted). Similarly it is possible that autoantibodies are generated against degenerating retinal tissue in AMD or brain tissue in AD. As cell surface antigens in retinal and brain cryosections are only poorly accessible to antibodies in immunostaining methods, it has been difficult to address this question experimentally. A great way forward, where relevant animal models for neurodegenerative disease are available, will be to examine the effects on the neurodegenerative process when these animals are crossed to antibody-deficient strains of mice such as the JhD mutant mouse that entirely lacks B cells.

    Lastly, the central importance of complement proteins in determining AMD susceptibility raises an important question. Could the exponential increase in age-dependent susceptibility to neurodegenerative disease be caused by an age-related change in the complement system? Possibly complement protein levels might increase with age, or important complement inhibitors might decrease with age. While the identification of polymorphisms in genes encoding complement proteins is an important step toward understanding AMD and further supports the potential long-suggested relevance of complement proteins in AD, it also raises many new questions for future studies.

    References:

    . Drusen complement components C3a and C5a promote choroidal neovascularization. Proc Natl Acad Sci U S A. 2006 Feb 14;103(7):2328-33. PubMed.

    . Role of complement and complement membrane attack complex in laser-induced choroidal neovascularization. J Immunol. 2005 Jan 1;174(1):491-7. PubMed.

    . Complement activation induces dysregulation of angiogenic factors and causes fetal rejection and growth restriction. J Exp Med. 2006 Sep 4;203(9):2165-75. PubMed.

    . Neurovascular mechanisms of Alzheimer's neurodegeneration. Trends Neurosci. 2005 Apr;28(4):202-8. PubMed.

    . Alzheimer's disease and angiogenesis. Lancet. 2003 Feb 15;361(9357):605-8. PubMed.

    . Induction of angiogenesis in the beta-amyloid peptide-injected rat hippocampus. Neuroreport. 2005 Feb 8;16(2):129-32. PubMed.

    . The Alzheimer's A beta -peptide is deposited at sites of complement activation in pathologic deposits associated with aging and age-related macular degeneration. Proc Natl Acad Sci U S A. 2002 Sep 3;99(18):11830-5. PubMed.

    . The potential role of amyloid beta in the pathogenesis of age-related macular degeneration. J Clin Invest. 2005 Oct;115(10):2793-800. PubMed.

    . Potential role of microglia in retinal blood vessel formation. Invest Ophthalmol Vis Sci. 2006 Aug;47(8):3595-602. PubMed.

  3. The complement system plays important roles not only in the neurodegeneration seen in AD and AMD, but also during normal brain aging. We reported that the activation of the early components of the complement appears to target myelin and oligodendrocytes in the aged rhesus monkey brain. Interestingly, the complement activation was also observed in the young brain, but to a much lesser extent. These findings suggest that only an exaggerated complement activation is detrimental, but that low-level activation may be physiologic and could be crucial for the turnover of myelin (Duce et al., 2006.)

    References:

    . Activation of early components of complement targets myelin and oligodendrocytes in the aged rhesus monkey brain. Neurobiol Aging. 2006 Apr;27(4):633-44. Epub 2005 Jun 29 PubMed.

  4. There is considerable data suggesting that complement plays a role in the progression of AD, including an animal model that we published in 2004. In that study, Tg2576 animals (APP) were crossed with C1q-deficient mice to effectively eliminate activation of the classical pathway of complement. The pathology of APPQ-/- was compared with that of APP mice and B6SJL controls at 3-16 months of age by immunohistochemistry and Western blot analysis. While at younger ages (3-6 months) when no plaque pathology was present, no significant differences were seen; at older ages (12 and 16 months), the level of activated glia surrounding the plaques was significantly lower in the APPQ-/- mice. In addition, although Tg2576 mice showed a progressive decrease in synaptophysin and MAP2 in the CA3 area of hippocampus compared with control B6SJL at 9, 12, and 16 months, the APPQ-/- mice had significantly less of a decrease in these markers at 12 and 16 months. Interestingly, the APP and APPQ-/- mice developed comparable total amyloid and fibrillar β amyloid in frontal cortex and hippocampus.

    The data are consistent with a deleterious role for complement at ages when the fibrillar plaques (that have been shown to activate complement in vitro, and to be selectively associated with complement components in human AD brain) are present, most likely through the activation of the classical complement cascade and the enhancement of inflammation. There are observations that imply that complement can also have a protective effect in the brain. There are indeed multiple points within the complement cascade in which protective effects can be exerted. Some of these were discussed in recent reviews, but others are likely to be uncovered as the use of additional animal models and modulators of complement activity are tested.

    References:

    . Absence of C1q leads to less neuropathology in transgenic mouse models of Alzheimer's disease. J Neurosci. 2004 Jul 21;24(29):6457-65. PubMed.

    . The double-edged flower: roles of complement protein C1q in neurodegenerative diseases. Adv Exp Med Biol. 2006;586:153-76. PubMed.

  5. Besides providing further evidence that the complement cascade is involved in neurodegeneration, these articles have broader implications for the field of genetics and disease mapping. Through a combination of linkage mapping (1-5) and association studies (6-8), the original reports declared that a tyrosine-to-histidine substitution at position 402 accounted for 20-50 percent of the overall risk of developing age-related macular degeneration (AMD). However, in the current reports, the coding change does not appear to have the largest effect. This has two implications for the field of neurogenetics. First, that it should not be assumed that examining single coding changes in genes is sufficient. With the recent boom in whole genome studies, this should no longer be as much of a problem for the field. But it does suggest that perhaps a re-examination of current association results is in order. Alzgene reports that after meta-analysis of 341 genes, 20 variants in 13 genes are significant. Approximately one-third of those are coding changes; however, in light of the recent findings with AMD, perhaps rather than trying to further replicate these specific polymorphisms, whole genes should be analyzed using HapMap data. Second, these studies also imply that for common diseases, it is genes and their alleles that matter, not individual polymorphisms. This is one of the reasons haplotype analysis can be more powerful than examining single SNPs. Within the first reports, the coding polymorphism divided the population into risk and non-risk; however, the current reports now identify multiple risk and protective CFH alleles. Thus, within the population, it is the collection of SNPs, not any single change, which creates risk (or in this case, protection). This seems obvious, but it was impossible to efficiently examine the genetic content of genes as a whole until recently. Again, with the advent of full genome scans, there is now the possibility to capture complete information about the genome and perform more quantitative analysis of phenotypes rather than just treating all individuals as risk and non-risk.

    References:

    . Age-related macular degeneration: a high-resolution genome scan for susceptibility loci in a population enriched for late-stage disease. Am J Hum Genet. 2004 Mar;74(3):482-94. PubMed.

    . Dissection of genomewide-scan data in extended families reveals a major locus and oligogenic susceptibility for age-related macular degeneration. Am J Hum Genet. 2004 Jan;74(1):20-39. PubMed.

    . Age-related macular degeneration--a genome scan in extended families. Am J Hum Genet. 2003 Sep;73(3):540-50. PubMed.

    . A genomewide scan for age-related macular degeneration provides evidence for linkage to several chromosomal regions. Am J Hum Genet. 2003 Oct;73(4):780-90. PubMed.

    . Age-related maculopathy: a genomewide scan with continued evidence of susceptibility loci within the 1q31, 10q26, and 17q25 regions. Am J Hum Genet. 2004 Aug;75(2):174-89. PubMed.

    . Complement factor H polymorphism and age-related macular degeneration. Science. 2005 Apr 15;308(5720):421-4. PubMed.

    . Complement factor H variant increases the risk of age-related macular degeneration. Science. 2005 Apr 15;308(5720):419-21. PubMed.

    . Complement factor H polymorphism in age-related macular degeneration. Science. 2005 Apr 15;308(5720):385-9. PubMed.

Make a Comment

To make a comment you must login or register.

References

News Citations

  1. Highs and Lows: Overactive Synuclein Promoter Linked to Parkinson’s, Tau Loss to Mental Retardation

Paper Citations

  1. . Drusen deposits associated with aging and age-related macular degeneration contain nonfibrillar amyloid oligomers. J Clin Invest. 2006 Feb;116(2):378-85. PubMed.

Further Reading

Papers

  1. . Complement factor H variant increases the risk of age-related macular degeneration. Science. 2005 Apr 15;308(5720):419-21. PubMed.
  2. . Complement factor H polymorphism in age-related macular degeneration. Science. 2005 Apr 15;308(5720):385-9. PubMed.
  3. . Complement factor H polymorphism and age-related macular degeneration. Science. 2005 Apr 15;308(5720):421-4. PubMed.
  4. . Genetics. Was the Human Genome Project worth the effort?. Science. 2005 Apr 15;308(5720):362-4. PubMed.
  5. . A common haplotype in the complement regulatory gene factor H (HF1/CFH) predisposes individuals to age-related macular degeneration. Proc Natl Acad Sci U S A. 2005 May 17;102(20):7227-32. PubMed.

Primary Papers

  1. . Common variation in three genes, including a noncoding variant in CFH, strongly influences risk of age-related macular degeneration. Nat Genet. 2006 Sep;38(9):1055-9. PubMed.
  2. . CFH haplotypes without the Y402H coding variant show strong association with susceptibility to age-related macular degeneration. Nat Genet. 2006 Sep;38(9):1049-54. PubMed.