The March 25 Nature Genetics details the largest set of human genomes ever sequenced. Senior author Kari Stefansson and colleagues from deCODE genetics, a subsidiary of Amgen based in Reykjavik, Iceland, sequenced whole genomes of more than 2,600 Icelanders. They found rare genetic variants that lend new insights into the intricacies of human genetics and disease, and have predicted the presence of those variants in more than 100,000 other compatriots. Their results are described in four papers. One reveals that rare loss-of-function variants in ABCA7 double a person's chances of getting Alzheimer’s disease. Previous genome-wide association studies had implicated this gene in AD, but not definitively. “This study really nails down ABCA7 as a key modulator of Alzheimer’s disease,” said Alan Tall, Columbia University, New York. “Importantly, it shows that loss of function of ABCA7 is responsible for that association.”

A Long Ancestry.

Citizens of Iceland can trace their lineage back more than 1000 years. [By Abraham Ortelius, in 1590]

Because Iceland has a modest, relatively isolated population that arose from a small number of settlers, genetic variants are likelier to be passed down through generations there, rather than getting lost as they might in larger, mixed populations. This means rare variants that associate with disease might occur at a higher frequency and be easier to find. DeCode has banked on this, and on Iceland's long and detailed genealogical history. 

For this study, deCode scientists sequenced the entire genomes of 2,636 volunteers, including both healthy individuals and those with various disorders. The researchers unearthed about 1.5 million base insertions or deletions and almost 20 million single-nucleotide polymorphisms (SNPs)—9.3 million of which were rare. Most of the rare variants that occurred in coding regions of the DNA had not been found previously. The authors then compared those whole-genome sequences with partial sequences obtained by genotype-array analysis of a much larger group of volunteers. These arrays pick up common SNPs. Knowing which SNPs are normally inherited en masse, and pinpointing where the rare variants fall in those blocks of DNA, the authors imputed, in genetic parlance, the occurrence of those rare variants in a total of 101,584 people, and by extrapolation almost 300,000 of their close relatives, both living and deceased. About 320,000 people live in Iceland.

The authors correlated those variants with various diseases. As reported in one of the papers, first author Stacy Steinberg and colleagues looked for rare loss-of-function variants in genomic regions that previously had been implicated in Alzheimer's disease by genome-wide association studies (Apr 2011 news; Lambert et al., 2013). Among 3,419 cases and more than 151,000 controls, they found eight rare variants within the ABCA7 gene that associated with the disease. These included nonsense and missense mutations, frameshifts, and intron splice-site mutations. In all cases the researchers concluded that the mutant gene made no protein. On average, people with these variants had twice the chance of getting AD. Steinberg found no rare variants in other GWAS genes that correlated with AD in their population.

The researchers then looked for these same loss-of-function ABCA7 polymorphisms in cohorts from Finland, Germany, Norway, and the United States. They found six out of the eight, and they turned up more often in people with AD. On average, people carrying these rare variants were 1.73 times likelier to have the disease than those who had the common alleles. None of these new variants matched the common SNPs around the ABCA7 gene locus that emerged in GWAS. However, because these rare variants all cause a loss of function, it suggests that the GWAS SNPs would also lead to reduced levels of the protein, wrote the authors.

ABCA7, short for ATP-binding cassette transporter A7, belongs to the “A” subfamily of ABC transporters that ferry lipids across membranes. The central nervous system expresses the protein in large amounts, especially in microglia. An in vitro study first suggested that ABCA7 plays a part in AD by enhancing lipid transport and reducing Aβ processing (Chan et al., 2008). Later, GWAS backed up that hypothesis when they revealed that SNPs in ABCA7 associated with AD (Hollingworth et al., 2011). ABCA7 variants also appeared to intensify plaque burden and memory decline (Shulman et al., 2013Carrasquillo et al., 2015). However, it was unclear if SNPs in ABCA7 affected the gene itself or a neighbor, and whether these variants increased or decreased the normal amount of protein. “We now know from this paper that ABCA7 is the correct gene to study [in this locus], but we still don’t know how it relates to Alzheimer’s disease,” said Gerard Schellenberg, University of Pennsylvania, Philadelphia.

In cellular models, ABCA7 loads phospholipids and cholesterol onto ApoA-I or ApoE (Abe-Dohmae et al., 2004). However, ABCA7 knockout mice have no ApoE changes, suggesting they have no lipid transport deficits (Kim et al., 2013). Other evidence points to ABCA7 supporting phagocytosis. Its ortholog in C. elegans facilitates phagocytosis of dying cells (Wu and Horvitz, 1998). Likewise, ABCA7 enhances macrophage phagocytosis of apoptotic cells in mice (Jehle et al., 2006). ABCA7 expression also ticks up in response to phagocytosis in mouse and human cell lines (Iwamoto et al., 2006). Stefansson and colleagues found no genetic evidence for synergism between ABCA7 variants and those in either the ApoE or the TREM2 genes.

Do these results point to a therapeutic target? Schellenberg was unsure. Since the data suggest that lowering ABCA7 is detrimental, they imply that boosting the protein or its activity could be beneficial. “It is relatively easy to inhibit a protein, but a lot harder to activate it,” said Schellenberg. More interesting than harmful loss-of-function variants, such as these ABCA7 polymorphisms, are protective loss-of-function changes, he said. They point to cases where inhibition may be desirable, and that could lead to good drug targets. Stefansson predicted such protective variants will arise from deCODE's analysis.

Nilufer Ertekin-Taner of the Mayo Clinic in Jacksonville, Florida, found it interesting that of all the potential candidate genes that emerged from GWAS studies on late-onset AD, the deCODE study suggests that only ABCA7 harbors rare loss-of-function variants. “The results hint that for the other GWAS hits, the functional culprit may not be rare deleterious variants with strong effect sizes,” she told Alzforum. “We may need to look for alternative types of functional variants, such as those in regulatory regions, to identify the full spectrum of genetic variation that underlies the AD GWAS signals.”

The findings in the ABCA7 paper make up only one part of a larger study. In another paper, first authors Daniel Gudbjartsson, Hannes Helgason, and colleagues detail other genetic variants that associate with different diseases. For instance, a frameshift deletion in MYL4, which encodes a myosin essential light chain, correlated with early onset atrial fibrillation. Likewise, missense and frameshift mutations in the ATP-binding cassette, sub-family B member 4 (ABCB4) heightened risk for liver and gallstone diseases. Other findings had less to do with disease and more to do with general human genetics. For instance, in another paper, co-first authors Helgason and Patrick Sulem reported that 7.7 percent of the Icelandic population studied had a gene that was completely knocked out. These homozygous loss-of-function mutations are equivalent to mouse knockouts. Stefansson said they plan to phenotype these individuals and see how the missing genes relate to disease and physiology. Last, but not least, a paper from Agnar Helgason and colleagues estimated the mutation rate of the Y chromosome, and predicted that the most recent common male ancestor to modern humans lived about 239,000 years ago.

DeCode cannot share individual data from the study because of restrictions laid out by Icelandic law. However, the company does contribute to other studies by sending out their summary statistics, such as effect sizes and P-values. Interested collaborators can also travel to Iceland to gain local access to the data. The company’s methods have previously paid off for AD. Their scientists found a variant of the amyloid precursor protein (APP) that protects against AD, as well as a rare mutation in the microglial activation protein TREM2 that triples risk for the disease (Jul 2012 news; Nov 2012 news).—Gwyneth Dickey Zakaib

Comments

  1. Loss-of-function variants in ABCA7 highlight an important role in Alzheimer’s disease

    Early in vitro studies suggested that ABCA7 may play a protective role in Alzheimer’s disease (AD) (Chan et al., 2008). Interest in the association of ABCA7 with AD has grown rapidly due to the identification of ABCA7 SNPs that confer increased risk for AD (for example, see Hollingworth et al., 2011; Reitz et al., 2013; Shulman et al., 2013; Lambert et al., 2013; Carrasquillo et al., 2015; Yu et al., 2015). This new publication from Steinberg et al. provides further compelling evidence that ABCA7 is a clinically relevant contributor to AD risk. The authors identified ABCA7 loss-of-function variants in an Icelandic population that conferred increased AD risk (OR = 2.12). Although these variants may be rare, the study provides important new information highlighting the functional importance of ABCA7 in the brain. The overall conclusions suggest that reduced levels of the ABCA7 protein underlie the increase in AD risk associated with ABCA7 variants. Given the highly reproducible findings from several studies that confirm an association between ABCA7 and AD, it is now important to focus on the true biological function of ABCA7 in the brain. In a study of the association between the top 10 GWAS “hits” for AD risk and plaque pathology, ABCA7 was one of four genes found to be significantly associated with plaque burden (Shulman et al., 2013). Although numerous pathological processes may be influenced by ABCA7 in the brain, an assessment of its role in the regulation of plaque pathology appears to be the most logical place to start.

    A recent study of ours provided evidence that loss of ABCA7 function has a profound impact on amyloid plaque pathology (Kim et al., 2013). We generated amyloidogenic J20 AD mice that were Abca7-deficient (J20/A7-/-) and discovered that their hippocampal amyloid plaque levels doubled compared to J20 littermates (Kim et al., 2013). We also showed that the clearance of oligomeric amyloid-β (Aβ) was impaired in phagocytic cells derived from Abca7-deficient mice as compared to wild-type-mouse-derived phagocytes (Kim et al., 2013). The questions that now need to be addressed are related to the identification of the precise mechanisms by which the ABCA7 protein regulates AD risk, and whether therapeutic manipulation of the pathways involved can be exploited to treat patients.

    Although ABCA7 is a member of the “A” subfamily of ATP-binding cassette transporters that were initially characterized by their capacity to transport lipids across membranes (Kim et al., 2013), unlike its closest homologue (ABCA1), ABCA7 does not appear to play a major role in transporting cholesterol across cell membranes to lipid-free ApoE (Chan et al., 2008). Indeed, in the absence of Abca7 (in our J20/A7-/- mouse experiments), we found no evidence for a reduction in brain ApoE levels that would imply inadequate ApoE lipidation (Kim et al., 2013). ABCA7 functions that are not related to lipid transport should therefore be considered in the context of AD.

    In addition to its similarity with ABCA1, ABCA7 shares sequence homology (24 percent identity, 43 percent similarity) with the C. elegans CED-7 protein (Jehle et al., 2006). CED-7 plays a crucial role in phagocytosis and it is recognized that ABCA7 also regulates phagocytosis in several mammalian cell types (Jehle et al., 2006; Iwamoto et al., 2006). These findings are in accord with the high expression of ABCA7 in macrophages and microglia (Kaminski et al., 2000; Kim et al., 2006). As ABCA7 is highly expressed in the brain (Kim et al., 2013; Kim et al., 2008), and this is particularly due to microglial expression (Kim et al., 2006), a function for ABCA7 in the phagocytic clearance of Aβ could explain the highly reproducible association of ABCA7 variants with increased AD risk. If a phagocytic role for ABCA7 is conclusively demonstrated (for example, in studies utilising myeloid cell conditional Abca7-deficient AD mice), a new AD therapeutic target could arise by virtue of a direct stimulation of oligomeric Aβ clearance by microglia. Such a pathway would also have implications for current Aβ immunotherapeutic strategies that ultimately depend on microglial phagocytic clearance of Aβ-antibody complexes. In showing that loss-of-function variations in ABCA7 confer AD risk, the new findings from Steinberg et al. place questions regarding the true function of ABCA7 protein in the brain at center stage.

    References:

    . ATP-binding cassette transporter A7 regulates processing of amyloid precursor protein in vitro. J Neurochem. 2008 Jul;106(2):793-804. PubMed.

    . Common variants at ABCA7, MS4A6A/MS4A4E, EPHA1, CD33 and CD2AP are associated with Alzheimer's disease. Nat Genet. 2011 May;43(5):429-35. PubMed.

    . Variants in the ATP-binding cassette transporter (ABCA7), apolipoprotein E ϵ4,and the risk of late-onset Alzheimer disease in African Americans. JAMA. 2013 Apr 10;309(14):1483-92. PubMed.

    . Genetic susceptibility for Alzheimer disease neuritic plaque pathology. JAMA Neurol. 2013 Sep 1;70(9):1150-7. PubMed.

    . Meta-analysis of 74,046 individuals identifies 11 new susceptibility loci for Alzheimer's disease. Nat Genet. 2013 Dec;45(12):1452-8. Epub 2013 Oct 27 PubMed.

    . Late-onset Alzheimer's risk variants in memory decline, incident mild cognitive impairment, and Alzheimer's disease. Neurobiol Aging. 2015 Jan;36(1):60-7. Epub 2014 Aug 4 PubMed.

    . Association of Brain DNA methylation in SORL1, ABCA7, HLA-DRB5, SLC24A4, and BIN1 with pathological diagnosis of Alzheimer disease. JAMA Neurol. 2015 Jan;72(1):15-24. PubMed.

    . Deletion of Abca7 increases cerebral amyloid-β accumulation in the J20 mouse model of Alzheimer's disease. J Neurosci. 2013 Mar 6;33(10):4387-94. PubMed.

    . Role of ATP-binding cassette transporters in brain lipid transport and neurological disease. J Neurochem. 2008 Mar;104(5):1145-66. PubMed.

    . ATP-binding cassette transporter A7 enhances phagocytosis of apoptotic cells and associated ERK signaling in macrophages. J Cell Biol. 2006 Aug 14;174(4):547-56. PubMed.

    . ABCA7 expression is regulated by cellular cholesterol through the SREBP2 pathway and associated with phagocytosis. J Lipid Res. 2006 Sep;47(9):1915-27. Epub 2006 Jun 20 PubMed.

    . Identification of a novel human sterol-sensitive ATP-binding cassette transporter (ABCA7). Biochem Biophys Res Commun. 2000 Jul 5;273(2):532-8. PubMed.

    . Quantitation of ATP-binding cassette subfamily-A transporter gene expression in primary human brain cells. Neuroreport. 2006 Jun 26;17(9):891-6. PubMed.

    View all comments by Brett Garner
  2. Steinberg and colleagues provide new compelling evidence for the genetic association between loss-of-function variants in the lipid transporter ABCA7 and increased risk of Alzheimer’s disease. Whole-genome sequences of 2,636 Icelanders were imputed into 104,220 long-ranged phased individuals and association testing was performed with 3,419 people with AD and 151,805 population controls. The most significant association, for both loss-of-function and missense variants, was for ABCA7, which had a combined odds ratio of 2.12 (p=2.2 x 10-13), and is distinct from the common variant previously reported to be associated with AD risk. Intriguingly, there was no evidence of interaction of AD-associated alleles of either APOE or TREM2.

    How ABCA7 contributes to AD risk is not yet entirely understood, and this now becomes a critically important question. ABCA7 is a member of the A subclass of ATP-binding cassette transporters that function to transport lipids across cellular membranes (Kim et al., 2008). ABCA7 is highly expressed in the central nervous system, and is particularly abundant in microglia. In cellular models, ABCA7 can function to promote cholesterol and phospholipid efflux to either ApoE or ApoA-I lipid acceptors, leading to hypotheses regarding its potential role in lipoprotein biogenesis. However, the ancestral role of ABCA7 may be more aligned with a role in regulating phagocytic activity. Evidence supporting this function includes in vitro studies showing that ABCA7 can promote phagocytosis in cell lines, and that ced7, a functional orthologue of ABCA7 in C. elegans, is essential for engulfment of apoptotic cells (Iwamoto et al. 2006; Jehle et al., 2006). Intriguingly, ABCA7 has been shown to preferentially flip phosphatidylserine from the cytoplasmic to the exocytoplasmic leaflet of membranes in an energy-dependent manner (Quazi and Molday, 2013), raising the hypothesis that ABCA7 may contribute to the presentation of “eat-me” signals in cells destined for engulfment.

    Regulation of the innate immune system is intimately regulated with that of lipid metabolism, particularly for professional phagocytes that ingest lipid-rich debris. Lipidomic approaches have begun to reveal distinct lipid mediator signatures in human phagocytes, which may influence both lipid trafficking and phagocytic activities across the M1-M2 macrophage activation spectrum (Dalli and Serhan, 2012; Chinetti-Gbaquidi et al., 2011).

    The increasing evidence that genetic variants in ABCA7 contribute to the risk of AD raises several important questions for future studies. Understanding the mechanisms by which ABCA7 in relevant cell types regulates phagocytic clearance of damaged cells is clearly a high-priority area. It will also be important to determine whether ABCA7 activity contributes to clearance of cellular debris after acute brain insult.

    References:

    . Role of ATP-binding cassette transporters in brain lipid transport and neurological disease. J Neurochem. 2008 Mar;104(5):1145-66. PubMed.

    . ABCA7 expression is regulated by cellular cholesterol through the SREBP2 pathway and associated with phagocytosis. J Lipid Res. 2006 Sep;47(9):1915-27. Epub 2006 Jun 20 PubMed.

    . ATP-binding cassette transporter A7 enhances phagocytosis of apoptotic cells and associated ERK signaling in macrophages. J Cell Biol. 2006 Aug 14;174(4):547-56. PubMed.

    . Differential phospholipid substrates and directional transport by ATP-binding cassette proteins ABCA1, ABCA7, and ABCA4 and disease-causing mutants. J Biol Chem. 2013 Nov 29;288(48):34414-26. Epub 2013 Oct 4 PubMed.

    . Specific lipid mediator signatures of human phagocytes: microparticles stimulate macrophage efferocytosis and pro-resolving mediators. Blood. 2012 Oct 11;120(15):e60-72. Epub 2012 Aug 17 PubMed.

    . Human atherosclerotic plaque alternative macrophages display low cholesterol handling but high phagocytosis because of distinct activities of the PPARγ and LXRα pathways. Circ Res. 2011 Apr 15;108(8):985-95. Epub 2011 Feb 24 PubMed.

    View all comments by Cheryl Wellington
  3. First, these data essentially confirm the common ABCA7 GWAS SNP as a legitimate genetic risk for Alzheimer's disease. Overall, this story is a nice integration of the rare variant/robust risk and common variant/modest risk thought process. SNPs with large effects on complex human disease are likely to be of two kinds: rare SNPs with large impact, or common SNPs with modest impact. ABCA7 now has examples of both types.

    Second, the data support the role of microglia in AD because ABCA7 has been linked with phagocytosis (Jehle et al., 2006; Kim et al., 2013; Tanaka et al., 2011). The alternative ABCA7 mechanism involves lipid transport, but the more compelling data support phagocytosis. That said, there are macrophage data that ABCA7 moves into the membrane forming the phagocytic cup (Jehle et al., 2006); perhaps ABCA7 functions to transport lipids to help form this cup. Overall, the result here that loss of ABCA7 increases AD risk fits nicely with the emerging role of microglia in AD.

    Third, the question of what this paper means for AD treatment and/or prevention arises. The odds ratio will put ABCA7 on the pharmacology map. The approach, though, is a bit more nebulous, because it’s harder to activate a protein than to inhibit it. That said, based on the idea that more ABCA7 is helpful, there are reports that statins, for example, can increase ABCA7 (Tanaka et al., 2011), so this could help resurrect statins as AD therapeutics. 

    References:

    . ATP-binding cassette transporter A7 enhances phagocytosis of apoptotic cells and associated ERK signaling in macrophages. J Cell Biol. 2006 Aug 14;174(4):547-56. PubMed.

    . Deletion of Abca7 increases cerebral amyloid-β accumulation in the J20 mouse model of Alzheimer's disease. J Neurosci. 2013 Mar 6;33(10):4387-94. PubMed.

    . HMG-CoA reductase inhibitors enhance phagocytosis by upregulating ATP-binding cassette transporter A7. Atherosclerosis. 2011 Aug;217(2):407-14. Epub 2011 Jun 23 PubMed.

    View all comments by Steven Estus

Make a Comment

To make a comment you must login or register.

References

News Citations

  1. Large Genetic Analysis Pays Off With New AD Risk Genes
  2. Protective APP Mutation Found—Supports Amyloid Hypothesis
  3. Enter the New Alzheimer’s Gene: TREM2 Variant Triples Risk

Paper Citations

  1. . Meta-analysis of 74,046 individuals identifies 11 new susceptibility loci for Alzheimer's disease. Nat Genet. 2013 Dec;45(12):1452-8. Epub 2013 Oct 27 PubMed.
  2. . ATP-binding cassette transporter A7 regulates processing of amyloid precursor protein in vitro. J Neurochem. 2008 Jul;106(2):793-804. PubMed.
  3. . Common variants at ABCA7, MS4A6A/MS4A4E, EPHA1, CD33 and CD2AP are associated with Alzheimer's disease. Nat Genet. 2011 May;43(5):429-35. PubMed.
  4. . Genetic susceptibility for Alzheimer disease neuritic plaque pathology. JAMA Neurol. 2013 Sep 1;70(9):1150-7. PubMed.
  5. . Late-onset Alzheimer's risk variants in memory decline, incident mild cognitive impairment, and Alzheimer's disease. Neurobiol Aging. 2015 Jan;36(1):60-7. Epub 2014 Aug 4 PubMed.
  6. . Human ABCA7 supports apolipoprotein-mediated release of cellular cholesterol and phospholipid to generate high density lipoprotein. J Biol Chem. 2004 Jan 2;279(1):604-11. Epub 2003 Oct 21 PubMed.
  7. . Deletion of Abca7 increases cerebral amyloid-β accumulation in the J20 mouse model of Alzheimer's disease. J Neurosci. 2013 Mar 6;33(10):4387-94. PubMed.
  8. . The C. elegans cell corpse engulfment gene ced-7 encodes a protein similar to ABC transporters. Cell. 1998 Jun 12;93(6):951-60. PubMed.
  9. . ATP-binding cassette transporter A7 enhances phagocytosis of apoptotic cells and associated ERK signaling in macrophages. J Cell Biol. 2006 Aug 14;174(4):547-56. PubMed.
  10. . ABCA7 expression is regulated by cellular cholesterol through the SREBP2 pathway and associated with phagocytosis. J Lipid Res. 2006 Sep;47(9):1915-27. Epub 2006 Jun 20 PubMed.

External Citations

  1. ABCA7

Further Reading

Papers

  1. . A multiancestral genome-wide exome array study of Alzheimer disease, frontotemporal dementia, and progressive supranuclear palsy. JAMA Neurol. 2015 Apr;72(4):414-22. PubMed.
  2. . Genome-wide association meta-analysis of neuropathologic features of Alzheimer's disease and related dementias. PLoS Genet. 2014 Sep;10(9):e1004606. Epub 2014 Sep 4 PubMed.
  3. . ABCA7 gene and the risk of Alzheimer's disease in Han Chinese in Taiwan. Neurobiol Aging. 2014 Oct;35(10):2423.e7-2423.e13. Epub 2014 May 14 PubMed.
  4. . ABCA7 in Alzheimer's Disease. Mol Neurobiol. 2014 May 31; PubMed.

Primary Papers

  1. . Loss-of-function variants in ABCA7 confer risk of Alzheimer's disease. Nat Genet. 2015 May;47(5):445-7. Epub 2015 Mar 25 PubMed.
  2. . Large-scale whole-genome sequencing of the Icelandic population. Nat Genet. 2015 Mar 25; PubMed.
  3. . Identification of a large set of rare complete human knockouts. Nat Genet. 2015 Mar 25; PubMed.
  4. . The Y-chromosome point mutation rate in humans. Nat Genet. 2015 Mar 25; PubMed.