Researchers believe the immune system plays a role in Alzheimer’s disease, but exactly how remains fuzzy. One way to clarify the picture would be to identify immune-related genes that affect AD risk. In the April 18 JAMA Neurology, researchers led by Rahul Desikan at the University of California, San Francisco, describe how they used a new method to unearth genetic polymorphisms linked to both AD and autoimmune disorders. They identified two immune-linked genetic variants that pump up risk for Alzheimer’s, as well as six that lower it. Both risk variants correlated with neurofibrillary tangles in AD brains, and one of them associated with steeper cognitive decline in cognitively impaired people.

“Shared biological pathways may underlie both autoimmune and Alzheimer’s disease,” first author Jennifer Yokoyama wrote to Alzforum. “This suggests that immune processes may directly contribute to Alzheimer’s pathology and disease progression, rather than just respond to them.”

In an accompanying editorial, Huntington Potter at the University of Colorado Anschutz Medical Center, Aurora, praised the approach. “Yokoyama et al. provide a major intellectual and technological advance in our understanding of the relationship between the immune system and Alzheimer's disease,” he wrote. Part of the advance lies in the method, which establishes a procedure for comparing risk genes between AD and any other disorder, Potter noted. In addition, the findings link immune-related disorders as a class to AD, which may provide clues to pathogenesis, he added.

While researchers have long seen signs of inflammation in Alzheimer’s brains, only in recent years has genetic evidence suggested that immune processes themselves might promote pathology. In particular, attention to the role of inflammation was renewed by the discoveries that variants in the microglial receptor TREM2 tripled AD risk and that CD33 variants impaired Aβ clearance (see Oct 2008 newsNov 2012 news; May 2013 webinar). Researchers have now turned up several more immune genes linked to AD, such as antigen-presenting proteins and genes involved in adaptive immunity (see Jul 2013 conference newsApr 2015 conference news). 

To search for additional immune genes, Desikan and colleagues employed a new method for identifying pleiotropic variants, i.e., those associated with multiple disorders (see Andreassen et al., 2013; Andreassen et al., 2013). By taking a handful of genetic variants that correlate with one disorder and analyzing them for links to a second, researchers increase the statistical power for finding an association. This is because such studies do not have to correct for the effect of making millions of comparisons, as typical GWAS do, and so can use a lower significance threshold (see Mar 2015 news). 

Using this method, the authors analyzed data from the International Genomics of Alzheimer’s Project (IGAP) stage 1 cohort, made up of 17,008 people with AD and 37,154 age-matched controls from four consortia in Europe and North America. Yokoyama and colleagues first considered subsets of SNPs linked to one of six autoimmune disorders: Crohn’s disease, ulcerative colitis, rheumatoid arthritis, type 1 diabetes, celiac disease, and psoriasis. In each case, they found that people with those SNPs carried more SNPs associated with Alzheimer’s disease than would be expected by chance. Conversely, people with AD SNPs carried more SNPs associated with those six immune disorders than they would by chance.

Next, the authors tested individual risk SNPs for each autoimmune disorder to find if any were significantly associated with AD. They turned up eight. Two of these heightened risk for both AD and the immune condition, while the other six lowered risk for both. In initial analyses, the authors focused on the two loci that raised risk, but the other variants deserve study as well, Yokoyama wrote to Alzforum. One of the two, rs2516049, associates with psoriasis and resides near the HLA-DRB5 gene. This codes for a subunit of a major histocompatibility complex (MHC) class II receptor. These proteins present antigens to T cells. The other SNP, rs12579988, boosts risk for Crohn’s and lies near the inositol polyphosphate multikinase (IPMK) gene. IPMK regulates cell signaling and gene expression and associates with inflammatory bowel disease as well, but exactly how it interacts with the immune system remains to be determined, Yokoyama noted.

To validate these SNPs, the authors looked for a relationship to pathology using the Alzheimer’s Disease Genetics Consortium database, which includes 4,914 brain autopsies and has associated genetic data (see Beecham et al., 2014). Both SNPs correlated with modestly worse tangle pathology, but had no relationship to amyloid plaques.

Mouse studies have found that the expression of many genes associated with AD changes when amyloid plaques, but not neurofibrillary tangles, are present, pointed out John Hardy at University College London (see Jan 2015 news). “[This] suggests perhaps that the order of events is plaque deposition, immune response, tangle deposition,” he wrote to Alzforum (see full comment below). Such a sequence would fit with the data linking these immune gene variants to increased tangles, but not plaques. Neuroinflammation and plaques are often found side by side in early disease (see Eikelenboom et al., 2000; Hoozemans et al., 2002). 

Since the SNPs associated with worse pathology, might they affect the clinical course of the disease? To answer this, the authors examined cognitive data from 723 participants in the Alzheimer’s Disease Neuroimaging Initiative. About half of ADNI participants have mild cognitive impairment, a quarter AD dementia, and the remainder are cognitively healthy. In the whole cohort, those who carried the SNP near HLA-DRB5 declined 50 percent more on cognitive tests over time than did those without the variant.

GWAS SNPs usually act as a tag for a nearby polymorphism that is the actual disease-modifying variant. That real McCoy can be difficult to identify. To see if the HLA-DRB5 and IPMK genes themselves might be the culprits, the authors examined data from the Gene Expression Omnibus dataset of 376 AD patients and 173 controls (see Zhang et al., 2013). IPMK expression was about 10 percent lower in AD patients, hinting that this gene might indeed play a role. HLA-DRB5 expression was unchanged.

In future work, the authors will explore how these genes might contribute to Alzheimer’s, as well as study the six protective SNPs. “[The findings] offer an inroad into the Alzheimer’s disease process and point to biological systems that might be amenable to therapeutic intervention,” Yokoyama wrote.—Madolyn Bowman Rogers


  1. It is clear that neuroimmune response is key to the genetics of Alzheimer’s disease, not just because of the identification of TREM2 as a risk locus but also from the systematic analysis of the GWAS hits (Jones et al., 2010). Interestingly, transgenic mouse work has shown that many of the GWAs hits, and TREM2, are the biggest responders to plaque deposition, and overlaying GWAS loci with the genes in the TREM2 module of co-expression shows just how important this process appears to be (Matarin et al., 2015). These genes do not respond as markedly in response to tangle deposition, suggesting perhaps that the order of events is plaque deposition, immune response, tangle deposition.


    . Genetic evidence implicates the immune system and cholesterol metabolism in the aetiology of Alzheimer's disease. PLoS One. 2010;5(11):e13950. PubMed.

    . A genome-wide gene-expression analysis and database in transgenic mice during development of amyloid or tau pathology. Cell Rep. 2015 Feb 3;10(4):633-44. Epub 2015 Jan 22 PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Four New Acts Debut on GWAS Screen
  2. Enter the New Alzheimer’s Gene: TREM2 Variant Triples Risk
  3. Pooled GWAS Reveals New Alzheimer’s Genes and Pathways
  4. Could Adaptive Immunity Set the Brakes on Amyloid?
  5. Tau Gene Confirmed to Amp Up Alzheimer’s Risk and Neurodegeneration
  6. Network Analysis Points to Distinct Effects of Amyloid, Tau

Webinar Citations

  1. Can Network Analysis Identify Pathological Pathways in Alzheimer’s

Paper Citations

  1. . Improved detection of common variants associated with schizophrenia by leveraging pleiotropy with cardiovascular-disease risk factors. Am J Hum Genet. 2013 Feb 7;92(2):197-209. Epub 2013 Jan 31 PubMed.
  2. . Improved detection of common variants associated with schizophrenia and bipolar disorder using pleiotropy-informed conditional false discovery rate. PLoS Genet. 2013 Apr;9(4):e1003455. Epub 2013 Apr 25 PubMed.
  3. . 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.
  4. . Neuroinflammation and Alzheimer disease: clinical and therapeutic implications. Alzheimer Dis Assoc Disord. 2000;14 Suppl 1:S54-61. PubMed.
  5. . The pathological cascade of Alzheimer's disease: the role of inflammation and its therapeutic implications. Drugs Today (Barc). 2002 Jun;38(6):429-43. PubMed.
  6. . Integrated systems approach identifies genetic nodes and networks in late-onset Alzheimer's disease. Cell. 2013 Apr 25;153(3):707-20. PubMed.

Further Reading

Primary Papers

  1. . Association Between Genetic Traits for Immune-Mediated Diseases and Alzheimer Disease. JAMA Neurol. 2016 Jun 1;73(6):691-7. PubMed.
  2. . New Genetic Links Between Alzheimer Disease and Immune-Mediated Diseases Revealed by Overlapping Genome-wide Association Study Hits. JAMA Neurol. 2016 Jun 1;73(6):638-40. PubMed.