Alzheimer's disease: Putting the pieces together with integrative genomics

National Human Genome Research Institute

National Institutes of Health
U.S. Department of Health and Human Services


Genome Advance of the Month

Alzheimer's disease: Putting the pieces together with integrative genomics

By Elizabeth Burke, Ph.D.
Intramural Postdoctoral Fellow, NHGRI
Illustrated human head with DNA helices
Alzheimer's disease - a neurological disorder causing progressive dementia, disorientation and behavioral changes - will affect more than 5 million Americans this year. While five percent of those with Alzheimer's disease develop it between the ages of 30-65 as a result of any one of several rare, inherited, single-gene mutations, the large majority of affected individuals develop a non-familial form after the age of 65, called late-onset Alzheimer's disease (LOAD). 
 
In comparison to the early-onset form, the underlying cause of LOAD is much more complex; it is thought to be caused by a combination of several genetic and non-genetic risk factors. Genetic risk factors refer to common human genetic variations, or alleles, that increase a person's chance of developing a disease without directly causing it. While each risk factor is not sufficient to cause the disease on its own, multiple risk factors can combine their modest individual effects to develop LOAD.
 
Though advancing age is currently the strongest known risk factor for LOAD, the most influential genetic factor to be identified is one of three common alleles for the gene apolipoprotein E (APOE) that is referred to as APOE4. A person who inherits the APOE4 allele from only one parent has a three-fold increase in LOAD risk, whereas a person that inherits APOE4 from both parents is ten times more likely to develop LOAD.  Despite this strong association, it has remained unclear how APOE4 contributes to the disease.  July's Genome Advance of the Month describes a study, published in the August 1, 2013, issue of Nature, which combined several genomic methods to identify important regulatory processes that link the common genetic variation APOE4 to the development of LOAD.
 
The team of researchers from Columbia University began their investigation by using publically available data that contained gene expression measurements from a specific region of autopsied brains called the cerebral cortex, which is responsible for controlling the intellectual functions (learning, memory, etc.) that are affected in Alzheimer's disease.  The gene expression measurements provided a record of the number of times each gene sequence is copied or 'transcribed' into messenger RNA (a code that contains instructions for making the protein) within the cerebral cortex, allowing the researchers to see whether the presence of APOE4 causes brain cells to express higher or lower levels of certain genes.
 
To this end, they compared the expression level of all genes in the genome that had been transcribed into messenger RNA (transcriptome) in the cerebral cortex of APOE4-negative, healthy individuals to that of both APOE4-negative LOAD patients and APOE4-positive, healthy individuals.  This comparison allowed the team to examine the effects of LOAD disease and the APOE4 allele on gene expression independently of one another.  Interestingly, many of the gene expression changes found in healthy APOE4 carriers matched the gene expression changes seen in LOAD patients. This suggests that individuals that carry the APOE4 allele have a specific set of gene expression changes in their brain that promotes LOAD (APOE4/LOAD expression profile), thereby increasing their risk of developing the disease.
 
To identify key regulatory genes that may be responsible for initiating this transcriptome-wide change in gene expression, the researchers performed a second analysis that focused on changes in co-expression, or simultaneous expression, of gene pairs. An instance of a change in co-expression would be if expression of both gene A and gene B were normally high in APOE4-negative tissue, but APOE4-positive tissue showed low expression of gene A and high expression of gene B.  If gene A's co-expression with several other genes was also changed, this would suggest gene A is an important regulatory gene.
 
Using this approach, the researchers sought to determine which genes were most altered in their co-expression with the set of genes in the APOE4/LOAD expression profile. The two most highly ranked candidate genes were RNF219, which has been associated with changes in cognitive performance, and SV2A, which is a well-described regulator of protein transport in brain cells.  Several subsequent experiments performed in human cells verified that these two genes also play critical regulatory roles in the processing and transport of amyloid precursor protein (APP) within brain cells that contain APOE4.  This is extremely relevant as a small portion of APP is cut off to generate a smaller protein fragment called amyloid beta, which is known to accumulate and form clumps in the brain cells of Alzheimer's patients as well as APOE4 carriers.  Taken together, these identified regulatory genes link the common genetic risk factor APOE4 to the disease-causing brain changes seen in LOAD. 
 
Although researchers are still missing a few pieces in this puzzle, the work presented here has significantly advanced understanding of the genetic mechanisms that increase LOAD risk.  Not only did this study determine that the APOE4 allele increases an individual's risk of LOAD through a specific set of gene expression changes, but it also identified two genes that play an important role in regulating APOE4-dependent amyloid beta production.  Overall, these impressive results provide a major steppingstone for future Alzheimer's research and demonstrate the power of an integrative genomics approach when it's applied to the study of complex, multifactorial diseases like LOAD.
 
Read the article: Rhinn H, Fujita R, Qiang L, Cheng R, Lee JH, Abeliovich A.  Integrative genomics identifies APOE ?4 effectors in Alzheimer's disease. Nature, 500(7460): 45-50. 2013.  PMID: 23883936. [PubMed]

Top of page

Posted: August 21, 2013