At the Icahn Institute for Data Science and Genomic Technology, we’re dedicated to studying Late Onset Alzheimer's Disease (LOAD), the most common––and lesser studied––form of Alzheimer’s.
For the past 30 years, clinicians, epidemiologists, and geneticists have focused primarily on unusual forms of Alzheimer's that begin early (usually around age 50) and cluster in families. From these decades of research, three significant discoveries have been made:
- There are three genes that may have mutations guaranteeing a heritable form of Alzheimer's––contributing to about 3 percent of all cases.
- A single gene acts as a risk factor for, and plays some role in, at least half of all Alzheimer's cases.
- There are 20 other risk factor genes, each with their own contribution to risk.
The first two findings, four genes in total, can be used clinically. The last finding, the other 20 genes, cannot. However, all 24 genes converge on a pathway beginning with amyloid and proceeding to tangles, inflammation, cell injury, and death.
Unfortunately, these risk factors and candidate genes don't tell us much about the cascade signaling pathways underlying Alzheimer’s. In order to develop effective disease modifying or preventive therapies, we need to understand how the disease begins. Therefore, we need to innovative new approaches that will help us identify these causal mechanisms.
Our researchers, Bin Zhang, PhD, Jun Zhu, PhD, and Eric Schadt , PhD, have pioneered a breakthrough approach––known as multi-scale network analysis––that integrates of all the data accumulated from previous genetic studies with gene expression data from the brains of patients with common forms of Alzheimer's. This integration of genetics and gene expression is the essence of multiscale network analysis. This gene expression data also helps amplify tiny genetic signals that would otherwise be overlooked.
A 2013 paper, published in Cell, by Dr. Zhang and colleagues, reports the discovery of a network of genes involved in inflammation response as the top ranked pathway causally linked to LOAD.
These human brain gene networks reveals multiple aspects of molecular modifications in LOAD across multiple brain regions. This new understanding of key pathways and genes involved in LOAD offers valuable insights to develop potential therapies for the disease.
Zhang et al. identified an inflammatory protein, known as TYROBP, as the key driver of the discovered inflammation network. TYROBP interacts with another molecule TREM2, which was also linked to Alzheimer's, a study in the New England Journal of Medicine reported.
Together, these two discoveries implicate that the TREM2-TYROBP pathway may be at the center of common forms of Alzheimer's, playing a key role in amyloid clearance and amyloid related inflammation. Recently, another inflammatory pathway molecule, CR1, was recently shown to be linked to Alzheimer's but rather than increasing brain amyloid, the risky form of CR1 reduces brain amyloid.
This paper in Cell, along with the previous publications about TREM2 and CR1, has lead researchers to revisit the inflammatory hypothesis of Alzheimer's. This hypothesis held favor for many years, fell out of favor, and is now being revisited in clinical trials. We expect to see much more attention paid to inflammation in the next five to ten years.
This connection may help us understand the long lag time between appearance of amyloid on brain scans and the appearance of clinical symptoms. An individual's inflammatory response could well play a role in that progression, and the appropriate anti-inflammatory drug––given after amyloid is detected, before symptoms begin––could be an important part of dementia prevention.
As a next step, our team plans to validate key causal networks and evaluate drugs that will affect the TREM2-TYROBP pathway, and other key pathways, as potential therapies for Alzheimer's disease.
Our team is also working to apply their gene network modeling approach to other neurodegenerative diseases.
Learn more about our Alzheimer's disease research