Skip to main content
Premium Trial:

Request an Annual Quote

Alzheimer's Disease Gene Expression Varies by Brain Cell Type

NEW YORK – New research is revealing how gene expression shifts in Alzheimer's patients vary from one brain cell type to the next — not only in neurons, but also in non-neuronal cell types.

"Perhaps the answer to treating Alzheimer's lies in understanding how these non-neuronal cells are affected during disease," co-first author Alexandra Grubman, a researcher affiliated with Monash University and the Monash Biomedicine Discovery Institute, said in a statement.

As they reported online today in Nature Neuroscience, Grubman, senior author Jose Polo, and colleagues from Australia and Singapore Medical School used a method called DroNc-Seq to perform single-nucleus RNA sequencing on more than 13,200 nuclei from half a dozen samples from the entorhinal cortex region of the brain from individuals with Alzheimer's and as many unaffected controls. Their results revealed Alzheimer's-related changes in gene expression in specific brain cell types, highlighting altered genetic networks as well as particularly important cell populations in this brain region.

"Notably, our data reveal complex patterns of expression changes for multifold Alzheimer's disease genes within or across specific cell populations," Grubman and her co-authors wrote, noting that "[t]his complexity should be taken into account to enhance our interpretation of genetic discoveries in Alzheimer's disease."

Genetic variants in APOE and other genes have been implicated in late-onset Alzheimer's disease through genome-wide association studies and other analyses, the team explained. But less is known about the relative contributions that different cell subpopulations make to Alzheimer's disease risk and development, despite evidence pointing to a potential role for genes and pathways involved in endocytosis, microglial function, and neuronal connectivity. 

"In light of the complex genetic contributions to [late-onset Alzheimer's disease], it is important to understand [late-onset Alzheimer's disease] GWAS gene involvement in cell-type-specific transcription factor networks that drive the transitions of cells from health to Alzheimer's disease states," the authors wrote.

The team suggested that it may be possible to gain a better understanding of the compartmentalization, regulatory shifts, and functional changes behind Alzheimer's disease by taking a closer look at the genes and cellular subpopulations showing altered expression profiles in certain cell subpopulations.

Starting from almost 14,900 cells, the researchers used a DroNc-Seq protocol with the 10x Genomics platform to generate high-quality nuclear RNA sequences for 13,214 cells from entorhinal cortex samples from six individuals with Alzheimer's disease and six unaffected controls matched for age and sex.

From there, the team identified cell clusters that roughly coincided with microglia, astrocyte, neuron, oligodendrocyte progenitor cell, oligodendrocyte, and endothelial cell types by combining the single-cell transcriptomes in combination with spatial information produced using an approach known as "uniform manifold approximation and projection."

The investigators reported that the Alzheimer's disease-related gene APOE was found at enhanced levels in a certain subpopulation of microglia cells, for example, but was dialed down in nuclei from other brain cell subpopulations, including the oligodendrocyte progenitor and astrocyte cell populations.

By folding in regulatory and genetic risk data, meanwhile, the investigators searched for the genetic changes that push certain cell subpopulations towards the dementia-causing neurodegenerative disease.

"These results provide insights into disease networks in the human brain," the authors wrote, explaining that the data offer opportunities to combine transcriptional network clues with genetic insights gleaned from GWAS and other studies. They added that these processes will continue becoming clearer as transcriptomic and genetic risk features are untangled in still larger patient populations.