It is commonly thought that neurons, especially excitatory ones, produce most of the Aβ in the brain. A paper in the March 5 Journal of Neuroscience suggests other cell types contribute as well. Scientists led by Sangram Sisodia, University of Chicago, report that brain cells such as microglia and astrocytes produce their fair share of the peptide and contribute to Aβ accumulation. “There is an emerging view that synaptic activity could be the source of most of the Aβ that is deposited or interfering with cognition,” said David Borchelt, University of Florida, Gainesville. “This is a nice piece of work to show that that Aβ can come from multiple sources.”

Past studies in mice have already shown that non-neuronal cells in the brain, such as astrocytes, can produce their own Aβ (see Zhao et al., 2011, and Lesné et al., 2003). However, although neurons make up only about 10 percent of the cells in the brain, scientists tend to focus on their Aβ production. A number of studies previously reported that neuronal activity can modulate Aβ levels in the central nervous system (CNS) (see Kamenetz et al., 2003Cirrito et al., 2005). Sisodia’s group wanted to know whether cells other than excitatory neurons contributed to the brain’s overall Aβ aggregation.

To find out, first author Karthikeyan Veeraraghavalu and colleagues used a Cre-loxP system to generate mice that expressed a mutated version of presenilin-1 (PS1∆E9) in all brain cells except excitatory neurons. PS1 forms the catalytic subunit of γ-secretase, which cleaves Aβ from its precursor (APP), and the mutant version is one cause of familial Alzheimer’s disease. In all CNS cells, the mice expressed the Swedish variant of APP, which is especially susceptible to amyloidogenic cleavage. In theory, this meant that all brain cells except excitatory neurons could produce Aβ. Control mice expressed both mutant genes in all CNS cells. 

Since they lacked mutant PS1 in the excitatory neurons, the neuronal PS1∆E9 knockouts made about 80 percent less of the catalytic subunit in the brain at seven months of age than did control mice. At this young age, knockouts had about 10 times fewer plaques than controls in the cortex and seven times fewer in the hippocampus. That suggested that while non-excitatory neurons do produce some Aβ, excitatory neurons contribute substantially to the aggregating Aβ pool early on. Surprisingly, by the time the mice were 10 to 12 months old, the knockouts had nearly caught up; their cortex and hippocampus contained only about 1.5-fold fewer plaques compared to controls. 

Aβ plaques in mouse brains. Courtesy, with permission, of Veeraraghavalu et al., The Journal of Neuroscience, 2014.

The authors are unsure why the plaque levels rose more quickly in the neuronal PS1∆E9 knockouts, but they offered two possibilities. First, cells other than excitatory neurons, such as astrocytes and microglia, become the primary source of Aβ later in life. Second, cytokines or growth factors secreted by these mutation-carrying, non-excitatory cells could encourage Aβ to aggregate as mice age. To demonstrate that these cells were making their own Aβ, the researchers cultured astrocytes, microglia, and neural progenitor cells from the mice. Western blots revealed Aβ secreted from all three. In support of the second hypothesis, the group previously documented altered levels of certain secreted factors between microglia isolated from adult mice that produced wild-type or mutant human PS1 (see Choi et al., 2008). Veeraraghavalu also raised the possibility that when glial cells proliferate with age or stress, they could secrete more Aβ.

Given previous studies that reported human microglia and astrocytes make their own Aβ (see Kamenetz  et al., 2003), it would make sense that these cells contribute to the aggregating pool of Aβ in humans, as they do in mice, Veeraraghavalu said. He said that in future, the group may knock out PS1∆E9 in different brain cell subtypes of mice to assess their relative contribution to the Aβ deposition and other markers of pathology, such as atrophy and glial proliferation. Sisodia pointed out that the results suggest researchers should differentiate patient-derived induced pluripotent stem cells into diverse brain cell types for more complete disease models.

“[The authors] make a great point that the AD field generally ignores Aβ generation in astrocytes and microglia,” wrote John Cirrito, Washington University in St. Louis, to Alzforum in an email. “This paper nicely highlights the need to consider Aβ generation from non-neuronal cells.” He pointed out that it will be important to determine whether production methods differ between cells types. However, he cautioned that the Cre-loxP system is imperfect, so it could still allow excitatory neurons to produce some small amount of PSEN1 in this experiment. Further, even though mutant PS1∆E9 is missing from those neurons, they nonetheless produce the endogenous protein, which could help produce Aβ. He added that it is still unclear how the proportion of Aβ produced by these other cell types compares to that of excitatory neurons.—Gwyneth Dickey Zakaib


Make a Comment

To make a comment you must login or register.

Comments on this content

No Available Comments


Paper Citations

  1. . The contribution of activated astrocytes to Aβ production: implications for Alzheimer's disease pathogenesis. J Neuroinflammation. 2011;8:150. PubMed.
  2. . Transforming growth factor-beta 1 potentiates amyloid-beta generation in astrocytes and in transgenic mice. J Biol Chem. 2003 May 16;278(20):18408-18. PubMed.
  3. . APP processing and synaptic function. Neuron. 2003 Mar 27;37(6):925-37. PubMed.
  4. . Synaptic activity regulates interstitial fluid amyloid-beta levels in vivo. Neuron. 2005 Dec 22;48(6):913-22. PubMed.
  5. . Non-cell-autonomous effects of presenilin 1 variants on enrichment-mediated hippocampal progenitor cell proliferation and differentiation. Neuron. 2008 Aug 28;59(4):568-80. PubMed.

Further Reading


  1. . Brain endothelial cells produce amyloid {beta} from amyloid precursor protein 770 and preferentially secrete the O-glycosylated form. J Biol Chem. 2010 Dec 17;285(51):40097-103. PubMed.

Primary Papers

  1. . Age-dependent, non-cell-autonomous deposition of amyloid from synthesis of β-amyloid by cells other than excitatory neurons. J Neurosci. 2014 Mar 5;34(10):3668-73. PubMed.