Astrocytes, one of the major glial cell types in the central nervous system, are required for normal development and functioning of neurons. In response to injury and disease, astrocytes change phenotype from their normal physiological role to one of many “reactive” states. Some of these states provide additional trophic support to neurons under stress, while others can lose support functions or become neurotoxic. We recently identified a reactive astrocyte subtype that forms around neuron cell bodies after acute injuries and in chronic neurodegeneration in both rodents and humans, and during normal aging. These astrocytes show dramatic functional deficits compared to physiological nonreactive astrocytes (e.g., loss of synapse formation and glutamate uptake).

Due to their ability to drive death of neurons, we predicted this substate to be neurotoxic. While the exact neurotoxin identity was still unknown, we showed that blocking formation of neurotoxic reactive astrocytes using genetic and pharmacological tools preserved neuron numbers after acute axonal injury, and notably in a mouse model of glaucoma. However, these previous methods lacked the fidelity to identify either a specific astrocyte-derived neurotoxin or the specific pathways mediating neuron death.

Using biochemical approaches, described below in preliminary data for this proposal, we have now isolated the astrocyte-derived toxic factor. We found that saturated lipids, in particular long-chain free fatty acids (FFAs), drive astrocyte-mediated neurotoxicity. Eliminating the formation of long-chain FFAs by astrocyte cell type–specific knockout of the saturated lipid synthesis enzyme ELOVL1 blocks this toxicity.

We have produced a conditional knockout mouse in which reactive astrocytes are unable to produce neurotoxic lipids. These mice show amelioration of RGC loss following acute axonal injury. Our data also suggest that the PERK-ATF3 apoptosis pathway is a key driver of neuron cell death following secretion of toxic FFAs by this substate of reactive astrocytes.

We hypothesize that controlling neurotoxic astrocyte reactivity can maintain neuron cell numbers and preserve neuron function and visual acuity. Here we address three main questions:

• Does the PERK-ATF3 pathway in neurons predominantly contribute to astrocyte-derived FFA-mediated cell death?
• Does block of astrocyte-derived neurotoxic lipids preserve neuronal numbers and function in mouse models of neurodegenerative disease?
• What genes/pathways in neurons impart susceptibility to astrocyte-mediated cell death? (ongoing and future directions)

This project will continue to involve collaborations across NYU Langone, including with the Rodent Behavior Laboratory and Genome Technology Center. Our long-term goal is to define pathways in neurotoxic reactive astrocytes and susceptible neuron populations that will provide a blueprint to enable development of effective therapeutic strategies for a range of neurodegenerative diseases like amyotrophic lateral sclerosis (ALS), Parkinson’s disease, Alzheimer’s disease, glaucoma, and many others.

Contact Us

For more information about this research, please contact principal investigator Shane A. Liddelow, PhD, at shane.liddelow@nyulangone.org.