Abstract
Neurite degeneration (ND) precedes cell death in many neurodegenerative diseases. However, it remains unclear how this compartmentalized cell death process is orchestrated in the central nervous system (CNS). The establishment of a CNS axotomy model (using modified 3D LUHMES cultures) allowed us to study metabolic control of ND in human midbrain-derived neurons without the use of toxicants or other direct disturbance of cellular metabolism. Axotomy lead to a loss of the NAD+ synthesis enzyme NMNAT2 within 2 h and a depletion of NAD+ within 4-6 h. This process appeared specific, as isolated neurites maintained ATP levels and a coupled mitochondrial respiration for at least 6 h. In the peripheral nervous system (PNS) many studies observed that NAD+ metabolism, in particular by the NADase SARM1, plays a major role in the ND occurring after axotomy. Since neither ferroptosis nor necroptosis, nor caspase-dependent apoptosis seemed to be involved in neurite loss, we investigated SARM1 as potential executioner (or controller). Knock-down or expression of a dominant-negative isoform of SARM1 indeed drastically delayed ND. Various modifications of NAD+ metabolism known to modulate SARM1 activity showed the corresponding effects on ND. Moreover, supplementation with NAD+ attenuated ND. As a third approach to investigate the role of altered NAD+ metabolism, we made use of the WLD(s) protein, which has been found in a mutant mouse to inhibit Wallerian degeneration of axons. This protein, which has a stable NMNAT activity, and thus can buffer the loss of NMNAT2, protected the neurites by stabilizing neurite NAD+ levels. Thus CNS-type ND was tightly linked to neurite metabolism in multiple experimental setups. Based on this knowledge, several new strategies for treating neurodegenerative diseases can be envisaged.