Motor axon loss is a cardinal symptom of amyotrophic lateral sclerosis (ALS). Axon loss can be driven
by a genetically encoded program in which the axon survival factors NMNAT2 and STMN2 inhibit the activity of
the axon destruction factor SARM1. Recent data suggest that this program of axon self-destruction may
contribute to pathology in ALS. First, aggregation of TDP-43, a hallmark of most ALS cases, results in the
selective loss of mRNA encoding functional STMN2, a key axon survival factor. Second, loss of SARM1
suppresses some neurodegenerative phenotypes in a mouse ALS model that expresses pathogenic human
TDP-43. Here we investigate the contribution of this axon degeneration pathway to ALS. We have defined the
mechanism of action of SARM1, demonstrating that it is the founding member of a new class of NAD-cleaving
enzymes. SARM1 enzyme activity is normally held in check via an autoinhibitory domain. Injury- or disease-
induced loss of NMNAT2 and STMN2 disinhibits SARM1, leading to rapid NAD+ depletion, metabolic
catastrophe, and axon fragmentation. Our structure-function studies of the SARM1 protein have identified
mutations with a range of consequences, from constitutively active variants that promote cell death and axon
loss, to dominant negative variants that are neuroprotective. These findings imply that human variants may exist
that either promote or protect against neurodegeneration, and that understanding the phenotypic consequences
of genetic variation requires functional studies of enzyme activity. In support of this hypothesis, we have identified
several rare SARM1 variants in ALS patients, but not in controls, that have constitutive NADase activity and
promote neuron death and axon loss. These variants also cause motor dysfunction and paralysis when
expressed in the mouse CNS, suggesting that activating SARM1 mutations may contribute to ALS pathogenesis.
Here we propose to define the function of SARM1 variants from ALS patients, controls, and the general
population. These studies will allow us to categorize SARM1 variants as putatively pro-degenerative,
neuroprotective, or neutral. In parallel, we will dissect the contribution of variation in components of the
programmed axon destruction pathway to ALS phenotypes, alone and in combination with known ALS genetic
risk-factors, in motor neurons differentiated from human induced pluripotent stem cells (iPSCs). Finally, we will
investigate neurodegeneration in a mouse knock-in model carrying a Sarm1 allele equivalent to a pro-
degenerative allele found in ALS patients, alone and in combination with a SOD1 model, based on a specific
patient genotype that we identified. We will attempt to suppress ALS phenotypes with SARM1 inhibition via a
proven gene therapy approach and with experimental small molecule inhibitors. Results of these studies will
establish the relationship between the SARM1-mediated axon destruction program and ALS, and build the
foundation to develop axoprotective therapeutics to treat this devastating disease.