Structure-function analysis of ceTIR-1/hSARM1 explains the lack of Wallerian axonal degeneration in C. elegans

Wallerian axonal degeneration (WD) does not occur in the nematode C. elegans, in contrast to other model animals. However, WD depends on the NADase activity of SARM1, a protein that is also expressed in C. elegans (ceSARM/ceTIR-1). We hypothesized that differences in SARM between species might exist and account for the divergence in WD. We first show that expression of the human (h)SARM1, but not ceTIR-1, in C. elegans neurons is sufficient to confer axon degeneration after nerve injury. Next, we determined the cryoelectron microscopy structure of ceTIR-1 and found that, unlike hSARM1, which exists as an auto-inhibited ring octamer, ceTIR-1 forms a readily active 9-mer. Enzymatically, the NADase activity of ceTIR-1 is substantially weaker (10-fold higher Km) than that of hSARM1, and even when fully active, it falls short of consuming all cellular NAD(+). Our experiments provide insight into the molecular mechanisms and evolution of SARM orthologs and WD across species.