Repository logo

Exploring the role of programmed axon death genes SARM1 and NMNAT2 in human disease



Change log


Ademi, Mirlinda 


Programmed axon death (PAD) is an evolutionary-conserved path of axon destruction initiated by injury and disease. Two proteins are known to profoundly influence axonal health: the pro-death SARM1 and its upstream regulator, the pro-survival NMNAT2. Notably, complete removal of SARM1 has been shown to rescue injured axons permanently in one case and delay axon loss in many others, alleviating animal models of Parkinson’s disease, traumatic brain injury, glaucoma, and other disorders. While mechanistic studies of these two proteins have been the focus of recent years, human genetics of PAD had been studied to a much lesser extent, despite growing interest and investment from several pharmaceutical companies to target SARM1 therapeutically. Consequently, the aim of this thesis was to address this knowledge gap and investigate how human gene variants in NMNAT2 and SARM1 influence axon vulnerability. Chapter 3 of this thesis revolves around the role of naturally-occurring SARM1 alleles in the human population many of which encode SARM1 loss-of-function (LoF) coding variants. They influence SARM1 enzymatic activity and axon survival functions, likely to reflect protective effects in disease. Importantly, two variants tested confer dominant LoF effects suggesting even stronger effects on disease risk. In chapter 4, biallelic NMNAT2 LoF alleles are described in two siblings affected by a severe polyneuropathy. In addition to the newly-identified patient variants, a small set of artificial and natural NMNAT2 missense variants is tested for pathogenic LoF of the protein. Thirdly, chapter 5 revolves around SARM1 missense variants found in the genomic cohorts UK Biobank and the 100,000 Genomes Project. Furthermore, it is investigated whether SARM1 LoF is underrepresented in disease cohorts, implying a protection against axonal insult. This is done by comparing frequencies of predicted and confirmed SARM1 LoF alleles between disease and control groups. Finally, I discuss how knowledge of these and other SARM1/NMNAT2 function-altering alleles will be useful for further establishing links between SARM1-dependent PAD and human pathologies that would be good candidates for the use of SARM1-blocking drugs in the future.





Coleman, Michael


axon vulnerability, axotomy, databases, gain-of-function, loss-of-function, missense mutation, neurodegeneration, neuroprotection, NMNAT2, programmed axon death, SARM1, Wallerian degeneration


Doctor of Philosophy (PhD)

Awarding Institution

University of Cambridge
MRC (2118602)
DTP studentship from the UK Medical Research Council Pinsent Darwin Trust