DBA/2J (D2) mice develop a form of chronic glaucoma with iris abnormalities, elevated intraocular pressure (IOP) and retinal ganglion cell (RGCs) loss.1 In order to better understand the early molecular events responsible for IOP-dependent RGC loss, Williams et al. used fluorescence-activated cell sorting to isolate RGCs from D2 mice with or without elevated IOP.2 The experimental group used nine-month-old D2 mice with elevated IOP, but before the onset of frank RGC loss. Controls animals included D2 mice at four months, before the onset of elevated IOP, and nine-month-old congenic D2 mice with a wildtype Gpnmb allele (D2-Gpnmb+) that do not develop elevated IOP. They then used RNA sequencing to analyze and compare the transcriptomes of the various RGC populations. The results showed that genes associated with mitochondrial stress responses and oxidative phosphorylation were differentially expressed, suggesting that mitochondrial dysfunction could be an early pathophysiologic event in response to elevated IOP. The authors went on to measure the level of the mitochondrial metabolite, nicotinamide adenine dinucleotide (NAD+), in D2 and control D2-Gpnmb+ mice and, in contrast to the above, found that whole-retinal NAD+ levels declined with age, independent of the presence of elevated IOP.
In other tissues, NAD+ levels are known to decrease with age and there is a rich body of literature on the role of NAD+ (and a class of NAD+-dependent enzymes known as sirtuins) in aging and neurodegeneration.3 Moreover, it is known that NAD+ depletion (and reduced sirtuin activity) can affect mitochondrial function and turnover and that genes (e.g., TBK1 and optineurin) that affect mitochondrial turnover can cause rare familial forms of glaucoma. Finally, NAD+ metabolites play a critical role in the SARM1-dependent genetic program responsible for axon degeneration after injury4 and axon injury at the optic nerve head is a key causal event in D2 glaucoma.5 Given this constellation of findings, the authors postulated that reduced NAD+ levels in aged RGCs might be necessary for elevated IOP to cause mitochondrial dysfunction, leading to axon degeneration and cell death.
To test this hypothesis, the authors turned to a biosynthetic precursor of NAD+ known as nicotinamide (NAM) or vitamin B3, which when given exogenously, can raise the levels of NAD+. Of note, vitamin B3 can also refer to a structurally-related precursor called niacin (nicotinic acid). Starting at either six months (before the onset of increased IOP) or at 9 months (after the onset of increased IOP), 550 or 2000 mg/kg/day of NAM was orally administered to D2 mice and the expected increase in retinal NAD+ concentrations was verified. They then examined eyes at 12 months, when the neurodegeneration is typically manifest, and showed a dose-dependent neuroprotection that was similar regardless of when the therapy was initiated. At the molecular and subcellular level, NAM supplementation reversed the gene expression changes normally seen with aging and glaucoma and improved mitochondrial morphology. At the cellular and tissue level, NAM led to preservation of RGC cell bodies, synaptic marker staining, electrophysiological activity and axons connected to the lateral geniculate nucleus/superior colliculus. The high (but not low) dose had an additional effect of blunting the IOP increase in D2 mice, confounding a direct effect of high dose NAM on RGCs. Finally, the authors were able to achieve near-complete protection by combining the low dose of NAM with overexpression of nicotinamide/nicotinic acid mononucleotide adenylyltransferase 1 (NMNAT1), an enzyme that converts NAM to NAD+.
Although it can be difficult to compare mouse and human dosing, the equivalent NAM dose in humans would far exceed the lethal concentration
Although the authors tested other models of RGC injury, it will be important to test this strategy in other animal models of glaucoma. Indeed, Wallerian degeneration slow, WLDS (a chimeric protein that includes NMNAT1) acts in a similar manner but was unable to provide sustained RGC protection in the rat laser model of glaucoma.6 Moreover, it may be the case that the pathway is more complicated as NMNAT1 can inhibit SARM1 independent of its ability to increase NAD+ levels.7 Finally, although it can be difficult to compare mouse and human dosing, the equivalent NAM dose in humans would far exceed the lethal concentration. Thus, it may be that other ways of modulating the pathway, either alone or in combination, may be more viable therapeutic options. Nonetheless, these results serve as the first proof-of-principle that modulation of NAD+-dependent signaling can protect RGC structure and function in the setting of elevated IOP.