Axon loss following injury and disease (Wallerian degeneration) is a regulated process of axon self-destruction (Coleman, 2005; Gerdts et al., 2016; Wang et al., 2012). The discovery of the Wallerian degeneration slow (Wlds) protein, which dramatically delays axon degeneration after injury, raised hopes that blocking this process would be useful in the treatment of neurological disorders (Conforti et al., 2014; Mack et al., 2001). The Wlds protein blocks axon degeneration by mislocalizing the nuclear nicotinamide adenine dinucleotide (NAD⁺) biosynthetic enzyme NMNAT1 into axons, thereby substituting for the loss of the labile axon maintenance factor NMNAT2 (Araki et al., 2004; Babetto et al., 2010; Gilley et al., 2010; Sasaki et al., 2010). The mechanism by which NMNAT enzymes block axon degeneration is unknown. These enzymes synthesize NAD⁺ from nicotinamide mononucleotide (NMN) and ATP, so two candidate mechanisms for how loss of NMNAT2 triggers axon degeneration have emerged: (1) the loss of NAD⁺ or (2) the accumulation of NMN. The striking axonal protection provided by neuronal expression of Escherichia coli NMN deamidase, which reduces NMN levels by converting it to nicotinic acid mononucleotide (NaMN), supports the model that NMN accumulation triggers axon degeneration (Di Stefano et al., 2015). Here we investigate NAD⁺ metabolism in healthy and injured axons through the measurement of steady-state metabolite levels and via the analysis of NAD⁺ metabolite synthesis and consumption (i.e. flux analysis). Surprisingly, we find that neither NAD⁺ loss nor NMN accumulation trigger axon degeneration. Instead, both NMNAT1 and NMN deamidase prevent axon degeneration via blocking the injury-induced NAD⁺ consumption that occurs following activation of the axodestructive molecule SARM1 (Gerdts et al., 2013, 2015; Osterloh et al., 2012).

To investigate whether NAD⁺ loss or NMN accumulation triggers axon degeneration, we assessed injury-induced axon degeneration and NAD⁺ metabolite levels in cultured DRG neurons in which the NAD⁺ biosynthetic pathway was perturbed at various steps (Figure 1a). To raise intracellular NMN levels we expressed the NMN biosynthetic enzyme NAMPT or applied nicotinamide riboside (NR) to neurons expressing NRK1, a nicotinamide riboside kinase that converts NR to NMN. To reduce NMN levels we treated neurons with the NAMPT inhibitor FK866 or expressed E. coli NMN deamidase (Di Stefano et al., 2015). We also tested the potent axoprotective cytoplasmic version of NMNAT1 (cytNMNAT1), which appears to substitute in axons for the short-lived NMNAT2 (Gilley et al., 2010, Sasaki et al., 2009a). The expression of these enzymes in axons was confirmed by Western blotting (Figure 1—figure supplement 1). Using our automated image analysis axon degeneration assay, and consistent with previous reports, all of these manipulations delayed axon degeneration although to dramatically different extents (Figure 1b,c and Figure 1—source data 1; Di Stefano et al., 2015; Sasaki et al., 2006, 2009b). FK866 and NR treatment provided only modest protection, delaying axon degeneration by ~6 to 9 hr. Neurons expressing NRK1 and treated with NR or expressing NAMPT showed strong axon protection for 24 to 48 hr. The strongest axonal protection was found in neurons expressing cytNMNAT1 or NMN deamidase, which both block axon degeneration for at least 3 days after axotomy.

Figure 1 with 1 supplement see all

Download asset Open asset

Figure 1—source data 1

Axonal degeneration index at 0, 9, 24, 48, and 72 hr post axotomy in the presence of various NAD⁺ biosynthesis manipulations (Figure 1c).

https://doi.org/10.7554/eLife.19749.003

Download elife-19749-fig1-data1-v2.xlsx

To correlate axon degeneration with levels of NAD⁺ metabolites, we measured baseline NMN, NaMN, NAD⁺, and nicotinic acid adenine dinucleotide (NaAD) from DRG neurons using LC-MS/MS before axonal transection. As previously described, FK866 led to a slow decline in both cellular NAD⁺ and NMN before the axons begin to fragment (Figure 2a and Figure 2—source data 1; Di Stefano et al., 2015). In contrast, NAMPT expression or pre-incubation with NR in the presence of NRK1 significantly increased both NMN and NAD⁺ levels. NMN deamidase significantly increased the levels of NaMN and NaAD while dramatically reducing NMN and NAD⁺ levels in DRG neurons (Figure 2b,c and Figure 2—source data 1). Despite reducing NAD⁺ levels to 11 ± 7% of control, NMN deamidase-expressing neurons showed no signs of cell death or axon degeneration, but instead displayed potent axonal protection (Figure 3c and Figure 3—source data 1) as previously reported (Di Stefano et al., 2015). The protection afforded by NMN deamidase was equivalent to that observed in neurons expressing cytNMNAT1 (Figure 3c and Figure 3—source data 1), however baseline levels of NMN and NAD⁺ were normal in neurons expressing cytNMNAT1 (Figure 2b,c and Figure 2—source data 1). Similarly, SARM1-deficient neurons had baseline levels of NMN, NAD⁺, NaMN, and NaAD that were equivalent to those of wildtype neurons (Figure 2—figure supplement 1 and Figure 2—source data 2, Gerdts et al., 2015). From these studies, it is clear that robust axonal protection can be observed in neurons that maintain low levels of NMN and NAD⁺ (NMN deamidase), normal levels of NMN and NAD⁺ (cytNMNAT1 or SARM1 knockout), or high levels of NMN and NAD⁺ (NAMPT or NRK1 + NR).

Figure 2 with 2 supplements see all

Download asset Open asset

Figure 2—source data 1

Steady state cellular NMN, NaMN, NAD⁺, and NaAD levels in the absence or presence of various NAD⁺ biosynthesis manipulations (Figure 2a (control or FK866) and Figure 2b,c (control or various NAD⁺ enzyme expressing neurons)).

Steady state axonal NAD⁺ (Figure 2d) and NMN (Figure 2e) levels at 0, 1, 2, 3, 6 hr post axotomy.

https://doi.org/10.7554/eLife.19749.006

Download elife-19749-fig2-data1-v2.xlsx

Figure 2—source data 2

Steady state NMN, NaMN, NAD⁺, and NaAD levels in wild-type or SARM1 KO DRG neurons (Figure 2—figure supplement 1).

https://doi.org/10.7554/eLife.19749.007

Download elife-19749-fig2-data2-v2.xlsx

Figure 2—source data 3

Steady state axonal total adenylate levels (ATP + ADP + AMP) and NMN normalized by total adenylate levels at 0, 1, 2, 3, 6 hr post axotomy (Figure 2—figure supplement 2).

https://doi.org/10.7554/eLife.19749.008

Download elife-19749-fig2-data3-v2.xlsx

Figure 3

Download asset Open asset

Figure 3—source data 1

Steady state cellular NMN, NaMN, NAD⁺, NaAD levels in control and NMN deamidase expressing neurons with or without NMN synthetase or NMN synthetase + NR (Figure 3a,b).

Axonal degeneration index at 0, 9, 24, 48, 72 hr post axotomy of control and NMN deamidase expressing neurons with or without NMN synthetase or NMN synthetase + NR (Figure 3c).

https://doi.org/10.7554/eLife.19749.012

Download elife-19749-fig3-data1-v2.xlsx

The above results demonstrate that changes in baseline levels of NMN and NAD⁺ are insufficient to trigger axon degeneration, so we next assessed changes in axonal NAD⁺ metabolites after axotomy (Figure 2d,e and Figure 2—source data 1). We plated neurons in such a manner that we could harvest axons separate from cell bodies and thereafter use LC-MS/MS to measure axon-specific metabolite levels. These studies showed a dramatic decline in NAD⁺ levels after injury in control axons, consistent with previous reports (Di Stefano et al., 2015; Gerdts et al., 2015; Wang et al., 2005). We also found a ~2-fold increase in the levels of NMN within the first two hours after injury. NMN levels declined back to those observed in uninjured axons by six hours post-axotomy. A previous report showed that sciatic nerve transection causes a continuous increase in nerve NMN concentration when it is normalized to the nerve adenylate pool (Di Stefano et al., 2015). We confirm this finding using purified axon preparations, however, we demonstrate that this is due to a rapid decline in the adenylate pool rather than an increase in NMN levels (Figure 2—figure supplement 2 and Figure 2—source data 3). We next assessed NAD⁺ metabolites in axons that are robustly protected from axotomy-induced degeneration to determine whether these treatments are correlated with a specific metabolite profile. In cells expressing the strongly axoprotective cytNMNAT1, axonal NAD⁺ and NMN were comparable to baseline levels in uninjured axons of control neurons and these metabolite levels did not change after axotomy. Expression of NMN deamidase reduced axonal NAD⁺ and NMN compared with control axons, and these metabolite levels did not change after axotomy. In contrast, axons from neurons expressing NAMPT had high levels of both NAD⁺ and NMN at baseline and at six hours after axotomy. In axons from neurons expressing NRK1 and treated with NR there was a dramatic increase of NAD⁺ and NMN levels at baseline. After injury, these axons showed a robust decrease in NAD⁺ but further increases in NMN. Indeed, NMN levels in axons from these neurons were 16-fold higher than in control axons (0.16 ± 0.04 µM vs. 0.01 ± 0.001 µM, Figure 2e and Figure 2—source data 1), yet these axons were strongly protected from degeneration (>48 hr). These results demonstrate that even within injured axons, elevated levels of NMN do not promote axon degeneration but rather can be compatible with robust axonal protection.

While these data are inconsistent with the hypothesis that elevated NMN induces axon degeneration, the finding that NMN deamidase both potently protects axons and leads to low levels of NMN does support this hypothesis (Di Stefano et al., 2015). Hence, we sought to determine whether maintaining low NMN levels is necessary for the axoprotective function of NMN deamidase. We approached this problem by attempting to maintain normal NMN levels in neurons expressing E. coli NMN deamidase and then assessing whether axons now degenerate after axotomy. To this end, we utilized an enzyme from Francisella tularensis called NMN synthetase that catalyzes the conversion of NaMN to NMN (i.e., the reverse of the reaction catalyzed by E. coli NMN deamidase) (Figure 1a, Figure 1—figure supplement 1). When we expressed NMN synthetase alone in neurons no changes in any of the measured metabolites were observed (Figure 3a,b and Figure 3—source data 1), and injured axons from these neurons degenerated normally (Figure 3c and Figure 3—source data 1). However, when NMN synthetase was co-expressed along with NMN deamidase, the normal NMN deamidase-catalyzed reductions in NMN and NAD⁺ as well as the increases in NaMN and NaAD are largely abolished (Figure 3a,b and Figure 3—source data 1). Furthermore, the addition of the NMN precursor NR to neurons co-expressing F. tularensis NMN synthetase and E. coli NMN deamidase raised axonal NMN well above control levels (Figure 3a and Figure 3—source data 1). Despite this elevation in NMN levels under these conditions, NMN deamidase continued to provide robust axon protection (Figure 3c and Figure 3—source data 1). In addition, we also found that protection of injured axons did not correlate with levels of NAD⁺, NaMN, or NaAD. Hence, neither a reduction in NMN nor the accumulation of NaMN/NaAD is necessary for NMN deamidase mediated axonal protection. These findings are inconsistent with the hypothesis that the axonal NMN level is a driver of axon degeneration.

Following axotomy, axonal NAD⁺ levels decline. This decline could be due to loss of NAD⁺ biosynthesis via degradation of the labile axonal NMNAT isoform NMNAT2 (Gilley et al., 2010), and/or to an increase in NAD ⁺ degradation due to activation of the pro-degenerative molecule SARM1 (Gerdts et al., 2015). To interrogate the processes of NAD⁺ synthesis and degradation independently, we developed an axonal NAD⁺ flux assay. The nicotinamide (Nam) in the DRG culture medium was largely replaced with deuterium-labeled nicotinamide (D4-Nam: 2,4,5,6-deutrium Nam, 300 µM) to achieve a 10-fold excess over the original Nam concentration in culture medium). The addition of up to 1 mM of this NAD⁺ precursor did not alter the time course of axon degeneration (Figure 4—figure supplement 1 and Figure 4—source data 2) and allowed us to identify its labeled-NAD⁺ derivative using LC-MS/MS. We added labeled Nam at time zero, and followed its conversion by quantifying non-labeled (light) and labeled (heavy) NAD⁺ in axons using LC-MS/MS at various times after addition. Conversion of the NAD⁺ pool from light to heavy was observed without delay, demonstrating that D4-Nam replaces Nam for NAD⁺ synthesis (Figure 4—figure supplement 2). The increase in heavy NAD⁺ following D4-Nam addition gives a measure of the NAD⁺ biosynthetic rate, while the loss of light NAD⁺ reflects the NAD⁺ consumption rate. In uninjured axons steady-state NAD⁺ (summation of heavy and light NAD⁺) was constant due to ongoing and balanced NAD⁺ synthesis (6.9 ± 1.4% per hour) and NAD⁺ consumption (8.2 ± 3.1% per hour) (Figure 4a). Next we compared basal NAD⁺ flux in uninjured axons under various axoprotective conditions. NAD⁺ biosynthesis enzymes were expressed and axonal NAD⁺ consumption rates were measured at 4 hr post D4-Nam addition. Absolute NAD⁺ consumption rates were normalized to ATP levels, which are constant in uninjured axons. We found that axon NAD⁺ consumption in uninjured axons was similar to wild type levels in SARM1 KO neurons as well as neurons overexpressing cytNMNAT1. NAD⁺ consumption was lower in axons from NMN deamidase-expressing neurons and higher in axons from NAMPT-expressing neurons (Figure 4—figure supplement 3 and and Figure 4—source data 2). Since NAD⁺ levels are at equilibrium in these uninjured axons, these changes in NAD⁺ consumption are balanced by equivalent changes in NAD⁺ production. Hence, basal levels of NAD⁺ flux are well correlated with steady-state NAD⁺ levels (lower in NMN deamidase-expressing neurons and higher in NAMPT-expressing neurons) but not with the axonal protection phenotypes (Figure 1c and Figure 1—source data 1, Figure 2b,c and Figure 2—source data 1, Figure 4—figure supplement 3). These results are expected since NMN deamidase inhibits NAD⁺ synthesis by reducing the levels of NMN whereas NAMPT promotes NAD⁺ synthesis by increasing the levels of NMN. NMN is the rate limiting metabolite in the NAD ⁺biosynthetic pathway (Revollo et al., 2004).

Figure 4 with 4 supplements see all

Download asset Open asset

Figure 4—source data 1

Relative axonal NAD⁺ consumption (Figure 4b) and synthesis (Figure 4c) rates in axons expressing various NAD⁺ biosynthesis enzymes with or without NR at 0, 4 hr (consumption), or 2 hr (synthesis) after axotomy.

https://doi.org/10.7554/eLife.19749.014

Download elife-19749-fig4-data1-v2.xlsx

Figure 4—source data 2

Axonal degeneration index at 0, 9 and 24 hr post axotomy in the presence or absence of 1 mM Nicotinamide (Figure 4—figure supplement 1).

https://doi.org/10.7554/eLife.19749.015

Download elife-19749-fig4-data2-v2.xlsx

Figure 4—source data 3

Absolute Axonal NAD⁺ consumption rates in axons expressing various NAD⁺ biosynthesis enzymes (Figure 4—figure supplement 3).

https://doi.org/10.7554/eLife.19749.016

Download elife-19749-fig4-data3-v2.xlsx

To analyze NAD⁺ flux following injury, D4-Nam was added at the time of axotomy. In axotomized axons, steady-state NAD⁺ levels declined as previously reported. We now show that this is due to a large increase in the NAD⁺ consumption rate (from 8.5 ± 3.8% per hour prior to injury to 21.7 ± 1.6% per hour after injury: Figure 4b and Figure 4—source data 1, Figure 4—figure supplement 4). Surprisingly, in the first two hours after injury, there was also a modest increase in NAD⁺ biosynthesis (control: 6.6 ± 3.7% per hour vs. post-axotomy: 14.0 ± 6.9% per hour; Figure 4a, c and Figure 4—source data 1). This is likely due to the activation of NAMPT, the rate-limiting enzyme in the NAD⁺ biosynthetic pathway, which is negatively regulated by NAD⁺. The decline in NAD⁺ relieves this feedback inhibition and results in increased NAMPT activity (Elliott and Rechsteiner, 1982), which is also the likely explanation for the transient increase in axonal NMN after injury (Figure 2d and Figure 2—source data 1). Although the NAD⁺ synthesis rate was increased during the first two hours after axotomy (Figure 4c and Figure 4—source data 1), total NAD⁺ declined due to the larger increase in the NAD⁺ consumption rate (Figure 4a). After two hours post-axotomy, heavy NAD⁺ declined at the same rate as light NAD⁺ (Figure 4a), indicating a complete loss of NAD⁺ biosynthesis, likely due to loss of the labile NMNAT2 in the severed axon (Gilley et al., 2010). Hence, flux analysis demonstrates that injury-induced NAD⁺ depletion is due to a major increase in NAD⁺ consumption that begins shortly after the axon damage and before the loss of axonal NAD⁺ synthesis (Figure 4c and Figure 4—source data 1). This surprising finding highlights the utility of flux analysis to provide mechanistic insights that were not appreciated in prior studies of steady-state NAD⁺ levels.

We hypothesized that SARM1 may be required for this injury-dependent increase in NAD⁺ consumption because we previously demonstrated that SARM1 is required for the steady-state loss of NAD⁺ following injury in vivo and that SARM1 activation in primary cultured neurons leads to NAD⁺ degradation (Gerdts et al., 2015). To test this hypothesis, we applied NAD⁺ flux analysis to SARM1-deficient neurons. The NAD⁺ consumption rate in SARM1-deficient axons was comparable to control axons prior to injury (Figure 4a, Figure 4—figure supplement 3 and Figure 4—source data 2). However in contrast to the dramatic increase in NAD⁺ consumption in wildtype neurons after injury, there was no significant change in NAD⁺ consumption rate after injury in SARM1 mutant axons (uninjured: 6.3 ± 2.4% per hour vs. SARM1 mutant: 7.9 ± 3.7% per hour after axotomy). Hence, SARM1 is required for the injury-induced increase in NAD⁺ consumption (Figure 4a,b and Figure 4—source data 1, Figure 4—figure supplement 4). This supports and extends our previous finding that SARM1 is responsible for initiating events that lead to the depletion of axonal NAD⁺ following axotomy (Gerdts et al., 2015).

Since SARM1 is required for axon degeneration and axonal NAD⁺ depletion, we next assessed whether other manipulations that affect axon survival do so by blocking injury-induced SARM1-dependent axonal NAD⁺ depletion (Figure 4b and Figure 4—source data 1). The overexpression of NAMPT, which leads to modest axon protection, did not block injury-induced SARM1-dependent NAD⁺ consumption. In contrast, expression of either cytNMNAT1 or NMN deamidase completely eliminated SARM1-dependent NAD⁺ consumption after axotomy (Figure 4b and Figure 4—source data 1). This is a striking result that likely explains the profound axoprotective effects of overexpressing either NMNAT or NMN deamidase —both block the ability of SARM1 to trigger NAD⁺ degradation following axotomy. Loss of SARM1 or gain of NMNAT or NMN deamidase all lead to profound and long-lasting axonal protection following injury, and all three block NAD⁺ degradation following injury. In contrast, treatments that provide modest axon protection (e.g. NAMPT) likely do so by transiently countering injury-induced NAD⁺ degradation through their ability to elevate intracellular NAD⁺ and thus maintain axonal NAD⁺ levels above a critical threshold for a longer period after injury.

We have shown that a reduction in NMN is not required to block axon degeneration (Figure 1–3 and Figure 3—source data 1), however the role of NMN in SARM1 activation has not been investigated. Using injury-activated NAD⁺ consumption as an assay of SARM1 pathway activity, we tested whether maintaining low NMN levels is necessary for the blockade of SARM1 pathway activation by cytNMNAT1 or NMN deamidase. Intracellular NMN was raised by the addition of NR to cytNMNAT1 expressing neurons (5.3 ± 2.9 fold) or NMN deamidase and NMN synthetase expressing neurons (3.7 ± 3.1-fold) and SARM1-dependent NAD⁺ consumption was measured. In both cases, SARM1-triggered axonal NAD⁺ degradation after injury is fully inhibited by cytNMNAT1 or NMN deamidase despite high levels of NMN (Figure 4b and Figure 4—source data 1). Hence, a reduction in NMN is not necessary for blockade of SARM1 pathway activation by NMNAT1 or NMN deamidase.

These findings support a model in which axotomy activates SARM1, which then triggers events that lead to a catastrophic loss of NAD⁺ followed by axon degeneration. Here we identify two distinct mechanisms that block axon degeneration (Figure 5). The most potent protection comes from overexpression of cytNMNAT1 or NMN deamidase, both of which block SARM1-dependent NAD⁺ depletion in the injured axon. By inference, NMNAT2 also works via this mechanism, thus when its levels decline in the injured axon, SARM1 becomes activated. This biochemical demonstration that NMNAT enzymes inhibit SARM1 activation is consistent with a recent study showing that the embryonic lethality of NMNAT2 knockout mice is rescued by the loss of SARM1. This provides strong genetic evidence that NMNAT2 functions upstream of SARM1 (Gilley et al, 2015, Figure 5). Weaker axonal protection is provided by overexpression of NAMPT or by treating neurons expressing NRK1 with NR. These latter manipulations do not block SARM1-dependent NAD⁺ consumption, but do lead to a large increase in steady-state NAD⁺ levels prior to axotomy that likely maintains NAD⁺ above a critical threshold for an extended period of time after SARM1 activation (Figure 5). As in other studies (Sasaki et al., 2009b, Di Stefano et al., 2015), we also observed that FK866, a NAMPT inhibitor, delays axon degeneration. We do not have an adequate explanation for this phenomenon; however, it should be noted that FK866 provides much weaker protection than that afforded by NAMPT (Figure 1, and Figure 1—source data 1).

Figure 5

Download asset Open asset

In this work we demonstrate that blockade of SARM1 activation and axonal protection is not due to an increase in NAD⁺ or a decrease in NMN. How then do NMNAT enzymes (and NMN deamidase) block SARM1-stimulated axon degeneration and NAD⁺ loss? One possibility is that these enzymes physically interact with SARM1 to inhibit its ability to activate axon degeneration pathways including those involved in NAD⁺ loss. However, NMNAT enzymes from many species, including the archean Methanocaldococcus jannaschii, provide dramatic axonal protection despite their minimal amino acid sequence similarity, thus making direct interaction with SARM1 an unlikely mechanism (Sasaki et al., 2009b). Second, the necessity for increased NAD⁺ levels in a specific axonal compartment is an attractive explanation, however there is no evidence for this mechanism. Protein transduction experiments with cytNMNAT1 show rapid and strong axon protection, and it seems unlikely that a mutant nuclear protein would end up in a specific axonal compartment necessary to promote axon protection (Sasaki et al., 2009a). Finally, mutations in both NMNAT enzymes and NMN deamidase that destroy their enzymatic activity also block their axon protective capabilities (Araki et al., 2004, Sasaki et al., 2009b, Di Stefano et al., 2015). This strongly suggests that metabolites whose levels are altered by NMNAT and/or NMN deamidase activity regulate SARM1 function. NMN was such a candidate metabolite because it is a substrate for both NMNAT enzymes and NMN deamidase; however, our studies show that NMN levels do not regulate NAD⁺ loss and axon degeneration, both SARM1-induced activities. Further investigations to identify other cellular metabolites modulated by NMNAT or NMN deamidase may help elucidate the mechanism of SARM1 regulation and axonal degeneration.

These studies support the hypothesis that SARM1-dependent NAD⁺ consumption is the central biochemical event in the axonal degeneration program. Further, they explain the phenotype of the wlds mutant mouse as one caused by the continued inhibition of SARM1 by the longer-lived Wlds protein that compensates for the loss of NMNAT2. Finally, these findings suggest that inhibitors of SARM1 or agents that boost NAD⁺synthesis are likely to be valuable approaches for blocking axonal degeneration in the injured or diseased nervous system.

1. 1. Araki T
   2. Sasaki Y
   3. Milbrandt J

   (2004) Increased nuclear NAD biosynthesis and SIRT1 activation prevent axonal degeneration

   Science 305:1010–1013.

   https://doi.org/10.1126/science.1098014

   • Google Scholar
2. 1. Babetto E
   2. Beirowski B
   3. Janeckova L
   4. Brown R
   5. Gilley J
   6. Thomson D
   7. Ribchester RR
   8. Coleman MP

   (2010) Targeting NMNAT1 to axons and synapses transforms its neuroprotective potency in vivo

   Journal of Neuroscience 30:13291–13304.

   https://doi.org/10.1523/JNEUROSCI.1189-10.2010

   • Google Scholar
3. 1. Coleman M

   (2005) Axon degeneration mechanisms: commonality amid diversity

   Nature Reviews Neuroscience 6:889–898.

   https://doi.org/10.1038/nrn1788

   • Google Scholar
4. 1. Conforti L
   2. Gilley J
   3. Coleman MP

   (2014) Wallerian degeneration: an emerging axon death pathway linking injury and disease

   Nature Reviews Neuroscience 15:394–409.

   https://doi.org/10.1038/nrn3680

   • Google Scholar
5. 1. Di Stefano M
   2. Nascimento-Ferreira I
   3. Orsomando G
   4. Mori V
   5. Gilley J
   6. Brown R
   7. Janeckova L
   8. Vargas ME
   9. Worrell LA
   10. Loreto A
   11. Tickle J
   12. Patrick J
   13. Webster JR
   14. Marangoni M
   15. Carpi FM
   16. Pucciarelli S
   17. Rossi F
   18. Meng W
   19. Sagasti A
   20. Ribchester RR
   21. Magni G
   22. Coleman MP
   23. Conforti L

   (2015) A rise in NAD precursor nicotinamide mononucleotide (NMN) after injury promotes axon degeneration

   Cell Death and Differentiation 22:731–742.

   https://doi.org/10.1038/cdd.2014.164

   • Google Scholar
6. 1. Elliott GC
   2. Rechsteiner MC

   (1982) Evidence for a physiologically active nicotinamide phosphoribosyltransferase in cultured human fibroblasts

   Biochemical and Biophysical Research Communications 104:996–1002.

   https://doi.org/10.1016/0006-291X(82)91348-1

   • Google Scholar
7. 1. Gerdts J
   2. Sasaki Y
   3. Vohra B
   4. Marasa J
   5. Milbrandt J

   (2011) Image-based screening identifies novel roles for IkappaB kinase and glycogen synthase kinase 3 in axonal degeneration

   Journal of Biological Chemistry 286:28011–28018.

   https://doi.org/10.1074/jbc.M111.250472

   • Google Scholar
8. 1. Gerdts J
   2. Summers DW
   3. Sasaki Y
   4. DiAntonio A
   5. Milbrandt J

   (2013) Sarm1-mediated axon degeneration requires both SAM and TIR interactions

   Journal of Neuroscience 33:13569–13580.

   https://doi.org/10.1523/JNEUROSCI.1197-13.2013

   • Google Scholar
9. 1. Gerdts J
   2. Brace EJ
   3. Sasaki Y
   4. DiAntonio A
   5. Milbrandt J

   (2015) SARM1 activation triggers axon degeneration locally via NAD⁺ destruction

   Science 348:453–457.

   https://doi.org/10.1126/science.1258366

   • Google Scholar
10. 1. Gerdts J
    2. Summers DW
    3. Milbrandt J
    4. DiAntonio A

    (2016) Axon Self-Destruction: New Links among SARM1, MAPKs, and NAD+ Metabolism

    Neuron 89:449–460.

    https://doi.org/10.1016/j.neuron.2015.12.023

    • Google Scholar
11. 1. Gilley J
    2. Coleman MP

    (2010) Endogenous Nmnat2 is an essential survival factor for maintenance of healthy axons

    PLoS Biology 8:e1000300.

    https://doi.org/10.1371/journal.pbio.1000300

    • Google Scholar
12. 1. Gilley J
    2. Orsomando G
    3. Nascimento-Ferreira I
    4. Coleman MP

    (2015) Absence of SARM1 rescues development and survival of NMNAT2-deficient axons

    Cell Reports 10:1974–1981.

    https://doi.org/10.1016/j.celrep.2015.02.060

    • Google Scholar
13. 1. Hara N
    2. Shibata T
    3. Osago H
    4. Yamada K
    5. Tsuchiya M

    (2014) Formation of [nicotinamide-²H₃]NAD⁺ from [²H₄]nicotinamide and [²H₄]nicotinic acid in human HepG2N cells and involvement of ²H/¹H exchange at the redox site of NAD⁺/NADH

    Journal of Nutritional Science and Vitaminology 60:17–21.

    https://doi.org/10.3177/jnsv.60.17

    • Google Scholar
14. 1. Hikosaka K
    2. Ikutani M
    3. Shito M
    4. Kazuma K
    5. Gulshan M
    6. Nagai Y
    7. Takatsu K
    8. Konno K
    9. Tobe K
    10. Kanno H
    11. Nakagawa T

    (2014) Deficiency of nicotinamide mononucleotide adenylyltransferase 3 (nmnat3) causes hemolytic anemia by altering the glycolytic flow in mature erythrocytes

    Journal of Biological Chemistry 289:14796–14811.

    https://doi.org/10.1074/jbc.M114.554378

    • Google Scholar
15. 1. Lois C
    2. Hong EJ
    3. Pease S
    4. Brown EJ
    5. Baltimore D

    (2002) Germline transmission and tissue-specific expression of transgenes delivered by lentiviral vectors

    Science 295:868–872.

    https://doi.org/10.1126/science.1067081

    • Google Scholar
16. 1. Mack TG
    2. Reiner M
    3. Beirowski B
    4. Mi W
    5. Emanuelli M
    6. Wagner D
    7. Thomson D
    8. Gillingwater T
    9. Court F
    10. Conforti L
    11. Fernando FS
    12. Tarlton A
    13. Andressen C
    14. Addicks K
    15. Magni G
    16. Ribchester RR
    17. Perry VH
    18. Coleman MP

    (2001) Wallerian degeneration of injured axons and synapses is delayed by a Ube4b/Nmnat chimeric gene

    Nature Neuroscience 4:1199–1206.

    https://doi.org/10.1038/nn770

    • Google Scholar
17. 1. Osterloh JM
    2. Yang J
    3. Rooney TM
    4. Fox AN
    5. Adalbert R
    6. Powell EH
    7. Sheehan AE
    8. Avery MA
    9. Hackett R
    10. Logan MA
    11. MacDonald JM
    12. Ziegenfuss JS
    13. Milde S
    14. Hou YJ
    15. Nathan C
    16. Ding A
    17. Brown RH
    18. Conforti L
    19. Coleman M
    20. Tessier-Lavigne M
    21. Züchner S
    22. Freeman MR

    (2012) dSarm/Sarm1 is required for activation of an injury-induced axon death pathway

    Science 337:481–484.

    https://doi.org/10.1126/science.1223899

    • Google Scholar
18. 1. Revollo JR
    2. Grimm AA
    3. Imai S

    (2004) The NAD biosynthesis pathway mediated by nicotinamide phosphoribosyltransferase regulates Sir2 activity in mammalian cells

    Journal of Biological Chemistry 279:50754–50763.

    https://doi.org/10.1074/jbc.M408388200

    • Google Scholar
19. 1. Sasaki Y
    2. Araki T
    3. Milbrandt J

    (2006) Stimulation of nicotinamide adenine dinucleotide biosynthetic pathways delays axonal degeneration after axotomy

    Journal of Neuroscience 26:8484–8491.

    https://doi.org/10.1523/JNEUROSCI.2320-06.2006

    • Google Scholar
20. 1. Sasaki Y
    2. Vohra BPS
    3. Baloh RH
    4. Milbrandt J

    (2009a) Transgenic mice expressing the Nmnat1 protein manifest robust delay in axonal degeneration in vivo

    Journal of Neuroscience 29:6526–6534.

    https://doi.org/10.1523/JNEUROSCI.1429-09.2009

    • Google Scholar
21. 1. Sasaki Y
    2. Vohra BPS
    3. Lund FE
    4. Milbrandt J

    (2009b) Nicotinamide mononucleotide adenylyl transferase-mediated axonal protection requires enzymatic activity but not increased levels of neuronal nicotinamide adenine dinucleotide

    Journal of Neuroscience 29:5525–5535.

    https://doi.org/10.1523/JNEUROSCI.5469-08.2009

    • Google Scholar
22. 1. Sasaki Y
    2. Milbrandt J

    (2010) Axonal degeneration is blocked by nicotinamide mononucleotide adenylyltransferase (Nmnat) protein transduction into transected axons

    Journal of Biological Chemistry 285:41211–41215.

    https://doi.org/10.1074/jbc.C110.193904

    • Google Scholar
23. 1. Szretter KJ
    2. Samuel MA
    3. Gilfillan S
    4. Fuchs A
    5. Colonna M
    6. Diamond MS

    (2009) The immune adaptor molecule SARM modulates tumor necrosis factor alpha production and microglia activation in the brainstem and restricts West Nile Virus pathogenesis

    Journal of Virology 83:9329–9338.

    https://doi.org/10.1128/JVI.00836-09

    • Google Scholar
24. 1. Wang J
    2. Zhai Q
    3. Chen Y
    4. Lin E
    5. Gu W
    6. McBurney MW
    7. He Z

    (2005) A local mechanism mediates NAD-dependent protection of axon degeneration

    The Journal of Cell Biology 170:349–355.

    https://doi.org/10.1083/jcb.200504028

    • Google Scholar
25. 1. Wang JT
    2. Medress ZA
    3. Barres BA

    (2012) Axon degeneration: molecular mechanisms of a self-destruction pathway

    The Journal of Cell Biology 196:7–18.

    https://doi.org/10.1083/jcb.201108111

    • Google Scholar

Author details

1. Yo Sasaki

   Department of Genetics, Washington University School of Medicine, Saint Louis, United States

   Contribution

   YS, Designed and performed experiments and analyzed data and prepared the manuscript

   For correspondence

   ysasaki@genetics.wustl.edu

   Competing interests

   YS: May derive benefit from a licensing agreement with ChromaDex, which did not provide any support for this work

   "This ORCID iD identifies the author of this article:" 0000-0003-0024-0031

2. Takashi Nakagawa

   Frontier Research Core for Life Sciences, University of Toyama, Toyama, Japan

   Contribution

   TN, Designed and performed experiments and analyzed data

   Competing interests

   No competing interests declared.

3. Xianrong Mao

   Department of Genetics, Washington University School of Medicine, Saint Louis, United States

   Contribution

   XM, Designed and performed experiments

   Competing interests

   No competing interests declared.

4. Aaron DiAntonio

   Department of Developmental Biology, Washington University School of Medicine, Saint Louis, United States

   Contribution

   AD, Designed and supervise experiments and prepared the manuscript, Analysis and interpretation of data

   For correspondence

   diantonio@wustl.edu

   Competing interests

   No competing interests declared.

   "This ORCID iD identifies the author of this article:" 0000-0002-7262-0968

5. Jeffrey Milbrandt

   Department of Genetics, Washington University School of Medicine, Saint Louis, United States

   Contribution

   JM, Designed and supervise experiments and prepared the manuscript, Analysis and interpretation of data

   For correspondence

   jmilbrandt@wustl.edu

   Competing interests

   JM: May derive benefit from a licensing agreement with ChromaDex, which did not provide any support for this work

• 5,594

  views

• 978

  downloads

• 164

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.